Asna1/TRC40 Controls B-Cell Function and Endoplasmic Reticulum Homeostasis by Ensuring Retrograde Transport

110 Diabetes Volume 65, January 2016

Stefan Norlin,1 Vishal S. Parekh,1 Peter Naredi,2 and Helena Edlund1

Asna1/TRC40 Controls b-Cell Function and Endoplasmic Reticulum Homeostasis by Ensuring Retrograde Transport

Diabetes 2016;65:110–119 | DOI: 10.2337/db15-0699

Type 2 diabetes (T2D) is characterized by insulin re- and misfolded proteins accumulate within the ER, and ER sistance and b-cell failure. Insulin resistance per se, stress develops, leading to activation of the unfolded however, does not provoke overt diabetes as long as protein response (UPR). During the development of type 2 compensatory b-cell function is maintained. The in- diabetes (T2D), pancreatic b-cells initially compensate creased demand for insulin stresses the b-cell endo- for insulin resistance successfully by increasing insulin plasmic reticulum (ER) and secretory pathway, and ER biosynthesis and secretion. However, conditions that b stress is associated with -cell failure in T2D. The tail lead to sustained ER stress (i.e., prolonged and persistent recognition complex (TRC) pathway, including Asna1/ insulin resistance and/or failure to reestablish proper ER TRC40, is implicated in the maintenance of endomem- homeostasis) are implicated in the deterioration of b-cell brane trafficking and ER homeostasis. To gain insight function and the development of overt diabetes (1–3). into the role of Asna1/TRC40 in maintaining endomem- Thus, identification of key molecules and factors that en- brane homeostasis and b-cell function, we inactivated b2/2 fi ISLET STUDIES Asna1 b Asna1 sure proper membrane traf cking and ER homeostasis, in -cells of mice. We show that mice b develop hypoinsulinemia, impaired insulin secretion, and thereby -cell function and survival, is important to and glucose intolerance that rapidly progresses to overt gaining insight into the etiology of T2D. diabetes. Loss of Asna1 function leads to perturbed In yeast, the Guided Entry of Tail-anchored proteins plasma membrane-to-trans Golgi network and Golgi- (GET) pathway (i.e., the tail recognition complex [TRC] to-ER retrograde transport as well as to ER stress in pathway equivalent) is associated with a broad range of b-cells. Of note, pharmacological inhibition of retrograde phenotypes (4–9). The GET complex has been suggested transport in isolated islets and insulinoma cells mim- to genetically associate with endomembrane trafficking icked the phenotype of Asna1b2/2 b-cells and resulted pathways (10,11), and inactivation of the GET pathway in reduced insulin content and ER stress. These data results in ER stress and activation of the UPR (12). Mech- support a model where Asna1 ensures retrograde anistic studies, primarily in cell-free systems, have sug- transport and, hence, ER and insulin homeostasis in gested a role for Get3 and the mammalian homolog b-cells. Asna1 (also known as TRC40) in delivering tail-anchored (TA) proteins for posttranslational insertion into the ER through the CAML/WRB receptor complex (13–16). In Secretory proteins (e.g., insulin) are inserted into the agreement with the proposed role for the GET/TRC path- endoplasmic reticulum (ER) where they are posttransla- way in membrane trafficking within the secretory path- tionally modified, folded, and then trafficked further way, key regulators of membrane-mediated transport and through the endomembrane system. If the protein load protein translocation (e.g., soluble NSF attachment pro- exceeds the protein folding capacity of the ER, unfolded tein receptors [SNAREs] such as Sec22b and Sed5 as well

