Diabetes Page 2 of 46
1 Crucial Role of the SH2B1 PH Domain for the Control of
2 Energy Balance
3
4 Anabel Floresa, Lawrence S. Argetsingerb+, Lukas K. J. Stadlerc+, Alvaro E. Malagab, Paul B.
5 Vanderb, Lauren C. DeSantisb, Ray M. Joea,b, Joel M. Clineb, Julia M. Keoghc, Elana Henningc,
6 Ines Barrosod, Edson Mendes de Oliveirac, Gowri Chandrashekarb, Erik S. Clutterb, Yixin Hub,
7 Jeanne Stuckeyf, I. Sadaf Farooqic, Martin G. Myers Jr. a,b,e, Christin Carter-Sua,b,e, g *
8 aCell and Molecular Biology Graduate Program, University of Michigan, Ann Arbor, MI 48109, USA
9 bDepartment of Molecular and Integrative Physiology, University of Michigan, Ann Arbor, MI 48109, USA
10 cUniversity of Cambridge Metabolic Research Laboratories and NIHR Cambridge Biomedical Research Centre,
11 Wellcome Trust-MRC Institute of Metabolic Science, Addenbrooke's Hospital, Cambridge, UK
12 dMRC Epidemiology Unit, Wellcome Trust-MRC Institute of Metabolic Science, Addenbrooke's Hospital,
13 Cambridge, UK
14 eDepartment of Internal Medicine, University of Michigan, Ann Arbor, MI 48109, USA
15 fLife Sciences Institute and Departments of Biological Chemistry and Biophysics, University of Michigan, Ann Arbor,
16 MI 48109, USA
17 +Authors contributed equally to this work
18 gLead contact
19 *Correspondence: [email protected] 20
21 Running title: Role of SH2B1 PH Domain in Energy Balance
22 23 24 25 26 27 28
1
Diabetes Publish Ahead of Print, published online August 22, 2019 Page 3 of 46 Diabetes
29
30 Abstract
31 Disruption of the adaptor protein SH2B1 is associated with severe obesity, insulin resistance and
32 neurobehavioral abnormalities in mice and humans. Here we identify 15 SH2B1 mutations in
33 severely obese children. Four obesity-associated human SH2B1 mutations lie in the Pleckstrin
34 Homology (PH) domain, suggesting that the PH domain is essential for SH2B1’s function. We
35 generated a mouse model of a human variant in this domain (P322S). P322S/P332S mice exhibited
36 substantial prenatal lethality. Examination of the P322S/+ metabolic phenotype revealed late-
37 onset glucose intolerance. To circumvent P322S/P322S lethality, mice containing a 2-amino acid
38 deletion within the SH2B1 PH domain (ΔP317, R318; ΔPR) were studied. Mice homozygous for
39 ΔPR were born at the expected Mendelian ratio and exhibited obesity plus insulin resistance and
40 glucose intolerance beyond that attributable to their increased adiposity. These studies demonstrate
41 that the PH domain plays a crucial role in SH2B1 control of energy balance and glucose
42 homeostasis.
43
44 Introduction
45 Hyperphagia, severe obesity, insulin resistance and neurobehavioral abnormalities have
46 been reported in individuals with rare coding variants in the SH2B1 gene (1,2). Consistently, mice
47 null for Sh2b1 exhibit obesity, impaired glucose homeostasis, and often, aggressive behavior (3-
48 5). Transgenic expression of the isoform of SH2B1 (SH2B1) in the brain largely corrects the
49 obesity and glucose intolerance of otherwise Sh2b1-null mice (6), suggesting the importance of
50 brain SH2B1 for the control of energy balance and glucose homeostasis.
51 At the cellular level, SH2B1 is an intracellular adaptor protein that is recruited to
2 Diabetes Page 4 of 46
52 phosphorylated tyrosine residues on specific membrane receptor tyrosine kinases (e.g. receptors
53 for brain-derived neurotrophic factor (BDNF), nerve growth factor (NGF), insulin) and cytokine
54 receptor/JAK complexes (e.g. leptin receptor/JAK2) and enhances the function of these receptors
55 (7-13). The exact mechanism(s) by which it does so is unclear, although a variety of mechanisms
56 have been proposed. These include enhanced dimerization causing increased activation of the
57 kinase itself (14), stabilization of the active state of the kinase (15), decreased dephosphorylation
58 or increased complex formation of insulin receptor substrate (IRS) proteins bound to receptors or
59 receptor/JAK2 (16,17), regulation of the actin cytoskeleton (18), and activation of specific
60 pathways, including ERKs, Akt, and/or phospholipase C (19,10). Some of these receptors,
61 including the leptin, BDNF, and insulin receptors, play important roles in the central control of
62 energy expenditure and/or glucose homeostasis (20). SH2B1 has been shown to enhance BDNF-
63 and NGF-stimulated neurite outgrowth in PC12 cells (21), (13).
64 The four isoforms of SH2B1 ( ), which differ only in their COOH-termini, share
65 631 NH2-terminal amino acids. These amino acids possess a dimerization domain, a pleckstrin
66 homology (PH) domain, a src-homology 2 (SH2) domain, and nuclear localization and export
67 sequences (NLS and NES, respectively) (22-24) (Fig. 1A). The SH2 domain enables SH2B1
68 recruitment to specific phosphorylated tyrosine residues in activated tyrosine kinases (25). The
69 NLS and NES are essential for SH2B1 to shuttle between the nucleus, the cytosol, and the plasma
70 membrane (22,23) . The NLS combined with the dimerization domain enables SH2B1 to associate
71 with the plasma membrane (26). However, the function and importance of the SH2B1 PH domain
72 remains largely unknown. Four human obesity-associated mutations lie in the SH2B1 PH domain
73 (Fig. 1A), suggesting the importance of the PH domain in SH2B1 function. The PH domains of
74 some proteins bind inositol phospholipids to mediate membrane localization (27,28). However,
3 Page 5 of 46 Diabetes
75 90-95% of all human PH domains do not bind strongly to phosphoinositides and presumably
76 mediate other functions (29). Indeed, the PH domain of SH2B1 neither localizes to the plasma
77 membrane nor is required to localize SH2B1 to the plasma membrane (23,30). Here, we tested
78 the importance of the PH domain of SH2B1 in vivo by generating and studying mice containing
79 human obesity-associated (P322S) or engineered (in-frame deletion of P317 and R318 (ΔPR))
80 mutations in the SH2B1 PH domain. Our results demonstrate that the SH2B1 PH domain plays
81 multiple crucial roles in vivo, including for the control of energy balance and glucose homeostasis
82 and in in vitro studies, changes the subcellular distribution of SH2B1 and enhances NGF-
83 stimulated neurite outgrowth in PC12 cells.
84
85 Research Design and Methods
86 Human studies
87 The Genetics of Obesity Study (GOOS) is a cohort of over 7,000 individuals with severe obesity
88 with age of onset less than 10 years (31,32). Severe obesity is defined as a body mass index (kg/m2)
89 standard deviation score greater than 3 (United Kingdom reference population). Whole exome
90 sequencing and targeted resequencing were performed as in (33). All variants were confirmed by
91 Sanger sequencing (1). HOMA-IR was calculated using the formula HOMA-IR Score = (Fasting
92 insulin) x (Fasting glucose) / 405, which estimates steady state -cell function and insulin
93 sensitivity (34,35). All human studies were approved by the Cambridge Local Research Ethics
94 Committee. Each subject (or parent for those under 16 years) provided written informed consent;
95 minors provided oral consent.
96 Animal care
4 Diabetes Page 6 of 46
97 Animal procedures were approved by the University of Michigan Committee on the Use and Care
98 of Animals in accordance with AAALAC and NIH guidelines. Mice were bred at the University
99 of Michigan and housed in ventilated cages at 23°C on a 12-hour light (6:00- 18:00)/ 12-hour dark
100 cycle with ad libitum access to food and tap water except as noted. Mice were fed standard chow
101 (20% protein, 9% fat, PicoLab Mouse Diet 20 5058, # 0007689) or in Figs. 2H-M and
102 Supplementary Fig. 2F-K), HFD (20% protein, 20% carbohydrate, 60% fat, Research Diets,
103 D12492).
104 Mouse models, genotyping and gene expression
105 CRISPR/Cas9 genome editing was used to insert the P322S mutation into mice. The reverse
106 complement of the genomic Sh2b1 sequence in C57BL/6J mice (accession # NC_000073, GRC
107 m38) was used to design the reagents for CRISPR. The guides were designed using the website
108 described in (36). The mutations in the donor are summarized in Fig. 1A (details in Supplementary
109 Data). After testing, each guide/donor combination was injected into C57BL/6J oocytes by the
110 University of Michigan Transgenic Animal Model Core. P322S and PR founders were
111 backcrossed against C57BL/6J mice. The mice were genotyped as described (37) using primers
112 listed in Supplementary Table S1. The P322S and PR PCR products were digested with Xbal or
113 purified and sequenced. SH2B1 KO mice were obtained from Dr. Liangyou Rui (University of
114 Michigan) and genotyped according to (3). C57BL/6J mice used to invigorate our C57BL/6J
115 colony came from Jackson Laboratories.
116 Relative levels of SH2B1 gene expression were determined using RT-PCR (Supplementary Data).
117 Mouse body weight and food intake
118 Mice were individually housed and body weight and food consumption were assessed weekly.
119 Mouse glucose tolerance tests, insulin tolerance tests, and hormone levels
5 Page 7 of 46 Diabetes
120 Mice were fasted 9:00-13:00 for glucose tolerance tests (GTT) or 8:00-14:00 for insulin tolerance
121 tests (ITT). Glucose or human insulin was injected intraperitoneally and blood was collected from
122 the tail vein. Blood glucose levels were assessed using a Bayer contour glucometer. For plasma
123 insulin levels, mice were fasted 8:00-14:00 and tail blood assayed using an UltraSensitive Mouse
124 Insulin ELISA kit (Crystal Chem, Inc, # 90080). Tail blood (9:00-10:00) (Fig. 2F, 4G) or trunk
125 blood after sacrifice (10:00-13:00) (Fig. 6B) from fed mice was tested for leptin using a Mouse
126 Leptin ELISA kit (Crystal Chem, Inc, # 90030).
127 Mouse body composition, metabolic assessment, and tissue collection
128 Body composition was measured at room temperature in the morning or evening (Fig. 6A) using
129 a Minispec LF90 II Bruker Optics NMR analyzer (University of Michigan Animal Phenotyping
130 Core). To assess metabolic state, mice were single-housed for three days and then tested for 72
131 hours using a Comprehensive Lab Animal Monitoring System (CLAMS) (Columbus Instruments).
