Embed Size (px)
Citation preview
Time to eat:
Links between neuronal function and cellular phagocytosis
Elizabeth Stone
Submitted in partial fulfillment of the requirements for the degree of
Doctor of Philosophy under the Executive Committee
of the Graduate School of Arts and Sciences
COLUMBIA UNIVERSITY
2015
©2015
Elizabeth Stone
All rights reserved
ABSTRACT
Time to eat: Links between neuronal function and cellular phagocytosis
Elizabeth Stone
How do the brain and the immune system interact, and what are the consequences of this
interaction on the physiology of an organism during infection? The main focus of my
thesis is neuroimmune interaction, as studied in the following: (1) circadian regulation of
immune system function, specifically phagocytosis by immune cells during bacterial
infection; (2) the impact of circadian-regulated metabolism and feeding behavior on
immunity and host tolerance of bacterial infection; and (3) immune system function in
the context of Fragile X syndrome, a neurological disease known to cause circadian
dysregulation. To investigate the interactions between these complex physiologies, I use
the well-characterized and genetically tractable Drosophila melanogaster animal model.
Each topic is briefly introduced in Chapter 1. Chapter 2 focuses on the body of work
identifying the circadian regulation of the immune system, particularly phagocytosis, by
immune cells during bacterial infection. Chapter 3 highlights findings regarding how diet
and host metabolic state impact survival after infection. Chapter 4 illustrates phagocytic
immune cell defects both systemically and in the brain in the Drosophila model of
Fragile X syndrome. Lastly the conclusions discuss how these three works have built on
our fund of knowledge of neuroimmune interactions and the future implications for these
results.
i
TABLE OF CONTENTS LIST OF GRAPHS, IMAGES, AND ILLUSTRATIONS ........................................... iv
ACKNOWLEDGEMENTS .......................................................................................... viii
DEDICATION ................................................................................................................. xi
CHAPTER 1: INTRODUCTION .................................................................................... 1
An introduction to the fly .............................................................................................................. 2
An introduction to circadian regulation ........................................................................................ 3
The molecular clock is the timekeeper for circadian rhythms ...................................................... 5
Clock neurons and the master clock ............................................................................................. 9
An introduction to the innate immune system of the fly ............................................................. 12
Circadian regulation of immunity ............................................................................................... 16
Circadian regulation of metabolism and immunity ..................................................................... 18
An introduction to Fragile X syndrome ...................................................................................... 19
FMRP is an RNA-binding protein that functions as a translational repressor ............................ 21
Synaptic plasticity is altered in Fragile X syndrome .................................................................. 23
Novel therapeutics to treat Fragile X syndrome ......................................................................... 24
CHAPTER 2: The circadian clock protein Timeless regulates phagocytosis of
bacteria in Drosophila ..................................................................................................... 26
Summary ..................................................................................................................................... 27
Abstract ....................................................................................................................................... 30
Introduction ................................................................................................................................. 32
Materials and Methods ............................................................................................................ 34
Results/Discussion .................................................................................................................... 39
ii
Acknowledgements ..................................................................................................................... 58
CHAPTER 3: Host tolerance of infection is promoted by dietary glucose and amino
acids through insulin and TORC2 signaling ................................................................ 59
Summary ..................................................................................................................................... 60
Abstract ....................................................................................................................................... 63
Introduction ................................................................................................................................. 65
Materials and Methods ............................................................................................................ 68
Results ......................................................................................................................................... 77
Discussion ................................................................................................................................... 94
Supplemental Materials ............................................................................................................... 97
Author Contributions ................................................................................................................ 102
Acknowledgments ..................................................................................................................... 103
CHAPTER 4: Phagocytic immune cell defects in the Drosophila model of Fragile X
syndrome ........................................................................................................................ 104
Summary ................................................................................................................................... 105
Abstract ..................................................................................................................................... 107
Introduction ............................................................................................................................... 108
Materials and Methods .......................................................................................................... 112
Results ..................................................................................................................................... 118
Discussion ................................................................................................................................. 130
Chapter 5: CONCLUSIONS ........................................................................................ 135
Phagocytosis is circadian-regulated in Drosophila ................................................................... 136
The circadian regulation of feeding and metabolism affects survival of infection in
Drosophila ................................................................................................................................ 104
iii
The Drosophila model of Fragile X syndrome exhibits defects in phagocytosis ..................... 105
REFERENCES .............................................................................................................. 145
APPENDIX I ................................................................................................................. 163
APPENDIX II: Phagocytosis is upregulated in dFMR1 hypomorphs ..................... 164
Abstract ..................................................................................................................................... 165
Materials and Methods .......................................................................................................... 166
Results ..................................................................................................................................... 169
iv
LIST OF GRAPHS, IMAGES, AND ILLUSTRATIONS
CHAPTER 1: INTRODUCTION .................................................................................... 1
Fig. 1. The molecular clock is comprised of negative transcriptional feedback loops ................. 7
Fig. 2. Drosophila use three main immune mechanisms to fight pathogens .............................. 13
CHAPTER 2: The circadian clock protein Timeless regulates phagocytosis of
bacteria in Drosophila ..................................................................................................... 26
Fig. 1. Tim activity has significant consequences for immunity. .............................................. 40
Fig. 2. Tim regulates resistance to infection by S. pneumoniae and S. marcescens .................. 42
Fig. 3. Resistance against S. pneumoniae correlates with Tim activity. .................................... 43
Fig. 4. Tim protein does not regulate melanization or AMP gene expression. .......................... 47
Fig. 5. Tim protein regulates an early stage of phagocytosis of bacteria by hemocytes. ........... 50
Fig. 6. Bead inhibition of phagocytosis in wild type flies eliminates the difference in survival
kinetics with Tim mutants after S. pneumoniae infection. .......................................................... 53
Fig. S1. Tim protein does not regulate AMP gene expression. ................................................. 56
CHAPTER 3: Host tolerance of infection is promoted by dietary glucose and amino
acids through insulin and TORC2 signaling ................................................................ 59
Fig. 1. Period mutants exhibit greater tolerance than isogenic controls during infection with B.
cepacia ........................................................................................................................................ 78
Fig. 2. Per mutants have lower metabolic resources and eat more than wild type flies. ........... 80
Fig. 3. Dietary restriction does not increase infection tolerance of either Per mutants or wild
type. ............................................................................................................................................. 83
Fig. 4. Dietary glucose and amino acids increase tolerance of B. cepacia
infection. ..................................................................................................................................... 85
v
Fig. 5. Insulin-like signaling mediates the glucose contribution to infection tolerance and the
TORC2 complex mediates the effects of dietary amino acids. ................................................... 88
Fig. 6. The effect of amino acids on survival of infection is facilitated by gut
microbiota. ..................................................................................................................... 92
Fig. S1. Antimicrobial peptides (AMPs) are induced by B. cepacia infection to similar
expression levels in Per mutants and wild type. ........................................................... 97
Fig. S2. Rapamycin reduces survival of B. cepacia infection. ..................................... 98
Fig. S3. Expression of Tsc1 and Tsc2 transcripts are induced by heat- shock. ............. 99
Fig. S4. Amino acid-stimulated tolerance is facilitated by the gut microbiota. .......... 100
Table S1. Primer sequences ........................................................................................ 101
CHAPTER 4: Phagocytic immune cell defects in the Drosophila model of Fragile X
syndrome ........................................................................................................................ 104
Fig. 1. dFMR1 mutants are less resistant to infections with specific pathogens. ...................... 119
Fig. 2. dFMR1 mutants exhibit altered immune system parameters. ....................................... 122
Fig. 3. Clearance of severed axons is delayed in dFMR1 mutants relative to wild-type
controls ...................................................................................................................................... 126
Fig. 4. Fewer activated glia are seen in dFMR1 mutants than in wild-type controls after axonal
severing ..................................................................................................................................... 127
Fig. 5. Time course of Draper expression after axonal severing shows that glial phagocytosis is
delayed but not abolished in dFMR1 mutants ........................................................................... 128
APPENDIX I ................................................................................................................. 163
Fig. 1. Per mutants are more tolerant of infection with P. aeruginosa, and starvation eliminates
this survival advantage .............................................................................................................. 163
vi
APPENDIX II: Phagocytosis is upregulated in dFMR1 hypomorphs ..................... 164
Fig. 1. dFMR1 hypomorphs survive longer than wild-type controls after infection with certain
pathogens. ................................................................................................................................. 169
Fig. 2. dFMR1 hypomorphs are more resistant to infection with S. pneumoniae and S.
marcenscens compared to wild-type controls ........................................................................... 170
Fig. 3. Antimicrobial peptide synthesis is not altered by reduced levels of dFMR1 ............... 172
Fig. 4. dFMR1 hypomorphs exhibit increased phagocytosis of bacteria by hemocytes ........... 173
Fig. 5. dFMR1 hypomorphs do not exhibit circadian variation in phagocytosis ...................... 174
Fig. 6. Inhibiting phagocytosis abolishes the survival benefit observed in dFMR1 hypomorphs
with S. pneumonia infection. ................................................................................................... 175
vii
ACKNOWLEDGEMENTS
Before anyone else, I need to thank my PhD mentor, Dr. Mimi Shirasu-Hiza. When I
walked into her office on that propitious day—almost five years ago now—to discuss
possible rotation projects, I had no forewarning that her scientific vision and ideas would
be so thoroughly compatible with my own. Since then, barely a week has gone by when I
have not popped into Mimi’s office simply to offer a new idea for an experiment or to
seek some feedback, only to find myself bouncing thoughts around with her for
sometimes hours on end. Without Mimi’s innovative perspectives, effective collaboration
and unwavering support, there is no way that I could have completed this work. I am
lucky to call Mimi my mentor and friend. Many, many thanks to you, Mimi.
To the Shirasu-Hiza lab members past and present, thank you for your continued
support, both scientifically and socially. Thanks to my fellow grad students, Victoria and
Vanessa, for your wildly entertaining stories and your help over the years with writing
and suggesting experiments. And Vanessa, my many thanks for your help with the
Listeria infections and melanization while I was pregnant. My daughter Rebecca thanks
you too. And Emily, your help with the hemocyte phagocytosis for the dFMR1 project
was crucial. I could not have gotten so far without you. And lastly, to all of the techs
and undergrads past and present, I can’t thank you enough for your help with ordering,
stocks, and all of the other things that come up in lab. Adam, Albert, Paul, Charlotte, and
especially Reed, you guys are awesome. And Reed, I can’t thank you enough for taking
over the dFMR1 project and putting so much blood, sweat, and tears into it. I know you
have had a love-hate relationship with this project since you were an undergrad, but see?
Isn’t it exciting?? Take care of my (non-biological) baby. It’s now in your very capable
viii
hands, and I know you will make the most of the project in my absence. And Jenn
Ziegenfuss—I know you’re not a member of the Shirasu-Hiza lab, but thank you so much
for all of your help troubleshooting glial phagocytosis with us. The project wouldn’t have
gotten to this point without your help.
I am so very grateful to my thesis committee members, Dr. Wes Grueber, Dr. Ai
Yamamoto, and Dr. Ron Liem. The three of you have had invaluable suggestions and
have kept me from spinning my wheels for too long on one aspect of my project or
another. Each of you has offered a unique contribution. There are many reasons why I
am especially thankful for Wes. Not only did you share your Drosophila and imaging
expertise and let me use your confocal, but you also for agreed to serve as a co-mentor
for my NRSA fellowship. Wes, you have been unbelievably supportive throughout my
PhD; thank you for your continuing mentorship. I also owe a big thanks to Dr. Tom
Jongens for agreeing to serve as my outside reader, serving as a co-mentor for my NRSA
fellowship, and offering vital advice on the immunity aspect of my dFMR1 project.
My many thanks to the MD-PhD program here at Columbia University. The
support from Dr. Patrice Spitalnik, Dr. Ron Liem, and Dr. Michael Shelanski has been
crucial to my success in the program. Thank you for all that you do. Jeffery Brandt is also
one of the best and most capable administrators that I have every encountered, and I can’t
thank you enough for your assistance with everything in the program, especially grant
submissions. Also, a huge thank you to my MD/PhD cohort and graduate friends outside
of the Shirasu-Hiza lab for your continuing love and support.
I have spent the last 26+ years either learning in school or working as a research
technician for a University, and I have to thank many educators throughout the years who
ix
have supported and encouraged my love of learning. Starting early, Mrs. Judy Lieb, my
Seminar teacher at Lynnewood Elementary School, was instrumental in fostering a
curiosity for puzzles and mysteries. I will never forget how Mrs. Sharon Levin and Mr.
Mark Levy from Akiba Hebrew Academy had confidence in me, even if I didn’t. It was
at Barnard College where my professors and classmates created an environment for me to
thrive and learn the confidence that I carry with me today. I am also very grateful to Dr.
Jen Darnell, for whom I was a research technician for three years after college. Jen
taught me the importance of controls and how to conduct the most rigorous experiment
for a particular question.
Without the continuing love and support from my family and friends, I would not
be here today. To my parents—I love you both very much, and I cannot thank you
enough for all that you have done for me at every stage of my life. For as long as I can
remember, my dad has nurtured my interest in science, and although he tried his hardest
to deter me, I am following in his footsteps. My mom has been enormously encouraging
and has taught me the importance of inclusiveness and compassion for others and to be
generous with my time. While both of my parents have pushed me to try my hardest,
they also allowed me to choose my own path with my hobbies and my career. Thank
you, thank you, thank you, Mom and Dad. To my brothers, Eric and Martin—Eric, thank
you for giving me unrelenting tough love from the age of five until you went off to
college; you’ve taught me to stand up for myself. And Martin, thank you for protecting
me from Eric’s onslaught; you taught me that I could rely on the help of others. I am
extremely lucky to have you both in my life. Thank you, bros, and I love you both. Many
x
thanks to my Grandmom, Selma Kohn, who has been my number one fan and who has
encouraged me every step of the way.
Last but not least, I thank my husband, Zachary Jacobs, for his incredible support
through the years and my daughter, Rebecca Jacobs, for sharing Mommy with her
passion for science and medicine. It hasn’t always been easy, but Zachary has always
been there for me. From late night studying in medical school, editing my qualifying
exam and my grants, trying to make sense of my science, listening to endless practice
talks, and putting up with late nights and weekends in lab, I could not ask for a more
supportive partner. Zachary and Rebecca, I love both of you to the moon and back, and I
can’t thank the two of you enough.
xi
DEDICATION I dedicate this work to my daughter, Rebecca Jacobs. I hope that this work inspires you to
choose your own adventure, to try your hardest to be the best person you can be, and to
make an impact. I love you, munchkin.
1
CHAPTER 1:
INTRODUCTION
2
How do the brain and the immune system interact, and what are the consequences of this
interaction on the physiology of an organism during infection? The main focus of my
work is neuroimmune interaction. The three major projects that I will discuss here are:
(1) circadian regulation of immune system function, specifically phagocytosis by immune
cells during bacterial infection; (2) the impact of circadian-regulated metabolism and
feeding behavior on immunity and host tolerance of bacterial infection; and (3) immune
system function in the context of Fragile X syndrome, a neurological disease known to
cause circadian dysregulation. I used the well-characterized Drosophila melanogaster
animal model to investigate these projects. These projects are highly representative of the
major interests of the Shirasu-Hiza lab, which utilizes Drosophila as a model organism to
understand novel interaction between complex systemic physiologies. I briefly introduce
each topic in Chapter 1. Chapter 2 focuses on the body of work identifying the circadian
regulation of the immune system, particularly phagocytosis, by immune cells during
bacterial infection. Chapter 3 highlights findings on how diet and host metabolic state
impact survival after infection. In Chapter 4, I illustrate two phagocytic immune cell
defects in the Drosophila model of Fragile X syndrome. Lastly, I will discuss how these
three works have built on our fund of knowledge of neuroimmune interactions in the
conclusions.
An introduction to the fly
We use Drosophila melanogaster as a tool to study neuroimmune interactions. There is a
rich tradition of using Drosophila to explore heredity and genetic interactions, beginning
with Charles W Woodworth and Thomas Hunt Morgan in the early 20th century. Flies are
3
genetically tractable and there are very powerful genetic tools available at our disposal.
In addition to fast generation times, a large number of animals are available per
experiment, increasing the power for statistical significance. Many of the applicable
circadian, metabolic, immunity, and neurological disease pathways are highly conserved,
both molecularly and functionally, between Drosophila and mammals [1-4]. Moreover,
these pathways and physiologies are often simplified in Drosophila. For example,
Drosophila has only an innate immune system while mammals have both an adaptive and
innate immune system, and typically there are also fewer gene family members [3]. Thus,
the fly is an ideal organism to address the complex questions that we ask in lab.
An introduction to circadian regulation
Circadian rhythms are ancient, evolutionarily conserved 24-hour fluctuations in
molecular, physiological, and behavioral processes. These daily fluctuations, while
influenced by external cues such as sunlight, continue even in the absence of external
cues, suggesting an internal “clock” that keeps time for the organism. The first historical
account of circadian regulation was reported in the fourth century B.C.E. when
Adrosthenes, a ship captain for Alexander the Great, recorded that the leaves of the
tamarind tree exhibited daily, regular movements [5]. It was not until approximately 2100
years later when the first recorded experiment testing endogenous rhythms took place; in
1729, Jean-Jacques d'Ortous de Mairan, an astronomer, observed that when he placed a
Mimosa pudica plant in the dark, the leaves still opened and closed with the same rhythm
as a plant subjected to natural light [5]. Plants exhibit diurnal rhythms by opening leaves
and flowers to maximize sunlight exposure during the day while closing their leaves and
4
flowers at night to minimize the surface area and water loss. Each species of animal has
its own rhythm that controls when the animal wakes, eats, and sleeps. Humans have very
specific circadian regulation of meal times, waking activity, and sleeping. Some of the
physiological and behavioral processes regulated by circadian rhythms include sleep,
learning and memory, feeding, metabolism, and immune system function [6-8]. The
molecular mechanisms governing the circadian regulation of these processes are still not
well understood.
Circadian regulation is set by numerous zeitgebers, or exogenous synchronizers
[9]. Zeitgebers are thought to reset the internal body clock to synchronize the internal
body clock to the environment. One powerful external cue is the cycling wavelengths and
brightness of light [9]. For diurnal animals, including humans, squirrels, butterflies, and
most lizards, the presence of light stimulates wakefulness. However, raccoons, owls, and
other nocturnal animals sleep when it is light and are active at night [10]. These animals
are presumed to have adapted to nocturnal schedules either to avoid predators or to hunt
prey that have adapted to wakefulness at night. Still other animals are most active at
dawn and at dusk but can display wakefulness during the day or night; these animals are
said to exhibit crepuscular rhythms. Some examples of animals that display crepuscular
rhythms are various insects, including Drosophila and mosquito species, bats in the
suborder Megachiroptera, pygmy rabbits, and species of deer [11-15]. Temperature
fluctuations and food intake are two other strong zeitgebers [16,17], and in humans and
Drosophila, it is has been hypothesized that social interactions can be as strong as the
other external cues described [18], but in humans it may be the effect of altered light
exposure instead of social interactions on the circadian cycle [18,19].
5
The molecular clock is the timekeeper for circadian rhythms
Drosophila have been crucial for our understanding of the molecular clock. The first
circadian regulatory gene was isolated through a forward genetic screen in Drosophila. In
their seminal report published in 1971, Seymour Benzer and Ronald Konopka described
three Drosophila mutants with abnormal eclosion rhythms and altered circadian
locomotor activity[20]. In populations of wild-type flies, eclosion rhythms and circadian
locomotor activity have periods of approximately 24 hours; Benzer and Konopka found
that one of their mutants was completely arrhythmic, one mutant had a short period
averaging 19 hours, and one mutant had a long period averaging 28 hours. Using elegant
classical genetics techniques, they showed that these three mutations mapped to the same
locus on the X chromosome, and they concluded that these are three different alleles of
the same gene, which they named period. These experiments were the first demonstration
of a circadian regulatory gene in any organism.
Since these experiments, PERIOD (PER) has been shown to be a critical part of
the basic, evolutionarily conserved four component molecular clock, which acts in a
negative transcriptional feedback loop to regulate circadian rhythm. In Drosophila, the
four essential components for circadian regulation are CLOCK (CLK), CYCLE (CYC),
PERIOD (PER) and TIMELESS (TIM) [21]. Rhythms are generated by negative
transcriptional feedback loops by the molecular clock. Briefly, the transcriptional
activators CLK and CYC, which have basic helix-loop-helix (bHLH) and PAS domains,
important for DNA binding and protein-protein interactions, respectively, activate the
expression of PER and TIM, along with many target genes, slowly throughout the day by
6
binding to enhancer elements (E-box) of their promoter regions [22]. Although TIM does
not have any appreciable DNA-binding domain, PER has a PAS domain, which is known
to interact with bHLH domains [23]. At night, when levels of PER and TIM are at their
peak, PER/TIM translocate to the nucleus together where they repress the binding of
CLK/CYC to the PER and TIM promotors [21], essentially inhibiting their own
transcription. TIM is then degraded in the presence of light [24], resulting in the release
of the inhibition of the transcription of per and tim. The levels of these proteins oscillate
with a period of approximately 24 hours, and light entrains, or regulates the period of, the
cycling of these proteins (Fig 1A).
There are many other enzymes and proteins that interact with the CLK and CYC
(Fig 1A). CLK/CYC also promote the transcription of many other genes. The proteins of
two of these genes, vrille (vri) and PAR domain protein 1ε (pdp1ε), act as a repressor and
activator, respectively, of CLK transcription [25]. VRI is a basic leucine zipper
transcription factor, while PDP1ε has a transcriptional activating domain [25]. These two
proteins work in opposition in a second negative feedback loop to regulate the mRNA
transcript levels of clk throughout the course of the day [26], but with little effect on CLK
protein expression. Clockwork orange (CWO) is another transcription factor that is
circadian regulated and is a bHLH repressor, preventing the transcription of the target
genes of CLK/CYC [26] (Fig 1B).
Numerous proteins, including kinases and phosphatases, either promote or reduce
the stability of PER and TIM (Fig 1B). For instance, doubletime (DBT) is a kinase that
phosphorylates PER and leads to its degradation by the proteasome [27]. Interestingly,
when PER is associated with TIM, DBT is unable to phosphorylate PER, preventing the
7
Figure 1. The molecular clock is comprised of negative transcriptional feedback loops. (A) Here is the basic negative feedback loop that makes up the molecular clock in Drosophila, which has the transcriptional regulators CLOCK (CLK) and CYCLE (CYC) binding to the promoters of the proteins PERIOD (PER) and TIMELESS (TIM). TIM and PER then block the association of CLK and CYC with the per and tim promoters. CLK and CYC also bind to the promoters of PAR domain protein 1ε (pdp1ε) and vrille (vri); the protein products of these genes, PDP1ε and VRI, promote or inhibit the transcription of clk, respectively, thus further modifying the feedback loop. Adapted from Zheng and Sehgal, 2008 [28].
8
(B) The post-translational modification of PER and TIM further regulate the circadian clock. TIM binds PER and prevents PER degradation. In the presence of light during the day, cryptochrome (CRY) binds TIM, targeting TIM for degradation through the E3 ligase JETLAG. The absence of TIM during the day causes PER to be phosphorylated by doubletime (DBT), and the E3 ligase supernumerary limbs (SLIMB) then targets PER for degradation by the proteasome. Protein phosphatase 2A (PP2A) and protein phosphatase I (PP1) dephosphorylate PER and TIM, respectively, stabilizing them from degradation. Additionally, the transcription factor and bHLH repressor clockwork orange (CWO) binds blocks the association of CLK and CYC to their DNA targets. Adapted from Zheng and Sehgal, 2008 [28]. (C) The mammalian homologs of the molecular clock are regulated in a similar fashion. Mammalian CLK and BMAL (the CYC homology) drive the expression of mPer and mCry (the TIM homolog). Casein Kinase I, the functional homolog of DBT, phosphorylates BMAL, mPer, and mCry, targeting them for degradation. The F Box E3 ligase FBXL3 further targets mCry for degradation by the proteasome. Protein phosphatase 1 (PP1), on the other hand, dephosphorylates mPer and mCry, stabilizing them from degradation. Adapted from Ko and Takahashi, 2006 [29].
degradation of PER [30]; thus TIM stabilizes PER. The E3 ligase supernumerary limbs
(SLIMB) targets phosphorylated PER for degradation by the proteasome [27]. Shaggy
(SGG), the fly homolog of glycogen synthase kinase 3β, phosphorylates TIM [31]. Two
phosphatases, protein phosphatase 2 (PP2) [32] and protein phosphatase 1 (PP1) [33]
dephosphorylate and stabilize PER and TIM, respectively, which prevents their
9
degradation. In the presence of blue light, the photoreceptor crytochrome (CRY) signals
for the degradation of TIM through JETLAG, an E3 ligase, and the proteasome [34].
These proteins all contribute to the length and timing of the rhythms of the molecular
clock.
Many of the same players are also involved in the molecular clock in mammals
(Fig 1C). For instance, CLOCK and BMAL1 (the mammalian homolog of CYC)
promote the transcription of the mammalian PER homologues (mPer1 and mPer2) and
crytochromes (mCry1 and mCry2), in the place of TIM. Casein kinase 1 (CK1), most
homologous to DBT, phosphorylates BMAL1 [35], mPer [36], and mCry [35] and targets
them for degradation. PP1 dephosphorylates and promotes the stability of mPer [37].
FBXL3, an E3 ligase F box protein, targets mCry for degradation by the proteasome [38].
Just as circadian regulation in Drosophila is made up of two negative
transcriptional feedback loops, the mammalian clock also has a second transcriptional
loop, in which CLK and BMAL1 drive the transcription of two retinoic acid-related
orphan nuclear receptors, Rorα and Rev-erbα [29]. RORα and REV-ERBα then
translocate into the nucleus and bind to segments of the Bmal1 promoter containing
retinoic acid related orphan response elements (ROREs). RORα promotes the
transcription of Bmal1 while REV-ERBα inhibits Bmal1 transcription.
Clock neurons and the master clock
Although the molecular clock is expressed in the brain and in many other tissues, specific
neurons in the brain are the master timekeeper [39]. In the brain, clock expression is
restricted to specific regulatory areas. These neurons are collectively called the central
10
clock, and they are thought to synchronize and regulate all other clocks throughout the
body, ultimately entraining an organism to the exogenous environment. The central clock
in Drosophila is governed by 150 so-called clock neurons [39]. Clock neurons are
centrally located in the brain and express the molecular clock described above. There are
7 groups of clock neurons in the brain: dorsal neurons (DN1, DN2, and DN3), small and
large ventral lateral neurons (sLNv and lLNv), dorsal lateral neurons (LNd), and the
lateral posterior neurons (LPN).
A small subset of clock neurons in the Drosophila brain is important for
regulating increased activity in anticipation of the change in light during dawn and dusk.
The morning peak is controlled by the sLNv [24], which are active in the morning, while
the DN1 and LNd control the evening peak; these populations of neurons are referred to
as the M and E cells, respectively [40]. In addition to the molecular clock, the sLNv
secrete the neuropeptide pigment dispersing factor PDF and signal to the E cells in the
morning, while the E cells signal back to the M cells in the evening [24]. The presence of
PDF in the M cells as well as CRY drives the morning activity, while the PDF-negative
and CRY-positive E cells direct the evening activity [40]. PDF expression by the sNLv
has two other important roles beside dictating the timing of the morning anticipation:
PDF is used both to synchronize clock neurons within the brain to ensure that all neurons
are cycling with the same period and to regulate the output of clock neurons to the rest of
the body. For example, one output of PDF-secreting neurons sends projections to the
mushroom body, the fly structure analogous to the hippocampus, which is important for
the sleep-wake cycle of the fly and for learning and memory [21,41,42].
Similar to Drosophila, mammalian circadian regulation is directed by a central
11
clock and synchronizing signals regulate peripheral clocks [43]. The mammalian central
clock is located in the suprachiasmatic nucleus (SCN) of the hypothalamus. In the
presence of light, the retina signals to the SCN to entrain the molecular oscillators in the
SCN to the exogenous environment. The SCN then communicates to peripheral tissues
through endocrine signaling [44,45], neuronal signaling via the autonomic nervous
system [46], and temperature changes [47,48]. Interestingly, unlike peripheral tissues,
which require input from endocrine signaling and temperature regulation from the SCN,
intercellular synchronization in the SCN makes its molecular oscillator remarkably
resistant to disruptions from endocrine signaling or temperature changes [49]. Thus, the
job of the SCN is to be in sync with external light cues, while the clocks of peripheral
tissues are more malleable and can respond to the local environment [50].
With the exception of the testis [51], almost every other tissue of the body that has
been tested expresses a cycling clock [47]; an incomplete list of tissues with peripheral
clocks includes the heart, adrenal gland, ovary, muscle, kidney, spleen, lung, and liver.
As an example, the liver is entrained by four different mechanisms [52]. First, autonomic
nervous system innervation of the liver contributes to the cycling of Per1 and Per2 in
addition to two glucose enzymes, PEPCK and GLuT2 [46]. Second, hormonal secretion
modifies downstream targets in the liver. For instance, rhythmic secretion of
glucocorticoids alters the expression of approximately 60% of the transcriptome of the
liver through the transcription factor HNF4α [53]. Additionally, the expression of many
nuclear hormone receptors is cyclic in the liver [54], affecting metabolic function at
different times of the day. Third, the temperature of the animal can change the cycling of
circadian genes in the liver [55] and all other peripheral clocks that have been tested [56].
12
Finally, feeding during the organism’s sleep phase, but not during the organism’s wake
cycle, strongly shifts the cycling of the liver, but not the SCN [57].
An introduction to the innate immune system of the fly
Drosophila innate immunity has at least three well-characterized branches that work
together to fight pathogens [58]: (1) antimicrobial peptide production, (2) reactive
oxygen species synthesis resulting in melanization, and (3) phagocytosis, or the
engulfment of pathogens by hemocytes. Induction of each of these arms of the
Drosophila innate immune system occurs following infections with different pathogens.
After systemic infection with bacteria or fungi, two different signaling cascades,
the Toll and imd pathways, culminate in the induction of the Drosophila antimicrobial
peptides in the fat body, analogous to the mammalian liver [58]. The Toll pathway, first
discovered in Drosophila as a developmental patterning gene [59], was published in the
pivotal work by Bruno Lemaitre and Jules Hoffman, where they reported that the Toll
pathway is important for immune defense against both fungal and bacterial infections
[60]. Fungi and bacteria expressing lys-type peptidoglycan (generally Gram positive
bacteria) preferentially activate the Toll pathway via specific peptidoglycan receptor
proteins (PGRPs) and activation of the receptor Toll, while bacteria expressing DAP-type
peptidoglycan (generally Gram negative bacteria) activate the immunity deficiency (imd)
pathway [60] via different PGRP receptors and other downstream effectors [61] (Fig 2A).
The two downstream NF-κb family members, Dif/Cactus and Relish, respectively, and
result in the transcription of antimicrobial peptides and other genes. Typically,
13
Figure 2. Drosophila use three main immune mechanisms to fight pathogens. (A) Antimicrobial peptides, produced in response to Toll and imd signaling, block pathogen growth by disrupting pathogen membranes [61]. In the presence of Gram positive bacteria and fungi, the Toll signaling cascade is
vertebrates
DD DD
DD dMyD88
Tube Pelle
DIF
Cactus
Drosomycin Other AMPs
DD DD
MyD88
IRAK
NF- B
I B
Proinflammatory cytokines
Drosophila
Toll TLR
Nucleus
PGRP
DD
IMD
DREDD dFADD
Relish
Diptericin Other AMPs
Proinflammatory cytokines
vertebrates Drosophila
Kenny (dIKK- ) Ird5
(dIKK- )
TNFR
DD
RIP
Caspase8 FADD
IKK-
IKK-
TNF-
NF- B
I B
A Toll/TLR signaling IMD/TNF- signaling
Adapted from Hoffmann and Reichhart, 2002
lysosome
Microbe recognition
ROS production Melanin deposition
Phenol Quinones
Phenoloxidase
B Melanization
fly
Microbe recognition
Internalization
Phagosome maturation
Immune cell
lysosome
C Phagocytosis
14
activated through the Toll-induced signaling complex (TISC), comprised of the Toll receptor (mammalian homologs are the Toll-like receptors, or TLRs) and 3 death domain (DD) containing proteins: dMyD88 (homologous to the mammalian MyD88), Tube, and Pelle, which is an IL-IR-associated kinase (IRAK) protein (homologous to mammalian IRAK). Cactus (mammalian homolog IκB), which is normally associated with Dif (mammalian homolog NF-κB), is phosphorylated and degraded, freeing Dif to translocate to the nucleus and transcribe certain antimicrobial peptides, including Drosomycin. Gram negative bacteria, on the other hand, bind to various PGRP receptors (homologous to mammalian TNF-α receptor, or TNFR), which binds to the DD-containing IMD (mammalian homolog RIP), along with DREDD (mammalian homolog Caspase 8) and dFADD (mammalian homolog FADD) to activate the imd pathway. Together with Kenny (mammalian homolog IKK-γ) and Ird5 (mammalian homolog IKK-β), DREDD and dFADD target Relish (another Drosophila homolog of the mammalian NF-κB) for cleavage. Cleaved Relish translocates to the nucleus to transcribe numerous antimicrobial peptides, including Diptericin. Adapted from Hoffmann and Reinhhart, 2002 [3]. (B) The melanization cascade produces reactive oxygen species, which are toxic to pathogens, and culminates in the deposition of melanin. The enzyme phenoloxidase over many steps converts phenol into quinones that are toxic to bacteria and results in the deposition of melanin, which is also used for encapsulation of parasites and clot formation in wounded flies [62]. (C) The process of phagocytosis has three basic steps: microbe recognition, internalization or engulfment of the pathogen into a phagosome, and the maturation of the phagosome by fusion with lysosomes. The resultant phagolysosome compartment is acidified, and the pathogen is destroyed.
researchers in the field infect Drosophila with M. luteus, which is Gram positive, and E.
coli, which is Gram negative, to induce the Toll pathway and the imd pathway,
respectively [63].
The mammalian homologues, the Toll-Like Receptor family (TLRs), were then
discovered in mice and their innate immune functions were investigated in the influential
work by Medzhitov et, al (1997) [64], Poltorak et al (1998) [65], and Hashino et al (1999)
[66]. The mammalian homologues of these pathways, the TLR pathway and interleukin-1
receptor (IL-1R) pathway, which are most similar to the Drosophila Toll pathways, and
the TNFα pathway, which is analogous to the Drosophila imd pathway, are very well
conserved (Fig 2A). In mammals, these pathways represent a crucial link between the
innate and adaptive arms of the immune system.
Melanization is an immune mechanism used by the fly to sequester and kill
pathogens by a combination of encapsulation and the generation of reactive oxygen
species (ROS) [67]. After wounding or infection, serine proteases, including MP1 and
15
MP2, activate a proteolytic cascade in which the enzyme Hayan cleaves pro-
phenoloxidase is into phenoloxidase, which then oxidizes phenols into quinones in the
hemolymph of the fly [68,69] . The quinones themselves and other ROS intermediates of
the proteolytic cascade are toxic to pathogens. Additionally, the cascade ends with the
deposition of melanin, which is not only toxic to pathogens, but it also isolates pathogens
from spreading throughout the hemolymph (Fig. 2B). Because the process of
melanization is also toxic to the fly, the induction of melanization is under tight control
[68]. There are three different protease inhibitors called serpins (Spn27A, Spn28D, and
Spn77Ba) that inhibit phenoloxidase activation to keep melanization in check. To induce
melanization in the lab, we use L. monocytogenes and S. typhimurium, and we can
observe melanin deposits through the cuticle [70].
