View
4
Download
0
Category
Preview:
Citation preview
1
Thyroid hormone coordinates pancreatic islet maturation during the zebrafish larval to
juvenile transition to maintain glucose homeostasis
Hiroki Matsuda1*
, Sri Teja Mullapudi1, Yuxi Zhang
2, Daniel Hesselson
2,3, and Didier Y. R.
Stainier1*
Author affiliation:
1. Max Planck Institute for Heart and Lung Research, Department of Developmental
Genetics, Bad Nauheim, Germany
2. Garvan Institute of Medical Research, Diabetes and Metabolism Division, Sydney,
Australia
3. St. Vincent’s Clinical School, UNSW Australia, Sydney, Australia
*Corresponding authors
Hiroki Matsuda (hiroki.matsuda.b5@tohoku.ac.jp)
Didier Stainier (Didier.Stainier@mpi-bn.mpg.de)
Department of Developmental Genetics
Max Planck Institute for Heart and Lung Research.
Ludwigstrasse 43, 61231 Bad Nauheim, Germany
Phone: +49 (0) 6032 705-1333
Fax:+49 (0) 6032 705-1304
Running title: TH regulates pancreatic islet maturation
Page 1 of 48 Diabetes
Diabetes Publish Ahead of Print, published online July 11, 2017
2
Abstract
Thyroid hormone (TH) signaling promotes tissue maturation and adult organ
formation. Developmental transitions alter an organism's metabolic requirements, and it
remains unclear how development and metabolic demands are coordinated. We used the
zebrafish as a model to test whether and how TH signaling affects pancreatic islet maturation,
and consequently glucose homeostasis, during the larval to juvenile transition. We found that
exogenous TH precociously activates the β cell differentiation genes pax6b and mnx1, while
downregulating arxa, a master regulator of α cell development and function. Together these
effects induced hypoglycemia, at least in part by increasing insulin and decreasing glucagon
expression. We visualized TH target tissues using a novel TH responsive reporter line and
found that both α and β cells become targets of endogenous TH signaling during the larval to
juvenile transition. Importantly, endogenous TH is required during this transition for the
functional maturation of α and β cells in order to maintain glucose homeostasis. Thus, our
study sheds new light on the regulation of glucose metabolism during major developmental
transitions.
Page 2 of 48Diabetes
3
Introduction
Organ development often involves a two-step process: the formation of a functionally
immature organ during embryogenesis followed by maturation into the adult form. This
second step most often takes place during the postembryonic/postnatal developmental period,
when plasma thyroid hormone (TH) concentrations are high. In rodents, these major
developmental changes occur during the suckling to weaning transition, and their digestive
organs undergo functional and morphological changes during this period (1), correlating with
the dietary switch from mother’s milk to solid foods. It has been proposed that Xenopus
metamorphosis and the zebrafish larval-juvenile transition are functionally equivalent to
mammalian weaning, based on cellular and molecular events in their digestive organs during
these periods (2-7).
The pancreas is an organ in the digestive system that undergoes morphological and
functional changes during the suckling to weaning transition. For example, the weight of the
rat pancreas increases 5-6 fold during this period (8). In addition, pancreatic acinar cells
become functionally mature and start to secrete the full suite of digestive enzymes, including
trypsin and amylase, into the duodenum during the weaning period (1). In the endocrine
pancreas, β cell mass increases (8), and more organized, functional islets appear during this
period (8-13) as β cells acquire the ability to secrete insulin in response to glucose around the
second week after birth (9). β cell secretory functions continue to improve well beyond
weaning periods (9), although the underlying mechanisms are not well understood.
The thyroid hormone T3 (triiodothyronine) signals through the nuclear Thyroid
Hormone Receptor (THR). THR functions as a transcriptional repressor in the absence of TH
and a transcriptional activator in the presence of TH. It has been reported that ligand bound
THR can directly activate Mafa expression and that exogenous TH can induce MAFA
Page 3 of 48 Diabetes
4
dependent glucose-responsive insulin secretion in the rat pancreas (14). Furthermore, TH is
used in the most effective protocols developed to date to generate glucose responsive insulin
secreting cells from pluripotent cells (15, 16). These data implicate TH signaling in β cell
maturation. Furthermore, TH can also induce endocrine cell differentiation from ductal cells
in E12.5 mouse pancreatic explants (17) and stimulate acinar cell proliferation in the adult
mouse (18). It has therefore been hypothesized that TH regulates multiple aspects of
pancreas development, including endocrine cell differentiation, β cell maturation, and the size
of the exocrine compartment. However, it remains unclear exactly when and where TH
signaling is activated during pancreas development.
The zebrafish (Danio rerio) has emerged as a powerful model to study mechanisms of
pancreas development and β cell differentiation. Zebrafish pancreatic development is similar
to that in mammals in terms of cellular events and molecular pathways (19-21). For example,
in zebrafish, bilateral pdx1 positive pancreatic primordia appear by 14 hours post fertilization
(hpf), and the dorsal bud emerges from these primordia by 22 hpf. At 40 hpf, the ventral bud
is visible, and by 52 hpf, it has fused with the dorsal bud; the principal islet, originally
derived from the dorsal bud, is organized like a mammalian islet, with α and δ cells
surrounding a β cell core. Here we investigated the spatiotemporal regulation of TH
signaling during zebrafish pancreatic development, focusing on the endocrine lineages. We
show that TH stimulates insulin expression, through upregulation of pax6b and mnx1. In
addition, we found that TH inhibits the expression of both glucagon paralogs by upregulating
pax4 expression and downregulating arxa expression. Together these effects, at least in part,
decrease glucose levels in zebrafish larvae. To visualize TH target tissues, we generated a
TH reporter line and found that both α and β cells are direct targets of TH signaling during
and after the larval-juvenile transition. Additionally, using a hypothyroid zebrafish model,
we found that TH can regulate glucose homeostasis by modulating the expression of
Page 4 of 48Diabetes
5
endocrine genes during the larval-juvenile transition. In adults, TH continues to regulate α
and β cell function as well as glucose homeostasis. These results indicate that TH signaling
is involved in the development and maturation of α and β cells during postembryonic
development, and that it maintains glucose homeostasis, at least in part, by modulating the
insulin to glucagon ratio throughout the vertebrate lifecycle.
Research Design and Methods
Zebrafish lines
All zebrafish husbandry was performed under standard conditions in accordance with
institutional and national ethical and animal welfare guidelines. A 1000x stock solution of 10
µM triiodothyronine (T3) in 5 mM NaOH was made and stored at – 20 ºC until use. Larvae
were treated with the indicated concentration of T3 in egg water for the indicated number of
days at 28°C. Larvae incubated in 5 µM NaOH egg water were used as controls. For adult
zebrafish, 10 µl of 10 nM T3 was injected intraperitoneally once a day for three consecutive
days. 10 µl of 7 µM NaOH in egg water were used for control injections.
We used the following transgenic lines: Tg(ins:Luc2;cryaa:mCherry)gi3
, abbreviated
ins:Luc2; Tg(ins:H2BGFP;ins:dsRED)s960
(22), abbreviated ins:H2BGFP; Tg(-
4.0ins:eGFP)Zf5
(23), abbreviated ins:eGFP; Tg(gcga:eGFP)ia1
(24), abbreviated gcga:eGFP;
Tg(P0-pax6b:eGFP)ulg515
(25), abbreviated pax6b:eGFP; TgBAC(neurod1:eGFP)nl1
(26),
abbreviated neurod1:eGFP; Tg(-2.6mnx1:GFP)ml59
(27), abbreviated mnx1:eGFP;
Tg(tg:nVenus-2a-nfsB)wp.rt8
(28), abbreviated tg:nVenus-2a-nfsB.
Nitroreductase-mediated cell ablation
To ablate the thyroid follicles of tg:nVenus-2a-nfsB fish, we incubated 20 days post
fertilization (dpf) larvae for 48 h in 10 mM Metronidazole (Mtz) with 1% DMSO, or 1%
Page 5 of 48 Diabetes
6
DMSO alone as a control. Thyroid ablations of adult fish were performed on transgenic fish
injected intraperitoneally with 15 µl of 30 mM Mtz or DMSO as a control.
