Articles in PresS. Am J Physiol Endocrinol Metab (March 10, 2015). doi:10.1152/ajpendo.00540.2014
1
Depletion of suppressor of cytokine signaling-1a causes hepatic
2
steatosis and insulin resistance in zebrafish
3 4 5
Ziru Dai1, 2, ¶, Hualin Wang3, ¶, Xia Jin2, Houpeng Wang3, Jiangyan He2, Mugen Liu1, Zhan
6
Yin1,2, Yonghua Sun3,*, Qiyong Lou2,*
7 8
1
9
and Technology, Center for Human Genome Research, Huazhong University of Science and
Key Laboratory of Molecular Biophysics of Ministry of Education, College of Life Science
10
Technology, Wuhan, Hubei 430074, P. R. China
11
2
12
3
13
Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan, Hubei, 430072, China
Key Laboratory of Aquatic Biodiversity and Conservation of the Chinese Academy of Sciences, State Key Laboratory of Freshwater Ecology and Biotechnology, Wuhan, Hubei, 430072, China
14 15
Running title: Lipodystrophy caused by socs1a depletion in zebrafish
16
¶
17
*
18
0086-27-68780069), Yonghua Sun, Email:
[email protected] (Tel. 0086-27-68780235, Fax:
19
0086-27-68780123).
These authors contribute equally to this work. Corresponding
authors:
Qiyong
Lou,
Email:
[email protected]
(Tel.
20 21
Key words: zebrafish, socs1a, gene targeting, hepatic steatosis, insulin resistance
22
Word count: Abstract: 245 words; Main text: 4847 words; Figure: 6
23
Copyright © 2015 by the American Physiological Society.
and
Fax:
24
Abstract
25
Suppressor of cytokine signaling-1a (SOCS1a) is a member of the suppressor of cytokine
26
signaling family, a group of related molecules that mediate the negative regulation of the
27
JAK-STAT pathway. Here, we depleted SOCS1a using the transcription activator-like (TAL)
28
effector nucleases (TALENs) technique to understand its physiological roles in zebrafish.
29
Although elevated levels of JAK-STAT5 activation and erythropoiesis have been observed in
30
socs1a-deficient zebrafish, these animals exhibited normal growth during the early stages.
31
Socs1a-deficient zebrafish began to grow slowly with certain mortalities after 20 days post
32
fertilization (dpf), while the heterozygous socs1a-deficient zebrafish exhibited enhanced
33
somatic growth. Decreased adiposity, hepatic steatosis, and insulin resistance were observed
34
in our socs1a-deficient adult zebrafish, which is similar to the lipodystrophy phenotypes
35
observed in mammals. Comparative transcriptomic analyses revealed elevated levels of
36
gluconeogenesis, lipolysis and hypoxia-inducible response and decreased activities of
37
lipogenesis and glycolysis in the hepatocytes of socs1a-deflicient adult zebrafish. Evident
38
mitochondrial dysfunction has also been observed in hepatocytes from socs1a-deficient
39
zebrafish. Taken together, our results suggest that the negative regulatory roles of SOCS1a on
40
JAK-STAT5 signaling may be involved in the suppression of the erythropoiesis and growth
41
hormone activities, which was also reflected with the fact of the enhanced somatic growth
42
performance observed in the heterozygous socs1a-deficient fish. The differences in the effects
43
caused by SOCS1a depletion on insulin sensitivity, lipid metabolism and inflammatory
44
responses between zebrafish and mammalian models observed here may reflect differences
45
between the functional mechanisms of SOCS members in terrestrial mammals and aquatic
46
teleosts.
47
48
INTRODUCTION
49
The emergency of various secreted signals and their receptors, including various cytokines
50
and hormones, enable cells in multicellular organisms to respond to distinct cues. The
51
JAK-STAT signaling pathway, which is one of the core transmission systems that mediate the
52
extracellular signals for the activation of many key signal cascades involved in various
53
physiological processes. Many cytokines or hormones that involves in many inflammatory
54
cytokine, GH, and prolactin signaling pathways have been proposed (32). The JAK-STAT
55
pathway can be negatively regulated at multiple levels. First, protein tyrosine phosphatases
56
can remove phosphate from receptors and activated STAT molecules. More recently, members
57
from a small protein family identified as suppressors of cytokine signaling (SOCS) have been
58
found to form negative feedback loops to inhibit JAK-STAT signaling (1). Based on mouse
59
genetic studies, distinguishable physiological roles for several members of the SOCS family
60
have been delineated. Among the mouse SOCS proteins, SOCS1 has been shown to be mainly
61
involved in IFNγ responses and prolactin signaling, whereas SOCS3 plays regulatory roles in
62
leukemia inhibitory factor and IL-6 signaling and placental development (15). The
63
observation of gigantism in SOCS2-deficient mice provides clear evidence that SOCS2 is a
64
negative regulator of GH signaling (29). Moreover, there is a growing body of evidence
65
suggesting a role for SOCS proteins in insulin signaling (4, 17, 24, 42).
66
Somatic growth is a tightly regulated process that is dependent on GH signaling and
67
action. Related to its role on postnatal growth promotion, GH has many other potent effects,
68
including lipid, glucose, and mineral metabolism (35). GH binds to its cognate receptor and
69
activates a number of transcription factors, including STAT5, mediated through intracellular
70
JAK2 activation (25). In mammals, the activation of JAK2-STAT5 signaling via GH leads to
71
the transcriptional activation of a wide range of target genes, including IGF-I, anti-apoptotic
72
genes, and SOCS2. Insulin is a peptide hormone produced by β-cells in the pancreas. Its main
73
function is to promote the absorption of glucose from the blood to other tissues and the
74
storage of fat for use as energy (17). Normally, GH also influences insulin signaling by
75
antagonizing the action of insulin on glucose and lipid homeostasis in diverse tissues (7). In
76
mammal animal models, chronic GH excess may promote insulin resistance by increasing
77
hepatic glucose production and triglyceride storage (10, 21).
78
The zebrafish SOCS family contains at least 12 members (http://zfin.org/). Additional
79
genome duplication also occurs in the stem lineage of teleost fishes to yield duplicate
80
members of SOCS genes in the zebrafish genome corresponding to some of the mammal
81
SOCS members (34). Zebrafish socs1a has been found to exhibit significant expression in the
82
liver (20). The expression of zebrafish socs1a may be induced by LPS and poly I:C,
83
suggesting its potential roles as an inhibitor of IFN signaling (30). However, it has also been
84
reported that socs1a may also be stimulated significantly in the livers of GH- or GHR-
85
overexpressing transgenic zebrafish, indicating its involvement in teleost GH signaling
86
pathways (13, 39).
87
Utilizing a dozen of powerful genetically modified mouse models, the physiological
88
functions of each SOCS family member have been carefully defined (32). However, there is
89
no similar genetic model for non-mammalian species. The specificity within the SOCS family
90
of proteins in an in vivo teleost model has not yet been elucidated. To address these questions,
91
we have generated zebrafish that lack the SOCS1a protein. These animals exhibited a
92
relatively healthy juvenile growth. However, the socs1a-null zebrafish begin to grow slowly
93
once they reach adulthood with certain casualties. The loss of socs1a in zebrafish was also
94
found to result in high phosphorylation levels of STAT5, high plasma levels of insulin, hepatic
95
steatosis, and decreased adipose tissue in adults. In additional, elevated levels of
96
gluconeogenesis and lipolysis in the socs1a-deficient liver have been revealed through
97
comparative transcriptomic studies of the zebrafish liver. In contrast, there are no significant
98
variations in the expression levels of the typical inflammatory cytokines, such as IFN-γ, IL-1,
99
IL-6, and TNFα, between the wild-type control and their socs1a-deficient siblings. Taken
100
together, our studies demonstrate that socs1a gene deficiency in zebrafish results in decreased
101
insulin sensitivity and hepatic steatosis with chronic excess-activated GH signaling.