1Umeå Centre for Molecular Medicine, Umeå University, Umeå, Sweden This article contains Supplementary Data online at http://diabetes 2Department of Surgery, Institute of Clinical Sciences, Sahlgrenska Academy, .diabetesjournals.org/lookup/suppl/doi:10.2337/db15-0699/-/DC1. University of Gothenburg, Gothenburg, Sweden © 2016 by the American Diabetes Association. Readers may use this article as Corresponding author: Helena Edlund, [email protected]. long as the work is properly cited, the use is educational and not for profit, and Received 25 May 2015 and accepted 28 September 2015. the work is not altered. diabetes.diabetesjournals.org Norlin and Associates 111 as Sec61b and RAMP4) have been proposed as protein using the DDCq method. TBP was used as an internal ref- clients for this pathway (17). Recently, Get3/Asna1 was erence gene. For detailed information, see Supplementary also shown to function, under oxidative stress conditions, Data. as a molecular chaperone that binds unfolded proteins to Cell Culture, Isolation, In Vitro Culture of Islets prevent their irreversible aggregation (18). In Caenorhab- Islet isolation and insulin secretion were performed ditis elegans, Asna1 function is required for larval growth essentially as previously described (24). For islet insulin and resistance to cisplatin, an oxidative stress–inducing secretion experiments, five equally sized islets were incu- anticancer drug (19,20). bated in CMRL-1066 (#21530; Gibco) supplemented with A mechanistic role for Asna1 in mammalian cells, 10% FBS (#10500; Gibco) at 37°C for 2 h. The islets were however, has not been functionally assessed in vivo equilibrated in ubiquitin (UB) buffer (2.8 mmol/L glucose, because global inactivation of Asna1 in mice results in em- 0.1% BSA) at 37°C for 1 h and then transferred to UB bryonic lethality (21). To explore a potential role for Asna1 buffer containing either 2.8 mmol/L glucose, 16.8 mmol/L in mammalian cells, we generated b-cell–specific Asna1 b2 2 b2 2 glucose, or 30 mmol/L KCl and incubated at 37°C for an mutant mice, denoted Asna1 / mice. Asna1 / mice additional 1 h (Supplementary Data). For Retro-2 treat- displayed pancreatic hypoinsulinemia, impaired insulin se- b2 2 ment experiments, MIN6 cells were passaged 1:3, cultured cretion, and early onset diabetes. b-Cells of Asna1 / for 48 h, and exposed to Retro-2 for 24 h, and islets were mice showed impaired retrograde transport, reduced cultured for 48 h after isolation and exposed to Retro-2 insulin content, and ER stress. Moreover, we show that for an additional 48 h. UB buffer (103) was prepared as Retro-2–mediated pharmacological inhibition of retrograde follows: NaCl 14.6 g, KCl 880 mg, CaCl2 Å; H O 376 mg, transport per se in isolated islets and insulinoma cells leads 2 MgCl Å; 6H O 488 mg, and HEPES 11.9 g was dissolved to decreased insulin content and ER stress. Thus, in addi- 2 2 in 200 mL H O. Upon dilution, pH was set at 7.35 and tion to identifying a role for Asna1 to ensure retrograde 2 0.1% BSA was added (ICN #105033, fatty acid free). transport as well as insulin and ER homeostasis in b-cells, the findings provide independent evidence for a role for Brefeldin A–Induced Retrograde and Anterograde retrograde transport in regulating b-cell function. Transport Assays Isolated islets were first incubated for 1 h in CMRL-1066 RESEARCH DESIGN AND METHODS supplemented with 10% FCS. For the COPI-independent Mouse Strains and Generation of Asna1flox Mice Golgi-to-ER retrograde transport assay, islets were then m A detailed description of the generation and genotyping transferred to media containing brefeldin A (BFA) 0.5 g/mL, fi of the conditional Asna1 allele is described in the Supple- and islets were removed and xed after 0, 2.5, 5, 10, mentary Data. Briefly, two loxP sites flanking exon 2 of 20, and 40 min. For the Golgi anterograde transport as- Asna1 were inserted by a recombination strategy essen- say, islets that had been incubated 1 h in CMRL-1066 tially as previously described (22). CRE recombinase– supplemented with 10% FCS were transferred to media mediated deletion of the intervening exon 2 is predicted with BFA and incubated for an additional 1 h. Islets were to result in translational termination after exon 1. ERAI then washed and incubated in medium without BFA, re- fi mice (23) were provided by the RIKEN BioResource Cen- moved, and xed after 0, 30, 45, 60, and 120 min. ter through the National BioResource Project of the Statistical Analyses MEXT, Japan. The animal studies were approved by the All numerical data are presented as mean 6 SEM. All Institutional Animal Care and Use Committee of Umeå statistical analyses were performed by heteroscedastic University and were conducted in accordance with the two-tailed Student t test. P , 0.05 was considered statis- guidelines for the care and use of laboratory animals. tically significant. Glucose Tolerance and Insulin Secretion Tests RESULTS Glucose tolerance test (GTT) and glucose-stimulated Loss of Asna1 Function in b-Cells Leads to Diabetes insulin secretion (GSIS) were performed on overnight- To elucidate the functional role of Asna1 in vivo, we – fasted (15 17 h) and sedated (Hypnorm and Dormicum) generated b-cell–specific deletion of Asna1 in mice by fl fl mice after intraperitoneal injection of glucose 2 g/kg body breeding Asna1 ox/ ox mice (Supplementary Fig. 1A–H) weight. Area under the curve (AUC) was calculated accord- with Ins1+/Cre mice (i.e., mice where the gene encoding ing to the trapezoidal rule (Supplementary Data). the CRE recombinase was inserted in one of the Ins1 Western Blot Analysis alleles [25,26]). Asna1 gene expression was progressively reduced between 2 and 4 weeks in islets of ;4-week-old Western blot expression data were normalized using b2/2 GAPDH, b-actin, or a-tubulin expression. For detailed Asna1 mice (Supplementary Fig. 2A), and conse- quently, Asna1 protein levels were decreased in information and antibodies, see Supplementary Data. b2 2 Asna1 / islets at 4 weeks (Supplementary Fig. 2B). Quantitative RT-PCR Analyses The residual Asna1 expression at 4 weeks likely predom- All quantitative RT-PCR (qRT-PCR) data are presented as inantly reflects expression in non-b islet cells, although we fold expression relative to the control sample and calculated cannot exclude a potential minor contribution from a few 112 Asna1 Ensures Retrograde Transport in b-Cells Diabetes Volume 65, January 2016 b-cells that had not yet fully deleted both Asna1 alleles at b2 2 this stage. Nonfasted glucose levels of Asna1 / mice were already mildly increased from ;2 weeks of age, and b2 2 Asna1 / mice progressed to overt diabetes between 6 and10weeksofage(Fig.1A).Theincreaseinnonfasting hyperglycemia and development of diabetes was more b2 2 pronounced in Asna1 / males (Fig. 1A)thanin b2 2 Asna1 / females (Supplementary Fig. 3). These findings show a requirement for Asna1 in the maintenance of b-cell function and glucose homeostasis. Subsequent experiments were performed on 3- to 4-week-old mice (i.e., before the onset of overt diabetes) using a mix of males and females. To assess how Asna1 ensures b-cell function, we analyzed glucose tolerance and GSIS in response to exogenous, intraperitoneal administration of glucose (i.e., b2 2 GTT) in Asna1 / mice. The targeted insertion of Cre into one of the Ins1 alleles, Ins1+/Cre; global inactivation of fl 2 one Asna1 flox allele, Asna1 ox/ ;orb-cell conditional in- b 2 activation of one Asna1 flox allele, Asna1 +/ , affected neither glucose tolerance nor GSIS (Supplementary Fig. b 2 fl fl fl 4A–F). Thus, Asna1 +/ , Asna1+/ ox, Asna1 ox/ ox,and b 2 b Asna1 +/ mice, collectively denoted Asna1 ctrl,were b 2 all used as controls. In contrast to Asna1 +/ litter- b2 2 mates, 3-week-old Asna1 / mice exhibited impaired glucose tolerance (Fig. 1B) as well as reduced insulin secretion in response to both glucose (Fig. 1C) and the insulin secretagogue arginine (Fig. 1D). Of note, glucose tolerance and insulin secretion were affected in both males and females at this stage (Supplementary Fig. 5A–F). Together, these data show that loss of Asna1 in b-cells leads to diabetes due to impaired b-cell function.