Results
T3 stimulates insulin promoter activity, reduces glucose levels and improves glucose
tolerance
To determine the effect of thyroid hormone on pancreatic preproinsulin (insulin)
expression, transgenic insulin:luciferase2 (Tg(ins:Lyc2;cryaa:mCherry)gi3
) larvae were
treated with T3 from 4 to 7 dpf (Fig. 1A and B). T3 significantly increased insulin promoter
activity at 10 nM, but showed no significant effects at 0.1, 1 or 100 nM (Fig. 1B). Glucose
levels were reciprocally reduced in a T3 dose-dependent manner (Fig. 1C). During the 4 to 7
day developmental window, 10 nM T3 was the optimal concentration for activation of insulin
expression and glucose reduction. To further define when T3 acts, larvae were treated with
10 nM T3 starting at 4 dpf and insulin promoter activity and glucose levels were analyzed at
5, 6, 7 and 8 dpf (Fig. 1D). Insulin promoter activity was enhanced at 7 and 8 dpf (Fig. 1E),
while glucose levels were reduced at 6,7 and 8 dpf (Fig. 1F). These results indicate that
insulin expression is T3 responsive by 7 dpf and that our 4 to 7 dpf treatment conditions
induce a hypoglycemic phenotype.
Given the correlation between insulin upregulation and glucose reduction in these
experiments, we next tested glucose tolerance in T3 treated larvae. At 7 dpf, after 3 days of
T3 treatment, larvae were exposed to glucose for 60 mins and glucose levels were analyzed at
0, 30, 60 and 90 mins after glucose treatment (Fig. 1G). Glucose levels were unaltered in T3
treated larvae immediately after the glucose challenge (0 min). Strikingly, at 30, 60 and 90
mins after the challenge, glucose levels were lower in T3-treated larvae compared to non-
Page 6 of 48Diabetes
7
treated controls (Fig. 1H). These data indicate that T3 increases the glucose disposal capacity
of zebrafish larvae.
T3 stimulates insulin expression but does not alter β cell number
Given that TH stimulated insulin promoter activity in ins:luc2 lines, we next tested
whether TH stimulates endogenous insulin expression using qPCR. Zebrafish has two insulin
genes, preproinsulin (insulin) and preproinsulin b (insulin b). We focused our analysis on
insulin, because insulin b expression is minimal after 2 dpf. Starting T3 treatment at 4 dpf,
insulin expression was analyzed at 5, 6 and 7 dpf. Although T3 did not significantly affect
endogenous insulin expression at 5 and 6 dpf, enhanced insulin expression was observed at 7
dpf (Fig. 2A). Next, we analyzed whether TH alters the endogenous insulin expression
pattern using in situ hybridization. Zebrafish larvae were treated with T3 from 4 dpf and
analyzed at 7 dpf. T3 did not induce ectopic insulin expression (Fig. 2B and C). Furthermore,
using a similar treatment strategy with ins:eGFP reporter fish, we did not observe eGFP
expression outside the pancreas (Fig. 2D and E). Notably, T3 increased ins:eGFP expression
intensity in the principal islet compared to controls (Fig. 2F-H). To determine whether an
expanded β cell mass contributed to the increased insulin expression in T3-treated larvae, we
used a nuclear localized ins:H2BGFP reporter to quantify β cell number. Interestingly, T3
did not alter the number of β cells (Fig. 2I-K). Taken together, these data suggest that T3
treatments of zebrafish larvae enhance insulin expression without increasing β cell number in
zebrafish.
T3 represses glucagon expression and reduces α cell number
As glucose homeostasis is also under the control of Glucagon secreted from
pancreatic α cells, we investigated whether TH signaling regulates α cell function. We
Page 7 of 48 Diabetes
8
treated wild-type larvae with T3 starting at 4 dpf and analyzed the expression levels of
glucagon a (gcga) and glucagon b (gcgb) at 5, 6 and 7 dpf by qPCR. T3 significantly
repressed the expression of both gcga and gcgb at 5, 6 and 7 dpf (Fig. 3A). We next treated
larvae with T3 starting at 4 dpf and investigated its effect on the expression pattern of gcga
and gcgb at 7 dpf by in situ hybridization. T3 treatment reduced gcga and gcgb expression in
pancreatic islets (Fig. 3B) and no ectopic expression was observed (Fig. 3B and data not
shown). Consistent with the effect on endogenous gcga expression, T3 suppressed the
expression of the gcga:eGFP transcriptional reporter (Fig. 3C). Interestingly, the number of
gcga:eGFP positive α cells was modestly reduced in T3 treated larvae (26 +/- 2 in controls;
21+/- 2 in T3 treated, Fig. 3E). Thus, TH impairs α cell differentiation and/or proliferation in
addition to negatively regulating glucagon expression.
To determine whether the reduction in glucagon expression was caused by negative
feedback from elevated paracrine Insulin signaling upon T3 treatment, we generated insulin
(ins, previously known as insa) mutants using CRISPR/Cas9 mediated genome editing. ins-/-
larvae exhibit a complete absence of detectable Insulin protein in their pancreatic islet (data
not shown), and although they appear to develop apparently normally during early stages,
mutant animals die by around 10-11 dpf due to metabolic defects (data not shown). We
treated ins -/- larvae with T3 at 4 dpf and analyzed gcga and gcgb expression levels by qPCR
at 6 dpf. Both gcga and gcgb expression were reduced after T3 treatment (Supplemental
Figure 1). These results indicate that T3 represses glucagon expression through an Insulin-
independent pathway.
T3 stimulates pax6b, mnx1 and pax4 expression and represses arxa expression.
To understand how T3 regulates pancreatic islet expression of insulin, gcga and gcgb,
we investigated the effects of T3 treatment on upstream endocrine transcription factor genes.
Page 8 of 48Diabetes
9
We used transcriptional eGFP reporters for pax6b, neurod1 and mnx1, which encode known
regulators of endocrine differentiation and/or maturation (Fig. 4A-F). T3 treatment increased
eGFP intensity of the pax6b and mnx1, but not of the neurod1, reporter transgenes within the
principal islet (Fig. 4A-F). Furthermore qPCR analysis revealed that T3 treatment
upregulated endogenous expression of pax6b, mnx1 and pax4, and downregulated expression
of arxa (Fig. 4G).
Pancreatic α and β cells are targets of thyroid hormone during and after larval
development.
We have shown that supplemental T3 stimulated insulin expression and repressed
glucagon expression leading to reduced glucose levels. We next asked whether and when
endogenous T3 signaling was activated in pancreatic α and β cells. To address this question,
we generated a TH responsive zebrafish line to visualize T3 target tissues in vivo. The
biological effects of T3 are mediated by signaling through THRs, which bind to TH response
elements (TRE) to regulate downstream target genes. Three copies of Xenopus laevis
THbzipTREs (6xTRE) were inserted upstream of candidate minimal promoters driving eGFP
to visualize TH signaling (Supplemental Figure 2A). To identify the optimal configuration,
four different minimal promoters (cfos, bglob1, tk and Pmini) were tested by transient
injection assay for their response to T3 treatment (Supplemental Figure 2B). The construct
incorporating the bglob1 minimal promoter showed the highest proportion of eGFP+
embryos after T3 treatment (Supplemental Figures 2C and D). Next, we confirmed that T3
responsiveness was mediated by the TREs. We compared 6xTRE to bglob1:eGFP plasmids
lacking TREs, containing 2xTREs, or containing mutant TREs that cannot bind THR
(6xTREs(mut)) (Supplemental Figure 2E), and found that intact TREs were essential for the
T3-dependent stimulation of 6xTRE-bglobe:eGFP expression (Supplemental Figure 2F).
Page 9 of 48 Diabetes
10
A stable transgenic 6xTRE-bglob1:eGFP reporter was used to identify cells
responsive to endogenous TH in the pancreas (Fig. 5A). First, to validate the reporter, we
confirmed that exogenous T3 induced ubiquitous eGFP expression during larval stages
(Supplemental Figures 3A, D and E). Before 20 dpf (Fig. 5B and G), including early larval
stages (Supplemental Figures 3B and C), eGFP positive cells were not detected in the
pancreas (Supplemental Figure 3C). By 25 dpf, eGFP expression was detected in 71 % of
Insulin positive cells (Fig. 5C and G, white arrows), but only 0.7 % of Glucagon positive
cells (red arrows). By 30 dpf, eGFP expression was also observed in 86 % of Glucagon
positive cells (Fig. 5D and G). In the adult pancreas, TH reporter expression was observed in
89 % of Insulin and 95 % of Glucagon positive cells (Fig. 5E and G). These results indicate
that endogenous TH signaling is active during and after the larval-juvenile transition in
differentiated α and β cells.
Thyroid hormone regulates glucose metabolism by stimulating insulin expression and
suppressing glucagon expression during the larval-juvenile transition.