132 O2 consumption (VO2), CO2 production (VCO2), X-activity, and Z-activity were collected in 20-
133 minute bins. The final 24 hours of recordings are presented. Mice were sacrificed (10:00-13:00)
134 using decapitation under isoflurane. Trunk blood was collected and the serum stored at -80C.
135 Tissues were collected, weighed, cryopreserved in liquid nitrogen and stored at -80C.
136 Immunoblotting
137 Frozen tissues were lysed in L-RIPA lysis buffer. Equal amounts of proteins were immunoblotted
138 with antibody to SH2B1 (SH2B1) (Santa Cruz Biotechnology, sc-136065, RRID:AB_2301871)
139 (1:1000 dilution) or -tubulin (Santa Cruz Biotechnology, sc-55529, RRID:AB_2210962) (1:1000
140 dilution) as described in (19). For immunoprecipitations, tissue lysates containing equal amounts
141 of protein were incubated with SH2B1 (1:100) and immunoprecipitated and immunoblotted as
6 Diabetes Page 8 of 46
142 in (19). PC12 cells (ATCC) were cultured and treated as in (19). Briefly, they were grown in
143 PC12 medium A (RPMI 1640, 5% FBS, 10% HS) in 10-cm dishes coated with rat tail type I
144 collagen (Corning, #354236). Cells were transfected and 24 hours later, incubated overnight in
145 deprivation medium (RPMI 1640, 2% HS and 1% FBS) before being lysed and immunoblotted
146 with SH2B1.
147 Live cell imaging
148 The indicated construct was transiently transfected into 293T cells or PC12 cells. Cells were
149 treated and live cell images captured by confocal microscopy using Olympus FV 500 laser
150 scanning microscope and Fluoview version 5.0 software as in (19).
151 Neurite outgrowth
152 For Fig. 3D, PC12 cells were plated in 6-well collagen-coated dishes, transiently transfected as
153 indicated for 24 hours and incubated overnight in deprivation medium. Cells were treated and
154 neurite outgrowth determined as in (19). For Fig. 1B, PC12 cells were treated as in (19) with
155 modifications described in Supplementary Data.
156 Structural modeling and ClustalW analysis
157 A structural model for human SH2B1 was created by overlaying the PH domain sequence of
158 human SH2B1 onto the mouse NMR structure of APS (PDB ID 1v5m) using the PyMOL
159 Molecular Graphics System, version 2.2.3. ClustalW alignments were performed using
160 LaserGene, version 14.0.0 (DNASTAR, Madison, WI). Functional homology was defined as
161 residues that match the consensus within 1 distance unit using the PAM250 mutation probability
162 matrix.
163 Statistics
7 Page 9 of 46 Diabetes
164 All non-human analyses were carried out using GraphPad Prism software. Body weight, GTTs,
165 and ITTs were analyzed by 2-way ANOVA followed by Fisher’s Least Significant Difference
166 post-test. Food intake was analyzed by linear regression. Significance of the deviation of birthrate
167 from the expected Mendelian ratio was assessed using Chi-Square Test. For other physiological
168 parameters, experimental animals were compared with their WT littermates by two-tailed
169 Student’s t test. Neurite outgrowth was analyzed by a two-tailed Student’s t test. For all
170 comparisons, *P <0.05 was considered significant.
171 Data and Resource Availability
172 Any raw datasets generated during the current study are available from the corresponding author
173 on reasonable request, with all reagent and analytical details are included in the published article
174 (and its online supplementary files). The mouse models generated and analyzed during the current
175 study are available from the corresponding author upon reasonable request.
176
177 Results
178 Identification and characterization of 15 rare human variants in SH2B1
179 Using exome sequencing, targeted resequencing and Sanger sequencing of 3000
180 individuals exhibiting severe obesity before the age of 10 (33), we identified 15 rare variants in
181 SH2B1 in 16 unrelated individuals (Table 1, Fig. 1A). Eleven variants are newly identified while
182 four of these (R227C, R270W, E299G, V209I) have been previously reported in other obese
183 individuals but not well characterized (4,38,39). Fourteen of these variants are in the first 631
184 amino acids shared by all isoforms of SH2B1. The mean (+/- SD) Body Mass Index Standard
185 Deviation Score of variant carriers was 4.0 +/- 0.6. The 15th variant causes the G638R mutation
186 in the C-terminal tail unique to the isoform of SH2B1. A number of the SH2B1 variant carriers
8 Diabetes Page 10 of 46
187 had HOMA-IR (Homeostatic Model Assessment of Insulin Resistance (34,35)) scores of more
188 than 1.9, indicating insulin resistance and increased risk of type 2 diabetes (40). Some of the
189 HOMA-IR scores, including those for people carrying three mutations in or near the PH domain
190 (G238C, R270Q, and M388V), were particularly high. A spectrum of neurobehavioural
191 abnormalities, including learning difficulties, dyspraxia, hyperactivity/inattention,
192 aggression/emotional lability, anxiety, and autistic traits, were detected in all of the individuals
193 for whom behavioral information was available (Table 1). In the neurite outgrowth assay, 7 of
194 these 15 rare variants impaired the ability of SH2B1 to stimulate NGF-induced neurite
195 outgrowth (Fig. 1B), suggesting that many of the mutations negatively impact on the neuronal
196 function of SH2B1. Because the variants are found in multiple domains in SH2B1 and
197 throughout the SH2B1 sequence, it is not surprising that individuals with different SH2B1
198 variants have different phenotypes. These newly characterized mutations add support to SH2B1
199 being an important regulator of human body weight, insulin sensitivity and behavior.
200 Interestingly, four of the human obesity-associated SH2B1 mutations lie in the PH domain of
201 SH2B1, suggesting the importance of the PH domain in the ability of SH2B1 to regulate energy
202 balance, glucose metabolism and behavior.
203 Developmental lethality in mice homozygous for the SH2B1 P322S mutation
204 To gain insight into the role of the PH domain of SH2B1 in energy balance and glucose
205 metabolism, we studied the effect of the P322S human obesity-associated SH2B1 PH domain
206 mutation in mice. We chose the P322S mutation because of its strong association with obesity in
207 the proband family (1); the conservation of P322 across mammals and with the SH2B1 family
208 member SH2B2/APS; its predicted disruptive effect on PH domain function by Provean
209 (RRID:SCR_002182) and PolyPhen (RRID:SCR_013189) analysis; and P322S-dependent
9 Page 11 of 46 Diabetes
210 deficiencies in SH2B1 function observed in cultured cells (1). We used CRISPR/Cas9-based
211 genome editing to introduce P322S into Sh2b1 in C57BL/6J mice (Fig. 1C). DNA sequencing
212 confirmed germline transmission of the P322S edit (Supplementary Fig. S1A). The P322S
213 mutation does not affect the mRNA levels for any of the Sh2b1 isoforms in the examined tissues
214 (brain, liver and heart) (Supplementary Figs. S1B, S1C). Neither does the P322S mutation alter
215 SH2B1 protein levels or isoform selection in brain tissue (Fig. 1D, Supplementary Fig. S1D).
216 However, we found that homozygous (P322S/P322S) mice are born at much less than the expected
217 Mendelian ratio (Fig. 1E), suggesting that the P322S mutation disrupts SH2B1 PH domain
218 function in a manner that interferes with embryo implantation and/or development. Consistent with
219 this, preliminary data from timed pregnancies reveal that at day E17, the homozygous embryos
220 (P322S/P322S) are also present at less than the expected Mendelian ratio.
221 Mice heterozygous for P322S exhibit altered glucose tolerance, but not energy balance
222 In addition to the difficulty of producing sufficient P322S/P322S mice for study, the high
223 rate of embryonic lethality in P322S/P322S mice suggested that the surviving P322S/P322S mice
224 might have underlying poor health, which could interfere with the analysis of their metabolic
225 phenotype. For these reasons, and because human obesity is linked with heterozygosity for P322S
226 (1), we studied energy balance and glucose homeostasis in heterozygous (P322S/+) male (Fig. 2)
227 and female (Supplementary Fig. S2) mice. We found no difference in food intake, body weight or
228 adiposity between wild-type (WT) and P322S/+ mice fed standard chow (9% fat) or a high-fat diet
229 (HFD) (60% fat). However, in contrast to their WT littermates, 28-week-old HFD-fed P322S/+
230 male and female mice display glucose intolerance in an intraperitoneal glucose tolerance test
231 (GTT). Neither insulin concentrations nor the response to an insulin tolerance test (ITT) are altered
232 in the P322S/+ animals compared to littermate controls, however. These findings suggest that the
10 Diabetes Page 12 of 46
233 PH domain of SH2B1 is important for SH2B1 function, including for the control of glucose
234 homeostasis, but that the resultant metabolic phenotype is less penetrant in the heterozygous state
235 in mice than it is in humans. We thus sought to study mice homozygous for mutations in the
236 SH2B1 PH domain
237 Deletion of P317 and R318 in the PH domain alters the subcellular localization of SH2B1
238 Because of the early lethality of P322S/P322S mice, we examined the function of another
239 SH2B1 mutation containing a two-amino acid deletion (ΔP317,R318) (ΔPR) within the PH
240 domain of SH2B1 (Fig.1C, Supplementary Fig. S1E). This mutation arose as a separate line during
241 the generation of the P322S mice.
242 When transiently expressed as Green Fluorescent Protein (GFP) fusion proteins in PC12
243 cells, SH2B1 and SH2B1 ΔPR demonstrate similar expression levels (Fig. 3A), suggesting that
244 PR does not destabilize the protein. However, while GFP-SH2B1 localizes primarily to the
245 plasma membrane and cytoplasm in 293T and PC12 cells (as previously shown (22,26)), SH2B1
246 PR localizes primarily to the nucleus (Fig. 3B, C). The nuclear localization of SH2B1 PR
247 suggests that the PR mutation alters SH2B1 nuclear cycling to favor retention in the nucleus. We
248 predicted that this altered localization would alter the cellular function of SH2B1 PR. Indeed,
249 SH2B1-dependent NGF-stimulated neurite outgrowth in PC12 cells is decreased in cells
250 expressing SH2B1 PR (Fig. 3D). Thus, disruption of the PH domain by the PR mutation alters
251 the subcellular distribution of SH2B1 and impairs the ability of SH2B1 to enhance neurotrophic
252 factor-induced neurite outgrowth.