102 103
MATERIALS AND METHODS
104
Zebrafish husbandry. AB-line zebrafish were maintained at a 14-h light/10-h dark
105
rhythm in circulated water at 28.5°C (22). The zebrafish were fed newly hatched brine shrimp
106
and TetraMin Tropical Fish food flakes (Tetra, Germany) three times a day. The fat
107
composition in both food supplies are 15% and 8%, respectively (8). Embryos were obtained
108
through natural spawning and cultured at 28.5°C in Ringer’s solution. Their developmental
109
stages were determined according to hours post fertilization (hpf) at 28.5°C or following the
110
morphological features previously described (22). All of the procedures for experimental
111
animal manipulation were approved by the Animal Research and Ethic Committee of the
112
Institute of Hydrobiology of the Chinese Academy of Sciences.
113
Socs1a knockout via TALENs. The construction and the sequence-specific TALEN
114
effector repeats were performed following the procedures described by Huang et al. (18). The
115
four basic single unit vectors NI, NG, NN and H that recognized ATGC were assembled using
116
NheI and SpeI. To generate capped mRNA containing DNA-binding TALEN repeats and the
117
FokI endonuclease domain, the TALEN expression vectors were linearized with NotI and
118
transcribed using a Sp6 mRNA kit (mMessage mMachine kit, Ambion USA, AM1340).
119
Capped mRNA was injected into wild-type embryos at the one- or two-cell stage.
120
Approximately 10 pooled F0 embryos were lysed for genomic DNA isolation, and the target
121
region was then amplified and digested with StuI. The primers are listed below in Table 1. The
122
remaining mutation-positive larvae were raised to adulthood and outcrossed with wild-type
123
fish. The F1 larvae were analyzed by StuI digestion and genomic DNA and cDNA sequencing.
124
The F1 transgenic larvae were raised to adulthood and self-crossed to obtain F2 homozygous
125
offspring. The fin genomic DNA was extracted for genotyping. Two independent socs1a
126
mutant lines were obtained (Fig. 1). However, since the second mutant line (mutant line 2,
127
M2) has been generated much later, so most of the assays were performed with the fish of the
128
mutant line 1 (M1) unless specifically stated.
129
Messenger RNA synthesis and microinjection. The full-length zebrafish socs1a cDNA
130
flanked with two restriction digestion sites in the primers was amplified and sub-cloned into
131
the Psp64 construct. The construct was linearized with EcoRI and transcribed using the
132
SP6mRNA kit (mMessage mMachine kit, Ambion USA, AM1340). Sos1a-deficient female
133
and socs1a-deficient male zebrafish were crossed to obtain socs1a-deficient embryos. Half of
134
the embryos were injected with socs1a mRNA in PBS solution at the one- or two-cell stage,
135
and the other half were injected with PBS saline solution as controls. Five days after injection,
136
specimens were collected for subsequent experimentation.
137
Histological analysis and Oil red-O staining. The frozen livers were cryosectioned, and
138
the sections were subjected to Hematoxylin and Eosin staining. The frozen sections were also
139
stained with Oil red O to visualize the fat deposits in the liver as previously described (2). The
140
stained sections were visualized under a light field, and images were captured at different
141
magnifications.
142
Blood sugar measurement in zebrafish. An OneTouch UltraVue (LifeScan) glucose
143
meter was used for the measurement of blood glucose, which needs 1 µl of sample for each
144
measurement. Blood collection was performed as previously described (11). The zebrafish
145
were fed a chow diet regularly, and the zebrafish were fasted as previously described (11). For
146
the glucose treatment assay, the glucose-treated zebrafish were placed in 5% glucose solution
147
for 8 h and then rinsed in Ringer’s solution for 10 min (11).
148
Triglyceride (TG) content Assay. Triglyceride assay Kit was purchased from BioAssay
149
Systems (ETGA-200, CA94545 USA). The liver samples from the adult zebrafish at 120 dpf
150
stage were solubilized in ethanoic KOH and then neutralization with MgCl2 as described
151
(31). The assay was performed following the instructions supplied by the manufacturer.
152
Optical density of the samples was measured using SPECTRAMAX M2 (Molecular Device,
153
USA). Each 5 individuals per gender type were used for each genotype. The experiment was
154
repeated once.
155
Western blot analysis. Protein extraction buffer was purchased from Thermo (Thermo
156
89990, USA), and protein extraction was performed according to the manufacturer’s
157
instructions. STAT5.1 antibody was purchased from Sigma (SAB2102320, Sigma; used at
158
1:1000 dilution). Phosphorylated Stat5.1 antibody was purchased from Millipore (05-495,
159
Millipore; used at 1:1000 dilution). Akt antibody was purchased from CST (#4691, Cell
160
Signaling Technology; used at 1:1000 dilution). Phosphorylated Akt antibody was purchased
161
from CST (#4060, Cell Signaling Technology; used at 1:1000 dilution). SDS-PAGE was
162
performed as described previously (27).
163
Total RNA extraction and gene expression levels quantified with real-time PCR analysis.
164
Total RNA was extracted using RNeasy Mini Kit (Qiagen, Germany). The quality of the total
165
RNA was verified with a microspectrophotometer (Eppendorf) and through electrophoresis.
166
Total RNA was reverse-transcribed with MMLV reverse transcriptase (Thermo).
167
Complementary DNA was diluted 1:50 prior to use. SYBR Green mix was purchased from
168
Toyobo Biotech (Japan). The primers used for real-time PCR are listed in Table 1. To analysis
169
the gene expression at larval stages or adult stages, total RNA samples were extracted from at
170
least 40 larvae each genotype, or at least 4 adults each genotype. The expression level of the
171
β-actin gene was used as the internal control for normalization. The experiments were
172
performed at least for 3 duplications.
173
Transcriptome analysis. The total RNA were extracted from the liver tissues of
174
socs1a-deficient and control zebrafish from the M1 mutant line at the 90-dpf stage. The
175
constructions of the cDNA library, and cDNA sequencing and comparison were performed by
176
BGI Co. Ltd. The expression analyses were performed as previously described (16).
177
Whole-mount in situ hybridization. The riboprobe of zebrafish gata1 was previously
178
described (33). Antisense RNA probes labeled with digoxigenin-UTP (Roche, Germany) were
179
synthesized and used for whole-mount in situ hybridization as previously described (27).
180
ELISA. The protein extraction buffer for ELISA was prepared as follows: 10 mM Tris-Cl
181
pH 7.4, 0.1 mM EDTA, 10 mM sucrose, and 0.8% NaCl. The volume of the protein extraction
182
buffer was added to 10 ml per gram of liver tissue (wet weight). The liver was homogenized
183
using an ultrasonic processor on ice and centrifuged at 2000×g for 10 min. ELISA was
184
performed following the instructions supplied by the manufacturer. ELISA kits for zebrafish
185
insulin (H203) and glucagon (H183) were purchased from Nanjing Jiancheng Bioengineering
186
Institute (Nanjing, China).
187
Statistical analysis. T tests were performed for each experiment. Each result represents
188
the mean of at least three independent experiments. The error bars represent the standard
189
deviations. The p values were calculated and are indicated in the figure legends.
190 191
RESULTS
192
Generation of socs1a-deficient zebrafish. The TALEN-based gene-targeting procedure
193
was used for the depletion of the socs1a gene zebrafish according to a previously published
194
method (18). Based on sequence information of zebrafish socs1a (NM_001003467), the
195
targeting region designed for socs1a was located at the second exon of the gene locus, and
196
both arms were 18 bp in length. The spacer length between the recognition arms was a 17-bp
197
fragment with a StuI site (Fig. 1A), which could be efficiently used for testing the mutation
198
within the spacer region. To test the efficiency of the depletion, the targeting region of the
199
socs1a locus was amplified from the genomic DNA of F0 embryos extracted at 3 dpf and then
200
subjected to digestion with StuI. Part of the amplified fragment could not be cut with StuI,
201
which indicates the existence of an indel within the targeting region after application of the
202
depletion procedure. When the F0 zebrafish reached adulthood, positive individuals were
203
identified by tail genomic DNA amplification and StuI digestion as a F0 founders. Subsequent
204
StuI digestion screening for individual juveniles revealed that the offspring of heterozygous
205
parents included fish of each of the three expected genotypes in M1 individuals (Fig. 1B). The
206
sequence analysis of the tail genomic DNA and transcripts of socs1a in homozygous mutant
207
fish confirmed that two independent mutant lines have been successfully generated as M1 and
208
M2 lines. In these mutants, the deletion of a single nucleotide, namely cytosine (C), or an
209
extra cytosine insertion in the StuI restriction site were found in M1 or M2 mutants
210
respectively (Fig. 1C), which results in premature termination of the mutant SOCS1a
211
containing 63 (M1) or 118 (M2) amino acids, of which only the first 22 amino acid residues
212
are identical to those of native zebrafish (Fig. 1D). After extraction of the total RNA from the
213
whole larvae body, the expression levels of the mutant and native forms of socs1a were
214
increased in socs1a-deficient fish than those of the wild-type control fish (Fig. 1E). This
215
might be results from the feedback compensatory regulation on socs1a transcription due to
216
the loss of the functional SOCS1a in vivo.