And lastly, phagocytosis by hemocytes, or specialized immune cells in the
circulatory system, results in the engulfment and subsequent degradation/digestion of
pathogens and injured or dead cells in Drosophila [71]. Briefly, specialized receptors on
the hemocyte’s surface recognize the pathogen and signal for actin cytoskeletal
rearrangement to begin engulfment of the pathogen into an intracellular vesicle called the
phagosome. When the pathogen is completely engulfed, the phagosome fuses with
cellular lysosomes, acidifying the resulting phagolysosome compartment. The contents of
the resulting phagolysosome are then digested and broken down (Fig 2C). In lab, we use
fluorescent beads injected into flies to measure phagocytic activity by hemocytes [63].
The molecular mechanisms of phagocytosis are highly conserved between
vertebrates and lower organisms. Indeed, much of what we know about the molecular
mechanisms of phagocytosis comes from studies of invertebrates, unicellular organisms,
16
or tissue culture cells in vitro [71]. For example, RNAi screens of S2 cells, which are
immortalized phagocytic immune cells from Drosophila, have identified many of the
transmembrane receptors, cytoskeletal components, and signaling pathways required cell-
autonomously for phagocytosis by immune blood cells [72].
Circadian regulation of immunity
Sleep and infection have long been thought to be associated with one another. For
instance, sleep is crucial for healing after an illness or stressor [73]. During acute illness,
patients often sleep longer with an irregular cycle [73]. It has been demonstrated in
mammals that many cytokines affect sleep; pro-inflammatory cytokines generally
promote sleep while anti-inflammatory cytokines inhibit sleep [74], suggesting a link
between sleep and immune parameters. For example, the inflammatory cytokines IL-1α,
IL-1β, IL-6, and TNF-α all promote sleep in mammals [75,76]. Additionally, sleep
deprivation and sleep fragmentation lead to altered immune parameters in mammals, with
altered cytokine and chemokine expression and changes in the circadian regulation of
circulating leukocytes [73,77]. There have been a limited number studies examining how
levels of circulating blood cells in humans are affected after sleep deprivation. Two of
these studies demonstrated that the maximum number of circulating B and T cells and
monocytes was delayed significantly and that natural killer cells cycled out of phase with
a much smaller amplitude of oscillation when compared with controls with normal
amounts of sleep [78,79]. These studies suggest a link between sleep and immune
parameters.
The powerful genetic tools available in Drosophila make the organism an ideal
17
model to study the molecular interaction between sleep and immunity. Numerous genes
involved in Toll and imd signaling were found to be upregulated in a sleep deprivation
model, including Relish, the NF-κB homolog [80]. In addition, the circadian and sleep
mutant cyc01 exhibited increased levels of many immunity genes, including Relish, even
when their immune systems were not challenged [80]. Lastly, Relish mutants were found
to have disrupted sleep [80]. These results all suggest that sleep and immunity are linked
in Drosophila.
Sleep is a major behavioral output of circadian regulation but is certainly not the
only physiology regulated by the molecular clock. Although there have been many
reports demonstrating an association between sleep and cytokine expression/circulating
leukocyte numbers in mammals and sleep and immunity gene expression in Drosophila,
the molecular and functional connections between the immune system and circadian
components themselves have not been well understood. In a paper published in 2007, Dr.
Shirasu-Hiza, while working in Dr. David Schneider’s lab, found that when wild-type
flies were challenged with pathogens, these flies lost their circadian regulation [81].
Moreover, Tim01 and Per01 mutant flies lacking essential components of the molecular
clock were found to die more quickly than wild-type controls when infected with two
different pathogens, S. pneumoniae and L. monocytogenes. These results illustrate that
the lack of circadian regulation leads to altered immune system function. I expand on
these results in Chapter two, examining the specific immune mechanism underlying
defects in survival of infection of Tim01 mutants. In the Chapter summary, I discuss my
specific contributions to this work.
18
Circadian regulation of metabolism and immunity
There are two types of mechanisms that organisms employ to fight pathogens:
resistance and tolerance [82]. An organism uses resistance mechanisms to kill and clear
bacteria, resulting in a reduction of the pathogenic burden on the organism. Resistance
mechanisms are generally thought of as the traditional immune system defenses; for
example, the Drosophila immune system defenses that I described above, antimicrobial
peptides, melanization, and phagocytosis, are all resistance mechanisms. Tolerance
mechanisms give an organism the ability to withstand the pathogenic affects of an
infection, either from damage accumulated by the microbe or the defenses mounted by
the organism itself [83].
Tolerance is not very well characterized, but two examples of tolerance
mechanisms are feeding behaviors and metabolism. Feeding can affect an organism’s
survival to particular pathogens. For instance, in Drosophila, reducing nutrient intake
improves survival when flies are infected with S. typhimurium, E. coli, and E. caratova
but decreases survival rates when flies are infected with L. monocytogenes [84,85].
Additionally, when infected with M. marinum or L. monocytogenes, reduced survival
rates are associated with a decrease in metabolic stores [86,87]. In general, the
contributions of specific nutrients and metabolic pathways on tolerance mechanisms have
yet to be elucidated.
Feeding behavior and the expression of metabolic genes are circadian-regulated;
both Drosophila and mouse circadian mutants display changes in metabolism and altered
feeding behaviors [88,89]. Although we and others have demonstrated that resistance
phenotypes against particular pathogens is circadian-regulated [63,81,90], it is unknown
19
whether circadian regulation of these two tolerance mechanisms, feeding and
metabolism, will alter immunity against infection.
Chapter 3 contains a manuscript in revisions describing work in which we
examine how altering feeding behavior and diet in Drosophila can affect tolerance to the
fast and lethal infection with the human pathogen, B. cepecia. In the Chapter summary, I
describe my specific contributions to this manuscript.
Introduction to Fragile X Syndrome
Many neurological diseases are associated with circadian dysregulation, including
Alzheimer disease, Parkinson Disease, epilepsy, stroke, multiple sclerosis, and autism. It
is not yet clear how circadian dysregulation or immune system function affects disease
progression or symptoms. Here I studied immune system function in a Drosophila model
for Fragile X syndrome, a disorder known be associated with circadian dysregulation.
This section will introduce Fragile X syndrome, while Chapter 4 will describe my results.
Fragile X syndrome is the most common monogenic cause of intellectual
disability and autism, occurring in approximately 1:4000 males and 1:8000 females [91].
About 70% of patients with Fragile X syndrome also have autistic features [92], and
patients with Fragile X syndrome make up about 5-8% of all causes of autism [93,94].
Fragile X syndrome often presents in patients with difficulties in learning and memory;
behavioral problems, including anxiety, aggression, and attention; circadian
dysregulation; seizures; and other non-neurological symptoms, such as frequent ear
infections and gastrointestinal symptoms[92,95,96]. Patients with the disease often have
long facies, long limbs, and prominent ears; male patients with the disease also have
macroorchidism [96]. Loss of FMR1 in humans and animal models causes neurons to
20
display excessive growth of dendritic spines and defects in synaptic plasticity, but the
mechanism underlying the neuronal dysfunction is unknown [97,98].
In the vast majority of cases, Fragile X syndrome is caused by a triplicate repeat
CGG expansion in the promoter region of the gene FMR1. When there are more than 200
CGG repeats, the promoter region is hypermethylated and the FMR1 gene is
subsequently silenced [99]. Two other disorders occur from the premutation at this gene
locus from an intermediate number, between 55 and 200, of CGG repeats: Fragile X-
associated Tremor/Ataxia Syndrome (FXTAS), which is characterized by intellectual
disabilities and motor symptoms; and Fragile X-associated Premature Ovarian
Insufficiency (FXPOI), which results in the termination of menstruation in a woman
younger than 40 years old. FXTAS and FXPOI have been reviewed by Wheeler, et al
(2013) [100]. The FMR1 gene encodes an RNA-binding protein called Fragile X Mental
Retardation Protein (FMRP), which regulates the translation of many targets. FMRP has
two mammalian homologues, FXR1 and FXR2 (for Fragile X Related 1 and 2), and these
two proteins have some overlapping RNA-binding properties with FMRP [101]. FMRP is
ubiquitously expressed, but is enriched in neurons and in the testis. The role of FMRP has
been widely studied and fairly well characterized in neurons.
Fragile X syndrome was first described as an X-linked disorder when Martin and
Bell examined the phylogeny of a large family affected with the disease (1943) [102].
FMR1 was first cloned in 1991 by Verkerk, et al [103], and was found to be an RNA
binding protein, which was suggested by the presence of the RNA-binding KH domains
and RGG box [104-106]. In 1996, two independent groups showed that FMRP is
associated with ribosomes [107,108], and FMRP was later found to be a translational
21
repressor [109]. The vast majority of subsequent studies have focused on the absence of
FMRP on protein synthesis in neurons; the role that FMRP plays in other tissues and cell
types has not been well characterized.
The Drosophila and mouse models for Fragile X syndrome are often utilized to
study molecular, physiological, and behavior phenotypes caused by the absence of
FMRP, as well as the consequences of pharmacological interventions. Both of these
animal models recapitulate many of the symptoms seen in human patients with Fragile X
syndrome, including defects in learning and memory and circadian dysregulation
[110,111]. Mice, like humans, express FMRP and the two homologs, FXR1 and FXR2
[112]. Drosophila, on the other hand, has only one FMRP family member: dFMR1[4].
Drosophila dFMR1 has a high degree of similarity and identity [4]. FMRP shares
functional overlap with dFMR1, as driving the expression of human FMRP in a
Drosophila dFMR1 null animal rescued protein levels in the brain and synaptic structure
[113]. Most of the discoveries discussed below were conducted using these animal
models.
FMRP is an RNA-binding protein that functions as a translational repressor
FMRP has three RNA-binding domains that are crucial for RNA binding: 2 KH domains
(KH1 and KH2) and an RGG box. Although in vitro studies have illustrated that the KH
domains have been shown to bind to RNA kissing complexes [114] and the RGG box
binds to G-quartets [115], an in vivo assay examining where FMRP binds its mRNA
targets has shown that FMRP binds without any particular bias to the RNA coding
sequences [109]. In neurons, FMRP is closely associated with polyribosomes and
22
functions to stall polyribosomal translation of the mRNAs to which it binds [109]. The
RNA-binding ability of the KH2 domain is critical for this association, as a missense
mutation in the KH2 domain, I304N [116], abrogates the association of FMRP with
polyribosomes in sole human patient with this mutation[117] and a mouse model of this
mutation [116,118]. This patient has a particularly severe case of the disease [117]. In the
absence of FMRP, there is an increase in protein synthesis in the brain in general [119],
and, administration of an inhibitor of protein translation rescued long-term memory
dysfunction in a Drosophila model of Fragile X syndrome [120]. These findings
highlight the importance of FMRP’s role in stalling polyribosomal translation on
neuronal function.
To try to understand FMRP function in neurons, many studies have been
conducted to elucidate the mRNA targets to which FMRP binds. The first study
examining FMRP targets was conducted by Brown, et al (2001) and used
immunoprecipitation and microarrays to find that roughly 4% of mRNA transcripts in the
brain are potential FMRP targets [121]. Another early study found that dFMR1 binds
futsch mRNA [122], which is the Drosophila homolog of MAP1B, one of the targets that
Brown, et al highlighted. Zhang, et al (2001) determined that the absence of dFMR1
causes increased levels of FUTSCH protein, and that double dfmr1 null/futsch
hypomorph rescues defects in synaptic architecture in the fly [122]. While working as a
technician in the Darnell lab prior to my PhD, we found 842 mRNA targets enriched in
the pre- and postsynaptic compartments to which FMRP binds in vivo by utilizing a
stringent crosslinking-immunoprecipitation (CLIP) protocol and high-throughput RNA
sequencing [109,123]. FMRP binds to a selective group of targets. For instance, FMRP
23
interacts with NMDA receptor and mGluR subunits and downstream partners, but FMRP
does not bind to the AMPA receptor or its downstream affectors [109]. These results
suggest that ameliorating protein synthesis may rescue the morphological and behavioral
phenotypes observed in the absence of FMRP.
Synaptic plasticity is altered in Fragile X syndrome
On the cellular level, patients with Fragile X syndrome and animal models of the disease
have immature, thin dendritic spines on their post-synaptic neurons in contrast to the
more mature, mushroom-shaped spines found in controls [97,124]. The dendritic spine
density seen in Fragile X syndrome models and in patients suggests that synapses are not
being appropriately strengthened or matured.
Neurons rely on local protein translation in dendrites and axon terminals for
proper synaptic plasticity. Two forms of synaptic plasticity, long-term potentiation
(LTP), which strengthens synapses through activity, and long-term depression (LTD),
which weakens synaptic strength, both require rapid protein synthesis in postsynaptic
dendrites [125,126]. As FMRP acts as a translational repressor [127] that interacts with
many mRNAs in the pre- and postsynaptic compartment [109], it follows that the absence
of FMRP leads to disordered synaptic plasticity. While LTP is not affected in the absence
of FMRP [128], LTD was enhanced in the absence of FMRP [129]. In wild-type
conditions, excitatory metabotropic glutamate receptor (mGluR) stimulation, one
mechanism important for LTD, causes an increase in the translation of FMRP [130].
Indeed, Darnell et al (2011) found that the mGluR receptor itself is an FMRP target,
along with many downstream affecters of mGluR signaling [109]. Interestingly, even
24
though LTD generally requires protein synthesis to occur, Nosyreva and Huber found
that in the presence of a translational inhibitor, FMRP mutants still exhibited enhanced
LTD [131]. As FMRP knockouts exhibit increased LTD, these results suggest that the
absence of FMRP causes excessive protein translation downstream of mGluR signaling,
causing the increased LTD and altered synaptic plasticity. These results led to the mGluR
theory of Fragile X syndrome [132].
Novel therapeutics to treat Fragile X syndrome
In light of the recent studies examining how the loss of FMRP affects neurons, there are
some new therapeutics that are being tested on animal models and on human patients.
One drug in particular, CTEP, is an mGluR inhibitor that has shown to rescue a number
of phenotypes observed in mice, including, learning and memory and the dendritic spine
defect [133]. It is notable that treatment with CTEP in this study began in 4-5 week old
mice, which is after the mice already began to exhibit the Fragile X phenotypes [133].
These results suggest that CTEP may be a viable drug for patients with Fragile X
syndrome, even after they develop symptoms. In fact, there are a number of mGluR
inhibitors that are currently in drug trials [133].
Another avenue for therapeutics is general protein synthesis inhibition. When
dFMR1 mutant flies were treated with the protein synthesis inhibitors cyclohexamide and
puromycin, the defect observed in learning was ameliorated [120]. Although these drugs
are too toxic for use in humans, these results demonstrate that inhibiting protein synthesis
in general is a possible route for drug treatment. One such drug in clinical trials is
minocycline, the tetracycline analog that is able to cross the blood brain barrier
25
[134,135]. Minocycline is known to work to inhibit MMP activity and is speculated to
reduce protein translation by stalling ribosomes [111]. Minocycline treatment was found
to reduce MMP-9 in mammals [136] and MMP-1 in flies [137], which are increased in
the murine and Drosophila models of Fragile X syndrome, respectively. Treatment with
minocycline rescued neuronal architecture in both mice and flies and behavioral
phenotypes in mice in the absence of FMRP. These results suggest that the increase in
MMP activity contributes to symptoms in Fragile X syndrome.
While much is known about neuronal defects in Fragile X syndrome, defects in
other cell types, including cells important for immune system function, have not been
well-characterized. Using the Drosophila model of Fragile X syndrome, I examined both
systemic immune system function and the function of glia, the resident immune cells in
the brain. Chapter 4 contains a manuscript in preparation where I describe my results that
the absence of dFMR1 causes altered systemic immune system function and immune
system dysfunction in the brain. In the Chapter summary, I describe my specific
contributions to this manuscript.
26
CHAPTER 2:
The circadian clock protein Timeless regulates phagocytosis of bacteria
in Drosophila
Published:
Stone, E. F., Fulton, B. O., Ayres, J. S., Pham, L. N., Ziauddin, J., & Shirasu-Hiza, M. M.
(2012). The Circadian Clock Protein Timeless Regulates Phagocytosis of Bacteria in
Drosophila. PLoS Pathogens, 8(1), e1002445. doi:10.1371/journal.ppat.1002445.s001
27
SUMMARY
Shift work, jet lag, and sleep disturbances from noise and light pollution are an
unfortunate consequence of living in a modern society. Shift work and jet lag lead to
altered circadian rhythms, which are the natural, endogenous oscillations of many
molecular, physiological, and behavioral processes with a period of approximately 24
hours. While it seems intuitive that disruptions of sleep and circadian rhythm affect our
immune system function and overall health, the underlying immune mechanisms are not
understood. One focus of our lab is to investigate circadian regulation of immune system
function using Drosophila as a model. We asked what happens to the fly’s immune
system function when the molecular clock, which is responsible for proper circadian
regulation, is genetically silenced.
The molecular components responsible for circadian regulation were first
discovered in Drosophila. Briefly, the transcription factors Clock and Cycle (Clk and
Cyc), translocate into the nucleus and bind to the promoter regions of the Period and
Timeless (Per and Tim) genes, leading to their expression. As the Per and Tim proteins
are expressed, they also heterodimerize and translocate to the nucleus, blocking the
association of Clk and Cyc with the promoter regions for Per and Tim genes. These four
transcription factors cycle in every cell and cause the expression of many other genes to
oscillate throughout the course of the day. When any component of the molecular clock is
silenced, this negative transcriptional feedback loop no longer functions, leading to
circadian dysregulation.
In this study, we examined the well-characterized Tim01 mutant, a null mutant
exhibiting complete loss of circadian regulation, to characterize the impact of loss of the
28
molecular clock on the three main branches of the fly’s innate immune system function in
Tim mutants.
In addition to data analysis and intellectual contributions to writing the paper, I
made three major scientific contributions to this paper. First I analyzed the immune
mechanism of melanization in Tim mutants and wild-type controls (Fig. 4A)
Melanization is one of the three main mechanisms of the fly’s innate immune system.
Briefly, when flies are wounded or infected with bacteria or parasites, there is induction
of a proteolytic cascade resulting in the production of reactive oxygen species, which are
toxic to bacteria and parasites, and the deposition of melanin. I examined melanization in
Tim mutants and their wild-type controls at different times during the day, and I found
that Tim mutants do not exhibit differences in melanization compared to wild-type
controls. Additionally, I found that wild-type flies do not exhibit differences in
melanization after infection during their subjective day compared to their wild-type flies
infected during their subjective night. These results suggest that the immune mechanism
of melanization is not circadian-regulated in Drosophila.
Finally, I conducted the critical experiment demonstrating that the immune
mechanism of phagocytosis underlies differences in survival of infection between Tim
mutants and wild-type controls. We had previously found that Tim mutants are more
sensitive to S. pneumoniae infection. Other studies have illustrated that Drosophila fight
S. pneumoniae infection with phagocytosis, or the engulfment of bacteria by hemocytes,
which are immune cells in the fly [138]. In this paper, we showed that Tim mutants
exhibit reduced phagocytosis compared to wild-type controls. We then hypothesized that
the absence of Tim reduces phagocytosis of S. pneumoniae, resulting in reduced survival
29
after infection. To test this, I inhibited phagocytosis in wild-type and Tim mutant flies by
by injection of 0.2 µm polystyrene beads, which are phagocytosed and inhibit subsequent
phagocytosis. As a control, I injected identical cohorts with buffer alone (PBS). A few
days later, I infected both bead-injected and PBS-injected controls with S. pneumoniae.
The major result of these studies was that bead inhibition of phagocytosis completely
ablated the difference in survival typically observed between Tim mutants and wild-type
controls (Fig 6). These results suggest that the survival defect of Tim mutants after S.
pneumoniae infection is due to a defect in phagocytosis.
My contributions to this work were published in the following article: Stone, E.
F., Fulton, B. O., Ayres, J. S., Pham, L. N., Ziauddin, J., & Shirasu-Hiza, M. M. (2012).
The Circadian Clock Protein Timeless Regulates Phagocytosis of Bacteria in Drosophila.
PLoS Pathogens, 8(1), e1002445. doi:10.1371/journal.ppat.1002445.s001
30
ABSTRACT
Survival of bacterial infection is the result of complex host-pathogen interactions. An
often-overlooked aspect of these interactions is the circadian state of the host.
Previously, we demonstrated that Drosophila mutants lacking the circadian regulatory
proteins Timeless (Tim) and Period (Per) are sensitive to infection by S. pneumoniae.
Sensitivity to infection can be mediated either by changes in resistance (control of
microbial load) or tolerance (endurance of the pathogenic effects of infection). Here we
show that Tim regulates resistance against both S. pneumoniae and S. marcescens. We
set out to characterize and identify the underlying mechanism of resistance that is
circadian-regulated. Using S. pneumoniae, we found that resistance oscillates daily in
adult wild-type flies and that these oscillations are absent in Tim mutants. Drosophila
have at least three main resistance mechanisms to kill high levels of bacteria in their
hemolymph: melanization, antimicrobial peptides, and phagocytosis. We found that
melanization is not circadian-regulated. We further found that basal levels of AMP gene
expression exhibit time-of-day oscillations but that these are Tim-independent; moreover,
infection-induced AMP gene expression is not circadian-regulated. We then show that
phagocytosis is circadian-regulated. Wild-type flies exhibit up-regulated phagocytic
activity at night; Tim mutants have normal phagocytic activity during the day but lack
this night-time peak. Tim appears to regulate an upstream event in phagocytosis, such as
bacterial recognition or activation of phagocytic hemocytes. Interestingly, inhibition of
phagocytosis in wild type flies results in survival kinetics similar to Tim mutants after
infection with S. pneumoniae. Taken together, these results suggest that loss of circadian
31
oscillation of a specific immune function (phagocytosis) can have significant effects on
long-term survival of infection.
32
INTRODUCTION
Survival of bacterial infection is the result of complex host-pathogen interactions. An
often-overlooked aspect of these interactions is the circadian state of the host. Most
metazoans undergo daily, dynamic changes in their physiology that are regulated by well-
characterized circadian machinery. At its core, the circadian clock is composed of four
transcriptional regulators paired as two distinct heterodimers. In Drosophila, Period and
Timeless form one heterodimer and Clock and Cycle form the other. These two
heterodimers engage in an autoregulatory negative feedback loop that causes circadian
oscillations in the expression levels of target genes [139]. Microarray analyses of
different vertebrate tissues, such as heart, liver, or spleen-derived macrophages, have
shown that approximately 5-10% of total gene expression in each tissue is circadian-
regulated [140,141]. These oscillations in transcription are thought to underlie circadian
changes in the organism’s physiology and behavior.
Reflecting the pervasiveness of circadian regulation, the Drosophila circadian
mutants Timeless (Tim) and Period (Per) have pleiotropic phenotypes. They exhibit loss
of circadian rhythms in locomotor activity, eclosion, male courtship, and oxidative stress
response [142-144]. These circadian mutants are also sensitive to infection by at least
two bacterial pathogens and resistant to infection by another [81,145]. The circadian-
regulated mechanisms underlying these changes in immunity were not known.
Two types of mechanisms can affect survival after infection: resistance and
tolerance [146]. Resistance mechanisms control microbial growth, while tolerance
mechanisms allow the host to endure the pathogenic effects of infection, including the
host’s response to infection. In Drosophila, known resistance mechanisms to control
33
microbial growth in the hemolymph include antimicrobial peptide synthesis, reactive
oxygen species generation, and phagocytosis by immune cells. Previously, microarray
analysis comparing wild type and circadian mutants suggested that expression of immune
signaling molecules that induce resistance mechanisms such as Imd (Immune deficient)
may be circadian-regulated [147,148]. Here we test the hypothesis that circadian mutants
are sensitive to infection due to changes in mechanisms of resistance and investigate
specific resistance mechanisms for circadian regulation.
It has been recently shown that in mice, many immune cell functions, such as
cytokine secretion by macrophages and natural killer (NK) cells, undergo oscillatory
time-of-day variation [141,149]. Phagocytosis by immune cells such as macrophages and
neutrophils has been reported to change with time of day, though it is not clear if this
oscillation is diurnal (due to photoperiod) or circadian (regulated by circadian proteins)
[150-153]. Phagocytosis of photoreceptors by pigment cells in the vertebrate retina has
also been shown to be circadian-regulated [154]. Thus we further hypothesized that
phagocytosis of bacteria by immune cells in Drosophila is also circadian-regulated and
that decreased phagocytic activity may significantly contribute to the sensitivity of
circadian mutants to certain types of bacterial infection.
34
MATERIALS AND METHODS Fly and bacterial strains
The Canton S (CS) wild-type and CS tim01 lines have been previously described [81].
Infections were performed with the following bacteria as described [155]: Streptococcus
pneumoniae strain SP1, a streptomycin-resistant variant of D39 and gift from Elizabeth
Joyce in Stan Falkow’s laboratory at Stanford University [156]; Listeria monocytogenes
strain 10403S, gift from Julie Theriot at Stanford University [157]; Serratia marcescens
strain DB1140, gift from Man Wah Tan at Stanford University [158]; Salmonella
typhimurium strain SL1344, gift from Falkow lab at Stanford University [159];
Burkholderia cepacia ATCC strain 2541; Escheria coli strain DH5-alpha; and
Micrococcus luteus ATCC strain 4698. S. pneumoniae was frozen at OD600 0.15 in 1 ml
aliquots with 10% glycerol, pelleted upon thawing, rediluted three-fold in fresh BHI
(Brain Heart Infusion media, Difco), and incubated (standing) for 1.5-2 hour at 37˚C with
5% CO2 to OD600 0.15 for injection into flies, which were incubated at 29˚C.
L. monocytogenes was grown in standing BHI, overnight at 37˚C, and diluted to OD600 of
0.1 for injection into flies, which were incubated at 25˚C. S. marcescens was grown in
shaking BHI, overnight at 37˚C, and diluted to OD600 ranging from 0.1 to 0.6 for
injection into flies, which were incubated at 29˚C. S. typhimurium was grown in standing
LB, overnight at 37˚C, and diluted to OD600 of 0.1 for injection into flies, which were
incubated at 29˚C. B. cepacia was grown in standing BHI, overnight at 29˚C, and diluted
to OD600 of 0.01 to 0.001 for injection into flies, which were incubated at 18˚ or 25˚C. E.
coli was grown in shaking LB, overnight at 37˚C, and diluted to OD600 0.1 for injection
35
into flies, which were incubated at 25˚C. M. luteus was grown in standing LB, overnight
at 30˚C, and diluted to OD600 0.1 for injection into flies, which were incubated at 25˚C.
Injection
All experiments, including injections, were performed with male flies, 5-7 days post-
eclosion. Flies were raised at 25°C, 55-65% humidity on yeasted molasses food in a 12h
light/dark cycle. For injections, flies were anesthetized with CO2. Injections were
carried out with a pulled glass capillary needle. A Picospritzer III (Parker-Hannifan) or
custom-made Tritech microinjector (Tritech) was used to inject 50 nL of liquid into each
fly. Volume was calibrated by measuring the diameter of the expelled drop under oil.
Injections comparing ZT07 (day) and ZT19 (night) were conducted in a dark room using
red safety lights as light sources.
Survival assays
60-85 flies per genotype per condition were assayed for each survival curve and placed in
3 vials of dextrose food with approximately 20 flies each. In each experiment, 20-40
flies of each line were also injected with media as a wounding control. Death was
recorded daily. Data was converted to Kaplan-Meier format using custom Excel-based
software called Count the Dead. Survival curves are plotted as Kaplan-Meier plots and
statistical significance is tested using log-rank analysis using GraphPad Prism software.
All experiments were performed at least three times and yielded similar results.
CFU determination
36
Following challenge with microbes, six individual flies were collected at each time point.
These flies were homogenized, diluted serially and plated on appropriate media (tryptic
soy blood agar for S. pneumoniae, LB for all others). Statistical significance was
determined using non-parametric two-tailed t-tests. All experiments were performed at
least three times and yielded similar results.
Melanization assays
Flies were injected with S. pneumoniae (OD600 0.1) or L. monocytogenes (OD600 0.1) and
incubated at 29˚C. On day 2 of the S. pneumoniae infection or day 3 of the L.
monocytogenes infection, flies were visually inspected for deposits of melanin resulting
from the proteolytic cascade that generates reactive oxygen species. Injection with media
typically causes a wound-induced melanization response, a small deposit of melanin at
the site of injection within 24 hours. Infection with certain pathogenic bacteria such as L.
monocytogenes or S. typhimurium also causes a disseminated melanization response, with
large spots of melanin deposition observed elsewhere than the site of injection: under the
cuticle or in deeper tissue on both dorsal and ventral sides of the abdomen, thorax, and
head. Flies were quantified for both number and size of spots.
Antimicrobial peptide gene expression analysis
Flies were injected with 50 nL of bacterial culture at OD600 0.10 (S. pneumoniae, M.
luteus, or E. coli), or media alone (BHI, Difco). Following injection, flies were placed in
vials containing molasses food and incubated at 25˚C for five hours at the same light:
dark cycle to which they had been entrained. Three groups of six flies were homogenized
37
in Trizol and stored at -80˚C until processed. RNA was isolated using a standard Trizol
preparation and the samples treated with DNAse (Invitrogen). Fermentas RevertAid First
Strand cDNA synthesis kit was used to produce the cDNA following the manufacturer’s
protocol. Quantitative RT-PCR was performed using a Stratagene Mx3000 qPCR
machine, with Roche FastStart Universal SYBR Green Master (Rox) mix and the
following primer sets: Drosomycin, Diptericin, Rpl1 [160]. Total mRNA concentration
was normalized using Rpl1 expression. Differences in the infection-induced gene
expression were calculated by normalizing to basal gene expression in uninfected CS
(day sample) taken at the same time point; p-values were obtained by Mann-Whitney
test. The data shown in Figures 4 and S1 pool the results of three independent trials, each
having three biological replicates of six flies in each sample for each condition.
Phagocytosis assays
5-7 day old male flies were injected with 50 nL of 20 mg/ml pHrodo-labeled S. aureus or
E. coli in water (Molecular Probes, cat# A10010 and P35361). The flies were allowed to
phagocytose the particles for 30-60 min. The wings of the flies were removed and flies
were pinned with a minutien pin onto a silicon pad. Fluorescence images were taken of
the dorsal surface using epifluorescent illumination with a Leica MZ3 microscope fitted
with an ORCA camera (Hamamatsu). Images were captured with Openlab (Improvision)
software. The exposure was set such that the brightest images had a very small number
of saturated pixels. The experiment was repeated three times with 6-10 flies for each
treatment.
38
Bead inhibition of phagocytosis
Fluorescent 0.2 um polystyrene beads (Molecular Probes, cat# F13080) were injected
into hemolymph to inhibit phagocytosis as previously described [161]. Briefly, 200 ul of
beads in solution were washed three times in sterile water and resuspended in 20 ul final
volume. 5-7 day old male flies were injected with 50 nL of bead solution. To confirm
that phagocytosis was inhibited with this protocol, the in vivo phagocytosis assay was
performed as described above. Phagocytosis was completely inhibited within 2 days of
bead injection. Flies were then injected with S. pneumoniae at OD600 ranging from 0.05-
0.15 (see Survival Assays, above).
Accession numbers
Accession numbers are listed below for the following genes, which were examined in this
study. Accession numbers were obtained from www.uniprot.org: Timeless (P49021);
Drosomycin (P41964); Diptericin (P24492)
39
RESULTS / DISCUSSION
Circadian immune phenotypes are pathogen-specific
The fly immune response to infection is highly complex and pathogen-specific.
Previously, we showed that Timeless (Tim) null mutants die more quickly than wild-type
flies when infected with S. pneumoniae [81]. Another group showed that Tim mutants
die less quickly than wild-type flies when infected with P. aeruginosa, suggesting a
complex immune phenotype that is pathogen-specific [162]. To further understand the
immune phenotype of Tim mutants, we infected Tim mutants with three other pathogenic
bacteria (S. marcescens, B. cepacia, and S. typhimurium). We compared Tim mutant
survival time with wild-type flies and confirmed a complex immune phenotype (Figure 1,
A-D). Tim mutants died more quickly than wild-type flies when infected with S.
marcescens and died with wild-type kinetics when infected with S. typhimurium and B.
cepacia. Tim mutants showed no difference in sensitivity to wounding alone. Thus, these
data suggest that Tim activity has different consequences for immunity against different
pathogens.
Tim regulates resistance to specific bacterial infection
We next asked if Tim regulates mechanisms of resistance or tolerance against specific
bacterial pathogens [70]. We functionally distinguished between these two types of
immune defense by comparing the bacterial loads of Tim mutant and wild-type flies after
infection with four different bacterial pathogens (Figure 2, A-D). We found that loss of
Tim protein causes a loss of resistance against two pathogens. Tim mutants die faster
than wild-type flies when infected with S. pneumoniae and S. marcescens and have high
40
Figure 1. Tim activity has significant consequences for immunity. Shown here are Kaplan-Meier survival curves comparing Tim null mutants (orange circles) and wild-type control flies (green squares). Tim mutants die more quickly after infection with (A) S. pneumoniae (p<0.0001) or (B) S. marcescens (p<0.0001). Tim mutants die at the same rate as wild-type flies after infection with (C) B. cepacia (p=0.1001), (D) S. typhimurium (p=0.1707), or (E) medium alone (p=0.9836). p-values were obtained by log-rank analysis.
B
Media alone:Tim mutant is like wild type
0 10 20 30 40Time (days)
S. marcescens:Tim mutant is more sensitive
A
WildtypeTim mutant
S. pneumoniae:Tim mutant is more sensitive
E
Time (days)
Perc
ent S
urvi
val
100
75
50
25
00 2 4 6 8 10 12
p < 0.0001***
p = 0.9836n.s.
Time (days)
p < 0.0001***
0 2 4 6 8 10 12 14
Perc
ent S
urvi
val
100
75
50
25
0
D B. cepacia:Tim mutant is like wild type
p = 0.1707n.s.
C S. typhimurium:Tim mutant is like wild type
Time (days)
p = 0.1001n.s.
Perc
ent S
urvi
val
5 10 15 20 250
100
75
50
25
0
Time (days)
Perc
ent S
urvi
val
100
75
50
25
00 2 4 6 8 10 12
Perc
ent S
urvi
val
100
75
50
25
0
41
bacterial loads. This suggests that Tim mutants are less able to control microbial growth
and lack resistance to these two pathogens. Tim mutants had no effect on survival or
bacterial growth during infection with S. typhimurium and B. cepacia.
These results demonstrate that Tim regulates resistance mechanisms for specific
bacterial pathogens. Different physiologies will be important for defense against each
type of pathogen; thus specific microbes can be used to probe different aspects of the
immune system. Circadian proteins like Tim are known to regulate many different types
of behavior and up to 5-10% of the transcriptome of different tissues. If specific
physiologies that contribute to the defense against certain pathogens are circadian-
regulated in Drosophila, then Tim mutants will present immune phenotypes when
infected with those pathogens. Thus we can use our knowledge about the defenses
required for different pathogens to identify those that are circadian-regulated.
Light-induced degradation of Tim protein in wild-type flies causes sensitivity to S.
pneumoniae.
We chose to focus on Tim mutants and their dramatic defect in resistance against S.
pneumoniae. We hypothesized that Tim mutants might lack resistance because of a
developmental defect (such as malformation of immune tissues) or because of a direct
effect of circadian disruption on immunity in the adult fly. To distinguish between these
two possibilities, we disrupted Tim protein in genetically wild-type adult flies using
constant light (Figure 3A). Blue light (~380 to 475 nm) causes complete and rapid
ubiquitin-mediated degradation of Tim protein [163]; normally, this allows the internal
42
molecular clock to be synchronized with external light cues [142]. Flies reared in
constant
Figure 2. Tim regulates resistance to infection by S. pneumoniae and S. marcescens. Mutants lacking resistance, or bacterial killing, have shorter survival times and greater bacterial loads than wild-type flies. Tim null mutants lack resistance to (A) S. pneumoniae and (B) S. marcescens. Tim mutants have bacterial loads similar to wild-type flies when infected with (C) B. cepacia and (D) S. typhimurium. p-values were obtained by two-tailed, unpaired t-test; all experiments were performed with day-time injections. light exhibit complete loss of circadian rhythm [164]. Here we use constant exposure to
light to induce a Tim-null phenocopy in adult wild-type flies, circumventing any
developmental requirements for Tim protein. Wild-type flies were infected with S.