Analysis of the TH reporter line revealed that pancreatic TH signaling is active in
islets during the larval-juvenile transition. Therefore, we next investigated how endogenous
TH signaling affects α cell and β cell development and glucose homeostasis during the larval-
juvenile transition. Towards this end, we used the NTR/Mtz system to ablate the thyroid in
tg:nVenus-2a-nfsb animals (22, 29). We treated tg:nVenus-2a-nfsB fish with Mtz at 20 dpf
and analyzed their phenotypes at 30 dpf (Fig. 6A). Overall eGFP expression of the TH
reporter was reduced, and TH signaling was not detected in the islets of Mtz treated animals
at 30 dpf (Fig. 6B). Having established that thyroid ablation eliminated TH signaling in
endocrine cells, we next analyzed its functional consequences. Islet morphology was
unaltered in thyroid-ablated zebrafish (Fig. 6C). Interestingly we found that compared to the
Page 10 of 48Diabetes
11
unablated controls, the thyroid-ablated zebrafish exhibited elevated glucose levels (Fig. 6D),
that insulin expression levels were reduced (Fig. 6E), and that glucagon expression levels
were increased (Fig. 6E). Furthermore, expression levels of neurod1, pax6b, mnx1 and pax4
were reduced, and expression levels of arxa were increased in the digestive organs (liver,
pancreas and intestine) in thyroid-ablated zebrafish (Fig. 6E). These results suggest that
endogenous TH regulates glucose homeostasis during the larval-juvenile transition by
modulating the expression of endocrine genes including insulin and glucagon.
T3 regulates glucose metabolism by stimulating Insulin secretion and suppressing
Glucagon secretion in adults
Analysis of the TH reporter line revealed that pancreatic TH signaling is active in
adult islets. Therefore, we next investigated how TH signaling affects α cell and β cell
function and glucose metabolism in adults. 10 µl of a 10 nM T3 solution was injected
intraperitoneally (IP) daily for three consecutive days. After the third injection, the fish were
starved for 24 h and then challenged with an IP injection of 10 µl of a 2.5% glucose solution.
Blood glucose was measured after 45 mins (Fig. 7A). Under basal conditions as well as after
glucose challenge, T3 pretreatment reduced blood glucose levels (Fig. 7B). Next, pancreata
were dissected and the expression levels of insulin, glucagon a and glucagon b were analyzed
by qPCR. T3 upregulated expression of insulin, and downregulated expression of glucagon a
(Fig. 7C), consistent with the effects observed in larvae. T3 did not affect the expression
levels of glucagon b (Fig. 7C). To address whether altered gene expression levels led to
functional effects on hormone secretion, serum Insulin and Glucagon levels were quantified
by dot-blot. Serum Insulin levels were higher in animals treated with T3 compared to
controls (Fig. 7D and F). In contrast, serum Glucagon levels were reduced in fish treated
with T3 compared to controls (Fig. 7E and F). In addition, we investigated how TH signaling
Page 11 of 48 Diabetes
12
affects α cell and β cell function and glucose metabolism in adults by analyzing hypothyroid
animals. Mtz was injected intraperitoneally in tg:nVenus-2a-nfsb fish. Beginning two days
following the Mtz injection, the fish were starved for an additional 24 h and then challenged
with a 10 µl IP injection of 1% glucose (Fig. 7G). Under basal conditions as well as after
glucose challenge, blood glucose levels were elevated in hypothyroid zebrafish (Fig. 7H).
The expression levels of insulin, glucagon a and glucagon b in the pancreas of hypothyroid
zebrafish were analyzed by qPCR. The expression levels of insulin were reduced while the
expression levels of both glucagon a and glucagon b were significantly increased (Fig. 7I).
Furthermore, serum Insulin levels were reduced (Fig. 7J and L), and serum Glucagon levels
were elevated in hypothyroid zebrafish (Fig. 7K and L). Thus, TH mediated effects on
pancreatic endocrine hormone expression induce an anti-hyperglycemic islet secretion profile
that improves glucose tolerance.
Discussion
In this study, we investigated the role of TH signaling during pancreas development in
vivo. We found that TH signaling activates insulin expression in β cells and inhibits
glucagon expression in α cells. TH also increases the expression of a transcriptional network
including the pax6b and mnx1genes, whose protein products regulate insulin expression.
Pax6, a pan-endocrine factor, directly regulates β cell specific mnx1 expression through a
regulatory element in the mnx1 promoter (27). While mnx1 is expressed specifically in β
cells (27), pax6b is expressed in all endocrine cell types including β cells (25), and both
Mnx1 and Pax6b regulate insulin expression. Our data are consistent with a model whereby
TH induction of pax6b expression induces mnx1 expression to drive insulin expression in β
cells. Furthermore, TH increases the expression of pax4 and inhibits the expression of arxa,
whose protein products regulate glucagon expression. Arx and Pax4 are mutually
Page 12 of 48Diabetes
13
antagonistic factors that modulate α cell development and glucagon expression (30). Our
data suggest that TH can tilt this balance in favor of Pax4, thus leading to a decrease in α cell
number and glucagon expression.
The TH/Pax6 pathway also appears to be at play in mammalian tissues. Exogenous
TH signaling can stimulate Insulin expression in db/db mice (31), and Pax6 expression is
reduced in TH deficient mice during fetal neurogenesis (32). Thus Pax6 might be one of the
common targets of TH signaling in pancreatic and neural tissues and might explain the
coordinated effects of TH signaling during embryogenesis, tissue differentiation, and organ
maturation. However, it remains to be investigated whether TH mediated Pax4 regulation is
also at play in mammalian tissues.
TH modulates the differentiation and function of various tissues including the
pancreas. However, it has been unclear when and in which cell types TH signals. Our data
using a novel TH responsive transgenic signaling reporter line reveal that in zebrafish, β cells
become TH responsive by 25 dpf and α cells by 30 dpf. These stages correspond to a critical
larval-juvenile transition, when zebrafish acquire an adult morphology and physiology (2, 3,
7, 33). This period is thought to be equivalent to the suckling to weaning transition in
mammals when various tissues undergo morphological and functional changes (2-7). During
this transition, rodent β cells acquire progressively higher glucose sensitivity and secretory
capacity (9). Our data indicate that islet endocrine cells are competent to respond to TH
during larval stages and that their maturation is stimulated by exposure to TH during and
after the larval-juvenile transition. Although TH signaling has been implicated in β cell
maturation (14), it is unclear to what extent endogenous TH signaling contributes to
mammalian pancreatic development and function. TH signaling is first detected in zebrafish
β cells during the larval-juvenile transition and remains active in adult islets. We found that
Page 13 of 48 Diabetes
14
TH signaling increases pax6 expression, which is required for maintaining the mature β cell
state in rodent islets (34-36). Thus, TH signaling may play a conserved role in the functional
maturation of β cells as well as the maintenance of β cell function in adult islets. In mammals,
Pax6 promotes and maintains β cell maturation by inducing a transcriptional network that
includes MAFA (34-37). However, in zebrafish we did not detect mafa expression in
pancreatic β cells (Supplemental Figure 4). Therefore, in zebrafish, Pax6b may use MafA-
independent pathways to induce β cell maturation. A complete understanding of the diversity
of TH-dependent regulatory mechanisms in vertebrates may reveal novel therapeutic
strategies to promote β cell maturation.
Interestingly, our data indicate that islet endocrine cells are competent to respond to
exogenous T3 during larval stages, but that endogenous T3 levels are not sufficient to
stimulate islet cells before 25 dpf. In addition, during development, β and α cells become TH
responsive at different stages (β cells by 25 dpf and α cells by 30 dpf), despite the fact that
both cell types have the competence to sense and respond to exogenous T3. Thus, islet cells
may have different thresholds for endogenous TH signaling. TH is synthesized in the
thyroid as thyroxine (T4), an inactive form for THR activation. Next, T4 is converted into T3,
a potent THR activator, by type 1 and type 2 deiodinases (D1 and D2, respectively) in
peripheral target tissues, and finally converted into the inactive T2 form by a type 3
deiodinase (D3) (38). In rat islets, D1 protein levels are much lower at postnatal day 7
compared to adult (14). In contrast, D3 shows a reciprocal pattern (14). Thus, expression of
key TH metabolic enzymes may explain the tissue specificity of the TH response.
Furthermore, in zebrafish, TH production dramatically increases during the larval-juvenile
transition (39). Thus, elevation of TH production/release and elevation of T3 by changes in
TH metabolism in target peripheral tissues may produce unique TH responses in each
tissue/cell type during the larval-juvenile transition. In addition, it has been reported that
Page 14 of 48Diabetes
15
Thrα and Thrβ have different expression patterns in rat islets (14). Thrβ is also regulated by
TH and marks mature tissues. Thus, understanding the differences between THRA and B
may provide additional information necessary to understand the molecular mechanisms of
tissue specific transcriptional regulation and tissue maturation by TH.
Neonatal β cells differ from adult β cells in the expression of key metabolic genes
including those involved in mitochondrial ATP production (11). TH has been shown to
stimulate mitochondrial biogenesis and function through activation of PGC-1 and NRF1
expression (40). TH-stimulated rat islets show an increase in pgc1a expression at P7
although it is unclear which cell types are TH responsive (14). Thus, TH signaling may
increase Insulin secretion in β cells through stimulation of mitochondrial function.