217
Conserved functions of zebrafish SOCS1a in Jak-Stat and GH signaling. SOCS proteins
218
have been well recognized as negative regulators of JAK-STAT signaling (1, 32). To confirm
219
the physiological effects of socs1a depletion in zebrafish in vivo, the status of the activation of
220
JAK-STAT signaling was analyzed. Elevated levels of STAT5 phosphorylation were observed
221
in socs1a-deficient zebrafish compared with wild-type control fish at 5 days post fertilization
222
(dpf, Fig. 2A). The application of AG490, an inhibitor of JAK, attenuated the excessive
223
activation of STAT5.1 phosphorylation in the mutant fish, suggesting that the superactivation
224
of STAT5.1 is mediated through JAK (Fig. 2A). It has been suggested that zebrafish SOCS1a
225
plays a regulatory role in erythropoiesis and inflammatory gene expression (30, 33).
226
Consistent with this notion, obvious elevated levels of erythropoiesis have been found in
227
socs1a-deficient fish during the early stages, as determined through an assay of gata1
228
expression at the 24-hpf stage (Fig. 2B) and the Hbbe1.1 (hemoglobin beta embryonic-1.1)
229
protein presence at the 5 dpf stage (Fig. 2C). In addition, elevated levels of erythropoiesis in
230
socs1a-deficient fish may be prevented by the application of AG490, suggesting the
231
involvement of JAK-STAT signaling in the function of zebrafish SOCS1a on erythropoiesis.
232
However, the excessive levels of erythropoiesis in socs1a-deficient fish tend to be moderate
233
during adulthood because only modestly increased numbers of red blood cells were observed
234
in the mutants compared with their wild-type control counterparts at 90 dpf stage (Fig. 2D).
235
To determine the inflammatory status in the socs1a-deficient larvae, the transcriptional
236
expression levels of the several typical inflammatory cytokines have been quantified with the
237
real-time PCR analyses. No significant elevation of the expression levels in socs1a-deficient
238
larvae compared with those of the control fish has been observed at 5 dpf stage (Fig. 2E).
239
At the hatchery, socs1a-deficient fish were undistinguishable from their normal siblings
240
(Fig. 3A), but by the 20-dpf stage, socs1a-deficient fish began to grow slowly compared with
241
their normal siblings (Fig. 3B). The socs1a-deficient fish became ill and died between 40 and
242
120 dpf (Fig. 3B). Most of the socs1a-deficient adult fish exhibited a smaller body size
243
compared with their wild-type control siblings raised in the same tank (Fig. 3D).
244
Hepatic steatosis and decreased adipose tissue in socs1a-deficient zebrafish. For the
245
purpose of healthy examination, the socs1a-deficient zebrafish and their wild-type siblings
246
were dissected. Decreased adipose tissue in the viscera was detected in the socs1a-deficient
247
zebrafish compared with the wild-type fish (Fig. 4A, C, and 4D). Because the liver is the
248
major organ involved in lipid metabolism and homeostasis (36), the pathological features of
249
the liver were examined by Oil red-O staining. Severe lipid accumulation and hepatic
250
steatosis were observed in the liver of socs1a-deficient zebrafish (Fig. 4E-H). Furthermore,
251
the TG content of the liver tissue was measured. As expected, the TG contents in both male
252
and female liver of socs1a-deficient zebrafish from both M1 and M2 were increased
253
compared with their wild control at 120 dpf stage (Table 2).
254
Comparative liver transcriptome analysis through RNA-Seq analysis. Comparative
255
transcriptomics of liver tissues from socs1a-deficient zebrafish and wild-type control fish at
256
90 dpf was performed using the RNA-Seq technique. Based on the RNA-Seq analysis results,
257
some of the transcripts encoding key molecules involved in metabolism were observed to
258
exhibit significantly differential expression (Table 3). We then selected 11 of these key genes
259
as representatives to confirm the expression patterns reflected by the RNA-Seq analysis.
260
These representative transcripts are insulin, insulin receptor b (insrb), phosphoenolpyruvate
261
carboxykinase (pck), glucose-6-phosphatase c family a1 (g6pca.1), hexokinase domain
262
containing 1 (shxk1), glucokinase (gk), O family member of forkhead transcription factors
263
(foxO1a), apolipoprotein A-IV (apoA4), low-density lipoprotein receptor (ldlr), stearoyl-CoA
264
desaturase (scd) and hypoxia induced factor, 1-alpha like (hif1αl). Our real-time RT-PCR
265
results confirmed that all of the 11 transcripts exhibited similar tendencies to those observed
266
in their expression patterns obtained through RNA-Seq analysis (Fig. 5A, Table 3). Among
267
the list of differentially expressed transcripts obtained based on the RNA-Seq analysis and our
268
real-time PCR results, Pck and g6pca1, rate-limiting enzymes of gluconeogenesis, were
269
significantly up-regulated in the liver from socs1a-deficient zebrafish compared with
270
wild-type control fish (Fig. 5A, Table 3). The expression level of a key molecule that
271
promotes the activity of lipolysis, foxO1a, was elevated in the socs1a-deficient liver (Fig. 5A).
272
In contrast, the transcription levels of shxk1, gk, and scd, which are key enzymes involved in
273
glycolysis and lipogenesis, were downregulated in the socs1a-deficient livers (Fig. 5A, Table
274
3). In addition, several genes involved in the hypoxia-induced responsive cascade, such as
275
hif1αl (Fig. 5A), max interactor 1(mxi-1), hypoxia inducible domain family, member 1A
276
(higd1a), angiopoietin-like 4 (Angptl4, or pgar), and heme oxygenase 1a (hmox1, or HO-1),
277
were also increased significantly in the socs1a-deficient liver (Table 3). In addition, the
278
transcriptional levels of many typical inflammatory cytokines, such as IFNγ, IL1b, IL6,
279
TNF-α and IL10, were either unchanged or downregulated in the socs1a-deficient liver
280
compared with the wild-type liver tissues (Table 3, or data not shown).
281
Insulin resistance in socs1a-deficient hepatocytes. The expression levels of insulin and
282
its receptor, insrb, have been found to be upregulated in the socs1a-deficient liver (Fig. 5A,
283
Table 3). The protein levels of insulin in the hepatocytes and plasma were measured. ELISA
284
of hepatic and plasma samples revealed significantly elevated levels of insulin in the
285
socs1a-deficient plasma and hepatocytes compared with those of the wild-type control fish
286
(Fig. 5B, 5C). However, the levels of glucagon in the liver and plasma were similar between
287
socs1a-deficient and wild-type control fish (Fig. 5D, 5E). However, under normal feeding,
288
starvation, and high-glucose conditions, the plasma levels of glucose in the socs1a-dificient
289
fish and wild-type control fish were similar (Fig. 5F). Furthermore, the levels of the
290
phosphorylated AKT levels decreased significantly in the hepatic tissue of socs1a-deficient
291
zebrafish compared with those of the control fish determined by the Western Blot analysis
292
(Fig. 5G), which might be associated with the insulin resistance in socs1a-deficient fish.
