Activity-regulated retinoic acid

signaling in olfactory sensory neurons

Hande Login

Department of Molecular Biology

Umeå 2014

Responsible publisher under swedish law: the Dean of the Medical Faculty

This work is protected by the Swedish Copyright Legislation (Act 1960:729)

ISBN: 978-91-7601-064-8

ISSN: 0346-6612

Cover photo: A picture of double OMP (red) and Cyp26B1 (green) immunostaining

of olfactory epithelium. All cells are visualized by nuclear staining (blue). Taken by

Hande Login.

Electronic version is available at http://umu.diva-portal.org/

Printed by: Print & Media Umeå, SWEDEN 2014

To my mother, my father and my husband who are the most inspiring people in my world.

“I hear I forget. I see I remember.

I do I understand.”

/Confucius

i

TABLE OF CONTENTS

TABLE OF CONTENTS i

ABSTRACT ii

ABBREVIATIONS iv

PAPERS IN THIS THESIS vi

INTRODUCTION 1

OLFACTION 1

MOUSE MAIN OLFACTORY SYSTEM 2

The Olfactory Epithelium 3

The Odorant Receptors 5

Olfactory Signal Transduction Pathway 8

Zonal Organization of Olfactory Epithelium 9

Olfactory Bulb 12

Formation of Glomerular Map 13

RETINOIC ACID 18

Retinoic Acid Metabolism 18

Function of Retinoic Acid 19

APP PROCESSING 25

Functions of APP Processing 26

AIMS 28

RESULTS and DISCUSSIONS 29

Overview 29

The effect of RAR signaling on homogeneous glomeruli formation

and maintenance. 29

RAR-mediated survival of OSNs 31

Neuronal activity- regulated RA metabolism 32

The link between neuronal activity-regulated Cyp26B1 and BACE1 33

The importance of zonal gradient of RA bioavailability for OR

expression patterns. 35

Cyp26B1 gain of function results in increased OSN turnover in the VL

OE. 36

Cyp26B1 overexpression inhibits the OSN differentiation 37

Odor-induced Cyp26B1 expression mediates neuronal competition

for cell survival 37

Differences between OMP-dnRAR and OMP-Cyp mice 39

CONCLUSIONS 40

ACKNOWLEDGEMENTS 41

REFERENCES 43

ii

ABSTRACT

The aim of the studies included in the thesis is to better understand the

interplay between neuronal activity-dependent gene regulation and the

bioactive vitamin A metabolite all-trans-retinoic acid (RA) during postnatal

development, refinement and maintenance of precise neuronal connectivity

using the olfactory sensory neuron (OSN) in the olfactory epithelium (OE) of

genetically modified mice as a model. We show that:

Inhibition of RA receptor (RAR)-mediated transcription in OSNs reduces

expression of the olfactory cyclic nucleotide-gated (CNG) ion channel, which

is required for odorant receptor (OR)-mediated stimulus transduction. This,

results in increased OSN death and errors in precise connectivity. The

increased cell death may be a consequence of reduced intrinsic excitability

and/or reduced influx of Ca2+ ions while the errors in connectivity may be

due to altered OR-dependent expression of axonal guidance proteins, such as

Kirrel-2 and Neuropilin-1.

Expression of the RA catabolic enzyme Cyp26B1 in OSNs is positively

regulated by RAR-mediated transcription as well as sensory stimulation in a

CNG channel-dependent manner. This shows that neuronal activity and local

vitamin A metabolism are parts of novel regulatory feedback loop controlling

precise connectivity and neuronal survival. The feedback loop may be a form

of homeostatic plasticity in response to global changes in neuronal activity.

BACE1, an enzyme is implicated in Alzheimer´s disease, and Cyp26B1 are

inversely regulated by CNG channel-dependent sensory stimulation.

Cyp26B1 expression is switched on at birth, forms a topographic expression

gradient in OE and inhibits BACE1 expression into an inverse counter

gradient. Taken together, these results reveal a novel neuronal activity-

dependent mechanism by which sensory stimuli can shape spatial gene

expression via altered RA bioavailability.

Increased Cyp26B1 expression stimulates turnover of OSNs during adult

neurogenesis by a non-cell-autonomous mechanism. The gradient of

Cyp26B1 expression correlates with spatially-regulated diversification of

OSNs into subpopulations that express different subsets of OR genes.

Cyp26B1 expression influences spatial OR diversification of OSNs by two

different mechanisms. In the ventrolateral OE, Cyp26B1 inhibits OR

expression by blocking OSN differentiation at a stage that may be associated

with the cell intrinsic mechanism regulating OR gene choice. In the

dorsomedial OE, the expression frequency of some ORs is unaltered while

iii

other increases, presumably as a consequence of neuronal activity-

dependent competition. A probable function of graded and activity-

dependent Cyp26B1 expression is to form a topographic partitioning of the

olfactory sensory map into functional domains, which gradually differ from

each other with regard to experience-driven plasticity and neurogenic

potential along the dorsomedial-ventrolateral axis of OE.

iv

ABBREVIATIONS

Aβ β amyloid

AD Alzheimer´s disease

AP action potential

A-P anterior-posterior

AC3 adenylate cyclase 3

APP amyloid precursor protein

cAMP cyclic adenosine 3´, 5´-monophosphate

BACE1 β site amyloid precursor protein cleaving enzyme 1

BL basal lamina

BG bowmans gland

Casp-3 caspase-3

CaM calmodulin

CaMKII calmodulin kinase II

CNG cyclic nucleotide- gated

CNGA2 cyclic nucleotide gated channel subunit A2

CNS central nervous system

CRABP cellular retinoic acid binding protein

DM dorso-medial

dnRAR dominant negative retinoic acid receptor

GBC globose basal cell

GL glomerular layer

GC granule cell

HBC horizontal basal cell

INP immediate neuronal precursor

LP lamina propria

LTD long term depression

LTP long term potentiation

MC mitral cell

Nrp neuropilin

NQO1 NADPH: quinone oxidoreductase 1

OB olfactory bulb

OC olfactory cortex

OE olfactory epithelium

OEC olfactory ensheating cell

OMP olfactory marker protein

OR odorant receptor

OSN olfactory sensory neuron

PD postnatal day

PNS peripheral nervous system

PKA protein kinase A

v

PPAR peroxisome proliferator-activated receptors

RA retinoic acid

RALDH retinaldehyde hydrogenase

RARE retinoic acid response element

RDH retinol dehydrogenase

RBP retinol binding protein

RAR retinoic acid receptor

RXR retinoic X receptor

Sus sustentacular cells

TH tyrosine hydroxylase

Sema semaphorin

VAD vitamin A deficient

VL ventrolateral

Z1-4 odorant receptor expression zones 1-4

vi

PAPERS IN THIS THESIS

This thesis is based on the following published articles and manuscript,

which will be referred to in the text by their Roman numerals (I-III).

I. OZTOKATLI, H*., HORNBERG, M*., BERGHARD, A. & BOHM,

S. 2012. Retinoic acid receptor and CNGA2 channel signaling are

part of a regulatory feedback loop controlling axonal

convergence and survival of olfactory sensory neurons. FASEB J,

26, 617-27.

This article is reprinted with permission from the publisher.

II. LOGIN, H*., BUTOWT, R*. & BOHM, S. 2014. Activity-

dependent and graded BACE1 expression in the olfactory

epithelium is mediated by the retinoic acid metabolizing enzyme

Cyp26B1. Brain Structure and Function, 1-15. doi:

10.1007/s00429-014-0783-z.

The final publication is available at Springer via

http://dx.doi.org/10.1007/s00429-014-0783-z.

III. LOGIN, H., HÅGLIN, S. & BOHM, S. Spatial differences in

activity-dependent neurogenesis and neuronal diversification in

the olfactory epithelium is regulated by the retinoic acid-

metabolizing enzyme Cyp26B1. Manuscript.

* These authors contributed equally.

1

INTRODUCTION A key question regarding all sensory maps is whether sensory experience

can modulate the neuronal circuitry and if so by which mechanism? During

my PhD studies, I investigated the connection between innate determinants

(i.e. genetically controlled molecular signals) and neuronal activity-

dependent plasticity during the formation and maintenance of correct

connectivity. Vitamin A derivatives (retinoids), especially all-trans-Retinoic

Acid (RA) is one of the innate determinants, a signaling molecule that

regulates the development of the nervous system.

RA-dependent genes control the pattern formation of the embryonic

nervous system. This regulation is orchestrated by spatiotemporal

expression of both RA synthetic (RALDHs: retinaldehyde dehydrogenases)

and degrading (Cyp26s: cytochrome P 450 26) enzymes. The exact function

and mechanisms that maintain an uneven distribution of RA signaling in the

different regions of the adult nervous system such as the hippocampus,

retina, and the olfactory epithelium remain unknown. The accumulated data

from many studies suggest interplay of RA metabolism and activity-

dependent neuro-plasticity in adults.

The sensory neurons of the olfactory system are easily accessible due to

their location in the nose. This accessibility, together with ongoing

regeneration, axon elongation and synaptic targeting make these neurons a

perfect model to study the processes of neuronal survival and axon targeting

in adult animals. The specific focus of my thesis is to understand the

crosstalk between RA and neuronal activity for the formation and

maintenance of the olfactory sensory map in the mouse.

OLFACTION

A recent report published in Science magazine claimed that based on

psychophysical tests human can discriminate at least 1 trillion different

olfactory stimuli (Bushdid et al., 2014). Even though this number is

significantly higher than the number of colors that human eye can

discriminate (2.3 million to 7.5 million), olfaction is not as essential as

vision. However for the majority of the animal kingdom, the ability to detect

and discriminate odorants in the environment is essential for survival.

Olfaction plays an important role in maternity behavior, recognition of

predator/prey and potential mates to reproduce. Also it is a key sensory

modality to identify food and taste perception (Shipley and Ennis, 1996).

Since olfactory pathways and limbic system are closely related, scents can

affect memory and generate emotions.

2

In mammals olfactory system is composed of two different parts: (i): the

accessory olfactory system which mostly detects pheromones that are non-

volatile chemical factors triggering intra/inter species social responses and

(ii) the main olfactory system which detects most of the volatile odors in the

environment. Nevertheless, only the mouse main olfactory system will be

described further in the introduction since only this part of the olfactory

system was used as a model in my study.

MOUSE MAIN OLFACTORY SYSTEM

In mouse main olfactory system, inhaled odorant molecules contact the

olfactory epithelium (OE), which contains numerous olfactory receptors

(ORs). These ORs are G protein coupled membrane proteins that are located

in the cilia of bipolar olfactory sensory neurons (OSNs). One OR can bind

more than one odorant molecules with different affinities. The chemical

signal generated give rise to depolarization upon binding of odorant

molecules that compose an odor, to their specific ORs. The depolarization

may result in action potentials that propagate in the axons of the OSN which

are part of the peripheral nervous system (PNS). The axons pass through the

small holes of cribriform plate and reach the olfactory bulb (OB) which is

part of the central nervous system (CNS) and is located in the anterior part

of the forebrain. After the odor information that reached OB is processed by

local circuits, it is transferred to different parts of the olfactory cortex (Mori

et al., 1999, Shepherd, 2004) (fig 1).

Figure 1. Schematic representation of the mouse main olfactory system.

Two components of the main olfactory system: Olfactory epithelium (OE) and

olfactory bulb (OB) are shown in red. OE lines in the nasal cavity and OB is located

in the anterior part of the forebrain. After chemical stimuli are converted to

electrical stimuli, the signal follows the way represented with arrows. First axons of

OSN project to OB and then secondary neurons in OB project their axons to OC

(olfactory cortex)

3

The Olfactory Epithelium

The olfactory epithelium which is located in the nasal cavity is a

neuroepithelium (fig. 1; 2A, B). Histological observations showed three

different major cell types in the OE. The cell bodies of these different cell

types are located in different layers: (i) the basal cells attached to the basal

lamina (BL), (ii) the OSNs in the middle and (iii) the sustentacular cells

(Sus) closest to the nasal lumen (fig. 2C). The underlying lamina propria

(LP) contains connective tissue including fibroblasts, blood vessels, olfactory

ensheating cells (OECs) that surround the axons and Bowmand´s gland (BG)

that produce mucus that covers OE (fig. 2C).

Figure 2. Anatomy and

histology of Olfactory

Epithelium (OE). A) The

location of OE in the nasal cavity

is shown in red. B) A coronal

section of OE. C) close up of the

area in the OE marked in red

square in B. Different cell types

in OE are illustrated. Basal

proliferating cells; horizontal

basal cells (HBCs, in yellow) and

globose basal cells (GBCs, in

blue) are located very close to basal lamina (BL). The somas of Sustentacular cells

(Sus, in pink) are in the most apical position. Both mature and immature OSNs

(OSNm in red, OSNi in green) are in between Sus and basal cells. In the lamina

propria under BL, the axons are surrounded by OEC (purple) and are close contact

with fibroblasts (brown). Bowman´s glands (BG) extend through the epithelium.

This layer organization of OE does not only reflect different cell functions

but also different cell maturation steps. Indeed, the basal cells are

C Nasal lumen

A B

4

proliferating progenitors and they are composed of two different

populations; horizontal basal cells (HBC) and globose basal cells (GBC).

HBCs are attached to the basal lamina, whereas GBCs are situated just above

HBCs (fig 2C). The middle layer is characterized by the OSNs that express

ORs. OSNs are found in different maturation stages because of the constant

turnover (Graziadei and Graziadei, 1979). Immature OSNs are positioned

close to the basal cells. They express neuron specific growth associated

proteins GAP43 and stathmin/SCG10 (Verhaagen et al., 1989, Pellier-

Monnin et al., 2001). On the other hand, the mature OSNs that express the

Olfactory Marker Protein (OMP) reside on top of immature OSNs. The

mature OSNs extend their cilia to the nasal lumen and they establish

synaptic connections with the cells located in the OB (Keller and Margolis,

1976). In fact, it is believed that the mature OSNs are the cells connecting OE

to central nervous system. Finally, the most apical layer of OE contains

sustentacular cells (Sus) that are glia-like supporting cells (fig. 2C).

Regenerative Capacity of Olfactory Epithelium

Like many exposed tissues that are in direct contact with the environment,

the OE in the nasal cavity frequently encounters toxins, environmental or

physical trauma. Hence the cells in OE can easily be damaged. Fortunately,

unlike the majority of CNS, the OE is able to regenerate constantly

throughout life. Indeed, new neurons are continuously generated to replace

old or damaged neurons (Murray and Calof, 1999, Schwob, 2002).

Understanding the reasons and mechanisms of the regenerative capability of

the olfactory epithelium can be useful to find out why neurogenesis is not

common in adult CNS and how it can be stimulated.

As it is mentioned above, the basal cells are proliferating cells and can

differentiate to OSNs. Basal cells are divided into two populations: GBCs and

HBCs. The identity of the stem cell of the OE is still not known. GBCs

proliferate actively and are composed of the transit amplifying and

immediate neuronal precursor classes express neuronal differentiation

markers MASH1 and neurogenin-1, respectively. (Beites et al., 2005, Gordon

et al., 1995, Schwob, 2002) (fig. 3). On the other hand, HBCs divide rarely,

express keratin but do not express any neuronal differentiation markers

(Carter et al., 2004). Massive disruption of the OE by methidium bromide,

an olfactotoxic reagent, induces HBCs division to ensure regeneration of

both OSNs and non-neuronal cells such as sustentacular and Bowmand´s

gland. Altogether, those observations, strongly suggest that HBCs are the

potential multipotent stem cells that become activated in response to injury.

However, under normal neuronal turnover conditions or selective neuronal

loss GBC progenitors ensure the OSN regeneration (Leung et al., 2007,

B

5

Duggan and Ngai, 2007) of the OE. The easy accessibility as well as the

continuous proliferation makes the OE a potential source of neuronal stem

cell for therapeutic applications in regenerative medicine.

Figure 3. Regeneration

of OSNs. Horizontal basal

cells (HBCs, yellow) and

Globose basal cells (blue)

proliferate with the

capacity of self-renewal.

Rapid proliferating transit

amplifying Mash1 +

globose cells (dark blue)

give rise to Neurogenin-1 +

immediate neuronal

precursor cells (INP, light

blue). The asymmetric

division of INP population give rise to OSNs. By time the immature OSN will get

mature and be located upper position in the OE.

The Odorant Receptors

ORs are G protein coupled receptors (GPCRs) also known as seven-

transmembrane domain receptors that bind to odorant molecules. ORs

constitute the largest protein family in mammals. Approximately 1300 and

350 functional members of this family are found in mouse and human

respectively (Buck and Axel, 1991, Young et al., 2002, Mombaerts, 1999). OR

genes have intronless coding regions and are clustered into up to 100 genes

broadly spread in 40-100 genomic locations (Young and Trask, 2002). Each

OSN expresses only one OR gene which gave the so called one cell-one

receptor rule. The location of OSNs expressing a given OR is restricted to

only a specific region or zone in the OE. OSNs with the same OR identity

project their axons to the olfactory bulb and converge to a few anatomically

discrete neuropils, called glomeruli. Thus each glomerulus has one OR

identity. Most ORs bind multiple odorant molecules and more than one

receptor can bind to the same chemical compound with different affinity

(Firestein, 2001, Mombaerts, 2004).