Mitochondrial ATP production is also implicated in the inhibition of Glucagon secretion by α
cells (41). Thus, increased mitochondrial metabolic activity in TH-responsive islet cells
could explain the hyperinsulinemia and hypoglucagonemia that we observed in TH-treated
adults. Further analysis of the link between TH signaling and mitochondrial function may
provide key insights into the regulation of islet maturation and function.
TH induced stimulation of Insulin secretion and inhibition of Glucagon secretion led
to reduced fasting glucose levels and increased glucose tolerance in larvae and adults.
Unravelling the regulation of TH signaling will help establish its role in the pathophysiology
of diseases including diabetes. In human, both hypothyroidism and hyperthyroidism increase
the incidence of diabetes (42). However, preclinical studies indicate that TH therapy
attenuates hyperglycemia and Insulin resistance in the db/db diabetic mouse model (31).
Therefore, exogenous TH administration within the physiological range may have therapeutic
applications. The acquisition of TH responsiveness during the functional maturation of islets
Page 15 of 48 Diabetes
16
during the larval-juvenile transition provides an in vivo platform to develop
antihyperglycemic therapies modulating this pathway.
Acknowledgement
This work was supported by funds from the Max Planck and EU society to D.Y.R.S.
No potential conflicts of interest relevant to this article were reported.
H.M. and D.Y.R.S. conceived the project. H.M. S.T.M, D. H. and D.Y.R.S.
contributed to discussion of data and manuscript preparation. H.M., S.T.M., and Y.Z
generated data. H.M. and D.Y.R.S. are the guarantors of this work and, as such, had full
access to all the data presented in the study and take responsibility for the data and the
accuracy of the data analysis.
We thank Dirk Meyer (University of Innsbruck) for the mnx1:eGFP line, David
Parichy (University of Washington) for the tg:nVenus-2a-nfsB line, Yu Hsuan Carol Yang,
Michelle Collins, Jasmin Gäbges and Einat Blitz (MPI-BN) for critical reading of the
manuscript, Sabine Fischer for the animal protocols, all members of the Stainier lab for
helpful discussion and the fish facility staff for fish care.
References
1. Henning SJ. Postnatal development: coordination of feeding, digestion, and
metabolism. Am J Phsyol 1981;241:199-214.
2. Crosnier C, Vargesson N, Gschmeissner S, Ariza-McNaughton L, Morrison A, Lewis J.
Delta-Notch signalling controls commitment to a secretory fate in the zebrafish
intestine. Development 2005;132:1093-1104.
Page 16 of 48Diabetes
17
3. Matsuda H, Shi YB. An essential and evolutionarily conserved role of protein arginine
methyltransferase 1 for adult intestinal stem cells during postembryonic development.
Stem Cells 2010;28:2073-2083.
4. Sirakov M, Plateroti M. The thyroid hormones and their nuclear receptors in the gut:
from developmental biology to cancer. Biochim Biophys Act 2011;1812:938-946.
5. Ishizuya-Oka A, Shi YB. Evolutionary insights into postembryonic development of
adult intestinal stem cells. Cell Biosci 2011;1:37.
6. Ninov N, Hesselson D, Gut P, Zhou A, Fidelin K, Stainier DY. Metabolic regulation of
cellular plasticity in the pancreas. Curr Biol 2013;23:1242-1250.
7. Stolovich-Rain M, Enk J, Vikesa J, Nielsen FC, Saada A, Glaser B, Dor Y. Weaning
triggers a maturation step of pancreatic β cells. Dev Cell 2015;32:535-545.
8. Scaglia L, Cahill CJ, Finegood DT, Bonner-Weir S. Apoptosis participates in the
remodeling of the endocrine pancreas in the neonatal rat. Endocrinology
1997;138:1736-1741.
9. Bliss CR, Sharp GW. Glucose-induced insulin release in islets of young rats: time-
dependent potentiation and effects of 2-bromostearate Am J Physiol 1992;263:E890-
896.
10. Aguayo-Mazzucato C, Koh A, El Khattabi I, Li WC, Toschi E, Jermendy A, Juhl K,
Mao K, Weir GC, Sharma A, Bonner-Weir S. Mafa expression enhances glucose-
responsive insulin secretion in neonatal rat beta cells. Diabetogia 2011;54:583-593.
11. Jermendy A, Toschi E, Aye T, Koh A, Aguayo-Mazzucato C, Sharma A, Weir GC,
Sgroi D, Bonner-Weir S. Rat neonatal beta cells lack the specialised metabolic
phenotype of mature beta cells. Diabetologia 2011;54:594-604.
Page 17 of 48 Diabetes
18
12. Blum B, Hrvatin SS, Schuetz C, Bonal C, Rezania A, Melton DA. Functional beta-cell
maturation is marked by an increased glucose threshold and by expression of urocortin
3. Nat Biotechnol 2012;30:261-264.
13. Borden P, Houtz J, Leach SD, Kuruvilla R. Sympathetic innervation during
development is necessary for pancreatic islet architecture and functional maturation.
Cell Rep 2013; 4:287-301.
14. Aguayo-Mazzucato C, Zavacki AM, Marinelarena A, Hollister-Lock J, El Khattabi I,
Marsili A, Weir GC, Sharma A, Larsen PR, Bonner-Weir S. Thyroid hormone
promotes postnatal rat pancreatic β-cell development and glucose-responsive insulin
secretion through MAFA. Diabetes 2013;62:1569-1580.
15. Pagliuca FW, Millman JR, Gurtler M, Segal M, Van Dervort A, Ryu JH, Peterson QP,
Greiner D, Melton DA. Generation of functional human pancreatic β cells in vitro. Cell
2014;159(2):428-439.
16. Rezania A, Bruin JE, Arora P, Rubin A, Batushansky I, Asadi A, O'Dwyer S,
Quiskamp N, Mojibian M, Albrecht T, Yang YH, Johnson JD, Kieffer TJ. Reversal of
diabetes with insulin-producing cells derived in vitro from human pluripotent stem cells.
Nat Biotechnol 2014;32:1121-1133.
17. Aïello V, Moreno-Asso A, Servitja JM, Martín M. Thyroid hormones promote
endocrine differentiation at expenses of exocrine tissue. Exp Cell Res 2014;322:236-
248.
18. Kowalik MA, Perra A, Pibiri M, Cocco MT, Samarut J, Plateroti M, Ledda-Columbano
GM, Columbano A. TRbeta is the critical thyroid hormone receptor isoform in T3-
induced proliferation of hepatocytes and pancreatic acinar cells. J Hepatol 2010;53:686-
692.
Page 18 of 48Diabetes
19
19. Field HA, Dong PD, Beis D, Stainier DY. Formation of the digestive system in
zebrafish. II. Pancreas morphogenesis. Dev Biol 2003;261:197-208.
20. Kinkel MD, Prince VE. On the diabetic menu: zebrafish as a model for pancreas
development and function. Bioessays 2009; 31:139-152.
21. Tiso N, Moro E, Argenton F. Zebrafish pancreas development. Mol Cell Endocrinol
2009;312:24-30.
22. Curado S, Anderson RM, Jungblut B, Mumm J, Schroeter E, Stainier DY. Conditional
targeted cell ablation in zebrafish: A new tool for regeneration studies. Dev Dyn
2007;236:1025-1035.
23. Huang H, Vogel SS, Liu N, Melton DA, Lin S. Analysis of pancreatic development in
living transgenic zebrafish embryos. Mol Cell Endocrinol 2001;177:117-124.
24. Zecchin E, Filippi A, Biemar F, Tiso N, Pauls S, Ellertsdottir E, Gnugge L, Bortolussi
M, Driever W, and Argenton F. Distinct delta and jagged genes control sequential
segregation of pancreatic cell types from precursor pools in zebrafish. Dev Biol
2007;301: 192‐204.
25. Delporte FM, Pasque V, Devos N, Manfroid I, Voz ML, Motte P, Biemar F, Martial JA,
Peers B. Expression of zebrafish pax6b in pancreas is regulated by two enhancers
containing highly conserved cis-elements bound by PDX1, PBX and PREP factors.
BMC Dev Biol 2008;8:53.
26. Obholzer N, Wolfson S, Trapani JG, Mo W, Nechiporuk A, Busch-Nentwich E, Seiler
C, Sidi S, Söllner C, Duncan RN, Boehland A, Nicolson T. Vesicular glutamate
transporter 3 is required for synaptic transmission in zebrafish hair cells. J Neurosci
2008;28: 2110-2118.
Page 19 of 48 Diabetes
20
27. Arkhipova V, Wendik B, Devos N, Ek O, Peers B, Meyer D. Characterization and
regulation of the hb9/mnx1 beta-cell progenitor specific enhancer in zebrafish. Dev
Biol
2012;365: 290–302.