293
These observations suggest that the typical phenomenon of insulin resistance was observed in
294
the socs1a-deficient fish.
295
Enhanced somatic growth in the heterozygous socs1a-deficient zebrafish. Contrast to the
296
homozygous socs1a-deficient mutant zebrafish, the heterozygous socs1a mutant zebrafish
297
from both independent lines seems to be healthy. Interestingly, their somatic growth
298
performances have been enhanced compared with their wild-type control siblings (Fig. 6A,
299
6B). Moderated increased levels of the phosphorylated Stat5.1 in the heterozygous
300
socs1a-deficient zebrafish liver have been also observed through Western Blot analyses
301
(Figure 6C, D), which indicating an enhanced GH signaling activation in the heterozygous
302
socs1a-deficient zebrafish as well. However, unlike their homozygous individuals, a partially
303
depletion of the socs1a allele could promote the somatic growth which results an integrated
304
gigantism effects.
305 306
DISCUSSION
307
SOCS1-null mice develop complex fatal neonatal defects featuring fatty degeneration
308
and necrosis of the multiorgan failure as a result of inflammatory infiltration, which appear to
309
be largely due to hyperresponsiveness to IFN-γ (38). The typical hepatosteatosis phenotype in
310
adult SOCS1-null mice could only be induced by a high-fat diet (HFD) after the early death
311
of SOCS1-null mice rescued with a RAG2-deficient background (12). The neutralization of
312
IFNγ in SOCS1/IFNγ-double-null mice improved their whole-body insulin sensitivity due to
313
enhanced insulin action in the liver and greater suppression of hepatic glucose production (19).
314
The zebrafish SOCS1a has been suggested to play roles in hematopoiesis and the IFN
315
signaling pathway similar to those of its mammalian counterpart. It has been demonstrated
316
that the functions of SOCS1a could be mediated by the JAK2-STAT5 pathway (30, 33). The
317
results of this study demonstrated elevated levels of phosphorylated STAT5 and hemoglobin
318
mediated via JAK activation in socs1a-null zebrafish compared with the wild-type control
319
fish during their early and adult stages, which confirmed the roles of SOCS1a in activation of
320
the JAK-STAT5 pathway and hematopoiesis (Fig. 2B, 2C). Based on our RNA-seq analyses,
321
although a moderate upregulation of IFNγ expression was observed in socs1a-deficient fish
322
during the early stages (Fig. 2E), no significant changes of the numbers of the adult
323
erythrocytes, and the expression levels of typical inflammatory cytokines, such as ifn-γ, il-1b,
324
il-6, and tnfα, in adult hepatocytes were observed between socs1a-deficient and control
325
zebrafish (Fig. 2D; Table 3).
326
The up-regulated expression levels of socs1a have been reported in GH over-expressing
327
transgenic zebrafish previously, suggesting that SOCS1a could be also a negative modulator
328
of the somatotrophic axis in zebrafish (13, 39). GH signaling is known to stimulate
329
gluconeogenesis, lipolysis and adipose tissue mobilization in mammals and teleosts (26, 28,
330
36). A central role of the JAK2-STAT5 pathway in mediating the growth-stimulating action of
331
GH signaling has been reported in teleosts (23). The expression levels of Foxo1a, pck and
332
G6Pase, a key regulator and two enzymes controlling the gluconeogenesis, were significantly
333
increased in the liver of socs1a-deficient zebrafish compared with those of the control fish
334
(Fig. 5A). Moreover, a significant decrease in the size of adipose tissue was observed in the
335
socs1a-deficient zebrafish compared with the control fish (Fig. 4A, C and D), which is
336
opposite against the phenotype of increased adiposity in gh-deficient zebrafish (28). In
337
addition, the enhanced somatic growth performance has been observed in the heterozygous
338
socs1a-deficient fish (Fig. 6A). Taken together, all these findings revealed that socs1a
339
deficiency results in a chronic excess activation of GH signaling in zebrafish. In contrast, the
340
expression levels of pck and G6Pase either remained unchanged in SOCS1-null mice or
341
significantly decreased in SOCS1/IFNγ-double-null mice (12, 19), providing further evidence
342
regarding the differences in the physiological functions and signaling mediated by zebrafish
343
SOCS1a and mouse SOCS1.
344
Energy metabolism and somatic growth are closely coordinated in animals, and the liver
345
is a key metabolic organ that governs body energy metabolism (36). With an evident defect in
346
somatic growth in the homozygous socs1a-deficient adult zebrafish (Fig. 3), the dynamical
347
status of the liver was our focus in this study. At 90 dpf, liver steatosis was observed in
348
socs1a-deficient fish (Fig. 4E-H; Table 2), as could also be observed in SOCS1-, SOCS2-,
349
and SOCS3-deficient mice only in the presence of HFD (12, 37, 42). Moreover, significantly
350
increased plasma insulin levels were observed in socs1a-null fish, and the mutant fish did not
351
exhibit improved plasma glucose clearance compared with the control fish (Fig. 5). However,
352
the increased levels of insulin in socs1a-deficient fish did not stimulate lipogenesis and
353
glycolysis in the liver, which was reflected through comparative transcriptomics of liver
354
tissues from control and mutant fish (Fig. 5F). This finding suggests that socs1a depletion
355
impairs hepatic insulin sensitivity in socs1a-deficient zebrafish (5, 36). However, the
356
impaired insulin sensitivity could be achieved by the over-expression of SOCS1 or SOCS3 in
357
the liver transgenic mouse models (40). Mouse SOCS1 deficiency does not prevent
358
HFD-induced insulin resistance in combination with SOCS1/RAG2 deficiency (12). SOCS2
359
deletion in mice can protect against hepatic steatosis but worsens insulin resistance only with
360
HFD feeding (42). Notably, the phenotype observed in socs1a-deficient zebrafish is clearly
361
different from the SOCS1-, SOCS2-, and SOCS3- knockout mouse phenotypes (12, 37, 40,
362
42), with enhanced liver steatosis and impaired insulin sensitivity under normal feeding
363
conditions, suggesting the differences in the regulation of hepatic metabolism between these
364
two vertebrate models.
365
The liver lies at the crossroad of lipid metabolism by actively taking up lipids associated
366
with remnant particles as well as FFA and NEFA originating from adipose tissue lipolysis. As
367
a result of enhanced lipolysis and gluconeogenesis activities in the socs1a-deficient liver,
368
these activities should consume a large amount of oxygen to produce energy (14). It is
369
conceivable that lipid metabolism is related to hypoxia sensing, particularly in aquatic
370
vertebrates. Unlike SOCS1-, SOCS2-, and SOCS3-deficient mouse models, the expression
371
level of hif1al was increased significantly in the liver of the socs1a mutant fish (Fig. 5A).
372
Hif-mediated metabolic approaches appear to be important for the maintenance of local
373
oxygen homeostasis in the liver by limiting either oxygen consumption or be promoting fatty
374
acid oxidation to glycolysis and blocking mitochondrial biogenesis (14). Due to the highly
375
activated GH signaling in socs1a mutant fish, glycolysis activity is inhibited. In fact, the
376
transcriptional expression of several key enzymes involved in glycolysis, such as shxk1 and
377
gk, were decreased in the socs1a mutant liver (Fig. 5A). In addition, up-regulated levels of
378
several key hypoxia-inducible genes, such as mxi-1, higd1a, Angptl4, and hmox1, other than
379
hif1al (Fig. 5A), have been observed in the socs1a-deficient liver (Table 3). These molecules
380
have been reported to be involved in hypoxia-inducible responses, and important regulators
381
for mitochondrial biogenesis (43), physiological stress (3), LPL activity in peripheral tissues
382
(9), and cytoprotection against oxidative stress during liver ischemia (41). Thus, our results
383
above reveal that the depletion of socs1a causes a complex metabolic syndrome, including
384
locally hypoxia stress, progressive hepatic steatosis, and insulin resistance in fish. Based on
385
the observed loss of visceral and subcutaneous fat, hepatic steatosis and hepatic insulin
386
resistance, our socs1a-deficient zebrafish display a phenotype similar to that observed in
387
mammalian models with lipodystrophy (6).