Insulin Biogenesis Is Perturbed in Asna1b2/2 b-Cells To elucidate the mechanism underlying b-cell failure in b2 2 Asna1 / mice, we next determined pancreatic islet area and b-cell number. Pancreatic islet area and b-cell number were both normal (Fig. 2A and B), whereas pancreatic b2 2 insulin content was reduced by ;60% in Asna1 / b2 2 mice (Fig. 2C), suggesting that Asna1 / mice develop diabetes as a consequence of reduced amounts of insulin. In mice, insulin is encoded by two highly homologous genes Ins1 and Ins2, and as expected, Ins1 expression was nearly significantly (P = 0.055, n = 4) reduced by ;50% in islets of mice carrying the Ins1Cre knockin allele (Supplementary Fig. 6A). However, because of the low expression of Ins1 relative to Ins2, the reduced Ins1 expression did not in itself affect total insulin (Ins1 + Ins2) expression or insulin protein content in Ins1+/Cre mice Figure 1—Asna1b2/2 mice develop diabetes. A: Nonfasted glucose (Supplementary Fig. 6A and B). Islet proinsulin and in- bctrl b2/2 ; levels in Asna1 and Asna1 males (n =6–26). B and C: Blood sulin content, however, was reduced by 70% in B C fi B C b2/2 b+/2 glucose ( ) and plasma insulin ( ) pro les and AUC ( and ) Asna1 islets compared with Asna1 islets (Fig. during GTT of 3-week-old Asna1b+/2 (n = 13) and Asna1b2/2 (n = 2D) without a corresponding difference in total insulin 12) mice. D: Plasma insulin secretion profiles and AUC of 4-week- b+/2 b2/2 (Ins1 + Ins2) mRNA levels (Supplementary Fig. 7), thus old Asna1 (n = 12) and Asna1 (n = 14) mice during arginine- 6 P < providing evidence for a posttranscriptional reduction in stimulated insulin secretion. Data are mean SEM. * 0.05, b2 2 **P < 0.01, ***P < 0.001 (Student t test). ns, not significant. insulin biogenesis in b-cells of Asna1 / mice. In contrast, pancreatic content of the hormone islet amyloid polypeptide, which is stored and cosecreted with insulin, diabetes.diabetesjournals.org Norlin and Associates 113

b2 2 vivo, isolated Asna1 / islets secreted less insulin in response to both glucose and the membrane depolarizer KCl (Fig. 2F). Normalization of the amount of secreted insulin- to-total insulin content, to adjust for the reduced insulin b2 2 content in Asna1 / islets, however, revealed that the b2 2 relative insulin secretion potential of Asna1 / islets was similar to that of control islets (Fig. 2G). Of note, plasma levels of insulin and proinsulin showed an increased b2 2 proinsulin/insulin ratio in Asna1 / mice (Supplemen- tary Fig. 8C), suggesting that proinsulin secretion is rela- b2 2 tively greater than insulin secretion in Asna1 / mice. The expression of genes involved in glucose uptake and metabolism (i.e., Glut2 and Gck) was reduced by ;70% and ;30%, respectively (Supplementary Fig. 9A). The expression of both Glut2 and Gck however, has been observed to be negatively affected by hyperglycemia (27), leaving open the possibility that the reduced expression of these genes b2 2 in Asna1 / islets may be secondary to the modest hyperglycemia observed at the stage of islet isolation (i.e., 3–4 weeks of age). The expression of genes involved in membrane depolarization (Sur1 and Kir6.2)andin- sulin exocytosis (Rab3a, Rab3b, Rab27, Snap25, Syt7, Stx1a, Stx4a, Vamp2,andVamp3)wasessentiallynormal with the exception of Rab3b expression, which was reduced by ;40% (Supplementary Fig. 9B). Taken together, b2 2 these ﬁndings suggest that Asna1 / mice develop insulin insufﬁciency and diabetes largely as a consequence of impaired insulin biogenesis, although it is possible that the reduced expression of Glut2, Gck, and Rab3b may contrib- ute to the impaired GSIS observed in vivo.