28. McMenamin SK, Bain EJ, McCann AE, Patterson LB, Eom DS, Waller ZP, Hamill JC,
Kuhliman JA, Eisen JS, Parichy DM. Thyroid hormone-dependent adult pigment cell
lineage and pattern in zebrafish. Science 2014;345:1358-1361.
29. Pisharath H, Rhee JM, Swanson MA, Leach SD, Parsons MJ. Targeted ablation of beta
cells in the embryonic zebrafish pancreas using E. coli nitroreductase. Mech Dev
2007;124:218-29.
30. Djiotsa J , Verbruggen V, Giacomotto J, Ishibashi M, Manning E, Rinkwitz S,
Manfroid I, Voz ML, Peers B. Pax4 is not essential for beta-cell differentiation in
zebrafish embryos but modulates alpha-cell generation by repressing arx gene
expression. BMC Dev Biol 2012;12:37.
31. Lin Y, Sun Z. Thyroid hormone potentiates insulin signaling and attenuates
hyperglycemia and insulin resistance in a mouse model of type 2 diabetes. Br J
Pharmacol 2011;162:597-610.
32. Mohan V, Sinha RA, Pathak A, Rastogi L, Kumar P, Pal A, Godbole MM. Maternal
thyroid hormone deficiency affects the fetal neocorticogenesis by reducing the
proliferating pool, rate of neurogenesis and indirect neurogenesis. Exp Neurol
2012;237:477-488.
33. Parichy DM, Elizondo MR, Mills MG, Gordon TN, Engeszer RE. Normal table of
postembryonic zebrafish development: staging by externally visible anatomy of the
living fish. Dev Dyn 2009;238:2975-3015.
Page 20 of 48Diabetes
21
34. Gosmain Y, Katz LS, Masson MH, Cheyssac C, Poisson C, Philippe J. Pax6 is crucial
for β-cell function, insulin biosynthesis, and glucose-induced insulin secretion. Mol
Endocrinol 2012; 26:696–709
35. Hart AW, Mella S, Mendrychowski J, van Heyningen V, Kleinjan DA. The
developmental regulator Pax6 is essential for maintenance of islet cell function in the
adult mouse pancreas . PLoS One 2013; 8:e54173
36. Ahmad Z, Rafeeq M, Collombat P, Mansouri A. Pax6 inactivation in the adult pancreas
reveals ghrelin as endocrine cell maturation marker. PLoS One 2015;10:e0144597.
37. Nishimura W, Rowan S, Salameh T, Maas RL, Bonner-Weir S, Sell SM, Sharma A.
Preferential reduction of β cells derived from Pax6–MafB pathway in MafB deficient
mice. Dev Biol 2008; 314:443–456
38. Guo C, Chen X, Song H, Maynard MA, Zhou Y, Lobanov AV, GladyshevVN, Ganis
JJ, Wiley D, Jugo RH, Lee NY, Castroneves LA, Zon LI, Scanlan TS, Feldman HA,
Huang SA. Intrinsic Expression of a Multiexon Type 3 Deiodinase Gene Controls
Zebrafish Embryo Size. Endocrinology 2014;155:4069–4080.
39. Brown DD. The role of thyroid hormone in zebrafish and axolotl development. Proc
Natl Acad Sci U S A 1997;94:13011-13016.
40. Weitzel JM, Iwen KA. Coordination of mitochondrial biogenesis by thyroid hormone.
Mol Cell Endocrinol 2011;342:1-7.
41. Quesada I, Tudurí E, Ripoll C, Nadal A. Physiology of the pancreatic alpha-cell and
glucagon secretion: role in glucose homeostasis and diabetes. J Endocrinol.
2008;199:5-19.
42. Wang C. The Relationship between Type 2 Diabetes Mellitus and Related Thyroid
Diseases. J Diabetes Res 2013:390534.
Page 21 of 48 Diabetes
22
Figure legends
Figure 1. T3 stimulates insulin expression, reduces glucose levels, and improves glucose
tolerance. (A-C) Zebrafish were treated with 0, 0.1, 1, 10 or 100 nM T3 from 4 to 7 dpf and
analyzed at 7 dpf for insulin promoter activity (B) and for glucose levels (C). Note that
insulin promoter activity was significantly enhanced by 10 nM T3 treatments, and that
glucose levels were reduced by 10 and 100 nM T3 treatments. (D-F) Zebrafish were treated
with 10 nM T3 from 4 dpf and analyzed for insulin promoter activity (E) and glucose levels
(F) at 5, 6, 7 and 8 dpf. Note that insulin promoter activity was enhanced at 7 and 8 dpf by
T3 treatment (E), and that glucose levels were reduced at 6, 7 and 8 dpf by T3 treatment (F).
(G and H) Larvae were first treated with 10 nM T3 from 4 to 7 dpf, then with 5 % glucose for
1 hour and finally analyzed for glucose levels. Glucose tolerance was compared after
treatment with and without T3 (H). Values of bioluminescence signal were derived from 5
wells per condition, with 3 animals per well. Values of glucose were derived from 5 wells
per condition, with extracts from 20 animals per well. *, P<0.05; **, P<0.01 compared to
controls by Turkey-Kramer HSD test after ANOVA; ns, not significant. AU, arbitrary units.
Figure 2. T3 enhances insulin expression levels, but does not change insulin expression
pattern or β cell numbers. (A) Expression of insulin (ins) mRNA was compared between
control and T3-treated larvae by qPCR. Endogenous ins expression was enhanced after 3
days of T3 treatment (7 dpf). (B and C) Expression pattern of ins was compared between
control (B) and T3-treated (C) larvae at 7 dpf by in situ hybridization. ins expression was
detected only in pancreatic β cells (arrow), but not in other tissues in control (B) and T3-
treated (C) larvae. (D-H) Tg(ins:eGFP) animals were treated with 10 nM T3 starting at 4 dpf
and analyzed at 7 dpf. eGFP expression mimicked the endogenous ins expression pattern in
control (D) and T3-treated (E) larvae. However, eGFP intensity was enhanced after T3
Page 22 of 48Diabetes
23
treatment (D-H). (I-K) Tg(ins:H2BGFP) animals were treated with 10 nM T3 starting at 4
dpf and analyzed at 7 dpf. Note that the β cell number did not change with T3 treatment (K).
Scale bars in B, C, D and E, 200 µm; scale bars in F,G, I and J, 10 µm. **, P<0.01 compared
to controls by Turkey-Kramer HSD test after two-way ANOVA in A. *, P<0.05 compared to
controls by Student’s t-test in H and K. ns, not significant.
Figure 3. T3 represses glucagon expression. (A) Expression of glucagon a (gcga) and
glucagon b (gcgb) mRNA was compared after treatment with and without T3 by qPCR. The
expression of gcga and gcgb was reduced at 5, 6 and 7 dpf by T3 treatment. (B) Expression
patterns of gcga and gcgb were compared after treatment with and without T3 by in situ
hybridization. Both gcga and gcgb expression levels appeared to be reduced after T3
treatment. (C and D) Tg(gcga:eGFP) animals were treated with 10 nM T3 starting at 4 dpf
and the eGFP expression levels in the islets were analyzed at 7 dpf. Note that eGFP intensity
was decreased after T3 treatment (D). (E) Tg(gcga:eGFP) positive cell number was
compared after treatment with and without T3. Note that the α cell number was decreased
after T3 treatment. Scale bar in B, 100 µm; scale bar in C, 10 µm. **, P<0.01 compared to
controls by Turkey-Kramer HSD test after two-way ANOVA in A. *, P<0.05 compared to
control by Student’s t-test in D and E. ns, not significant.
Figure 4. T3 modulates the expression of endocrine differentiation markers. (A-F)
Tg(pax6b:eGFP) (A and B), Tg(neurod1:eGFP) (C and D) and Tg(mnx1:eGFP) (E and F)
animals were treated with T3 to investigate its effect on three genes (pax6b, neurod1 and
mnx1) involved in endocrine cell differentiation. Note that eGFP intensities were enhanced
in Tg(pax6b:eGFP) and in Tg(mnx1:eGFP) animals, but not in Tg(neurod1:eGFP) animals
after T3 treatment. (G) qPCR analysis showed that T3 treatment enhances the expression of
Page 23 of 48 Diabetes
24
pax6b, mnx1, and pax4 and reduces the expression of arxa at 6 dpf. Scale bars, 10 µm. *,
P<0.05; **, P<0.01 compared to controls by Student’s t-test. ns, not significant.
Figure 5. Pancreatic α and β cells are targets of thyroid hormone signaling during and
after larval development. (A) Schematic diagram of thyroid hormone reporter construct.