388
Our data provide strong evidence that SOCS1a acts as an essential negative regulator of
389
GH signaling mediated by JAK-STAT5 in zebrafish. However, due to differences in
390
physiological and metabolic features between the terrestrial mammal model and the aquatic
391
vertebrate model, the gigantism phenotype of the SOCS2-null mouse has not been achieved in
392
SOCS1a-deficient fish, which actually progressively develop a complex metabolic syndrome.
393
Although changes in the expression patterns of typical inflammatory cytokines are not
394
observed in the socs1a mutants, several inflammation-responsive genes have been found to be
395
differentially expressed in the livers between wild-type and socs1a-deficient fish (Table 3).
396
Therefore, chronic, low-grade inflammation may not be excluded as a cause of the hepatic
397
steatosis and insulin resistance observed in the mutant fish. Zebrafish SOCS1a has been
398
suggested to function in immune response and GH signaling, which is consistent with our
399
observation that socs1a-deficient fish exhibit characteristics of both GH and hepatic steatosis
400
phenotypes. However, compared with mouse genetic models, socs1a depletion in zebrafish
401
did not recapitulate the phenotypes of SOCS1-, SOCS2-, or SOCS3-deficient mice. In this
402
aspect, our existing data also cannot rule out the compensatory roles of zebrafish SOCS1b in
403
our socs1a-deficient models. More detailed analyses of the unique metabolic features of
404
teleosts from mammal animal models and of the major targeting signaling pathway for each
405
teleost SOCS member will further clarify the metabolism principles applied in fishery and the
406
physiological roles of teleost SOCS members.
407 408
GRANTS
409
This study was supported by funds obtained from the National Basic Research Program of
410
China (973 Program, 2014CB138602) and Hi-Tech Research and Development Program of
411
China (863 Program, 2012AA022402) to ZY, Natural Science Foundation of China (No.
412
31222052) to YHS.
413 414
DISCLOSURES
415
The authors declare no conflict of interest.
416 417
REFERENCES
418
Primary Sources
419
Secondary Sources
420
Uncategorized References
421 422 423 424
1.
Ahmed SF, and Farquharson C. The effect of GH and IGF1 on linear growth and skeletal
development and their modulation by SOCS proteins. J Endocrinol 206: 249-259, 2010. 2.
Amali AA, Rekha RD, Lin CJ, Wang WL, Gong HY, Her GM, and Wu JL. Thioacetamide induced
liver damage in zebrafish embryo as a disease model for steatohepatitis. J Biomed Sci 13: 225-232,
425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468
2006. 3.
Ameri K, Rajah AM, Nguyen V, Sanders TA, Jahangiri A, DeLay M, Donne M, Choi HJ, Tormos KV,
Yeghiazarians Y, Jeffrey SS, Rinaudo PF, Rowitch DH, Aghi M, and Maltepe E. Nuclear Localization of the Mitochondrial Factor HIGD1A during Metabolic Stress. Plos One 8: 2013. 4.
Banks AS, Li J, McKeag L, Hribal ML, Kashiwada M, Accili D, and Rothman PB. Deletion of SOCS7
leads to enhanced insulin action and enlarged islets of Langerhans. J Clin Invest 115: 2462-2471, 2005. 5.
Chen G, Zhang Y, Lu D, Li NQ, and Ross AC. Retinoids synergize with insulin to induce hepatic Gck
expression. Biochem J 419: 645-653, 2009. 6.
Cortes VA, Cautivo KM, Rong SX, Garg A, Horton JD, and Agarwal AK. Leptin ameliorates insulin
resistance and hepatic steatosis in Agpat2(-/-) lipodystrophic mice independent of hepatocyte leptin receptors. J Lipid Res 55: 276-288, 2014. 7.
Davidson MB, Shen DC, Venkatesan N, and Sladen G. In vivo insulin antagonism but evanescent
in vitro tissue effect in rats with growth hormone-secreting tumors. J Endocrinol Invest 10: 569-574, 1987. 8.
De Clercq P, Arijs Y, Van Meir T, Van Stappen G, Sorgeloos P, Dewettinck K, Rey M, Grenier S,
and Febvay G. Nutritional value of brine shrimp cysts as a factitious food for Orius laevigatus (Heteroptera: Anthocoridae). Biocontrol Science and Technology 15: 467-479, 2005. 9.
Dijk W, and Kersten S. Regulation of lipoprotein lipase by AngptI4. Trends Endocrin Met 25:
146-155, 2014. 10. Dominici FP, Argentino DP, Munoz MC, Miquet JG, Sotelo AI, and Turyn D. Influence of the crosstalk between growth hormone and insulin signalling on the modulation of insulin sensitivity. Growth Horm IGF Res 15: 324-336, 2005. 11. Eames SC, Philipson LH, Prince VE, and Kinkel MD. Blood Sugar Measurement in Zebrafish Reveals Dynamics of Glucose Homeostasis. Zebrafish 7: 205-213, 2010. 12. Emanuelli B, Macotela Y, Boucher J, and Ronald Kahn C. SOCS-1 deficiency does not prevent diet-induced insulin resistance. Biochem Biophys Res Commun 377: 447-452, 2008. 13. Figueiredo MA, Mareco EA, Silva MDP, and Marins LF. Muscle-specific growth hormone receptor (GHR) overexpression induces hyperplasia but not hypertrophy in transgenic zebrafish. Transgenic Res 21: 457-469, 2012. 14. Goda N, and Kanai M. Hypoxia-inducible factors and their roles in energy metabolism. Int J Hematol 95: 457-463, 2012. 15. Greenhalgh CJ, Rico-Bautista E, Lorentzon M, Thaus AL, Morgan PO, Willson TA, Zervoudakis P, Metcalf D, Street I, Nicola NA, Nash AD, Fabri LJ, Norstedt G, Ohlsson C, Flores-Morales A, Alexander WS, and Hilton DJ. SOCS2 negatively regulates growth hormone action in vitro and in vivo. J Clin Invest 115: 397-406, 2005. 16. He W, Dai X, Chen X, He J, and Yin Z. Zebrafish pituitary gene expression before and after sexual maturation. J Endocrinol 221: 429-440, 2014. 17. Howard JK, and Flier JS. Attenuation of leptin and insulin signaling by SOCS proteins. Trends Endocrinol Metab 17: 365-371, 2006. 18. Huang P, Xiao A, Zhou M, Zhu Z, Lin S, and Zhang B. Heritable gene targeting in zebrafish using customized TALENs. Nat Biotechnol 29: 699-700, 2011. 19. Jamieson E, Chong MM, Steinberg GR, Jovanovska V, Fam BC, Bullen DV, Chen Y, Kemp BE, Proietto J, Kay TW, and Andrikopoulos S. Socs1 deficiency enhances hepatic insulin signaling. J Biol Chem 280: 31516-31521, 2005.
469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512
20. Jin HJ, Xiang LX, and Shao JZ. Identification and characterization of suppressor of cytokine signaling 1 (SOCS-1) homologues in teleost fish. Immunogenetics 59: 673-686, 2007. 21. Kim YD, Li T, Ahn SW, Kim DK, Lee JM, Hwang SL, Kim YH, Lee CH, Lee IK, Chiang JY, and Choi HS.