Col
ony-
Form
ing
Uni
tsPe
r Fly
(CFU
s)
B
D
S. marcescens:Tim mutant lacks resistance
107
105
103
106
105
104
103
Col
ony-
Form
ing
Uni
tsPe
r Fly
(CFU
s)
n.s. not significant p < 0.001 p < 0.0001*****
Days Post Injection
WildtypeTim mutant
1 20
*** **n. s.
B. cepacia:Tim mutant is like wild type
S. pneumoniae:Tim mutant lacks resistance
Col
ony-
Form
ing
Uni
tsPe
r Fly
(CFU
s)
A
C
1 20
*** **n. s.
Days Post Injection0 1 2 3
106
104
102
108
Days Post Injection
Col
ony-
Form
ing
Uni
tsPe
r Fly
(CFU
s) 106
104
102
108 n. s. n. s. n. s. n. s.
S. typhimurium:Tim mutant is like wild type
n. s. n. s. n. s. n. s.
Days Post Injection0 1 2 3
43
pneumoniae and incubated either in darkness or constant light. We found that flies
incubated after infection in constant light died faster than those incubated in darkness
A
C
Time (days)
Perc
ent S
urvi
val
100
75
50
25
00 1 2 3 4 5 6
Light-induced degradation of Tim protein makesWT flies sensitive to infection
B
Time (days)
sleep/wakecycle
time
Light-induced sleep/wake rhythms do not rescueTim mutant sensitivity to infection
Time (days)
expectedTim protein
levels
time
WT flies injected during the day, when Tim proteinis low, are sensitive to infection
expectedTim protein
levels
injection
time
WT flies in DARKWT flies in LIGHTTim mutants in DARKTim mutants in LIGHT
LightDark
Photoperiod
0 1 2 3 5 6 74
Perc
ent S
urvi
val
100
75
50
25
0
WT inj in DARK phaseWT inj in LIGHTphaseTim inj in DARK phaseTim inj in LIGHT phase
Perc
ent S
urvi
val
100
75
50
25
0masked*
WT in DARKWT in LIGHT:DARKTim in DARK
*Tim in LIGHT:DARK0 1 2 3 5 6 74
injection
injection
44
Figure 3. Resistance against S. pneumonia correlates with Tim activity. Shown here are Kaplan-Meier survival curves comparing wild-type flies and Tim mutants infected with S. pneumoniae and subjected to different light conditions as detailed in the schematic diagrams. (A) Light-induced degradation of Tim protein in wild-type flies results in increased sensitivity to S. pneumoniae. Infected wild-type flies incubated in constant light were more sensitive than wild-type flies incubated in darkness (p<0.0001). Tim mutants incubated in constant light died as rapidly as did Tim mutants incubated in darkness (p=0.3404). (B) Wild-type flies were more sensitive to S. pneumoniae when infected at a time of day when Tim protein is low. Wild-type flies were less sensitive when infected 6 hours after lights off (DARK) than 6 hours after lights on, when Tim protein is minimally and maximally expressed, respectively (p<0.0001). Tim mutant sensitivity was not rescued by infection 6 hours after lights off (DARK) (p=0.4134). (C) Light-induced sleep-wake rhythms (a startle response called “masking”) do not rescue Tim mutant sensitivity to infection. Tim mutant sensitivity was not rescued by incubation in cycling light:dark conditions compared to incubation in the dark. In fact, both Tim and wild-type flies were more sensitive to infection in cycling light:dark conditions than in the dark (p=0.0125 and p<0.0001 respectively).
(p<0.0001). As a negative control, we treated Tim null mutants and found that they were
not further sensitized by constant light, suggesting that Tim is epistatic to and therefore
downstream of light treatment. These results are consistent with the hypothesis that Tim
protein regulates resistance to S. pneumoniae in the adult fly and also demonstrate that
environmental and genetic disruption of circadian regulation alters immunity.
Wild-type flies are more sensitive to S. pneumoniae when infected at a time of day
when Tim protein expression is low.
If Tim protein positively regulates resistance in the adult fly and if Tim protein levels
oscillate over the circadian day, we predicted that the resistance of wild-type flies should
also oscillate over the circadian day. To test this, we infected two sets of flies, light-
entrained in anti-phase with each other, at times that correspond to minimal and maximal
Tim protein expression. Tim protein levels are lowest approximately 7 hours after lights
turn on (Zeitgeber 07 or ZT07/day) and highest at approximately 7 hours after lights off
45
(ZT19/night) [165]. Consistent with minimal and maximal Tim expression, wild-type
flies infected during the day (ZT07) died significantly faster than those infected at night
(ZT19) (Figure 3B). Tim null mutants showed no difference in survival time based on
time of infection. These results suggest that increased Tim expression at night positively
regulates a mechanism(s) of resistance required to fight infection by S. pneumoniae.
Loss of resistance seen in Tim mutants cannot be rescued by a sleep/wake cycle
induced by environmental light cues
“Masking” is a phenomenon in which flies lacking endogenous circadian rhythm exhibit
circadian-like, light-stimulated locomotor activity and sleep-wake patterns due to external
light cues (startle response) [166]. Thus Tim mutants in masking light/dark conditions
exhibit a locomotor response to light stimulus even in the absence of Tim protein. We set
out to ask if masking would restore the immune function of Tim mutants back to that of
wild-type flies. To test this, we infected wild-type flies and Tim mutants with S.
pneumoniae and incubated them either in the dark (relying on endogenous circadian
rhythm) or in circadian light-dark conditions (masking) (Figure 3C). Masking conditions
did not rescue the immune phenotype of Tim mutants infected with S. pneumoniae,
suggesting that this phenotype is not rescued by light-stimulated locomotor activity but
requires the normal expression of Tim protein.
The melanization response is not circadian-regulated
We next examined the three main Drosophila immune responses for circadian regulation:
melanization (generation of reactive oxygen species, or ROS), anti-microbial peptide
46
(AMP) synthesis, and phagocytosis by hemocytes. We tested each immune response for
time-of-day differences in wild type and Tim mutants and found evidence for circadian
regulation of phagocytosis but not melanization or AMP gene induction. This is
consistent with our survival data for S. pneumonia, as phagocytic hemocytes are crucial
to control the growth of S. pneumonia in Drosophila [138].
We first tested if the melanization response is circadian-regulated. Insects react to
injury and some bacterial infections with an enzymatic cascade that generates toxic
reactive oxygen species (ROS) and results in the deposition of melanin, visible through
the cuticle as dark black spots [70]. If melanization is circadian-regulated, we would
expect to see a difference in melanized spot formation between wild-type flies and Tim
mutants as well as between wild-type flies at different times of day, but not between Tim
mutants at different times of day. We assayed melanization in wild type and Tim mutants
infected with S. pneumoniae, but did not observe a systemic melanization response,
consistent with published reports [138,167]. We then assayed the melanization response
of flies infected with L. monocytogenes. Flies exhibit a strong melanization response
after L. monocytogenes infection and require melanization enzymes for resistance against
this pathogen [70]. We found no significant difference in melanized spot formation
between wild-type flies and Tim mutants or between wild-type flies at different times of
day (Figure 4A). These data suggest that melanization is not regulated by Tim.
AMP gene expression induced by septic injury is not circadian-regulated
We next tested if antimicrobial peptide (AMP) gene expression is circadian-regulated.
Expression of AMP genes in response to infection is the best characterized immune
47
response of Drosophila; these small peptides secreted by the fat body are thought to
control microbial growth by mechanisms such as disrupting bacterial membranes.
Previously, others had found that expression of several immunity signaling genes
Figure 4. Tim protein does not regulate melanization or AMP gene expression. (A) Wild type and Tim mutants injected with L. monocytogenes at ZT07 (DAY) or ZT19 (NIGHT) have similar levels of melanization. p-values for pair-wise comparisons were not significant: WT (DAY) vs. Tim (DAY), p=0.3551; WT (NIGHT) vs. Tim (NIGHT), p=1.000; WT (DAY) vs. WT (NIGHT), p=0.4031; Tim (DAY) vs. Tim (NIGHT), p=0.8717. p-values were obtained by unpaired, two-tailed t-test; n = 18-20 flies per genotype per condition. (B) Basal (uninduced) levels of Drosomycin expression were increased at ZT11 relative to ZT05, ZT17, ZT23 in both wild type and Tim mutants, suggesting that this change in expression is not mediated by Tim. (C) Infection-induced levels of Drosomycin expression were not significantly different after injection of wild type and Tim mutants with M. luteus at ZT05 (DAY) or ZT 17 (NIGHT). Every pair-wise comparison resulted in p-values greater than 0.05 (not significant) by Mann-Whitney test. (D) Infection-induced levels of Diptericin expression were not significantly different after injection of wild type and Tim mutants with E. coli at ZT05 (DAY) or ZT17 (NIGHT). Every pair-wise comparison resulted in p-values greater than 0.05 (not significant) by Mann-Whitney test.
A
C
Mel
anin
spo
ts/fl
yR
elat
ive
Expr
essi
on(D
roso
myc
in)
WTDAY
TimDAY
WTNIGHT
TimNIGHT
40
0
10
20
30
WTDAY
TimDAY
WTNIGHT
TimNIGHT
6
4
2
0
B
Rel
ativ
e ex
pres
sion
(Dro
som
ycin
)
WildtypeTim mutant
0 5 10 15 20
6
4
2
0
DR
elat
ive
expr
essi
on(D
ipte
ricin
)
WTDAY
TimDAY
WTNIGHT
TimNIGHT
2000
6000
4000
0
10000
8000
ZT
48
controlling AMP gene expression such as Rel and Imd is circadian-regulated
[145,147,148,162]. Here we measured the basal (uninduced) and pathogen-induced
expression of two representative AMP genes, Drosomycin (Drs) and Diptericin (Dpt),
which are widely used as reporters for induction of the Toll and imd signaling pathways,
respectively [168].
We first compared the basal (uninduced) expression of Drs and Dpt in wild type
and Tim mutants at four time points around the circadian cycle using qRT-PCR: ZT05,
ZT11, ZT17, and ZT23. We found that both wild type and Tim mutants exhibited higher
basal expression of the Toll-regulated AMP Drs at ZT11 than other time points (Figure
4B). This result suggests that basal expression of Drs varies with time of day but that this
difference in expression is not mediated by Tim protein. This time-of-day difference in
basal expression is statistically significant but very small relative to infection-induced
expression levels (Figure 4 C, D). Basal expression of Dpt was not significantly different
at any time of day in either wild type or Tim mutants (Figure S1A).
We next measured Drs and Dpt expression after S. pneumoniae infection in wild
type and Tim mutants at ZT05 and ZT17, approximately when wild-type flies exhibit
differences in survival and Tim mutants do not (see Figure 3). If AMP induction were
circadian-regulated, we would expect to see a difference in AMP expression between
wild-type flies at ZT05 and ZT17 and between wild type and Tim mutants at ZT17 but
not between Tim mutants at ZT05 and ZT17. We found that S. pneumoniae-induced
expression of Drs and Dpt was not significantly different at either time of day in either
wild type or Tim mutants (Figure S1B, C).
49
S. pneumoniae is not typically used to induce AMP expression and did not cause
the dramatic induction of AMP gene expression seen with other bacteria relative to media
alone [169]. Thus we next examined circadian-regulation of AMP expression in response
to two bacteria, M. luteus and E. coli, that have been most extensively used to probe
AMP induction and more generally, to quantify signaling through the Toll pathway
(activated by M. luteus) or the Imd pathway (activated by E. coli) [168]. We examined
Drs and Dpt induction in wild-type flies and Tim mutants infected at either ZT07 or ZT19
with either M. luteus or E. coli and collected six hours later for qRT-PCR analysis.
Again, we found that infection-induced expression levels of Drs and Dpt were not
significantly different at these time points in wild type and Tim mutants (Figure 4C,D).
Taken together, these data suggest that, for both Drs and Dpt, basal levels of expression
and short-term expression induced by these bacteria (S. pneumonia, M. luteus, E. coli) are
not circadian-regulated.
Tim regulates phagocytosis of pathogens by immune cells
Finally, we tested for circadian regulation of the Drosophila immune response of
phagocytosis—the physical engulfment and destruction of bacteria by specialized
immune cells. Phagocytosis is typically assayed by measuring the internalization of
fluorescently-labeled bacteria such as S. aureus or E. coli. Here, we measured phagocytic
activity at different times of day in both wild-type flies and Tim mutants. If phagocytosis
is regulated by Tim protein, we predict that phagocytic activity will be high at night and
low during the day in wild-type flies but will not vary with time of day in Tim mutants.
We assayed phagocytic activity in adult flies by injecting dead S. aureus labeled with
50
pHrodo, a pH-sensitive rhodamine dye that fluoresces in acidic environments. When
pHrodo-labeled bacteria are phagocytosed and processed into acidic lysosomes, the
phagocytic cell emits red fluorescence and can be imaged through the dorsal surface of
live, intact flies. We found that phagocytic activity was significantly
Figure 5. Tim protein regulates an early stage of phagocytosis of bacteria by hemocytes. (A) Wild-type flies in the night phase of the circadian cycle have more phagocytic activity than
WT day WT night
WT assayed in DAYWT assayed at NIGHT Tim assayed in NIGHT
Tim assayed in DAY
A
B
pHrodo-labeledS. aureus
0
400
800
1200
1600
WT Tim WT TimS. aureus E. coli
p = 0.51 p = 0.74
Phag
ocyt
ic a
ctiv
ity a
sar
ea o
f flu
ores
cenc
e (p
ixel
s)
p = 0.63
p < 0.001**
n. s. n. s. n. s.
51
wild-type flies in the day phase. Shown here are images of the dorsal surface of representative flies after injection of dead S. aureus labeled with the fluorophore, pHrodo, which emits fluorescence in acidic environments such as the lysosomes. Hemocytes that have phagocytosed these bacteria become fluorescent. Dashed lines indicate the region enlarged in the inset. (B) Phagocytic activity for wild type and Tim mutant flies was quantified by measuring areas of fluorescence. Wild-type flies exhibited significant circadian differences in phagocytosis of S. aureus (p<0.001) but not E. coli (p=0.74). This circadian difference is not present in Tim mutant flies with either S. aureus (p=0.51) or E. coli (p=0.63). p-values were obtained by unpaired, two-tailed t-test. higher at ZT19 (night) than ZT07 (day) in wild-type flies. Phagocytic activity did not
vary with the circadian cycle in Tim mutants (Figure 5A, B). Thus, phagocytic activity
oscillates with circadian rhythm in vivo in wild-type flies but not in Tim mutants,
consistent with the hypothesis that Tim protein up-regulates phagocytosis of S. aureus at
night. If phagocytosis is required to clear S. aureus, these data are consistent with
previous studies demonstrating that wild-type flies exhibit higher rates of survival when
infected with S. aureus at night than when infected during the day [162].
To test if circadian-regulated phagocytosis is bacteria-specific, we then assayed
phagocytic activity with a different type of bacteria. We injected wild type and Tim
mutants with pHrodo-labeled E. coli at ZT07 (day) and ZT19 (night) (Figure 5B). In
contrast to injections of S. aureus, wild-type flies did not exhibit a circadian difference in
phagocytosis of E. coli. Tim mutants also exhibited no circadian difference in
phagocytosis of E. coli. These results suggest that Tim regulates bacteria-specific
phagocytosis by immune cells.
Phagocytosis can be generally described as three steps: receptor-mediated
substrate recognition and binding, particle engulfment, and phagosome maturation.
Fluorescently-labeled S. aureus and E. coli have often been used in phagocytosis assays
to determine whether specific phagocytic components discriminate between different
types of bacteria. Many cellular steps of phagocytosis subsequent to substrate recognition
52
do not appear to be bacteria-specific. For example, inhibition of b-COP, thought to be
involved in phagosome maturation, leads to decreased phagocytosis of both S. aureus and
E. coli [170]. In contrast, some molecular components responsible for phagocyte
recognition of bacteria are bacteria-specific. For example, inhibition of the phagocytic
receptor PGRP-LC leads to decreased binding and phagocytosis of E. coli but not S.
aureus [170]. Another phagocytic receptor, PGRP-Sc1a, mediates phagocytosis of S.
aureus but not E. coli [171]. Thus our finding that Tim up-regulates phagocytosis of S.
aureus but not E. coli suggests that Tim regulates a bacteria-specific step of this process
such as substrate recognition or binding.
Bead inhibition of phagocytosis eliminates the survival difference between wild type
and Tim mutants after infection with S. pneumoniae
Because phagocytosis is crucial in defense against S. pneumoniae [138], these results
suggest that differences in phagocytic activity might contribute to the difference in
survival after S. pneumoniae infection between wild type and Tim mutants. Thus we
tested if inhibition of phagocytic activity would decrease these survival differences.
Phagocytosis can be inhibited in vivo by injection of polystyrene beads; phagocytes
engulf these beads and are unable to phagocytose subsequent injections of fluorescently-
labeled bacteria [161]. Thus we compared the survival kinetics of wild type and Tim
mutants infected with S. pneumoniae with or without bead pre-injection (Figure 6). As
described above, Tim mutants are highly sensitive to S. pneumoniae infection relative to
wild type flies. Consistent with published results, we found that bead pre-injection
increased sensitivity of wild type flies. Bead pre-injection of wild type flies decreased
53
survival rate to those similar to Tim mutants, suggesting that inhibition of phagocytosis is
sufficient to recapitulate Tim mutant sensitivity. Consistent with this, we also found that
bead pre-injection did not increase the sensitivity of Tim mutants to S. pneumoniae,
suggesting that phagocytosis is already impaired in these flies. Interestingly, in some
Figure 6. Bead inhibition of phagocytosis in wild type flies eliminates the difference in survival kinetics with Tim mutants after S. pneumoniae infection. Shown here are Kaplan-Meier survival curves comparing wild-type flies and Tim mutants infected with S. pneumoniae with and without pre-injection of beads to inhibit phagocytosis. Infected wild-type flies pre-injected with beads were more sensitive than wild-type flies without pre-injection (p<0.0001) and had similar survival kinetics as Tim mutants (p=0.0502). Infected Tim mutants had identical survival kinetics with or without bead pre-injection (p=0.1327). In this experiment infected, bead-injected wild-type flies were more sensitive than infected, bead-injected Tim mutants (p<0.0001); however, this result varied from experiment to experiment. experiments, bead-inhibited wild type flies appear to be more sensitive than bead-
inhibited Tim mutants, suggesting that the presence of Tim protein in the absence of
phagocytosis may have a negative effect on survival. Taken together, these results
suggest that Tim-mediated phagocytosis plays an important role in survival of S.
pneumoniae and that loss of Tim is equivalent to total inhibition of phagocytosis, though
Tim mutants are still able to phagocytose. Inhibition of phagocytosis is thought to inhibit
Wild type + beadsTim + beadsWild type, no beadsTim, no beads
Perc
ent S
urvi
val
0 1 2 3 540
25
100
75
50
Time (days)
54
phagocytes’ ability to signal the presence of infection; perhaps Tim also plays a role in
these downstream signaling events. Importantly, our results do not rule out the possibility
of multiple roles for Tim protein in immunity against bacterial infection, including S.
pneumoniae infection.
Circadian regulation of immunity has long been reported anecdotally in humans and other
animals. Here we show that Tim protein has complex, pathogen-specific effects on
immunity, regulating resistance to infection in Drosophila. Resistance against S.
pneumoniae is upregulated by Tim in the adult fly and is not simply a diurnal response
(i.e., activated by light/dark cycles even in the absence of circadian regulatory protein
function). We also identified phagocytosis as a specific circadian-regulated immune
mechanism. This is the first demonstration of a link between Tim protein and the cellular
immune response in vivo. Our data suggests that circadian regulation affects phagocytic
immune cells at an early stage of specific pathogen recognition and makes a significant
contribution to survival of infection.
This work has implications for people whose occupations disrupt their circadian
regulation such as night-shift workers, flight attendants, and hospital staff. Recently, it
was shown that chronic jet lag (shifts in circadian rhythm) in mice did not cause loss of
sleep or increased stress but did cause circadian dysregulation of the innate immune
system and dramatic vulnerability to LPS-induced endotoxemic shock [172].
Furthermore, mice infected with S. pneumoniae during their rest phase died significantly
less quickly than mice infected during their active phase [173]. In modern society,
circadian dysregulation and exposure to bacterial infection at night are likely not
55
uncommon. This work demonstrates that disruption of circadian oscillations in
phagocytosis can have significant adverse effects on resistance against infection.
56
SUPPLEMENTAL INFORMATION
Figure S1. Tim protein does not regulate AMP gene expression. (A) Basal (uninduced) levels of Diptericin expression were very low and not significantly different at different times of day in wild type and Tim mutants. Every pair-wise combination resulted in p-value greater than
AR
elat
ive
Expr
essi
on(D
ipte
ricin
)4
0
1
2
3
WildtypeTim mutant
0 5 10 15 20ZT
B
Rel
ativ
e Ex
pres
sion
(Dro
som
ysin
)
WTDAYS. p.
TimDAYS. p.
WTNIGHTS. p.
TimNIGHTS. p.
WTDAYBHI
TimDAYBHI
WTNIGHT
BHI
TimNIGHT
BHI
30
0
10
20
C
Rel
ativ
e Ex
pres
sion
(Dip
teric
in)
8000
0
2000
4000
6000
WTDAYS. p.
TimDAYS. p.
WTNIGHTS. p.
TimNIGHTS. p.
WTDAYBHI
TimDAYBHI
WTNIGHT
BHI
TimNIGHT
BHI
57
0.05 (not significant) by Mann-Whitney test. (B) Drosomycin expression levels induced by injection of S. pneumoniae or media (BHI) are not circadian-regulated. Wild type and Tim mutants were injected at ZT05 (DAY) or ZT 17 (NIGHT). Pair-wise comparisons of flies injected with S. pneumoniae resulted in p-values greater than 0.05 (not significant) by Mann-Whitney test. Tim mutants injected with media during the day exhibited significantly less Drosomycin expression than wild-type flies injected with media during the day or night or Tim mutants injected at night; all other pair-wise comparisons of flies injected with media did not show significant differences by Mann-Whitney test. S. pneumoniae-infected flies exhibited higher levels of Drosomycin expression than BHI-injected controls; p<0.01 by Mann-Whitney test. (C) Infection-induced levels of Diptericin expression were not significantly different after injection of wild type and Tim mutants with E. coli at ZT05 (DAY) or ZT17 (NIGHT). Pair-wise comparisons within the set of flies injected with S. pneumoniae resulted in p-values greater than 0.05 (not significant) by Mann-Whitney test. S. pneumoniae-infected and BHI-injected flies did not exhibit significant differences in Diptericin expression; p>0.05 by Mann-Whitney test.
58
ACKNOWLEDGMENTS
We thank Elizabeth Joyce, Stan Falkow, Man Wah Tan, and Julie Theriot for providing
bacteria; we thank Gerard Karsenty and the Karsenty lab for use of their qRT-PCR
machine, Gary Struhl for his fluorescence microscope, and Laura Johnston and Andrew
Tomlinson for microscopy equipment; and we thank Joel Shirasu-Hiza for help with
computer programming. David Schneider, Joe Takahashi, Jen Zallen, Tarun Kapoor, and
Julie Canman provided insightful comments on this manuscript. Marc Dionne, Paul
Taghert, and Amita Sehgal provided invaluable discussion, as did members of the
Schneider lab and other members of the Shirasu-Hiza lab (Victoria Allen, James Cheong,
Albert Kim, Paul Kim, Yang “Sunny” Li, Reed O’Connor, Adam Ryan). We thank the
reviewers for useful comments and experimental suggestions.
59
CHAPTER 3:
Host tolerance of infection is promoted by dietary glucose and amino
acids through insulin and TORC2 signaling
This manuscript is being revised for resubmission to Current Biology:
Victoria W. Allen, Reed M. O’Connor, Clarice G. Zhou, Vanessa M. Hill, Elizabeth F.
Stone, Royce B. Park, Keith R. Murphy, Julie C. Canman, William W. Ja, and Mimi M.
Shirasu-Hiza
60
SUMMARY
Sepsis is extraordinarily life threatening, with a mortality rate spanning 35-56% [174].
Sepsis and its related syndromes, severe sepsis and septic shock, are prevalent
worldwide, and accounted for 4.2% of hospital stays in 2009 in the US alone [175]. The
proper treatment for patients with sepsis is still debated and rife with controversy,
possibly contributing to the high morbidity and mortality rates [176]. While it is known
that sepsis syndrome results from the body’s response to numerous infections, the role of
the circadian regulation and the host’s metabolism on sepsis syndrome remains unknown.
To test the circadian regulation of a sepsis-like infection and the metabolic
consequences, we created two models for sepsis by infecting wild-type and Period (Per)
mutant Drosophila with high bacterial loads of P. aeruginosa and B. cepecia, creating an
overwhelming disease burden in the fly, to investigate the contribution of the host
metabolic state on survival after infection. These results are currently being revised for
resubmission to Current Biology.
I made four contributions to this paper. First, I pioneered the sepsis model in lab
by infecting Drosophila with a high bacterial load of P. aeruginosa. When I infected Per
mutants and wild-type controls with P. aeruginosa, I found that Per mutants survive
significantly longer after infection compared to wild-type controls (Appendix I, Fig. 1A-
B). Because this is a fast infection and the fly’s feeding behavior is circadian-regulated,
we speculated that timing of feeding contributed to the improved survival of Per mutants.
We hypothesized that Per mutants irregular feeding behaviors contributed to their
improved survival after infection. When I placed wild-type and Per flies on a restricted
diet and then infected the flies with P. aeruginosa, diet-restricted Per mutants survived
61
like diet-restricted wild-type flies (Appendix I, Fig 1C-D). My second contribution
demonstrated that feeding is the key physiology impacting Per tolerance to infection.
Although we had decided to use B. cepecia for infections in the final paper, the P.
aeruginosa results provided a framework and a phenotype for the paper on which the rest
of the paper was modeled.
The following two experiments that I had conducted were included in the paper.
For my third contribution, I had investigated whether phagocytosis was the mechanism
behind the tolerance phenotype seen in Per mutants after infection with B. cepecia (Fig
1H). I injected wild-type and Per mutant flies with either small polystyrene beads to
inhibit phagocytosis or PBS alone as a control. The fly’s hemocytes phagocytose these
beads and are subsequently unable to phagocytose bacteria for the duration of the fly’s
life. These flies were then injected with B. cepecia two or three days later after bead
injection, and survival was counted. While bead-inhibition reduced survival times in
both wild-type and Per mutants, we found that bead inhibition did not abolish the survival
advantage that we had observed in Per mutants flies. These results suggest that there is
not a phagocytosis advantage in Per mutants with B. cepecia infection.
We had found that Per mutants have reduced energy stores, Per mutants eat more
than wild-type controls, and that Per mutants have reduced survival with a restricted diet.
We examined which nutrient signaling pathways contribute to survival after infection.
One of the major nutrient signaling pathways in the fly is insulin-like signaling, and the
phosphorylation of Akt (p-Akt), which is downstream of insulin-like signaling, is readout
for glucose-mediated insulin-like signaling. For my final contribution, I tested the
presence of p-Akt in wild-type and Per mutants by Western blot analysis, and I found that
62
Per mutants have increased p-Akt compared to wild-type controls, suggesting increased
insulin-like signaling (Fig 5C). These results demonstrated that increased insulin-like
signaling is correlated with improved survival after B. cepecia infection.
Those two experiments showing that phagocytosis is not important for the
increased tolerance in Per mutants after infection with B. cepecia and that Per mutants
have increased p-Akt signaling were included in the following manuscript, which is
currently being revised for resubmission to Current Biology: Allen V. W., O’Connor R.
M., Zhou C. G., Hill V. M., Stone E. F., Park R. B., Murphy K. R., Canman J. C., Ja W.
W., and Shirasu-Hiza M. M. Host tolerance of infection is promoted by dietary glucose
and amino acids through insulin and TORC2 signaling.
63
ABSTRACT
Background: Nearly all metazoans undergo dynamic circadian-regulated changes in their
behavior and physiology. Currently it is not known how circadian-regulated behaviors
impact immunity against infection. There are two broad categories of defense against
bacterial infection: resistance mechanisms to control microbial growth and tolerance
mechanisms to withstand the pathogenic effects of infection.
Results: Our study of the Drosophila circadian mutant Period, which is behaviorally
arrhythmic, led us to identify a novel link between increased feeding behavior and
increased host tolerance of infection by B. cepacia, a bacterial pathogen of rising
importance in hospital-acquired infection. We further found that infection tolerance in
wild-type animals is stimulated by transient exposure to dietary glucose and amino
acids. Glucose-stimulated tolerance is mediated by the insulin-like signaling pathway
and can be induced by feeding or direct glucose injection. Using glucose injections, we
identified a surprisingly narrow window for glucose stimulation of tolerance, between 2
hours before and after infection. Amino acids-stimulated tolerance is mediated by TOR
Complex 2 (TORC2), not TORC1--an unexpected result considering the known role of
TORC1 as an amino acid sensor. In contrast to glucose, amino acids must be
administered by ingestion to stimulate tolerance and is facilitated by the host microbiota.
Conclusion: Taken together, this work demonstrates that circadian mutants can be used
as a tool to dissect physiologies important for survival of infection and indicates that the
nature, quantity, and timing of dietary intake on the day of infection by B. cepacia can
64
make a significant difference in long-term survival.
65
INTRODUCTION
Nearly all metazoans undergo daily, dynamic changes in their behavior and physiology
regulated by well-characterized, evolutionarily conserved circadian mechanisms [139].
Currently it is not known how circadian-regulated behaviors may impact immunity
against infection. The circadian clock is composed of four transcriptional regulators
paired as two heterodimers engaged in an autoregulatory transcriptional negative
feedback loop [142]. In Drosophila, Clock and Cycle form one heterodimer and Timeless
(Tim) and Period (Per) form the other. Clock and Cycle proteins are transcriptional
activators that activate expression of Tim and Per as well as hundreds of tissue-specific
target genes [21,139,177]. Circadian oscillations in gene expression are thought to cause
circadian oscillations in physiological function and, ultimately, organismal behavior.
One evolutionarily conserved, circadian-regulated physiology is immunity against
infection [63,81]. For both flies and vertebrates, innate immunity is the first line of
defense against infection. Drosophila lack adaptive immune components such as T cells
and B cells and rely solely on innate immune responses to survive infection [178].
Evolutionary conservation extends to the two primary Drosophila immune signaling
pathways, the Toll and imd (immune deficiency) pathways [179]. Flies and vertebrates
employ several similar innate immune effectors to kill bacteria, including: phagocytosis
by immune cells; reactive oxygen species generation, which in the fly is caused by
melanization; and secreted antimicrobial peptides (AMPs) with bacteriocidal properties.
Resistance is only one type of immune mechanism by which host organisms
combat bacterial infection. Resistance mechanisms control bacterial proliferation during
the course of infection, thus reducing the effects of infection by reducing the host’s
66
pathogen burden. A second distinct and complementary mechanism of immunity is
termed tolerance [146,180]. Tolerance mechanisms allow the organism to survive the
pathological effects of infection—whether caused by the microbe itself or by the host
immune response—without necessarily decreasing bacterial load [181,182].
Infection tolerance mechanisms are not well understood but include feeding and
metabolism. In Drosophila, decreased survival of infection is associated with decreased
metabolic stores during infection with two different bacterial pathogens, M. marinum or
L. monocytogenes [86,183]. The effect of feeding behavior on infection is also pathogen-
specific: decreased feeding increases survival of S. typhimurium, E. coli, and E. caratova
infections but decreases survival of L. monocytogenes infection [184,185]. In many
cases, the precise nutrients important for survival and the underlying molecular signaling
pathways have not been identified.
Both feeding behavior and expression of metabolic genes are known to be
circadian-regulated and both fly and mouse circadian mutants exhibit metabolic disorders
and changes in feeding behavior [186,187]. While we and others have shown previously
that host resistance against specific pathogens is circadian-regulated, it has not yet been
shown whether loss of circadian regulation of metabolism and feeding behavior will
affect immunity against infection [63,81,90].
Here we exploit a rapid, lethal infection of Drosophila with the human pathogen
B. cepacia to examine how small, acute differences in feeding behavior and diet can
impact infection tolerance. We show that infection tolerance is stimulated by an acute
flux of both dietary glucose and amino acids at the time of infection and that these effects
are respectively mediated by the insulin-like signaling pathway and the TORC2 signaling
67
pathway. Injected glucose can stimulate tolerance when administered within two hours
before or after infection; in contrast, amino acids must be ingested and their effects are
mediated by the fly's microbiota. This work suggests that what and how much the host
eats on the day of infection can make a significant difference in their long-term survival.
68
MATERIALS AND METHODS
All infection experiments were performed at least three times and yielded similar results.
p-values were obtained by unpaired t-test unless otherwise noted.
Fly lines
Per01mutants were compared to an isogenic white-eyed (w1118) Canton S strain. Oregon
R flies were used in experiments examining specific dietary components. dFOXO
mutants were trans heterozygotes generated by crossing dFOXO21 and dFOXO25
heterozygous parentals (gift of Marc Dionne [86]). UAS-Tsc1/Tsc2 (gift of Marc Tatar
[188]) homozygous males were crossed to w1118; Gal80-ts; tub-Gal4/TM6c (gift of Marc
Dionne) virgins and maintained at 18º through development. To induce activation of the
transgenes, flies were placed at 29º between 24 and 48 hours prior to infection.
Transgene induction after 24 hours at 29º was confirmed by qRT-PCR (see below, also
Fig. S3). RicTOR null mutants (gift of Stephen Cohen [189]) were derived by imprecise
excision of a p-element inserted into the RicTOR coding sequence. Controls for these
mutants were derived by precise excision of the same p-element. Axenic (germ free) OR
flies were generated as described below.
Fly cultures and media
Flies were bred and raised on standard food containing cornmeal, molasses, and yeast.
2.10 kg of cornmeal, 0.96 kg of agar, and 1.1 L of molasses were combined with 36 L
(summer) or 40 L (winter) distilled water. The mixture was boiled for 25 minutes, then
let cool for 45 minutes, or until the temperature reached 65°C or less. Methylparaben
69
(Tegosept, 50 g in 300 mL ethanol) and 180 mL of propionic acid were then added as
antifungal agents. Sugar-only foods contained 1% (w/v) agar (Fisher #A360-500 or Alfa
Aesar #A10752) and different concentrations (w/v) of glucose (Fisher #BP350-1 or
#BP350-500) dissolved in ddH20. Amino acid-containing food contained 1% agar, 5%
glucose, and 2% (w/v) amino acid mixture per the manufacturer’s instruction (Sunrise
Science 1360-030). For experiments involving diets other than standard food, flies were
switched to the experimental food ~24 hours before infection.
Bacterial cultures
Infections were performed as previously described [155] with Burkholderia
cepacia (ATCC strain #25416). B. cepacia was grown overnight at 29°C in 5 mL
standing Brain Heart Infusion (Teknova), resuspended in sterile PBS (Invitrogen
#003000), and diluted to an OD600 of 0.025 or 0.0275 for injection into flies (short
infection at 29°C) or OD600 of 0.1 (long infection at 18°C).