(B-E) Thyroid hormone reporter expression was compared with Glucagon and Insulin
expression in islets at 20 (B), 25 (C) and 30 (D) dpf and in adults (E) by
immunohistochemistry. Note that strong eGFP signal could be detected in both Glucagon
(red arrows) and Insulin (white arrows) positive cells at 30 dpf and in adults. Weak eGFP
signal in Insulin positive cells, but not in Glucagon positive cells, was also present at 25 dpf
(white arrows). (G) Ratio of TRE:eGFP positive cells to Glucagon positive cells as well as
Insulin positive cells. Values were derived from 5 fishes per developmental time point, with a
minimum of 3 sections per fish. Scale bars, 5 µm. *, P<0.05; **, P<0.01 compared to 20 dpf
by Turkey-Kramer HSD test after ANOVA.
Figure 6. Endogenous thyroid hormone regulates endocrine marker expression and
glucose levels during the larval to juvenile transition. (A) Schematic detailing the time
course of Mtz treatment in Tg(tg:nVenus-2a-nfsB) line to generate hypothyroid zebrafish.
(B) Thyroid hormone reporter expression was reduced in 30 dpf Tg(tg:nVenus-2a-
nfsB)/Tg(6xTRE-bglob1:eGFP) fish after Mtz treatment. (C) Morphology of islets in
hypothyroid zebrafish at 30 dpf. No morphological differences could be detected in the
pancreatic islets between control and Mtz-treated animals. (D) Free glucose level
measurements at 30 dpf show that hypothyroid zebrafish exhibited elevated glucose levels.
(E) qPCR analysis of 30 dpf animals treated with or without Mtz reveals that thyroid ablation
resulted in a marked decrease of ins, neurod1, pax6b, mnx1, and pax4 expression, while the
Page 24 of 48Diabetes
25
expression of gcga and arxa was significantly increased. Triangles indicate principal islet.
Arrows point to secondary islets. Scale bars, 200 µm. *, P<0.05; **, P<0.01 compared to
controls by Student’s t-test. ns, not significant.
Figure 7. Thyroid hormone signaling regulates glucose homeostasis by stimulating
Insulin secretion and repressing Glucagon secretion in adult zebrafish. (A) Schematic
time course of T3 and glucose injections. (B) Blood glucose levels in control and T3 treated
adult fish before and after glucose injections. (C) insulin and glucagon expression in adult
pancreas by qPCR after T3 treatment. (D) Plasma Insulin levels after T3 treatment compared
to control. (E) Plasma Glucagon levels after T3 treatment compared to control. (F)
Quantification of D and E. (G) Schematic time course of Mtz and glucose injections. (H)
Blood glucose levels in Tg(tg:nVenus-2a-nfsB) animals treated with Mtz before and after
glucose injection. (I) insulin and glucagon expression in adult pancreas by qPCR in
Tg(tg:nVenus-2a-nfsB) animals treated with Mtz. (J) Plasma Insulin levels in Tg(tg:nVenus-
2a-nfsB) animals treated with Mtz. (K) Plasma Glucagon levels in Tg(tg:nVenus-2a-nfsB)
animals treated with Mtz. (L) Quantification of J and K. *, P<0.05; **, P<0.01 compared to
controls by Student’s t-test. ns, not significant.
Page 25 of 48 Diabetes
4 5 6 7 80
2
4
6
8
age (dpf)
norm
aliz
ed g
luco
se le
vels
controlT3
-60 0 30 60 900
1
2
3
time after glucose treatment (mins)
norm
aliz
ed g
luco
se le
vels
controlT3
4 5 6 7 80
5
10
15
age (dpf)
controlT3
contr
ol
0.1 nM 1 n
M10
nM
100 n
M0
1
2
contr
ol
0.1 nM 1 n
M10
nM
100 n
M01234567
+T3
A
B
Matsudaetal.,Figure1
C
nsns *
*
+T3
D
E F
****
* **
3 4 5 60 7dpf
4 5 6 70 8dpf
+5%glucosefor1hr
G3 4 5 60 7dpf
+T3
+glucose
H
* * ***
*
insp
romoteractivity
(bioluminescence,AU)
nsns
**ns
norm
alize
dglucoselevels
insp
romoteractivity
(bioluminescence,AU)
Page 26 of 48Diabetes
control T3
0
20
40
60
num
ber o
f ins
:H2B
GFP
+ ce
lls
control T3
0
1
2
3
4
5
ins:
eGFP
inte
nsity
5 6 70
1
2
3
4
age (dpf)
ins
mR
NA
leve
ls
controlT3
ins:eG
FP
control +T3
A B
F
DH
*
*
control
+T3
C
E
G
ins:H2
BGFP
I J
Kcontrol +T3ns
Matsudaetal.,Figure2
20/20
20/20
n=16 n=15
n=15 n=15
0
1
ins
ins
Page 27 of 48 Diabetes
control T3
0
10
20
30
40
num
ber
of gcga:
eGFP
+ ce
lls
control T3
0
1
2
3
gcga
:GFP
inte
nsity
5 6 7 5 6 70
1
mR
NA
leve
ls
controlT3
gcga
:eGF
P
control +T3
* * * *
A
C D
gcga gcgb(dpf)
B
control
+T3
gcga gcgb
*
Matsudaetal.,Figure3
n=18 n=16
**
E
n=18 n=16
16/16 17/17
18/20 17/21
*
01
*
Page 28 of 48Diabetes
neurod1
pax6b
mnx1
pax4
arxa
0
1
2
3
mR
NA
leve
ls
controlT3
control T3
0
1
2
3
4
mnx1:
eGFP
inte
nsity
control T3
0
1
2
3
neurod1:
eGFP
inte
nsity
control T3
0
1
2
3
pax6b:
eGFP
inte
nsity
neurod
1:eG
FPpa
x6b:eG
FP
control +T3
control +T3
*
ns
mnx1:eG
FP
**control +T3
A B
C D
E F
G
Matsudaetal.,Figure4
n=15 n=15
n=18 n=15
n=15 n=15
01
01
01
*
**
* *
Page 29 of 48 Diabetes
20 25 30adult 20 25 30
adult
0
20
40
60
80
100
30dpf
20dpf
A
B
adult
D
25dpf
C
DAPIa-eGFPa-Insulina-Glucagon
DAPIa-eGFPa-Insulina-Glucagon
DAPIa-eGFPa-Insulina-Glucagon
DAPIa-eGFPa-Insulina-Glucagon
6xTR
E-bg
lob1
:eGF
P6xTR
E-bg
lob1
:eGF
PMatsudaetal.,Figure5
eGFPbglob16xTRE-bglob1:eGFP
E
6xTHbzipTREs
G
α cell β cell
*
** **** **
num
bero
feG
FP+
cells
/tota
l cel
lnum
ber(
%)
Page 30 of 48Diabetes
insgcgagcgb
neurod1pax6bmnx1pax4arxa
0
1
2
3
mR
NA
leve
ls
controlMtz
control Mt
z0
1
2
norm
aliz
ed g
luco
se le
vels
+10mMMtz
20 220 30dpf
A
Btg:nVenus-2A-nfsb/6xTRE-bglob1:eGFP
10/10
10/10
**
Matsudaetal.,Figure6
D
a-Insulin
control
Mtz
Tg:nVenus-2A-nfsb
a-Insulin
control
Mtz
C ins:eGFP gcga:eGFP
E
* *
* * * *
*
Page 31 of 48 Diabetes
Insulin
Glucagon
0
1
2
prot
ein
leve
ls
controlMtz
ins gcga
gcgb
0
1
2
mR
NA
leve
ls
controlMtz
-glucose
+glucose
0
200
400
600
Blo
od g
luco
se (m
g/dl
)
controlMtz
Insulin
Glucagon
024681012
prot
ein
leve
ls
controlT3
ins gcga
gcgb
0
1
2
3
4
mR
NA
leve
ls
controlT3
-glucose
+glucose
0
200
400
600
Blo
od g
luco
se (m
g/dl
)
controlT3
IPinjectionof2.5%glucose45mins
IPinjectionofT3 nofeeding
0 1 2 3daysA
B C
*
* *
*ns
controlT3
α-Insulin
α-Glucagon
controlT3
D E F *
**
Matsudaetal.,Figure7
G
IPinjectionof1%glucose60mins
IPinjectionofMtz nofeeding
0 1 2days
**
*
H I
J K L
****
*
α-Insulin
controlMtz
α-Glucagon
controlMtz
tg:nVenus-2A-nfsb
*
**
Page 32 of 48Diabetes
1
Supplemental Figure 1. T3 regulates glucagon expression in an Insulin-independent
manner. Expression levels of gcga and gcgb mRNA were assessed in 6 dpf insulin mutant
larvae after or without T3 treatment. The expression levels of gcga and gcgb were reduced
after 2 days of T3 treatment.