Orphan
nuclear
receptor
small
heterodimer
partner
negatively
regulates
growth
hormone-mediated induction of hepatic gluconeogenesis through inhibition of signal transducer and activator of transcription 5 (STAT5) transactivation. J Biol Chem 287: 37098-37108, 2012. 22. Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, and Schilling TF. Stages of embryonic development of the zebrafish. Dev Dyn 203: 253-310, 1995. 23. Kittilson JD, Jones E, and Sheridan MA. ERK, Akt, and STAT5 are Differentially Activated by the Two Growth Hormone Receptor Subtypes of a Teleost Fish (Oncorhynchus Mykiss). Front Endocrinol (Lausanne) 2: 30, 2011. 24. Krebs DL, Uren RT, Metcalf D, Rakar S, Zhang JG, Starr R, De Souza DP, Hanzinikolas K, Eyles J, Connolly LM, Simpson RJ, Nicola NA, Nicholson SE, Baca M, Hilton DJ, and Alexander WS. SOCS-6 binds to insulin receptor substrate 4, and mice lacking the SOCS-6 gene exhibit mild growth retardation. Mol Cell Biol 22: 4567-4578, 2002. 25. Lanning NJ, and Carter-Su C. Recent advances in growth hormone signaling. Rev Endocr Metab Disord 7: 225-235, 2006. 26. Leena S, Shameena B, and Oommen OV. Studies on the effect of growth hormone in vivo and in vitro on lipogenic enzymes and transaminases in a teleost Anabas testudineus (Bloch). Endocr Res 25: 341-355, 1999. 27. Lou Q, He J, Hu L, and Yin Z. Role of lbx2 in the noncanonical Wnt signaling pathway for convergence and extension movements and hypaxial myogenesis in zebrafish. Biochimica et biophysica acta 1823: 1024-1032, 2012. 28. McMenamin SK, Minchin JE, Gordon TN, Rawls JF, and Parichy DM. Dwarfism and increased adiposity in the gh1 mutant zebrafish vizzini. Endocrinology 154: 1476-1487, 2013. 29. Metcalf D, Greenhalgh CJ, Viney E, Willson TA, Starr R, Nicola NA, Hilton DJ, and Alexander WS. Gigantism in mice lacking suppressor of cytokine signalling-2. Nature 405: 1069-1073, 2000. 30. Nie L, Xiong R, Zhang YS, Zhu LY, Shao JZ, and Xiang LX. Conserved inhibitory role of teleost SOCS-1s in IFN signaling pathways. Dev Comp Immunol 43: 23-29, 2014. 31. Norris AW, Chen L, Fisher SJ, Szanto I, Ristow M, Jozsi AC, Hirshman MF, Rosen ED, Goodyear LJ, Gonzalez FJ, Spiegelman BM, and Kahn CR. Muscle-specific PPARγ-deficient mice develop increased adiposity and insulin resistance but respond to thiazolidinediones. The Journal of Clinical Investigation 112: 608-618, 2003. 32. O'Sullivan LA, Liongue C, Lewis RS, Stephenson SEM, and Ward AC. Cytokine receptor signaling through the Jak-Stat-Socs pathway in disease. Mol Immunol 44: 2497-2506, 2007. 33. O'Sullivan LA, Noor SM, Trengove MC, Lewis RS, Liongue C, Sprigg NS, Nicholson SE, and Ward AC. Suppressor of cytokine signaling 1 regulates embryonic myelopoiesis independently of its effects on T cell development. J Immunol 186: 4751-4761, 2011. 34. Opazo JC, Butts GT, Nery MF, Storz JF, and Hoffmann FG. Whole-Genome Duplication and the Functional Diversification of Teleost Fish Hemoglobins. Mol Biol Evol 30: 140-153, 2013. 35. Press M. Growth hormone and metabolism. Diabetes Metab Rev 4: 391-414, 1988. 36. Rui LY. Energy Metabolism in the Liver. Compr Physiol 4: 177-197, 2014. 37. Sachithanandan N, Fam BC, Fynch S, Dzamko N, Watt MJ, Wormald S, Honeyman J, Galic S, Proietto J, Andrikopoulos S, Hevener AL, Kay TW, and Steinberg GR. Liver-specific suppressor of
513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534
cytokine signaling-3 deletion in mice enhances hepatic insulin sensitivity and lipogenesis resulting in fatty liver and obesity. Hepatology 52: 1632-1642, 2010. 38. Starr R, Metcalf D, Elefanty AG, Brysha M, Willson TA, Nicola NA, Hilton DJ, and Alexander WS. Liver degeneration and lymphoid deficiencies in mice lacking suppressor of cytokine signaling-1. Proc Natl Acad Sci U S A 95: 14395-14399, 1998. 39. Studzinski ALM, Almeida DV, Lanes CFC, Figueiredo MD, and Marins LF. SOCS1 and SOCS3 are the main negative modulators of the somatotrophic axis in liver of homozygous GH-transgenic zebrafish (Danio rerio). Gen Comp Endocr 161: 67-72, 2009. 40. Ueki K, Kondo T, Tseng YH, and Kahn CR. Central role of suppressors of cytokine signaling proteins in hepatic steatosis, insulin resistance, and the metabolic syndrome in the mouse. Proc Natl Acad Sci U S A 101: 10422-10427, 2004. 41. Yun N, Cho HI, and Lee SM. Impaired autophagy contributes to hepatocellular damage during ischemia/reperfusion: heme oxygenase-1 as a possible regulator. Free Radic Biol Med 68: 168-177, 2014. 42. Zadjali F, Santana-Farre R, Vesterlund M, Carow B, Mirecki-Garrido M, Hernandez-Hernandez I, Flodstrom-Tullberg M, Parini P, Rottenberg M, Norstedt G, Fernandez-Perez L, and Flores-Morales A. SOCS2 deletion protects against hepatic steatosis but worsens insulin resistance in high-fat-diet-fed mice. FASEB J 26: 3282-3291, 2012. 43. Zhang H, Gao P, Fukuda R, Kumar G, Krishnamachary B, Zeller KI, Dang CV, and Semenza GL. HIF-1 inhibits mitochondrial biogenesis and cellular respiration in VHL-deficient renal cell carcinoma by repression of C-MYC activity. Cancer Cell 11: 407-420, 2007.
535
Figure legends
536
Figure 1. Disruption of zebrafish socs1a gene by TALENs. A: the endogenous socs1a gene is
537
shown (top), and the exons are depicted as boxes. The coding region is shown in black. The
538
targeting arms aimed at the coding region are underlined with a red line and are shown in red
539
font, and the flanking region in the middle is shown in black font. Two independent mutant
540
lines have been obtained. One nucleotide deletion or one nucleotide insertion in the targeting
541
region has been found in mutant line 1 (M1) or line 2 (M2) respectively. B: a representative
542
genotyping for StuI digestion patterning of the PCR products amplified from the genomic
543
DNA samples from M1 socs1a-/-, socs1a+/- and socs1a+/+ individuals. C: a C deletion in the
544
targeting region of the M1 mutant socs1a site (arrow) and a C insertion in the targeting region
545
of the M2 mutant socs1a site (arrow). D: the diagram shows a truncated protein, in which the
546
first 22 amino acids (shown in green) are identical to those of the wild-type SOCS1a protein
547
and which contains 41 (M1) or 96 (M2) miscoding amino acids (shown in red) predicted from
548
the sequences of the mutant forms of the socs1a transcripts in the two independent mutant
549
lines. E: the relative levels of the wild-type and mutant socs1a transcripts from socs1a+/+ and
550
socs1a-/- larvae at 5 dpf were assayed by quantitative RT-PCR.
551 552
Figure 2. Effects of zebrafish socs1a disruption on JAK-STAT5 activation, erythropoiesis,
553
and inflammatory gene expression. A: elevated levels of phosphorylated STAT5.1 in socs1a-/-
554
fish compared with wild-type control fish at 5 dpf stage. This highly elevated level of
555
STAT5.1 phosphorylation could be attenuated with the addition of native socs1a mRNA or
556
AG490, an inhibitor of JAK kinases. B: an upregulated expression level of gata1, an early
557
erythroid marker, was found in socs1a-/- embryos (bottom) compared with wild-type embryos
558
(upper) at 24 hpf. C: elevated levels of Hbbe1 in socs1a-/- fish compared with the wild-type
559
larvae were detected through western blot at 5 dpf. The enhanced erythropoiesis could also be
560
attenuated by the addition of AG490. D: a modest increase in the numbers of erythrocytes was
561
observed in socs1a-/- adults compared with wild-type adults, as determined through the red
562
blood cell counts at 90 dpf. E: no significantly elevated expression levels of the inflammatory
563
genes ifnγ, il1b, nfkb2, nkkbiab, il6 and tnfα were observed in socs1a-/- larvae compared with
564
their control siblings at 5 dpf.