Figure 2—Reduced pancreatic insulin content in Asna1b2/2 mice. Loss of Asna1 in b-Cells Leads to ER Stress A: Quantification of pancreatic islet cell area in 3-week-old On a systemic level, Get3/Asna1 has been associated with Asna1bctrl (n =4)andAsna1b2/2 (n = 4) mice. B:Quantification bctrl maintenance of ER homeostasis (10,12,28), thus consti- of b-cell fraction in islets of 3-week-old Asna1 (n =5)and Asna1b2/2 (n = 5) mice. C: Total pancreatic insulin content in 3- tuting a likely intersection point for Asna1 TA-targeting week-old Asna1bctrl (n = 10) and Asna1b2/2 (n = 5) mice. D and E: activity and b-cell function. The expression and localiza- Proinsulin and insulin content (D) and proinsulin/insulin ratio (E)in tion of markers for the ER-Golgi intermediate compart- islets from 3–4-week-old Asna1+/2 and Asna1b2/2 mice (n =6–7). F +/2 ment (ERGIC53), cis Golgi (Gm130), trans Golgi network and G: Insulin secretion from islets isolated from Asna1 and Asna1b2/2 mice incubated at 2.8 and 16.8 mmol/L glucose and (TGN46), proinsulin vesicles, endosome (EEA1), and lyso- 30 mmol/L KCl (F) and insulin secretion normalized to islet insulin some (Lamp1) compartments all appeared normal in b2 2 content (G)(n = 3). Data are mean 6 SEM. *P < 0.05, **P < 0.01, Asna1 / islets (Fig. 3A), whereas the expression of P < t *** 0.001 (Student test). the ER stress-response chaperones BiP (Grp78) and Grp94, as judged by KDEL immunostaining, appeared more intense (Fig. 3A). Moreover, strong nuclear ATF4 was unaffected (Supplementary Fig. 7B). No significant expression was observed (Fig. 3B) and IRE1a activity, difference in islet hormone content was observed when monitored using ERAI reporter mice (23) on an b2 2 comparing males and females (Supplementary Fig. 7C). Asna1 / background, was enhanced (Fig. 3C). Addition- The pancreatic proinsulin/insulin ratio was normal in ally, transmission electron microscopy revealed dilated ER b2 2 b2 2 Asna1 / islets (Fig. 2E), and inhibition of protein deg- cisterna in a subset of Asna1 / b-cells at 5–6 weeks of radation using proteasomal and lysosomal inhibitors had age (Supplementary Fig. 9C). The expression of UPR similar effects on proinsulin and insulin content in genes, including BiP (Hspa5, grp78), Grp94 (Hsp90b1), b2 2 b 2 Asna1 / and Asna1 +/ islets (Supplementary Fig. 8A DnaJc3, Ero1lb, Erp29, Pdia4 (Erp72), Edem2, HRD, and B). These results argue against an increased rate of Herpud1, Sel1l, Atf3, Atf4, Chop10 (Ddit3), and Trib3,were, b2 2 (pro)insulin degradation as the underlying mechanism for with the exception of Ero1lb, all increased in Asna1 / b2 2 the reduced insulin content in Asna1 / b-cells. In agree- islets (Fig. 3D). Together, these findings, particularly the ment with the perturbed glucose tolerance and impaired activation of all three branches of the UPR, suggest that glucose- and arginine-induced insulin secretion observed in loss of Asna1 in b-cells leads to ER stress. 114 Asna1 Ensures Retrograde Transport in b-Cells Diabetes Volume 65, January 2016

Figure 3—Loss of Asna1 in b-cells results in ER stress. A and B: Immunostaining of pancreatic sections from Asna1b+/2 and Asna1b2/2 mice using KDEL, Gm130 (green), ERGIC53 (red), TGN46 (red), proinsulin (green), EEA1 (green), and Lamp1 (red) antibodies (A) and ATF4 antibodies (B). DAPI (blue) indicates nuclei. C: Activation of the Ire1a UPR pathway as monitored by ERAI-GFP reporter activity on an Asna1b+/2 and Asna1b2/2 background. D: qRT-PCR expression analyses of the indicated UPR genes in islets isolated from 3-week-old Asna1bctrl and Asna1b2/2 mice (n =6–8). Scale bar = 10 mm(A) and 25 mm(B and C). Data are mean 6 SEM. *P < 0.05, **P < 0.01, ***P < 0.001 (Student t test). ERAD, ER-associated degradation; GFP, green ﬂuorescent protein; Proins, proinsulin vesicles.