Based on phylogenetic analyses, the OR genes are divided into two classes:

class I and class II. In mouse, class I ORs (10% of functional ORs) are mostly

expressed by the OSNs located in dorsal part of the OE, whereas the OSNs

that express class II are distributed throughout the OE (Niimura and Nei,

2007, Tsuboi et al., 2006, Zhang et al., 2004). Also, axonal projections to the

OB mirror the organization in the OE. In other words, OSNs expressing class

6

I ORs extend their axons to the dorsal part of the OB while OSNs expressing

class II ORs extend their axons to all parts of the OB (Kobayakawa et al.,

2007, Tsuboi et al., 2006).

Odorant Receptor Expression and Regulation

Each OSN selects one OR gene and its expression is limited to one allele

(Chess et al., 1994). This monogenic and monoallelic features in the receptor

choice creates a great neuronal variety. Moreover, this OR singularity is the

core of OR-instructed axonal projection of OSNs and olfactory sensory map.

The mechanism behind singular OR expression is still elusive even

though a considerable amount of research projects are developed on this

topic. There are two major and difficult questions to answer, the first one:

“How neurons choose a single OR from many loci? And, the second one:

“How this singularity is maintained during the life time of an OSN?

A first model has been proposed in which the OR gene choice is

deterministic. In this model, it is believed that, a specific combination of

transcription factors bind to different regulatory DNA regions upstream of a

specific OR. Based on this hypothesis, DNA arrangements are irreversible

and assign the OR choice of the OSN (Rospars et al., 1996). However, data

from other studies argue against this hypothesis. Indeed, when mice were

cloned from nuclei extracted from differentiated OSNs that express a given

OR, the OSNs of those cloned mice display normal singular expression from

full OR repertoire (Eggan et al., 2004, Li et al., 2004, McClintock, 2010).

This result strongly indicates that OR selection does not require permanent

DNA arrangements.

Lately, an alternative model has been proposed to explain the one neuron-

one receptor rule. This model named stochastic model suggests a regulation

process implicating a cis-acting regulatory region located upstream of OR

gene clusters. This locus control region (LCR) controls multiple OR-genes

clustered at a specific locus. It is assumed that LCR can bind only one

promoter and activate the expression of one specific OR at a time by physical

interaction. Promoter choice of LCR is stochastic. This hypothesis is

supported by the discovery of an LCR region, called “H” that is located

upstream of the MOR28 OR gene cluster (Lomvardas et al., 2006, Serizawa

et al., 2003). Moreover it has been reported that homeodomain factors,

Lhx2, Emx2 and O/E family proteins bind to their motifs in the OR promoter

regions (Hirota et al., 2007, Wang et al., 2004). Therefore the interaction

between those transcription factors and LCR may trigger a reorganization of

the chromatin structure and finally, an activation of one OR promoter at a

time by physical interaction (Serizawa et al., 2003).

Many studies provide data that support the stochastic model but the

physical restriction of ORs into four different regions in OE along its

7

dorsomedial- ventrolateral (DM-VL) axis (Miyamichi et al., 2005, Ressler et

al., 1993, Vassar et al., 1993) shows that OR gene choice is not fully

stochastic. For instance, the location of the basal cell population may dictate

the OR choice of the OSN. Indeed it has been reported that transcription

factors such as Msx1 and Foxg1 are expressed in a gradient that follows DM-

VL organization of the ORs in the OE (Duggan and Ngai, 2007, Norlin et al.,

2001).The levels of these transcription factors could be cell lineage-

dependent, suggesting that OR gene choice can be regulated by this

positional information within OE. Both deterministic and stochastic models

speculate on the OR choice but neither of these models are enough to explain

how this single OR choice is maintained throughout the OSN life.

Concerning this point, a negative feedback regulation by the functional OR

has been identified as a mechanism for this lifetime OR monogamy of the

OSNs. According to the current model when a functional OR gene is

expressed, the expression of other ORs is consequently inhibited (Serizawa

et al., 2003). The components and targets of this negative feedback

regulation are still unknown. However, several observations strongly support

a feedback regulation process that would be critical for maintenance of OR

singularity. For instance, it has been shown that OSNs expressing non-

functional ORs containing deletions or frame shift mutations start to express

a second OR that is functional (Feinstein and Mombaerts, 2004, Lewcock

and Reed, 2004, Serizawa et al., 2003, Shykind et al., 2004). Moreover

endogeneous OR expression is attenuated by ectopic OR expression

(Fleischmann et al., 2008, Nguyen et al., 2007). Altogether these data

suggest OR-mediated feedback is also important for quality control. Thus if

one OSN chooses a non-functional pseudo-gene, a new OR-selection process

will take place until the OSN expresses a functional OR which is critical for

maturation of the OSN. It was observed that immature OSNs can switch

their ORs (Bader et al., 2010, Shykind et al., 2004). Switching must be

restricted in time to an early window in the development of sensory neurons,

prior to axons migration to specific glomeruli in OB. According to the recent

studies, epigenetic modifications by LSD1, -a histone demethylase, play an

important role in stabilizing OR expression. LSD1 is expressed in early steps

of OR choice. When a functional OR is encoded, expression of adenylate

cyclase III, a component of the signal transduction pathway in OSNs, is

upregulated which leads reduction in LSD1 levels. The downregulation of

LSD1 prevents activation of other OR genes. ACIII allows correct targeting of

axons, synapse formation and maturation of OSNs (Lyons et al., 2013).

Another study reported that OR clusters were compact with high levels of

methylated H3 (H3K9me3) and H4 (H420me3) before OR transcription.

Thus according to their model, OR silencing is followed by OR expression

and the basis of monogenic and monoallelic OR expression relays on the

escape from chromatin-mediated silencing (Magklara et al., 2011).

8

Another interesting findings is that in neonatal mice, the OSNs located in

the septal organ (a small part of the epithelium located at the anterior and

ventral part of the septum) express multiple ORs. However by sensory

stimulation/neuronal activity those OSNs with multiple ORs are eliminated

with an unknown mechanism during postnatal development. In the same

study sensory deprivation in young adult mice by unilateral naris closure

(UNO) experiments showed that the frequency of OSNs with multiple OR is

high in neonatal mice. Based on these data, the authors proposed that for a

subset of OSNs, the maintenance of the one neuron-one receptor rule

depends on stimulus induced neuronal activity (Tian and Ma, 2008).

Olfactory Signal Transduction Pathway

The olfactory signal transduction pathway corresponds to a cascade that

transforms the information from the chemical stimuli to electrical signal that

will reach the brain. This pathway is initiated in the cilia by binding of an

odorant molecule to an OR. This binding activates a specific heterotrimeric

guanine nucleotide protein (Golf) that subsequently enhances cyclic AMP

(cAMP) production by the adenylate cyclase type III (ACIII). Increased level

of intracellular cAMP leads to opening of cyclic nucleotide gated channel

(CNG) and influx of cations (Na+ and Ca+2). Increased Ca+2 levels depolarize

the cell membrane and activate the Cl- channels. Because of relatively high

intracellular Cl- concentration in OSNs under resting conditions, opening of

the channel will result in Cl- efflux. This will eventually result in further

depolarization of the cell membrane which will trigger the opening of voltage

gated channels and the formation of action potential (AP) afterwards (fig 4).

The olfactory signal transduction pathway can rapidly adapt its sensitivity to

stimulation via Ca+2- dependent negative feedback regulation. Ca+2 binds to

Calmodulin (CaM) and form a complex that reduce the binding affinity of the

CNG channel to cAMP. Moreover this complex is involved in activation of

Calmodulin Kinase II (CaMKII) that phosphorylates ACIII resulting in cAMP

level reduction, and finally inhibition of depolarization (Matthews and

Reisert, 2003, Zou et al., 2009) (fig 4).

In mammals, 6 CNG channel genes code for four alpha subunits (CNGA1-

4) and two beta subunits (CNGB1 and CNGB3). In OSNs the CNG channels

are heterotetramers which are composed of two CNGA2, one CNGA4 and

one CNGB1b subunits. CNGA2 is actually the functional subunit of the

channel that is responsible of the signal transduction process, whereas the

other subunits modulate the channel properties (Bradley et al., 2005). It has

been shown that CNGA2 knockout mice are anosmic for most of the

odorants, highlighting CNGA2 importance for signal transduction (Brunet et

al., 1996, Lin et al., 2004). Moreover, the CNG channel does not only

9

contribute to odor-induced neuronal activity but also to the resting

conductance/membrane potential at basal cAMP levels (Pun and Kleene,

2003). CNG channels are also located in the axons of OSNs and they provide

local regulation of Ca+2 levels and neurotransmitter release (Murphy and

Isaacson, 2003). Altogether, these different studies show a critical role of

CNG channel for signal transduction in cilia and also for synaptic regulations

in axon terminals.

Figure 4. Olfactory signal transduction pathway in cilia. Binding of an

odorant molecule to an OR activates the Golf and AC3 leading to increased levels of

cellular cAMP and the opening of the CNG channel respectively. The cation influx

via the channel will activate the Cl- channel. Cl- efflux depolarizes the cell

membrane. If the depolarization is over the threshold, action potential (AP) is

formed. Intracellular Ca+2 level is also involved in the negative feedback regulation

by inhibiting cAMP production and cAMP-CNG interaction for adaptation.

Zonal Organization of Olfactory Epithelium

It has been mentioned earlier in the introduction that the olfactory

epithelium (OE) has a relatively organized topography. In fact, the OE is

divided into four topographically distinct zones based on the expression

patterns of different ORs (fig. 5A). A certain OR is expressed only in one

specific zone. The zones are organized along dorsomedial (DM) -

ventrolateral (VL) axis of OE. OSNs expressing Zone 1 (Z1) ORs are located

in the most DM position; whereas OSNs with Zone 4 (Z4) ORs are located in

the most VL position (fig. 5B). Projections of the axons of OSNs also follow a

spatial organization along dorsal-ventral axis in the OB (fig 5C). The

mechanism controlling specification and zone projections of OSN is elusive.

In several parts of the nervous system i.e., retina, cerebral cortex, during

10

early development cell specificity is determined by combination of intrinsic

and extrinsic signaling molecules (Edlund and Jessell, 1999). The fate of

OSN is probably determined by the same way when OSN stem cells are

generated. Due to constant neurogeneration in adults, the cells in the OE

need to express the signaling molecules for the OR specificity and the genes

that are necessary for correct projections throughout life. In adult mice zone-

specific and graded patterns of gene expression that correlate with the

organization of the OR zones have been identified (fig. 5c). Zone specific

genes are expressed in either Z1 or Z2-4. For instance, quinone detoxifying

enzyme NQO1 expression is found in Z1 OSNs, whereas neural cell adhesion

molecule 2 (NCAM2) and RA synthesizing enzyme RALDH1 are expressed in

z2-4 OSNs and sustentacular cells respectively (Alenius and Bohm, 1997,

Gussing and Bohm, 2004, Peluso et al., 2012). It has been shown that the

expression of Msh homeobox 1 (Msx1) which is involved in determining the

location of neural induction and retinoic acid synthesizing enzyme-3

(RALDH3) genes follows a VLhigh -DMlow gradient in the basal cells. A similar

gene expression gradient was detected for semaphoring receptor Neuropilin-

2 and another retinoic synthesizing enzyme RALDH2 respectively in OSNs

and in the perineural cells (the cells surround the axons). Expression of Bone

morphogenetic protein (BMP) –type 1 receptor (ALK6), which binds to Msx1

key regulator BMP, in sustentacular cells showed a reverse gradient (Norlin

et al., 2001, Peluso et al., 2012)(fig. 5C). The fact that those genes are

involved cell specification and axonal guidance, strongly suggests that

positional information from the local environment in OE may dictate a zone

specific expression of ORs and axonal guidance molecules by OSNs for

formation of correct sensory map.

A B

11

Figure 5. Zonal organization of OE and olfactory sensory map. A)

A schematic representation of mouse OE divided in 4 zones following a

dorsomedial-ventrolateral axis. B) in situ hybridizations with Zone1 (Z1)

and Zone 4 (Z4) ORs on a OE coronal tissue section (positive labelling

appears in white). Note the difference in spatial location of the zones and

the scattered distribution of a specific OR expression within a restricted

zone. The Z1 OR is expressed in the most dorsomedial part of the OE,

whereas the Z4 OR is expressed in the most ventrolateral part of the OE. C)

OSNs expressing the same OR converge to the same glomeruli. The OSNs

with different OR identities are shown in different colors. Axons from Z1

will project to the dorsal area of the olfactory bulb, while axons from Z 2-4

project more ventrally. In the OE, genes that have gradient patterns that

follow the zonal organization (shown in triangles) and zone specific

expression confined to Zone1 or Zone 2-4 (in bars) have been identified.

(D=Dorsal, V=Ventral, M=Medial, L=Lateral)

Involvement of neuronal activity in the expression patterning of OR

throughout the OE has been reported, however the mechanism behind

remains unclear. Recently Ma and his colleagues investigated the

expressions of different OR genes by in situ hybridization when the mice

were sensory deprived by UNO. They reported that OSNs that belong to

different zones were affected differently. The density for OSNs with Z4 ORs

was increased, whereas the density for OSNs with Z1 ORs decreased (Zhao et

al., 2013). These results indicate a putative influence of neuronal activity in

ORs pattern formation.

D

V

C

12

Another important question is: What is the functional importance of zonal

organization of OE along DM-VL axis? It has been observed that turnover in

VL part is higher than DM part of the OE and this difference develops

postnatally by neuronal activity (Vedin et al., 2009). Also, it has been

reported that DM connectivity is more sensitive to global inhibition of

activity during postnatal development (Yu et al., 2004, Zheng et al., 2000a).

Moreover, Sakano and his colleagues showed that DM OSNs mediate innate

responses to aversive and predator odorants, whereas VL OSNs which are

responsible to mediate conditionally learned odorant associations link to the

same odorants (Cho et al., 2011, Kobayakawa et al., 2007). Collectively these

results suggest that zonal organization of OE mediates spatial differences in

neurogenesis and neuroplasticity associated which is associated with

different forms of activity-dependent learning and memory.

Olfactory Bulb

The main olfactory bulb (OB) is a bilateral structure located in the

anterior part of the forebrain (fig. 1, 6A). It is the first synaptic target of OSN

axons. OB is constituted by different cell types that are organized in layers

(fig. 6B,C). The axons of OSNs pass through the cribriform plate (part of the

bone separating the brain from nasal cavity) to form the nerve layer around

the OB from which they penetrate the glomerular layer. Thereafter, OSN

axons converge to anatomically discrete neuropils, named glomeruli where

they make synapses with mitral/tufted cells (Graziadei and Graziadei, 1979).

The mitral/tufted cells extend their dendrite to receive input from only one

glomerulus. The information is processed/modified by inhibitory

interneurons called periglomerular cells that are located around the

glomeruli and granular cells which reside deepest layer of the bulb (Pinching

and Powell, 1971). The mitral cells in turn send their axons to olfactory

cortex in brain where the information is further transmitted to higher

cortical and limbic areas for conscious and emotional perception of odor

(Restrepo et al., 2009) (fig. 6C).

In mouse main olfactory system odor information is detected by

approximately 1200 different types of odorant receptors and each individual

OSN is specified to express only one defined OR (Buck and Axel, 1991,

Ressler et al., 1993). OSNs expressing the same OR, project their axons to

the same glomeruli (defined as one glomeruli-one receptor rule) where they

make connections with mitral/tufted cells. OR- specific axonal convergence

from ~1200 OR-specific OSN subpopulations generates a sensory map that

contains as many pairs of medial and lateral OR-specific glomeruli at

stereotyped positions. Each odorant will activate a unique and specific

combination of activated glomeruli in the OB. This topographical

13

organization as well as the odorants binding features by ORs allows

mammalian brain to detect and discriminate different kinds of odorants

(Korsching, 2002).

Formation of Glomerular Map

The olfactory map is established by a combination of genetically

determined targeting and activity-dependent refinement processes(Chen and

Flanagan, 2006). According to the current model, at early developmental

stage, OSNs are guided to approximate destination in the OB by the dorsal-

ventral patterning. As mentioned before dorsal-ventral positioning of the

projections depend on the anatomical location of the OSNs in the OE. OSNs

expressing Z1 ORs extend their axons to the dorsal part of the OB, while

OSNs with Z4 ORs project their axons to the ventral part of the OB (Alenius

and Bohm, 1997, Ressler et al., 1993, Vassar et al., 1993). Studies have shown

that two sets of repulsive ligands/receptors, which are Slits/Robo2 and

Semaphorin 3F/ Neuropilin-2 play role in guidance of axonal projections to

dorsal or ventral parts of the OB (Norlin et al., 2001, Takeuchi et al., 2010).

After axonal penetration to the glomerular layer, the anterior-posterior (A-P)

position of the glomeruli is determined by odor- independent OR signaling

(fig. 8A). It has been shown that ORs conduct intrinsic activity during

absence of odorant molecules to bind. This odor independent OR signaling

leads to production of basal level of cAMP via activated AC3. Each OR

generates a specific level of cAMP which constitutes a signal for relative axon

guidance molecules. Such regulation was observed for Neuropilin-1 (Nrp-1),

and its repulsive ligand Sema3A via cAMP-dependent protein kinase A

(PKA) signaling. The posterior part of the OB is occupied by the axons of the

OSNs with high levels of cAMP, whereas the OSNs with low levels of cAMP

project their axons to the anterior part of the OB (fig. .8A). Studies showed

that when intrinsic OR activity is inhibited the olfactory sensory map is

severely perturbed (Chesler et al., 2007, Imai and Sakano, 2009, Imai et al.,

2006, Nakashima et al., 2013).