Injections
All experiments, including injections, were performed with age-matched male flies, 5–10
days post-eclosion. All flies except Gal80-ts; tub-Gal4>UAS-Tsc1/Tsc2 flies were raised
at 25°C, 55–65% humidity on standard food and entrained to a 12h light/dark cycle in a
Darwin Chambers incubator for at least 3 days prior to infection. Gal80-ts; tub-
Gal4>UAS-Tsc1/Tsc2 flies were raised at 18ºC through development. For transgene
induction, flies were shifted to 29ºC for 24 hours prior to infection. For injections, flies
were lightly anesthetized with CO2. Injections were carried out with a 10µL Drummond
70
Scientific glass capillary needle (#3-000-210- G), machine-pulled by a Sutter Instrument
Co. machine (Model P-30). A custom-modified Tritech microinjector was used to inject
50 nL of liquid into each fly. Volume was calibrated by measuring the diameter of an
expelled drop in halocarbon oil 700 (Sigma, H8898) under a layer of mineral oil (Sigma,
M8410). All injections were performed between Zeitgeber time (ZT) 7.5 and ZT 10.5 to
minimize variability from circadian effects on immunity. After injection with B. cepacia
flies were incubated at 29°C, except for long infections and heatshock control
experiments in which flies were incubated at 18°C. For co-injection of amino acids, B.
cepacia was resuspended in either PBS or a solution of amino acids dissolved in PBS
(low dose 11 µg/mL or high dose 220 µg/ml). For co-injection of rapamycin, flies were
injected with B. cepacia suspended in either buffer alone or in buffer containing 0.2
mg/mL rapamycin (CALBIOCHEM #553210).
Survival assays
Except where otherwise noted, approximately 60 male flies per genotype per condition
were assayed for each survival curve and divided into vials with approximately 20 flies
per vial. In each experiment, approximately 20 flies of each line were also injected with
media to control for the effects of wounding alone. Flies were injected between ZT 7.5
and ZT 10.5 and incubated at 29˚C in 12:12 light:dark conditions. Death was assayed by
visual inspection and recorded the next day every hour or more frequently. Data were
converted to Kaplan-Meier format using custom Excel-based software called Count the
Dead (J. Shirasu-Hiza). Survival curves were plotted as Kaplan-Meier survival plots and
statistical significance was tested by log-rank analysis using GraphPad Prism software.
71
Median survival times were compared in order to determine percentage-differences in
survival time.
Starvation assay
Using the DAM5 system (TriKinetics), age-matched 5-7 day old male flies were
incubated at 25º, 55-65% humidity in a 12:12 LD:DL cycle on media containing 1% agar
alone. For each genotype, n=16 flies per experiment. Flies were counted as dead
following an absence of beam crossings for the remainder of the experiment. p-values
for survival curves were obtained by log-rank analysis. Similar results were obtained
when starvation experiments were performed using visual inspection instead of DAM5 to
confirm death.
Bacterial load quantitation
Following challenge with microbes, approximately six individual flies were collected at
each time point. Time points chosen typically included just after infection (0h); mid-
infection (6h); and late in infection (12-16h) just before flies began to die. Flies were
homogenized in PBS, diluted serially and plated on standard LB agar plates. Statistical
significance was determined using unpaired t tests for 0 hour time points, while
subsequent time points were tested with non-parametric Mann-Whitney U tests, which
does not assume that samples have normal distribution. To ensure that flies were not
already infected with bacteria and that the buffer was not contaminated, 3 PBS-injected
flies were collected at the beginning of the experiment, homogenized and plated on LB
agar plates. Typically no colonies were observed for uninfected flies.
72
Quantitative real-time RT-PCR
RNA was extracted from three or four samples of 5-6 adult flies using TRIZOL
(Invitrogen) according to the manufacturer’s directions. RNA was treated with DNAse I
(Invitrogen, #18068-015), followed by reverse transcription using the First Strand cDNA
Synthesis Kit (Thermo Scientific, #K1622), priming with random hexamers.
Quantitative PCR was performed in 25 µL reactions with Roche Fast Start Universal
SYBR Green Master Mix (#04913850001) on a Stratagene Mx3000P or a Bio-Rad CFX
Connect Real-Time System. The cycling conditions used were as follows: Hold 95°C for
10 minutes, then 40 cycles of 95°C for 15s, 58°C for 30s, 72°C for 30s, then a final
sequence of 95°C for 1 minute, 57° for 30s, and 95° for 30s. All calculated gene
expression values were initially normalized to the transcript level of ribosomal protein 4,
Rpl1, prior to further analysis. Primer sequences are listed in Supplemental Table 1.
Melanization assays
Two groups of flies were injected with B. cepacia (OD600 0.005 – 0.05) and incubated at
18°C. One day after infection, flies from one group were each visually inspected for
wound site melanization, a small deposit of melanin at the site of injection. Three to
twelve days after infection, flies from the second group were visually inspected for the
systemic melanization response, characterized by one or more melanin deposits located
anywhere on the body other than at the injection site. p-values were obtained by paired t-
test for three independent trials.
73
Bead inhibition of phagocytosis
Phagocytosis was inhibited by injecting fluorescent 1.0 µm polystyrene beads (Molecular
Probes, cat# F13080) into hemolymph as previously described [63]. 200 µl of beads in
solution were washed three times in sterile PBS and resuspended in PBS for a final
volume of 30 µl. 7-10 day old male flies were injected with either 100 nL of bead
solution or buffer (PBS) as a control. To confirm that phagocytosis was inhibited with
this protocol, an in vivo phagocytosis assay was performed 2-3 days post bead inhibition
on a small sample of bead- and PBS-injected flies. Briefly, flies were injected with 50 nL
of 20 mg/mL pHrodo-labeled S. aureus in PBS (Molecular probes, cat# A10010). Flies
phagocytosed the bacteria for 20-30 minutes. The dorsal aspects of the flies were
superglued to coverslips, and the dorsal surface was imaged using epifluorescence
illumination with a Nikon upright fluorescent microscope with HQ CCD using Nikon
Elements software. Phagocytosis was completely inhibited within 2 days of bead
injection. Flies were then injected with B. cepecia at an OD600 of 0.025 (see Survival
Assays, above). p-values for survival curves were obtained by log-rank analysis.
Metabolic storage assays
For metabolic assays, samples were prepared by homogenizing 8 adult male flies (5-10
days old) in 200 ul Tris-EDTA (TE; 10 mM Tris, 1 mM EDTA, pH 8) + 0.1% Triton X-
100, then freezing at -20ºC. Triglyceride levels were measured as previously described
[190]. Triglyceride levels were blanked against reactions lacking lipase, which measured
the baseline levels of glycerol in the samples. Glucose and glycogen were measured as
described [86]. Briefly, the glucose and trehalose levels were measured using a glucose
74
oxidase reagent (Pointe Scientific) dissolved in ddH2O. Glycogen was measured using
the same reagent supplemented with 1 U/mL amyloglucosidase (Sigma, 10115-1G-F).
This reaction was blanked against the glucose reaction for each given sample. All
reactions were performed on 96-well tissue culture plates with 0.2 ml reaction mixture +
0.02 ml sample. Plates were incubated at 37º for one hour, and then the absorbance for
each well was measured at 490 nm using a BioRad Model 680 Microplate Reader. The
values obtained for each metabolic reserve were normalized to the average weight of flies
of that given genotype. Average weight was determined by measuring the mass of 8
samples of 50 adult male flies from each genotype. The resulting ratio of wild type to
Per mass was 1.054; in no case did this adjustment convert a result from being
insignificant to significant, nor vice versa. These values were then normalized to the
mean obtained from the wild type values and plotted with the normalized SEM.
Feeding assays
CAFE assays and 32P feeding assays were performed as previously described [191-193].
Briefly, for the radiolabeled feeding assay, flies were maintained on medium
supplemented with α-32P-dCTP for 24 h and then collected for liquid scintillation to
measure accumulated 32P. Aliquots of radiolabeled food were used to convert
scintillation counts to volume. For the CAFE assay, flies were fed liquid food solutions
of yeast extract and sucrose in glass microcapillaries that facilitated the direct
measurement of volume consumed.
Protein extraction and Western blotting
75
Four whole flies were homogenized in 80 µL of extraction buffer (10 mM HEPES, 100
mM KCl, 5% glycerol, 0.1% Triton, and 0.5 mM phenylmethylsulfonyl fluoride).
Samples were stored with 20 µL of loading buffer (20% glycerol, 10% SDS, 250 mM
Tris-HCl pH 6.8, 5% 2-mercaptoethanol, bromphenol blue), and processed as previously
described [194]. Briefly, samples were resolved on a 4-15% SDS polyacrylamide tris-
glycine gel, transferred overnight at 4°C at 22 volts, and immunoblotted using standard
procedures. Blots were blocked using 5% Carnation Instant Nonfat Dry Milk in PBS.
Antibodies were diluted in 5% milk in PBS. p-Akt, α-mouse, and α-rabbit antibodies
(Cell Signaling #4054S, #7076, #7074S) were used at a dilution of 1:1000. The actin
antibody was purchased from Abcam (ab8224) and used at a concentration of 1:10,000.
Immobilon Western (Millipore #WBKLS0500) was used as a chemiluminiscent
substrate. Fiji (a distribution of Image J) quantification of phospho-Akt signal was
performed by measuring the mean intensities of the phospho-Akt bands, which were
normalized to the mean intensities of the actin controls for each sample.
Preparation of germ-free flies
Standard food was autoclaved, added sterilely to autoclaved vials, and allowed to solidify
under UV light for several hours. Vials were then covered with parafilm and left under
UV light overnight, then plugged with autoclaved flugs the next day. These vials were
stored in a plastic tray that was treated with 70% ethanol, 10% bleach, and UV light
which was placed in a similarly treated Sterilite plastic storage bin. Axenic (germ-free)
eggs were prepared using a modified version of a previously published method [195].
Briefly, 5-8 males and 25-40 females were allowed to mate on petri dishes containing
76
food prepared by autoclaving 20g agar (Fisher #A360-500), 50g sucrose (source), 1L
ddH2O, and 8-10 drops blue food coloring (source) for 30 minutes. This food was
allowed to cool overnight under UV lights and was then partially covered with yeast
pellets before flies were added. These flies were allowed to breed overnight in a 25º
incubator with a 12:12 light:dark cycle. Before 10 am the following morning, flies were
removed from the dishes using bleached tools to prevent contamination. In a biohazard
hood, agar plates were washed with either 10% bleach (axenic) or ddH2O (conventional)
for approximately 2.5 minutes. Next, the contents of each dish were added to a nylon
mesh filter basket that was treated with bleach and ethanol. Embryos were then washed
with either 10% bleach (axenic) or ddH2O (conventional). Both sets were then washed
with ddH2O to remove bleach and yeast pellets. Embryos were then transferred to sterile
food using a sterile paintbrush. Both conventional and axenic flies were grown and
maintained under sterile conditions.
77
RESULTS
Period mutants exhibit increased tolerance when infected with B. cepacia.
We found that arrhythmic Per mutants survived longer than wild type flies when infected
with the human pathogen Burkholderia cepacia, which has previously been characterized
using Drosophila [155,196-198]. This increased survival was seen following both fast
and slow infection conditions: either a low bacterial dose at low temperature or a high
dose at high temperature (Fig.1A and 1B, p<0.0001 for both). To determine if this
increased survival was due to changes in resistance or tolerance, we measured bacterial
loads in individual flies at different time points for both fast and slow infections. For
both conditions at each time point, we found that wild type and Per mutants carried
equivalent bacterial loads (Fig.1C and 1D, p>0.05 for each time point). This result
suggests that the enhanced survival of Per mutants is not due to greater resistance, but
due to greater tolerance.
Known resistance mechanisms do not explain increased survival of infection.
To confirm that Per mutants do not have increased resistance to B. cepacia, we
quantitatively analyzed specific resistance mechanisms following infection. We
compared wild type and Per mutants infected with B. cepacia for the three well-
characterized mechanisms of resistance in Drosophila: antimicrobial peptide (AMP)
induction, melanization, and phagocytosis. We found no significant differences in B.
cepacia-induced AMP expression between wild type and Per mutants (Fig. 1E, 1F,
Supplemental Fig. 1A-E). We also found no difference in the B. cepacia-induced
melanization response between wild type and Per mutants (Fig. 1G). Wound-site
78
melanization was unchanged and, as previously published, B. cepacia does not induce
Fig. 1: Period mutants exhibit greater tolerance than isogenic controls during infection with B. cepacia. Per mutants survived longer than wild type during A) a long infection (Per, n=78; WT, n=77) and B) a short infection (Per, n=57; WT, n=64) with B. cepacia. Per mutants and wild type flies had similar bacterial loads over time following a C) long infection (n≥4 flies/time
n.s. n.s.
0 5 6
n.s.
C D
Bacte
ria
l lo
ad
/fly
B. cepacia, 29°C
Hours post-infectionHours post-infection
A B
0
25
50
75
100
Hours post-infection
0
25
50
75
100
p < 0.0001
Days post-infection
p < 0.0001
WT
Per
Rela
tive e
xpre
ssio
n
50
0
100
Hours post-infection
0 6 12
n.s. n.s.n.s.
Toll activation:
DrosomycinE Imd activation:
Diptericin
Hours post-infection
1000
0
2000
F
B. cepacia, 18°C B. cepacia, 29°C
0 6 12
n.s. n.s.n.s.
Pe
rce
nt
surv
iva
l
n.s. n.s.
0 4 14
n.s.
102
104
106
108P
erc
en
t surv
iva
lB
acte
rial lo
ad
/fly
Re
lative e
xpre
ssio
n
Figure 1: Allen et al., 2014
G
0
50
100
wound systemic
Melanization assay:
wound site vs. systemic
n.s. n.s.
Perc
ent
of flie
s
H Bead-inhibition
of phagocytosis
Hours post-infection
Perc
en
t su
rviv
al p < 0.0001
0
25
50
75
100
20 25 30
Pre-injection with:
beads, Per
beads, WT PBS, WT
PBS, Per
WT
Per
0 5 10 15 0 5 10 15 20 25
B. cepacia, 18°C
102
104
106
79
point) and D) short infection (n=6 flies/time point) with B. cepacia; p-values were obtained by unpaired t-test (0h) and non-parametric Mann-Whitney test (other time points). Consistent with a tolerance phenotype, antimicrobial peptide (AMP) induction was not different between Per mutants and wild type flies after B. cepacia infection as shown by: E) Drosomycin and F) Diptericin. Other AMPs are shown in Fig. S1. G) Per mutants and wild type flies did not exhibit differences in systemic and injection site melanization after B. cepacia infection (3 trials, n=17-22 flies/trial/genotype). H) Inhibition of phagocytosis by bead pre-injection did not remove Per mutant survival advantage over wild type after B. cepacia infection (Per, n=76 with beads, n=81 with buffer; wild type, n=81 with beads, n=80 with buffer; p<0.0001 for all pair-wise curve comparisons). p-values for survival curve comparisons were obtained by log-rank analysis; all others were obtained by unpaired t-test unless otherwise noted. n.s.=p>0.05.
systemic melanization [70]. Inhibition of phagocytosis by bead pre-injection decreased
survival of both Per and wild type controls, yet Per mutants still survived significantly
longer than wild type flies (Fig. 1H). This result suggests that phagocytosis is not
responsible for the increased tolerance of Per mutants. Taken together, these results
indicate that tolerance, and not resistance, leads to increased survival of Per mutants after
B. cepacia infection.
Per mutants have decreased energy storage.
We next wanted to determine the mechanism(s) leading to increased tolerance in Per
mutants during B. cepacia infection. It has been shown that decreased survival of other
bacterial infections, M. marinum and L. monocytogenes, is associated with loss of
metabolic stores [86,183]. Moreover, metabolic gene expression is known to be
circadian-regulated [186,187] and regulated by immunity genes [199,200]. These
findings raise the possibility that the increased tolerance of Per mutants during infection
is due to increased metabolic stores. If so, uninfected Per mutants would be more
resistant to starvation than wild-type controls. In fact, we found the opposite: Per
mutants are more sensitive to starvation than wild-type controls (Fig. 2A, p<0.0001),
80
suggesting that Per
Fig. 2. Per mutants have lower metabolic resources and eat more than wild type flies. A) Uninfected Per mutants were more sensitive to starvation than uninfected wild type flies (Per n=15; WT n=12). B) Quantification of metabolic storage levels comparing uninfected Per mutants and wild type flies (n=12 for both) revealed that Per mutants had lower levels of triglycerides (p=0.0004) and glycogen (p=0.0007) and similar levels of primary circulating sugars. C) 16 hours after infection with B. cepacia, Per mutants relative to wild type (n=12 for both) had lower levels of triglycerides and similar levels of glycogen and primary circulating sugars. D) Schematic of the radioactive food consumption assay. E) In the radioactive food assay, Per mutants ate ~14% more than wild type (Per, n=9; WT, n=9, p=0.016). F) Schematic of the CApillary FEeder (CAFE) assay. G) Using the CAFE assay, Per mutants ate ~23% more than wild type (Per, n=24; WT, n=21; p=0.0344). n.s.=p>0.05; p-values were obtained by unpaired t-test.
AP
erc
en
t su
rviv
al
Hours post-starvation
p < 0.0001
0
25
50
75
100
20 30 40
B
Me
tab
olic s
tore
s (
au
)
TG GLY TH +
GLU
Uninfected
0.00
0.25
0.50
0.75
1.00
n.s.*** ***
Figure 2: Allen et al., 2014
Starvation assay C
Me
tab
olic s
tore
s (
au
)
Infected with B. cepacia
***n.s.n.s.
0.00
0.25
0.50
0.75
1.00
TG GLY TH +
GLU
D
FCAFE assay
32P-labeled
food
Radioactive food assay
Measure 32P with
scintillation counter
32P
Day 1
Day 2
Measure
change
in liquid
E
0.0
0.5
1.0
Fe
ed
ing
(µ
l/fly/d
ay)
*
WT Per
Fe
ed
ing
(µ
l/fly/d
ay)
G
0.0
0.2
0.4
0.6 *
WT Per
WT
Per GLY = Glycogen
TG = Trigylcerides
p < 0.001***
n.s. ≥ 0.05p TH = Trehalose
GLU = Glucoseagar onlyPer starved
WT starved
81
mutants have less energy storage than wild type. To directly test this, we measured three
major types of energy storage: fat (triglycerides), glycogen, and primary circulating
sugars (trehalose and glucose). Consistent with their increased sensitivity to starvation,
uninfected Per mutants exhibited significantly lower levels of triglycerides (p=0.0004)
and glycogen (p=0.0007) and similar levels of trehalose and glucose (p=0.7065) when
compared to wild type (Fig. 2B).
An alternative metabolic explanation for the increased tolerance in Per mutants is
that these mutants may lose energy less quickly than wild type and have greater energy
stores late in infection. Both Per mutants and wild type lost energy stores during
infection, but Per mutants still had lower energy stores than wild type (Fig. 2C). At 16
hours post-infection, triglyceride levels in Per mutants were reduced relative to wild type
(~70% of wild type, p=0.0001), with levels of circulating sugars and glycogen similar to
wild type (p=0.9314 and 0.484, respectively). These data suggest that the increased
tolerance of Per mutants are not due to increased energy stores or decreased energy usage
during infection.
Per mutants exhibit increased feeding behavior.
We reasoned that Per mutants, because they have low energy stores, might eat more than
wild type animals and that this increased feeding behavior itself could enhance infection
tolerance. To directly investigate the feeding behavior of Per mutants, we compared the
nutrient intake of Per mutants and wild type flies by measuring consumption of 32P-
labeled, standard molasses food over 24 hours (Fig. 2D, 2E) [192,193]. In addition, we
measured the consumption of liquid food containing sugar and yeast extract using the
82
Capillary Feeder (CAFE) assay (Fig. 2F, 2G) [191,193]. In the 32P-labeled food assay,
Per mutants ate 14% more than wild type controls; in the CAFE assay, Per mutants ate
23% more than wild type controls (Fig. 2F, p = 0.016; Fig. 2G, p = 0.034). Thus, despite
having decreased energy storage before and during infection, Per mutants exhibit
significantly greater food intake than wild type.
Nutrient availability enhances infection tolerance of B. cepacia.
If the increased survival of Per mutants is due to increased feeding, then decreasing
nutrient intake by dietary restriction should abolish the enhanced survival time of Per
mutants in response to B. cepacia infection. To induce dietary restriction, we used a low
sugar, protein-free diet containing only water, agar, and 1% glucose. We found that even
short-term dietary restriction (~24 hours before infection) decreased survival time for
both wild type (Fig. 3A, p<0.0001) and Per mutants (Fig. 3B, p<0.0001). Per mutants
consistently survived an average of 22% longer than wild type flies when fed standard
food (20/20 experiments, comparison of median survival times). In contrast, diet-
restricted Per mutants either had no survival advantage over wild type (4/12
experiments), survived significantly less well than wild type (2/12 experiments), or
survived an average of only 7% longer than wild type (6/12) (Fig. 3C, 3G). Bacterial
loads remained unchanged under these feeding conditions (Fig. 3D, 3E, 3F; p>0.05 for all
time points). Together, these results show that short-term dietary restriction reduces
infection tolerance and support the hypothesis that the increased feeding behavior of Per
mutants on the day of infection contributes to their increased tolerance of infection with
B. cepacia.
83
Fig. 3. Dietary restriction does not increase infection tolerance of either Per mutants or wild type A-C) Dietary restriction decreased survival time after infection for both A) wild type (standard food n=66, DR n=63) and B) Per mutants (standard food n=59, DR n=62). C) Dietary restriction eliminated the consistent survival advantage of Per mutants over wild-type flies (Per n=62, WT n=63). Dietary restriction did not alter bacterial load for D) wild type (n≥5 flies/time point) or E) Per mutants (n≥5 flies/timepoint, n>0.05, all time points); moreover, F) diet-restricted wild type and Per mutants had similar bacterial loads (n≥5 flies/time point). p-values were obtained by unpaired t-test (0h) and non-parametric Mann-Whitney test (other time points). G) Multiple experimental trials showed that, on standard food, Per mutants always survived B. cepacia infection longer than wild type but not after dietary restriction; survival outcome was judged significant when p<0.05. p-values for survival curve comparisons were obtained by log-rank analysis; all others were obtained by unpaired t-test unless otherwise noted. n.s.=p>0.05; fed=standard diet, DR=diet-restricted (1% glucose).
Figure 3: Allen et al., 2014
0 6 12
A
D
Bac
teria
l loa
d/fly
Per
cent
sur
viva
l
Hours post-infection
WT: Fed vs. DRp < 0.0001
0
25
50
75
100
15 20 25 30
Hours post-infection
2
4
6
10
10
10
108
0 126
n.s. n.s.n.s. E
Hours post-infection
Per
cent
sur
viva
l
B Per mutant: Fed vs. DR
15 20 25 300
25
50
75
100 p < 0.0001
102
104
106
Hours post-infection0 6 12
Bac
teria
l loa
d/fly
2
4
6n.s. n.s.n.s.
Per
cent
sur
viva
lB
acte
rial l
oad/
fly
F
Hours post-infection
WT vs. Per: DRn.s.
C
15 20 25 300
25
50
75
100
10
102
104
6
Hours post-infection
G
n.s. n.s.n.s.
Standard diet:WT fed Per fed
Diet-restricted:WT DRPer DR
Standard diet(n = 20)
Outcomes of B. cepacia infection: WT vs. Per (number of trials)
DR(n = 12)
Per and WT same (p>0.05)WT survives longer (p<0.05)
Per survives longer (p<0.05)
Per vs. WT survival
84
Dietary glucose and amino acids enhance infection tolerance
To evaluate the specific dietary components that contribute to tolerance of infection, we
supplemented a restricted diet with defined nutrients (Fig. 4A). Because Per mutants
display pleiotropic defects in metabolism and other circadian-regulated physiologies
[177], we focused on wild type flies. We first tested if increasing glucose content in the
food could fully complement the restricted diet, which contains 1% glucose. We
compared the effects of standard food with protein-free, agar-based food containing 1%,
5%, 10%, or 15% glucose (Fig. 4B). Wild-type flies survived longest on standard food
and shortest when switched to 1% glucose 24 hours before infection (Fig. 4B, p<0.0001
comparing standard food or 1% glucose with any other condition). Increasing glucose
concentration from 1% to 5% increased survival time (Fig. 4B, p<0.0001) without
changing bacterial loads (Fig. 4D, p>0.05 for all time points), but glucose concentrations
higher than 5% did not further increase survival time (4B, p>0.05 for any pair-wise
comparison of 5%, 10%, and 15% glucose). Moreover, none of these glucose-only diets
was sufficient to increase infection survival time to that observed on standard food
(p<0.0001 comparing any glucose-only diet with standard food). Thus, glucose enhances
infection tolerance but glucose alone is not sufficient for optimal survival of infection.
This result suggests that other components in standard food also contribute to survival of
infection.
In addition to sugar, standard food also contains a complex mixture of lipids,
proteins, vitamins, and other nutrients derived from yeast and cornmeal ingredients. We
tested whether 5% glucose supplemented with amino acids was sufficient to substitute for
standard food (Fig. 4A). We compared survival of flies fed three different diets: 5%
85
glucose, 5% glucose plus amino acids, and standard food. A diet of 5% glucose plus
Fig. 4: Dietary glucose and amino acids increase tolerance of B. cepacia infection A) Schematic of dietary conditions: wild-type flies were raised on standard food and switched 24 hours before B. cepacia infection to fresh standard food, glucose diets (B,D) or glucose diet plus amino acids (C,E). B) Increasing glucose concentration (5%, 10%, 15%) increased survival time relative to 1% glucose diet (p<0.0001 in all cases) and caused similar survival kinetics compared
86
to each other (p>0.05 in all cases). C) Supplementing 5% glucose with amino acids increased survival time significantly longer than the glucose-only diets (p<0.0001 in all cases) and was sufficient for survival kinetics similar to standard food (p>0.05). There was no difference in bacterial load comparing flies fed D) 1% vs. 5% glucose (n=6 flies/time point) or E) 5% glucose vs. 5% glucose plus amino acids (n=6 flies/time point); p-values were obtained by unpaired t-test (0h) and non-parametric Mann-Whitney test (other time points). F) Wild-type flies survived longer when injected 1.5 hours before infection with 50 nL of 5% glucose (n=21) than with PBS control (n=18, p=0.0007). G) Multiple trials showed that 5% glucose injection enhanced survival relative to buffer injection when administered within 2 hours before or after the time of infection. Survival outcome was judged significant when p<0.05. p-values for survival curve comparisons were obtained by log-rank analysis; all others were obtained by unpaired t-test unless otherwise noted. n.s.=p>0.05. amino acids 24 hours prior to infection resulted in significantly longer survival time
compared to 5% glucose alone (Fig. 4C, p<0.0001), with no change in bacterial load (Fig.
4E, all time points p>0.05). In fact, 5% glucose plus amino acids was sufficient to
increase survival time as long as standard food (Fig. 4C, p=0.3039). Taken together, these
results suggest that glucose and dietary amino acids alone are sufficient to promote
tolerance of infection, and that acute exposure to both within ~24 hours of B. cepacia
infection is necessary for optimal survival.
Glucose is required around the time of infection for increased host tolerance.
We set out to precisely characterize the required timing of the glucose contribution to
infection tolerance. To identify the time and place where glucose acts to increase
tolerance, we developed a method to mimic acute glucose intake. We found that a 50 nL
injection of 5% glucose administered directly into the circulatory system of diet-restricted
flies significantly increased infection survival time relative to injection of buffer alone
(Fig. 4F, p <0.0001). This amount of glucose is approximately equivalent to the quantity
ingested by a single fly in a 1-hour period (as calculated from prior feeding experiments;
Fig. 2D-G).
87
We found that glucose injection was only effective if administered within a
specific time window centered around the time of infection with B. cepacia. Glucose
injections within 2 hours before or after the time of infection led to a significant increase
in survival (Fig. 4G). In contrast, injections performed 2-4 hours before or after infection
had no effect or a negative effect on survival (Fig. 4G). This time window is
unexpectedly narrow, consistent with an acute rather than chronic effect of diet upon
infection tolerance. These results suggest that acute glucose intake acts to stimulate a
specific signaling pathway that increases immune tolerance when administered at or
around the time of infection.
FOXO activity mediates a glucose-activated increase in infection tolerance.
We set out to identify the signaling pathways underlying the nutrient contributions to
tolerance. Because insulin-like signaling (ILS) is important for glucose homeostasis
[201], we hypothesized that ILS mediates the glucose contribution to infection tolerance.
Phosphorylation of dAkt, a downstream component of the ILS pathway, is indicative of
glucose-activated insulin-like signaling (Fig. 5D) [202]. Consistent with their increased
feeding, we found that Per mutants exhibit increased levels of phospho-dAkt relative to
wild type flies by Western blot analysis (Fig. 5C). Thus activation of insulin-like
signaling is correlated with increased tolerance of B. cepacia.
To directly test whether insulin signaling mediates glucose-stimulated tolerance,
we infected mutant flies lacking dFOXO, a critical downstream component of the ILS
pathway (Fig. 5D). Because dFOXO activation results in negative feedback to inhibit the
ILS pathway, loss of dFOXO function causes constitutive activation of insulin-like
88
signaling. If dietary glucose alters survival through activation of ILS, then dFOXO
Fig. 5: Insulin-like signaling mediates the glucose contribution to infection tolerance and the TORC2 complex mediates the effects of dietary amino acids. A) dFOXO mutants showed no significant difference in survival kinetics when fed 1% or 5% glucose 24 hours before infection (1%, n=20; 5%, n=23). B) dFOXO mutants survived
89
significantly longer when fed 5% glucose supplemented with amino acids (a.a.) compared to 5% glucose alone (glucose alone, n=23; with a.a., n=11). C) Per mutants exhibit 1.7x higher levels of phospho-Akt than wild-type flies as determined by Western blot analysis. D) Schematic of the insulin-like signaling pathway. E) Flies co-injected with buffer at the time of infection survived longer when fed 5% glucose plus amino acids (n=52) than 5% glucose alone (n=70). F) Inhibition of TORC1 by co-injection of rapamycin at the time of infection failed to abolish the survival benefit conferred by dietary amino acids (glucose alone, n=59; with a.a., n=67). G) Gal80-ts; tub-Gal4>UAS-Tsc1/Tsc2 flies raised at 18ºC (not over-expressing Tsc1/2, see Fig. S3) exhibited increased survival time in response to dietary amino acids (glucose alone, n=71; with a.a, n=65). H) Gal80-ts; tub-Gal4>UAS-Tsc1/Tsc2 flies raised at 29ºC (over-expressing Tsc1/2, see Fig. S3) exhibited increased survival time in response to dietary amino acids (glucose alone, n=63; with a.a., n=59). I) RicTOR01 control flies exhibited increased survival time in response to dietary amino acids (both, n=71). J) RicTOR01 mutants did not exhibit increased survival time in response to dietary amino acids (glucose alone, n=50; with a.a., n=51). K) Schematic of the TOR pathway. n.s.=p>0.05.; a.a.=amino acids.
mutants will exhibit no difference in survival whether fed 1% or 5% glucose. Consistent
with this, dFOXO mutants fed 1% glucose or 5% glucose showed no difference in
survival after B. cepacia infection (Fig. 5A, p=0.2974).
In contrast, defective insulin-like signaling had no effect on increased survival
induced by dietary amino acids. Infected dFOXO mutants given dietary amino acids
survived longer when infected with B. cepacia than those deprived of amino acids (Fig.
5B, p<0.001), similar to wild-type flies. Together, these results indicate that the insulin-
like signaling pathway mediates the effects of dietary glucose, but not dietary amino
acids, in increasing tolerance of B. cepacia infection.
TORC2 mediates the amino acid contribution to infection tolerance.
We next asked which molecular pathways might mediate the effects of amino acids in
increasing tolerance of B. cepacia infection. The canonical sensor of amino acid
availability is the TOR signaling pathway, specifically the TORC1 pathway [203]. The
TOR protein forms two distinct complexes: TOR Complex 1 (TORC1) is inhibited by the
90
small molecule rapamycin and monitors amino acid availability; TOR Complex 2
(TORC2) is rapamycin-insensitive and is better characterized as a regulator of the actin
cytoskeleton (Fig. 5K) [204,205].
To determine whether TORC1 signaling plays a role in amino-acid stimulated
tolerance, we tested whether tolerance is inhibited by rapamycin [204,205]. We found
that injecting 9.6 ng of rapamycin per fly (approximately equivalent to the mammalian
dose of 16 mg/kg, [206]) significantly decreased survival of infection (Fig. S2, rapamycin
vs buffer, p<0.0001 on food with or without amino acids). This effect is likely due to
TORC1-dependent resistance mechanisms, as seen during P. aeruginosa infection
(personal communication, S. Pletcher). We found that, when injected with either
rapamycin or vehicle, flies survived infection significantly longer when fed dietary amino
acids (Fig. 5E, 5F, p<0.0001 in both cases). This result suggests that the rapamycin-
sensitive, canonical amino acid sensor TORC1 is not involved in amino acid-stimulated
tolerance of B. cepacia infection.
To confirm that TORC1 signaling is not the key mediator of amino acid-
stimulated infection tolerance, we compared the effects of perturbing another molecular
component of this pathway, Tsc1/2. Tsc1/2 activity inhibits Rheb, thus negatively
regulating TORC1 signaling (Fig. 5K) [207]. We found that mutants overexpressing
Tsc1/2 (Fig. S3A,B) still responded to dietary amino acids with increased survival similar
to controls following B. cepacia infection (Fig. 5G,H p<0.0001). These results further
confirm that the effect of amino acids on survival of infections is independent of TORC1
signaling.
We next tested whether dietary amino acids enhance infection tolerance through
91
TORC2 signaling. In TORC2, TOR complexes with RicTOR (the Rapamycin-insensitive
companion of TOR) and Sin1 (Fig. 5K) [208]. We compared mutants carrying a
transposon-mediated deletion in the RicTOR gene with control flies in which this
transposon has been precisely excised from the RicTOR gene. While control flies
responded to dietary amino acids with increased survival of infection (Fig. 5I, p<0.0001),
RicTOR mutants exhibited the same survival time with or without dietary amino acids
(Fig. 5J, p>0.05). These results therefore suggest that TORC2, but not TORC1, is
required for amino acid-mediated tolerance of infection.
Amino acid contribution to survival depends on the gut microbiota.
To characterize the mechanism underlying amino acid-stimulated tolerance, we asked
whether direct injection of amino acids can increase survival time as well as ingestion of
amino acids, similar to glucose (Fig. 6A, schematic). We found that injection of amino
acids at two different concentrations at different times points before or during infection
did not improve survival time (Fig. 6B, p=0.0031 with injection of a high dose of amino
acids decreasing survival time; Fig. 6C, low amino acid dose, p>0.05; and Fig. S4A and
S4B, showing amino acid injections 2h before and after infection, respectively, p>0.05
for both). Flies injected with buffer were still able to respond to ingested amino acids
(Fig. 6C, p<0.0001). One possibility is that amino acids need to be administered in a
more sustained fashion than glucose in order to enhance tolerance. Another possibility is
that amino acids need to be ingested by the animal through the gut. If the latter is true,
we hypothesized that the gut microbiota may be required for amino acid-stimulated
infection tolerance.
92
To test this, we modified a previously published method [195] to raise germ-free
Fig. 6. The effect of amino acids on survival of infection is facilitated by gut microbiota. A) Schematic of experimental set-up for B and C. Conventionally raised flies are either injected with amino acids or have amino acids included in their diet. Injection of amino acids in a high dose (B, n=69) or low dose (C, n=43) does not improve survival when compared to buffer alone (B, n=64, p=0.0031; C, n=47, p>0.05). As a control, buffer-injected flies are still able to respond to dietary amino acids with enhanced survival time (C, n=25). D) Experimental set-up for E and F. Conventionally-raised wild-type flies’ and germ-free wild-type flies’ survival of infection are compared on diets with and without amino acids. E) Germ-free flies survive infection similarly to
93
conventionally-raised flies with normal microbiota when amino acids are not present in the diet (germ-free, n=38; conventional, n=54, p>0.05). F) When exposed to dietary amino acids, flies with intact microbiota exhibit enhanced survival significantly greater than germ-free flies (germ-free, n=39; conventional, n=61). p-values obtained by log-rank analysis. flies lacking their microbiota. We then compared the effects of dietary amino acids on
survival to the effects on conventional flies with intact microbiota (Fig. 6D, schematic).