253 Obesity, hyperphagia and disrupted glucose homeostasis in mice homozygous for the SH2B1
254 PR mutation
11 Page 13 of 46 Diabetes
255 We examined the phenotype of the mice containing the PR mutation, with the hope that
256 this mutation might produce a less dramatic reproductive phenotype than that observed with P322S
257 in the homozygous state, allowing us to examine the effects of the PR mutation on energy balance
258 and glucose homeostasis in homozygous mice. As with the P322S mutation, PR does not affect
259 the mRNA levels for any of the Sh2b1 isoforms in the tissues tested (brain or heart) (Fig. 4A). At
260 the protein level, the PR mutation did not alter the relative levels of the different isoforms in
261 brain tissue although levels of SH2B1 protein were somewhat reduced (Fig. 4B). Importantly, in
262 contrast to P322S/P322S mice, PR/PR homozygous mice are born and survive at the expected
263 Mendelian frequency (Supplementary Fig. S1F), permitting us to examine the effect of this SH2B1
264 PH domain mutation in the homozygous state.
265 PR/PR mice fed standard chow exhibit significantly increased body weight compared
266 to their WT littermates (Fig. 4C, D). By 20 weeks of age (Fig. 4C), PR/PR male mice are 15
267 grams (>40%) heavier than their WT littermates, while female PR/PR mice are approximately
268 9 grams (~35%) heavier than their littermate controls. It should be noted that we do not think that
269 the reduced levels of SH2B1 protein in the PR/PR mice can account for the increased obesity
270 detected in PR/PR mice because heterozygote SH2B1-/+ mice are not obese (5). Body length
271 was not significantly different in preliminary studies (Supplementary Fig. S1G-H). Overall
272 adiposity (Fig. 4E, F), as well as circulating leptin concentrations (Fig. 4G), are increased in
273 PR/PR homozygotes, but not lean body mass (Fig. 4H). The heterozygous (PR/+) male and
274 female mice show no significant increase in adiposity (Fig. 4F). However, PR/+ males have a
275 slight increase in circulating leptin levels (Fig. 4G), suggesting that in males, even a single copy
276 of the PR mutation may be sufficient to produce a minor effect on energy balance.
12 Diabetes Page 14 of 46
277 Increased food intake (assessed at 18-20 weeks) is observed in PR/PR male and female
278 mice compared to their WT and PR/+ littermates (Fig. 5A), while oxygen consumption (at 11-
279 12 weeks of age) (Fig. 5B), respiratory exchange ratio (RER) (data not shown), and locomotor
280 activity (data not shown) are not altered. Based on these data and the previous finding that SH2B1
281 KO mice are obese primarily as a consequence of increased food intake (5), we think it most likely
282 that the PR mutation causes obesity in mice primarily as a consequence of increasing food intake,
283 rather than decreasing energy expenditure.
284 Glucose tolerance and insulin resistance of PR/PR mice 285 286 We initially examined parameters of glycemic control in PR mice at 18-19 weeks of age.
287 In homozygous PR/PR mice, hyperglycemia at baseline was evident (Fig. 5C), as well as
288 impaired glucose tolerance (male and female mice) and insulin resistance (male mice) in
289 intraperitoneal GTT and ITT, respectively (Fig. 5D, E). Male heterozygous PR/+ mice (like
290 P322S/+ mice) also display impaired glucose tolerance (Fig. 5D), although other parameters of
291 glucose homeostasis are not different from littermate controls.
292 Because the disruption of glucose homeostasis in the aged PR/PR mice presumably
293 results (at least in part) from their increased adiposity, we sought to examine glucose homeostasis
294 in young pre-obese mice to define any adiposity-independent effects of SH2B1 PR on glucose
295 homeostasis. We examined the adiposity of younger PR/PR mice to determine an age at which
296 we might examine glucose homeostasis without it being confounded by increased adiposity. At
297 11-12 weeks of age, adiposity is already increased in male PR/PR mice, but not detectably
298 increased in female PR/PR mice (Fig. 6A). By 7 weeks, leptin levels are increased in male, but
299 not female, PR/PR mice (Fig. 6B). We thus examined glucose homeostasis in PR/PR mice
300 at 8 weeks of age, revealing hyperinsulinemia and glucose intolerance (with unchanged insulin
13 Page 15 of 46 Diabetes
301 tolerance) in both male and female PR/PR mice (Fig. 6C-F). The hyperinsulinemia and glucose
302 intolerance in the presence of unchanged leptin and adiposity in young pre-obese females suggest
303 that the PR mutation interferes with glucose homeostasis independently of adiposity. Taken
304 together, our results suggest that in obese PR mice, the PR mutation likely interferes with
305 glucose homeostasis both independently of adiposity and secondary to the effects of adiposity on
306 energy balance.
307
308 Discussion
309 The identification of four human obesity-associated SH2B1 variants in the PH domain of
310 SH2B1, the fact that the three individuals tested had HOMA-IR scores suggesting severe insulin
311 resistance and risk of type 2 diabetes, and the fact that for the three individuals whose behavior
312 was documented, all displayed behavioral abnormalities ((1,2) and this paper), highlights the
313 importance of the PH domain in SH2B1 function. The lethality of the human obesity-associated
314 P322S mutation in the PH domain of SH2B1 in homozygous P322S/P322S mice demonstrates the
315 importance of this mutation for SH2B1 function in vivo. Similarly, the obesity and diabetes
316 observed in PR/PR mice highlight the importance of the SH2B1 PH domain for SH2B1-
317 mediated metabolic control. The adiposity-independent glucose intolerance of young PR/PR
318 female mice prior to the onset of obesity as well as of P322S/+ and PR/+ mice also reveals the
319 importance of SH2B1 and its PH domain for the control of glucose homeostasis independent of
320 body weight, as previously suggested from the phenotype of humans bearing mutations in SH2B1
321 (1,2).
322 Based on the increased food intake of PR/PR mice and the findings in the SH2B1 KO
323 mice (5), we think it likely that the increased body weight of the PR/PR mice is due to impaired
14 Diabetes Page 16 of 46
324 function of SH2B1 in the hypothalamus. However, SH2B1 is also expressed in the periphery.
325 The increased insulin concentrations with glucose intolerance in young, non-obese female
326 PR/PR mice suggest alterations in tissues that control glucose uptake. However, the presence
327 of glucose intolerance despite the increased insulin levels is consistent with the islets of PR/PR
328 mice having an impaired ability to fully compensate for those alterations (41,42).
329 While humans heterozygous for P322S exhibit severe obesity, P322S/+ mice display mild
330 glucose intolerance only in aged, HFD-fed animals. Because the region surrounding P322 is
331 conserved between mice and humans (Fig. 7A), it is unlikely that the more modest phenotype of
332 P322S/+ mice compared to humans reflects species differences that result in structural changes in
333 the SH2B1 PH domain, per se, but rather that PH domain binding partners may have different
334 tolerances for P322S in mice and humans, and/or that human physiology adapts more poorly to
335 the resultant alterations in SH2B1 function. Consistent with the importance of the PH domain for
336 the function of SH2B family members, at least 9 point mutations (E208Q/E, A215V, G220V/R,
337 A223V, G229S, D234N, F287S) have been identified in the PH domain of the SH2B1 orthologue
338 SH2B3/Lnk in patients with myeloproliferative neoplasms (43-48) (Fig. 7A).
339 To gain insight into how the SH2B1 P322S mutation or deletion of residues P317 and R318
340 in SH2B1 might regulate the function of the PH domain in SH2B1, we performed Clustal analysis
341 of SH2B family members and analyzed a model of SH2B1 based on the NMR structure of the PH
342 domain of the SH2B1 family member APS (49). Clustal analysis of the PH domains of the SH2B
343 family members reveals that the PH domains are highly conserved (Fig. 7A). In the model,
344 residues P317, R318 and P322S in SH2B1 are on an exterior surface of the PH domain (Fig. 7B
345 and Video1). This surface is presumably a binding interface that interacts with either another
346 region in SH2B1 or another protein. Another human obesity-associated mutation in SH2B1
15 Page 17 of 46 Diabetes
347 (E299G), as well as five of the human myeloproliferative neoplasm-associated mutations in
348 SH2B3/Lnk (G220V/R, A223V, G229S, and D234N) are in close proximity to P317, R318 and
349 P322 in SH2B1. In addition, eight of the human mutations (E208Q, A215V, G220V/R, G229S,
350 D234N in SH2B3/Lnk and E299G, P322S in SH2B1) as well as P317 and R318 in SH2B1 are on
351 or in close proximity to this putative protein binding interface (Fig. 7 and Video1).
352 The number of human patient mutations in SH2B1 and SH2B3 in this region of the PH
353 domain suggests that small structural changes in this region due to mutation or other modification
354 have the potential to produce substantial functional consequences. Because the residues
355 corresponding to P317 and R318 in SH2B1 are on the surface of the PH domain and do not
356 substantially change the direction of the loop, the P317, R318 deletion in SH2B1 would shorten
357 the loop but not severely damage the overall structure. However, the deletion would be expected
358 to diminish stabilization of the turn provided by the predicted pi-pi stacking between residues P317
359 and F309 in SH2B1. In addition, the deletion would be expected to alter the shape and electrostatics
360 of the interface surface in the region of P317 and R318 in SH2B1.