b2 2 The normal pancreatic proinsulin/insulin ratio (Fig. transport in Asna1 / islets, we first made use of BFA, 2E) together with the normal localization of proinsulin to which 1) inhibits COPI-dependent traffic by blocking the vesicular structures adjacent to the trans Golgi network assembly of the COPI vesicle coat and 2) collapses the b2 2 (TGN) in Asna1 / b- cells (Fig. 3A) suggest that proin- Golgi into the ER through the alternative Rab6-dependent sulin folding, transport, and processing are largely unaf- route (32). Similar to HeLa cells with impaired Rab6 ac- b2 2 b2 2 fected in Asna1 / b-cells. Moreover, treatment of islets tivity (32,33), Asna1 / islets showed a delayed and in- with the chemical chaperones 4-PBA and TUDCA, which are complete collapse of the Golgi compartment when known to improve ER folding capacity (29,30), failed to exposed to BFA (Fig. 4A and Supplementary Fig. 11A normalize islets proinsulin and insulin content (Supplemen- and B). These data suggest that the Rab6-dependent ret- tary Fig. 10A and B) and to alleviate ER stress (Supplemen- rograde transport route between the Golgi and ER b2 2 b2 2 tary Fig. 10C and D) in Asna1 / islets. Taken together, is impaired in Asna1 / b-cells. In contrast, rebuilding these results argue against impaired folding of proinsulin as of the Golgi after BFA washout appeared largely unal- b2 2 the underlying cause for ER stress in Asna1 / islets. tered, suggesting that anterograde transport was normal b2 2 in Asna1 / islets (Fig. 4B). Loss of Asna1 Leads to Impaired Retrograde We next assayed plasma membrane (PM)-to-TGN retro- Transport in b-Cells grade transport by monitoring the uptake and transport of Get3/Asna1 has been suggested to genetically interact fluorescently labeled Ricin toxin (594-RiTx), which is trans- with the retromer complex, the COG complex, and the ported through endosomes, TGN, and cis Golgi into the ER COPI coatomer complex, implying a potential role for afterendocytosis(34).Indispersedb-cells from control Get3/Asna1 in retrograde transport (11). Retrograde islets, 594-RiTx accumulated in the TGN after 90 min transport between the Golgi and the ER are mediated (Fig. 4C). In contrast, accumulation of 594-RiTx in the b2 2 by both COPI coatomer-coated vesicles and Rab6- TGN was significantly reduced by 20% in Asna1 / b-cells dependent membrane tubules (31). To assess retrograde (Fig. 4C and D), providing evidence that PM-to-TGN diabetes.diabetesjournals.org Norlin and Associates 115

we reasoned that Asna1 might regulate retrograde transport by ensuring proper localization of Stx5, Stx6, and/or Vamp3 to the cis Golgi, TGN, and endosomes, respectively. Consistent with this notion, Stx5, Stx6, and b2 2 Vamp3 proteins were barely detectable in Asna1 / b-cells, whereas the localization of the Golgi TA protein giantin was unaltered (Fig. 5A). The reduced Golgi localization of Stx5 and Stx6 was already observed in b2 2 ;1-week-old Asna1 / mice (Supplementary Fig. 12), whereas the loss of Vamp3 was observed ﬁrst after 3 weeks of age (Fig. 5A), suggesting that the loss of Vamp3 may be secondary to impaired retrograde transport and/or deterioration of b-cell function. The protein levels of the major 35-kDa Golgi Stx5 b2 2 isoform and Stx6 were largely unaltered in Asna1 / islets (Fig. 5B), although there was a tendency for the minor 42-kDa Stx5 ER isoform (37) to be slightly reduced. Together, these results show that Stx5 and Stx6 are not degraded but likely become mislocalized or redistributed b2 2 from their respective Golgi compartment in Asna1 / b-cells. Stx5 was successfully coimmunoprecipitated when coexpressed with Myc-tagged Asna1 in insulinoma cell lines (Fig. 5C), leaving open the possibility that analogous to Get3/Asna1-Sed5 interaction (28), Asna1 mediates membrane insertion of Stx5 in b-cells. The localization of other proposed Asna1 targets, such as the ER resident proteins Sec61b and Sec22b (Fig. 5D) as well as the post-ER TA-SNAREs Vamp2 and Stx1a, which are implicated in insulin vesicle exocytosis (Fig. 5E), revealed a normal localization and expression, arguing against a critical role for Asna1 in the biogenesis of these TA proteins in b-cells. Taken together, these results provide evidence of a requirement for Asna1 function in ensuring Golgi localization of Stx5 and Stx6 in b-cells.

Inhibition of Retrograde Transport in b-Cells Results in Reduced Insulin Biogenesis and ER Stress Figure 4—Retrograde transport is impaired in Asna1b2/2 islets. A To test whether the activation of the UPR and/or reduced b2 2 and B: Time course of retrograde collapse of Golgi into the ER insulin content observed in Asna1 / b-cells might be following exposure to BFA (A) and rebuilding of cis Golgi following b+/2 secondary to impaired retrograde transport, we used the BFA washout (B) monitored by Gm130 staining (green) in Asna1 and Asna1b2/2 islets (n = 3). DAPI (blue) indicates nuclei. C and D: small molecule inhibitor Retro-2 that inhibits retrograde Colocalization analyses (C) and quantiﬁcation (D) of endocytosed transport from early endosomes (EEs) to the TGN (38). 2 2 594-Ricin toxin (red) and the TGN marker TGN46 (green) in dis- Of note, analogous to the phenotype of Asna1 / b-cells, Asna1bctrl Asna1b2/2 persed and islets following a 90-min chase exposure of isolated wild-type islets to Retro-2 resulted in (n =4–5). Arrowheads indicate Golgi remnants in A. Scale bar = 10 mm(A and B) and 5 mm(C). Data are mean 6 SEM. **P < 0.01 not only reduced Golgi localization of Stx5 but also, albeit (Student t test). less pronounced, reduced Stx6 Golgi localization (Fig. 6A). 2 2 However, as observed in Asna1 / b-cells, the levels of the 35-kDa isoform of Stx5 were largely unaltered, whereas the less abundant 42-kDa expression tended to be reduced b2/2 b retrograde transport is impaired in Asna1 -cells. These in Retro-2–treated islets (Supplementary Fig. 13A). Thus, results demonstrate that Asna1 is required for sustained similar to that previously observed in HeLa cells (38), ex- retrograde transport to both the TGN and the ER. posure of islets to Retro-2 results in a redistribution of Loss of Asna1 Leads to Mislocalization of Syntaxins Stx5 from the Golgi compartment. Moreover, and like 2 2 Involved in Retrograde Transport Asna1 / b-cells, COPI-independent Golgi-to-ER retro- Vesicle transport and membrane recycling among the PM, grade transport was impaired in Retro-2–treated islets endosome, and TGN critically depends on TA-SNAREs, (Supplementary Fig. 13C–E). Because ex vivo cultivation such as syntaxin (Stx) 5, Stx6, and Vamp3 (35,36). Thus, of isolated islets in itself provokes ER stress, the functional 116 Asna1 Ensures Retrograde Transport in b-Cells Diabetes Volume 65, January 2016