Supplemental Figure 2. Generation of thyroid hormone reporter zebrafish. (A)
Schematic diagram of thyroid hormone reporter construct. Four different minimal promoters
(cfos, bglob1, tk and Pmini) were tested as candidates to build the optimal reporter construct.
(B) Time course of transient injection assays to compare reporter activities in zebrafish
embryos. (C and D) Results of transient injection assays. Note that the construct with the
bglob1 minimal promoter showed the highest reporter activation upon T3 treatment. (E)
Schematic diagrams of thyroid hormone reporter constructs. (F) Results of transient injection
assays. Note that the 6xTRE-bglob1:eGFP construct was used to generate the thyroid
hormone reporter zebrafish line.
Supplemental Figure 3. T3 induces ubiquitous eGFP expression in Tg(6xTRE-
bglob1:eGFP) larvae. (A-E) Tg(6xTRE-bglob1:eGFP) animals were treated with T3 starting
at 4 dpf and eGFP expression analyzed at 7 dpf. Note that eGFP was strongly induced by T3
treatment. C and E show higher magnification images after dissecting out the digestive
organs. p, pancreas; l, liver; i, intestine. Dotted lines in C outline the pancreas. Scale bars,
200 µm.
Supplemental Figure 4. mafa expression in adult zebrafish pancreas. Two color in situ
hybridization was performed with anti-insulin and anti-mafa probes. We could not detect any
overlapping expression of mafa (blue) and insulin (brown), indicating that in zebrafish mafa
is not expressed in adult beta cells, but only in the exocrine pancreas.
Page 33 of 48 Diabetes
2
Supplemental Material
Luciferase assay
After T3 treatment, samples were incubated in Steady Glo (Promega) as described
previously (1). Bioluminescence signal was analyzed by FLUOstar Omega (BMG
LABTECH).
Glucose measurements and treatments
After T3 treatment, 20 larvae or 10 juveniles per group were ground up and free
glucose levels were determined by using a glucose assay kit (BioVision) as described
previously (1). For the glucose tolerance tests, samples were incubated in 5% glucose in egg
water for 1 hr and then free glucose was measured using the glucose assay kit. For the graphs
shown in Figure 1 and Figure 6D, each sample was normalized to a control sample. For
blood glucose measurements, adult zebrafish were anesthetized in tricaine and then cut by the
anal fin with a knife. Whole blood was analyzed immediately with a glucometer (Abbott).
Glucose treatment of adult zebrafish was performed by injecting 10 µl of 1 or 2.5 % glucose
intraperitoneally.
qRT-PCR
Whole animals were used for RNA isolation from larval zebrafish. For RNA
isolation from juvenile zebrafish, the gut, liver and pancreas were dissected out. For RNA
isolation form adult zebrafish, the pancreases were dissected out. Total RNA from control
and T3-treated larvae was extracted by TRIZOL reagent (Ambion) and then cleaned with
RNA Clean & Concentrator-5 (ZYMO RESEARCH) to remove DNA contamination.
Reverse transcription/polymerase chain reaction (RT–PCR) was performed by using
SuperScript II First-Strand Synthesis System (Invitrogen) according to the manufacturer’s
instructions with 500 ng total RNA as a template.
Page 34 of 48Diabetes
3
qRT-PCR with 2xDynamo Color Flash SYBR Green master mix (Thermo Scientific)
was carried out to quantify gene expression levels on a CFX connect Real-time System (Bio
Rad) with gene-specific primers (Supplemental Table 1). Each sample was normalized to a
control gene (actb). Cq values are listed in Supplemental Table 2.
In situ hybridization
Partial cDNA’s encoding zebrafish insulin, glucagon a and glucagon b were obtained
by reverse transcription (RT)-PCR with RNA extracted from zebrafish at 3 days post
fertilization (dpf) with gene-specific primers (Supplemental Table 1). insulin PCR products
were cloned into pCRII-TOPO vector (Invitrogen) and verified by sequencing. glucagon a
and glucagon b PCR products were cloned into pGEM-T-easy vector (Promega) and verified
by sequencing. To synthesize antisense RNA probes, these plasmids were linearized with
BamHI for insulin, SacII for glucagon a, and SpeI for glucagon b and transcribed with T7
RNA polymerase for insulin and glucagon b and Sp6 RNA polymerase for glucagon a
(Roche Applied Science). Whole-mount in situ hybridizations were performed as described
previously (2).
Fluorescence imaging and immunohistochemistry
Fluorescence images were acquired with a LSM700 or LSM780 laser scanning
confocal microscope (Carl Zeiss), and fluorescence intensity for each transgenic line was
calculated using the ZEN software. Cell numbers were counted manually for ins:H2BGFP or
using the Imaris software (Bitplane) for gcga:eGFP.
Immunohistochemistry was performed as described previously (3). The following
primary antibodies were utilized at the indicated dilutions in 1% FBS in PBS-containing
0.1% Triton X-100 (PBTx): guinea pig anti-Insulin (Thermo Scientific) at 1:300, monoclonal
mouse anti-Glucagon (Sigma Aldrich) at 1:100 and chicken anti-eGFP (Thermo Scientific) at
Page 35 of 48 Diabetes
4
1:500. AlexaFluor-conjugated secondary antibodies (Invitrogen) were used at 1:500-1:1000
dilution with 1% FBS in PBTx.
Plasmid construction, Transient injection assays and transgenesis
To generate Tg(ins:Luc2;cryaa:mCherry)gi3
, luc2 was placed under the control of a
1.1 kb insulin promoter element. The -0.5cryaa:mcherry transgenesis marker was inserted
into the vector in the reverse orientation.
For the thyroid hormone responsive element, we utilized the promoter from the
Xenopus laevis TH/bzip gene, which contains two THR response elements (4). To produce
the reporter plasmid 6xTHbzipTREs-cFos:eGFP, TREs-cFos element containing 3 copies of
TH/bzip promoter (6 TREs) and a mouse cfos minimal promoter was amplified and inserted
into the cFos:eGFP vector (5). Each 6xTHbzipTREs-bglob1:eGFP, 6xTHbzipTREs-tk:eGFP,
and 6xTHbzipTREs-Pmini:eGFP was generated by replacing the cFos promoter of
6xTHbzipTREs-cFos:eGFP with a rabbit bglob1 minimal promoter (6), and a HSV thymidine
kinase minimal promoter (tk) minimal promoter (7) and synthetic mini promoter (Pmini)
promoter (8). bglob1:eGFP was generated by replacing the 6xTHbzipTREs-Pmini region of
6xTHbzipTREs-Pmini:eGFP with a bglob1 minimal promoter. To produce 2xTHbzipTREs-
bglob1:eGFP, a copy of the TH/bzip promoter was replaced by the 6xTHbzipTREs part of
6xTHbzipTREs-bglob1:eGFP. To produce 6xTHbzipTREs (mut)-bglob1:eGFP, a mutated
TH/bzip TREs -bglob1 element, containing 3 copies of a mutated TH/bzip promoter (4) and a
bglob1 promoter was amplified and placed instead of the 6xTHbzipTREs-cFos region of the
6xTHbzipTREs-cFos:eGFP construct.
One-cell stage zebrafish embryos were microinjected with each promoter construct
using a micromanipulator. The injected embryos were transferred to culture dishes and
reared in egg water at 28 °C. Injected embryos were treated with T3 starting at 24 hpf and
analyzed for eGFP signal at 60 hpf.
Page 36 of 48Diabetes
5
To generate Tg(ins:Luc2;cryaa:mCherry)gi3
, the ins:Luc2;cryaa:mCherry plasmid
was co-injected with I-SceI meganuclease into one-cell stage embryos as described
previously (9). To generate Tg(6xTRE-bglob1:eGFP)bns91Tg
, the 6xTHbzipTREs-
bglob1:eGFP plasmid was co-injected with tol2 mRNA into one-cell stage embryos as
described previously (10). Multiple stable lines were established. A single representative
line was used for all experiments.
Plasma Insulin and Glucagon detection
After making a cut at the level of the anal fin with a knife, 1 µl of blood was taken per
fish and placed into a tube containing 5 µl heparin. The blood of 10 fish per condition was
pooled and centrifuged at 13700 g at 4 ºC for 15 mins to separate the cells and plasma. An
equal volume of plasma sample was used for dot blotting. Dot blots were performed as
described previously (11). For Insulin and Glucagon detection, guinea pig anti-Insulin
(1:200) (Thermo Scientific) and monoclonal mouse anti-Glucagon (1:600) (Sigma-Aldrich)
antibodies were used. The membrane was analyzed with a ChemiDoc MP imaging system
(Bio Rad).
Supplemental references
1. Gut P, Baeza-Raja B, Andersson O, Hasenkamp L, Hsiao J, Hesselson D, Akassoglou K,
Verdin E, Hirschey MD, Stainier DY. Whole-organism screening for gluconeogenesis
identifies activators of fasting metabolism. Nat Chem Biol 2013;9: 97–104.