A B

14

Figure 6. Anatomy and histology of Olfactory Bulb (OB). A) The location of

OB in the forebrain. B) A coronal section of OB (marked with dashed line in A) where

all nuclei were stained with Hoechst showing different histological layers. Glomerular

layer (GL) consists of glomeruli (marked with red circles and pointed out with red

arrow). Mitral cell layer (MCL) and granule cell layer (GCL) are pointed out with blue

and green arrows, respectively. C) A schematic drawing of neuronal circuitry in OB.

The OSNs project their axons to their corresponding glomeruli (blue, red, green

OSNs to blue, red and green glomerulus respectively). Within the glomeruli axons

make synapses with mitral cells (MC). After the information is modified by granule

cells (GC) and periglomerular cells (PGCs) that is located around glomeruli, MCs

send their axons to olfactory cortex (OC).

Neuronal Activity Dependent Axonal Refinement

After OSN axons are sorted to approximate positions in glomerular layer,

the connections are refined by activity dependent processes. By definition,

refinement corresponds to neuronal activity dependent remodeling of

connections (synapses). Many studies have shown that odor-evoked

neuronal activity is important for axonal refinement of OSNs. It has been

found that in early postnatal development many glomeruli are mixed

(heterogeneous) which means those glomeruli are innervated by axons of

OSNs with different OR identities. However, over time by odor evoked-

neuronal activity those glomeruli are eliminated and axons with different OR

identities segregate and find their correct targets (Fig 7). It has been shown

that in adult mice which were operated neonatally to close one of the naris,

glomerular refinement in the ipsilateral olfactory bulb was attenuated (Zou

et al., 2004). Likewise, elimination of ectopic glomeruli is inhibited in a

transgenic mouse line in which OSNs initially project their axons to wrong

C

15

targets. In addition, it has been reported that the time course of glomerular

refinement is decreased when the mice were stimulated with high levels of

odorants (Kerr and Belluscio, 2006). In addition to unilateral occluded mice,

genetically modified mice deficient in CNGA2 were analyzed to understand

the mechanism of neuronal activity for olfactory sensory map formation. In

CNGA2 deficient mice, the overall glomerular map seems normal (Baker et

al., 1999, Lin et al., 2000, Zheng et al., 2000a). With this observation it was

assumed that channel activity is not important for glomerular refinement.

Later CNGA2 knockout mice were used to study enhanced competition

between inactive and active OSNs. The CNGA2 gene is on the X chromosome

and due to dosage compensation, female CNGA2+/- mice have an OE that is a

mosaic for active and inactive OSNs. As a consequence of competition of

neuronal space, axons of the same OR identity segregate into CNGA2+ and

CNGA2- glomeruli (Serizawa et al., 2006, Zheng et al., 2000a). By using

sensory deprived (one naris closed) and CNGA2 deficient mice, Sakano and

his colleagues showed that some of the cell adhesion/repulsion molecules

such as Kirrel-2/-3 and EphrinA5/Eph-A5 are transcribed in a CNG activity-

dependent manner. Odor induced neuronal activity upregulates Kirrel2 and

Eph-A5 expression, whereas it inhibits Kirrel-3 and EphrinA5 expression in

a CNG-dependent manner (fig. 8B). More interestingly, in the same study

they showed that the expression levels of those molecules are determined by

OR types. Thus, different ORs generate different neuronal activities that

subsequently assign the expression levels of axon sorting molecules. The

genes encoding axonal guidance molecules that contribute the A-P

positioning of the glomeruli are not regulated in a CNG-dependent manner

(Serizawa et al., 2006). These results suggest that both odor- independent

and odor- dependent OR signaling pathways are involved in axonal targeting

of OSNs. Odor-independent OR signaling pathway results in regulation of

axonal guidance molecules responsible for A-P positioning of glomeruli

while odor dependent OR mediated signaling leads the regulation of axonal

guidance molecules which refines the glomeruli (fig. 8A, B).

16

Figure 7. Neuronal activity dependent glomerular refinement. In early

development the mice glomeruli are innervated by axons with different odorant

receptor identities. Blue and red OSNs project their axons to both of the glomeruli,

giving two “mixed” (heterogeneous) glomeruli. Postnatally with the neuronal

activity those blue and red axons refine and segregate to their correct targets

(blue/red glomeruli).

In vertebrates visual system, before any visual experience, spontaneous

waves of action potentials are generated triggering the formation of the

terminal zone by retinal ganglion cell axons (McLaughlin et al., 2003). OSNs

also generate spontaneous firing in the absence of odorants using the basal

cAMP levels (Nakashima et al., 2013, Rospars et al., 1994). Analyses of

anosmic CNGA2 deficient mice, which fail to exhibit odorant-evoked

responses, revealed that in these mice the axonal wiring seems to be normal

(Baker et al., 1999, Belluscio et al., 1998, Lin et al., 2000). Based on these

observations, many studies focused on the importance of spontaneous

activity for glomerular formation. It was shown that axonal convergence is

perturbed severely in inward rectifier K+ channel (Kir2.1) overexpressing

mouse. The OSNs of this mouse line are hyperpolarized and this process

results in inhibition of both odor-evoked and spontaneous action potentials

(Yu et al., 2004). Recently it has been found like in Drosophila (Hallem et al.,

2004) that, the spontaneous firing rate is regulated by OR choice of the OSN

(Connelly et al., 2013). Moreover they showed that OSNs that are expressing

inactive OR don´t generate spontaneous activity.

17

Figure 8. Odorant Receptor (OR) -mediated axonal guidance. According to

the current model there are two types of OR-mediated signaling that are involved in

axonal guidance of OSNs. Odor-independent OR-signaling play a role in anterior-

posterior (A-P) glomerular positioning by regulating axonal guidance molecules that

are downstream of PKA-signaling, like Nrp-1 and PlexinA1. Odor dependent OR-

signaling play a role in glomerular refinement by regulating axonal guidance and

adhesion molecules downstream of the CNG channel, like Kirrel-2/-3,

EphrinA5/Eph-A5

Neuronal Activity-Dependent Survival of Olfactory Sensory

Neurons

Many studies proposed that the survival of OSNs as well as the sensory

map organization and maintenance are impacted by neuronal activity.

Indeed, in absence of any odorant dependent activity (e.g. closed naris), all

OSN types stably coexist. However under normal conditions, the accessibility

to odorant molecules creates a competition between activated (the ones that

they bind to their odorant molecule) and inactivated (the ones that do not

bind to their odorant molecule). This competition actually triggers

elimination of the inactivated OSNs. In other words “Use it or lose it” dogma

is applicable on olfactory system.

In fact, the effect of activity triggered competition on correct axonal

targeting and cell survival was first proposed in 1960s by David Hubel and

Torsten Wiesel. Their hypothesis was based on an elegantly designed

experiment where they showed that stimulation from the environment

influence the development of the brain area responsible for the visual signal

processing. Using a cat as a model, they showed that sensory deprivation in

one eye during critical period inhibits ocular dominance development in

visual cortex. Monocular deprivation reduced the width of the ocular

columns for the closed side (Hubel and Wiesel, 1964). In the same way, in

CNGA2 mosaic mice the OSNs without functional channel are eliminated by

time. This phenotype can be explained by a competition between the odorant

activated and inactivated OSNs. However when there is global activity

deprivation, provided by naris occlusion, both CNGA2 expressing and not

expressing OSNs survive (Zhao and Reed, 2001). These results suggest that

CNG affect the survival via an activity dependent mechanism.

Other studies showed that not only odor-induced neuronal activity but

also spontaneous activity plays a role in the survival of OSNs under

competitive environment. Gogos and colleagues reported that tetanus toxin

(blocks neurotransmitter release) or Kir2.1 (attenuates of AP formation by

hyperpolarizing OSNs) expression in small populations of OSNs results in

more severe consequences regarding survival and axonal targeting compared

to a situation in which all OSNs are equally weakened (Yu et al., 2004).

18

The molecular mechanisms linking neuronal activity to the survival of

OSNs are elusive. It has been shown that odorant stimulation increase the

lifespan of OSN by inducing MAPK/CREB-dependent transcriptional

pathway (Watt et al., 2004). Recently Santaro and Dulac discovered that the

histone variant H2Be which is involved in chromatin remodeling, is

expressed in an OR in a neuronal activity dependent manner. The levels of

H2Be expression in an OSN depend on the identity of the co-expressed OR.

Gain and loss of function experiments displayed that this histone variant is

important for OR patterning and olfactory function. Since H2Be expression

starts after OR expression, the OR choice is not affected by H2Be levels.

Instead, according to the suggested model, the active OSNs which are

stimulated by odorant molecules live longer and the OSNs that are activated

infrequently have a short lifespan. These findings may explain the

mechanism behind the elimination of the silent OSNs (Santoro and Dulac,

2012).

RETINOIC ACID

Retinoic acid (RA) is a small lipophilic signaling molecule derived from

Vitamin A (retinol). In most of the cases RA act as a ligand for nuclear

receptors, inducing the transcriptional activity of target genes that play a

critical role in many biological processes during embryonic development

such as development of the nervous system including the olfactory system

(LaMantia et al., 1993, Maden, 2007).

Retinoic Acid Metabolism

Animals, including human obtain Vitamin A, like the other vitamins by

dietary intake. Retinol (Vitamin A) in the bloodstream is bound to retinol

binding protein (RBP), and this complex is taken up by RA producing cells.

Once inside the cell, the retinol binds to cellular retinol binding protein

(CRBP), which facilitates the 2 sequential oxidative steps. First retinol is

converted to retinaldehyde by the retinol/alcohol dehydrogenases (RDH)

and retinaldehyde is converted to RA afterward by retinaldehydrogenases

(RALDH-1, -2, -3)(Duester, 2000). RA binds to cellular retinoic acid binding

proteins (CRABP-1, -2). After the synthesis there are a few possible scenarios

that RA may follow. RA can (i) activate transcription in the nucleus, (ii) exert

non-genomic actions, (iii) diffuse to neighboring cells (paracrine signaling)

or finally (iv) be inactivated by degradation (fig. 9).

Mostly RA exerts its biological functions via by binding canonical RA

receptors (RARs) in the nucleus. RARs are ligand inducible transcription

factors belonging to the superfamily of nuclear hormone receptors and can

19

be found in different variants (RARα, β, γ) (Chambon, 1996). In fact, RARs

form a heterodimer with retinoid X receptor (RXR) and this complex when

associated to RA activates the expression of RA-regulated genes. RAR/RXR

heterodimer binds to a RA response element (RARE) which is, a regulatory

DNA element present in the promoter region of RA regulated genes

triggering a corepressors/coactivators exchange to initiate the transcription

(Germain et al., 2002). RA is delivered to RARs by CRABP-2. Recent studies

also mention the presence of an alternative receptor for RA, which is called

PPARβ/δ. This receptor, like RAR also makes heterodimer with RXR to get

functional. However for PPAR- mediated transcription, RA is delivered to

PPARβ/δ by the fatty acid binding protein 5 (FABP5) instead of CRABP-2

(Berger and Moller, 2002). It has been suggested that activation of those

receptors by RA has different effects and the ratio between CRABP-2 and

FABP5 assign which receptor will be activated (Schug et al., 2007, Yu et al.,

2012).

Although RA is known as transcriptional regulator, recent studies showed

that RA and RARα regulate translation in hippocampal neurons (Chen and

Napoli, 2008, Chen et al., 2008).

Alternatively to autocrine manner, RA can also be release from the

synthesizing cells and diffuse to adjacent cells. In rodent hippocampus, the

RA synthesized by RALDH-2 is not enough to support RA dependent

processes. It has been discovered that meninges which is located very close

to dentate gyrus, a part of hippocampus with high neurogenesis and

responsible for memory formation, is providing the RA for hippocampus

(Goodman et al., 2012).

The cellular level of RA is regulated by the ratio between RA synthesizing

and degrading enzymes. Cyp26s are the RA degrading enzymes that

inactivate RA in cells that do not need RA. RA is delivered to Cyp26s by

CRABP-1. The most common degradation products are 4-oxo-RA, 4-OH-RA

and 5-6-epoxy-RA. It is believed that those degradation products are not

metabolically active. However it has been reported that some RA

degradation products have biological activities (Gaemers et al., 1996,

Pijnappel et al., 1993).

Function of Retinoic Acid

From old civilizations, ancient Egypt medicine is one of the most

advanced medicines. Several documents reported that they were using liver,

which is the richest source of Vitamin A, to cure the night blindness. Without

knowing, they were the first who established a connection between vision

and Vitamin A. Centuries later, it has been experimentally shown that

Vitamin A deficiency results in vision, growth and reproduction defects

(Mark et al., 2006). After the discovery of Vitamin A in 1913, many studies

20

revealed that Vitamin A plays an important role in neurogenesis, apoptosis,

patterning of embryo and organogenesis. Actually, the physiologic effects of

Vitamin A are mediated by its biologically active derivative, RA.

Figure 9. Metabolic pathway of Retinoic acid (RA) and its signaling in the

cell. In the bloodstream, retinol (Vitamin A) binds to the retinol binding protein

(RBP) and thereafter it is internalized by RA generator cells. Once in the cytoplasm,

the retinol binds to cellular retinol binding protein (CRBP) and is converted to RA

by two oxidation steps; first it is converted to retinaldehyde by

retinoldehydrogenases (RDHs) and then to RA by retinaldehydrogenases

(RALDHs). After RA binds to to cellular retinoic acid binding protein (CRABPs), it

can : 1. Be transported to nucleus to mediate transcription 2. Exert non genomic

actions 3. Diffuse to neighboring cells and involved in paracrine manner cell

signaling or 4. Be degraded by Cyp26 family.

In the developing nervous system, RA has two main roles: neurogenesis

and control of neuronal patterning. The first function has been extensively

studied using in vitro models. It is now known that RA induces the

differentiation of different types of neurons and glia (Gottlieb and Huettner,

1999, Maden and Holder, 1991) by activating the transcription of many genes

such as genes encoding transcription factors, cell signaling molecules,

enzymes and cell surface receptors (Maden, 2006). As a patterning factor,

RA is involved in both the A-P and D-V patterning of the neural plate and

neural tube (the structures that CNS originates from). RA cooperates with

21

well-studied signaling molecules WNTs and fibroblast growth factors (FGFs)

for (A-P) patterning of neural plate to organize both posterior hindbrain and

the anterior spinal cord (Liu et al., 2001, Maden, 2002, Melton et al., 2004).

RA signaling deficiency results in undeveloped posterior hindbrain and an

abnormal anterior spinal cord (Maden et al., 1996, Wilson et al., 2004).

Furthermore, studies have shown that RA is synthesized by RALDH2 in the

posterior mesoderm, whereas RA is degraded by Cyp26B1 in the anterior

mesoderm creating a RA gradient. It is plausible that this RA gradient is

involved in hindbrain patterning (Glover et al., 2006, Reijntjes et al., 2004).

Regulation of patterned gene by a molecular gradient has been described

before. For example, in spinal cord neurons are classified by homeobox

transcription factor expression profiles. Dorsal patterning is controlled by a

gradient of bone morphogenetic proteins (BMPs) while ventral patterning is

regulated by sonic hedgehog (Shh). RA cooperates with those genes to

regulate D-V axis development. In RA- depleted chicks, the spinal cord

ventral signaling is increased whereas the dorsal signaling is decreased. This

change in the signaling pattern results in absence of interneurons (Diez del

Corral and Storey, 2004, Wilson and Maden, 2005).

Neurogenesis is the second major role of RA in the developing nervous

system. Indeed, several studies explored the effect of RA on stem cell

progeny differentiation. For instance, RA induces the differentiation of

embryonic stem cells in culture to Pax-6 expressing neural progenitors (Bibel

et al., 2007). However the effect of RA on proliferation depends on cell types.

It has been shown that RA inhibits proliferation in many different cell types.

Actually, some medical treatments have taken advantage of the anti-

proliferative properties of RA to treat diseases such as leukemia, breast and

lung cancers (Clarke et al., 2004, Crowe, 2002). Also, it has been shown the

exposure to RA inhibits proliferation of progenitors in the subgranular zone

of the hippocampus where adult neurogenesis take place (Crandall et al.,

2004). However, in some tissues RA appears to promote cell proliferation. It

was reported that skin tumor formation can be stimulated by RA. In this

study, it has been shown that proliferation of basal skin cells was increased

when RA was administered topically to the skin (Schmuth et al., 2007).

Later, it was suggested that the effect of RA on proliferation depends on

which one of two different RA nuclear receptors, RAR or PPARβ/δ is

activated (Schug et al., 2007). According to the data shown in this study RA

can either induce cell proliferation by activating PPAR β/δ (like in skin cells)

or inhibit proliferation via RAR (like in cancer cells).

It is known that RA degrading Cyp26 enzymes have important

developmental functions. For instance, it has been found that in Xenopus, 4-

oxo-RA, a RA catabolic product generated by Cyp26B1, binds to RARs to

control the A-P positional specification during early embryogenesis

(Pijnappel et al., 1993). Moreover the same metabolite induce

22

spermatogenesis in mouse (Gaemers et al., 1996). In most of the cases

expressions of Cyp26 and RA synthesizing enzymes are mutually exclusive.

This non-overlapping expression can generate a RA gradient. Such uneven

distribution of RA metabolism plays a significant regulatory role for pattern

formation of different part of the nervous system such as embryonic

hindbrain and retina (Maden, 2007, Leung et al., 2012, Sen et al., 2005).