361 Because SH2B1 from humans and mice share 95% sequence identity, with only one
362 conservative difference (S325T) near P322 (Fig. 7A), the structures in mouse and human are
363 expected to be nearly identical. Therefore, the more modest phenotypes of P322S/+ mice compared
364 to humans are likely due to the affinity of the PH domain binding partners. The mouse binding
365 partners may be able to accommodate the P322S mutation better than humans, and/or human
366 physiology adapts more poorly to the resultant alterations in SH2B1 function. Given the different
367 phenotypes produced by the P322S and ΔPR mutations in mice, we postulate that the two
368 mutations alter the structure of the SH2B1 PH domain in different ways to produce distinct changes
369 in cell physiology. That relatively small changes in the PH domain, predicted to have only minor
16 Diabetes Page 18 of 46
370 effects on PH domain structure, cause a rather profound effect on SH2B1 localization at the
371 cellular level and energy balance and glucose homeostasis at the whole animal level provides some
372 of the first real evidence of the importance of the PH domain in SH2B1 function. While SH2B1
373 has been shown to cycle through the nucleus, it is generally found at the plasma membrane and in
374 the cytoplasm (22,23,26). The accumulation of SH2B1 PR in the nucleus indicates that the
375 PR deletion greatly alters the ratio between nuclear import and nuclear export of SH2B1.
376 Consistently, the PR mutation as well as many of the other human obesity-associated SH2B1
377 variants impair the ability of SH2B1 to promote neurotrophic factor-induced neurite outgrowth
378 of PC12 cells. Because neurite outgrowth in PC12 cells shares many properties with the formation
379 of axons and/or dendrites (50), and because the SH2B1 KO have impaired leptin signaling, it will
380 be important in the future to examine the impact of SH2B1 PH domain changes on the structure
381 of neurons that control energy balance.
382
383 Acknowledgements
384 This work was supported by NIH grants R01-DK54222 and R01-DK107730 (to C. Carter-Su) and
385 predoctoral fellowships to A. Flores from the University of Michigan (Rackham Merit
386 Fellowship), Systems and Integrative Biology Training Program (NIH T32-GM8322), and the
387 Howard Hughes Medical Institute (Gilliam Fellow for Advanced Study). Mouse body
388 composition measurements were partially supported by P30-DK020572 (Michigan Diabetes
389 Research Center), P30-DK089503 (Michigan Nutrition Obesity Research Center), and U2C-
390 DK110678 (Michigan Mouse Metabolic Phenotyping Center). Generation of the CRISPR mice
391 was partially supported by the Mouse Genetics Core of the Michigan Diabetes Research Center
392 (P30-DK020572). Studies in humans were supported by the Wellcome Trust (207462/Z/17/Z),
17 Page 19 of 46 Diabetes
393 NIHR (National Institute for Health Research) Cambridge Biomedical Research Centre, Bernard
394 Wolfe Health Neuroscience Endowment (all to ISF), and Wellcome Trust (WT206194, to IB). We
395 thank Drs. Malcolm Low, Miriam Meisler, Lei Yin, Xin Tong, Liangyou Rui, and Stephanie Bielas
396 for helpful discussions and Dr. Rui (University of Michigan) for the gift of the SH2B1 KO strain.
397 We acknowledge the Wellcome-MRC Institute of Metabolic Science (IMS) Translational
398 Research Facility (TRF) and Imaging Core Facility, both supported by a Wellcome Strategic
399 Award [100574/Z/12/Z], the University of Michigan DNA Sequencing Core for DNA sequencing,
400 and Dr. Thomas Saunders, Galina Gavrilina and Dr. Wanda Filipiak of the University of Michigan
401 Transgenic Animal Model Core as well as the Michigan Diabetes and Research Center Molecular
402 Genetics Core for help making the mouse models. We are indebted to the participants and their
403 families for their participation and to the physicians involved in the Genetics of Obesity Study
404 (GOOS) (www.goos.org.uk). Parts of this work have been presented in preliminary form as poster
405 or short oral presentations at: 2015 Gordon Research Conference: Neurotrophic Factors, Newport,
406 RI; 2016 Keystone Symposia – Axons: From Cell Biology to Pathology, Santa Fe, NM; 2017
407 Keystone Symposia Meeting: Neuronal Control of Appetite, Metabolism, and Weight,
408 Copenhagen, Denmark; 2017 National SACNAS Conference, Salt Lake City, UT; 2018 Gordon
409 Research Conference: Molecular and Cellular Neurobiology, Hong Kong, China; 2018 National
410 SACNAS Conference, San Antonio, TX; 2018 American Physiological Society: Experimental
411 Biology, San Diego, CA; 2019 Keystone Symposia: Functional Neurocircuitry of Feeding
412 Disorders, Banff, Canada. The views expressed are those of the author(s) and not necessarily
413 those of the NHS, the NIHR, or the NIH.
414
18 Diabetes Page 20 of 46
415 Author contributions
416 A.F., L.S.A., I.S.F, M.G.M., and C.C-S. developed the concept, designed experiments, and
417 interpreted the data. A.F. and L.S.A. designed and generated the mice. A.F. directed experiments,
418 conducted experiments, analyzed data, and prepared the manuscript. C.C-S., L.S.A., I.S.F. and
419 M.G.M. made revisions to the manuscript. A.M., L.D., G.C., and Y.H. helped measure body
420 weight and food intake (Figs. 2A-B, H-I, 4C, 5A, Supplementary Figs. S2A-B, F-G). A.M. helped
421 re-genotype the mice. P.V. conducted neurite outgrowth experiments (Fig. 3D) and helped with
422 experiments for Figs. 3A and C. R.M.J. conducted preliminary experiments for Fig. 3B. E.S.C.
423 maintained mouse colonies and helped genotype mice and collect blood samples (Fig. 4G and 6B).
424 J.M.C. made the GFP-SH2B1 WT and GFP-SH2B1 PR constructs (Fig. 3). J.S. and L.S.A.
425 analyzed the model of the PH domain (Fig. 7 and Video1). L.K.S and E.M.d.O. characterized the
426 human mutations in cells (Fig. 1B); J.M.K., E.H and I.S.F. performed the clinical studies in
427 mutation carriers (Figs. 1A, Table 1); I.S.F. and I.B. performed the genetic studies (Figs. 1A, Table
428 1). All authors approved the final content. C.C.S. and I.S.F. are the guarantors of this work and,
429 between them, had full access to the combined data in the study and take responsibility for the
430 integrity of the data and the accuracy of the data analysis for their respective studies.
431
432 Disclosure summary: The authors have nothing to disclose.
433
434
435
436 References
437 1. Doche MD, Bochukova EG, Su HW, Pearce L, Keogh JM, Henning E, Cline JM, Dale A, 438 Cheetham T, Barroso I, Argetsinger LS, O'Rahilly SO, Rui L, Carter-Su C, Farooqi IS.
19 Page 21 of 46 Diabetes
439 2012 SH2B1 mutations are associated with maladaptive behavior and obesity. J Clin 440 Invest. 122(12):4732-4736. 441 2. Pearce LR, Joe R, Doche MD, Su HW, Keogh JM, Henning E, Argetsinger LS, 442 Bochukova EG, Cline JM, Garg S, Saeed S, Shoelson S, O'Rahilly S, Barroso I, Rui L, 443 Farooqi IS, Carter-Su C. 2014 Functional characterisation of obesity-associated variants 444 involving the alpha and beta isoforms of human SH2B1. Endocrinology. 445 9(Sept;155):3219-3226. 446 3. Duan C, Yang H, White MF, Rui L. 2004 Disruption of SH2-B causes age-dependent 447 insulin resistance and glucose intolerance. Mol Cell Biol. 24(17):7435-7443. 448 4. Jiang L, Su H, Keogh JM, Chen Z, Henning E, Wilkinson P, Goodyer I, Farooqi IS, Rui 449 L. 2018 Neural deletion of Sh2b1 results in brain growth retardation and reactive 450 aggression. FASEB journal : official publication of the Federation of American Societies 451 for Experimental Biology. 32(4):1830-1840. 452 5. Ren D, Li M, Duan C, Rui L. 2005 Identification of SH2-B as a key regulator of leptin 453 sensitivity, energy balance and body weight in mice. Cell Metab. 2:95-104. 454 6. Ren D, Zhou Y, Morris D, Li M, Li Z, Rui L. 2007 Neuronal SH2B1 is essential for 455 controlling energy and glucose homeostasis. J Clin Invest. 117(2):397-406. 456 7. Kurzer JH, Argetsinger LS, Zhou Y-J, Kouadio J-L, O'Shea JJ, Carter-Su C. 2004 457 Tyrosine 813 is a site of JAK2 autophosphorylation critical for activation of JAK2 by 458 SH2-Bβ. Mol Cell Biol. 24(10):4557-4570. 459 8. Li Z, Zhou Y, Carter-Su C, Myers Jr. MG, Rui L. 2007 SH2B1 enhances leptin signaling 460 by both Janus kinase 2 Tyr813 phosphorylation-dependent and -independent mechanisms. 461 Mol Endocrinol. 21(9):2270-2281. 462 9. Nelms K, O'Neill TJ, Li S, Hubbard SR, Gustafson TA, Paul WE. 1999 Alternative 463 splicing, gene localization, and binding of SH2-B to the insulin receptor kinase domain. 464 Mammalian Genome. 10(12):1160-1167. 465 10. Qian X, Riccio A, Zhang Y, Ginty DD. 1998 Identification and characterization of novel 466 substrates of Trk receptors in developing neurons. Neuron. 21(5):1017-1029. 467 11. Riedel H, Wang J, Hansen H, Yousaf N. 1997 PSM, an insulin-dependent, pro-rich, PH, 468 SH2 domain containing partner of the insulin receptor. J Biochem. 122(6):1105-1113. 469 12. Rui L, Carter-Su C. 1999 Identification of SH2-Bβ as a potent cytoplasmic activator of 470 the tyrosine kinase Janus kinase 2. Proc Nat Acad Sci USA. 96(13):7172-7177. 471 13. Shih CH, Chen CJ, Chen L. 2013 New function of the adaptor protein SH2B1 in brain- 472 derived neurotrophic factor-induced neurite outgrowth. PloS one. 8(11):e79619. 473 14. Qian X, Ginty DD. 2001 SH2-B and APS are multimeric adapters that augment TrkA 474 signaling. Mol Cell Biol. 21(5):1613-1620. 475 15. Kurzer JH, Saharinen P, Silvennoinen O, Carter-Su C. 2006 Binding of SH2-B family 476 members within a potential negative regulatory region maintains JAK2 in an active state. 477 Mol Cell Biol. 26(17):6381-6394. 478 16. Duan C, Li M, Rui L. 2004 SH2-B promotes insulin receptor substrate 1 (IRS1)- and 479 IRS2-mediated activation of the phosphatidylinositol 3-kinase pathway in response to 480 leptin. J Biol Chem. 279(42):43684-43691. 481 17. Morris DL, Cho KW, Zhou Y, Rui L. 2009 SH2B1 enhances insulin sensitivity by both 482 stimulating the insulin receptor and inhibiting tyrosine dephosphorylation of insulin 483 receptor substrate proteins. Diabetes. 58(9):2039-2047.