We confirmed germ-free status by testing for the presence of bacterial 16S ribosomal
genes using PCR analysis (Fig. S4C). On the protein-free 5% glucose diet, germ-free and
conventional flies had identical survival times after infection with B. cepacia, showing
that germ-free flies are not generally immunocompromised against this infection (Fig.
6E, p<0.05). However, dietary amino acids enhanced the survival of conventional flies
significantly more than germ-free flies (Fig. 6F, p<0.0001). Dietary amino acids still
increased the survival of germ-free flies relative to glucose-only diet (Fig. S4D),
suggesting that the microbiota are not exclusively required for amino acid-mediated
survival. Taken together, these results suggest that amino acids are optimally
administered by feeding to increase infection tolerance; the gut microbiota may help to
mediate absorption of these amino acids or may increase survival of infection by a
different mechanism.
94
DISCUSSION
By examining a circadian mutant with increased infection tolerance against B. cepacia,
we first identified increased feeding as a behavior that contributes to increased tolerance.
We found that two specific nutrients, glucose and amino acids, fully substitute for
standard food in promoting optimal tolerance of rapid infection with B. cepacia. We
found that the glucose contribution to survival is mediated by insulin-like signaling;
consistent with this, two previous studies have implicated ILS in Drosophila tolerance of
M. marinum and resistance against viral infections [86,209] and two other studies have
found that induction of immunity gene expression leads to decreased insulin signaling in
wild-type flies [199,200]. Our work strengthens the hypothesis that insulin-like signaling
may be a general feature of Drosophila immunity, irrespective of the specific type of
infection. Our data further suggest a time and location for this insulin activation, in that
an acute increase in circulating glucose around the time of infection can increase overall
survival time. Thus, what and how much a fly ingests near the time of infection has a
significant effect on its survival of that infection.
Our results also show that dietary amino acids can increase a host’s tolerance.
This effect does not appear to be mediated by the canonical amino acid sensing TORC1
pathway but through the less well-understood TORC2 pathway. From our rapamycin
experiments, TORC1 appears to play a role in immunity against B. cepacia independent
of amino acid-stimulated tolerance. We hypothesize that this is a TORC1-mediated
resistance mechanism, similar to an effect recently observed with the related bacterial
pathogen, P. aeruginosa (S. Pletcher, personal communication). In vertebrates,
rapamycin is a well-charactarized immunosuppressant due to its inhibition of mTORC1.
95
A role for TORC2 in host tolerance has not been previously described. TORC2 is
thought to be stimulated by growth factors and PI3K [205]. Interestingly, in both yeast
and mammalian cells, TORC2 is activated by association with the ribosome but not by
translation [210,211]; and evidence exists that TORC2 may be stimulated by amino acids
[212]. In other systems, downstream targets of TORC2 include components of the
cytoskeletal system, as well as Akt and SGK1, and are thought to play a role in tissue-
specific morphology [204,213]. In our system, dietary amino acids appear to activate
TORC2 and enhance host tolerance, an intriguingly novel finding. While in most cases,
immune effects of TOR are thought to act through TORC1, recent evidence suggests that,
in mouse embryonic fibroblasts, RicTOR inhibits TLR-stimulated cytokine expression
[214]. Thus RicTOR may contribute to host tolerance by suppressing harmful effects of
the host immune response. The direct targets of TORC2 that are relevant for infection
tolerance remain unknown and their identification will be an important goal of future
studies.
Our results also show that in Drosophila, amino-acid mediated tolerance is
facilitated by the gut microbiota. Recent work has shown that the gut microbiota can
affect the pathogenesis of diabetes [215], obesity, and atherosclerosis [216]. Here we
show that the fly's microbiota also play a role in tolerance of a systemic infection. This
role could include a contribution to the fly's development so that it can respond properly
to amino acids [217-219] or the gut microbiota may be required for absorption of specific
amino acids by the gut epithelia, as is postulated for branched-chain amino acids [219].
Alternatively, the fly’s microbiome itself may be altered or stimulated by amino acids to
produce molecules or interact with the gut epithelium in a way that enhances infection
96
tolerance. The Drosophila system described here will be useful in distinguishing
between these models.
The cellular and molecular mechanisms that promote host tolerance of infection
are not well-understood [82]. B. cepacia is a significant opportunistic bacterial pathogen,
particularly in hospital settings. This hospital-acquired infection is associated with high
rates of mortality, up to 50% for severe strains, and is often antibiotic-resistant [220,221].
Understanding the mechanisms that are stimulated by acute glucose and dietary amino
acids and lead to increased tolerance will help to identify targets for pharmacological
treatments. The Drosophila model of infection may therefore prove useful in further
dissection of acute, glucose and amino acid-stimulated host tolerance of this human
pathogen.
97
Fig. S1. Antimicrobial peptides (AMPs) are induced by B. cepacia infection to similar expression levels in Per mutants and wild type. There is no difference in AMP expression between Per mutants and wild type (WT) observed before or after B. cepacia infection. Shown here are expression levels measured by qPCR for five antimicrobial peptides (AMPs): Attacin (A), Cecropin (B), Defensin (C), Drosocin (D), and Metchnikowin (E). p-values obtained by unpaired t-test; n.s.=p>0.05.
0
400
800
0
100
200
300
0
40
80
A BAttacin
Hours post-infection0 6 12
n.s. n.s.n.s.800
400
0
E
Hours post-infection0 6 12
n.s. n.s.n.s.Metchnikowan
Cecropin
Hours post-infection0 6 12
n.s.n.s.n.s.
D Drosocin
Hours post-infection0 6 12
n.s. n.s.n.s.
1000
500
0
C
Hours post-infection0 6 12
n.s. n.s.n.s.Defensin
WTPer
Figure S1: Allen et al., 2014 (Related to Fig. 1)
Rel
ativ
e ex
pres
sion
Rel
ativ
e ex
pres
sion
98
Fig. S2. Rapamycin reduces survival of B. cepacia infection. Flies injected with 9.6 ng rapamycin at the time of infection (n=59) died more quickly following B. cepacia infection than flies injected with control buffer (n=72, p<0.0001).
99
Fig. S3. Expression of Tsc1 and Tsc2 transcripts are induced by heat- shock. After heat-shock treatment at 29°C for 0, 24, 48, or 72 hours (left to right), adult Gal80-ts; tub-Gal4>UAS-Tsc1/Tsc2 flies showed increased expression of A) Tsc1 and B) Tsc2. Transcript levels for each sample were normalized to total RNA level by comparison to ribosomal protein 1 (rpl4). C) Schematic of the TOR pathway.
100
Fig. S4. Amino acid-stimulated tolerance is facilitated by the gut microbiota. Amino acids must be ingested (enter through the gut) to stimulate host tolerance. Amino acids injected 2 hours prior to infection (A) or post- infection (B) did not improve survival outcome (p>0.05 for both). C) Bacterial 16S ribosomal DNA is not detected after PCR in germ-free flies, but is present in conventionally-raised flies. D) Survival curves for all 4 conditions for germ-free fly experiments: conventionally-raised flies fed glucose, conventionally-raised flies fed glucose and amino acids, germ-free flies fed glucose, and germ-free flies fed glucose and amino acids. Dietary amino acids improve survival of germ-free flies (p<0.0001) to a lesser extent than they improve survival of conventionally- raised flies (also p<0.0001).
101
Table S1. Primer sequences.
Gene Left primer Right primer
AttA CACAATGTGGTGGGTCAGG GGCACCATGACCAGCATT
CecA1 TCTTCGTTTTCGTCGCTCTC CTTGTTGAGCGATTCCCAGT
Def TTCTCGTGGCTATCGCTTTT GGAGAGTAGGTCGCATGTGG
Dipt ACCGCAGTACCCACTCAATC CCCAAGTGCTGTCCATATCC
Dro CCATCGAGGATCACCTGACT CTTTAGGCGGGCAGAATG
Drs GTACTTGTTCGCCCTCTTCG
CTTGCACACACGACGACAG
Mtk TCTTGGAGCGATTTTTCTGG TCTGCCAGCACTGATGTAGC
Rpl1 TCCACCTTGAAGAAGGGCTA TCCACCTTGAAGAAGGGCTA
Tsc1 GTAAACACACCTTGTCCAAGCAGC TGACAGATGGATAGACGGAACCAC
Tsc2 ACACATAGACAACAACGAGAGG
AAGAGATATCATGGCAGGATGC
102
AUTHOR CONTRIBUTIONS
Experiments were conceived by MSH with input from VWA and RMO. VWA, RMO,
and CGZ performed all survival and bacterial load assays after B. cepacia infection;
VWA and RMO analyzed these experiments. In addition, VWA performed and analyzed
the AMP induction experiments and starvation assay; RMO performed and analyzed the
metabolic storage assays. EFS performed and analyzed the bead inhibition experiments
and Western blot analysis. VMH performed and analyzed the melanization experiments.
RBP performed the axenic (germ-free) fly experiments. KRM and WWJ performed the
radioactive and CAFE feeding experiments. MSH, JCC, VWA, and RMO wrote the
manuscript and constructed the figures.
103
ACKNOWLEDGMENTS
We thank all members of the Shirasu-Hiza lab for support and advice; in particular, we
thank Emily Marcinkevicius for insightful comments on experiments and this manuscript.
We also thank Shawn Jordan and Tim Davies for critical reading of this manuscript. We
are grateful to Joel Shirasu-Hiza for assistance in writing our custom analysis software,
Count the Dead; Marc Dionne and David Schneider for support and productive
discussions; and Adam Ryan, Albert Kim, and Paul Kim for technical support. We thank
Gerard Karsenty for use of his qRT-PCR machine, microplate reader, and film developer;
and Rodney Rothstein for use of his Nanodrop. This work was supported by: NIH
1F31NS080673 (EFS); NIH DP2 OD008773 (JCC); NIH R21 DK092735 (WWJ); a
Hirschl Career Award from the Irma T. Hirschl Foundation (MSH); and NIH R01
GM105775 (MSH).
104
CHAPTER 4:
Phagocytic immune cell defects in the Drosophila model of Fragile X syndrome
Work in progress:
Elizabeth F. Stone, Reed M. O’Connor, Emily Marcinkevicius, Vanessa M. Hill, Jennifer S. Ziegenfuss, Clarice G. Zhou, Wesley B. Grueber, and Mimi M. Shirasu-Hiza
105
SUMMARY
Having found significant differences in immune system function in the absence of
circadian regulation, I next wanted to know how the immune system functions in the
Drosophila model of Fragile X syndrome, a disease that is known to be associated with
circadian dysregulation. Fragile X syndrome is the most common single locus genetic
cause of intellectual disability and autism and results from the loss of the RNA-binding
protein FMRP that regulates the translation of a whole host of mRNA targets. In addition
to circadian dysregulation, patients and animal models of the disease exhibit defects in
learning and memory, altered synaptic plasticity, and increased dendritic arborization.
Altered immune system parameters observed in patients with Fragile X syndrome and
autism suggests an altered immune system function in these diseases, but immunity has
not been well characterized, either in humans or in animal models. Here I ask, how does
the loss of dFMR1, the Drosophila homolog of FMRP, affect immune system function
both in the body and in the brain?
Dr. Mimi Shirasu-Hiza and I conceived of all of the experiments enclosed in this
chapter and I have written the manuscript with input from Dr. Shirasu-Hiza. I initially
conducted all the systemic immunity experiments described in this chapter using dFMR1
transheterozygous hypomorphs, made by crossing flies with the dFMR1B55 allele to flies
with the null dFMR1 Δ50M allele and compared them to their isogenic wild-type control
(see Appendix II). At the time I was unable to generate dFMR1 null animals in lab. I
have since optimized husbandry conditions to obtain dFMR1 null animals by raising the
dFMR1 mutants and wild-type controls on food containing low amounts of glutamate
[222]. To avoid recessive background mutations, I used a transheterozygous null for all
106
of the experiments listed in this chapter. The two dFMR1 mutant alleles used for these
experiments were dFMR1Δ50M and dFMR13 and their isogenic wild-type controls.
Others in the lab have also contributed to this work by optimizing experiments,
repeating experiments, and assisting in data analysis. I conducted all of the systemic
immunity experiments in this chapter with dFMR1 hypomorphs (see Appendix II) as well
as null mutants (Fig 1A-G). Vanessa Hill conducted the L. monocytogenes infection and
the melanization experiment on the dFMR1 null animals (Fig 1C, 1G, 2A-B). Reed
O’Connor and I performed the AMP experiment (Fig 2E-F). I performed all of the initial
hemocyte phagocytosis experiments, with subsequent trials performed in collaboration
with Dr. Emily Marcinkevicius (Fig 2H-I). I performed the genetic crosses for necessary
fly lines and initiated the glial phagocytosis assays in Figs 3, 4, and 5, with some of the
subsequent trials performed in collaboration with Reed O’Connor, who was invaluable in
optimizing the immunostaining and confocal microscopy conditions.
107
ABSTRACT
Autism spectrum disorder has become a major national health crisis. Most research on
autism spectrum disorder has focused on neuronal defects; however, recent evidence
suggests that immune system dysfunction correlates with this neurological disease. The
role of glia, or non-neuronal immune cells in the brain, remains unclear. Here we
examine a Drosophila model of Fragile X syndrome, the most common genetic cause of
autism, for immune cell defects. Fragile X syndrome is caused by loss of FMR1 function
and flies lacking the Drosophila homolog dFMR1 display neurological phenotypes
similar to patients with Fragile X syndrome, including dendritic over-arborization,
defects in learning and memory, and loss of circadian regulation. We found that dFMR1
mutants have defective immune responses to bacterial pathogens, and that these defects
are the result of aberrant phagocytic activity in the flies’ hemocytes. Additionally, we
show that these mutants exhibit defective phagocytosis by ensheathing glia in the brain in
response to neuronal injury: glia in dFMR1 mutants show significant delays in clearance
of axonal debris and in their expression of the critical phagocytic receptor Draper. These
results represent the first in vivo study demonstrating a defect in a specific glial function
in a model of Fragile X syndrome. Thus our work establishes a new model of
neuroimmune dysregulation in an autism spectrum disorder model, demonstrating a
possible role for glia in pathogenesis, with potential future implications for therapeutic
approaches to the disease.
108
INTRODUCTION
The prevalence of autism spectrum disorders has reached a modern crisis, occurring at a
rate of 1 in 68 children [223]. The vast majority of autism cases occur sporadically, but
the most common known monogenic cause of intellectual disability and autism is Fragile
X syndrome. Thus Fragile X syndrome is a highly utilized model for studying autism and
intellectual disability. In Fragile X syndrome, transcriptional silencing of the FMR1 gene
leads to absence of the protein FMRP, an mRNA binding protein and translational
regulator. FMRP is ubiquitously expressed but is particularly abundant in neurons [91].
Loss of FMR1 in humans and animal models causes neurons to display excessive growth
of dendritic spines and defects in synaptic plasticity, but the mechanism underlying the
neuronal dysfunction is unknown [132].
One factor that may contribute to the neurological symptoms of autism is immune
system dysfunction. Autism is strikingly correlated with both maternal infection and
autoimmune diseases during pregnancy; this suggests that prenatal immune challenge in
the mother may cause immune misregulation in the fetus, contributing to the
development of autism. In support of this hypothesis, numerous studies have
demonstrated that children with autism express auto-antibodies with reactivity against
brain-specific antigens [224], and others have suggested an association between specific
human leukocyte antigen alleles and autism [225]. Fragile X syndrome is also associated
with altered immune system function, including elevated proinflammatory cytokine
levels in the blood [226] and gastrointestinal disorders [95]. However, whether altered
immune system function in the body is a cause or consequence of neuronal dysfunction
remains unclear.
109
Recent studies have also suggested a role for glia, which are the resident immune
cells in the brain, in the etiology of Fragile X syndrome and autism. FMR1 is expressed
in mouse glia during embryonic development as well as postnatally [227]. FMR1 mouse
mutants exhibit significant and often subtle defects in different subtypes of glia, including
a decreased number of radial glial cells during neocortical neurogenesis [228]; atypical
activated reactive astrocytes in specific brain regions [229]; decreased glutamate
transporter expression in astroglial cells [230]; and delayed myelination due to impaired
oligodendrocyte precursor cells [231]. In an in vitro co-culture assay, FMRP knockout
neurons grown with wild-type glia developed a wild-type architecture, while wild-type
neurons grown with FMRP knockout glia developed a morphology characteristic of
FMRP knockout neurons. These results demonstrate that glia are disordered in the
absence of FMRP and may play an important role in regulating neuronal structure. One
of the main cellular functions of glia is phagocytosis, or the engulfment of extracellular
material, which is mediated by phagocytic receptors and cytoskeletal rearrangement [22-
24]. Still, the role of FMR1 in glial phagocytosis has not yet been examined.
In both vertebrates and invertebrates, it has become increasingly clear that
phagocytosis by glia is important for neuronal structure and function [232-234]. Rather
than being passive structural cells or vacuum cleaners that merely clear the extracellular
spaceof debris, glia are now understood to play an active role in shaping their local
environment. Glia contribute to the restructuring of neuronal circuits by phagocytosing
synaptic elements and newborn cells [235]. Recently, phagocytic properties of microglia
have been implicated in a mouse model of Rett syndrome, which in humans causes
autism spectrum disorder [236]. This suggests that neuronal disease may be caused by a
110
defect in glial phagocytosis. Glial phagocytosis is additionally known to be required for
dendritic pruning. Because FMR1 mutants exhibit defects in dendritic pruning leading to
an over-arborization of dendritic branching, glial phagocytosis in these mutants could be
dysregulated or absent. Glial phagocytosis has not yet been examined in any model of
Fragile X syndrome.
The Drosophila dFMR1 mutant is a well-established model for studying specific
aspects of Fragile X syndrome. Drosophila has a single FMR1 family member (dFMR1),
which is most similar to vertebrate FMRP with a high degree of amino acid sequence
similarity [237]. In many key parameters, loss of dFMR1 in Drosophila phenocopies
mammalian models of Fragile X syndrome. Similar to vertebrate models, Drosophila
dFMR1 mutants display neurological symptoms: defects in learning, memory, and
circadian regulation [120], as well as similar neuroanatomical defects and synaptic
dysfunction [238]. dFMR1 is also functionally conserved on the molecular level: dFMR1
regulates the translation of Drosophila homologs of putative targets regulated by
mammalian FMRP [121,122]. However, while much research has been undertaken to
examine the neuronal defects in animal models of Fragile X syndrome, immune system
function in other cell types—particularly in non-neuronal, glial cell types in the brain, of
dFMR1 mutants has not been investigated.
This is the first study to investigate specific immune mechanisms both in the body
and in the brain using the Drosophila model of Fragile X syndrome. Here we identify
several pathogens to which dFMR1 mutants are more susceptible, and we show that
dFMR1 mutants exhibit elevated basal expression of anti-microbial peptides but reduced
phagocytic activity. These results are the first demonstration of a cellular immune system
111
defect in the Drosophila model for Fragile X syndrome. After observing a hemocyte
phagocytosis defect in the body, we next investigated glial phagocytosis in dFMR1
mutants using an axonal injury model. We found that dFMR1 mutants exhibit reduced
clearance of axonal remnants after injury. We observed a decrease in the induction of
Draper, a receptor crucial for phagocytosis, in dFMR1 mutants after injury, suggesting
that glia are less activated after injury in dFMR1 mutants. This is the first study to
demonstrate a defect in a specific glial function in vivo in dFMR1 mutants, namely glial
phagocytosis. Thus, as glial phagocytosis is crucial for proper dendritic pruning and
synaptic function, these results suggest that glia may play a critical role in the
pathogenesis of Fragile X syndrome.
112
MATERIALS AND METHODS
Fly and bacterial strains
In the following experiments, we generated flies transheterozygous for two dFMR1 nulls
alleles, dFMR1Δ50M (a kind gift from D. Zarnescu [122]) and dFMR13 (a kind gift from T.
Jongens [239]), and their isogenic controls. Both of these alleles were outcrossed to their
isogenic controls (Oregon R and iso31b, respectively) for at least 6 generations. The
control flies were transheterozygous for OreR and iso31b. To label hemocytes with GFP,
we recombined a hemocyte-specific Gal4 driver (hmlΔ-Gal4) with UAS-GFP(2x) and
crossed this to dFMR1 and control lines. To label a specific population of olfactory
neurons with GFP, we crossed the OR85e>mCD8::GFP reporter (a kind gift from M.
Freeman [240]) into dFMR1 and control lines.
Infections were performed with the following bacteria as previously described
[63]: Streptococcus pneumoniae strain SP1, a streptomycin-resistant variant of D39 and
gift from Elizabeth Joyce in Stan Falkow’s laboratory at Stanford University [156];
Listeria monocytogenes strain 10403S, gift from Julie Theriot at Stanford University
[157]; Serratia marcescens strain DB1140, gift from Man Wah Tan at Stanford
University [158]; Escheria coli strain DH5-alpha; and Micrococcus luteus ATCC strain
4698. S. pneumoniae was frozen at OD600 from 0.15-0.4 in 1 ml aliquots with 10%
glycerol, pelleted upon thawing, rediluted in BHI (Brain Heart Infusion media, Difco) to
an OD600 0.12 for injection into flies, which were incubated at 29˚C after injection.
L. monocytogenes was grown in standing BHI, overnight at 37˚C, and diluted to OD600 of
0.1 for injection into flies, which were incubated at 25˚C. S. marcescens was grown in
shaking BHI overnight at 37˚C and diluted to OD600 ranging from 0.1 to 0.6 for injection
113
into flies, which were incubated at 29˚C. E. coli was grown in shaking LB, overnight at
37˚C, and diluted to OD600 0.1 for injection into flies, which were incubated at 25˚C. M.
luteus was grown shaking in BHI, overnight at 37˚C, and diluted to OD600 0.1 for
injection into flies, which were incubated at 29˚C.
Fly rearing conditions
Flies were raised at 25°C, 55-65% humidity onyeasted low glutamate food in a 12h
light/dark cycle. The recipe for the low glutamate food was from Chang, et al, 2008
[222]. Briefly, the low glutamate food contains equal parts 6.25% v/v cornmeal, 6.25%
v/v molasses, 25.75 g/L Baker’s yeast, and 9.25 g/L agar. Flies collected for survival and
hemocyte phagocytosis experiments were maintained on standard dextrose food. The
recipe for standard dextrose food is as follows: 38 g/L cornmeal, 20.5 g/L yeast, 85.6 g/L
dextrose, 7.1 g/L agar. All infection experiments were performed with male flies, 5-7
days post-eclosion. Flies collected for glial phagocytosis assays were maintained on low
glutamate food and were kept on low glutamate food after maxillary palp excisions.
Injections
Flies were anesthetized on CO2 pads. Injections were performed using a custom Tritech
microinjector (Tritech) and pulled glass capillary needles. 50 nL of liquid were injected
into each fly, as calibrated by measuring the diameter of the expelled drop under oil.
Survival assays
114
As described previously [63], 60-85 flies per genotype per condition were assayed for
each survival curve and placed in 3 vials of standard dextrose food with approximately 20
flies each. In each experiment, 20-40 flies of each line were also injected with media as a
wounding control. Death was recorded hourly or daily depending on the virulence of the
pathogen. Data was converted to Kaplan-Meier format using custom Excel-based
software called Count the Dead. Survival curves are plotted as Kaplan-Meier plots and
statistical significance is tested using log-rank analysis using GraphPad Prism software.
All experiments were performed at least three times and yielded similar results.
Bacterial load quantitation
Following microbial challenge, six individual flies were collected at each time point.
These flies were homogenized, diluted serially and plated on appropriate media (tryptic
soy blood agar for S.pneumoniae, LB agar for all others). Statistical significance was
determined using non-parametric Mann-Whitney tests. All experiments were performed
at least three times and yielded similar results.
Melanization assay
The melanization assay was performed as described previously [63]. Briefly, flies were
injected with L. monocytogenes (OD600 0.1) and incubated at 29˚C. 5 days after infection,
flies were visually inspected for melanin deposits. Each deposit was assigned a point
value based on its estimated size. The point scale ranged form 0.5 to 4, where 4
rpresented the largest deposits. The point values for each fly were totaled to give a
melanization index. Separate melanization indices were calculated for melanization at
115
the injection site, which results from wounding, and disseminated melanization, which
results from the systemic response to infection.
Antimicrobial peptide gene expression
Flies were injected with 50 nL of bacterial culture at OD600 0.10 (M. luteus and E. coli).
Following injection, flies were placed in vials containing dextrose food and incubated at
25˚C for 6, 24, and 48 hours. Three groups of six flies after each time point were
homogenized in Trizol and stored at -80˚C until processed. RNA was isolated using a
standard Trizol preparation and the samples treated with DNAse (Invitrogen). Fermentas
RevertAid First Strand cDNA synthesis kit was used to produce the cDNA following the
manufacturer’s protocol. Quantitative RT-PCR was performed using a Bio-Rad CFX
Connect system, with Roche FastStart Universal SYBR Green Master (Rox) mix and the
following primer sets: Drosomycin, Diptericin, Rpl1 [160]. Total mRNA concentration
was normalized using Rpl1 expression. Differences in the infection-induced gene
expression were calculated by normalizing to basal gene expression at the same time
point; p-values were obtained by Mann-Whitney test.
Hemocyte phagocytosis assay
5-7 day old male flies were injected with 50 nL of 20 mg/ml pHrodo-labeled S. aureus in
PBS (Molecular Probes, cat# A10010). The flies were allowed to phagocytose the
particles for 35-45 minutes. The dorsal surface of the thorax was glued onto cover glass
(VWR, cat# 3406)with Loctite brand super glue. Images were taken using epifluorescent
illumination with a Nikon Eclipse E800 microscope fitted with a Photometrics Cool
116
Snaps HQ2 camera. Images were captured with Nikon Elements software. The
experiment was repeated three times with 5-6 flies for per genotype. Images were
thresholded to generate ROIs corresponding to fluorescent phagocytic compartments.
The same threshold was used for all images within a given experiment. ROI pixel areas
and intensities were calculated using Nikon Elements software and summed to generate a
single intensity value for each fly, called the sum intensity. Significance was calculated
between genotypes using the student t-test. To elucidate whether dFMR1 mutants had
fewer hemocytes or hemocytes that were less active, we performed a phagocytosis assay
on flies expressing a UAS-GFP under the hemocyte-specific promoter hmlΔ-Gal4,
driving hemocyte-specific GFP expression.
Glial phagocytosis assay and immunohistochemistry
Glial phagocytosis assays were conducted as previously described [241] with some
adjustments. The maxillary palps of 7 to 10-day-old OR85e>mCD8::GFP; dFMR1
mutants and wild-type controls were excised to sever the olfactory neurons. Flies were
collected and decapitated at various time points after wounding, and the fly heads were
fixed for 40 minutes at room temperature in 4% paraformaldehyde in PBS + 0.1% Triton
X-100 (PTX). Fly heads were washed 5 times in PTX and brains were dissected in ice
cold PTX. Brains were blocked in 4% normal donkey serum in PTX and incubated in
primary antibodies at 4˚C for 22-24 hours. Primary antibody was diluted in PTX with 2%
normal donkey serum and 0.02% sodium azide; the primary antibody was reused up to
three times and stored at 4˚C for up to 4 weeks. The following primary antibodies and
dilutions were used: rabbit anti-Draper (a kind gift from M. Freeman, 1:500 [242]);
117
chicken anti-GFP (Abcam, #ab13970, 1:1000). Brains were then washed 5 times over the
course of 1 hour at room termperature in PTX and incubated at 4˚C for 22-24 hours.
Secondary antibodies were diluted in PTX with 2% normal donkey serum and 0.02%
sodium azide and made fresh for each experiment. The following secondary antibodies
and dilutions were used in this study: Rhodamine-Red X donkey anti-rabbit IgG (Jackson
Immunoresearch, # 711-295-152, 1:200) and AlexaFluor 488 donkey anti-chicken IgG
(Jackson, # 703-545-155 1:200). Brains were mounted onto coverslips that had been
coated with poly-L-lysine and Kodak Photoflow (10 mL poly-L-lysine and 36 uL
Photoflow), dehydrated by incubating in increasing concentrations of alcohol for 5
minutes (50%, 75%, 95%, 100%, 100%), and incubated in two different solutions of
100% xylenes for 10 minutes each. Coverslips were mounted onto slides with DPX
(Electron Microscopy Service # 13510) and slides dried overnight in the dark at room
temperature. Slides were stored at 4˚C before imaging.
Confocal microscopy was performed using a Zeiss LSM 510 Meta with 488nm
and 533nm lasers and an AxioImager Z1 upright microscope with a PlanNeoFL 40X/1.3
oil objective. Slices with a depth of 8 bit were taken from the anterior to posterior
extremities of the bilateral glomeruli, as indicated by GFP staining. The sum of the mean
fluorescent intensity, using both the GFP and rhodamine red channels, of the middle three
slices of each glomerulus were quantified and normalized to background separately using
ImageJ.
118
RESULTS
dFMR1 mutants are sensitive to infections with specific pathogens
The innate immune system of Drosophila is composed of three main immune
mechanisms that help to kill pathogens: melanization, antimicrobial peptide (AMP)
production, and phagocytosis. Drosophila use specific immune mechanisms to defend
against different bacterial pathogens. To assess whether the loss of dFMR1 causes
specific defects in immunity, we first compared the survival kinetics of dFMR1 mutants
and wild-type isogenic controls after infection with four different pathogens known to
elicit immune system activation. We found that dFMR1 mutants died faster than wild-
type controls after infection with S. pneumoniae (Fig 1A, p < 0.0001), S. marcescens (Fig
1B, p < 0.0001), and P. aeruginosa (Fig 1D, p < 0.0001). Survival rates of dFMR1
mutants and wild-type controls did not differ after infection with L. monocytogenes (Fig
1C, p = 0.1702). These results suggest that dFMR1 mutants have defects in specific
immune mechanisms, which cause sensitivity to particular infections.
Decreased survival could occur through defects in resistance, defined as the
ability to kill and clear bacteria, or through defects in tolerance, defined as the ability to
withstand the pathogenic affects of infection. To determine if dFMR1 mutants were less
resistant to infection, we next examined bacterial loads in dFMR1 mutants and wild-type
controls at different time points after infection. We found that dFMR1 mutants have more
bacteria per fly compared to wild-type controls at specific time points after infection with
S. pneumoniae (Fig 1E) and S. marcenscens (Fig 1F), indicating dFMR1 mutants were
less able to kill and clear bacteria. That is, dFMR1 mutants are less resistant to these
pathogens. As expected, dFMR1 mutants and wild-type controls had no consistent
119
Fig. 1. dFMR1 mutants are less resistant to infections with specific pathogens. (A-D) Shown here are Kaplan-Meier survival curves of wild-type controls (WT, black circles) and dFMR1 mutants (dFMR1, purple circles) after injection with select pathogens. dFMR1 mutants died more quickly than wild-type controls after infection with (A) S. pneumoniae (p < 0.0001), (B) S. marcescens (p < 0.0001), and (D) P. aeruginosa (p < 0.0001). dFMR1 had the same survival kinetics after infection with (C) L. monocytogenes (p = 0.1702). (E-H) dFMR1 mutants have higher bacterial loads, in other words are less resistant, to infections with (E) S. pneumoniae and (F) S. marcescens. dFMR1 mutants have no consistent difference in bacterial loads after infections with (G) L. monocytogenes and (H) P. aeruginosa. p-values for survival
101
102
103
104
105
6.0 6.5 7.0 7.5 8.00
25
50
75
100
103
105
107
0 5 10 15
A B
0
25
50
75
100
Days post-infection
0
25
50
75
100
p < 0.0001
Days post-infection
p < 0.0001
WT
dFMR1
S. pneumoniae S. marcescens
Perc
ent surv
ival
Perc
ent surv
ival
Figure 1: Stone et al., 2014
n.s. n.s.
0 3 18
**
E F
Bacte
rial lo
ad/fly
S. marcescens
Hours post-infectionHours post-infection
n.s.
0 24 48
n.s.
Bacte
rial lo
ad
/fly
S. pneumoniae
0 2 4 6
0 3 6 9
n.s. n.s.
0 24 48
n.s.
G H
Bacte
rial lo
ad/fly
P. aeruginosa
Hours post-infectionHours post-infection
n.s. n.s.
0 2 4
n.s.
105
106
107
108
Bacte
rial lo
ad
/fly
L. monocytogenes
C D
0
25
50
75
100
Hours post-infection
p < 0.0001
Days post-infection
n.s.
L. monocytogenes P. aeruginosa
Perc
ent surv
ival
Perc
ent surv
ival
102
103
104
p = 0.0043
**
p = 0.005
104
106
120
curves were obtained by log-rank analysis. p-values for bacterial load analyses were obtained by Mann-Whitney nonparametric t-tests. difference in bacterial loads after infection with L. monocytogenes (Fig 1G), a pathogen
that produced no survival difference between these two groups. In contrast, dFMR1
mutants and wild-type controls had equivalent bacterial loads after P. aeruginosa
infection (Fig 1H), even though dFMR1 mutants were more sensitive to this infection.
Similar bacterial loads with reduced survival after infection with P. aeruginosa suggest
that dFMR1 mutants are less tolerant of P. aeruginosa compared to wild-type controls; in
other words, dFMR1 mutants are less able to withstand the pathogenic effects of this
infection. Because the fly uses different immune mechanisms to fight these specific
pathogens and dFMR1 mutants have different survival kinetics depending on the
pathogen, these data suggest that only specific immune mechanisms are regulated by
dFMR1.
dFMR1 mutants exhibit a reduced melanization response
While it is difficult to investigate a defect in tolerance, a phenomenon that is not well
understood, there are many established assays for measuring particular resistance
mechanisms. Thus, we next sought to identify which specific resistance mechanisms are
regulated by dFMR1. The three best-characterized immune system mechanisms that flies
use to combat infections are antimicrobial peptide synthesis, melanization, and
phagocytosis by hemocytes, which are analogous to primitive macrophages. To
determine if any of the three immune mechanisms are regulated by the dFMR1, we next
121
examined each of these three immune mechanisms in dFMR1 mutants and wild-type
controls.
We first tested whether the loss of dFMR1 leads to changes in the antimicrobial
peptide response, the most well-characterized branch of the Drosophila innate immune
system. Antimicrobial peptides (AMPs) are small peptides that have bacteriocidal
properties and are produced by the Drosophila fat body in response to bacterial and
fungal infections. The well-conserved Toll and imd signaling pathways culminate in
AMP induction and are mainly induced by Gram positive and Gram negative bacterial
infections, respectively. In order to determine if dFMR1 mutants have altered AMP
induction, we examined mRNA levels of two different AMPs, Drosomycin (Drs) and
Diptericin (Dpt), which are induced after activation of the toll and imd signaling
pathways, respectively. Although AMP induction is known to occur following P.
aeruginosa infection [243], the rapid time course of this infection is not suitable for
assaying long-term AMP induction. Instead, we examined basal levels of Drs and Dpt in
uninfected flies and in flies injected with M. luteus and E. coli, used most often to induce
the toll and imd pathways, respectively. Basal levels of Drs and Dpt were increased in
dFMR1 mutants (Fig 2A, p = 0.0370 and Fig 2B, p = 0.0010, respectively), but there
were no significant differences in Drs and Dpt expression at 5 and 24 hours after
infection in dFMR1 mutants and wild-type controls (Figs 2A and 2B). These results
suggest that loss of dFMR1 causes an increase in basal levels of AMP expression but
does not alter AMP response after infection in Drosophila.
122
Fig. 2. dFMR1 mutants exhibit altered immune system parameters.