Thus, it has been shown that spatial distribution of enzymes that synthesize

RA and degrading RA divides the chick retina into 3 different regions along

D-V axis of chick retina: a ventral part with high RA concentration, a dorsal

region with low RA concentration and a transitional region (Sen et al.,

2005). It has been suggested that this graded RA is required for D-V

patterning of optic cup (McCaffery et al., 1999, Wagner et al., 2000).

Several studies showed that RA is not only important for the CNS

development, but also functions as a regulator of neuroplasticity. In

hippocampus RA is important for homeostatic synaptic plasticity to keep a

steady neuronal network which is necessary in learning and memory (Aoto et

al., 2008, Chen and Napoli, 2008, Maghsoodi et al., 2008, Wang et al.,

2011). It has been shown that when hippocampal neurons are silenced by

tetradotoxin, the Ca+2 levels are decreased in inactive postsynaptic terminals

and this leads to RA synthesis which is followed by production of new AMPA

receptors to increase excitability of the neurons (Aoto et al., 2008). Different

studies by using vitamin A deficiency (VAD) by diet revealed that

hippocampal long term potentiation (LTP) and depression (LTD),

neurogenesis, all depend on Vitamin A (Misner et al., 2001, Cocco et al.,

2002, Jacobs et al., 2006). The defects were reversible with RA treatment.

VAD rats produce a weaker LTP. Recently it has been reported that VAD

attenuates RARα expression is causing downregulation of an N-methyl-D-

aspartate (NMDA) receptor subunit via a non-transcriptional mechanism

and ultimately an inhibition of hippocampal neuronal Ca+2 excitability. As a

consequence of LTP weakness, active learning and spatial memory in adult

rats are affected (Ghenimi et al., 2009, Jiang et al., 2012).

Similarly to VAD rats, the defect in spatial memory, LTP weakness and

RAR downregulation are observed in ageing rats. Again, these effects can be

suppressed by RA administration (Enderlin et al., 1997, Etchamendy et al.,

2003). Based on these observations, it is tempting to make a correlation

between RA signaling defect and age related neuronal death, cognitive

failure, or even the neurodegenerative diseases. Studies have shown that in

Amyotrophic lateral sclerosis (ALS) patients, RARα and RALDH2 expression

are downregulated (Corcoran et al., 2002). The hallmarks of Alzheimer´s

disease (AD), which are accumulation of amyloid-β (Aβ) peptide in cerebral

vessels, downregulation of RARα, lack of choline acetiltransferase (CHAT) in

the forebrain cortical neurons, are the observations collected from VAD rats

(Corcoran et al., 2002). Also it has been found that RA induces the α

23

secretase, ADAM10 which strengthen LTP , therefore it is involved in a

mechanism that protect from AD (Fahrenholz and Postina, 2006). Finally it

has been found that AD patients present the same symptoms observed in

VAD rats (e.g.; downregulation of RARα and RALDH2 expression.

Altogether these observations suggest that the components of RA signaling

and metabolism can be potential targets for treatment of aging and

neurodegenerative diseases.

The Function of RA in the Olfactory System

As it was mentioned earlier, the RA concentration gradient generated by

differential expression of RA synthetic enzymes (RALDH-1, -2, -3) is

important for pattern formation during development of nervous system

(Appel and Eisen, 2003, Ross et al., 2000). This morphogenic property of

RA is also true for olfactory system. RA governs the formation of nasal

structure, olfactory epithelial organization and mesenchymal and epithelial

tissue interaction together with other signaling molecules such as FGF8,

sonic hedgehog (shh) and BMPs. It has been shown that in the early stage of

embryogenesis during olfactory system development, RALDH-2 expression

in the frontonasal mesenchyme leads to a local production of RA (Bhasin et

al., 2003). Such a local RA synthesis is important for mesenchymal and

epithelial interaction which leads to expression of RA signaling molecules;

i.e. RARs, CRABPs (LaMantia et al., 1993). Also, it has been shown that mice

lacking RALDH-3 die neonatally because of severe abnormalities of the nasal

region (Dupe et al., 2003). Moreover, Pax6 mutant mice that are not able to

synthesize RA in the frontonasal mesenchyme, are characterized by a

disturbed OE and a mucosa that is unorganized and undifferentiated

(LaMantia et al., 2000). Another interesting result from the same study

displays that the differentiation of the lateral part of OE is more affected that

the medial part. DMlow-VLhigh differential RALDH-1, -2 and-3 expressions

suggest an important role of RA in olfactory mucosa (Niederreither et al.,

2002). Also, it has been shown that RALDH-3-produced RA synthesis is

essential for OB development. In RALDH-3 deficient mice the number of

GABAergic neurons in OB declined drastically probably due to a decrease of

differentiation (Chatzi et al., 2011). Analyses of RA-reporter mice showed

that RA activates a population of OSNs in the embryonic OE in the beginning

of neurogenesis (Rawson and LaMantia, 2006). Another study showed that

in OSNs there is a temporal correlation between RA-responsive gene

expression and the production of RARs (Whitesides et al., 1998).

The data from different research groups show that retinoid metabolism is

active in postnatal rodent olfactory tissues. By using in situ hybridization

and immunohistochemical analyses, it was shown that all three RALDH

24

enzymes (RALDH-1, -2, -3) expressed in the adult OE but in different cell

types. RALDH-1 is expressed in both the sustentacular cells of the VL OE and

the cells adjacent to the OSN axons including OECs and fibroblasts located

in the LP. A stronger signal was detected in the VL region in LP as in OE.

Concerning RALDH-2 expression, DMlow-VLhigh pattern was observed in the

fibroblasts. The RALDH3 expressing fibroblasts are located in the most

superficial layer of the LP and are concentrated in the VL region of the OE

(Peluso et al., 2012). In fact, RALDH-1, RALDH-2 and RALDH-3 expressing

cells are mutually exclusive (Niederreither et al., 2002, Norlin et al., 2001,

Peluso et al., 2012). Transcriptomic analyses revealed that the nuclear RA

receptor variants (RARs and RXRs) are expressed in mouse olfactory

mucosa (Zhang, 1999). Likewise RALDHs RA binding proteins CRABP1 and

CRABP2 expression is mutually exclusive in the postnatal rodent OE.

Immunohistochemical analyses revealed that CRABP1 is located in

immature OSNs while CRABP2 is located in a subset of proliferating basal

cells. CRABP2 expression in the olfactory epithelium is extremely interesting

since this protein is found in the adult tissues with ongoing differentiation

such as; skin, testes, ovarian and uterine tissue (Bailey and Siu, 1990, Eller et

al., 1994, Zheng et al., 2000b). In these tissues CRABP2 probably promote

RA-dependent differentiation (Bucco et al., 1997). On the other hand

CRABP1 is expressed broadly in many tissues where it might participate in

discarding the excess amount of RA by interacting with catabolic enzymes,

such as Cyp26s (Napoli, 1999).

There are many in vitro and in vivo evidences supporting the importance

of RA for maintenance and regeneration of OSNs. RA administration affects

the olfactory precursor cell which increases in vitro outgrowth of axons

(Haskell and LaMantia, 2005, Whitesides et al., 1998). Also, in VAD rats, it

was shown that the number of mature OSNs that express OMP declined

while proliferating basal cell population is increased (Asson-Batres et al.,

2003). Furthermore, our group analyzed the importance of RA signaling for

survival of OSNs by phenotypically characterizing a transgenic mouse line,

which expresses dominant negative RAR (dnRAR) under transcriptional

control of OMP. OMP-dnRAR mice RA binds to dnRAR/RXR complexes that

are incapable to activate RA-regulated genes in all postmitotic OSNs. The

data showed that the inhibition of RAR-mediated RA signaling leads the

death of mature OSN population, whereas death does not induce a

compensatory increase in neurogenesis (Hagglund et al., 2006). Altogether,

these findings suggest that RA signaling is crucial for survival of OSNs and

maintenance of correct olfactory sensory map.

25

APP PROCESSING

The amyloid precursor protein (APP) has been the core of AD research

since its discovery approximately 30 years ago. APP processing produces the

β-amyloid (Aβ) peptide which plays a critical role in AD pathogenesis. AD is

the most common cause of dementia among the elderly people. AD patients

suffer from memory loss and their cognitive abilities are progressively

diminished. Deposition of β-amyloid plaques, neurofibrillary tangles in the

brain and neurodegeneration are the pathological hallmarks of AD (Hardy

and Selkoe, 2002).

Mammalian APP family of proteins are highly expressed in the neurons

located in the brain but can be found in other tissues as well. APP is an

integral membrane protein and can be processed by two different pathways

called amyloidogenic pathway and non- amyloidogenic pathway. β-amyloid

is a peptide of 38-43 residues that is generated from APP processing by

amyloidogenic pathway containing β -and γ- secretase enzyme activities.

First, the β-secretase (BACE1) cleaves APP at the Aβ peptide N-terminus and

the soluble ectodomain APPsβ is released. Later on, the γ-secretase cleaves

the remaining membrane anchored part of the protein generating the Aβ

peptide. On the other hand APP can be processed by non-amyloidogenic

pathway containing α -and γ- secretase enzymes. In this pathway, APP is

cleaved within the Aβ peptide by the α-secretase to produce APPsα which is

soluble. Afterwards, the P3 fragment is released as a result of the γ-secretase

(Vassar, 2004) (fig. 10).

26

Figure 10. Amyloid precursor protein (APP) processing. The extracellular

domain of APP is cleaved by either α secretase which leads to non-amyloidogenic

pathway (on the right) or by β secretase (BACE1) which leads to amyloidogenic

pathway (on the left) seen in Alzheimer´s disease. In both of the pathways large N-

terminal ectodomains (APPsβ after β cleavage and APPsα after α cleavage) are

released the remaining C-terminal membrane-attached part of APP is cleaved by γ

secretase. In non amylodic pathway this cleavage leads to generation of a small

peptide called P3, while the cleavage in amylodic pathway result in β amyloid (Aβ)

peptide formation. Aβ domain is shown in red.

Functions of APP Processing

Besides its pathological role in AD, APP itself and APP-derived peptides

have many physiological functions under normal conditions. For instance it

has been shown that APPsα promotes neurite outgrowth, synaptogenesis and

cell adhesion (Gakhar-Koppole et al., 2008, Mattson, 1997). In vivo studies

demonstrated that APPsα promotes learning and memory in rodents by

inducing NMDA receptor- mediated currents and increased long term

potentiation (Taylor et al., 2008). On the other hand it has been reported

that APPsβ acts as a ligand for the death receptor 6 (DR6), a member of the

tumor necrosis factor (TNF) receptors superfamily that induces caspase-6

mediated death. This death pathway is involved in pruning of axon branches

in synapses during the development of both motor neuron and retinal axons

(Nikolaev et al., 2009).

As mentioned earlier and depicted in fig. 10, the β secretase, known as β-

site APP cleaving enzyme 1 (BACE1) is necessary for the production of Aβ

peptide which is associated to AD pathology. Studies with BACE1 knockout

mice revealed that this deletion is not lethal and does not affect the fertility

(Roberds et al., 2001). Nevertheless, when those BACE1 deficient mice are

crossed with APP transgenic mice that develop amyloid plaques, Aβ

production is attenuated and Aβ dependent memory deficits are no longer

observed (Laird et al., 2005, McConlogue et al., 2007). Therefore BACE1 is

considered as a key therapeutic target for AD. The physiological functions of

BACE1 are still unclear. Phenotypes observed in BACE1 deficient mice such

as hypomyelination, seizures and axonal mistargeting suggest that BACE1

may be involved in the processing other substrates than APP. BACE1 is

concentrated in axonal termini and it has been proposed that this

localization is functionally relevant (Yan and Vassar, 2014). Moreover, a

proteomics study displayed that axon guidance molecules are substrates of

BACE1 (Hemming et al., 2009). It has been found that BACE1 deficiency

result in axon guidance defect in olfactory system and hippocampus (Cao et

al., 2012a, Rajapaksha et al., 2011, Hitt et al., 2012). Vassar and Albers

27

groups independently reported that correct OR-specific glomerular

formation is disturbed in both BACE1 deficient and Aβ overexpressing mice.

Furthermore, Vassar and his colleagues showed that the axonal mistargeting

defect by BACE1 inhibition also takes place in hippocampus. Later, it was

discovered that neuronal cell adhesion molecule, CHL1 is not processed in

BACE1 knockout mice. This result could explain the mistargeting phenotype

in OB and in hippocampus (Hitt et al., 2012).

The link between APP and neurodegeneration is still unknown. Role of β-

amyloid plague in neuronal loss is debatable and still not clear. Even though

β-amyloid deposit is considered as a hallmark of AD, the cognitive

impairment does not correlate with the amount of plaques (Hardy and

Selkoe, 2002). It is important to note that olfactory dysfunction occurs

during AD onset and APP processing is involved in OSNs axonal targeting.

Belluscio and colleagues reported that the in mouse, OSNs that are

overexpressing humanized APP that contain familial AD mutation (hAPP)

also express activated caspase-3 apoptotic marker despite the absence of

extracellular β-amyloid plaques. This neurodegeneration caused error in the

olfactory neuronal circuitry and impaired olfactory behavior (Cheng et al.,

2013). These finding may explain the link between olfactory dysfunction and

AD.

Another interesting observation is that APP processing is regulated by

neuronal activity. Ex vivo studies of hippocampal slices suggests that

neuronal activity promotes the endocytosis of surface APP, enhancing the

accessibility of APP to BACE1 and γ-secretase in endosomes. Also, acute

application of oligomeric human Aβ peptide to hippocampal cell cultures

alters LTP and LTD (Li et al., 2011). However, clear in vivo data of neural

activity-dependent endogenous APP processing are lacking. Recently it has

been shown by naris occlusion experiments that neuronal activity suppress

BACE1 expression in OSNs in mouse olfactory system (Cao et al., 2012a).

These findings suggest a link between APP processing and neuronal activity

but the mechanism behind is elusive.

28

AIMS

During my PhD studies, I have studied the olfactory sensory map in

genetically modified mice. Previous results obtained in the project have

shown that locally produced RA plays an important role for cell survival of

chemosensory neurons in olfactory neuroepithelia in the nose of postnatal

and adult mice. The overall aim has been to unravel the molecular links that

connects RA signaling with neuronal activity-dependent gene expression,

cell survival and formation of precise connectivity. A specific focus has been

to understand the relationship between spatial differences in vitamin A

metabolism and spatial differences in activity-dependent mechanisms with

importance for the functional division of the olfactory sensory map along its

DM-VL axis.

The Specific Aims were to:

Identify and characterize a previous unanticipated crosstalk between

RAR signaling and basal neuronal activity that regulates cell survival as

well as formation and maintenance of functionally precise connections

between peripheral chemosensory neurons and target neurons in the

brain. (Paper I, published article).

Identify and characterize a novel mechanism by which RA-degradation

mediates stimulus-induced expression of the AD-associated β-secretase

(BACE1) enzyme, in a graded manner along the DM-VL axis of the

olfactory sensory map. (Paper II, published article).

Identify and characterize a novel mechanism by which stimulus-induced

RA-degradation differentially mediates neurogenesis and pattern

formation of response properties (OR gene choice) along the DM-VL axis

of the olfactory sensory map. (Paper III, manuscript)

29

RESULTS and DISCUSSIONS

In this section, the results from my thesis work are presented in three

separated parts that corresponds to three different manuscripts. Each

manuscript presentation starts with a summary of the results that is followed

by a discussion considering both my results and the published data in the

literature. The figures that are referred in each section can be found in the

corresponding manuscript.

Overview

The first manuscript that was published in the FASEB Journal in 2012,

describes a novel homeostatic feedback mechanism involving retinoic acid

metabolism and olfactory signal transduction pathway. We showed that this

feedback mechanism affects the maintenance of precisely connected OSNs in

olfactory sensory map.

The second manuscript that was published in Brain Structure and

Function Journal in April 2014, presents the data obtained when I

investigated the link between the neuronal activity-dependent RA

metabolism and BACE1 expression in OSNs.

Finally in the third manuscript I addressed the role of neuronal activity–

dependent degradation of RA for both the neurogenesis and cell specification

in postnatal and adult olfactory system.

The effect of RAR signaling on homogeneous glomeruli formation and maintenance.

(Paper I)

During formation and maintenance precise connectivity within a sensory

map, the principles and mechanisms that connect innate determinants (i.e.

extracellular signaling molecules) and neuronal activity-dependent plasticity

are poorly understood. One signaling molecule that regulate both neuronal

plasticity in adults and neural circuit formation during development is all-

trans retinoic acid (RA) (Maden, 2007). RA regulates gene expression via

heterodimers of nuclear transcription factors, retinoic acid receptors (RARs)

and retinoid X receptors (RXRs). To address the role of RA signaling in

OSNs, we generated mice that express a dominant negative RAR

(hRARα403) transgene selectively in postmitotic OSNs. The dnRAR

transgene which does not encode the AF2 ligand-dependent transactivation

domain, forms inactive dnRAR-RXR heterodimers and inhibits transcription

30

mediated by all RAR isoforms. We use a 10 kb DNA fragment that contains

the promoter of the olfactory marker protein (OMP) gene to drive the

transgene expression selectively into both mature and immature OSNs.

Previous work has shown that RA-responsive element (RARE) - dependent

transcription in OSNs is inhibited by OMP promoter-driven dnRAR

expression (Hagglund et al., 2006).

Analyses of the transgenic dnRAR mice Kirrel-2 expression was

downregulated. Kirrel-2 is a cell adhesion molecule that regulates odorant

receptor (OR) specific axonal convergence. We also observed increased

expression of Nrp-1 which plays role in repulsive interactions with both glial

cells and other axons in the nerve layer. Altered levels of Kirrel-2 and Nrp-1

correlated with increased number of heterogeneous glomeruli that are

formed incorrectly by axons of OSNs expressing different ORs. Both Kirrel-2

and Nrp-1 are involved in correct axonal targeting and their expressions are

regulated by OR-mediated signaling pathways (Imai and Sakano, 2009).