20 Diabetes Page 22 of 46
484 18. Diakonova M, Gunter DR, Herrington J, Carter-Su C. 2002 SH2-Bβ is a Rac-binding 485 protein that regulates cell motility. J Biol Chem. 277(12):10669-10677. 486 19. Joe RM, Flores A, Doche ME, Cline JM, Clutter ES, Vander PB, Riedel H, Argetsinger 487 LS, Carter-Su C. 2018 Phosphorylation of the Unique C-Terminal Tail of the Alpha 488 Isoform of the Scaffold Protein SH2B1 Controls the Ability of SH2B1alpha To Enhance 489 Nerve Growth Factor Function. Mol Cell Biol. 38(6). 490 20. Timper K, Bruning JC. 2017 Hypothalamic circuits regulating appetite and energy 491 homeostasis: pathways to obesity. Dis Model Mech. 10(6):679-689. 492 21. Rui L, Herrington J, Carter-Su C. 1999 SH2-B is required for nerve growth factor- 493 induced neuronal differentiation. J Biol Chem. 274(15):10590-10594. 494 22. Chen L, Carter-Su C. 2004 Adapter protein SH2-Bβ undergoes nucleocytoplasmic 495 shuttling: implications for nerve growth factor induction of neuronal differentiation. Mol 496 Cell Biol. 24(9):3633-3647. 497 23. Maures TJ, Chen L, Carter-Su C. 2009 Nucleocytoplasmic shuttling of the adapter 498 protein SH2B1β (SH2-Bβ) is required for nerve growth factor (NGF)-dependent neurite 499 outgrowth and enhancement of expression of a subset of NGF-responsive genes. Mol 500 Endocrinol. 23:1077-1091. 501 24. Yousaf N, Deng Y, Kang Y, Riedel H. 2001 Four PSM/SH2-B alternative splice variants 502 and their differential roles in mitogenesis. J Biol Chem. 276(44):40940-40948. 503 25. Maures TJ. Molecular mechanisms by which adapter protein SH2B1β facilitates NGF- 504 dependent neuronal differentiation. Cellular and Molecular Biology. Vol Ph.D. Ann 505 Arbor: University of Michigan; 2008:261. 506 26. Maures TJ, Su H-W, Argetsinger LS, Grinstein S, Carter-Su C. 2011 Phosphorylation 507 controls a dual function polybasic NLS in the adapter protein SH2B1β to regulate its 508 cellular function and distribution between the plasma membrane, cytoplasm and nucleus. 509 J Cell Sci. 124(Pt9):1542-1552. 510 27. Baumeister MA, Rossman KL, Sondek J, Lemmon MA. 2006 The Dbs PH domain 511 contributes independently to membrane targeting and regulation of guanine nucleotide- 512 exchange activity. Biochem J. 400(3):563-572. 513 28. Klein DE, Lee A, Frank DW, Marks MS, Lemmon MA. 1998 The pleckstrin homology 514 domains of dynamin isoforms require oligomerization for high affinity phosphoinositide 515 binding. J Biol Chem. 273(42):27725-27733. 516 29. Lemmon MA. 2003 Phosphoinositide recognition domains. Traffic. 4(4):201-213. 517 30. Park WS, Heo WD, Whalen JH, O'Rourke NA, Bryan HM, Meyer T, Teruel MN. 2008 518 Comprehensive identification of PIP3-regulated PH domains from C. elegans to H. 519 sapiens by model prediction and live imaging. Mol Cell. 30(3):381-392. 520 31. Farooqi IS, Keogh JM, Yeo GS, Lank EJ, Cheetham T, O'Rahilly S. 2003 Clinical 521 spectrum of obesity and mutations in the melanocortin 4 receptor gene. N Eng J Med. 522 348(12):1085-1095. 523 32. Wheeler E, Huang N, Bochukova EG, Keogh JM, Lindsay S, Garg S, Henning E, 524 Blackburn H, Loos RJ, Wareham NJ, O'Rahilly S, Hurles ME, Barroso I, Farooqi IS. 525 2013 Genome-wide SNP and CNV analysis identifies common and low-frequency 526 variants associated with severe early-onset obesity. Nat Genet. 45(5):513-517. 527 33. Hendricks AE, Bochukova EG, Marenne G, Keogh JM, Atanassova N, Bounds R, 528 Wheeler E, Mistry V, Henning E, Korner A, Muddyman D, McCarthy S, Hinney A, 529 Hebebrand J, Scott RA, Langenberg C, Wareham NJ, Surendran P, Howson JM,
21 Page 23 of 46 Diabetes
530 Butterworth AS, Danesh J, Nordestgaard BG, Nielsen SF, Afzal S, Papadia S, Ashford S, 531 Garg S, Millhauser GL, Palomino RI, Kwasniewska A, Tachmazidou I, O'Rahilly S, 532 Zeggini E, Barroso I, Farooqi IS, Understanding Society Scientific G, Consortium E-C, 533 Consortium UK. 2017 Rare Variant Analysis of Human and Rodent Obesity Genes in 534 Individuals with Severe Childhood Obesity. Sci Rep. 7(1):4394. 535 34. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. 1985 536 Homeostasis model assessment: insulin resistance and beta-cell function from fasting 537 plasma glucose and insulin concentrations in man. Diabetologia. 28(7):412-419. 538 35. Muniyappa R, Lee S, Chen H, Quon MJ. 2008 Current approaches for assessing insulin 539 sensitivity and resistance in vivo: advantages, limitations, and appropriate usage. Am J 540 Physiol Endocrinol Metab. 294(1):E15-26. 541 36. Ran FA, Hsu PD, Wright J, Agarwala V, Scott DA, Zhang F. 2013 Genome engineering 542 using the CRISPR-Cas9 system. Nat Protoc. 8(11):2281-2308. 543 37. Truett GE, Heeger P, Mynatt RL, Truett AA, Walker JA, Warman ML. 2000 Preparation 544 of PCR-quality mouse genomic DNA with hot sodium hydroxide and tris (HotSHOT). 545 Biotechniques. 29(1):52, 54. 546 38. Li Q, Cao X, Qiu H-Y, Lu J, Gao R, Liu C, Yuan M-X, Yang G-R, Yang J-K. 2016 A 547 three-step programmed method for the identification of causative gene mutations of 548 maturity onset diabetes of the young (MODY). Gene. 588(2):141-148. 549 39. Sheng L, Liu Y, Jiang L, Chen Z, Zhou Y, Cho KW, Rui L. 2013 Hepatic SH2B1 and 550 SH2B2 regulate liver lipid metabolism and VLDL secretion in mice. PLoS One. 551 8(12):e83269. 552 40. van der Aa MP, Fazeli Farsani S, Kromwijk LA, de Boer A, Knibbe CA, van der Vorst 553 MM. 2014 How to screen obese children at risk for type 2 diabetes mellitus? Clin 554 Pediatr (Phila). 53(4):337-342. 555 41. Chen Z, Morris DL, Jiang L, Liu Y, Rui L. 2014 SH2B1 in beta-cells promotes insulin 556 expression and glucose metabolism in mice. Molecular endocrinology. 28(5):696-705. 557 42. Chen Z, Morris DL, Jiang L, Liu Y, Rui L. 2014 SH2B1 in beta-cells regulates glucose 558 metabolism by promoting beta-cell survival and islet expansion. Diabetes. 63(2):585-595. 559 43. Hurtado C, Erquiaga I, Aranaz P, Migueliz I, Garcia-Delgado M, Novo FJ, Vizmanos JL. 560 2011 LNK can also be mutated outside PH and SH2 domains in myeloproliferative 561 neoplasms with and without V617FJAK2 mutation. Leuk Res. 35(11):1537-1539. 562 44. Lasho TL, Mudireddy M, Finke CM, Hanson CA, Ketterling RP, Szuber N, Begna KH, 563 Patnaik MM, Gangat N, Pardanani A, Tefferi A. 2018 Targeted next-generation 564 sequencing in blast phase myeloproliferative neoplasms. Blood Adv. 2(4):370-380. 565 45. McMullin MF, Cario H. 2016 LNK mutations and myeloproliferative disorders. Am J 566 Hematol. 91(2):248-251. 567 46. Oh ST, Gotlib J. 2010 JAK2 V617F and beyond: role of genetics and aberrant signaling 568 in the pathogenesis of myeloproliferative neoplasms. Expert Rev Hematol. 3(3):323-337. 569 47. Pardanani A, Lasho T, Finke C, Oh ST, Gotlib J, Tefferi A. 2010 LNK mutation studies 570 in blast-phase myeloproliferative neoplasms, and in chronic-phase disease with TET2, 571 IDH, JAK2 or MPL mutations. Leukemia. 24(10):1713-1718. 572 48. Tefferi A, Lasho TL, Guglielmelli P, Finke CM, Rotunno G, Elala Y, Pacilli A, Hanson 573 CA, Pancrazzi A, Ketterling RP, Mannarelli C, Barraco D, Fanelli T, Pardanani A, 574 Gangat N, Vannucchi AM. 2016 Targeted deep sequencing in polycythemia vera and 575 essential thrombocythemia. Blood Adv. 1(1):21-30.