Figure 5—Stx5 and Stx6 are mislocalized in Asna1b2/2 islets. A: Immunostaining of pancreatic sections from Asna1b+/2 and Asna1b2/2 mice using antibodies against Stx5 (red), giantin (red), Gm130 (green), Stx6 (green), TGN46 (red), insulin (green), and Vamp3 (red). DAPI (blue) indicates nuclei (n = 3). Insets show individual color channels of selected regions (boxes). *Non-b-cells. B: Representative immunoblots and quantiﬁcation of Stx5 (35- and 42-kDa isoforms) and Stx6 protein levels in islets isolated from 4–5-week-old Asna1bctrl and Asna1b2/2 mice (n = 3). C: Representative immunoblots showing anti-Myc immunoprecipitation of Myc-tagged Asna1 (top panel: Asna1 [39 kDa], Myc-Asna1 [40 kDa] and coprecipitation of Stx5; bottom panel: 35- and 42-kDa Stx5 isoforms) from MIN6 cells transfected with Myc- tagged Asna1 or LacZ and Stx5 constructs as indicated (n = 3). Dashed lines indicate ﬁlter cuts. D and E: Immunostaining of pancreatic sections from Asna1b+/2 and Asna1b2/2 mice with Sec61b (red), Sec22b (red) (D), and Gm130 (green) and Stx1a (red), insulin (red), and Vamp2 (green) (E) antibodies. Scale bar = 10 mm. au, arbitrary unit; IB, immunoblot; IP, immunoprecipitation.

effects of Retro-2 on UPR were assessed on day 2 after iso- cells, insulin expression was not significantly reduced (Fig. lation (i.e., at a stage when the acute activation of ER 6H). Taken together, these data show that inhibition of stress genes is somewhat dampened) (Supplementary retrograde transport in b-cells results in ER stress and Fig. 14). Retro-2–exposed islets displayed reduced insulin impaired insulin biogenesis. Additionally, these findings content by ;30% and enhanced expression of UPR genes, provide evidence that the reduced insulin content and ER indicating ER stress (Fig. 6B and C). Additionally, insulin stress observed in Asna1-deficient b-cells are a consequence expression was significantly reduced in Retro-2–exposed of impaired PM/EE-to-TGN retrograde transport. islets (Fig. 6D), thus likely contributing to the observed reduction in insulin content in these islets. The subcellu- DISCUSSION lar location of Stx5 was unaffected by the ER stress/UPR This study shows a critical role for Asna1 in ensuring activators tunicamycin and thapsigargin (Supplementary b-cell function. Loss of Asna1 in b-cells of mice results Fig. 15), providing evidence that the mislocalization of in pancreatic hypoinsulinemia, impaired insulin secretion, Stx5 is not secondary to ER stress. and early onset diabetes. Additionally, b-cells of b2 2 To independently assess the effects of Retro-2 on Asna1 / mice showed impaired PM-to-TGN as well insulin biogenesis and UPR activation in b-cells, obviating as Golgi-to-ER retrograde transport, ER stress, and mis- potential confounding effects of ER stress induced by iso- localization of Stx5 and Stx6. Of note, we also show lation and ex vivo cultivation of islets, we next exposed that inhibition of retrograde transport at the level of MIN6 cells to Retro-2. Exposure of MIN6 cells to Retro-2 EE-to-TGN in isolated islet and insulinoma cells results also resulted in reduced Golgi localization, but not expres- in impaired Golgi-to-ER retrograde transport, decreased sion, of Stx5 (Fig. 6E and Supplementary Fig. 13B), insulin content, and ER stress. Thus, the findings provide reduced insulin content, and enhanced expression of evidence that Asna1 is required in b-cells to ensure retro- UPR genes (Fig. 6F and G). However, although insulin grade transport, which in turn appears to be essential for ER content was reduced by ;30% in Retro-2–treated MIN6 homeostasis and proinsulin biogenesis. Additionally, the diabetes.diabetesjournals.org Norlin and Associates 117