Previous analyses of CNGA2 deficient mice as well as sensory deprived mice

have shown that Kirrel-2 is regulated by neuronal activity in a CNG channel-

dependent manner. Kirrel-2 expression was downregulated both in the OE of

CNG channel deficient mice and in OE ipsilateral to the closed naris

(Serizawa et al., 2006) .

Importantly, we found a mechanistic connection between RAR signaling

and activity-dependent regulation of Kirrel-2. Expression of dnRAR reduces

the expression of the CNG channel in OSN cilia and axons. In addition the

results indicate that this perturbed mechanism in the transgenic mice

leading to cell death and glomeruli that are incorrectly innervated by axons

of more than one OR identity. We suggested that formation of heterogeneous

glomeruli in dnRAR mice is a consequence of altered CNG channel activity.

According to our data, reduced RAR-mediated signaling triggers

accumulation of heterogeneous glomeruli which has more than one OR

identity. It is likely that dnRAR interferes with Kirrel-2 and Nrp-1 expression

and thereby inhibits the mechanism controlling formation and maintenance

of OR-specific glomeruli.

Interestingly, increase in glomerular heterogeneity coincides with an

increase in OSN death in the second postnatal week. Several studies indicate

that OR populations that are activated by odorants have a survival advantage

(Zhao and Reed, 2001, Zou et al., 2004, Santoro and Dulac, 2012, Watt et al.,

2004). It is known that heterogeneous glomeruli are eliminated by an

activity-dependent mechanism during the first 2 months old postnatal life

(Kerr and Belluscio, 2006, Zou et al., 2004). We observed an increased

number of heterogeneous glomeruli in postnatal dnRAR mice. One

possibility is that the increased death of mature OSNs results in decreased

competition due to a reduced total number of OSNs. Moreover, the lack of

increased neurogenesis in response to the increased death of mature OSNs

31

may underlie the maintenance of heterogeneous glomeruli in dnRAR

transgenic mice. It has been shown that inhibition of OSN turnover following

sensory deprivation by naris occlusion inhibits elimination of heterogeneous

glomeruli (Zou et al., 2004).

RAR-mediated survival of OSNs (Paper I)

During integration of a neuron into a circuitry, synaptic input induced-

activity together with spontaneous neuronal activity support neuronal

survival (Lin et al., 2010). The analyses of unilateral naris closed mice show

that sensory deprivation does not have any effect on dnRAR- induced Casp-3

activation. This result suggests that dnRAR-mediated death does not depend

on odorant stimulation. We propose that increased death in dnRAR is a

consequence of reduced excitability due to a decreased number of CNG

channel. In a recent study, the authors analyzed the phenotype of OMP-

Kir2.1 mice that have OSNs overexpressing a hyperpolarizing potassium

channel (Kir2.1). Kir-2.1 overexpression, leads to a decreased excitability of

OSNs and an increased death of mature OSNs. Moreover , the increased

death of OSNs in OMP-Kir2.1 mice do not result in an increased

compensatory neurogenesis (Zhao et al., 2013). This phenotype is similar to

what is observed in the olfactory epithelium of OMP-dnRAR mice. This

phenotypic similarity between OMP-Kir2.1 and dnRAR mice most likely

relates to the fact that Kir2.1 overexpression as well as CNG reduced

expression in dnRAR mice hence decrease OSN excitability.

We propose that decreased number of CNG channels in an OSN makes the

neuron more dependent on the OR activity to keep Ca+2 levels necessary for

the regulation of many enzymes and transcription factors which are

necessary for survival (Greer and Greenberg, 2008). Defect RAR signaling

may reduce the CNG channel activity of most of the OSN subpopulations to a

level that is below the threshold necessary for the survival. This may create a

competition between the subpopulations of OSNs which triggers apoptosis of

the neurons that lost the competition. We detected high levels of activated

Casp-3 in the glomeruli that express relatively low levels of Kirrel-2.

Previously it was published that Kirrel-2 expression is a direct readout of OR

dependent CNG channel activity (Serizawa et al., 2006). We propose that

OSNs mediating low odor-independent CNG channel activity in dnRAR mice

are the “losers” in the competition for survival and they are eliminated.

C

32

Neuronal activity- regulated RA metabolism

(Paper I) By In situ hybridization analyses of unilateral naris closed mice as well as

CNGA2 deficient mice, we showed that RA catabolizing enzyme Cyp26B1 is

regulated by neuronal activity in a CNG-dependent manner. This result

together with RAR mediated CNG expression; suggest a feedback loop that

may be a mechanism for homeostatic plasticity, i.e. adaptation to global

inhibition of neuronal activity. In the absence of odor stimulation, OSNs may

increase cell–intrinsic excitability by increasing the number of CNG channel

(fig. 11). The overall aim of this regulation is to survive and stay connected

with the right targets in the absence of stimulating odorants. Thus, we

suggest that correct axonal convergence and survival of the OSNs depend on

RA when there is no odor stimulation. We propose that in the absence of odor stmulation (during infection, for

instance), RA is involved in a homeostatic control mechanism to preserve

neuronal activity to a basal level. This function of RA in OSNs is very similar

to what has been described in RA-dependent homeostatic plasticity in

hippocampal neurons. An RA and RARα-dependent AMPA receptor

expression which strengthens the synapses is promoted by inhibition of

activity in these neurons by tetrodotoxin (Aoto et al., 2008).

33

Figure 11. Model of RA-CNG channel-Cyp26B1 Gene Regulatory feedback

loop. When odor stimulation is absent (when you caught a cold), RA degradation

by Cyp26B1 is reduced due to low channel activity. Increased level of RA and RAR

mediated signaling then leads to increased number of the CNG channel which will

result in increased cell intrinsic excitability. Also, the expression of CNG

downstream genes, such as Kirrel-2 which is involved in correct axonal targeting

will be maintained. According to this model, in the absence of odor stimulation,

RAR-dependent gene expression is involved in both cell survival and maintenance of

correct OR-specific axonal convergence.

The link between neuronal activity-regulated Cyp26B1 and BACE1

(Paper II)

Many neurodegenerative disorders arise from both a defect activity-

dependent neuroplasticity and a failure of maintaining connectivity.

Interestingly, the inability to detect and discriminate odors is one of the early

symptoms of Alzheimer´s disease. Recently, it has been shown that β site

amyloid precursor protein (APP)–cleaving enzyme 1 (BACE1) and the

neurotoxic Aβ peptide influence the mechanisms regulating olfactory

sensory map assembly (Cao et al., 2012a, Cao et al., 2012b, Rajapaksha et al.,

2011). Another interesting observation is that BACE1 mRNA has a DMlow-

VLhigh expression gradient in the olfactory epithelium. Also, BACE1 protein

level in the axon termini follows DMlow-VLhigh gradient in the OB. Moreover,

in a study using unilateral naris occluded mice, it was showed that neuronal

activity inhibits BACE1 expression in OSNs. Thus, BACE1 expression is

inversely regulated by neuronal activity. Several studies have shown that

inhibition of neuronal activity differentially disrupts the formation and/or

maintenance of OR-specific glomeruli in a DM>VL manner (Nakashima et

al., 2013, Yu et al., 2004, Zheng et al., 2000a). Similarly, such a pattern is

also observed in BACE1 deficient mice. Altogether, these observations

strongly suggest that an activity dependent expression gradient of BACE1

might influence precise OR-specific connectivity and neuronal survival. In

the second study we aimed to identify the mechanisms behind neuronal

activity-dependent BACE1 expression.

By immunohistochemical analyses, we detected the RA degrading enzyme

Cyp26b1, in the cytoplasm, dendrites and axons of OSNs that are scattered

mostly in the apically located mature OSNs. In postnatal mice, by in situ

hybridization analyses, we showed that Cyp26B1 is expressed in a DMhigh-

VLlow manner, which is the inverse to the gradient of BACE1 expression.

34

Immunohistochemical analyses of male CNGA20/− mice, with OSNs that are

all deficient in CNGA2 function, showed inverse levels of expression of

Cyp26B1 and BACE1 in cell bodies and glomeruli, respectively. Results from

immunohistochemical analyses of unilateral naris occluded mice showed

that Cyp26B1-positive OSNs were not present in OE ipsilateral to the closed

naris. Taken together these results indicated that Cyp26B1 and BACE1

protein levels in OSN are regulated in an inverse manner in response to

sensory deprivation and CNG channel deficiency.

In order to determine the onset expression of Cyp26B1, we analyzed the

OE of embryonic and newborn (PD1) mice. Cyp26B1 expression was first

evident at PD1 mice. This result was in agreement with naris closure

experiment, and indicated that the onset of Cyp26B1 expression correlates

with the onset of respiration and passage of air/odorants through the nasal

cavity.

In light of the inverse spatial and neuronal activity-dependent regulation

of BACE1 and Cyp26B1 expression, we hypothesized that BACE1 expression

in OSNs is regulated by neuronal activity-dependent and Cyp26B1-mediated

RA degradation. To address this hypothesis, we analyzed whether the

overexpression of Cyp26B1 reduces BACE1 expression. Specifically, we used

the OMP promoter to generate transgenic mouse line (OMP-Cyp26B1) that

constitutively express high levels of Cyp26B1 in OSNs. In situ hybridization

analyses showed that in the OMP-Cyp26B1 mice, both DM and VL OSNs

expressed elevated Cyp26B1 mRNA levels compared to littermate controls.

In accordance with the prediction we found that BACE1 mRNA expression

was abolished in the transgenic mice. Moreover, we found a downregulation

of BACE1 protein in the glomerular layer in the transgenic mice.

Quantifications of relative fluorescence signal intensities showed a

significant downregulation in axon termini. Similarly, we observed the

reduction in BACE1 protein levels in glomerular layer of OMP-dnRAR mice.

Thus, we confirmed that RA signaling regulates BACE1.

The spatial and scattered expression in the OE of Cyp26B1 resembles the

expression pattern of individual ORs. This similarity together with onset of

Cyp26B1 expression after birth suggest that a temporal asynchrony of

Cyp26B1 expression in OSN subpopulations that respond to specific

environmental odors causes the scattered and spatial pattern of Cyp26B1

expression.

Under normal conditions BACE1 is expressed in both mature and

immature OSNs. It is puzzling how scattered Cyp26B1-positive OSNs can

downregulate BACE1 expression in a much larger population of immature

and mature OSNs. One working model is that Cyp26B1 degrades diffusible

RA produced by RALDH-positive cells in the olfactory mucosa. This

Cyp26B1 may function as a “RA sink” and alter gene expression in

neighboring cells in a non-cell-autonomous manner.

35

In summary we have identified a novel link between neural activity

dependent Vitamin A metabolism and BACE1 expression. This link is

important for the formation and maintenance of precise connectivity during

postnatal development. Mutually exclusive expression and inverse response

to neuronal activity of Cyp26B1 and BACE1, suggest that RA probably

control the genes that are involved in conversion of sensory experience into

changes in connectivity.

The importance of zonal gradient of RA bioavailability for OR expression patterns.

(Paper III)

The formation and maintenance of the olfactory sensory map depends on

neuronal activity and follows the “use it or lose it” principle. Based on this

principle, competition between active and inactive neurons results in

elimination of the inactive or, silent neurons from the system. Molecular

pathways and extracellular signals that are involved in neuronal activity–

dependent competition between neurons for correct connectivity and cell

survival are largely unknown.

Spatiotemporal regulated RA metabolism has been shown to play a

significant regulatory role for pattern formation of the embryonic spinal

cord, hindbrain and retina (Glover et al., 2006, Sen et al., 2005). In adults

spatial differences in RA metabolism remains in the dentate gyrus of

hippocampus, spinal cord and thalamus (Leung et al., 2012, Goodman et al.,

2012, Sen et al., 2005). The function of this uneven distribution of RA in

these regions is elusive. A study of hippocampus suggests that spatial

differences in RALDH-2 and Cyp26B1 influence adult neurogenesis

(Goodman et al., 2012).

OR expression patterns in the olfactory epithelium are divided into 4

zones along the DM-VL axis. How OR zones are formed and maintained is

not known. One possibility is that each zone contains progenitor cells that

generate OSNs which express a defined set of ORs. We and others have

identified genes that are expressed asymmetrically in a manner that

correlates with the zonal organization. Such genes are involved in RA

synthesis (RALDH-1, -2,-3) and degradation (Cyp26B1). Cyp26b1 and RA

synthesizing enzymes are expressed in counter gradient in olfactory mucosa.

In paper III, we investigated the expression patterns of selected OR genes

by using in situ hybridization in 2 weeks old OMP-Cyp26B1 mice.

Quantifications of different OR-specific OSNs subpopulations showed that

the frequencies of different ORs were altered in OMP-Cyp26B1 compared to

control mice. The results indicate that the number of DM OSNs (Z1 and Z2)

increases, whereas VL OSNs (Z4) number reduces (Fig 2A, B). Also by using

36

Z1 marker, NQO1, we compared the length of Z1 region (NQO1 positive) to

Z2-4 region (NQO1 negative). We observed a significant increase of Z1 length

while Z2-4 length decreased. These results indicate that reduced RA

bioavailability “dorsalize” the olfactory epithelium. Abolishment of DMlow-

VLhigh expression gradient of BACE1 in OMP-Cyp26B1 mice is in line with a

dorsalized OE in the OMP-Cyp26B1 mice described earlier (Paper II).

Recently, an analysis of OR expression following one month sensory

deprivation by unilateral naris occlusion has revealed that OR frequencies

are altered in an inverse way to what we observe in OMP-Cyp26B1 mice

(Zhao et al., 2013). According to this publication, sensory deprivation

increases the Z4OR frequencies in the OE while Z1-3 OR frequencies are

either decreased or not affected. The inverse result regarding OR expression

obtained from unilateral naris occluded and OMP-Cyp26B1 mice can be

explained by the fact that sensory deprivation attenuates Cyp26B1

expression.

Cyp26B1 gain of function results in increased OSN turnover in the VL OE.

(Paper III)

The neurogenesis was increased in 2 weeks old transgenic mice. More

precisely, both the number of proliferating cells and immature OSNs were

increased. We also detected increased number of Casp-3 positive cells in

transgenic mice. Both proliferation and death is more pronounced in the

neurogenic VL region. These results are consistent with a model in which

Cyp26B1-mediated RA degradation results in an increased neurogenesis,

accumulation of immature OSNs and increased cell death. Interestingly, the

OSNs that reside in VL part of olfactory epithelium diminish over time. Six

months old mice epithelium in that region is lacking of OSNs. In light of the

increased turnover, we propose that inhibited RA signaling leads to

depletion of adult neuronal stem/progenitor cell pool in VL region of OE. As it was mentioned in introduction, RA inhibits proliferation in many

different cell types. Actually, some medical treatments have taken advantage

of the anti-proliferative properties of RA to treat diseases such as leukemia,

breast and lung cancers (Clarke et al., 2004, Crowe, 2002). It has been

shown the exposure to RA inhibits proliferation of progenitors in the

subgranular zone of the hippocampus where adult neurogenesis take place

(Crandall et al., 2004). Also, results from Vitamin A deficient (VAD) rats

(Asson-Batres et al., 2003) and RALDH3 deficient mice embryos (Paschaki

et al., 2013) support the idea that RA inhibits proliferation in the OE. An

important question is: How expression of Cyp26B1 in the OSNs stimulates

the progenitor cell proliferation? As we discussed in the second paper,

37

Cyp26B1 can act as a sink for diffusible RA and increased cell division might

be mediated by a non-cell-autonomous mechanism. Such a cell-

nonautonomous function of Cyp26b1 enzymatic activity has been suggested

for RA-regulated dentate gyrus neurogenesis and for embryonic nervous

system pattern formation (Maden, 2007, Goodman et al., 2012)

Cyp26B1 overexpression inhibits the OSN differentiation

(Paper III)

We detected the reduction of Z4 OR expression before Casp-3 activation

and without reduction of OSNs in VL region of the olfactory epithelium. In

addition we found an increased number of SCG10-positive immature OSNs

as well as an increased total number of cells. These results suggest that the

number of OR-positive VL OSNs is reduced in the epithelium with an

increased number of SCG10-positive immature OSNs. This indicates that the

transgenic mice have an increased number of immature OSNs that do not

express any OR since they are blocked at a differentiation stage prior to OR

expression. Many in vitro studies revealed that RA is involved in the

induction and maintenance of neural differentiation (Maden, 2007).

Remarkably, we found that adenyl cyclase 3 (ACIII), which is a component of

the signal transduction pathway, forms a DMhigh to VLlow expression gradient

in the transgenic mice. This is an important observation since recent studies

have shown that ACIII expression promotes both OSN differentiation and

stabilization of OR gene choice (Lyons et al., 2013, Dalton et al., 2013). Thus,

reduced expression of ACIII in VL OSNs in OMP-Cyp mice support the idea

that the differentiation of OSNs is inhibited at a stage that is associated with

onset of OR expression.