22 Diabetes Page 24 of 46
576 49. Li H, Tochio N, Koshiba S, Inoue M, Kigawa T, Yokoyama S. 2003 Solution Structure 577 of the Pleckstrin Homology Domain of Mouse APS. RCSB Protein Data Bank. 578 50. Vaudry D, Stork PJ, Lazarovici P, Eiden LE. 2002 Signaling pathways for PC12 cell 579 differentiation: making the right connections. Science. 296(5573):1648-1649. 580 51. Karczewski KJ, Francioli LC, Tiao G, Cummings BB, Alföldi J, Wang Q, Collins RL, 581 Laricchia KM, Ganna A, Birnbaum DP, Gauthier LD, Brand H, Solomonson M, Watts 582 NA, Rhodes D, Singer-Berk M, Seaby EG, Kosmicki JA, Walters RK, Tashman K, 583 Farjoun Y, Banks E, Poterba T, Wang A, Seed C, Whiffin N, Chong JX, Samocha KE, 584 Pierce-Hoffman E, Zappala Z, O'Donnell-Luria AH, Vallabh Minikel E, Weisburd B, 585 Lek M, Ware JS, Vittal C, Armean IM, Bergelson L, Cibulskis K, Connolly KM, 586 Covarrubias M, Donnelly S, Ferriera S, Gabriel S, Gentry J, Gupta N, Jeandet T, Kaplan 587 D, Llanwarne C, Munshi R, Novod S, Petrillo N, Roazen D, Ruano-Rubio V, Saltzman 588 A, Schleicher M, Soto J, Tibbetts K, Tolonen C, Wade G, Talkowski ME, Neale BM, 589 Daly MJ, MacArthur DG. 2019 Variation across 141,456 human exomes and genomes 590 reveals the spectrum of loss-of-function intolerance across human protein-coding genes. 591 bioRxiv.531210. 592
593
23 Page 25 of 46 Diabetes
594 Table 1. Phenotypes seen in carriers of rare variants in SH2B1 SH2B1 MAF Age Gender BMI Leptin HOMA Variant (%) (years) (sds) (ng/ml) -IR delay (S) lability (+) lability Anxiety (+) Dyspraxia (+) inattention (+) inattention Hyperactivity / Autistic traits (+) Speech and language Speech and language Learning difficulties (+) Learning Aggression / Emotional Aggression / Emotional
R44C 0.005 4.7 F 29 25.1 2.7 + S + + + + + (c.130C>T) (4.7) G59W 0 * 13.3 M 33 63.3 2.5 - + - + + - (c.175G>T) (3.2) S141F 0 * 7.7 F 29 (c.442C>T) (3.6) K150R† 0.019 2.8 F 25 (c.449A>G) (4.2) R167H 0.002 11.4 M 38 85.4 2.8 (c.500G>A) (3.7) V209I 0.002 3.4 M 27 18.4 1.7 + S - - - - + (c.625G>A) (5.1) R227C 0.011 15.5 F 44 - - + - + - (c.679C>T) (3.9) R227C 0.011 7.5 M 29 + S - + + + + (c.679C>T) (4.0) G238C ‡ 8.1 M 31 56.9 4.5 + + + + + + (c.712G>T) (3.9) R270Q 0.002 9.5 F 37 76.3 5.2 (c.809G>A) (4.0) R270W 0.001 4.5 M 25 4.7 + S - + + - + (c.808C>T) (4.3) E299G 0.002 16.1 F 40 50.5 2.5 - - + + - - (c.896A>G) (3.6) M388V ‡ 16.5 M 41 118 9.3 (c.1162A>G) (3.7) G516S 0.002 11 M 29 + - - + + + (c.1546G>A) (3.1) S616P 0.019 7.0 F 26 27.9 1.3 (c.1846T>C) (3.4) G638R ‡ 4.3 F 31 47.5 3.3 - - - - + - (c.1912 G>A) (5.3) 595 596 MAF, Minor Allele Frequency (%) in Non-Finnish European (NFE) gnomAD exomes (51); "0", variant 597 detected in gnomAD exomes with MAF=0 in the NFE sample; * = variant not present in NFE samples but 598 present in samples from other ethnic groups; "†", variant not found in any gnomAD exomes; "‡", 599 homozygous variant in a proband from a consanguineous family. Gender (M; male; F; female); BMI, Body 600 Mass Index = weight in kg/height in m2 with age- and gender-adjusted BMI standard deviation score shown 601 in brackets. When fasting plasma glucose and insulin were available, HOMA-IR (Homeostatic Model 602 Assessment of Insulin Resistance) was calculated. Scores in young obese children are difficult to interpret
24 Diabetes Page 26 of 46
603 given the paucity of control data. However, in adults, HOMA-IR >1.9 indicates insulin resistance. 604 Behavioral data was gathered from medical history. "+", indicates presence of a phenotype. "-", indicates 605 absence of a phenotype. “S” indicates speech and language delay. Blank cell = indicates insufficient 606 information to record a phenotype. 607
608
609
610
611
612
613
614
615
616
617
618
619
620 Figure legends
621 Fig 1: Identification of SH2B1 variants and generation of SH2B1 P322S mice. A) Human
622 SH2B1 protein (NP_001139268). P, proline-rich region; DD, dimerization domain; NLS, nuclear
623 localization sequence; NES, nuclear export sequence; PH, pleckstrin homology domain. Amino
624 acid residues for newly characterized human obesity-associated SH2B1 variants and the previously
625 characterized variant P322S are shown. B) SH2B1 mutations impair the ability of SH2B1 to
626 enhance neurite outgrowth. PC12 cells transiently co-expressing GFP and either empty
627 pcDNA3.1(+) vector (-), human SH2B1 WT or SH2B1 mutant were treated with 20 ng/ml rat
628 NGF for 3 days after which neurite outgrowth was assessed. R555E lacking an intact SH2 domain
25 Page 27 of 46 Diabetes
629 and the human mutation P322S reported previously were included as positive controls. GFP-
630 positive cells were scored for the presence of neurites two times the length of the cell body (≥ 400
631 cells/condition/experiment). The percentage of cells with neurites was determined by dividing the
632 number of GFP-positive cells with neurites by the total number of GFP-positive cells. Means ±
633 SEM, n=5. Each construct was compared to WT using a two-tailed Student’s t test. *P<0.05,
634 **P<0.01, ***P<0.001, ****P<0.0001. C) CRISPR-cas9 schematic for SH2B1 P322S gene
635 editing. RNA guide sequence, PAM sequence, and cut site for Cas9 are shown for guide 2. The
636 region of the 180 nt oligo donor template used to direct homology-directed repair in the vicinity
637 of the P317, R318 deletion and P322 is shown. Mutations in donor template that introduce the
638 C>T mutation to code for P322S and silent mutations to create a diagnostic XbaI site and disrupt
639 guide RNA binding following repair are highlighted in red. D) Proteins in brain tissue lysates
640 from SH2B1 WT, P322S/P322S, and KO male mice were immunoblotted with αSH2B1 or α-
641 tubulin. Migration of the 100 kDa protein standard and the four isoforms of SH2B1 are shown.
642 E) P322S/P322S mice from intercrosses of heterozygous mice are born at ~half of the expected
643 Mendelian ratio. n = 179 mice. Chi-Square Test, *P<0.05.
644
645 Fig 2: The P322S mutation in SH2B1 leads to impaired glucose homeostasis in male mice
646 challenged with a high fat diet. Panels A-G: male mice fed standard chow. A) Body weight was
647 assessed at weeks 4-29. n=8 (WT), 16 (P322S/+). B) Food intake was assessed at weeks 5-25 and
648 cumulative food intake graphed. n=7 (WT), 8 (P322S/+). C) Body fat mass was determined at
649 week 30. Percent fat mass was determined by dividing fat mass by body weight. n=9 (WT), 16
650 (P322S/+). D) GTT was assessed in 28-week-old mice. After a 4-hour fast, mice were injected
651 intraperitoneally with D-glucose (2 mg/kg of body weight). Blood glucose was monitored at
26 Diabetes Page 28 of 46
652 indicated times. n=8 (WT), 14 (P322S/+). E) ITT was assessed in 29-week-old mice. After a 6-
653 hour fast, mice were injected intraperitoneally with insulin (1 IU/kg of body weight). Blood
654 glucose was monitored at indicated times. n=8 (WT), 15 (P322S/+). F) At week 30, serum from
655 P322S/+ and control mice was assayed for leptin. n=5 (WT), 8 (P322S/+). G) 13-week-old mice
656 were fasted overnight (18:00 – 9:00) and insulin levels were determined. n=5. Panels H-M: male
657 mice fed a HFD: H) Starting at week 6, body weight of mice was assessed weekly. n=7 (WT), 12
658 (P322S/+). I) Food intake in mice was measured during week 27. n=7 (WT), 12 (P322S/+). J)
659 Body fat mass was determined at week 30. n=7 (WT), 11 (P322S/+). K) GTT was assessed as in
660 Panel 2D at 28 weeks and blood glucose monitored at times indicated. n=7 (WT), 12 (P322S/+).
661 L) ITT was assessed as in Panel 2C at 29 weeks and blood glucose monitored at the times
662 indicated. n=6 (WT), 10 (P322S/+). M) At week 30, mice were fasted overnight and insulin levels
663 were determined. n=6 (WT), 7 (P322S/+). For all comparisons: Means ± SEM, * P<0.05,
664 **P<0.01, n.s. = not significant.
665
666 Fig 3: Disruption of the PH domain changes the subcellular localization of SH2B1 and
667 impairs the ability of SH2B1 to enhance NGF-induced neurite outgrowth. A) Proteins in
668 whole cell lysates from PC12 cells transiently expressing the indicated GFP-SH2B1 were
669 immunoblotted with αSH2B1. Migration of molecular weight standards are on the left. B) Live
670 293T cells and C) PC12 cells transiently expressing GFP-SH2B1 WT or GFP-SH2B1PR were
671 imaged by confocal microscopy. D) PC12 cells transiently expressing GFP, GFP-SH2B1 WT or
672 GFP-SH2B1PR were treated with 25 ng/ml mouse NGF for 2 days after which neurite
673 outgrowth was assessed. GFP-positive cells were scored for the presence of neurites > two times
674 the length of the cell body (total of 300 cells/condition/experiment). The percentage of cells with
27 Page 29 of 46 Diabetes
675 neurites was determined by dividing the number of GFP-positive cells with neurites by the total
676 number of GFP-positive cells counted. Means ± SEM, *P<0.05, n=3.