Figure 6—Pharmacological inhibition of EE-to-TGN retrograde transport in islets and insulinoma cells mimics Asna1b2/2 phenotypes. A: Immunostaining of islet cells incubated with vehicle (DMSO) and Retro-2 (50 mmol/L) for 48 h using antibodies Stx5 (red), Gm130 (green), Stx6 (green), and TGN46 (red) (n = 4). Insets show individual color channels of selected regions (boxes). B: qRT-PCR analyses of UPR gene expression in islets incubated with vehicle (DMSO) and Retro-2 (50 mmol/L) for 48 h (n = 4). C and D: Insulin protein content (C) and qRT- PCR analysis of insulin expression (D) of islets incubated with vehicle (DMSO) and Retro-2 (50 mmol/L) for 48 h (n = 5). E: Immunocyto- chemical staining of MIN6 cells incubated with vehicle (DMSO) and Retro-2 (80 mmol/L) for 24 h using antibodies against Stx5 (red) and Gm130 (green) (n = 4). F: qRT-PCR analysis of UPR gene expression in MIN6 cells incubated with vehicle (DMSO) and Retro-2 (80 mmol/L) for 24 h (n = 4). G and H: Insulin protein content (G) and qRT-PCR of insulin expression (H) of MIN6 cells incubated with vehicle (DMSO) and Retro-2 (80 mmol/L) for 24 h (n = 4). DAPI (blue) indicates nuclei (A and E). Scale bar = 10 mm(A) and 50 mm(E). Data are mean 6 SEM. *P < 0.05, **P < 0.01, ***P < 0.001 (Student t test).

perturbed Golgi-to-ER retrograde transport in Retro-2– insufficient insulin folding capacity. Of note, we observed treated primary islets suggests that the impairment of this an increased plasma proinsulin/insulin ratio, suggesting b2 2 step in Asna1 / b-cells likely is secondary to the inhibi- that secretion of newly synthesized proinsulin is in- tion of retrograde transport at the level of EE-to-TGN. creased and raising the possibility that Asna1-dependent b2 2 The primary cause of diabetes in Asna1 / mice ap- EE-to-TGN retrograde transport normally counteracts pears to be insufficient production of insulin. The current leakage or premature secretion of proinsulin through data do not support a role for Asna1 in the posttransla- the endosomes and the constitutive-like pathway (40). tional targeting of insulin itself, such as has been de- Instead, we favor the idea that impaired retrograde b2 2 scribed for other short secretory proteins (39), because transport in Asna1 / mice provokes reduced insulin we did not observe an increase or alteration in pancreatic biogenesis as a negative consequence of UPR activation proinsulin levels. Although the activation of the UPR in- onproteintranslation(41).Additionally,therewasa dicates that ER homeostasis is perturbed in the absence of tendency, albeit nonsignificant, for insulin expression b2 2 Asna1 function, we found no evidence of impaired pro- levels to be reduced in Asna1 / mice, which would insulin folding or anterograde transport through the secre- also be consistent with a negative effect of UPR activa- tory pathway. Moreover, vesicular proinsulin localization tion (e.g., through IRE1a/XBP1andTrib3induction,on b2 2 appears unaltered in Asna1 / b-cells, pancreatic islet insulin transcription [42,43], or as a consequence of amyloid polypeptide content is unaffected, and insulin IRE1a/XBP1–mediated degradation of insulin mRNA maturation as well as exocytosis seem largely unaffected. [44,45]). Moreover, Retro-2–mediated impairment of Thus, activation of the UPR is likely not provoked by retrograde transport resulted in robust activation of 118 Asna1 Ensures Retrograde Transport in b-Cells Diabetes Volume 65, January 2016