Odor-induced Cyp26B1 expression mediates neuronal competition for cell survival

(Paper III)

In contrary to Z4, we found increased apoptosis that was followed by

increased proliferation in Z1. This finding suggests that increased OSNs

turnover in Z1 at one month age may be a consequence of reduced OSNs

lifespan. The increased/unaltered Z1 and 2 OR frequencies may be an

outcome of odor-dependent competition for cell survival. A recent study

showed that odor-stimulation increase the life span of OSNs by

downregulation of histone variant H2BE (Santoro and Dulac, 2012). Our

current hypothesis regarding the death observed in Z1 of OMP-Cyp26B1

mice is that odorant-induced Cyp26B1 can act as a sink for RA in the local

38

environment which may affect the survival of neighboring OSNs. It is

reasonable to consider that the OSNs that are activated (stimulated by an

odorant molecule) have longer lifespan due to e.g. downregulation of H2BE.

Also, activated OSNs express Cyp26B1 which leads to RA inactivation in the

surrounding extracellular space. This will cause the death of unstimulated

OSNs that require RA for survival (see paper I). Altogether, these

observations suggest that odor-induced Cyp26B1 may induce neuronal

competition for survival and compensatory neurogenesis.

The data presented in paper III suggest that RA regulates progenitor

proliferation, OSN differentiation and creates competition for cell survival

(fig. 12). We propose that DMhigh-VLlow expression pattern of Cyp26B1 is to

spatially restrict the influence of odor exposure on OR expression.

Accordingly, activity dependent Cyp26B1 expression facilitate experience-

driven OR expression in the DM olfactory epithelium and at the same time

inhibits genetically determined OR gene choice in the VL olfactory

epithelium. In addition, odor-induced Cyp26B1 will result in reduced life

time of unstimulated OSNs that require RA to maintain their intrinsic

excitability to life promoting levels (Paper I). The consequence of spatial

Cyp26B1 and RALDHs expressions is that the effect of RA on neurogenesis

will be most pronounced in Z4 while activity dependent competition for cell

survival will be most pronounced in Z1.

Figure 12. Proposed model for the functions of RA in OSNs. RA can (1)

inhibit proliferation of progenitor basal cells, (2) induce differentiation by regulating

ACIII expression which is involved in the OR gene choice mechanism and (3) support

the cell survival. According to our model, the effect of RA on proliferation and

differentiation is most pronounced in Z4, while activity dependent competition for

cell survival is most pronounced in Z1.

39

Differences between OMP-dnRAR and OMP-Cyp mice (Paper I, II, III)

The results presented in paper III show that the OE phenotype of OMP-

Cyp26B1 mice differs from the phenotype observed in OMP-dnRAR

transgenic mice. The differences in OE phenotype of OMP-Cyp26B1 and

OMP-dnRAR mice may look contradictory at first due to the fact that both

transgenes inhibits RA-dependent gene regulation and their expressions are

controlled by the same promoter. However increased RA degradation may

have more pleitropic effects and may also influence the regulation of genes

that are not RAR-mediated. For instance, it is plausible that the RA-

regulated transcription mediated by PPARβ/δ is affected. In fact PPARβ/δ

and FABP5 (binding protein that delivers RA to PPARβ/δ) are expressed in

the OE (data not shown). A previous study of OMP-dnRAR mice have shown

that mature OSNs die in all OR zones without any compensatory increase in

neurogenesis (Hagglund et al., 2006). As previously discussed, this

phenotype is similar to the one observed in the olfactory epithelium of

OMP-Kir2.1 mice that overexpress the hyperpolarizing potassium channel

(Kir2.1). This similar phenotype in OMP-Kir2.1 and dnRAR mice can be

explained by the fact that both reduced CNG channel expression in dnRAR

mice and Kir2.1 overexpression decrease OSN excitability (Zhao et al., 2013).

In contrary to a global death of OSNs in all OR zones, we observed different

alterations in OR expression in different OR zones in OMP-Cyp26B1 mice.

Zone-specific changes in OR expression has also been described in the OE

following sensory deprivation by naris closure. Reduced sensory input

increases the Z4OR populations and decreases or do not affect Z1-3 OR

populations(Zhao et al., 2013). In addition, sensory deprivation inhibits

neurogenesis (Zou et al., 2004). In OMP-Cyp26B1 mice that overexpress

Cyp26B1, we observe altered OR frequency phenotype in an inverse way;

reduced number in Z4 ORs and increased or not changed number in Z1 and

2 ORs. In contrary to sensory deprivation we detected an increased

neurogenesis in OMP-Cyp26B1 mice. On the other hand in OMP-dnRAR

mice, increased neurogenesis is not evident (Hagglund et al., 2006,

Hornberg et al., 2009). Previously our group published that in OMP-dnRAR

mice Cyp26B1 is inhibited (Hagglund et al., 2006). However in OMP-

Cyp26B1 mice Cyp261 is overexpressed. Unlike dnRAR, gain of Cyp26B1

activity may regulate gene expression in neighboring cells in a non-cell-

autonomous manner. In fact, lack of compensatory neurogenesis after death

of mature OSNs in dnRAR can be explained by inhibition of Cyp26B1

expression and increased RA concentration.

40

CONCLUSIONS

The findings presented in this thesis can be summarized as follows:

Inhibition of RAR-mediated transcription in OSNs is associated with reduced expression of an important component of the stimulus transduction pathway, the CNG channel. This results in increased death of mature OSNs and errors in OR-specific connectivity with target neurons in the olfactory bulb. The increased death of mature OSNs may be a consequence of reduced intrinsic excitability and/or reduced influx of Ca2+ ions.

Expression of the RA catabolic enzyme Cyp26B1 in OSNs is

positively regulated by RAR-mediated transcription as well as

sensory stimulation in a CNG channel-dependent manner. These

findings show that neuronal activity and local vitamin A metabolism

are parts of novel regulatory feedback loop controlling precise

connectivity and neuronal survival. The feedback loop may be a form

of homeostatic plasticity in response to global changes in neuronal

activity.

Cyp26B1 and BACE1, an enzyme that is implicated in Alzheimer´s disease, are inversely regulated by sensory stimuli as well as CNG channel activity in OSNs. Cyp26B1 inhibits BACE1 expression. Cyp26B1 expression is switched on at birth in a DMhigh-VLlow gradient that shapes BACE1 expression into a DMlow-VLhigh counter gradient. Taken together these results reveal a novel neuronal activity-dependent mechanism by which sensory stimuli can shape spatial gene expression via altered RA bioavailability.

Increased Cyp26B1 expression in OSN stimulates turnover of OSNs

during adult neurogenesis. The effect of Cyp26B1 on OSN turnover

is driven by non-cell-autonomous mechanisms such as increased

death of inactive OSNs and increased progenitor cell proliferation.

In addition, increased Cyp26B1 inhibits OSN differentiation at a

stage associated with OR gene choice. The combined effects of

increased Cyp26B1 expression results in an altered repertoire of

expressed OR genes. A probable function of graded and activity-

dependent Cyp26B1 expression may be to facilitate experience-

driven diversification of OSNs in the DM epithelium and at the same

time inhibit OR gene switching in the VL olfactory epithelium.

41

ACKNOWLEDGEMENTS

First, I would like to express my special appreciation and thanks to my

supervisor Staffan Bohm. Staffan, you have been a terrific mentor for me.

You taught me how to be optimistic and enthusiastic about research. I would

like to thank you for your listening and for giving me confidence about my

future in science. I would gladly admit that the trust you put in me, kept me

going during these years. Now I know that satisfaction after success is

priceless and I would like to pursue my professional life as a scientist. I hope

after 20 years from now when I talk about science, my eyes sparkle with

excitement like yours… Also, I would like to thank my co-supervisor Anna

Berghard. You taught me how to be critical and realistic about research. I

cannot count how many times you made get my feet back on the ground after

I got overexcited by a great result. I enjoyed every single conversation

(scientific/private) with you and I will always keep your advices in mind.

I would like also to thank Sara Wilson and Paolo Medini for nice

discussions in the journal club meetings and making me curious about

different neuroscientific questions. Jonathan Gilthorpe, thank you for

your guidance in photomicrography and signal intensity quantifications.

Hans Wolf-Watz and Åke Forsberg, thank you for your support and

care.

The present and former group members of Bohm and Berghard´s labs:

Viktoria, the queen of the lab, the master of immunostainings. I assigned

you to be a big sister of mine. You are a great listener and the most “correct”

person I have ever met. You witnessed many of my little harmless crimes and

be my inspiration for the hair styles I tried ;). Sofia, the next generation PhD

student, I have no doubt about that the project is in good hands. It was a big

pleasure to share the lab with you and also to know you as a person. One

more time, thanks for being my private translator, being brave to enjoy

Eurovision Song Contest and for marvellous birthday present! Pauline—the

youngest member of the group, I hope as a project student, you learn many

useful things for your future and I would like to thank you for introducing

stroop waffles to me. Maria H. and Fredrik, thanks for not leaving me

alone with dnRAR and Cyp projects.

Erik, Margarida and Edvin, who shared teaching experience with me.

It was extremely fun to work with you guys! I think we learned as much as

the students ;)

I want to thank the regular members of “the international lunch table”

(the most popular table of Molbiol department): Ala, Teresa, Marylise,

Bernard, Shahab, Nicole, Tomas and Helen. The best time of the day

was with you guys. Thanks to the people with whom I was spending small

breaks and sharing big laughter in the department: Coni, Saskia, Nabil,

42

Anne-Laure, Sveta, Bhupiss. I also really appreciated the little Turkish

community: Kemal, Ummuhan, Tugrul, Sarp and Hasan who helped

me feel less homesick.

Good old friends, the people that have left Umeå but not my heart. The

former “friday fika” members: Lina, Therese, Chaz and Linus. I enjoyed

our movie making sessions, extremely creative parties. Maria L. and

Christina, I wish we could have more time to spend at iksu and on the

dancefloor =). And Stefanie, my closest friend in Umeå. You cannot

imagine how much I missed our coffee time gossips, girly talks, indoor

walkings in the past year. Even you were far away, I always felt your hand on

my back. Thanks for dozens of postcards and gifts that made me feel like a

spoiled kid =)

Frédéric, my talented ghost writer =), I feel extremely lucky to have you

and share a life with you. Professionally and personally I learned many

important things from you. Thanks for all the motivation and inspiration

during my PhD. Je t´aime, askilotam!

Last, I would like to thank to my parents for their physical and

psychological support for 29 years. I hope they will never feel regret all the

sacrifices and exertions they have done for me. Anne, kar kis demeden, beni

yalniz birakmamak icin Umeå´ya defalarca geldin. Özlemisimdir diye

Turkiye´den yufka, gullac, tahin tasidin. Isten, dersten bunaldigimda

alisverise ciktik, gezdik, tozduk. Inan bana seninle gecirdigim zamanin tadini

baska kimsede bulamadim. Beni, sesimin tonundan iyi veya kötu oldugumu

anlayacak kadar iyi taniyan, yerlere göklere sigdiramayacak kadar cok seven,

her zaman dogruya yönlendiren cok zeki ve arslan kadar guclu bir annem

oldugu icin cok mutluyum. Baba, cocukken anlayamadigim kurallarini daha

yeni yeni anlamaya basliyorum. Hatta, gitgide kendimde seni buluyorum

desem yalan olmaz. Doktora sirasinda verdigin tavsiyeler hep kulagimda

cinladi. Senin gibi ileriyi görmeye calistim ve kapiyi hafifce tiklatan firsatlari

degerlendirdim. Edindigim bilgileri son damlasina kadar kullandim,

kendime guvendim, yaparim dedim ve doktoranin sonuna gelmeyi basardim.

Umarim ikinizin de emeklerinizi bosa cikarmam. Sizleri cok seviyorum.

43

REFERENCES ALENIUS, M. & BOHM, S. 1997. Identification of a novel neural cell

adhesion molecule-related gene with a potential role in selective axonal projection. J Biol Chem, 272, 26083-6.

AOTO, J., NAM, C. I., POON, M. M., TING, P. & CHEN, L. 2008. Synaptic signaling by all-trans retinoic acid in homeostatic synaptic plasticity. Neuron, 60, 308-20.

APPEL, B. & EISEN, J. S. 2003. Retinoids run rampant: multiple roles during spinal cord and motor neuron development. Neuron, 40, 461-4.

ASSON-BATRES, M. A., ZENG, M. S., SAVCHENKO, V., ADEROJU, A. & MCKANNA, J. 2003. Vitamin A deficiency leads to increased cell proliferation in olfactory epithelium of mature rats. J Neurobiol, 54, 539-54.

BADER, A., BAUTZE, V., HAID, D., BREER, H. & STROTMANN, J. 2010. Gene switching and odor induced activity shape expression of the OR37 family of olfactory receptor genes. Eur J Neurosci, 32, 1813-24.

BAILEY, J. S. & SIU, C. H. 1990. Purification of cellular retinoic acid-binding proteins types I and II from neonatal rat pups. Methods Enzymol, 189, 356-63.

BAKER, H., CUMMINGS, D. M., MUNGER, S. D., MARGOLIS, J. W., FRANZEN, L., REED, R. R. & MARGOLIS, F. L. 1999. Targeted deletion of a cyclic nucleotide-gated channel subunit (OCNC1): biochemical and morphological consequences in adult mice. J Neurosci, 19, 9313-21.

BEITES, C. L., KAWAUCHI, S., CROCKER, C. E. & CALOF, A. L. 2005. Identification and molecular regulation of neural stem cells in the olfactory epithelium. Exp Cell Res, 306, 309-16.

BELLUSCIO, L., GOLD, G. H., NEMES, A. & AXEL, R. 1998. Mice deficient in G(olf) are anosmic. Neuron, 20, 69-81.

BERGER, J. & MOLLER, D. E. 2002. The mechanisms of action of PPARs. Annu Rev Med, 53, 409-35.

BHASIN, N., MAYNARD, T. M., GALLAGHER, P. A. & LAMANTIA, A. S. 2003. Mesenchymal/epithelial regulation of retinoic acid signaling in the olfactory placode. Dev Biol, 261, 82-98.

BIBEL, M., RICHTER, J., LACROIX, E. & BARDE, Y. A. 2007. Generation of a defined and uniform population of CNS progenitors and neurons from mouse embryonic stem cells. Nat Protoc, 2, 1034-43.

BRADLEY, J., REISERT, J. & FRINGS, S. 2005. Regulation of cyclic nucleotide-gated channels. Curr Opin Neurobiol, 15, 343-9.

BRUNET, L. J., GOLD, G. H. & NGAI, J. 1996. General anosmia caused by a targeted disruption of the mouse olfactory cyclic nucleotide-gated cation channel. Neuron, 17, 681-93.

44

BUCCO, R. A., ZHENG, W. L., DAVIS, J. T., SIERRA-RIVERA, E., OSTEEN, K. G., CHAUDHARY, A. K. & ONG, D. E. 1997. Cellular retinoic acid-binding protein(II) presence in rat uterine epithelial cells correlates with their synthesis of retinoic acid. Biochemistry, 36, 4009-14.

BUCK, L. & AXEL, R. 1991. A novel multigene family may encode odorant receptors: a molecular basis for odor recognition. Cell, 65, 175-87.

BUSHDID, C., MAGNASCO, M. O., VOSSHALL, L. B. & KELLER, A. 2014. Humans can discriminate more than 1 trillion olfactory stimuli. Science, 343, 1370-2.

CAO, L., RICKENBACHER, G. T., RODRIGUEZ, S., MOULIA, T. W. & ALBERS, M. W. 2012a. The precision of axon targeting of mouse olfactory sensory neurons requires the BACE1 protease. Sci Rep, 2, 231.

CAO, L., SCHRANK, B. R., RODRIGUEZ, S., BENZ, E. G., MOULIA, T. W., RICKENBACHER, G. T., GOMEZ, A. C., LEVITES, Y., EDWARDS, S. R., GOLDE, T. E., HYMAN, B. T., BARNEA, G. & ALBERS, M. W. 2012b. Abeta alters the connectivity of olfactory neurons in the absence of amyloid plaques in vivo. Nat Commun, 3, 1009.

CARTER, L. A., MACDONALD, J. L. & ROSKAMS, A. J. 2004. Olfactory horizontal basal cells demonstrate a conserved multipotent progenitor phenotype. J Neurosci, 24, 5670-83.

CHAMBON, P. 1996. A decade of molecular biology of retinoic acid receptors. FASEB J, 10, 940-54.

CHATZI, C., BRADE, T. & DUESTER, G. 2011. Retinoic acid functions as a key GABAergic differentiation signal in the basal ganglia. PLoS Biol, 9, e1000609.

CHEN, N. & NAPOLI, J. L. 2008. All-trans-retinoic acid stimulates translation and induces spine formation in hippocampal neurons through a membrane-associated RARalpha. FASEB J, 22, 236-45.

CHEN, N., ONISKO, B. & NAPOLI, J. L. 2008. The nuclear transcription factor RARalpha associates with neuronal RNA granules and suppresses translation. J Biol Chem, 283, 20841-7.

CHEN, Y. & FLANAGAN, J. G. 2006. Follow your nose: axon pathfinding in olfactory map formation. Cell, 127, 881-4.

CHENG, N., BAI, L., STEUER, E. & BELLUSCIO, L. 2013. Olfactory functions scale with circuit restoration in a rapidly reversible Alzheimer's disease model. J Neurosci, 33, 12208-17.

CHESLER, A. T., ZOU, D. J., LE PICHON, C. E., PETERLIN, Z. A., MATTHEWS, G. A., PEI, X., MILLER, M. C. & FIRESTEIN, S. 2007. A G protein/cAMP signal cascade is required for axonal convergence into olfactory glomeruli. Proc Natl Acad Sci U S A, 104, 1039-44.