677
678 Fig 4: Disruption of the PH domain in SH2B1 results in obesity. A) mRNA was extracted from
679 brain and heart tissue of WT and PR/PR male mice. The migration of DNA standards (left) and
680 isoform-specific PCR products (right) are shown. B) Proteins in brain lysates from SH2B1 WT,
681 PR/PR, and KO male mice were immunoblotted with αSH2B1 and α-tubulin. The migration
682 of the 100 kDa protein standard (left) and the four known isoforms of SH2B1 and -tubulin (right)
683 are shown. C) Body weight was assessed weekly starting at week 4. Males: n= 9 (WT), 14 (PR/+),
684 9 (PR/PR). Females: n=7 (WT), 16 (PR/+), 11 (PR/PR). D) Representative SH2B1 WT
685 and PR/PR male mice (6 months). E) Perigonadal fat of representative SH2B1 WT and
686 PR/PR male littermates (6 months). F, H) Body fat and lean mass was determined at weeks 24-
687 26. Percent fat or lean mass was determined by dividing by body weight. Males: n=5 (WT, PR/+,
688 PR/PR). Females: n= 5 (WT), 11 (PR/+), 8 (PR/PR). G) At week 24-26, serum from
689 SH2B1 WT, PR/+, and PR/PR male and female mice was assayed for leptin. Males: n= 8
690 (WT, PR/+, PR/PR). Females: n=7 (WT), 6 (PR/+), 9 (PR/PR). For all comparisons:
691 Means ± SEM, * P<0.05, **P<0.01, and ***P<0.001, n.s. = not significant.
692
693 Fig 5: The PR mice exhibit increased food intake and reduced glucose tolerance and insulin
694 sensitivity A) Food intake was measured for weeks 18-20 and cumulative food intake graphed.
695 Males: n= 11 (WT), 10 (PR/+), 7 (PR/PR). Females: n=6 (WT), 9 (PR/+, PR/PR). Same
696 cohort of mice as Fig. 4. B) Energy expenditure was assessed at 11-12 weeks using CLAMS. VO2
697 was normalized to lean body mass (LBM). Males: n= 13 (WT), 10 (PR/+), 11 (PR/PR).
28 Diabetes Page 30 of 46
698 Females: n=11 (WT), 15 (PR/+), 13 PR/PR). C) At week 18, mice were fasted for 4 hours and
699 blood glucose was measured. Males: n= 8 (WT), 12 (PR/+), 10 (PR/PR). Females: n=7 (WT),
700 14 (PR/+), 10 (PR/PR). D) GTT was assessed at 18 weeks as in Fig. 2D and blood glucose
701 monitored at times indicated. Males: n= 8 (WT), 12 (PR/+), 10 (PR/PR). Females: n=7 (WT),
702 14 (PR/+), 10 (PR/PR). E) ITT was assessed at 19 weeks as in Fig. 2E and blood glucose
703 monitored at the times indicated. Males: n= 9 (WT), 14 (PR/+), 11 (PR/PR). Females: n=6
704 (WT), 13 (PR/+), 11 (PR/PR). Same cohort of mice as Fig. 4. For all comparisons: Means ±
705 SEM, * P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 when compared to WT littermates, n.s. =
706 not significant.
707
708 Fig 6: PR female mice exhibit reduced glucose tolerance prior to the onset of obesity. A)
709 Body fat mass was determined at week 11-12. Percent fat mass was determined by dividing the
710 mass by body weight. Males: n=13 (WT), 10 (PR/+), 11 (PR/PR). Females: n=11 (WT), 15
711 (PR/+), 13 (PR/PR). B) At week 7, serum from WT, PR/+, and PR/PR male mice was
712 assayed for leptin. Males: n=8 (WT, PR/PR), 7 (PR/+). Females: n=8 (WT), 5 (PR/+), 12
713 (PR/PR). C) 8-week-old mice were fasted for 6 hours and insulin levels were determined.
714 Males: n=11 (WT), 9 (PR/+), 10 (PR/PR). Females: n=9 (WT), 11 (PR/+), 13 (PR/PR).
715 D) At week 8, mice were fasted for 4 hours and blood glucose was measured. Males: n=13 (WT),
716 10 (PR/+), 11 (PR/PR). Females: n=10 (WT), 13 (PR/+), 12 (PR/PR). E) In a separate
717 study, GTT was assessed at 8 weeks as in Fig. 2D and blood glucose was monitored at the times
718 indicated. Males: n= 13 (WT), 10 (PR/+), 11 (PR/PR). Females: n=10 (WT), 13 (PR/+), 12
719 (PR/PR). F) ITT was assessed at 9 weeks as in Fig. 2E and blood glucose monitored at the times
29 Page 31 of 46 Diabetes
720 indicated. Males: n=11 (WT), 9 (PR/+), 10 (PR/PR). Females: n=9 (WT), 11 (PR/+), 13
721 (PR/PR). Same cohort of mice as Fig. 5b. For all comparisons: Means ± SEM, * P<0.05,
722 **P<0.01, ***P<0.001, ****P<0.0001 when compared to WT littermates, n.s. = not significant.
723
724 Fig 7: ClustalW analysis and modeling of the 3-D structure of the PH domain of SH2B1. A)
725 ClustalW of SH2B1, SH2B2/APS, and SH2B3/Lnk in the region included in the NMR structure
726 of SH2B2/APS. Homologous residues are highlighted in black, functionally homologous residues
727 are cyan. The PH domain is indicated by the blue line below the sequences. P317, R318 in SH2B1
728 and the residues in SH2B1 for which mutations are associated with obesity are indicated by
729 magenta ovals. The mutations in Lnk associated with myeloproliferative neoplasms are indicated
730 by orange ovals. The mutations are noted above the Clustal. Residues within 8Å of P317 in
731 SH2B1 (P240 in SH2B3/Lnk) are denoted by taupe ovals. Residues within 8Å of R318 in SH2B1
732 (K241 in SH2B3/Lnk) are denoted by purple ovals. Residues within 8Å of P322 in SH2B1 (P245
733 in SH2B3/Lnk) are denoted by green ovals. B) Still image for Video 1.
734
735 Video Legend
736 Video 1: Modeling of the 3-D structure of the PH domain of SH2B1. A model of human SH2B1
737 was created by overlaying the sequence of the PH domain of human SH2B1 onto the mouse
738 structure of APS. The two PH domain sequences are 75% similar (55.3% identical). The SH2B1
739 model is missing a single amino acid insertion at residue 296 and the 16-residue insertion at residue
740 262 which contains the R270W/Q human mutation in SH2B1. The amino acid sequence of residues
741 surrounding the P322 site is highly conserved between the two proteins. P322 is shown in yellow
742 sticks. The other human mutations in SH2B1 and P317, R318 are magenta sticks. The human
30 Diabetes Page 32 of 46
743 mutations in Lnk are orange sticks. Differences between mouse and human SH2B1 are brown (no
744 sticks). Oxygen atoms are red, nitrogen blue. -pleated sheets, -helices, connecting loops, and
745 hydrogen bonds in the region between the n-terminal end of -strand 2 and the loop containing
746 P317, R318 and P322 are indicated. The surface of the PH domain is tinted grey. P322 in SH2B1
747 is within 8Å of the site of the D234N mutation in SH2B3/Lnk. The turn that contains D234N also
748 contains the G229S mutation in SH2B3/Lnk and the E299G mutation in SH2B1. P317 and R318
749 in SH2B1 are within 8Å of the site of G220R/V and A223V in SH2B3/Lnk. This region is
750 stabilized by pi-pi stacking between P317 and F309 in SH2B1 [P248 and F240 in Lnk] and a
751 network of hydrogen bonds.
31 Page 33 of 46 Diabetes
Fig 1: Identification of SH2B1 variants and generation of SH2B1 P322S mice.
173x225mm (300 x 300 DPI) Diabetes Page 34 of 46
Fig 2: The P322S mutation in SH2B1 leads to impaired glucose homeostasis in male mice challenged with a high fat diet.
94x234mm (300 x 300 DPI) Page 35 of 46 Diabetes
Fig 3: Disruption of the PH domain changes the subcellular localization of SH2B1 and impairs the ability of SH2B1 to enhance NGF-induced neurite outgrowth. %"
175x126mm (300 x 300 DPI) Diabetes Page 36 of 46
Fig 4: Disruption of the PH domain in SH2B1 results in obesity
167x233mm (300 x 300 DPI) Page 37 of 46 Diabetes
Fig 5: The ΔPR mice exhibit increased food intake and reduced glucose tolerance and insulin sensitivity
159x230mm (300 x 300 DPI) Diabetes Page 38 of 46
Fig 6: ΔPR female mice exhibit reduced glucose tolerance prior to the onset of obesity.
141x235mm (300 x 300 DPI) Page 39 of 46 Diabetes
Fig 7: ClustalW analysis and modeling of the 3-D structure of the PH domain of SH2B1.