565 566
Figure 3. General phenotypic abnormality observed in socs1a-deficient zebrafish. A: normal
567
morphological features of socs1a+/+ (upper) and socs1a-/- (lower) larvae at the 3-dpf stage. B:
568
the obvious somatic growth retardation of socs1a-/- zebrafish began at the juvenile stage.
569
Asterisk (*) indicated significant difference. C: elevated mortality observed after 40 dpf. D:
570
the socs1a-/- adults (bottom) were significantly smaller in size compared with the wild-type
571
zebrafish adults (upper) raised in the same tank at the 90 dpf stage.
572 573
Figure 4. Hepatic steatosis and abnormal adiposity observed in socs1a-deficient adults. A:
574
representative fluorescent image of adipose tissue from adult zebrafish (90 dpf) stained with
575
Nile red obtained using an excitation wave length of 470 nm. Socs1a+/+ (Upper) and socs1a-/-
576
(lower) zebrafish were set in the same field. A marked decrease in the size of the
577
subcutaneous adipose tissue (white arrows) was observed in the socs1a-/- adults (bottom)
578
compared with the wild-type control fish (upper). B: diagram of the region used for sectioning.
579
C and D: Oil red O staining of adult zebrafish trunk sections. A socs1a+/+ sample is shown in
580
(C), and a socs1a-/- sample is shown in (D). E and F: representative images of Oil Red
581
O-stained sections from adult wild-type liver tissue at low (20×, E) and high (100×, F)
582
magnification. The high-magnification image observed in (F) is boxed in (E). G and H:
583
representative images of Oil Red O-stained sections from adult wild-type liver tissue at low
584
(20×, G) and high (100×, H) magnification. The high-magnification image observed in (H)
585
is shown as in panel (G). A markedly higher number of lipid drops are observed in the liver
586
tissue from socs1a-deficient adults compared with that from the control adults at the 90 dpf
587
stage. All of the assays were performed with at least three pairs of size-matched socs1a-/- and
588
socs1a+/+ adults, and similar results were obtained.
589 590
Figure 5. Real-time RT-PCR confirmation of the representative key signaling transcripts
591
identified from the comparative liver RNA-Seq transcriptomics analyses. RNA samples were
592
extracted from 4 livers of socs1a-/- zebrafish or control siblings at 90 dpf respectively. The
593
real-time RT-PCR measurements were performed with specific primers designed for insulin
594
(ins),
595
glucose-6-phosphatase c family a1 (g6pca.1), hexokinase domain containing 1 (shxk1),
596
glucokinase (gk), O family member of Forkhead transcription factors (foxO1a),
597
apolipoprotein A-IV (apoA4), low density lipoprotein receptor (ldlr), stearoyl-CoA desaturase
598
(scd), and hypoxia induced factor 1-alpha like (hif1αl). The relative transcript levels were
599
determined by real-time RT-PCR using β-actin as the internal standard. The expression levels
600
of some of the genes in the livers were assayed with the samples from both M1 and M2
insulin
receptor
b
(insrb),
phosphoenolpyruvate
carboxykinase
(pck),
601
mutant lines. B and C: the levels of insulin in the liver (B) and plasma (C) from socs1a-/- and
602
wild-type control adults were determined by ELISA. D and E: the levels of glucagon in the
603
liver (D) and plasma (E) samples from socs1a-/- and wild-type control adults were determined
604
by ELISA. F: blood glucose concentration in socs1a-/- and socs1a+/+ zebrafish under different
605
conditions at 90 dpf. The detailed conditions are described in the “Materials and Methods”
606
section. G: decreased levels of phosphorylated AKT in socs1a-/- fish compared with wild-type
607
control fish in hepatic tissue at 90 dpf. The data are expressed as the means ± SE from three
608
separate measurements. The asterisk (*) and double asterisks (**) indicate significant
609
differences at a (P < 0.05) and very significant differences (P< 0.01) with at least a one-fold
610
difference. The primers designed for the real-time PCR analysis are listed in Table 1.
611 612
Figure 6. Improved performance of somatic growth observed in the heterozygous
613
socs1a-deficient zebrafish. A and B: the measurements of the body weight (A) and body
614
length (B) of the socs1a+/- fish compared with their control siblings respectively. The
615
measurements for the socs1a+/- fish were assayed at 120 dpf stage for M1 (n=15 for the
616
control siblings, n=13 for the socs1a+/-) or at 185 dpf stage for M2 (n=9 for the control
617
siblings, n=21 for the socs1a+/-). The socs1a+/- fish and their control siblings from each mutant
618
line were raised in the same tank respectively. The slower overall growth of the individuals of
619
M2 than those of M1 observed might be due to different densities and culture conditions in
620
different aquariums. C: the levels of the phosphorylated-STAT5.1 in the adult hepatic samples
621
of the socs1a+/+, socs1a+/-, and their control sibling fish were quantified with Western Blot
622
analyses. D: Gray values of western blot signal were read by Image J.
623
624
Table 1. Primers used in the experiments product symbol
Gene ID
length
forward
reverse
genotyping
NM_001003467
378
GCCGGTTGGTCTTTCCTGTA
TTAATCCGGATGCTGACGGG
socs1a
NM_001003467
704
AACGACCGATGAGTCTCATC
TAATGCAACAAGTCTCGTCT
g6pca1
NM_001003512.1
104
ATCGCTGCACCTTACGAGAT
ACCCAGTGAAACACGCTCTC
foxO1a
NM_001077257.2
107
GCGGCAAAGAAAAAGCTGGC
TCATTGCTGTGGGAGTTCGG
Apof
XM_003198822.3
132
TTCTAGAAGAGCCAGTGAGTGC
TGAGGGAATGGAGCTTTGCC
ldlra
NM_001030283.1
117
ATACATACGCGTCATCCCCC
GCGCCCCTGATGCTGTAT
scd
NM_198815.2
143
CTCACACTCCTCTGGGCTTTT
AAGGCCATGGAGTTTCCGAT
igfbp1a
NM_173283.3
102
AGCGAGACAGCACCAGATCA
GTCGAATGGCTTTCCGTGC
insulin
NM_131056.1
147
GGTCGTGTCCAGTGTAAGCA
GGAAGGAAACCCAGAAGGGG
insulinR
NM_001123229.1
106
ACTGCTGGTCTTTCGGAGTG
CCTCCGTCCATGACGAACTT
pck
NM_214751.1
106
GGTCAACAACTGGCCCTGTA
CAGCAGTGAGTTTCCTCCGT
shxk1
NM_001115125.1
136
TCAGCTAATCTGGTGGCTGC
GCCTTTTGGGGTACTGTGGA
gk
NM_001045385.2
102
CACCGCTGACCTGCTATGAT
AGTCGGCCACTTCACATACG
hif1al
NM_200405
94
CGTCGAAAAGGCCCAGTTTG
CGAGCTCATAACACCTCCGT
il1b
NM_212844.2
150
TGGACTTCGCAGCACAAAATG
GTTCACTTCACGCTCTTGGATG
nfkb
NM_001001840.3
133
AAACAAGACGCAAGGAGCCC
GCTGAAGGAAACGTCATAGGC
nfkbiab
NM_199629.1
142
TGCACAGGAGCAGTGTAACG
GGTCAGGTGATAAGGCGTGT
il6
NM_001261449.1
98
ATGACGGCATTTGAAGGGGT
CGCGTTAGACATCTTTCCGTG
β-actin
NM_131031.1
99
GCCACCTTAAATGGCCTAGCA
GCCATACAGAGCAGAAGCCA
ifnγ
NM_212864.1
133
TGGGCGATCAAGGAAAACGA
TTGATGCTTTAGCCTGCCGT
tnfa
NM_212859.2
99
ATCATTTTGGCTGTGGGCCT
TGTGAGTCTCAGCACACTTCC
socs1a
625 626
Table 2. Comparison of liver triglyceride content between socs1a+/+ and socs1a-/- fish Mutant lines M1 M1 M1 M1 M2 M2 M2 M2
627 628 629
Genotypes socs1a+/+ socs1a-/socs1a+/+ socs1a-/socs1a+/+ socs1a-/socs1a+/+ socs1a-/-
Genders male male female female male male female female
Triglyceride content in the liver (μmol/g) 17.23±0.63 25.65±0.51 * 18.62±0.70g 24.80±0.69 * 20.32±0.49 39.25±0.53* 17.06±0.61 25.38±0.46*
*represents significant difference between socs1a+/+ and socs1a-/-（P value less than 0.05）.