CHESS, A., SIMON, I., CEDAR, H. & AXEL, R. 1994. Allelic inactivation regulates olfactory receptor gene expression. Cell, 78, 823-34.

CHO, J. H., PRINCE, J. E., CUTFORTH, T. & CLOUTIER, J. F. 2011. The pattern of glomerular map formation defines responsiveness to aversive odorants in mice. J Neurosci, 31, 7920-6.

45

CLARKE, N., GERMAIN, P., ALTUCCI, L. & GRONEMEYER, H. 2004. Retinoids: potential in cancer prevention and therapy. Expert Rev Mol Med, 6, 1-23.

COCCO, S., DIAZ, G., STANCAMPIANO, R., DIANA, A., CARTA, M., CURRELI, R., SARAIS, L. & FADDA, F. 2002. Vitamin A deficiency produces spatial learning and memory impairment in rats. Neuroscience, 115, 475-82.

CONNELLY, T., SAVIGNER, A. & MA, M. 2013. Spontaneous and sensory-evoked activity in mouse olfactory sensory neurons with defined odorant receptors. J Neurophysiol, 110, 55-62.

CORCORAN, J., SO, P. L. & MADEN, M. 2002. Absence of retinoids can induce motoneuron disease in the adult rat and a retinoid defect is present in motoneuron disease patients. J Cell Sci, 115, 4735-41.

CRANDALL, J., SAKAI, Y., ZHANG, J., KOUL, O., MINEUR, Y., CRUSIO, W. E. & MCCAFFERY, P. 2004. 13-cis-retinoic acid suppresses hippocampal cell division and hippocampal-dependent learning in mice. Proc Natl Acad Sci U S A, 101, 5111-6.

CROWE, D. L. 2002. Receptor selective synthetic retinoids as potential cancer chemotherapy agents. Curr Cancer Drug Targets, 2, 77-86.

DALTON, R. P., LYONS, D. B. & LOMVARDAS, S. 2013. Co-opting the unfolded protein response to elicit olfactory receptor feedback. Cell, 155, 321-32.

DIEZ DEL CORRAL, R. & STOREY, K. G. 2004. Opposing FGF and retinoid pathways: a signaling switch that controls differentiation and patterning onset in the extending vertebrate body axis. Bioessays, 26, 857-69.

DUESTER, G. 2000. Families of retinoid dehydrogenases regulating vitamin A function: production of visual pigment and retinoic acid. Eur J Biochem, 267, 4315-24.

DUGGAN, C. D. & NGAI, J. 2007. Scent of a stem cell. Nat Neurosci, 10, 673-4.

DUPE, V., MATT, N., GARNIER, J. M., CHAMBON, P., MARK, M. & GHYSELINCK, N. B. 2003. A newborn lethal defect due to inactivation of retinaldehyde dehydrogenase type 3 is prevented by maternal retinoic acid treatment. Proc Natl Acad Sci U S A, 100, 14036-41.

EDLUND, T. & JESSELL, T. M. 1999. Progression from extrinsic to intrinsic signaling in cell fate specification: a view from the nervous system. Cell, 96, 211-24.

EGGAN, K., BALDWIN, K., TACKETT, M., OSBORNE, J., GOGOS, J., CHESS, A., AXEL, R. & JAENISCH, R. 2004. Mice cloned from olfactory sensory neurons. Nature, 428, 44-9.

ELLER, M. S., HARKNESS, D. D., BHAWAN, J. & GILCHREST, B. A. 1994. Epidermal differentiation enhances CRABP II expression in human skin. J Invest Dermatol, 103, 785-90.

46

ENDERLIN, V., ALFOS, S., PALLET, V., GARCIN, H., AZAIS-BRAESCO, V., JAFFARD, R. & HIGUERET, P. 1997. Aging decreases the abundance of retinoic acid (RAR) and triiodothyronine (TR) nuclear receptor mRNA in rat brain: effect of the administration of retinoids. FEBS Lett, 412, 629-32.

ETCHAMENDY, N., ENDERLIN, V., MARIGHETTO, A., PALLET, V., HIGUERET, P. & JAFFARD, R. 2003. Vitamin A deficiency and relational memory deficit in adult mice: relationships with changes in brain retinoid signaling. Behav Brain Res, 145, 37-49.

FAHRENHOLZ, F. & POSTINA, R. 2006. Alpha-secretase activation--an approach to Alzheimer's disease therapy. Neurodegener Dis, 3, 255-61.

FEINSTEIN, P. & MOMBAERTS, P. 2004. A contextual model for axonal sorting into glomeruli in the mouse olfactory system. Cell, 117, 817-31.

FIRESTEIN, S. 2001. How the olfactory system makes sense of scents. Nature, 413, 211-8.

FLEISCHMANN, A., SHYKIND, B. M., SOSULSKI, D. L., FRANKS, K. M., GLINKA, M. E., MEI, D. F., SUN, Y., KIRKLAND, J., MENDELSOHN, M., ALBERS, M. W. & AXEL, R. 2008. Mice with a "monoclonal nose": perturbations in an olfactory map impair odor discrimination. Neuron, 60, 1068-81.

GAEMERS, I. C., VAN PELT, A. M., VAN DER SAAG, P. T. & DE ROOIJ, D. G. 1996. All-trans-4-oxo-retinoic acid: a potent inducer of in vivo proliferation of growth-arrested A spermatogonia in the vitamin A-deficient mouse testis. Endocrinology, 137, 479-85.

GAKHAR-KOPPOLE, N., HUNDESHAGEN, P., MANDL, C., WEYER, S. W., ALLINQUANT, B., MULLER, U. & CICCOLINI, F. 2008. Activity requires soluble amyloid precursor protein alpha to promote neurite outgrowth in neural stem cell-derived neurons via activation of the MAPK pathway. Eur J Neurosci, 28, 871-82.

GERMAIN, P., IYER, J., ZECHEL, C. & GRONEMEYER, H. 2002. Co-regulator recruitment and the mechanism of retinoic acid receptor synergy. Nature, 415, 187-92.

GHENIMI, N., BEAUVIEUX, M. C., BIRAN, M., PALLET, V., HIGUERET, P. & GALLIS, J. L. 2009. Vitamin A deficiency in rats induces anatomic and metabolic changes comparable with those of neurodegenerative disorders. J Nutr, 139, 696-702.

GLOVER, J. C., RENAUD, J. S. & RIJLI, F. M. 2006. Retinoic acid and hindbrain patterning. J Neurobiol, 66, 705-25.

GOODMAN, T., CRANDALL, J. E., NANESCU, S. E., QUADRO, L., SHEARER, K., ROSS, A. & MCCAFFERY, P. 2012. Patterning of retinoic acid signaling and cell proliferation in the hippocampus. Hippocampus, 22, 2171-83.

GORDON, M. K., MUMM, J. S., DAVIS, R. A., HOLCOMB, J. D. & CALOF, A. L. 1995. Dynamics of MASH1 expression in vitro and in vivo

47

suggest a non-stem cell site of MASH1 action in the olfactory receptor neuron lineage. Mol Cell Neurosci, 6, 363-79.

GOTTLIEB, D. I. & HUETTNER, J. E. 1999. An in vitro pathway from embryonic stem cells to neurons and glia. Cells Tissues Organs, 165, 165-72.

GRAZIADEI, P. P. & GRAZIADEI, G. A. 1979. Neurogenesis and neuron regeneration in the olfactory system of mammals. I. Morphological aspects of differentiation and structural organization of the olfactory sensory neurons. J Neurocytol, 8, 1-18.

GREER, P. L. & GREENBERG, M. E. 2008. From synapse to nucleus: calcium-dependent gene transcription in the control of synapse development and function. Neuron, 59, 846-60.

GUSSING, F. & BOHM, S. 2004. NQO1 activity in the main and the accessory olfactory systems correlates with the zonal topography of projection maps. Eur J Neurosci, 19, 2511-8.

HAGGLUND, M., BERGHARD, A., STROTMANN, J. & BOHM, S. 2006. Retinoic acid receptor-dependent survival of olfactory sensory neurons in postnatal and adult mice. J Neurosci, 26, 3281-91.

HALLEM, E. A., HO, M. G. & CARLSON, J. R. 2004. The molecular basis of odor coding in the Drosophila antenna. Cell, 117, 965-79.

HARDY, J. & SELKOE, D. J. 2002. The amyloid hypothesis of Alzheimer's disease: progress and problems on the road to therapeutics. Science, 297, 353-6.

HASKELL, G. T. & LAMANTIA, A. S. 2005. Retinoic acid signaling identifies a distinct precursor population in the developing and adult forebrain. J Neurosci, 25, 7636-47.

HEMMING, M. L., ELIAS, J. E., GYGI, S. P. & SELKOE, D. J. 2009. Identification of beta-secretase (BACE1) substrates using quantitative proteomics. PLoS One, 4, e8477.

HIROTA, J., OMURA, M. & MOMBAERTS, P. 2007. Differential impact of Lhx2 deficiency on expression of class I and class II odorant receptor genes in mouse. Mol Cell Neurosci, 34, 679-88.

HITT, B., RIORDAN, S. M., KUKREJA, L., EIMER, W. A., RAJAPAKSHA, T. W. & VASSAR, R. 2012. beta-Site amyloid precursor protein (APP)-cleaving enzyme 1 (BACE1)-deficient mice exhibit a close homolog of L1 (CHL1) loss-of-function phenotype involving axon guidance defects. J Biol Chem, 287, 38408-25.

HORNBERG, M., GUSSING, F., BERGHARD, A. & BOHM, S. 2009. Retinoic acid selectively inhibits death of basal vomeronasal neurons during late stage of neural circuit formation. J Neurochem, 110, 1263-75.

HUBEL, D. H. & WIESEL, T. N. 1964. Effects of Monocular Deprivation in Kittens. Naunyn Schmiedebergs Arch Exp Pathol Pharmakol, 248, 492-7.

IMAI, T. & SAKANO, H. 2009. Odorant receptor gene choice and axonal projection in the mouse olfactory system. Results Probl Cell Differ, 47, 57-75.

48

IMAI, T., SUZUKI, M. & SAKANO, H. 2006. Odorant receptor-derived cAMP signals direct axonal targeting. Science, 314, 657-61.

JACOBS, S., LIE, D. C., DECICCO, K. L., SHI, Y., DELUCA, L. M., GAGE, F. H. & EVANS, R. M. 2006. Retinoic acid is required early during adult neurogenesis in the dentate gyrus. Proc Natl Acad Sci U S A, 103, 3902-7.

JIANG, W., YU, Q., GONG, M., CHEN, L., WEN, E. Y., BI, Y., ZHANG, Y., SHI, Y., QU, P., LIU, Y. X., WEI, X. P., CHEN, J. & LI, T. Y. 2012. Vitamin A deficiency impairs postnatal cognitive function via inhibition of neuronal calcium excitability in hippocampus. J Neurochem, 121, 932-43.

KELLER, A. & MARGOLIS, F. L. 1976. Isolation and characterization of rat olfactory marker protein. J Biol Chem, 251, 6232-7.

KERR, M. A. & BELLUSCIO, L. 2006. Olfactory experience accelerates glomerular refinement in the mammalian olfactory bulb. Nat Neurosci, 9, 484-6.

KOBAYAKAWA, K., KOBAYAKAWA, R., MATSUMOTO, H., OKA, Y., IMAI, T., IKAWA, M., OKABE, M., IKEDA, T., ITOHARA, S., KIKUSUI, T., MORI, K. & SAKANO, H. 2007. Innate versus learned odour processing in the mouse olfactory bulb. Nature, 450, 503-8.

KORSCHING, S. 2002. Olfactory maps and odor images. Curr Opin Neurobiol, 12, 387-92.

LAIRD, F. M., CAI, H., SAVONENKO, A. V., FARAH, M. H., HE, K., MELNIKOVA, T., WEN, H., CHIANG, H. C., XU, G., KOLIATSOS, V. E., BORCHELT, D. R., PRICE, D. L., LEE, H. K. & WONG, P. C. 2005. BACE1, a major determinant of selective vulnerability of the brain to amyloid-beta amyloidogenesis, is essential for cognitive, emotional, and synaptic functions. J Neurosci, 25, 11693-709.

LAMANTIA, A. S., BHASIN, N., RHODES, K. & HEEMSKERK, J. 2000. Mesenchymal/epithelial induction mediates olfactory pathway formation. Neuron, 28, 411-25.

LAMANTIA, A. S., COLBERT, M. C. & LINNEY, E. 1993. Retinoic acid induction and regional differentiation prefigure olfactory pathway formation in the mammalian forebrain. Neuron, 10, 1035-48.

LEUNG, C., CHAN, S. C., TSANG, S. L., WU, W. & SHAM, M. H. 2012. Cyp26b1 mediates differential neurogenicity in axial-specific populations of adult spinal cord progenitor cells. Stem Cells Dev, 21, 2252-61.

LEUNG, C. T., COULOMBE, P. A. & REED, R. R. 2007. Contribution of olfactory neural stem cells to tissue maintenance and regeneration. Nat Neurosci, 10, 720-6.

LEWCOCK, J. W. & REED, R. R. 2004. A feedback mechanism regulates monoallelic odorant receptor expression. Proc Natl Acad Sci U S A, 101, 1069-74.

49

LI, J., ISHII, T., FEINSTEIN, P. & MOMBAERTS, P. 2004. Odorant receptor gene choice is reset by nuclear transfer from mouse olfactory sensory neurons. Nature, 428, 393-9.

LI, S., JIN, M., KOEGLSPERGER, T., SHEPARDSON, N. E., SHANKAR, G. M. & SELKOE, D. J. 2011. Soluble Abeta oligomers inhibit long-term potentiation through a mechanism involving excessive activation of extrasynaptic NR2B-containing NMDA receptors. J Neurosci, 31, 6627-38.

LIN, C. W., SIM, S., AINSWORTH, A., OKADA, M., KELSCH, W. & LOIS, C. 2010. Genetically increased cell-intrinsic excitability enhances neuronal integration into adult brain circuits. Neuron, 65, 32-9.

LIN, D. M., WANG, F., LOWE, G., GOLD, G. H., AXEL, R., NGAI, J. & BRUNET, L. 2000. Formation of precise connections in the olfactory bulb occurs in the absence of odorant-evoked neuronal activity. Neuron, 26, 69-80.

LIN, W., ARELLANO, J., SLOTNICK, B. & RESTREPO, D. 2004. Odors detected by mice deficient in cyclic nucleotide-gated channel subunit A2 stimulate the main olfactory system. J Neurosci, 24, 3703-10.

LIU, F., QI, H. L. & CHEN, H. L. 2001. Regulation of differentiation- and proliferation-inducers on Lewis antigens, alpha-fucosyltransferase and metastatic potential in hepatocarcinoma cells. Br J Cancer, 84, 1556-63.

LOMVARDAS, S., BARNEA, G., PISAPIA, D. J., MENDELSOHN, M., KIRKLAND, J. & AXEL, R. 2006. Interchromosomal interactions and olfactory receptor choice. Cell, 126, 403-13.

LYONS, D. B., ALLEN, W. E., GOH, T., TSAI, L., BARNEA, G. & LOMVARDAS, S. 2013. An epigenetic trap stabilizes singular olfactory receptor expression. Cell, 154, 325-36.

MADEN, M. 2002. Retinoic acid and limb regeneration--a personal view. Int J Dev Biol, 46, 883-6.

MADEN, M. 2006. Retinoids and spinal cord development. J Neurobiol, 66, 726-38.

MADEN, M. 2007. Retinoic acid in the development, regeneration and maintenance of the nervous system. Nat Rev Neurosci, 8, 755-65.

MADEN, M., GALE, E., KOSTETSKII, I. & ZILE, M. 1996. Vitamin A-deficient quail embryos have half a hindbrain and other neural defects. Curr Biol, 6, 417-26.

MADEN, M. & HOLDER, N. 1991. The involvement of retinoic acid in the development of the vertebrate central nervous system. Development, Suppl 2, 87-94.

MAGHSOODI, B., POON, M. M., NAM, C. I., AOTO, J., TING, P. & CHEN, L. 2008. Retinoic acid regulates RARalpha-mediated control of translation in dendritic RNA granules during homeostatic synaptic plasticity. Proc Natl Acad Sci U S A, 105, 16015-20.

MAGKLARA, A., YEN, A., COLQUITT, B. M., CLOWNEY, E. J., ALLEN, W., MARKENSCOFF-PAPADIMITRIOU, E., EVANS, Z. A.,

50

KHERADPOUR, P., MOUNTOUFARIS, G., CAREY, C., BARNEA, G., KELLIS, M. & LOMVARDAS, S. 2011. An epigenetic signature for monoallelic olfactory receptor expression. Cell, 145, 555-70.

MARK, M., GHYSELINCK, N. B. & CHAMBON, P. 2006. Function of retinoid nuclear receptors: lessons from genetic and pharmacological dissections of the retinoic acid signaling pathway during mouse embryogenesis. Annu Rev Pharmacol Toxicol, 46, 451-80.

MATTHEWS, H. R. & REISERT, J. 2003. Calcium, the two-faced messenger of olfactory transduction and adaptation. Curr Opin Neurobiol, 13, 469-75.

MATTSON, M. P. 1997. Cellular actions of beta-amyloid precursor protein and its soluble and fibrillogenic derivatives. Physiol Rev, 77, 1081-132.

MCCAFFERY, P., WAGNER, E., O'NEIL, J., PETKOVICH, M. & DRAGER, U. C. 1999. Dorsal and ventral retinal territories defined by retinoic acid synthesis, break-down and nuclear receptor expression. Mech Dev, 82, 119-30.