(A-B) dFMR1 mutants have increased basal levels of antimicrobial peptides. dFMR1 mutants exhibited increased basal levels of (A) Drosomycin (p = 0.0370) and (B) Diptericin (p = 0.0010) compared to wild-type controls, but infection-induced levels of Drosomycin and Diptericin were equivalent between dFMR1 mutants and wild-type controls. (C-E) dFMR1 mutants have reduced systemic melanization after infection with L. monocytogenes compared to wild-type controls. (C) Schematic of the melanization assay: 5 days after infection with L. monocytogenes, systemic melanization and melanin deposited at the wounding site were calculated. Briefly, for each fly, areas of melanin deposition were given a score from smallest to largest of 0.5 to 4. These scored were summed for each fly to generate the melanization index for each fly. (D) dFMR1 mutants have reduced systemic melanization when the melanization index was calculated (p < 0.0001) (E) but equivalent melanin deposition at wounding sites (p = 0.1949) after infection with L. monocytogenes. (F-H) dFMR1 mutants exhibit reduced phagocytic activity compared to wild-type controls. (F) Schematic of the phagocytosis assay: S. aureus labeled with the pH sensitive pHrodo fluorophore was injected into wild-type and dFMR1 mutants flies. pHrodo only fluoresces in acidic environments; only hemocytes that have phagocytosed the pHrodo-labeled bacteria fluoresces. The dashed line represents the
123
thorax where fluorescence was quantified. (G) Representative micrographs showing thoraces from a wild-type fly and a dFMR1 mutant. (H) Phagocytic activity in wild-type and dFMR1 mutants was quantified by measuring the sum of the intensity, which is a measurement generated from both the area and the intensity of each fluorescent puncta in the thorax of each fly. dFMR1 mutants exhibit significantly less phagocytic activity compared to wild-type controls (p = 0.0159). p-values for melanization and antimicrobial peptides were calculated by the Mann-Whitney nonparametric t-test. p-values for antimicrobial peptide were obtained by the Mann-Whitney nonparametric t-test. p-values for phagocytic activity were calculated using an unpaired t-test. dFMR1 mutants exhibit reduced systemic melanization after infection with L.
monocytogenes compared to wild-type controls
We next tested whether the loss of dFMR1 affects the melanization response. The process
of melanization produces reactive oxygen species that are toxic to pathogens and
parasites and leads to the deposition of melanin both systemically and at local sites of
wounding and infection. Melanization is the main defense used by Drosophila to fight
infections with L. monocytogenes [70]. To measure melanization, dFMR1 mutants and
wild-type controls were infected with L. monocytogenes and inspected for melanin
deposit five days after infection (Fig 2C). A melanization index was calculated for each
fly based on the number and size of melanin deposits present. Deposits at the injection
site, which primarily develop in response to wounding rather than infection, were scored
separately. Although we had previously found there to be no survival difference between
dFMR1 mutants and wild-type controls after infection with L. monocytogenes, we found
a significant decrease in systemic melanin deposition in dFMR1 mutants (Fig 2D, p <
0.0001). However, melanization at the injection site was comparable between the two
groups (Fig 2E, p = 0.1949), indicating that dFMR1 mutants are capable of producing
melanin but are specifically defective in depositing melanin systemically in response to a
pathogen. These results suggest that the loss of dFMR1 may decrease the systemic
melanization response after infection but not locally after wounding. Additionally, this
124
altered systemic melanization response does not affect survival after L. monocytogenes
infection. dFMR1 mutants may exhibit wild-type survival rates and bacterial loads after
infection with L. monocytogenes despite reduced rates of melanization in part because of
their increased basal rate of AMP induction, as Toll signaling has been found to protect
flies synergistically with melanization against L. monocytogenes [69].
dFMR1 mutants exhibit decreased phagocytosis of bacteria by hemocytes
Drosophila is particularly dependent on phagocytosis by hemocytes to combat certain
infections, including S. pneumoniae, S. marcescens, and S. aureus [138,155]. To
determine if dFMR1 regulates phagocytosis, we then measured phagocytosis of
fluorescently-labeled bacterial particles in vivo in dFMR1 mutants and wild-type controls.
Flies were injected S. aureus bioparticles conjugated to a pH-sensitive rhodamine
fluorophore. Therefore, only particles that are engulfed and processed in acidic
phagolysosomes generate fluorescent signal (Fig 2F). We imaged the dorsal surface of
the thorax, where fluorescent puncta, representing phagocytosed bacteria, were visible
through the cuticle (Fig 2F). We found a decrease in the intensity of fluorescence in
dFMR1 mutants when compared to wild-type controls (Fig 2F and Fig 2G, p = 0.0159).
This result show that the loss of dFMR1 causes decreased phagocytic activity by
hemocytes, suggesting that dFMR1 promotes phagocytosis.
Loss of dFMR1 results in a decrease in axonal clearance after axotomy
Having found a significant defect in hemocyte phagocytosis, we hypothesized that the
loss of dFMR1 may impact phagocytosis by glia, the phagocytic immune cells in the
125
brain. Glia and hemocytes use the same receptors and phagocytic pathways for
phagocytosis [244]. Glial phagocytosis is crucial for proper pruning during development
and for the removal of dead neurons [242,245]. To determine if dFMR1 mutants have a
defect in glial phagocytosis, we used a well-characterized assay developed by Marc
Freeman’s group to examine axonal clearance by glia after axotomy in dFMR1 mutants
[242]. Briefly, we genetically-labeled olfactory neurons that innervate the maxillary
palps with membrane-bound GFP. The cell bodies of these neurons reside in the
maxillary palp and the neurons project to glomeruli in the antennal lobes. After
surgically excising the maxillary palps, these olfactory neurons cannot survive without
their cell bodies, and the axonal remnants are phagocytosed by glia. With successful glial
phagocytosis, we see a decrease in GFP intensity in the glomeruli after maxillary palp
excision. GFP intensity is identical in unwounded dFMR1 mutants and wild-type
controls. We found that, after axotomy, dFMR1 mutants have less of a reduction of GFP
intensity in the glomeruli 18 hours after maxillary palp excision when compared to wild-
type controls (Fig 3A-B, p = 0.0078), suggesting that dFMR1 mutants have reduced
axonal clearance after axotomy. These results suggest that dFMR1 mutants may have a
defect in glial phagocytosis.
There is a decrease in the induction of Draper, the phagocytic receptor, after
axotomy in dFMR1 mutants
Because we saw a significant decrease in the clearance of axonal remnants in the dFMR1
126
Fig. 3. Clearance of severed axons is delayed in dFMR1 mutants relative to wild-type controls. (A) Shown here are representative micrographs of GFP-labeled glomeruli at 0 and 18 hours after axonal severing of wild-type (top) and dFMR1 mutants (bottom). (B) With quantification, dFMR1 mutants exhibit significantly higher levels of GFP-labeled axonal remnants at 18 hours after injury (p = 0.0078). GFP levels are similar in uninjured dFMR1 mutants and wild-type controls (p > 0.05 ). p-values were obtained by unpaired t-test. mutants, we next asked if there is a defect in ensheathing glia, the specific glial
population that phagocytoses axonal remnants after injury. Both hemocytes and glia
express Draper, a phagocytic receptor required for phagocytosis [246]. Draper
expression is induced early after injury and is associated with the degree of glial
phagocytosis [241]. To investigate if dFMR1 mutants have reduced glial activity in
addition to reduced clearance of axonal remnants, we tested Draper induction after injury
as a measure of glial activation. We found that there is less Draper induction 18 hours
after injury in dFMR1 mutants compared to wild-type controls (Fig 4A and B, p <
0.0001). These results suggest that glial activation after axonal wounding is impaired in
dFMR1 mutants compared to wild-type controls.
Figure 3: Stone et al., 2014
Uninjured Post-injury
WT
dFMR1
A
WT dFMR1 WT dFMR10
100
200
300
400
Mea
n in
tens
ity o
f G
FPim
mun
oflu
ores
cenc
e (a
u)
uninjured post-injury
**p = 0.0078
n.s.p > 0.05
B
127
Fig. 4. Fewer activated glia are seen in dFMR1 mutants than in wild-type controls after axonal severing. (A) Shown here are representative micrographs of GFP-labeled axons and Draper-expressing glia for uninjured wild-type and dFMR1 mutants and18 hours after injury. (B) Quantification shows that Draper induction is the same in uninjured wild-type and dFMR1 mutants but significantly lower in dFMR1 mutants after injury (p < 0.0001). p-values were obtained by unpaired t-test. Clearance of axonal remnants is delayed in dFMR1 mutants while Draper induction
is reduced at numerous time points after axotomy
Having found a defect in axonal clearance and Draper induction at 18 hours after
wounding, we next sought to determine whether glial phagocytosis in dFMR1 mutants is
inhibited or delayed after axotomy. To test this, we performed a time course, examining
Figure 4: Stone et al., 2014
A Uninjured Post-injury
WT
dFMR1
GFP GFPDRAPER DRAPER
B
WT dFMR1 WT dFMR10
255075
100125
uninjured post-injury
****p < 0.0001
n.s.p > 0.05
Mea
n in
tens
ity o
f D
rape
rim
mun
oflu
ores
cenc
e (a
u)
128
Fig. 5. Time course of Draper expression after axonal severing shows that glial phagocytosis is delayed but not abolished in dFMR1 mutants. (A) Shown here are representative micrographs of GFP-labeled axons for wild-type (top) and dFMR1 mutants (bottom) in uninjured flies, 12 hours, 18 hours, 24 hours, and 48 hours after injury. (B) Shown here are representative micrographs of Draper-expressing glia for wild-type (top) and dFMR1 mutants (bottom) in uninjured flies, 12 hours, 18 hours, 24 hours, and 48 hours after injury. (C) Quantification of GFP expression in wild-type and dFMR1 mutants shows significantly higher levels of GFP-labeled axonal remnants at 12 (p = 0.0026), 18 (p = 0.0062), and 48 hours (p = 0.0126) after wounding. Levels of GFP-labeled remnants were the same in uninjured flies (p = 0.1527), and
Figure 5: Stone et al., 2014
0 12 18 24 480
2500
5000
7500
10000
12500
Tota
l GFP
inte
nsity
C
Hours post-injury
**n.s. n.s.** *
0 12 18 24 480
2500
5000
7500
10000
12500
Tota
l DR
APER
inte
nsity
D
Hours post-injury
n.s. n.s. n.s.* *
A
WT
dFMR1
Uninjured 12 hours 18 hours 24 hours 48 hours
GFP
WT
dFMR1
Uninjured 12 hours 18 hours 24 hours 48 hours
DR
AP
ER
B
WTdFMR1
129
levels were not consistently higher at 24 hours (p = 0.3265 for this experiment). (D) Quantificiation shows that Draper induction is the same in uninjured wild-type and dFMR1 mutants (p = 0.8069) and in wild-type and dFMR1 mutants 12 (p = 0.4663) and 48 hours (p = 0.8844) after wounding. Draper induction is significantly lower in dFMR1 mutants 18 (p = 0.0256) and 24 hours (p = 0.0261) after wounding. p-values were obtained by unpaired t-test. GFP intensity of the olfactory neurons and Draper induction at different times after
maxillary palp excision. We found that dFMR1 mutants have significantly more GFP
intensity at 12 (p = 0.0026), 18 (p = 0.0062) and 48 hours (p = 0.0126) after maxillary
palp excision (Fig 5A). There was no consistent difference in GFP intensity at 24 hours
after wounding (p = 0.3265 for this particular experiment). These results show that
axonal remnants are cleared after axotomy in dFMR1 mutants, although the axonal
remnants persist longer in the mutants compared to wild-type controls.
We next investigated Draper induction at various times after axotomy to
determine if Draper induction is also delayed. We found that Draper induction is reduced
at 18 (p = 0.0256) and 24 hours (p = 0.0261) but is no different from wild-type levels by
48 hours post-axotomy (p = 0.8844). These results demonstrate that Draper induction is
not delayed but is moderately inhibited in dFMR1 mutants. Taken together, these
findings suggest that the loss of dFMR1 not only causes general immune system
dysregulation, but also results in a defect in phagocytosis by glia, the resident immune
cells in the brain.
130
DISCUSSION
In addition to neurological symptoms, patients with Fragile X syndrome and autism
spectrum disorders often present with other symptoms, including recurrent otitis media
[247] and gastrointestinal disease [248,249]. Patients with Fragile X syndrome also have
elevated cytokines levels [250], suggesting that they have increased inflammation.
Additionally, when the mothers of patients with Fragile X syndrome had aggravations of
an autoimmune disease or asthma during pregnancy, the children are more likely to have
seizures and tics, suggesting that immune activation affects the brain during development
[251]. The non-neurological symptoms in the absence of FMRP, the contribution of the
immune system on the neurological symptoms, and the etiology of these symptoms are
not well understood. Using the Drosophila model for Fragile X syndrome, we probed the
innate immune system function in the absence of dFMR1. We found the dFMR1 mutants
are more susceptible to infection with particular pathogens. When we examined the
three main branches of the fly’s innate immune system in dFMR1 mutants, we found that
the dFMR1 mutants have increased basal levels of antimicrobial peptides, reduced
reactive oxygen species production after infection, and decreased phagocytosis of
bacteria. Just as human patients with Fragile X syndrome and autism have altered
immune system function, these results demonstrate that dFMR1 mutants have altered
immune system function in all three branches of the fly’s innate immune system.
Glia are the phagocytic immune cells in the brain, and glial phagocytosis is
crucial for proper synaptic and dendritic pruning and for the clearance of axonal debris
[242,252]. Patients with Fragile X syndrome and many forms of autism have defects in
synaptic pruning and synaptic plasticity, but how the absence of FMRP affects glial
131
function on synaptic pruning and plasticity has not been previously studied. Using an
axonal injury model, we investigated ensheathing glial function in the absence of dFMR1.
We found that dFMR1 mutants exhibit reduced axonal clearance after injury. We next
examined the induction of Draper, a phagocytic receptor that is crucial for the activation
of glial phagocytosis, and we found that dFMR1 mutants have reduced Draper induction
after injury, corresponding with reduced glial activation. When we examined axonal
clearance at numerous times after injury, we found that glial phagocytosis is delayed but
is not abolished in dFMR1 mutants. These results suggest that there is a defect in glial
phagocytosis in the absence of dFMR1 mutants. This is the first in vivo evidence of
defects in a specific glial function, phagocytosis, in an animal model of Fragile X
syndrome.
To further characterize the glial phagocytosis phenotype, it would be important to
determine whether the glial phagocytosis defect stems from a defect in glial migration,
glial phagocytic activity, or a combination of these two possibilities. In order to
phagocytose the axonal remnants in the glomeruli, ensheathing glia need to migrate from
the neuropil and infiltrate the glomeruli. If glial motility is delayed, this could be due to
intracellular defects (such as defects in the molecular components of the glial
cytoskeleton or an altered expression of surface receptors) or extracellular defects (such
as defects in the extracellular matrix or defects in the secretion of glial enzymes, such as
metalloproteases, that allow motility through the extracellular matrix). This last
possibility is especially interesting because minocycline, a drug that inhibits
metalloproteases, has been shown to ameliorate some of the neuroanatomical and
neurological defects in animal models of autism spectrum disorder [136,137]. If a
132
migration defect is observed, we can next characterize ensheathing glia during migration
for their cytoskeletal components (including actin, microtubules, intermediate filaments,
and integrins). If, on the other hand, glial motility is unaffected and glial expression of
phagocytic components is delayed, this result would point to a different set of
mechanisms. For example, this defect in phagocytic receptor expression could be due to
defects in glial gene expression, protein translation, or protein trafficking to the
membrane. These results would begin to suggest a potential mechanism that dFMR1 is
employing in glia.
Another aspect of the glial phagocytosis defect that is essential to investigate is
the location in which dFMR1 is required for proper glial function. Both neurons and glia
are thought to actively participate in the process of axonal clearance after severing. After
axonal severing, axonal remnants putatively secrete two types of signals: a "find me"
signal that causes ensheathing glia to migrate specifically to the glomeruli and an "eat
me" signal that activates ensheathing glia to induce Draper expression [253]. The precise
nature and order of secretion of these signals is not known; nor is it known if there are
two discrete signals or a single signal that serves both functions. We hypothesize that
dFMR1 regulates either these neuronal signals and/or the glial response to these signals.
In order to account for the seemingly contradictory increased basal levels of
antimicrobial peptides (AMPs) but decreased hemocyte phagocytosis, we speculate that
the absence of dFMR1 induces an inappropriately primed and “distracted” immune
system that is less able to combat infections. Thus, when the immune system encounters a
pathogen, the immune system may not realize the threat and would then be unable to
combat the infection effectively. How the primed immune system function in dFMR1
133
mutants affects brain function is not yet know. However, recent studies have emphasized
the importance of communication between the gut microbiome, the immune system, and
brain function in mammals. Studies in wild-type mouse [254], rat [255], and rhesus
monkey [256] have indicated that maternal immune activation (MIA) during pregnancy
can cause autistic features in the offspring of these animals. These offspring have
increased inflammatory cytokine signaling and abnormalities in hematopoietic lineages in
both the serum and in different areas of the brain [257]. Furthermore, the offsprings’
microbiome was altered, consistent with the observed differences in the microbiota
between patients with autism and typically developing controls [257]. What causes the
altered microbiota composition and the consequence of this cellular defect on immune
system and neuronal functions have not been fully elucidated. It would be important to
examine whether the differences in immune system function in dFMR1 mutants also
extend to an altered gut microbiota in these mutants.
Using the power of Drosophila genetics, it would be beneficial to dissect the
mechanism behind the so-called gut-microbiota-brain axis. In mammals, the phagocytic,
antigen-presenting dendritic cells, an innate immune cell type, is the link between the
microbiota and the adaptive immune system. If these cells have a defect in phagocytosis,
proper communication between the gut microbiota and the immune system may not
occur. Additionally, proper signaling between the immune system and the developing
brain, mediated by cytokine secretion and circulating immune cells, may be altered,
resulting in the exacerbation of some of the symptoms seen in Fragile X syndrome and
autism. Our results highlight the utility of the Drosophila model of Fragile X syndrome
to examine the innate immune system, and one future question is whether the innate
134
immune system function in the brain apart from glia is dysregulated in the absence of
dFMR1.
Because Fragile X syndrome and autism are considered developmental disorders,
defects in phagocytosis seen during adulthood in dFMR1 mutants may reflect other, more
general defects in phagocytosis by glia during development. In both mouse and
Drosophila, there appear to be two waves of dFMR1 expression, one during development
and the other during adulthood. The wave of dFMR1 expression during development,
specifically the pupal stage, appears to be required for neuronal function in adulthood
[257,258]. The mushroom body, the neuronal structure important for learning and
memory analogous to the hippocampus in vertebrates, has morphological defects in
dFMR1 mutants during adulthood [259]. Interestingly, neurite pruning of gamma neurons
in the mushroom body has been shown to require proper phagocytosis by glia during
neurodevelopment in wild-type flies [252]. We speculate that the mushroom body defect
observed in dFMR1 mutants stems from a glial phagocytosis defect. Taken together,
these studies highlight the need for future experiments to examine immune system
function, particularly glial phagocytosis, on Fragile X syndrome and autism, with the
potential for new therapeutics for these diseases.
135
CHAPTER 5:
CONCLUSIONS
136
Many complex physiologies, including immune system function, feeding, metabolism,
and learning and memory, are regulated by circadian rhythms. Circadian rhythms are
generated by the aptly named molecular clock, whose daily oscillations originating in
clock neurons in the brain drive the expression of hundreds of genes both in the neurons
and in the body. However, the manner in which the molecular clock and circadian
rhythms regulate these other complex physiologies has yet to be elucidated. In this thesis,
I have used the Drosophila model organism to investigate the interaction between the
brain and immune system function. In Chapter 2, I showed that a specific immune
mechanism, phagocytosis, is altered in the absence of circadian regulation. In Chapter 3, I
presented evidence that circadian regulation of two tolerance mechanisms, feeding
behaviors and metabolism, impacts survival after two sepsis-like infections in Drosophila.
And finally, in Chapter 4 I demonstrated that the Drosophila model of Fragile X
syndrome, a neurological disease that is characterized by intellectual disability, autistic
features, and circadian dysregulation, exhibits altered phagocytosis both by immune cells
in the body and in the brain. Here I will summarize my findings and propose future
investigations.
Phagocytosis of bacteria is circadian-regulated in Drosophila
An unfortunate consequence of modern living is circadian dysregulation and is evident in,
for instance, jet lag and shift workers. An interaction between sleep and immune system
function has long been known [260], however, the mechanism behind this interaction has
not been well elucidated. Circadian regulation and innate immunity are well conserved
between Drosophila and mammals, and the genetic tools and assays available for
Drosophila making Drosophila a useful model organism to study the relationship
137
between the two physiologies. We had previously found that wild-type flies lose
circadian regulation when infected with pathogenic bacteria and that two circadian
mutants, per01 and tim01, are more sensitive to infection with particular pathogens
compared to wild-type controls [81]; however, it was not known which (if any) of the
immunity mechanisms is circadian regulated. I investigated this relationship by
examining particular immune system mechanisms using the genetically tractable
Drosophila circadian tim01 mutant.
As I reported here, we found that tim01 mutants are more sensitive to particular
infections, but not all infections tested, suggesting that only particular immune
mechanisms are circadian regulated. In order to fight against pathogens, Drosophila
innate immune system utilizes three main well-conserved immune mechanisms:
antimicrobial peptide synthesis, melanization, and phagocytosis. We found that
phagocytosis, but not antimicrobial peptide synthesis or melanization, is upregulated
during the night compared to during the day in wild-type flies and is reduced in the tim01
circadian mutant. The definition of a circadian-regulated physiology is a process that
exhibits variation of the course of the circadian cycle in wild-types, and that change is
abolished in circadian mutants; thus, our results suggested that phagocytosis is circadian
regulated in Drosophila. Additionally, we found that a defect in phagocytosis resulting
from the absence of circadian regulation affects survival after infection. These results
suggest that Tim protein regulates phagocytosis.
As the mechanism by which Tim specifically regulates phagocytosis remains
unknown, we have a significant opportunity to build on these findings. For instance,
determining if Tim is expressed in phagocytes themselves and required for wild-type
138
phagocytosis would suggest whether Tim acts cell autonomously or non-cell
autonomously to regulate phagocytosis. If Tim is found to work non-cell autonomously, it
would be important to screen for potential downstream effectors between the molecular
clock in the brain and the phagocytes in the body. In addition, we do not know what is
altered in phagocytes in the absence of Tim. For instance, Tim expression at different
times of the day may induce phagocytic surface receptors or other phagocytic machinery
to be differentially expressed, enhancing phagocytosis activity at certain times of the day.
Alternatively, Tim expression may lead to an increased number of phagocytes during
development, leading to increased numbers of phagocytes during the life of the fly, thus
increasing phagocytosis rate compared to tim01 mutants after infection. In the meantime,
the results published here and in PLoS pathogens [63] demonstrate that the absence of
circadian oscillation has a direct, negative effect on phagocytosis after infection,
contributing to poor survival outcomes.
The circadian regulation of feeding and metabolism affects survival of infection in
Drosophila
Sepsis and its related syndromes, severe inflammatory response syndrome and severe
sepsis, are lethal consequences of extreme immune system activation in the presence of
overwhelming infection. The best therapy available to patients with sepsis include
supportive care, such as fluid replenishment, antibiotic treatment, and isolating and
treating the source of the infection [176]. However, patients are still dying of sepsis at an
alarming rate [174], and the optimal treatment is controversial [176]. I pioneered a sepsis-
like infection in Drosophila to further characterize potential metabolic mediators during
139
infection with the aim that these data may deepen the understanding of sepsis.
Here I highlighted our findings that circadian regulation of two tolerance
mechanisms, feeding behaviors and metabolism, impacts survival after infection
in Drosophila. I show that per01 circadian mutants are more tolerant of rapidly lethal,
sepsis-like infections with P. aeruginosa and B. cepecia when compared to wild-type
controls, as they live longer but they have equivalent bacterial loads. These results
suggest that the per01 mutants are better able to tolerate the pathogenic effects of these
infections compared to wild-type flies. The improved survival in per01 mutants was found
to be due to a difference in feeding behavior, as uninfected per01 mutants consumed more
food over the course of the day and that starvation of infected per01 mutants and wild-
type controls eliminated the difference in survival. Increased feeding was demonstrated to
be beneficial for survival after infection with B. cepecia and we determined that glucose
and amino acids were necessary food components required for this survival benefit.
Interestingly, although glucose could be injected into the fly to improve survival, amino
acids needed to be delivered to the flies orally to improve survival, and the fly’s
microbiota mediated the improved survival. These results suggest a complicated
relationship between diet, metabolism and microbiota on survival after infection.
Possible future experiments include elucidating how the gut microbiota interacts
with immunity after infection, either by priming the fly to absorb amino acids during
development [261,262] or by promoting the absorption of specific amino acids that
would otherwise be unavailable to the fly [263]. Another important avenue of study
would be to compare the microbiome between wild-type and per01 mutants and clarify
which Genus or species of bacteria is required for microbiota-mediated immunity. Taken
140
collectively, these results published here contribute to our understanding of feeding and
metabolism on tolerance mechanisms after rapid, sepsis-like infections.
The Drosophila model of Fragile X syndrome exhibits defects in phagocytosis
Lastly, in addition to examining the altered phagocytic immune system function in
Drosophila circadian mutants, I also examined the immune system function in the
Drosophila model of a particular developmental disease, Fragile X syndrome, known to
have circadian dysregulation. Fragile X syndrome is the most common monogenic cause
of intellectual disability and autism. While the effect of Fragile X syndrome on the
immune system has not been well-characterized, patients with autism have altered
immune system parameters, including cytokine expression patterns [264]. I investigated
immune system function both systemically and in the brain in dFMR1 mutants, the
Drosophila model of Fragile X syndrome.
dFMR1 mutants were found to be more sensitive to infections with specific
bacteria but survived like wild-type flies when infected with other bacteria. Drosophila
use different immune mechanisms to fight against specific pathogens, and interestingly, I
found that dFMR1 mutants have defects in all three immune mechanisms, antimicrobial
peptide generation, melanization, which generates reactive oxygen species, and
phagocytosis. Although dFMR1 mutants have equivalent induction of antimicrobial
peptides after infection compared to wild-type controls, they exhibit an increase in the
basal level of two antimicrobial peptides. This result suggests that in an uninfected state,
the immune system function might be upregulated in dFMR1 mutants. We next examined
melanization in dFMR1 mutants with L. monocytogenes, which Drosophila use
melanization primarily to combat this particular infection [70]. Interestingly, dFMR1
141
mutants were found to exhibit decreased systemic melanization after infection with L.
monocytogenes compared to wild-type controls, even though dFMR1 mutants had
identical survival and bacterial loads compared to wild-type controls when infected with
that same pathogen. These results suggest that dFMR1 is required for proper systemic
melanization, although the altered melanization does not affect survival or bacterial load.
This melanization and survival result should be investigated further, not only with other
pathogens to explore whether the melanization phenotype is pathogen-specific or a
general defect in melanization but also to elucidate the mechanism behind this
unexpected phenotype. Because L. monocytogenes infection also leads to antimicrobial
peptide induction [69], the increased basal levels of antimicrobial peptides may keep
bacterial growth in check in lieu of proper melanization, and this relationship should be
examined. Expression levels and activities of key players in the melanization enzymatic
cascade should be also investigated. Additionally, the effects of increased basal levels of
antimicrobial peptide expression and decreased systemic melanization on neuronal
function should be explored.
When dFMR1 mutants were examined for macrophage phagocytic activity, they
were found to have decreased macrophage phagocytosis compared to wild-type controls.
These results suggest that dFMR1 promotes survival after infection and is require for
proper phagocytosis. However, there are many other questions about this phenotype that
should be explored. For instance, do dFMR1 mutants exhibit less phagocytic activity
and/or fewer total phagocytes compared to wild-type controls? Is there a developmental
defect in macrophage phagocytosis? Where is dFMR1 required for proper macrophage
phagocytosis: in the phagocytes themselves or another cell type? Do dFMR1 mutants
142
exhibit different patterns of macrophage distribution compared to wild-type controls,
making our assay examining phagocytosis through the dorsal cuticle a poor measure for
macrophage phagocytosis in these mutants? Is there altered circadian regulation of
phagocytosis in dFMR1 mutants? Additionally, do mammalian models of Fragile X
syndrome also have cellular phagocytosis immune cell defects? It is clear from these
questions that the characterization of the macrophage phagocytosis phenotype needs to be
expanded.
Next, I examined the absence of dFMR1 on glial function. Glia are the resident
immune cells in the brain, and glial phagocytosis has previously been found by others to
be important for proper neuronal connections, learning, and memory [233,265]. Using an
axonal injury model, I found that there is a delay in the clearance of axonal remnants
in dFMR1 mutants and that glia are less activated compared to wild-type controls. These
results suggest that the absence of dFMR1 leads to a defect in glial phagocytosis.
In additional to the experiments suggested in the Discussion of Chapter 3, further
examination of the glial phagocytosis phenotype after axotomy needs to be conducted.
There are three basic steps for glial phagocytosis of axonal remnants after axotomy:
presumed glial activation by an unknown “eat me” signal from the axonal remnants, glial
migration to the axonal remnants, and glial phagocytosis of the axonal remnants [266].
Using Drosophila genetics and the imaging techniques described in Chapter 3, whether
the cause of the defect in glial phagocytosis is due to reduced glial activation, glial
migration, or the glial phagocytosis machinery can be explored. Another important
avenue of study is to determine where dFMR1 is required for proper glial phagocytosis: is
it required in glia, neurons, or both? The results to these experiments would suggest the
143
mechanism behind the role of dFMR1 on glial phagocytosis after axotomy.
Whether this observed adult glial phagocytosis phenotype has any consequences
on neuronal structure or function during development remains unknown. Preliminary
studies suggest that, similar to the glial phagocytosis phenotype observed after axotomy
in adults, there is a delay in normal axonal pruning during larval and pupal development
in dFMR1 mutants, suggesting a defect in glial phagocytosis during development. The
developmental pruning of a population of neurons has been tested; these neurons are
located in the gamma lobe of the mushroom body, the structure in the fly brain important
for learning and memory. This population of neurons is pruned at specific times during
development [245,252]. Other preliminary studies highlight that these developmental
defects in glial phagocytosis may be ameliorated by minocycline, a tetracycline analog
that is presumed to inhibit matrixmetalloproteases that is currently in drug trials for the
treatment of Fragile X syndrome and autism [134,267]. If minocycline is found to restore
glial phagocytosis in dFMR1 mutants, then this would suggest that the defect in glial
phagocytosis may be due to matrixmetalloproteases or the extracellular matrix and would
lead to further avenues of study. These preliminary results suggest that the glial
phagocytosis phenotype extends to a defect during development, and these are the first
data suggesting that glial glial dysfunction may contribute to observed defects in neuronal
structure. Additionally, these results illustrate the need for the characterization of the role
of dFMR1 on glial function as it relates to neuronal dendritic pruning and synaptogenesis,
and thus neural function. If glial function on dendritic pruning and synaptogenesis is
found to be dysregulated in dFMR1 mutants, these results may help guide future
therapeutics for the disease.
144
As I explored in detail above, future studies are required to explore the role of
circadian regulation on systemic immune system function in Fragile X syndrome and
whether the altered systemic immunity impacts neuronal function. While neuronal
function in animal models of Fragile X syndrome has been investigated thoroughly, my
results reported here highlight the need for future studies on how glial function impacts
learning and memory in Fragile X syndrome. In particular, as glial phagocytosis is crucial
for proper synaptic pruning both during development and in adulthood [252,265], the
mechanism behind the glial phagocytosis defect needs to be clarified.
In summary, the findings discussed in this dissertation have built on previous
research by us and other groups and have demonstrated novel interactions between the
brain, metabolism, and immune system function. Each of these three projects has rich
possibilities for future work and discoveries for these complex physiologies with possible
future therapeutic consequences.
145
REFERENCES
1. Zylka MJ SL, Weaver DR, Reppert SM (1998) Three period homologs in mammals: Differential light responses in the suprachiasmatic circadian clock and oscillating transcripts outside of brain. Neuron 20: 1103-1110.
2. Zheng X, Sehgal A (2010) AKT and TOR signaling set the pace of the circadian pacemaker. Curr Biol 20: 1203-1208.
3. Hoffmann JA, Reichhart, J-M. (2002) Drosophila innate immunity: an evolutionary perspective. Nature Immunology 3: 121-126.
4. Wan L, Dockendorff TC, Jongens TA, Dreyfuss G (2000) Characterization of dFMR1, a Drosophila melanogaster Homolog of the Fragile X Mental Retardation Protein. Molecular and Cellular Biology 20: 8536-8547.
5. McClung CR (2006) Plant Circadian Rhythms. The Plant Cell 18: 792-803.
6. Kumar Jha P CE, Kalsbeek A. (2015) Circadian rhythms in glucose and lipid metabolism in nocturnal and diurnal mammals. Mol Cell Endocrinol.
7. Smarr BL JK, Driscoll JR, Kriegsfeld LJ (2014) A time to remember: the role of circadian clocks in learning and memory. Behav Neurosci 128: 283-303.
8. Curtis AM BM, Sassoone-Corsi P, O'Neill LA (2014) Circadian clock proteins and immunity. Immunity 40: 178-186.
9. Burke TM MR, Chinoy Ed, Snider JA, Bessman SC, Jung CM, Wright KP Jr. (2013) Combination of light and melatonin time cues for phase advancing the human circadian clock. Sleep 26: 1617-1624.
10. Muindi F ZJ, Colas D, Heller HC (2013) The acute effects of light on murine sleep during the dark phase: importance of melanopsin for maintenance of light-induced sleep. Eur J Neurosci 37: 1727-1736.
11. Bennie JJ DJ, Inger R, Gaston KJ (2014) Biogeography of time partitioning in mammals. Proc Natl Acad Sci U S A 111: 13727-13732.
12. Ensing EP CS, de Wijs FA, Lentferink DH, Ten Hoedt A, Boyce MS, Hut RA (2014) GPS based daily activity patterns in European red deer and North American elk (Cervus elaphus): indication for a weak circadian clock in ungulates. PLoS One 9: e106997.
13. Wang D OT, Mower J, Shimmin LC, Yim S, Honeycutt RL, Tsao H, Li WH. (2004) Molecular evolution of bat color vision genes. Mol Bio Evol 21: 295-302.
14. Zimmerman RH LL, Nishimura N, Galardo AK, Galardo CD, Arruda ME. (2013) Nightly biting cycles of malaria vectors in a heterogeneous transmission area of eastern Amazonian Brazil. Malar J 12: 262.
15. Rieger D FC, Popp J, Bichler D, Dittmann R, Helfrich-Förster C. (2007) The fruit fly Drosophila melanogaster favors dim light and times its activity peaks to early dawn and late dusk. J Biol Rhythms 22: 387-399.
146
16. Tahara Y SS (2013) Chronobiology and nutrition. Neuroscience 253: 78-88.
17. Nikhil KL GG, Ratna K, Sharma VK (2014) Role of temperature in mediating morning and evening emergence chronotypes in fruit flies Drosophila melanogaster. J Biol Rhythms 29: 427-441.
18. Bloch G HE, Levine JD, Schwartz WJ (2013) Socially synchronized circadian oscillators. Proc Biol Sci 280: 20130035.
19. Mistiberger RE SD (2004) Social influcences on mammalian circadian rhythms: animal and human studies. Biol Rev Camb Philos Soc 79: 533-556.
20. Konopka RJ, & Benzer, S. (1971). Clock mutants of Drosophila melanogaster. Proceedings of the National Academy of Sciences of the United States of America, 68(9), 2112–2116.
21. Allada R, Chung BY (2010) Circadian organization of behavior and physiology in Drosophila. Annu Rev Physiol 72: 605-624.
22. Darlington TKCtCLC-IToIOIpatS, 280(5369), 1599–1603. doi:10.1126/science.280.5369.1599.
23. Saez L, & Young, M. W. (1996). Regulation of nuclear entry of the Drosophila clock proteins period and timeless. Neuron, 17(5), 911–920.