630 631
Table 3. Some of differential expressed transcripts between the hepatic tissues of socs1a-/fish and their control siblings based on RNA-seq analyses Gene ID
Full name
Symbol
RPKM +/+
*
RPKM of -/-
log2
socs1a *
Ratio**
39.171289
93.342562
1.2527384
socs1a angptl4
of
ENSDARG00000035859
angiopoietin-like 4
ENSDARG00000086281
apolipoprotein A-IV a
APOA4
0.7967262
7.0850112
3.1526143
ENSDARG00000016773
cytokine
socs8
5.4190333
1.6627393
-1.704474
foxo1a
22.812925
53.834928
1.238691
g6pca
12.742255
25.232548
1.1302395
inducible
SH2-containing protein b ENSDARG00000063540 ENSDARG00000031616
forkhead box O1 a glucose-6-phosphatase
a,
catalytic ENSDARG00000068006
glucokinase (hexokinase 4)
gck
53.531774
0.6856726
-6.286732
ENSDARG00000027529
heme
hmox1
32.919674
77.315739
1.2318121
hkdc1
7.6791699
1.5390071
-2.318951
h6pd
65.751368
31.997154
-1.039077
hig1
9.860814
22.024901
1.1593569
hif1al
34.340549
100.65683
1.55146
oxygenase
(decycling) 1a ENSDARG00000038703
hexokinase
domain
containing 1 ENSDARG00000060153
hexose-6-phosphate
ENSDARG00000022303
HIG1
dehydrogenase hypoxia
inducible
domain family, member 1A ENSDARG00000041169
hypoxia-inducible factor 1, alpha subunit, like
ENSDARG00000071524
insulin receptor b
inrb
3.6456078
8.1925728
1.1681572
ENSDARG00000032768
interferon regulatory factor
irf1b
25.657533
12.24516
-1.067171
irf2bp2b
14.869047
7.2601528
-1.03424
irf7
2.6557538
16.457309
2.6315352
irf8
7.6835218
1.7315174
-2.149731
isg15
15.326053
98.491401
2.6840116
il1b
1.3716531
0.1920888
-2.83607
receptor,
il4r
7.8087661
3.8386256
-1.024505
10
receptor,
il10ra
1.8827663
0.7812326
-1.26903
2
receptor,
il2rga
3.163493
0.8667808
-1.867779
21
receptor,
ir21
2.8691533
1.3393382
-1.099105
lipoprotein
ldlr
105.53622
239.91985
1.1848143
1b ENSDARG00000006275
interferon regulatory factor 2 binding protein 2b
ENSDARG00000045661
interferon regulatory factor 7
ENSDARG00000056407
interferon regulatory factor 8
ENSDARG00000086374
interferon stimulated gene, 15 kDa
ENSDARG00000005419
interleukin 1, beta
ENSDARG00000031051
interleukin
4
tandem duplicate 1 ENSDARG00000041180
interleukin alpha
ENSDARG00000068858
interleukin gamma a
ENSDARG00000069961
interleukin
tandem duplicate 1 ENSDARG00000029476
low
density
receptor a ENSDARG00000040884
max
interactor
1,
mxi1
5.9676921
17.655519
1.5648742
pck1
63.775129
205.1901
1.6858953
ins
0.5725708
5.4124101
3.2407452
scd
389.23362
111.45556
-1.804168
tnfsf10
3.8694333
0.655503
-2.561448
tnfb
1.3868937
0.2158035
-2.684067
dimerization protein ENSDARG00000013522
phosphoenolpyruvate carboxykinase 1 (soluble)
ENSDARG00000035350
preproinsulin
ENSDARG00000033662
stearoyl-CoA
desaturase
(delta-9-desaturase) ENSDARG00000057241
tumor
necrosis
(ligand)
factor
superfamily,
member 10 ENSDARG00000013598
tumor necrosis factor b (TNF superfamily, member 2)
632 633 634 635
*RPKM: reads per kilobase of mRNA length per million mapped reads **socs1a mutant/socs1a control ***socs1a mutant/socs1a control
start codon
A
exon2
exon1
WT: atagaagatcgccaagtgagcactgaggcctcctcagacgtcttccaatcccaaagat M1: atagaagatcgccaagtgagcactgagg-ctcctcagacgtcttccaatcccaaagat (Δ1) M2: atagaagatcgccaagtgagcactgaggccctcctcagacgtcttccaatcccaaagat (+1)
B
-/-
+/-
+/-
+/+
+/+
M C
C deletion
M2
M1
D
E
22aa
SOCS1a +/+
2 201aa
SOCS1a -/- M1
63aa
SOCS1a -/- M2
118aa
C insertion
Relative expression of Socs1a
1.5 1 0.5
0 con
Figure 1
M1
M2
B
gata1
24 hpf
A P-STAT5.1
socs1a+/+
STAT5.1 β-actin
Numbers (106) cell counts(million)
D
Hbbe1 β-actin
E 4
Red blood cell counts
3 2 1 0
socs1a+/+ socs1a-/-
Figure 2
Expression levels (folds)
C
socs1a-/1.80 1.60 1.40 1.20 1.00 0.80 0.60 0.40 0.20 0.00
socs1a+/+ socs1a-/-
IFNγ
il1b
nfkb2 nfkbiab
IL6
TNFα
A
B *
0.5
socs1a+/+
socs1a-/-
50μm
C
100%
Body weight (gram)
3 dpf
0.45 0.35
*
0.3 0.25 0.2 0.15 0.1
socs1a+/+
0.05
socs1a-/-
80%
0
D
60%
20
40
60
90
dpf 120 socs1a+/+
40% socs1a+/+
20%
socs1a-/-
socs1a-/-
0% 0
20
Figure 3
40
50
60
70
80
90 100 110 120
dpf
5mm
90 dpf
Survival rate (%)
*
0.4
Socs1a+/+
A 60mm
Socs1a+/+
Socs1a-/-
C
D
Subcutaneous
Socs1a-/-
B 5mm
F
25um
3um
socs1a+/+
E
H
25um
3um
socs1a-/-
G
Figure 4
5mm
A Relative expression levels˄folds˅
18 16 14
**
12
socs1a+/+
**
**
10
M1
**
*
8
M2
**
6 4
*
*
*
2
*
** **
**
** **
0
Hepatic insulin concentration (mIU/g )
0.0400
*
Plasma Glucagon concentration(pg/g)
C
0.0300 0.0200 0.0100 0.0000 socs1a+/+
E
G6pca.1
F
3.500 3.000 2.500 2.000 socs1a+/+ socs1a-/-
GK
foxO1a
*
6.000 5.500 5.000 4.500
apoA4
LDLR
D
10
G
socs1a+/+ socs1a-/-
P-AKT
8 6
AKT
4 2
Β-actin
0
chow diet
fasted
glucose treated
Hif1al
3.00 2.00 1.00 0.00 socs1a+/+ socs1a-/-
socs1a+/+ socs1a-/-
12
SCD
4.00
4.000
socs1a-/-
4.000
Figure 5
shxk1
Hepatic glucagon concentration (pg/g)
pck
blood glucose˄mmol/L˅
B
insrb
Plasma insulin Concentration (mIu/mg)
ins
body weight(g)
A
*
0.5
socs1a+/+
0.4
socs1a+/-
0.3
*
2.9
B
*
socs1a+/+
2.8 body length(cm)
0.6
0.2
socs1a+/-
2.7
*
2.6 2.5 2.4
0.1
2.3
0 M1
M1
M2
M2
P-STAT5.1 STAT5.1 β-actin
Relative gray value
D
C
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 socs1a+/+ socs1a+/- socs1a-/-
Figure 6