2. Cerveny K, Houart C. Wholemount in situ Hybridization and Antibody Staining. Neural
Development and Genetics of the Zebrafish 2009:103-107.
3. Matsuda H, Parsons MJ, Leach SD. Aldh1-expressing endocrine progenitor cells regulate
secondary islet formation in larval zebrafish pancreas. PLoS One 2013;8:e74350.
Page 37 of 48 Diabetes
6
4. Furlow JD, Brown DD. In vitro and in vivo analysis of the regulation of a transcription
factor gene by thyroid hormone during Xenopus laevis metamorphosis. Mol Endocrinol
1999;13:2076-2089.
5. Fisher S, Grice EA , Vinton RM , Bessling SL , Urasaki A , Kawakami K , McCallion
AS. Evaluating the biological relevance of putative enhancers using Tol2 transposon-
mediated transgenesis in zebrafish. Nat Protoc 2006;1:1297–1305.
6. Parsons MJ, Pisharath H, Yusuff S, Moore JC, Siekmann AF, Lawson N, Leach SD.
Notch-responsive cells initiate the secondary transition in larval zebrafish pancreas. Mech
Dev 2009;126:898-912
7. Shimizu N, Kawakami K, Ishitani T. Visualization and exploration of Tcf/Lef function
using a highly responsive Wnt/β-catenin signaling-reporter transgenic zebrafish. Dev Biol
2012;370:71-85.
8. Weger BD, Weger M, Nusser M, Brenner-Weiss G, Dickmies T. A chemical screening
system for glucocorticoid stress hormone signaling in an intact vertebrate. ACS Chem
Biol 2012;7:1178-1183.
9. Thermes V, Grabher C, Ristoratore F, Bourrat F, Choulika A, Wittbrodt J, Joly JS. I-SceI
meganuclease mediates highly efficient transgenesis in fish. Mech Dev 2002;118:91-98.
10. Kawakami K. Transgenesis and gene trap methods in zebrafish by using the Tol2
transposable element. Methods Cell Biol 2004;77:201–224.
11. Olsen AS, Sarras MP Jr, Intine RV. Limb regeneration is impaired in an adult zebrafish
model of diabetes mellitus. Wound Repair Regen 2010;18:532-542.
12. Hesselson D, Anderson RM, Beinat M, Stainier DY. Distinct populations of quiescent
and proliferative pancreatic beta-cell identified HOTcre mediated labeling. Proc Natl
Acad Sci U S A 2009;106:14896-14901.
Page 38 of 48Diabetes
7
13. Ye L, Robertson MA, Hesselson D, Stainier DY, Anderson RM. Glucagon is essential
for alpha cell transdifferentiation and beta cell neogenesis. Development 2015;142:1407-
1417.
14. McCurley AT, Callard GV. Characterization of housekeeping genes in zebrafish: male-
female differences and effects of tissue type, developmental stage and chemical treatment.
BMC Mol Biol 2008;9: 102.
Page 39 of 48 Diabetes
gcga
gcgb
0
1
mR
NA
leve
ls
controlT3
Matsudaetal.,SupplementalFigure1
** *
Page 40 of 48Diabetes
6xTHbzipTREs
cfosbglob1tk
PminieGFPxxxminiP
A
B24 60hpf0
+T3
+T3
-T3
+T3
cfos
bglob1
0102030405060708090100
cfos
bglob1 tk
Pmininu
mbe
rofe
GFP+
embryos/total
embryos(%)
withoutT3
withT3
eGFPbglob1bglob1:eGFP
2xTRE-bglob1:eGFP eGFPbglob1
eGFPbglob16xTRE-bglob1:eGFP
E
eGFPbglob16xTRE(mut)-bglob1:eGFP
-T3
D
0102030405060708090100
noTRE
2xTR
E6xTR
E6xTR
E(m
ut)
numbe
rofe
GFP+
em
bryos/totalembryos(%
)
withoutT3
withT3
F
C
Matsudaetal.,SupplementalFigure2
cfos
bglob1 tk
Pmini
Page 41 of 48 Diabetes
4 7dpf0
+T3
control
T3
. . . ..... . . .......
......
. . .
p
li
A
B C
D E
Matsudaetal.,SupplementalFigure3
Page 42 of 48Diabetes
insmafa
Matsudaetal.,SupplementalFigure4
Page 43 of 48 Diabetes
Supplemental Table 1. primers sequence
(A) Primers for qPCR
primer name sequence (5'-->3') reference
insulin -f TTTAAATGCAAAGTCAGCCACCTCAG 12
insulin -r GGCTTCTTCTACAACCCCAAGAGAGA 12
gcga -f AAGGCGACAGCACAAGCACA 13
gcga -r GCCCTCTGCATGACGTTTGACA 13
gcgb -f GGAGACCAGGAGAGCACAAG
gcgb -r TGCAGGTA CGAGCTGACATC
neurod1 -f CTTTCAACACACCCTAGAGTTCCG
neurod1 -r GCATCATGCTTTCCTCGCTGTATG
pax6b -f GCCAGGACAACCAAATCAAGACG
pax6b -r GTGAAGGACGTTCTGTTTCTCTGC
mnx1 -f GAAGAGGAGCGGTGACACTC
mnx1 -r GGTCTGCACTTTTGGTGGAT
pax4 -f GAGAGAGCCTGCAGAGGAGA
pax4 -r TGTTGCTGAAAATGGCTCTG
arxa-f GGAGAGACTGCTGGACCAAG
arxa-r TTCATCACTCTGGCTGAACG
gapdh -f GTGGAGTCTACTGGTGTCTTC 14
gapdh -r GTGCAGGAGGCATTGCTTACA 14
(B) Primers for ISH probe
primer name sequence
insulin -f CCATATCCACCATTCCTCGCC
insulin -r CAAACGGAGAGCATTAAGG CC
gcga -f ATTTGCTGGTGTGTGTTGGA
gcga -r TAGTG TTTGGCACAGGGTGA
gcgb -f GAGAAAACCAGCGAGACGAC
gcgb -r GCATGATTCATACAGCACTGG
Page 44 of 48Diabetes
larval
Cq
5 dpf
actb 25.06
25.11
actb+ 24.99
24.88
ins 25.76
25.73
ins+ 25.9
25.66
6 dpf
actb 24.63
24.69
actb+ 24.53
24.6
ins 25.62
25.63
ins+ 25.77
25.58
7 dpf
actb 24.78
24.85
actb+ 24.96
24.93
ins 25.48
25.23
ins+ 24.2
23.92
5 dpf
actb 25.06
25.11
actb+ 24.99
24.88
gcga 26.42
26.67
gcga+ 27.65
27.42
gcgb 28
27.89
gcgb+ 28.39
28.5
6 dpf
Page 45 of 48 Diabetes
actb 24.63
24.69
actb+ 24.53
24.6
gcga 26.01
25.82
gcga+ 26.4
26.71
gcgb 28.26
28.16
gcgb+ 28.56
28.52
7 dpf
actb 24.78
24.85
actb+ 24.96
24.93
gcga 28.52
28.87
gcga+ 28.81
28.92
gcgb 29.63
29.61
gcgb+ 29.89
30.08
6 dpf
actb 28.5
28.49
actb+ 28.72
28.26
neurod1 29.21
29.25
neurod1+ 29.03
28.75
pax6b 28.87
28.61
pax6b+ 27.75
27.82
mnx1 30.84
31.25
mnx1+ 29.81
30.02
pax4 29.32
Page 46 of 48Diabetes
29.24
pax4+ 28.11
28.27
arxa 29.53
30.12
arxa+ 31.02
31.23
juvenile
30 dpf
actb 24.46
24.11
24.16
actb+ 24.74
24.39
24.3
ins 28.08
28.03
28.02
ins+ 29.13
28.72
28.71
gcga 34.85
35.07
34.9
gcga+ 33.64
33.19
33.26
gcgb 28.28
28.51
28.48
gcgb+ 28.06
28.31
28.17
neurod1 34.54
34.69
34.7
neurod1+ 35.27
35.62
35.37
pax6b 33
32.49
32.7
pax6b+ 34.88
Page 47 of 48 Diabetes
34.78
34.32
mnx1 33.26
33.25
32.75
mnx1+ 34.61
34.14
34.1
pax4 33.07
32.42
32.43
pax4+ 34.56
34.38
33.64
arxa 35.82
35.06
35.41
arxa+ 34.22
34.5
34.89
adult
actb 25.83
25.88
actb+ 25.75
25.79
ins 24
23.91
ins+ 23.15
22.73
gcga 23.16
23.25
gcga+ 24.23
24.3
gcgb 26.18
26.7
gcgb+ 26.15
26.22
Page 48 of 48Diabetes
Recommended