24. Dubruille R, Emery P (2008) A plastic clock: how circadian rhythms respond to environmental cues in Drosophila. Mol Neurobiol 38: 129-145.
25. Cryan JF, Dinan TG (2012) Mind-altering microorganisms: the impact of the gut microbiota on brain and behaviour. Nat Rev Neurosci 13: 701-712.
26. Kadener S, Stoleru, D., McDonald, M., Nawathean, P., & Rosbash, M. (2007). Clockwork Orange is a transcriptional repressor and a new Drosophila circadian pacemaker component. Genes & Development, 21(13), 1675–1686. doi:10.1101/gad.1552607.
27. Ko HW, Jiang, J., & Edery, I. (2002). Role for Slimb in the degradation of Drosophila Period protein phosphorylated by Doubletime. Nature, 420(6916), 673–678. doi:10.1038/nature01272.
28. Zheng XaAS (2008) Probing the Relative Importance of Molecular Oscillations in the Circadian Clock. Genetics 178: 1147-1155.
29. Ko CH, & Takahashi, J. S. (2006). Molecular components of the mammalian circadian clock. Human Molecular Genetics, 15 Spec No 2(Review Issue 2), R271–7. doi:10.1093/hmg/ddl207.
30. Price JL, Blau, J., Rothenfluh, A., Abodeely, M., Kloss, B., & Young, M. W. (1998). double-time is a novel Drosophila clock gene that regulates PERIOD protein accumulation. Cell, 94(1), 83–95.
31. Martinek S, Inonog, S., Manoukian, A. S., & Young, M. W. (2001). A role for the segment polarity gene shaggy/GSK-3 in the Drosophila circadian clock. Cell, 105(6), 769–779.
32. Sathyanarayanan S, Zheng, X., Xiao, R., & Sehgal, A. (2004). Posttranslational regulation of Drosophila PERIOD protein by protein phosphatase 2A. Cell, 116(4), 603–615.
147
33. Fang Y, Sathyanarayanan, S., & Sehgal, A. (2007). Post-translational regulation of the Drosophila circadian clock requires protein phosphatase 1 (PP1). Genes & Development, 21(12), 1506–1518. doi:10.1101/gad.1541607.
34. Koh KJRtDCCbPL-IDoTS, 312(5781), 1809–1812. doi:10.1126/science.1124951.
35. Eide EJ, Vielhaber, E. L., Hinz, W. A., & Virshup, D. M. (2002). The circadian regulatory proteins BMAL1 and cryptochromes are substrates of casein kinase Iepsilon. Journal of Biological Chemistry, 277(19), 17248–17254. doi:10.1074/jbc.M111466200.
36. Akashi M, Tsuchiya, Y., Yoshino, T., & Nishida, E. (2002). Control of Intracellular Dynamics of Mammalian Period Proteins by Casein Kinase I (CKI ) and CKI in Cultured Cells. Molecular and Cellular Biology, 22(6), 1693–1703. doi:10.1128/MCB.22.6.1693-1703.2002.
37. Gallego M, Kang, H., & Virshup, D. M. (2006). Protein phosphatase 1 regulates the stability of the circadian protein PER2. The Biochemical Journal, 399(1), 169. doi:10.1042/BJ20060678.
38. Siepka SM, Yoo, S.-H., Park, J., Song, W., Kumar, V., Hu, Y., et al. (2007). Circadian Mutant Overtime Reveals F-box Protein FBXL3 Regulation of Cryptochrome and Period Gene Expression. Cell, 129(5), 1011–1023. doi:10.1016/j.cell.2007.04.030.
39. Nitabach MN, Taghert PH (2008) Organization of the Drosophila circadian control circuit. Curr Biol 18: R84-93.
40. Stoleru D, Peng, Y., Agosto, J., & Rosbash, M. (2004). Coupled oscillators control morning and evening locomotor behaviour of Drosophila. Nature, 431(7010), 862–868. doi:10.1038/nature02926.
41. Lear BC, Zhang L, Allada R (2009) The neuropeptide PDF acts directly on evening pacemaker neurons to regulate multiple features of circadian behavior. PLoS Biol 7: e1000154.
42. Yoshii T, Wulbeck C, Sehadova H, Veleri S, Bichler D, et al. (2009) The neuropeptide pigment-dispersing factor adjusts period and phase of Drosophila's clock. J Neurosci 29: 2597-2610.
43. Partch CL, Green, C. B., & Takahashi, J. S. (2014). Molecular architecture of the mammalian circadian clock. Trends in Cell Biology, 24(2), 90–99. doi:10.1016/j.tcb.2013.07.002.
44. Dibner C, Schibler, U., & Albrecht, U. (2010). The Mammalian Circadian Timing System: Organization and Coordination of Central and Peripheral Clocks. Annual Review of Physiology, 72(1), 517–549. doi:10.1146/annurev-physiol-021909-135821.
45. Yang X, Lamia, K. A., & Evans, R. M. (2007). Nuclear receptors, metabolism, and the circadian clock. Cold Spring Harbor Symposia on Quantitative Biology, 72(1), 387–394. doi:10.1101/sqb.2007.72.058.
46. Cailotto C, Lei, J., van der Vliet, J., van Heijningen, C., van Eden, C. G., Kalsbeek, A., et al. (2009) Effects of Nocturnal Light on (Clock) Gene Expression in Peripheral Organs: A Role for the Autonomic Innervation of the Liver. PLoS One 4: e5650.
148
47. Brown SA, & Azzi, A. (2013). Peripheral Circadian Oscillators in Mammals. In Circadian Clocks (Vol. 217, pp. 45–66). Berlin, Heidelberg: Springer Berlin Heidelberg. doi:10.1007/978-3-642-25950-0_3.
48. Buhr ED, Yoo, S.-H., & Takahashi, J. S. (2010). Temperature as a universal resetting cue for mammalian circadian oscillators. Science, 330(6002), 379–385. doi:10.1126/science.1195262.
49. Liu AC, Welsh, D. K., Ko, C. H., Tran, H. G., Zhang, E. E., Priest, A. A., et al. (2007). Intercellular coupling confers robustness against mutations in the SCN circadian clock network. Cell, 129(3), 605–616. doi:10.1016/j.cell.2007.02.047.
50. Stratmann M, & Schibler, U. (2006). Properties, Entrainment, and Physiological Functions of Mammalian Peripheral Oscillators. Journal of Biological Rhythms, 21(6), 494–506. doi:10.1177/0748730406293889.
51. Alvarez JD, Chen, D., Storer, E., & Sehgal, A. (2003) Non-cyclic and developmental stage-specific expression of circadian clock proteins during murine spermatogenesis. Biology of Reproduction 69: 81-91.
52. Schmutz I, Albrecht, U., & Ripperger, J. A. (2012) The role of clock genes and rhythmicity in the liver. . Molecular and Cellular Endocrinology 349: 38-44.
53. Reddy AB, Maywood, E. S., Karp, N. A., King, V. M., Inoue, Y., Gonzalez, F. J., et al. (2007) Glucocorticoid signaling synchronizes the liver circadian transcriptome. Hepatology 45: 1478-1488.
54. Yang X, Downes, M., Yu, R. T., Bookout, A. L., He, W., Straume, M., et al. (2006) Nuclear receptor expression links the circadian clock to metabolism. Cell 126: 801-810.
55. Kornmann B, Schaad, O., Bujard, H., Takahashi, J. S., & Schibler, U. (2007) System-Driven and Oscillator-Dependent Circadian Transcription in Mice with a Conditionally Active Liver Clock. PLoS Biol 5: e34.
56. Schibler U RJ, Brown SA. (2003) Peripheral circadian oscillators in mammals: time and food. Journal of Biological Rhythms 18: 250-260.
57. Stokkan KA, Yamazaki, S., Tei, H., Sakaki, Y., & Menaker, M. (2001). Science, 291(5503), 490–493. (2001) Entrainment of the circadian clock in the liver by feeding. . Science 291: 490-493.
58. Y-J KTaK (2005) Overview of Innate Immunity in Drosophila. Journal of Biochemistry and Molecular Biology 38: 121-127.
59. Hashimoto C, Hudson, K. L., & Anderson, K. V. (1988). Cell, 52(2), 269–279. (1988) The Toll gene of Drosophila, required for dorsal-ventral embryonic polarity, appears to encode a transmembrane protein. Cell 52: 269-279.
60. Lemaitre B, Nicolas, E., Michaut, L., Reichhart, J. M., & Hoffmann, J. A. (1996) The dorsoventral regulatory gene cassette spätzle/Toll/cactus controls the potent antifungal response in Drosophila adults. Cell 86: 973-983.
149
61. Ferrandon D, Imler JL, Hetru C, Hoffmann JA (2007) The Drosophila systemic immune response: sensing and signalling during bacterial and fungal infections. Nat Rev Immunol 7: 862-874.
62. Ligoxygakis P PN, Ji C, Leclerc V, Duvic B, Belvin M, Jiang H, Hoffmann JA, and Reichhart J-M. (2002) A serpin mutant links Toll activation to melanization in the host defense of Drosophila. The EMBO Journa 21: 6330-6337.
63. Stone EF, Fulton BO, Ayres JS, Pham LN, Ziauddin J, et al. (2012) The circadian clock protein timeless regulates phagocytosis of bacteria in Drosophila. PLoS Pathog 8: e1002445.
64. Medzhitov R, Preston-Hurlburt, P., & Janeway, C. A. (1997) A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature 388: 394-397.
65. Poltorak A, He, X., Smirnova, I., Liu, M. Y., Van Huffel, C., Du, X., et al. (1998) Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282: 2085-2088.
66. Hoshino K, Takeuchi, O., Kawai, T., Sanjo, H., Ogawa, T., Takeda, Y., et al. (1999) Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product. . The Journal of Immunology 162: 3749-3752.
67. Jiravanichpaisal P, Lee BL, Soderhall K (2006) Cell-mediated immunity in arthropods: hematopoiesis, coagulation, melanization and opsonization. Immunobiology 211: 213-236.
68. Tang H (2009) Regulation and function of the melanization reaction in Drosophila. Fly 3: 105-111.
69. Binggeli O NC, Poidevin M, Lemaitre B. (2014) Prophenoloxidase activation is required for survival to microbial infections in Drosophila. PLoS Pathog 10: 1-15.
70. Ayres JS, Schneider DS (2008) A signaling protease required for melanization in Drosophila affects resistance and tolerance of infections. PLoS Biol 6: 2764-2773.
71. Stuart LM, Ezekowitz RA (2008) Phagocytosis and comparative innate immunity: learning on the fly. Nat Rev Immunol 8: 131-141.
72. Ulvila J, Vanha-Aho, L. M. & Ramet, M. (2011) Drosophila phagocytosis - still many unknowns under the surface. . APMIS 119: 651-662.
73. Imeri L, & Opp, M. R. (2009) How (and why) the immune system makes us sleep. Nature Reviews Neuroscience 10: 199-210.
74. Krueger JM, Obál, F. J., Fang, J., Kubota, T., & Taishi, P. (2001) The role of cytokines in physiological sleep regulation. Annals of the New York Academy of Sciences 933: 211-221.
75. Dantzer R (2001) Cytokine-Induced Sickness Behavior: Where Do We Stand? Brain Behavior and Immunity 15: 7-24.
150
76. Gemma C, Imeri, L., de Simoni, M. G., & Mancia, M. (1997). The American Journal of Physiology, 272(2 Pt 2), R601–6. (1997) Interleukin-1 induces changes in sleep, brain temperature, and serotonergic metabolism. The American Journal of Physiology 272: R601-606.
77. Hurtado-Alvarado G, Pavón, L., Castillo-García, S. A., Hernández, M. E., Domínguez-Salazar, E., Velázquez-Moctezuma, J., & Gómez-González, B. (2013) Sleep Loss as a Factor to Induce Cellular and Molecular Inflammatory Variations. Clinical and Developmental Immunology 2013: 1-14.
78. Dimitrov S, Lange, T., Nohroudi, K., & Born, J. (2007) Number and function of circulating human antigen presenting cells regulated by sleep. . Sleep 30: 401-411.
79. Born J, Lange, T., Hansen, K., Mölle, M., & Fehm, H. L. (1997). The Journal of Immunology, 158(9), 4454–4464. (1997) Effects of sleep and circadian rhythm on human circulating immune cells. . The Journal of Immunology 158: 4454-4464.
80. Williams JA, Sathyanarayanan, S., Hendricks, J. C., & Sehgal, A. (2007) Interaction between sleep and the immune response in Drosophila: a role for the NFkappaB relish. Sleep 30: 389-400.
81. Shirasu-Hiza MM, Dionne MS, Pham LN, Ayres JS, Schneider DS (2007) Interactions between circadian rhythm and immunity in Drosophila melanogaster. Curr Biol 17: R353-355.
82. Medzhitov R, Schneider DS, Soares MP (2012) Disease tolerance as a defense strategy. Science 335: 936-941.
83. Schneider DS, & Ayres, J. S. (2008) Two ways to survive infection: what resistance and tolerance can teach us about treating infectious diseases. Nature Reviews Immunology 8: 889-895.
84. Brown AE, Baumbach, J., Cook, P.E., and Ligoxygakis, P. (2009). PloS one 4, e4490 (2009) Short-term starvation of immune deficient Drosophila improves survival to gram-negative bacterial infections. PLoS one 4: e4490.
85. Ayres JS, and Schneider, D.S. (2009) The role of anorexia in resistance and tolerance to infections in Drosophila. PLoS Biol 7: e1000150.
86. Dionne MS, Pham LN, Shirasu-Hiza M, Schneider DS (2006) Akt and FOXO dysregulation contribute to infection-induced wasting in Drosophila. Curr Biol 16: 1977-1985.
87. Chambers MC, Song, K.H., and Schneider, D.S. (2012). PloS one 7, e50679. (2012) Listeria monocytogenes infection causes metabolic shifts in Drosophila melanogaster. Plos one 7: 50679.
88. Green CB, Takahashi, J.S., and Bass, J. (2008). The meter of metabolism. Cell 134, 728-742. (2008) The meter of metabolism. Cell 134.
89. Xu K, Zheng, X., and Sehgal, A. (2008). Regulation of feeding and metabolism by neuronal and peripheral clocks in Drosophila. Cell metabolism 8, 289-300. (2008) Regulation of feeding and metabolism by neuronal and peripheral clocks in Drosophila. Cell metabolism 8: 289-300.
151
90. Lee JE, Edery I (2008) Circadian regulation in the ability of Drosophila to combat pathogenic infections. Curr Biol 18: 195-199.
91. Jin P, & Warren, S. T. (2000) Understanding the molecular basis of fragile X syndrome. Human Molecular Genetics 9: 901-908.
92. Hernandez RN, Feinberg RL, Vaurio R, Passanante NM, Thompson RE, et al. (2009) Autism spectrum disorder in fragile X syndrome: a longitudinal evaluation. Am J Med Genet A 149A: 1125-1137.
93. Muhle R, Trentacoste, S. V., & Rapin, I. (2004). The genetics of autism. Pediatrics, 113(5), e472–86. (2004) The genetics of autism. Pediatrics 113: e472-486.
94. Schaefer GB, & Mendelsohn, N. J. (2008) Genetics evaluation for the etiologic diagnosis of autism spectrum disorders. Genetics in Medicine 10: 4-12.
95. Kidd SA, Lachiewicz A, Barbouth D, Blitz RK, Delahunty C, et al. (2014) Fragile X Syndrome: A Review of Associated Medical Problems. Pediatrics 134: 995-1005.
96. Warren ST, & Nelson, D. L. (1994) Advances in molecular analysis of fragile X syndrome. JAMA 271: 536-542.
97. Comery TA, Harris, J. B., Willems, P. J., Oostra, B. A., Irwin, S. A., Weiler, I. J., & Greenough, W. T. (1997) Abnormal dendritic spines in fragile X knockout mice: maturation and pruning deficits. Proceedings of the National Academy of Sciences of the United States of America 94: 5401-5404.
98. Hinton VJ BW, Wisniewski K, Rudelli RD. (1991) Analysis of neocortex in three males with the fragile X syndrome. 41 3.
99. Pietrobono R, Tabolacci, E., Zalfa, F., Zito, I., Terracciano, A., Moscato, U., et al. (2005) Molecular dissection of the events leading to inactivation of the FMR1 gene. Human Molecular Genetics 14: 267-277.
100. Wheeler AC, Bailey, D. B., Jr, Berry-Kravis, E., Greenberg, J., Losh, M., Mailick, M., et al. (2014). , 6(1), 1–14. (2014) Associated features in females with an FMR1 premutation. J Neurodev Disord 6: 1-14.
101. Darnell JC, Fraser CE, Mostovetsky O, Darnell RB (2009) Discrimination of common and unique RNA-binding activities among Fragile X mental retardation protein paralogs. Hum Mol Genet 18: 3164-3177.
102. Martin JP, & Bell, J. (1943) A PEDIGREE OF MENTAL DEFECT SHOWING SEX-LINKAGE. Journal of Neurology and Psychiatry 6: 154-157.
103. Verkerk AJ, Pieretti, M., Sutcliffe, J. S., Fu, Y. H., Kuhl, D. P., Pizzuti, A., et al. (1991) Identification of a gene (FMR-1) containing a CGG repeat coincident with a breakpoint cluster region exhibiting length variation in fragile X syndrome. cell 65: 905-914.
104. Ashley CT, Wilkinson, K. D., Reines, D., & Warren, S. T. (1993) FMR1 protein: conserved RNP family domains and selective RNA binding. Science 262: 563-566.
152
105. Siomi MC, Higashijima K, Ishizuka A, Siomi H (2002) Casein Kinase II Phosphorylates the Fragile X Mental Retardation Protein and Modulates Its Biological Properties. Molecular and Cellular Biology 22: 8438-8447.
106. Siomi H, Siomi, M. C., Nussbaum, R. L., & Dreyfuss, G. (1993) The protein product of the fragile X gene, FMR1, has characteristics of an RNA-binding protein. Cell 74: 291-298.
107. Eberhart DE, Malter, H. E., Feng, Y., & Warren, S. T. (1996) The fragile X mental retardation protein is a ribonucleoprotein containing both nuclear localization and nuclear export signals. Human Molecular Genetics 5: 1083-1091.
108. Khandjian EW, Corbin, F., Woerly, S., & Rousseau, F. (1996) The fragile X mental retardation protein is associated with ribosomes. Nature Genetics 12: 91-93.
109. Darnell JC, Van Driesche SJ, Zhang C, Hung KY, Mele A, et al. (2011) FMRP stalls ribosomal translocation on mRNAs linked to synaptic function and autism. Cell 146: 247-261.
110. McBride SM, Choi CH, Wang Y, Liebelt D, Braunstein E, et al. (2005) Pharmacological rescue of synaptic plasticity, courtship behavior, and mushroom body defects in a Drosophila model of fragile X syndrome. Neuron 45: 753-764.
111. Darnell JC, Klann E (2013) The translation of translational control by FMRP: therapeutic targets for FXS. Nat Neurosci 16: 1530-1536.
112. Zhang Y OCJ, Siomi MC, Srinivasan S, Dutra A, Nussbaum RL, Dreyfuss G. (1995) The fragile X mental retardation syndrome protein interacts with novel homologs FXR1 and FXR2. EMBO J 14: 5358-5366.
113. Coffee RL, Jr., Tessier CR, Woodruff EA, 3rd, Broadie K (2010) Fragile X mental retardation protein has a unique, evolutionarily conserved neuronal function not shared with FXR1P or FXR2P. Dis Model Mech 3: 471-485.
114. Darnell JC, Fraser CE, Mostovetsky O, Stefani G, Jones TA, et al. (2005) Kissing complex RNAs mediate interaction between the Fragile-X mental retardation protein KH2 domain and brain polyribosomes. Genes Dev 19: 903-918.
115. Darnell JC, Jensen, K. B., Jin, P., Brown, V., Warren, S. T., & Darnell, R. B. (2001) Fragile X mental retardation protein targets G quartet mRNAs important for neuronal function. Cell 107: 489-499.
116. Zang JB, Nosyreva ED, Spencer CM, Volk LJ, Musunuru K, et al. (2009) A mouse model of the human Fragile X syndrome I304N mutation. PLoS Genet 5: e1000758.
117. De Boulle K, Verkerk, A. J., Reyniers, E., Vits, L., Hendrickx, J., Van Roy, B., et al. (1993) A point mutation in the FMR-1 gene associated with fragile X mental retardation. Nature Genetics 3: 31-35.
118. Feng Y, Absher, D., Eberhart, D. E., Brown, V., Malter, H. E., & Warren, S. T. (1997) FMRP associates with polyribosomes as an mRNP, and the I304N mutation of severe fragile X syndrome abolishes this association. Molecular Cell 1: 109-118.
153
119. Qin M, Kang, J., Burlin, T. V., Jiang, C., & Smith, C. B. (2005) Postadolescent changes in regional cerebral protein synthesis: an in vivo study in the FMR1 null mouse. The Journal of Neuroscience : the Official Journal of the Society for Neuroscience 25: 5087-5095.
120. Bolduc FV, Bell K, Cox H, Broadie KS, Tully T (2008) Excess protein synthesis in Drosophila fragile X mutants impairs long-term memory. Nat Neurosci 11: 1143-1145.
121. Brown V, Jin, P., Ceman, S., Darnell, J. C., O'Donnell, W. T., Tenenbaum, S. A., Jin, X., Feng, Y., Wilkinson, K. D., Keene, J. D., Darnell, R. B., Warren, S. T. (2001) Microarray identification of FMRP-associated brain mRNAs and altered mRNA translational profiles in fragile X syndrome. Cell 107: 477-487.
122. Zhang YQ, Bailey, A. M., Matthies, H. J., Renden, R. B., Smith, M. A., Speese, S. D., Rubin, G. M., Broadie, K. (2001) Drosophila fragile X-related gene regulates the MAP1B homolog Futsch to control synaptic structure and function1-3. Cell 107: 591-603.
123. Ule J, Jensen, K., Mele, A., & Darnell, R. B. (2005) CLIP: A method for identifying protein–RNA interaction sites in living cells. . Methods 37: 376-386.
124. Irwin SA, Idupulapati M, Gilbert ME, Harris JB, Chakravarti AB, et al. (2002) Dendritic spine and dendritic field characteristics of layer V pyramidal neurons in the visual cortex of fragile-X knockout mice. Am J Med Genet 111: 140-146.
125. Malenka RC, Bear MF (2004) LTP and LTD: an embarrassment of riches. Neuron 44: 5-21.
126. Huber KM, Kayser, M. S., & Bear, M. F. (2000) Role for rapid dendritic protein synthesis in hippocampal mGluR-dependent long-term depression. Science 288: 1254-1257.
127. Li Z, Zhang, Y., Ku, L., Wilkinson, K. D., Warren, S. T., & Feng, Y. (2001) The fragile X mental retardation protein inhibits translation via interacting with mRNA. Nucleic Acids Research 29: 2276-2283.
128. Godfraind JM, Reyniers, E., De Boulle, K., D'Hooge, R., De Deyn, P. P., Bakker, C. E., et al. (1996) Long-term potentiation in the hippocampus of fragile X knockout mice. American Journal of Medical Genetics Part A, 64: 246-251.
129. Huber KM, Gallagher SM, Warren ST, Bear MF (2002) Altered synaptic plasticity in a mouse model of fragile X mental retardation. Proc Natl Acad Sci U S A 99: 7746-7750.
130. Weiler IJ, & Greenough, W. T. (1999) Synaptic synthesis of the Fragile X protein: possible involvement in synapse maturation and elimination. American Journal of Medical Genetics Part A 83: 248-252.
131. Nosyreva ED, & Huber, K. M. (2006). (2006) Metabotropic receptor-dependent long-term depression persists in the absence of protein synthesis in the mouse model of fragile X syndrome. . Journal of Neurophysiology 95: 3291–3295.
132. Bear MF, Huber KM, Warren ST (2004) The mGluR theory of fragile X mental retardation. Trends Neurosci 27: 370-377.
133. Michalon A SM, Ballard TM, Ozmen L, Spooren W, Wettstein JG, Jaeschke G, Bear MF, Lindemann L. Chronic Pharmacological mGlu5 Inhibition Corrects Fragile X in Adult Mice. Neuron 74: 49–56. .
154
134. Paribello C, Tao, L., Folino, A., Berry-Kravis, E., Tranfaglia, M., Ethell, I. M., & Ethell, D. W. (2010) Open-label add-on treatment trial of minocycline in fragile X syndrome. BMC Neurology 10: 91.
135. Kim H-S, & Suh, Y.-H. (2009) Minocycline and neurodegenerative diseases. Behavioural Brain Research 196: 168-179.
136. Bilousova TV, Dansie L, Ngo M, Aye J, Charles JR, et al. (2009) Minocycline promotes dendritic spine maturation and improves behavioural performance in the fragile X mouse model. J Med Genet 46: 94-102.
137. Siller SS, Broadie K (2011) Neural circuit architecture defects in a Drosophila model of Fragile X syndrome are alleviated by minocycline treatment and genetic removal of matrix metalloproteinase. Dis Model Mech 4: 673-685.
138. Pham LN, Dionne MS, Shirasu-Hiza M, Schneider DS (2007) A specific primed immune response in Drosophila is dependent on phagocytes. PLoS Pathog 3: e26.
139. Lowrey PL, Takahashi JS (2004) Mammalian circadian biology: elucidating genome-wide levels of temporal organization. Annu Rev Genomics Hum Genet 5: 407-441.
140. Storch KF, Lipan O, Leykin I, Viswanathan N, Davis FC, et al. (2002) Extensive and divergent circadian gene expression in liver and heart. Nature 417: 78-83.
141. Keller M, Mazuch J, Abraham U, Eom GD, Herzog ED, et al. (2009) A circadian clock in macrophages controls inflammatory immune responses. Proc Natl Acad Sci U S A 106: 21407-21412.
142. Hardin PE (2005) The circadian timekeeping system of Drosophila. Curr Biol 15: R714-722.
143. Beaver LM, Giebultowicz JM (2004) Regulation of copulation duration by period and timeless in Drosophila melanogaster. Curr Biol 14: 1492-1497.
144. Krishnan N, Davis AJ, Giebultowicz JM (2008) Circadian regulation of response to oxidative stress in Drosophila melanogaster. Biochem Biophys Res Commun 374: 299-303.
145. Williams JA, Sathyanarayanan S, Hendricks JC, Sehgal A (2007) Interaction between sleep and the immune response in Drosophila: a role for the NFkappaB relish. Sleep 30: 389-400.
146. Schneider DS, Ayres JS (2008) Two ways to survive infection: what resistance and tolerance can teach us about treating infectious diseases. Nat Rev Immunol 8: 889-895.
147. McDonald MJ, Rosbash M (2001) Microarray analysis and organization of circadian gene expression in Drosophila. Cell 107: 567-578.
148. Ceriani MF, Hogenesch JB, Yanovsky M, Panda S, Straume M, et al. (2002) Genome-wide expression analysis in Drosophila reveals genes controlling circadian behavior. J Neurosci 22: 9305-9319.
149. Mendez-Ferrer S, Lucas D, Battista M, Frenette PS (2008) Haematopoietic stem cell release is regulated by circadian oscillations. Nature 452: 442-447.
155
150. Kondo Y, Cahyaningsih U, Abe A, Tanabe A (1992) Presence of the diurnal rhythms of monocyte count and macrophage activities in chicks. Poult Sci 71: 296-301.
151. Kurepa Z, Rabatic S, Dekaris D (1992) Influence of circadian light-dark alternations on macrophages and lymphocytes of CBA mouse. Chronobiol Int 9: 327-340.
152. Hriscu ML (2005) Modulatory factors of circadian phagocytic activity. Ann N Y Acad Sci 1057: 403-430.
153. Roy B, Singh R, Kumar S, Rai U (2008) Diurnal variation in phagocytic activity of splenic phagocytes in freshwater teleost Channa punctatus: melatonin and its signaling mechanism. J Endocrinol 199: 471-480.
154. Nguyen-Legros J, Hicks D (2000) Renewal of photoreceptor outer segments and their phagocytosis by the retinal pigment epithelium. Int Rev Cytol 196: 245-313.
155. Schneider DS, Ayres JS, Brandt SM, Costa A, Dionne MS, et al. (2007) Drosophila eiger mutants are sensitive to extracellular pathogens. PLoS Pathog 3: e41.
156. Joyce EA, Kawale A, Censini S, Kim CC, Covacci A, et al. (2004) LuxS is required for persistent pneumococcal carriage and expression of virulence and biosynthesis genes. Infect Immun 72: 2964-2975.
157. Mansfield BE, Dionne MS, Schneider DS, Freitag NE (2003) Exploration of host-pathogen interactions using Listeria monocytogenes and Drosophila melanogaster. Cell Microbiol 5: 901-911.
158. Flyg C, Xanthopoulos K (1983) Insect Pathogenic Properties of Serratia marcescens. Passive and Active Resistance to Insect Immunity Studied with Protease-deficient and Phage-resistant Mutants. Journal of General Microbiology 129: 453-464.
159. Hoiseth SK, Stocker BA (1981) Aromatic-dependent Salmonella typhimurium are non-virulent and effective as live vaccines. Nature 291: 238-239.
160. Clark RI, Woodcock KJ, Geissmann F, Trouillet C, Dionne MS (2011) Multiple TGF-beta superfamily signals modulate the adult Drosophila immune response. Curr Biol 21: 1672-1677.
161. Elrod-Erickson M, Mishra S, Schneider D (2000) Interactions between the cellular and humoral immune responses in Drosophila. Curr Biol 10: 781-784.
162. Lee J-E, Edery I (2008) Circadian regulation in the ability of Drosophila to combat pathogenic infections. Curr Biol 18: 195-199.
163. Naidoo N, Song W, Hunter-Ensor M, Sehgal A (1999) A role for the proteasome in the light response of the timeless clock protein. Science 285: 1737-1741.
164. Konopka RJ, Pittendrigh C, Orr D (1989) Reciprocal behaviour associated with altered homeostasis and photosensitivity of Drosophila clock mutants. J Neurogenet 6: 1-10.
165. Peng Y, Stoleru D, Levine JD, Hall JC, Rosbash M (2003) Drosophila free-running rhythms require intercellular communication. PLoS Biol 1: E13.
156
166. Rieger D, Stanewsky R, Helfrich-Förster C (2003) Cryptochrome, compound eyes, Hofbauer-Buchner eyelets, and ocelli play different roles in the entrainment and masking pathway of the locomotor activity rhythm in the fruit fly Drosophila melanogaster. Journal of Biological Rhythms 18: 377-391.
167. Ayres J, Schneider D (2009) The role of anorexia in resistance and tolerance to infections in Drosophila. PLoS Biol 7: e1000150.
168. Lemaitre B, Reichhart JM, Hoffmann JA (1997) Drosophila host defense: differential induction of antimicrobial peptide genes after infection by various classes of microorganisms. Proc Natl Acad Sci U S A 94: 14614-14619.
169. De Gregorio E, Spellman PT, Tzou P, Rubin GM, Lemaitre B (2002) The Toll and Imd pathways are the major regulators of the immune response in Drosophila. Embo J 21: 2568-2579.
170. Ramet M, Pearson A, Manfruelli P, Li X, Koziel H, et al. (2001) Drosophila scavenger receptor CI is a pattern recognition receptor for bacteria. Immunity 15: 1027-1038.
171. Garver LS, Wu J, Wu LP (2006) The peptidoglycan recognition protein PGRP-SC1a is essential for Toll signaling and phagocytosis of Staphylococcus aureus in Drosophila. Proc Natl Acad Sci USA 103: 660-665.
172. Castanon-Cervantes O, Wu M, Ehlen JC, Paul K, Gamble KL, et al. (2010) Dysregulation of inflammatory responses by chronic circadian disruption. J Immunol 185: 5796-5805.
173. Feigin RD, San Joaquin VH, Haymond MW, Wyatt RG (1969) Daily periodicity of susceptibility of mice to pneumococcal infection. Nature 224: 379-380.
174. Liu V, Escobar, G. J., Greene, J. D., Soule, J., Whippy, A., Angus, D. C., & Iwashyna, T. J. (2014) Hospital Deaths in Patients With Sepsis From 2 Independent Cohorts. JAMA 312: 90.
175. Anne Elixhauser BFES (2011) Statistical Brief #122: Septicemia in U.S. Hospitals, 2009,. 1–13.
176. Marik PE (2014) Early Management of Severe Sepsis. . Chest 145: 1407-1418.
177. Lin Y, Han M, Shimada B, Wang L, Gibler TM, et al. (2002) Influence of the period-dependent circadian clock on diurnal, circadian, and aperiodic gene expression in Drosophila melanogaster. Proc Natl Acad Sci U S A 99: 9562-9567.
178. Brennan CA, Anderson KV (2004) Drosophila: the genetics of innate immune recognition and response. Annu Rev Immunol 22: 457-483.
179. Lemaitre B, Nicolas E, Michaut L, Reichhart JM, Hoffmann JA (1996) The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults. Cell 86: 973-983.
180. Repicky S, Broadie K (2009) Metabotropic glutamate receptor-mediated use-dependent down-regulation of synaptic excitability involves the fragile X mental retardation protein. J Neurophysiol 101: 672-687.
157
181. Raberg L, Sim D, Read AF (2007) Disentangling genetic variation for resistance and tolerance to infectious diseases in animals. Science 318: 812-814.
182. Ayres JS, Freitag N, Schneider DS (2008) Identification of Drosophila mutants altering defense of and endurance to Listeria monocytogenes infection. Genetics 178: 1807-1815.
183. Chambers MC, Song KH, Schneider DS (2012) Listeria monocytogenes infection causes metabolic shifts in Drosophila melanogaster. PLoS One 7: e50679.
184. Brown AE, Baumbach J, Cook PE, Ligoxygakis P (2009) Short-term starvation of immune deficient Drosophila improves survival to gram-negative bacterial infections. PLoS One 4: e4490.
185. Ayres JS, Schneider DS (2009) The role of anorexia in resistance and tolerance to infections in Drosophila. PLoS Biol 7: e1000150.
186. Xu K, Zheng X, Sehgal A (2008) Regulation of feeding and metabolism by neuronal and peripheral clocks in Drosophila. Cell Metab 8: 289-300.
187. Green CB, Takahashi JS, Bass J (2008) The meter of metabolism. Cell 134: 728-742.
188. Tapon N, Ito N, Dickson BJ, Treisman JE, Hariharan IK (2001) The Drosophila tuberous sclerosis complex gene homologs restrict cell growth and cell proliferation. Cell 105: 345-355.
189. Hietakangas V, Cohen SM (2007) Re-evaluating AKT regulation: role of TOR complex 2 in tissue growth. Genes & Development 21: 632-637.
190. McGowan MW, Artiss JD, Strandbergh DR, Zak B (1983) A peroxidase-coupled method for the colorimetric determination of serum triglycerides. Clin Chem 29: 538-542.
191. Ja WW, Carvalho GB, Mak EM, de la Rosa NN, Fang AY, et al. (2007) Prandiology of Drosophila and the CAFE assay. Proc Natl Acad Sci U S A 104: 8253-8256.
192. Ja WW, Carvalho GB, Zid BM, Mak EM, Brummel T, et al. (2009) Water- and nutrient-dependent effects of dietary restriction on Drosophila lifespan. Proc Natl Acad Sci U S A 106: 18633-18637.
193. Deshpande SA, Carvalho GB, Amador A, Phillips AM, Hoxha S, et al. (2014) Quantifying Drosophila food intake: comparative analysis of current methodology. Nat Methods 11: 535-540.
194. Emery P (2007) Protein extraction from Drosophila heads. Methods Mol Biol 362: 375-377.
195. Brummel T, Ching A, Seroude L, Simon AF, Benzer S (2004) Drosophila lifespan enhancement by exogenous bacteria. Proceedings of the National Academy of Sciences of the United States of America 101: 12974-12979.
196. Castonguay-Vanier J, Vial L, Tremblay J, Deziel E (2010) Drosophila melanogaster as a model host for the Burkholderia cepacia complex. PLoS One 5: e11467.