173x179mm (300 x 300 DPI) Diabetes Page 40 of 46
Online Supplementary Data Page 41 of 46 Diabetes
Supplementary Figure S1: Validation of the P322S and PR mice. A) Sequencing of DNA from mice from the N1 generation confirms germline transmission of the SH2B1 P322S mutation. The relevant region of DNA sequence from P322S/+ and WT littermates are shown. The codons for P322 (upper panel) and S322 (lower panel) are underlined. The peaks corresponding to the C >T mutation are denoted by asterisks. B) mRNA was extracted from brain, liver, and heart tissue of SH2B1 WT, P322S/+, and P322S/P322S 31-week old male mice. The migration of DNA standards (left) and isoform-specific PCR products (right) are shown. C) Location of forward and reverse primers used for PCR in Panel S1B. D) Proteins in lysates from brain tissues from WT, P322S/P322S and KO mice were immunoprecipitated using αSH2B1 and immunoblotted with αSH2B1. The migration of the 100 kDa protein standard (left) and four known isoforms of SH2B1 (right) are shown. E) The relevant DNA sequences from ΔPR/ΔPR and WT littermates are shown. The sequence corresponding to the codons for P317, R318 is underlined in blue. The location of the missing codons for P317, R318 in the Sh2b1 ΔPR/ΔPR mice is indicated by an arrow. F) The ΔPR mice are born at the expected (1 out of 4) Mendelian ratio. n = 183 mice. G and H) Body length was assessed at weeks 8-10. Males n=7 (WT), 9 (ΔPR/ ΔPR) and females n=5 (WT), 5 (ΔPR/ ΔPR). Diabetes Page 42 of 46
Supplementary Figure S2: The P322S SH2B1 mutation leads to impaired glucose homeostasis in female mice challenged with a HFD. Panels A-E: female mice fed standard chow. A) Body weight was assessed at weeks 4-29. n=10 (WT), 14 (P322S/+). B) Food intake was assessed at weeks 5-25 and cumulative food intake graphed. Females n=7 (WT), 9 (P322S/+). Page 43 of 46 Diabetes
C) Body fat mass was determined at week 30. Percent fat mass was determined by dividing fat mass by body weight. n=14 (WT, P322S/+). D) GTT was assessed at 28 weeks. After a 4-hour fast (9:00 – 13:00), mice were injected intraperitoneally with D-glucose (2 mg/kg of body weight). Blood glucose was monitored at indicated times. n=9 (WT), 14 (P322S/+). E) ITT was assessed at 29 weeks. After a 6-hour fast (8:00 – 14:00), mice were injected intraperitoneally with insulin (1 IU/kg of body weight). Blood glucose was monitored at indicated times. n=9 (WT), 13 (P322S/+). Panels F-K: female mice fed a HFD: F) Starting at week 6, body weight was assessed weekly. n=9 (WT), 20 (P322S/+). G) Food intake was measured during week 27. n=5 (WT), 14 (P322S/+). H) Body fat mass was determined at week 30. n=9 (WT), 20 (P322S/+). I) GTT was assessed at 28 weeks. Mice were fasted for 4 hours (9:00-13:00) and then injected intraperitoneally with D-glucose (2 mg/kg of body weight). Blood glucose was monitored at times indicated. Females: n=9 (WT), 20 (P322S/+). J) ITT was assessed at 29 weeks. Mice were fasted for 6 hours (8:00 – 14:00) and then injected intraperitoneally with human insulin (1 IU/kg of body weight). Blood glucose was monitored at the times indicated. n=9 (WT), 20 (P322S/+). K) At week 30, mice were fasted overnight and insulin levels were determined. n=8 (WT, P322S/+). For all comparisons: Means ± SEM, *P<0.05, **P<0.01, n.s. = not significant. Diabetes Page 44 of 46
Supplementary Table S1. Oligo / Primer Table Description Location of guide / Oligo / primer primer sequence * guide 1, reverse nt 3325-3344 5’-AAACcgtaggcgtcccgaccccgtC-3’ † complement guide 1, guide nt 3325-3344 5’-CACCGacggggtcgggacgcctacg-3’ ‡ § sequence strand Guide 2, reverse nt 3337-3356 5’-AAACgaccccgtctcagcattccc-3’ complement Guide 2, guide nt 3337-3356 5’-CACCgggaatgctgagacggggtc-3’ sequence strand Donor (sense nt 3264-3443 5’- strand) gaatggagggagatacagttagatttgacacaccaaaaactgattcttcttccctttcccgcgta ggcgtctagacctagactcagcatttcctgctctactattactgatgtccgcacagccacagcc ctagagatgcctgacagggagaacacgtttgtggttaaggtaggaacccca-3’ Donor (antisense nt 3264-3443 5’- strand) tggggttcctaccttaaccacaaacgtgttctccctgtcaggcatctctagggctgtggctgtgc ggacatcagtaatagtagagcaggaaatgctgagtctaggtctagacgcctacgcgggaaa gggaagaagaatcagtttttggtgtgtcaaatctaactgtatctccctccattc-3’ Genotyping SH2B1 nt 3156-3185 5’-tattgctgtcctgggttcagtgctaactgt-3’ P322S and ∆PR mice nt 3518-3547 5’-aagactcaaagcccccgacatatactcatc-3’ Genotyping SH2B1 nt 6280-6303 5’- ggaggcactggctcccatggtgtc-3’ KO mice nt 6460-6483 5’-gcaggatgacaagtgaggtgggag-3’ From neo cassette 5’-attcctcccactcatgatctatagatc-3’ SH2B1 gene nt 2387-2410 5’-ttcgatatgcttgagcacttccgg-3’ expression nt 2651-2671 5’-gcctcttctgccccaggatgt-3’ * Nucleotide location in reverse complement of genomic Sh2b1 (accession # NC_000073, GRC m38). † AAAC overhang for subcloning into a Bbs1 site with a C for binding to the complementary G. ‡ CACCG overhang for subcloning into the second Bbs1 site with a G for initiating transcription by U6 polymerase. § guide sequence is in lower case. Page 45 of 46 Diabetes
Supplementary Data
Generation of SH2B1 P322S and P317, R318 mice
Two sets of RNA guides (Guides 1 and 2) that cut within 40 nt of the codon for P322 and had a high inverse likelihood of off-target binding were tested. The oligos required for expressing the guides (Supplementary Table S1) (Integrated DNA Technologies IDT) were subcloned into the chimeric guide RNA expression cassette in the pX330 vector, which also contained a hSpCas9 expression cassette (1,2). The sequence in the region of the guide sequence was verified. The donor to direct homology-directed repair (HDR) spanned the 180 nt region from nucleotide 3264- 3443 of Sh2b1(from IDT). The donor template introduced the C>T mutation into the codon for P322 and silent mutations to create a diagnostic Xbal site and disrupt guide RNA binding following HDR (Fig. 1C). The donor and the pX330 plasmid containing the guide sequence were purified initially using the University of Michigan Transgenic Animal Model Core procedure and subsequently Qiaprep miniprep columns (Qiagen, # 27106) using filtered buffers (Anotop 0.02 m filters, Whatman, # 6809-1002). The University of Michigan Transgenic Animal Model Core tested the purified donor and guides in blastocysts. They then injected each guide/donor combination into ~75 oocytes from C57BL/6J mice, and implanted the oocytes into C57BL/6J mice. The resulting pups were genotyped and the sequence confirmed by DNA sequencing. Guide 1 produced 4 wild-type (WT) pups and 1 pup containing an indel. Guide 2 produced 2 pups containing the P322S edit and 4 pups containing indels. The same Guide 2 founder was used for all experiments. N2 mice were used for experiments in Fig. 2A-G and Supplementary Fig. S2A- E, N5 mice for Fig. 2H-M and Supplementary Fig. S2F-K. One of the indels encoded SH2B1 P317, R318 (PR). N3 mice were used for experiments in Figs. 4, 5A, 5C-E, N4 mice for experiments in Figs. 5B and 6.
Gene expression
Total RNA was extracted using TRIzol Reagent (Ambion, Life Technologies). cDNA was synthesized from the RNA using Taqman Reverse Transcription Reagent Kit (Applied Bio- Systems, # N808-0234). PCR was performed using GoTaq DNA polymerase (Promega, Catalog #M3005). Primers for genotyping SH2B1 P322S and ∆PR mice (Supplementary Table S1) and the PCR parameters (95°C 2 minutes, 95°C 15 seconds, 63°C 30 seconds, 72°C 1:20 minutes, repeat 42 cycles, 72°C 4 minutes, 4°C) were chosen to optimize detection of all four SH2B1 isoforms.
Mouse body weight and food intake
Body weight was assessed weekly at the same time and day of the week. To assess food intake, food was added and weighed at the start of each week. At the end of the week, the food that remained was removed and weighed, including any food in the bedding. Bedding was changed each week. Mice that nibble on the food but do not necessarily eat it, were excluded from the study. Food intake shown is cumulative intake. Food intake, glucose tolerance tests, insulin tolerance tests, and blood draws for determination of hormone levels were also performed on these cohorts of mice. Diabetes Page 46 of 46
Mouse body length
Body length was measured in mice anesthetized with isoflurane in order to avoid inaccuracy from movement. Body length was measured with a ruler from tip of nose to middle of anus as previously described in (3).
Plasmids
GFP-tagged rat SH2B1 (4) and SH2B1 P322S (5) (GenBank accession number NM_001048180) have been described previously. The PR deletion was introduced into GFP- SH2B1 using a QuikChange II site-directed mutagenesis kit (Stratagene) and the deletion confirmed by sequencing. For Fig. 1B, human SH2B1 in pcDNA3.1(+) (GenBank accession number NM_015503) was purchased from Origene (# RC222333) and GFP from Addgene (plasmid no. 13031). Human SH2B1 was mutated using QuikChange II site-directed mutagenesis and the sequence confirmed by Sanger sequencing (3730 DNA Analyser, Thermo Scientific).
Neurite outgrowth
For Fig. 1B, PC12 cells (from Harvey McMahon, Laboratory of Molecular Biology, Cambridge, UK) were grown in suspension and maintained in PC12 growth medium B (RPMI growth medium, 7.5% FBS, 7.5% HS). The PC12 cells were transferred into serum-free Opti-MEM medium (Thermo Scientific) and passed through a 21G needle four times to break up lumps of cells. The cells were seeded into collagen IV (Corning, #354429) coated 96-well plates at a density of 2500 cells per well. The next day, cells were co-transfected with 50 ng/well of each construct encoding GFP and either WT or mutant SH2B1 using Lipofectamine 2000 (Thermo Scientific). Four hours later, rat NGF (20 ng/ml) (R&D Systems) was added. After 72 hours, cells were photographed (EVOS FL Cell imaging system, ThermoFisher Scientific) under blue laser and the number of GFP-expressing cells were assessed for the presence of neurites greater than two times the length of the cell body using the ImageJ (6) ‘Cell Counter’ plugin (https://imagej.nih.gov).
Supplementary Data References
1. Ran FA, Hsu PD, Lin CY, Gootenberg JS, Konermann S, Trevino AE, Scott DA, Inoue A, Matoba S, Zhang Y, Zhang F. 2013 Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing specificity. Cell. 154(6):1380-1389. 2. Ran FA, Hsu PD, Wright J, Agarwala V, Scott DA, Zhang F. 2013 Genome engineering using the CRISPR-Cas9 system. Nat Protoc. 8(11):2281-2308. 3. Barnett SA. 1965 Genotype and environment in tail length in mice. Q J Exp Physiol Cogn Med Sci. 50(4):417-429. 4. Rui L, Herrington J, Carter-Su C. 1999 SH2-B is required for nerve growth factor- induced neuronal differentiation. J Biol Chem. 274(15):10590-10594. 5. Doche MD, Bochukova EG, Su HW, Pearce L, Keogh JM, Henning E, Cline JM, Dale A, Cheetham T, Barroso I, Argetsinger LS, O'Rahilly SO, Rui L, Carter-Su C, Farooqi IS. 2012 SH2B1 mutations are associated with maladaptive behavior and obesity. J Clin Invest. 122(12):4732-4736. Page 47 of 46 Diabetes
6. Schneider CA, Rasband WS, Eliceiri KW. 2012 NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 9(7):671-675.