MCCLINTOCK, T. S. 2010. Achieving singularity in mammalian odorant receptor gene choice. Chem Senses, 35, 447-57.

MCCONLOGUE, L., BUTTINI, M., ANDERSON, J. P., BRIGHAM, E. F., CHEN, K. S., FREEDMAN, S. B., GAMES, D., JOHNSON-WOOD, K., LEE, M., ZELLER, M., LIU, W., MOTTER, R. & SINHA, S. 2007. Partial reduction of BACE1 has dramatic effects on Alzheimer plaque and synaptic pathology in APP Transgenic Mice. J Biol Chem, 282, 26326-34.

MCLAUGHLIN, T., TORBORG, C. L., FELLER, M. B. & O'LEARY, D. D. 2003. Retinotopic map refinement requires spontaneous retinal waves during a brief critical period of development. Neuron, 40, 1147-60.

MELTON, K. R., IULIANELLA, A. & TRAINOR, P. A. 2004. Gene expression and regulation of hindbrain and spinal cord development. Front Biosci, 9, 117-38.

MISNER, D. L., JACOBS, S., SHIMIZU, Y., DE URQUIZA, A. M., SOLOMIN, L., PERLMANN, T., DE LUCA, L. M., STEVENS, C. F. & EVANS, R. M. 2001. Vitamin A deprivation results in reversible loss of hippocampal long-term synaptic plasticity. Proc Natl Acad Sci U S A, 98, 11714-9.

MIYAMICHI, K., SERIZAWA, S., KIMURA, H. M. & SAKANO, H. 2005. Continuous and overlapping expression domains of odorant receptor genes in the olfactory epithelium determine the dorsal/ventral positioning of glomeruli in the olfactory bulb. J Neurosci, 25, 3586-92.

MOMBAERTS, P. 1999. Seven-transmembrane proteins as odorant and chemosensory receptors. Science, 286, 707-11.

51

MOMBAERTS, P. 2004. Odorant receptor gene choice in olfactory sensory neurons: the one receptor-one neuron hypothesis revisited. Curr Opin Neurobiol, 14, 31-6.

MORI, K., NAGAO, H. & YOSHIHARA, Y. 1999. The olfactory bulb: coding and processing of odor molecule information. Science, 286, 711-5.

MURPHY, G. J. & ISAACSON, J. S. 2003. Presynaptic cyclic nucleotide-gated ion channels modulate neurotransmission in the mammalian olfactory bulb. Neuron, 37, 639-47.

MURRAY, R. C. & CALOF, A. L. 1999. Neuronal regeneration: lessons from the olfactory system. Semin Cell Dev Biol, 10, 421-31.

NAKASHIMA, A., TAKEUCHI, H., IMAI, T., SAITO, H., KIYONARI, H., ABE, T., CHEN, M., WEINSTEIN, L. S., YU, C. R., STORM, D. R., NISHIZUMI, H. & SAKANO, H. 2013. Agonist-independent GPCR activity regulates anterior-posterior targeting of olfactory sensory neurons. Cell, 154, 1314-25.

NAPOLI, J. L. 1999. Retinoic acid: its biosynthesis and metabolism. Prog Nucleic Acid Res Mol Biol, 63, 139-88.

NGUYEN, M. Q., ZHOU, Z., MARKS, C. A., RYBA, N. J. & BELLUSCIO, L. 2007. Prominent roles for odorant receptor coding sequences in allelic exclusion. Cell, 131, 1009-17.

NIEDERREITHER, K., FRAULOB, V., GARNIER, J. M., CHAMBON, P. & DOLLE, P. 2002. Differential expression of retinoic acid-synthesizing (RALDH) enzymes during fetal development and organ differentiation in the mouse. Mech Dev, 110, 165-71.

NIIMURA, Y. & NEI, M. 2007. Extensive gains and losses of olfactory receptor genes in mammalian evolution. PLoS One, 2, e708.

NIKOLAEV, A., MCLAUGHLIN, T., O'LEARY, D. D. & TESSIER-LAVIGNE, M. 2009. APP binds DR6 to trigger axon pruning and neuron death via distinct caspases. Nature, 457, 981-9.

NORLIN, E. M., ALENIUS, M., GUSSING, F., HAGGLUND, M., VEDIN, V. & BOHM, S. 2001. Evidence for gradients of gene expression correlating with zonal topography of the olfactory sensory map. Mol Cell Neurosci, 18, 283-95.

PASCHAKI, M., CAMMAS, L., MUTA, Y., MATSUOKA, Y., MAK, S. S., RATAJ-BANIOWSKA, M., FRAULOB, V., DOLLE, P. & LADHER, R. K. 2013. Retinoic acid regulates olfactory progenitor cell fate and differentiation. Neural Dev, 8, 13.

PELLIER-MONNIN, V., ASTIC, L., BICHET, S., RIEDERER, B. M. & GRENNINGLOH, G. 2001. Expression of SCG10 and stathmin proteins in the rat olfactory system during development and axonal regeneration. J Comp Neurol, 433, 239-54.

PELUSO, C. E., JANG, W., DRAGER, U. C. & SCHWOB, J. E. 2012. Differential expression of components of the retinoic acid signaling pathway in the adult mouse olfactory epithelium. J Comp Neurol, 520, 3707-26.

52

PIJNAPPEL, W. W., HENDRIKS, H. F., FOLKERS, G. E., VAN DEN BRINK, C. E., DEKKER, E. J., EDELENBOSCH, C., VAN DER SAAG, P. T. & DURSTON, A. J. 1993. The retinoid ligand 4-oxo-retinoic acid is a highly active modulator of positional specification. Nature, 366, 340-4.

PINCHING, A. J. & POWELL, T. P. 1971. The neuron types of the glomerular layer of the olfactory bulb. J Cell Sci, 9, 305-45.

PUN, R. Y. & KLEENE, S. J. 2003. Contribution of cyclic-nucleotide-gated channels to the resting conductance of olfactory receptor neurons. Biophys J, 84, 3425-35.

RAJAPAKSHA, T. W., EIMER, W. A., BOZZA, T. C. & VASSAR, R. 2011. The Alzheimer's beta-secretase enzyme BACE1 is required for accurate axon guidance of olfactory sensory neurons and normal glomerulus formation in the olfactory bulb. Mol Neurodegener, 6, 88.

RAWSON, N. E. & LAMANTIA, A. S. 2006. Once and again: retinoic acid signaling in the developing and regenerating olfactory pathway. J Neurobiol, 66, 653-76.

REIJNTJES, S., GALE, E. & MADEN, M. 2004. Generating gradients of retinoic acid in the chick embryo: Cyp26C1 expression and a comparative analysis of the Cyp26 enzymes. Dev Dyn, 230, 509-17.

RESSLER, K. J., SULLIVAN, S. L. & BUCK, L. B. 1993. A zonal organization of odorant receptor gene expression in the olfactory epithelium. Cell, 73, 597-609.

RESTREPO, D., DOUCETTE, W., WHITESELL, J. D., MCTAVISH, T. S. & SALCEDO, E. 2009. From the top down: flexible reading of a fragmented odor map. Trends Neurosci, 32, 525-31.

ROBERDS, S. L., ANDERSON, J., BASI, G., BIENKOWSKI, M. J., BRANSTETTER, D. G., CHEN, K. S., FREEDMAN, S. B., FRIGON, N. L., GAMES, D., HU, K., JOHNSON-WOOD, K., KAPPENMAN, K. E., KAWABE, T. T., KOLA, I., KUEHN, R., LEE, M., LIU, W., MOTTER, R., NICHOLS, N. F., POWER, M., ROBERTSON, D. W., SCHENK, D., SCHOOR, M., SHOPP, G. M., SHUCK, M. E., SINHA, S., SVENSSON, K. A., TATSUNO, G., TINTRUP, H., WIJSMAN, J., WRIGHT, S. & MCCONLOGUE, L. 2001. BACE knockout mice are healthy despite lacking the primary beta-secretase activity in brain: implications for Alzheimer's disease therapeutics. Hum Mol Genet, 10, 1317-24.

ROSPARS, J. P., LANSKY, P., TUCKWELL, H. C. & VERMEULEN, A. 1996. Coding of odor intensity in a steady-state deterministic model of an olfactory receptor neuron. J Comput Neurosci, 3, 51-72.

ROSPARS, J. P., LANSKY, P., VAILLANT, J., DUCHAMP-VIRET, P. & DUCHAMP, A. 1994. Spontaneous activity of first- and second-order neurons in the frog olfactory system. Brain Res, 662, 31-44.

ROSS, S. A., MCCAFFERY, P. J., DRAGER, U. C. & DE LUCA, L. M. 2000. Retinoids in embryonal development. Physiol Rev, 80, 1021-54.

53

SANTORO, S. W. & DULAC, C. 2012. The activity-dependent histone variant H2BE modulates the life span of olfactory neurons. Elife, 1, e00070.

SCHMUTH, M., WATSON, R. E., DEPLEWSKI, D., DUBRAC, S., ZOUBOULIS, C. C. & GRIFFITHS, C. E. 2007. Nuclear hormone receptors in human skin. Horm Metab Res, 39, 96-105.

SCHUG, T. T., BERRY, D. C., SHAW, N. S., TRAVIS, S. N. & NOY, N. 2007. Opposing effects of retinoic acid on cell growth result from alternate activation of two different nuclear receptors. Cell, 129, 723-33.

SCHWOB, J. E. 2002. Neural regeneration and the peripheral olfactory system. Anat Rec, 269, 33-49.

SEN, J., HARPAVAT, S., PETERS, M. A. & CEPKO, C. L. 2005. Retinoic acid regulates the expression of dorsoventral topographic guidance molecules in the chick retina. Development, 132, 5147-59.

SERIZAWA, S., MIYAMICHI, K., NAKATANI, H., SUZUKI, M., SAITO, M., YOSHIHARA, Y. & SAKANO, H. 2003. Negative feedback regulation ensures the one receptor-one olfactory neuron rule in mouse. Science, 302, 2088-94.

SERIZAWA, S., MIYAMICHI, K., TAKEUCHI, H., YAMAGISHI, Y., SUZUKI, M. & SAKANO, H. 2006. A neuronal identity code for the odorant receptor-specific and activity-dependent axon sorting. Cell, 127, 1057-69.

SHEPHERD, G. M. 2004. The human sense of smell: are we better than we think? PLoS Biol, 2, E146.

SHIPLEY, M. T. & ENNIS, M. 1996. Functional organization of olfactory system. J Neurobiol, 30, 123-76.

SHYKIND, B. M., ROHANI, S. C., O'DONNELL, S., NEMES, A., MENDELSOHN, M., SUN, Y., AXEL, R. & BARNEA, G. 2004. Gene switching and the stability of odorant receptor gene choice. Cell, 117, 801-15.

TAKEUCHI, H., INOKUCHI, K., AOKI, M., SUTO, F., TSUBOI, A., MATSUDA, I., SUZUKI, M., AIBA, A., SERIZAWA, S., YOSHIHARA, Y., FUJISAWA, H. & SAKANO, H. 2010. Sequential arrival and graded secretion of Sema3F by olfactory neuron axons specify map topography at the bulb. Cell, 141, 1056-67.

TAYLOR, C. J., IRELAND, D. R., BALLAGH, I., BOURNE, K., MARECHAL, N. M., TURNER, P. R., BILKEY, D. K., TATE, W. P. & ABRAHAM, W. C. 2008. Endogenous secreted amyloid precursor protein-alpha regulates hippocampal NMDA receptor function, long-term potentiation and spatial memory. Neurobiol Dis, 31, 250-60.

TIAN, H. & MA, M. 2008. Activity plays a role in eliminating olfactory sensory neurons expressing multiple odorant receptors in the mouse septal organ. Mol Cell Neurosci, 38, 484-8.

TSUBOI, A., MIYAZAKI, T., IMAI, T. & SAKANO, H. 2006. Olfactory sensory neurons expressing class I odorant receptors converge their axons on an antero-dorsal domain of the olfactory bulb in the mouse. Eur J Neurosci, 23, 1436-44.

54

VASSAR, R. 2004. BACE1: the beta-secretase enzyme in Alzheimer's disease. J Mol Neurosci, 23, 105-14.

VASSAR, R., NGAI, J. & AXEL, R. 1993. Spatial segregation of odorant receptor expression in the mammalian olfactory epithelium. Cell, 74, 309-18.

VEDIN, V., MOLANDER, M., BOHM, S. & BERGHARD, A. 2009. Regional differences in olfactory epithelial homeostasis in the adult mouse. J Comp Neurol, 513, 375-84.

VERHAAGEN, J., OESTREICHER, A. B., GISPEN, W. H. & MARGOLIS, F. L. 1989. The expression of the growth associated protein B50/GAP43 in the olfactory system of neonatal and adult rats. J Neurosci, 9, 683-91.

WAGNER, E., MCCAFFERY, P. & DRAGER, U. C. 2000. Retinoic acid in the formation of the dorsoventral retina and its central projections. Dev Biol, 222, 460-70.

WANG, H. L., ZHANG, Z., HINTZE, M. & CHEN, L. 2011. Decrease in calcium concentration triggers neuronal retinoic acid synthesis during homeostatic synaptic plasticity. J Neurosci, 31, 17764-71.

WANG, S. S., LEWCOCK, J. W., FEINSTEIN, P., MOMBAERTS, P. & REED, R. R. 2004. Genetic disruptions of O/E2 and O/E3 genes reveal involvement in olfactory receptor neuron projection. Development, 131, 1377-88.

WATT, W. C., SAKANO, H., LEE, Z. Y., REUSCH, J. E., TRINH, K. & STORM, D. R. 2004. Odorant stimulation enhances survival of olfactory sensory neurons via MAPK and CREB. Neuron, 41, 955-67.

WHITESIDES, J., HALL, M., ANCHAN, R. & LAMANTIA, A. S. 1998. Retinoid signaling distinguishes a subpopulation of olfactory receptor neurons in the developing and adult mouse. J Comp Neurol, 394, 445-61.

WILSON, L., GALE, E., CHAMBERS, D. & MADEN, M. 2004. Retinoic acid and the control of dorsoventral patterning in the avian spinal cord. Dev Biol, 269, 433-46.

WILSON, L. & MADEN, M. 2005. The mechanisms of dorsoventral patterning in the vertebrate neural tube. Dev Biol, 282, 1-13.

YAN, R. & VASSAR, R. 2014. Targeting the beta secretase BACE1 for Alzheimer's disease therapy. Lancet Neurol, 13, 319-29.

YOUNG, J. M., FRIEDMAN, C., WILLIAMS, E. M., ROSS, J. A., TONNES-PRIDDY, L. & TRASK, B. J. 2002. Different evolutionary processes shaped the mouse and human olfactory receptor gene families. Hum Mol Genet, 11, 535-46.

YOUNG, J. M. & TRASK, B. J. 2002. The sense of smell: genomics of vertebrate odorant receptors. Hum Mol Genet, 11, 1153-60.

YU, C. R., POWER, J., BARNEA, G., O'DONNELL, S., BROWN, H. E., OSBORNE, J., AXEL, R. & GOGOS, J. A. 2004. Spontaneous neural activity is required for the establishment and maintenance of the olfactory sensory map. Neuron, 42, 553-66.

55

YU, S., LEVI, L., SIEGEL, R. & NOY, N. 2012. Retinoic acid induces neurogenesis by activating both retinoic acid receptors (RARs) and peroxisome proliferator-activated receptor beta/delta (PPARbeta/delta). J Biol Chem, 287, 42195-205.

ZHANG, Q. Y. 1999. Retinoic acid biosynthetic activity and retinoid receptors in the olfactory mucosa of adult mice. Biochem Biophys Res Commun, 256, 346-51.

ZHANG, X., ROGERS, M., TIAN, H., ZHANG, X., ZOU, D. J., LIU, J., MA, M., SHEPHERD, G. M. & FIRESTEIN, S. J. 2004. High-throughput microarray detection of olfactory receptor gene expression in the mouse. Proc Natl Acad Sci U S A, 101, 14168-73.

ZHAO, H. & REED, R. R. 2001. X inactivation of the OCNC1 channel gene reveals a role for activity-dependent competition in the olfactory system. Cell, 104, 651-60.

ZHAO, S., TIAN, H., MA, L., YUAN, Y., YU, C. R. & MA, M. 2013. Activity-dependent modulation of odorant receptor gene expression in the mouse olfactory epithelium. PLoS One, 8, e69862.

ZHENG, C., FEINSTEIN, P., BOZZA, T., RODRIGUEZ, I. & MOMBAERTS, P. 2000a. Peripheral olfactory projections are differentially affected in mice deficient in a cyclic nucleotide-gated channel subunit. Neuron, 26, 81-91.

ZHENG, W. L., SIERRA-RIVERA, E., LUAN, J., OSTEEN, K. G. & ONG, D. E. 2000b. Retinoic acid synthesis and expression of cellular retinol-binding protein and cellular retinoic acid-binding protein type II are concurrent with decidualization of rat uterine stromal cells. Endocrinology, 141, 802-8.

ZOU, D. J., CHESLER, A. & FIRESTEIN, S. 2009. How the olfactory bulb got its glomeruli: a just so story? Nat Rev Neurosci, 10, 611-8.

ZOU, D. J., FEINSTEIN, P., RIVERS, A. L., MATHEWS, G. A., KIM, A., GREER, C. A., MOMBAERTS, P. & FIRESTEIN, S. 2004. Postnatal refinement of peripheral olfactory projections. Science, 304, 1976-9.