197. D'Argenio DA, Gallagher LA, Berg CA, Manoil C (2001) Drosophila as a model host for Pseudomonas aeruginosa infection. J Bacteriol 183: 1466-1471.
158
198. Schwager S, Agnoli K, Kothe M, Feldmann F, Givskov M, et al. (2013) Identification of Burkholderia cenocepacia Strain H111 Virulence Factors Using Nonmammalian Infection Hosts. Infection and Immunity 81: 143-153.
199. Clark RI, Tan SWS, Pean CB, Roostalu U, Vivancos V, et al. (2013) MEF2 Is an In Vivo Immune-Metabolic Switch. Cell 155: 435-447.
200. DiAngelo JR, Bland ML, Bambina S, Cherry S, Birnbaum MJ (2009) The immune response attenuates growth and nutrient storage in Drosophila by reducing insulin signaling. Proceedings of the National Academy of Sciences of the United States of America 106: 20853-20858.
201. Grewal SS (2009) Insulin/TOR signaling in growth and homeostasis: A view from the fly world. International Journal of Biochemistry & Cell Biology 41: 1006-1010.
202. Kockel L, Kerr KS, Melnick M, Bruckner K, Hebrok M, et al. (2010) Dynamic switch of negative feedback regulation in Drosophila Akt-TOR signaling. PLoS Genet 6: e1000990.
203. Staschke KA, Dey S, Zaborske JM, Palam LR, McClintick JN, et al. (2010) Integration of general amino acid control and target of rapamycin (TOR) regulatory pathways in nitrogen assimilation in yeast. J Biol Chem 285: 16893-16911.
204. Loewith R, Jacinto E, Wullschleger S, Lorberg A, Crespo JL, et al. (2002) Two TOR complexes, only one of which is rapamycin sensitive, have distinct roles in cell growth control. Molecular Cell 10: 457-468.
205. Wullschleger S, Loewith R, Hall MN (2006) TOR signaling in growth and metabolism. Cell 124: 471-484.
206. Lee N, Woodrum Cl Fau - Nobil AM, Nobil Am Fau - Rauktys AE, Rauktys Ae Fau - Messina MP, Messina Mp Fau - Dabora SL, et al. Rapamycin weekly maintenance dosing and the potential efficacy of combination sorafenib plus rapamycin but not atorvastatin or doxycycline in tuberous sclerosis preclinical models.
207. Huang J, Manning BD (2008) The TSC1-TSC2 complex: a molecular switchboard controlling cell growth. Biochemical Journal 412: 179-190.
208. Sarbassov DD, Ali SM, Kim DH, Guertin DA, Latek RR, et al. (2004) Rictor, a novel binding partner of mTOR, defines a rapamycin-insensitive and raptor-independent pathway that regulates the cytoskeleton. Current Biology 14: 1296-1302.
209. Xu J, Hopkins K, Sabin L, Yasunaga A, Subramanian H, et al. (2013) ERK signaling couples nutrient status to antiviral defense in the insect gut. Proc Natl Acad Sci U S A 110: 15025-15030.
210. Oh WJ, Wu CC, Kim SJ, Facchinetti V, Julien LA, et al. (2010) mTORC2 can associate with ribosomes to promote cotranslational phosphorylation and stability of nascent Akt polypeptide. Embo Journal 29: 3939-3951.
211. Zinzalla V, Stracka D, Oppliger W, Hall MN (2011) Activation of mTORC2 by Association with the Ribosome. Cell 144: 757-768.
159
212. Tato I, Bartrons R, Ventura F, Rosa JL (2011) Amino Acids Activate Mammalian Target of Rapamycin Complex 2 (mTORC2) via PI3K/Akt Signaling. Journal of Biological Chemistry 286: 6128-6142.
213. Dimitroff B, Howe K, Watson A, Campion B, Lee HG, et al. (2012) Diet and Energy-Sensing Inputs Affect TorC1-Mediated Axon Misrouting but Not TorC2-Directed Synapse Growth in a Drosophila Model of Tuberous Sclerosis. Plos One 7: 16.
214. Brown J, Wang HZ, Suttles J, Graves DT, Martin M (2011) Mammalian Target of Rapamycin Complex 2 (mTORC2) Negatively Regulates Toll-like Receptor 4-mediated Inflammatory Response via FoxO1. Journal of Biological Chemistry 286: 44295-44305.
215. Kootte RS, Vrieze A, Holleman F, Dallinga-Thie GM, Zoetendal EG, et al. (2012) The therapeutic potential of manipulating gut microbiota in obesity and type 2 diabetes mellitus. Diabetes Obesity & Metabolism 14: 112-120.
216. Caesar R, Fak F, Backhed F (2010) Effects of gut microbiota on obesity and atherosclerosis via modulation of inflammation and lipid metabolism. Journal of Internal Medicine 268: 320-328.
217. Lee WJ, Brey PT (2013) How Microbiomes Influence Metazoan Development: Insights from History and Drosophila Modeling of Gut-Microbe Interactions. In: Schekman R, editor. Annual Review of Cell and Developmental Biology, Vol 29. Palo Alto: Annual Reviews. pp. 571-592.
218. Shin SC, Kim SH, You H, Kim B, Kim AC, et al. (2011) Drosophila Microbiome Modulates Host Developmental and Metabolic Homeostasis via Insulin Signaling. Science 334: 670-674.
219. Storelli G, Defaye A, Erkosar B, Hols P, Royet J, et al. (2011) Lactobacillus plantarum Promotes Drosophila Systemic Growth by Modulating Hormonal Signals through TOR-Dependent Nutrient Sensing. Cell Metabolism 14: 403-414.
220. Lipuma JJ (2005) Update on the Burkholderia cepacia complex. Curr Opin Pulm Med 11: 528-533.
221. Kunz AN, Brook I (2010) Emerging resistant Gram-negative aerobic bacilli in hospital-acquired infections. Chemotherapy 56: 492-500.
222. Chang S, Bray SM, Li Z, Zarnescu DC, He C, et al. (2008) Identification of small molecules rescuing fragile X syndrome phenotypes in Drosophila. Nat Chem Biol 4: 256-263.
223. Developmental Disabilities Monitoring Network Surveillance Year 2010 Principal Investigators CfDCaPC (2014) Prevalence of autism spectrum disorder among children aged 8 years - autism and developmental disabilities monitoring network, 11 sites, United States, 2010. Morbidity and Mortality Weekly Report Surveillance Summaries (Washington, DC : 2002) 63: 1-21.
224. Wills S, Cabanlit, M., Bennett, J., Ashwood, P., Amaral, D., & van de Water, J. (2007) Autoantibodies in Autism Spectrum Disorders (ASD). Annals of the New York Academy of Sciences 1107: 79-91.
160
225. Mostafa GA, Shehab, A. A., & Al-Ayadhi, L. Y. (2013) The link between some alleles on human leukocyte antigen system and autism in children. Journal of Neuroimmunology 255: 70-74.
226. Ashwood P, Nguyen DV, Hessl D, Hagerman RJ, Tassone F (2010) Plasma cytokine profiles in Fragile X subjects: is there a role for cytokines in the pathogenesis? Brain Behav Immun 24: 898-902.
227. Pacey LK, Doering LC (2007) Developmental expression of FMRP in the astrocyte lineage: implications for fragile X syndrome. Glia 55: 1601-1609.
228. Saffary R, Xie Z (2011) FMRP regulates the transition from radial glial cells to intermediate progenitor cells during neocortical development. J Neurosci 31: 1427-1439.
229. Yuskaitis CJ, Beurel E, Jope RS (2010) Evidence of reactive astrocytes but not peripheral immune system activation in a mouse model of Fragile X syndrome. Biochim Biophys Acta 1802: 1006-1012.
230. Higashimori H, Morel L, Huth J, Lindemann L, Dulla C, et al. (2013) Astroglial FMRP-dependent translational down-regulation of mGluR5 underlies glutamate transporter GLT1 dysregulation in the fragile X mouse. Hum Mol Genet 22: 2041-2054.
231. Pacey LK, Xuan IC, Guan S, Sussman D, Henkelman RM, et al. (2013) Delayed myelination in a mouse model of fragile X syndrome. Hum Mol Genet 22: 3920-3930.
232. Logan MA, Hackett R, Doherty J, Sheehan A, Speese SD, et al. (2012) Negative regulation of glial engulfment activity by Draper terminates glial responses to axon injury. Nat Neurosci 15: 722-730.
233. Chung WS, Barres BA (2012) The role of glial cells in synapse elimination. Curr Opin Neurobiol 22: 438-445.
234. Blank T, Prinz M (2012) Microglia as modulators of cognition and neuropsychiatric disorders. Glia.
235. Tremblay ME, Stevens, B., Sierra, A., Wake, H., Bessis, A., Nimmerjahn, A. (2011) The role of microglia in the healthy brain. J Neurosci 31: 16064-16069.
236. Derecki NC, Cronk, J. C., Lu, Z., Xu, E., Abbott, S. B. G., Guyenet, P. G., & Kipnis, J. (2012). Nature, 483(7392), 105–109. doi:10.1038/nature10907 (2012) Wild-type microglia arrest pathology in a mouse model of Rett syndrome. Nature 483: 105-109.
237. Gao FB (2002) Understanding fragile X syndrome: insights from retarded flies. . Neuron 34: 859-862.
238. Morales J, Hiesinger, P. R., Schroeder, A. J., Kume, K., Verstreken, P., Jackson, F. R., Nelson, D. L., Hassan, B. A. (2002) Drosophila fragile X protein, DFXR, regulates neuronal morphology and function in the brain. Neuron 34: 961-972.
239. Dockendorff TC, Su, H. S., McBride, S. M. J., Yang, Z., Choi, C. H., Siwicki, K. K., Sehgal A., Jongens T. A. (2002). Drosophila lacking dfmr1 activity show defects in circadian output and fail to maintain courtship interest. Neuron, 34(6), 973–984. (2002) Drosophila lacking dfmr1 activity show defects in circadian output and fail to maintain courtship interest. Neuron 34: 973-984.
161
240. Ziegenfuss JS, Biswas R, Avery MA, Hong K, Sheehan AE, et al. (2008) Draper-dependent glial phagocytic activity is mediated by Src and Syk family kinase signalling. Nature 453: 935-939.
241. Ziegenfuss JS, Doherty J, Freeman MR (2012) Distinct molecular pathways mediate glial activation and engulfment of axonal debris after axotomy. Nat Neurosci 15: 979-987.
242. MacDonald JM, Beach MG, Porpiglia E, Sheehan AE, Watts RJ, et al. (2006) The Drosophila cell corpse engulfment receptor Draper mediates glial clearance of severed axons. Neuron 50: 869-881.
243. Lau GW, Goumnerov BC, Walendziewicz CL, Hewitson J, Xiao W, et al. (2003) The Drosophila melanogaster Toll Pathway Participates in Resistance to Infection by the Gram-Negative Human Pathogen Pseudomonas aeruginosa. Infection and Immunity 71: 4059-4066.
244. Kurant E (2011) Keeping the CNS clear: glial phagocytic functions in Drosophila. Glia 59: 1304-1311.
245. Awasaki T, Ito K (2004) Engulfing action of glial cells is required for programmed axon pruning during Drosophila metamorphosis. Curr Biol 14: 668-677.
246. Hashimoto Y TY, Sakurai K, Kutsuna M, Kurokawa K, Awasaki T, Sekimizu K, Nakanishi Y, Shiratsuchi A. (2009) Identification of lipoteichoic acid as a ligand for draper in the phagocytosis of Staphylococcus aureus by Drosophila hemocytes. J Immunol 183: 7451-7460.
247. Gallagher A, Hallahan B (2012) Fragile X-associated disorders: a clinical overview. J Neurol 259: 401-413.
248. Adams JB, Johansen LJ, Powell LD, Quig D, Rubin RA (2011) Gastrointestinal flora and gastrointestinal status in children with autism--comparisons to typical children and correlation with autism severity. BMC Gastroenterol 11: 22.
249. Utari A, Adams E, Berry-Kravis E, Chavez A, Scaggs F, et al. (2010) Aging in fragile X syndrome. J Neurodev Disord 2: 70-76.
250. Molloy CA, Morrow AL, Meinzen-Derr J, Schleifer K, Dienger K, et al. (2006) Elevated cytokine levels in children with autism spectrum disorder. J Neuroimmunol 172: 198-205.
251. Chonchaiya W, Tassone F, Ashwood P, Hessl D, Schneider A, et al. (2010) Autoimmune disease in mothers with the FMR1 premutation is associated with seizures in their children with fragile X syndrome. Hum Genet 128: 539-548.
252. Tasdemir-Yilmaz OE, Freeman MR (2014) Astrocytes engage unique molecular programs to engulf pruned neuronal debris from distinct subsets of neurons. Genes Dev 28: 20-33.
253. Doherty J, Logan MA, Tasdemir OE, Freeman MR (2009) Ensheathing glia function as phagocytes in the adult Drosophila brain. J Neurosci 29: 4768-4781.
254. Malkova NV, Yu CZ, Hsiao EY, Moore MJ, Patterson PH (2012) Maternal immune activation yields offspring displaying mouse versions of the three core symptoms of autism. Brain Behav Immun 26: 607-616.
162
255. Baharnoori M, Bhardwaj SK, Srivastava LK (2012) Neonatal behavioral changes in rats with gestational exposure to lipopolysaccharide: a prenatal infection model for developmental neuropsychiatric disorders. Schizophr Bull 38: 444-456.
256. Bauman MD, Iosif AM, Smith SE, Bregere C, Amaral DG, et al. (2014) Activation of the maternal immune system during pregnancy alters behavioral development of rhesus monkey offspring. Biol Psychiatry 75: 332-341.
257. Hsiao EY, McBride SW, Hsien S, Sharon G, Hyde ER, et al. (2013) Microbiota modulate behavioral and physiological abnormalities associated with neurodevelopmental disorders. Cell 155: 1451-1463.
258. Gatto CL, Broadie K (2009) Temporal requirements of the fragile x mental retardation protein in modulating circadian clock circuit synaptic architecture. Front Neural Circuits 3: 8.
259. Michel CI, Kraft R, Restifo LL (2004) Defective neuronal development in the mushroom bodies of Drosophila fragile X mental retardation 1 mutants. J Neurosci 24: 5798-5809.
260. Majde JA KJ (2005) Links between the innate immune system and sleep. J Allergy Clin Immunol 116: 1188-1198.
261. Shin SC, Kim, S.H., You, H., Kim, B., Kim, A.C., Lee, K.A., Yoon, J.H., Ryu, J.H., and Lee, W.J. (2011) Drosophila Microbiome Modulates Host Developmental and Metabolic Homeostasis via Insulin Signaling. . Science 334: 670-674.
262. Lee WJ, and Brey, P.T. (2013). How Microbiomes Influence Metazoan Development: Insights from History and Drosophila Modeling of Gut-Microbe Interactions. In Annual Review of Cell and Developmental Biology, Vol 29, Volume 29, R. Schekman, ed. (Palo Alto: Annual Reviews), pp. 571-592. (2013) How Microbiomes Influence Metazoan Development: Insights from History and Drosophila Modeling of Gut-Microbe Interactions.; R. Schekman e, editor: Palo Alto: Annual Reviews. 571-592 p.
263. Storelli G, Defaye, A., Erkosar, B., Hols, P., Royet, J., and Leulier, F. (2011). Lactobacillus plantarum Promotes Drosophila Systemic Growth by Modulating Hormonal Signals through TOR-Dependent Nutrient Sensing. Cell metabolism 14, 403-414. (2011) Lactobacillus plantarum Promotes Drosophila Systemic Growth by Modulating Hormonal Signals through TOR-Dependent Nutrient Sensing. Cell metabolism 14: 403-414.
264. Ashwood P, Wills S, Van de Water J (2006) The immune response in autism: a new frontier for autism research. J Leukoc Biol 80: 1-15.
265. Chung WS, Clarke LE, Wang GX, Stafford BK, Sher A, et al. (2013) Astrocytes mediate synapse elimination through MEGF10 and MERTK pathways. Nature 504: 394-400.
266. Corty MM, Freeman MR (2013) Cell biology in neuroscience: Architects in neural circuit design: glia control neuron numbers and connectivity. J Cell Biol 203: 395-405.
267. Pardo CA BA, Thurm A, Lee LC, Azhagiri A, Neville DM, Swedo SE. (2013) A pilot open-label trial of minocycline in patients with autism and regressive features. J Neurodev Disord 5: 9.
163
APPENDIX I
Fig. 1. Per mutants are more tolerant of infection with P. aeruginosa, and starvation eliminates this survival advantage. (A) Here is a Kaplan-Meier survival curve of wild-type and Per mutants that were infected with P. aeruginosa at an OD of 1.0. Per mutants survived significantly longer than wild-type controls (p > 0.0001). (B) Per mutants have equivalent bacterial loads compared to wild-type controls 3 and 6 hours after infection with P. aeruginosa. These results demonstrate that Per mutants are more tolerant of infection with P. aeruginosa. (C) Under starvation conditions, Per mutants lose their survival advantage after infection with P. aeruginosa (n.s., p = 0.1575). (D) Under starvation conditions, Per mutants still have equivalent bacterial loads compared to wid-type controls. These data suggest that the tolerance phenotype that Per mutants exhibit is a result of feeding or metabolic state.
5 6 7 8 9
0
25
50
75
100
Time (hours)
Perc
ent surv
ival
p = 0.1575
n.s.
6 7 8 9
0
25
50
75
100
Time (hours)
Perc
ent surv
ival
p < 0.0001
WT fed
Per fed
AAA
B
102
103
104
105
106
Bacte
rial lo
ad p
er
fly
0 3 6
Hours post infection
n.s. n.s. n.s.
B
103
104
105
106
107
Bacte
rial lo
ad p
er
fly
n.s. n.s. n.s.
0 3 6
Hours post infection
DC
WT starved
Per starved
164
APPENDIX II
Phagocytosis is upregulated in dFMR1 hypomorphs
Elizabeth Stone and Mimi Shirasu-Hiza
Data and observations unpublished.
165
ABSTRACT
Loss of function of FMR1 causes Fragile X syndrome, the most common cause of
intellectual disability and autism. Autism and Fragile X syndrome are associated with a
loss of circadian rhythms, or physiological oscillations with a period of approximately 24
hours, and altered immune system function; however, the molecular mechanisms driving
these symptoms are poorly understood. Using the established Drosophila model of
Fragile X syndrome, we can examine immune system function. We found that a dFRM1
hypomorph, the Drosophila homolog of FMR1, has increased phagocytic activity by
immune blood cells. Surprisingly, the phagocytic phenotype in dFMR1 hypomorphs is
the opposite of the phenotype observed in dFMR1 null flies, which exhibit reduced
phagocytic activity by immune blood cells. Previously, we showed that phagocytosis by
macrophages is circadian-regulated [44]. We report here that dFMR1 hypomorph mutants
have lost circadian regulation of phagocytosis by macrophages. These results are the first
demonstration that loss of circadian regulation in the context a neurological disease leads
to immune system dysfunction. We anticipate that our research will introduce a new
model of Fragile X syndrome with possible future implications for therapeutic
approaches to this and other neurological diseases.
166
Materials and Methods
Fly and bacterial strains
A transheterozygous Drosophila dFMR1 mutant was used in these experiments,
made from crossing the hypomorph dFMR1B55 mutant to the null dFMR1Δ50M. Infections
were performed with the following bacteria as described in Chapters 2, 3, and 4 [44].
Fly rearing conditions
Flies were raised as described in Chapter 4.
Injections
Flies were anesthetized on CO2 pads. Injections were performed using a custom
Tritech microinjector (Tritech) and pulled glass capillary needles. 50 nL of liquid were
injected into each fly, and the volume of each injection was calibrated by measuring the
diameter of the expelled drop under oil.
Survival assays
As described previously in Chapters 2, 3, and 4 [40], 60-85 flies per genotype per
condition were assayed for each survival curve and placed in 3 vials of either low
glutamate food or standard dextrose food with approximately 20 flies each. In each
experiment, 20-40 flies of each line were also injected with media as a wounding control.
Death was recorded daily. Data was converted to Kaplan-Meier format using custom
Excel-based software called Count the Dead. Survival curves are plotted as Kaplan-
Meier plots and statistical significance is tested using log-rank analysis using GraphPad
Prism software. All experiments were performed at least three times and yielded similar
results.
167
Bacterial load quantitation
Following microbial challenge, six individual flies were collected at each time
point. These flies were homogenized, diluted serially and plated on appropriate media
(tryptic soy blood agar for S.pneumoniae, LB agar for all others). Statistical significance
was determined using non-parametric Mann-Whitney tests. All experiments were
performed at least three times and yielded similar results.
Antimicrobial peptide gene expression
Flies were injected with 50 nL of bacterial culture at OD600 0.10 (M. luteus or E.
coli), or media alone (LB, Difco). Following injection, flies were placed in vials
containing dextrose food and incubated at 25˚C for 6, 24, and 48 hours. Three groups of
six flies after each time point were homogenized in Trizol and stored at -80˚C until
processed. RNA was isolated using a standard Trizol preparation and the samples treated
with DNAse (Invitrogen). Fermentas RevertAid First Strand cDNA synthesis kit was
used to produce the cDNA following the manufacturer’s protocol. Quantitative RT-PCR
was performed using a Stratagene Mx3000 qPCR machine, with Roche FastStart
Universal SYBR Green Master (Rox) mix and the following primer sets: Drosomycin,
Diptericin, and Rpl1. Total mRNA concentration was normalized using Rpl1
expression. Differences in the infection-induced gene expression were calculated by
normalizing to basal gene expression in uninfected wild-type sample taken at the same
time point; p-values were obtained by Mann-Whitney test.
Phagocytosis assay
168
5-7 day old male flies were injected with 50 nL of 20 mg/ml pHrodo-labeled S.
aureus in PBS (Molecular Probes, cat# A10010). The flies were allowed to phagocytose
the particles for 35-45 minutes. The wings of the flies were removed and flies were
pinned with a minutien pin onto a silicon pad. Fluorescence images were taken of the
dorsal surface using epifluorescent illumination with a Nikon Eclipse E800 microscope
fitted with an Photometrics Cool Snaps HQ2 camera. Images were captured with Nikon
Elements software. The exposure was set such that the brightest images had a very small
number of saturated pixels. The experiment was repeated three times with 8-14 flies for
each treatment.
Bead inhibition of phagocytosis
Fluorescent 0.2 um polystyrene beads (Molecular Probes, cat# F13080) were
injected into hemolymph to inhibit phagocytosis as previously described [138]. Briefly,
200 ul of beads in solution were washed three times in sterile water and resuspended in
20 ul final volume. 5-7 day old male flies were injected with 50 nL of bead solution. To
confirm that phagocytosis was inhibited with this protocol, the in vivo phagocytosis assay
was performed as described above. Phagocytosis was completely inhibited within 2 days
of bead injection. Flies were then injected with S. pneumoniae at OD600 ranging from 0.1
(see Survival Assays, above).
169
Results
dFMR1 hypomorph mutants are more resistant to infections with specific pathogens
Drosophila use specific immune mechanisms to defend against different bacterial
pathogens. By infecting dFMR1 hypmorph mutants with different pathogens, I can assess
whether the loss of dFMR1 causes specific defects in immunity. I first compared the
survival kinetics of dFMR1 mutants and wild-type isogenic controls after infection with
four different pathogens known to elicit immune system activation. I found that dFMR1
mutants survived significantly longer than wild-type controls after infection with S.
pneumoniae (Fig 1A, p < 0.0001), S. marcescens (Fig 1B, p < 0.0001), and L.
monocytogenes (Fig 1C, p < 0.0001). Survival rates of dFMR1 mutants and wild-type
controls did not differ after infection with P. aeruginosa (Fig 1C, p = 0.6898). These
results show that dFMR1 mutants survive longer than wild-type flies after particular
infections.
Figure 1. dFMR1 mutants survive longer than wild-type controls after infection with certain pathogens. The graphs shown here are Kaplan-Meier survival curves of wild-type (black squares) and dFMR1 mutant (dFMR1Δ50M/dFMR1B55, green triangles) flies. dFMR1 mutants survive significantly longer than wild-type flies after infection with (A) S. pneumonia (p < 0.0001) or (B) S. marcenscens (p < 0.0001). dFMR1 mutants have identical survival kinetics when compared to wild-type controls after infection with P. aeruginosa (C, p = 0.6898). dFMR1 mutants are more resistant to L. monocytogenes compared to wild-type controls (D, p < 0.0001). p-values are determined using log-rank analysis.
0 2 4 6 8 10 12
0
25
50
75
100
Time (days)
L. monocytogenes:
dFMR1 is resistant
D
p < 0.0001***
0 2 4 6 8 100
25
50
75
100
Time (days)
S. marcenscens:
dFMR1 is resistant
B
p < 0.0001***
S. pneumoniae:
dFMR1 is resistant
Infection with S. pneumoniae:
dFMR1 mutants are more resistant
0 5 10 150
25
50
75
100
Time (days)
Pe
rce
nt su
rviv
al
A
p < 0.0001***
7.0 7.5 8.0 8.5 9.0
0
25
50
75
100
Time (hours)
P. aeruginosa:
dFMR1 is like wild-type
C
p = 0.6898n.s.
n.s. not significant
*** p < 0.0001dFMR1 mutant
Wild-type
170
To determine if dFMR1 mutants were more resistant, meaning they were able to
kill and clear bacteria more efficiently, or if dFMR1 mutants were more tolerant, namely
if they were better able to withstand the pathogenic effects of infection, I next examined
bacterial loads in dFMR1 mutants and wild-type controls at different time points after
infection. I found that dFMR1 mutants have fewer bacterial colonies per fly compared to
wild-type controls at specific time points after infection with S. pneumoniae (Fig 2A) and
S. marcenscens (Fig 2B), indicating dFMR1 mutants were more resistant to these two
bacteria following infection when compared to wild-type controls. dFMR1 mutant flies
had similar bacterial loads after infection with L. monocytogenes (Fig 2C), dFMR1
mutants and wild-type controls had similar bacterial loads after P. aeruginosa infection
(Fig 2C). As the fly uses different immune mechanisms to fight these individual
pathogens and dFMR1 mutants have different survival kinetics compared to wild-type
flies depending on the pathogens, these data suggest that perhaps only specific immune
mechanisms are regulated by dFMR1.
Figure 2. dFMR1 mutants are more resistant to infection with S. pneumoniae and S. marcenscens compared to wild-type controls. After infection, dFMR1 mutants (green triangles) are better able to kill and clear (A) S. pneumonia and (B) S. marcenscens when compared to wild-type controls (black squares). dFMR1 mutants have similar bacterial loads after infection with (C) P. aeruginosa and (D) L. monocytogenes, regardless of nutritional state (data not shown). p-values were determined using Mann-Whitney non-parametric test.
n.s. not significant * p < 0.05 ** p < 0.005dFMR1 mutant
Wild-type
L. monocytogenes:dFMR1 mutants are like wild-type
n.s.
102103104105106107
n.s. n.s.
0 1 2Days post infection
DC
n.s. n.s. n.s.
103104105106107108
0 3 6Hours post infection
P. aeruginosa:dFMR1 mutants are like wild-type
S. pneumoniae:dFMR1 is resistant
100101102103104105106
Col
ony-
Form
ing
Uni
ts
per f
ly (C
FUs)
Hours post infection0 5.5 20
** *
A
n.s.
0 84
S. marcescens:dFMR1 is resistant
103
104
105
106 ** *
B
n.s.
Hours post infection
171
Antimicrobial peptide induction is not regulated by dFMR1
I next tested whether the loss of dFMR1 leads to changes in the antimicrobial peptide
response, the most well characterized branch of the Drosophila innate immune system.
Antimicrobial peptides (AMPs) are small peptides that have bacteriocidal properties and
are produced by the Drosophila fat body in response to bacterial and fungal infections.
The well conserved Toll and imd signaling pathways culminate in AMP induction and are
mainly induced by Gram positive and Gram negative bacterial infections, respectively. In
order to determine if dFMR1 mutants have altered AMP induction, we examined mRNA
levels of two different AMPs, Drosomycin (Drs) and Diptericin (Dpt), which are induced
after activation of the toll and imd signaling pathways, respectively. Although AMP
induction is known to occur following P. aeruginosa infection [222], the rapid time
course of this infection is not suitable for assaying long-term AMP induction. Instead, I
examined basal levels of Drs and Dpt in uninfected flies and in flies injected with M.
luteus and E. coli, used most often to induce the toll and imd pathways, respectively.
There were no significant differences in Drs and Dpt expression either basally or at 6, 24,
and 48 hours after infection in dFMR1 mutants and wild-type controls (Fig 3). These
results suggest that loss of dFMR1 does not alter AMP response in Drosophila.
dFMR1 hypomorph mutants exhibit increased phagocytosis of bacteria by
hemocytes
I measured phagocytosis in dFMR1 mutants and wild-type controls. Drosophila
hemocytes use phagocytosis to combat particular infections, including S. pneumoniae, S.
marcescens, and S. aureus. To determine if dFMR1 regulates phagocytosis, I measured
172
Figure 3. Antimicrobial peptide synthesis is not regulated by dFMR1 Here are representative graphs of the relative expression of (A) Drosomycin and (B) Diptericin assayed by qRT-PCR. dFMR1 hypomorph mutants were found in three trials not to have a statistically-significant difference at all time points assayed after infection in either (A) Drosomycin (6 hours: p = 1.0000; 24 hours: p = 0.1000; 48 hours: p = 0.4000) or (B) Diptericin (6 hours: p = 1.0000; 24 hours: p = 0.2000; 48 hours: p = 0.2000) compared to wild-type controls.
phagocytosis of fluorescently-labeled bacteria by hemocytes in live dFMR1 mutants and
wild-type controls. Briefly, to assay phagocytosis, I injected dFMR1 mutants and wild-
type controls with FITC-labeled, dead S. aureus. I allowed the hemocytes to phagocytose
the dead bacteria, and then we quenched the extracellular fluorescence; thus, the only
fluorescence observed would be due to intracellular, phagocytosed bacteria. I imaged the
dorsal surface of the flies, and fluorescent puncta, representing phagocytosed bacteria,
were visible through the cuticle (Fig 4A). I found an increased number of fluorescent
puncta and an increase in the area of fluorescence in dFMR1 mutants when compared to
wild-type controls (Fig 4B-C). These results show that the loss of dFMR1 causes an
increase in phagocytosis by hemocytes, suggesting that the presence of dFMR1 inhibits
phagocytosis.
Time course of relative Drosomycin expression
0
200
400
600
800
1000
6 hr 24 hr 48 hrGenotype/time point
Rel
ativ
e Ex
pres
sion
WTdFMR1
Time course of relative Diptericin expression
0
2000
4000
6000
8000
6 hr 24 hr 48 hrGenotype/time point
Rel
ativ
e Ex
pres
sion
A B
dFMR1 mutantWild-type
173
Figure 4. dFMR1 mutants exhibit increased phagocytosis of bacteria by hemocytes. (A) dFMR1 mutants and wild-type controls were injected with fluorescein-labeled, dead S. aureus, and, after an incubation during which hemocytes phagocytosed the fluorescent bacteria, extracellular fluorescence was quenched. Hemocytes with intracellular bacteria become fluorescent. The dashed line illustrates the dorsal surface of the flies imaged in the adjacent micrographs. (B) Shown here are representative images of phagocytosis in wild-type controls and dFMR1 mutants. The insets are enlarged regions represented by dashed lines. (C) Phagocytosis in wild-type and dFMR1 mutant flies was quantitated by counting the number of fluorescent puncta. dFMR1 mutant flies exhibit significantly more phagocytosis when compared to wild-type controls (p = 0.0047, obtained by unpaired, two-tailed t-test).
dFMR1 mutants have lost circadian regulation of phagocytosis.
One of the most striking phenotypes of dFMR1 mutants is loss of circadian
regulation, a phenotype also seen in human patients with Fragile X syndrome. I showed
in a previous publication that phagocytosis is circadian-regulated [40]. When I found that
dFMR1 mutants have altered phagocytic activity by their macrophages, I set out to test
whether this is because they have lost circadian regulation of phagocytosis. I tested this
hypothesis by performing phagocytosis assays as described above at different times of
day and night (Fig 5). As shown previously, wild-type flies exhibited less phagocytosis
during the day and more at night. In contrast, dFMR1 mutants exhibited similarly high
levels of phagocytosis during the day and night. This result suggests that dFMR1
mutants have macrophages stuck in the “night state”—that is, rather than oscillating
between low and high activity, they have constant high phagocytic activity. Thus dFMR1
appears to inhibit phagocytosis by macrophages.
174
Figure 5. dFMR1 mutants do no exhibit circadian changes in phagocytosis. Shown here are analyses of phagocytic activity. For each genotype, two groups were entrained to antiphase circadian rhythms before being assayed. Thus day and night conditions could be assayed simultaneously. (A) Wild-type exhibited circadian differences in phagocytosis, with upregulated activity at night as previously published. (B) In contrast, dFMR1 mutants did not exhibit these circadian differences.
Inhibiting phagocytosis prior to infection with S. pneumoniae abolishes the survival
benefit observed in dFMR1 mutants
I next asked whether the increase in phagocytosis observed in dFMR1 mutants
was responsible for their increase in survival time after S. pneumoniae infection. To
address this question, we inhibited phagocytosis in vivo and then assayed the survival
kinetics in dFMR1 mutants and wild-type controls. Briefly, we inhibited phagocytosis by
injecting flies with 0.2 mm polystyrene beads; hemocytes engulf these beads and
subsequently are unable to phagocytose bacteria 3 days post bead injection (Fig 6A).
Bead-inhibited flies were then infected with S. pneumoniae, and their survival rates were
compared to flies that were not bead-inhibited. Consistent with previous results showing
that bead-inhibiting flies decreases survival time after infection with S. pneumoniae [40],
we found that bead-inhibited wild-type and dFMR1 mutant flies died significantly more
quickly than uninhibited wild-type and dFMR1 mutant flies. We also found that bead-
inhibited dFMR1 mutants and wild-type controls had identical survival kinetics after
day night0
20
40
60
80 n.s.
dFMR1
# flu
ores
cent
pun
cta
day night0
20
40
60
80 *
WT
# flu
ores
cent
pun
cta
A B
175
infection with S. pneumoniae (Fig 6B). That is, bead inhibition caused dFMR1 mutants
to lose their survival advantage after S. pneumoniae infection. This result suggests that
the increased phagocytosis observed in dFMR1 mutants is responsible for the increase in
survival time after S. pneumoniae infection when compared to wild-type controls.
Figure 6. Inhibiting phagocytosis abolishes the survival benefit observed in dFMR1 mutants with S. pneumonia infection. (A) Phagocytosis was inhibited by injecting wild-type and dFMR1 mutants flies with polystyrene beads prior to infection with S. pneumoniae (“+ beads”); wild-type and dFMR1 mutant controls were injected with an equal volume of PBS (“no beads”). (B) The graph shown here is a Kaplan-Meier survival curve comparing wild-type and dFMR1 mutant flies with and without inhibited phagocytes. Uninhibited dFMR1 mutants survived significantly longer than wild-type uninhibited flies (p = 0.0144). Bead-inhibited wild-type flies were more sensitive to infection with S. pneumoniae when compared to uninhibited wild-type controls (p < 0.0001), and bead-inhibited dFMR1 mutants were also more sensitive when compared to dFMR1 uninhibited controls. However, bead-inhibited dFMR1 mutant flies had similar survival kinetics as bead-inhibited wild-type flies (p = 0.1779). p-values are determined using log-rank analysis.
Together, these results demonstrate that reduced amounts of dFMR1 improves
phagocytic activity compared to wild-type flies while the complete absence of dFMR1
worsens phagocytic activity compared to wild-type controls, suggesting that there is a
dosage effect of the amount of dFMR1 present in Drosophila for proper immune cell
function.
176