UPR genes and decreased insulin protein content in both context of retrograde transport and ER homeostasis. isolated islets and MIN6 cells and were accompanied, First, the effect of Retro-2 on Stx5 localization is acute albeit to a different extent, by suppression of insulin and complete, whereas the effect on Stx6 localization is transcription. Thus, although we cannot exclude that slower and less severe (Fig. 6A) (38), suggesting that Stx6 Retro-2 has additional UPR-independent effects on in- mislocalization may be secondary to the loss of Stx5 from sulin biogenesis, the reduced insulin content observed in the Golgi. Second, in yeast Get3/Asna1 mutants, Sed5 is Retro-2–treated islets and MIN6 cells appears to be sec- mislocalized, and the Bip/grp78 ER chaperone homolog ondary to UPR-mediated impairment of insulin biogen- Kar2 is abnormally secreted (10,28), which may reflect an esis at both the transcriptional and the translational impaired retention or retrograde transport of Kar2 (10). level. These findings also provide strong evidence that Importantly, overexpression of the Stx5 homolog Sed5 in impairment of retrograde transport in b-cells by Retro-2 Get3/Asna1 mutants rescues the Kar2 (Bip) secretion phe- in vitro and ex vivo and due to loss of Asna1 function in notype (28). Taken together, these data are consistent vivoleadstoactivationoftheUPRthatinturnnega- with a chain of events where inactivation of Asna1/ tively affects insulin biogenesis, which, in vivo, results in Get3, likely through mislocalization of Stx5/Sed5, per- insulin insufficiency and the development of diabetes turbs retrograde transport and, thereby, ER homeosta- (Supplementary Fig. 16). sis. In mammalian b-cells, impairment of retrograde Previous studies, primarily in cell-free systems, have transport appears to have the additional consequence outlined a role for Get3/Asna1/Asna1 in the targeting of of attenuating insulin biogenesis, thus leading to the TA proteins to the ER membrane receptor Get1/Get2 development of diabetes. (CAML/WRB in mammals), thus facilitating their in- Recently, an additional function distinct from its TA sertion into the ER membrane (15,16). The current data protein–targeting activity was described for Get3/Asna1 indicate that at least in b-cells, Golgi localization of the (18). Under oxidative stress conditions and independent TA-SNARE proteins Stx5 and Stx6 depend on Asna1 func- of ATP, Get3/Asna1 was shown to function as a molecular tion. Total levels of Stx5 and Stx6 mRNA and protein chaperone that binds unfolded proteins to prevent their b2 2 were unaltered in Asna1 / b-cells, suggesting that irreversible aggregation. The proposed dual role for Get3/ Stx5 and Stx6 become redistributed and/or mislocalized Asna1 in yeast is intriguing and implies a potential role for in the absence of Asna1 activity. In contrast, the localiza- maintenance of both ER and redox homeostasis. Future tion of the TA proteins Sec61b, Sec22b, giantin, Stx1, and studies are required to separate potential oxidative stress– Vamp2 was unaffected, suggesting that other chaperone induced chaperone activity of Asna1 from the TA protein– systems, such as the Hsc70/Hsp40 pair (17), compensate targeting function to fully understand how Asna1 ensures for the targeting of these TA proteins in b-cells lacking retrograde transport as well as insulin and ER homeostasis. Asna1 function. In agreement with such a notion, the In conclusion, we show that Asna1 is critical for b-cell yeast TA proteins Sbh1 and Sbh2 (i.e., the homologs of function. The results provide evidence that Asna1 plays a Sec61b) and Scs2 and Ysy6 all retain a certain level of ER role in ensuring retrograde transport along a PM-to-TGN localization in yeast GET1/GET2 mutants (28). Consistent and a Golgi-to-ER route. Impairment of these functions in with a potential direct role for Asna1 in the targeting of b-cells leads to ER stress, insulin insufficiency, and devel- Stx5 to the ER of b-cells, and like Get3/Asna1-Sed5 in- opment of diabetes. The findings suggest that maintenance teractions, we found that Asna1 physically interacts with and/or restoration of retrograde transport in b-cells may Stx5 in insulinoma cells upon coexpression of these pro- be therapeutically relevant for T2D. The study was per- teins. We cannot, however, exclude the possibility that formed, however, using genetically modified mice and iso- Stx5 and Stx6 are appropriately targeted in the absence lated mouse islets. Thus, although the data provide strong of Asna1 activity but become mislocalized or redistributed evidence for a role for Asna1 and retrograde transport in to other cellular compartments as a consequence of im- ensuring mouse b-cell function, a potential similar role for paired retrograde transport. ASNA1 and/or retrograde transport for b-cell function in Exposure of MIN6 insulinoma cells and isolated islets humans will require additional analyses. to the small molecule inhibitor Retro-2 closely mimics b2 2 key phenotypes of Asna1 / b-cells, including 1)reduced insulin content, 2) activated UPR, and 3)redis- Acknowledgments. The authors thank Elisabet Pålsson, Jurate tributed Stx5 and Stx6. Although the molecular targets Straseviciene, Fredrik Backlund, and Lisa Lundberg (Umeå Centre for Mo- ’ of Retro-2 are unknown and at what point the Retro-2 lecular Medicine) for technical assistance and members of the authors and Asna1 pathways intersect is unclear, the reduction of laboratory for technical instructions, suggestions, and helpful discussions. Funding. These studies were facilitated by support from the Strategic Stx5 in the cis Golgi and Stx6 in the TGN is likely to affect Research Program in Diabetes at Umeå University and supported by grants from Asna1-dependent retrograde transport. Small interfering the Swedish Research Council (521-2013-3215) and the Knut and Alice Wallen- – RNA mediated knockdown of Stx5 and inhibition of Stx6 berg Foundation (KAW 2010.0033). using blocking antibodies have both been shown to neg- Duality of Interest. H.E. is a cofounder, shareholder, and consultant of the atively affect EE-to-TGN recycling (35,36). However, two unlisted biotech company Betagenon AB. No other potential conflicts of interest observations argue for a primary role for Stx5 in the relevant to this article were reported. diabetes.diabetesjournals.org Norlin and Associates 119

Author Contributions. S.N. and V.S.P. contributed to the design and 22. Liu P, Jenkins NA, Copeland NG. A highly efficient recombineering-based method performance of experiments, data interpretation, discussion, and writing and for generating conditional knockout mutations. Genome Res 2003;13:476–484 editing of the manuscript. P.N. initiated the study, provided advice, and contrib- 23. Iwawaki T, Akai R, Kohno K, Miura M. A transgenic mouse model for uted to the discussion and review of the manuscript. H.E. initiated the study, monitoring endoplasmic reticulum stress. Nat Med 2004;10:98–102 designed and supervised the study, analyzed and interpreted the data, contrib- 24. Steneberg P, Rubins N, Bartoov-Shifman R, Walker MD, Edlund H. The FFA uted to the discussion, and wrote the manuscript. H.E is the guarantor of this receptor GPR40 links hyperinsulinemia, hepatic steatosis, and impaired glucose work and, as such, had full access to all the data in the study and takes homeostasis in mouse. Cell Metab 2005;1:245–258 responsibility for the integrity of the data and the accuracy of the data analysis. 25. Kone M, Pullen TJ, Sun G, et al. LKB1 and AMPK differentially regulate pancreatic b-cell identity. FASEB J 2014;28:4972–4985 References 26. Thorens B, Tarussio D, Maestro MA, Rovira M, Heikkilä E, Ferrer J. Ins1(Cre) 1. Chan JY, Luzuriaga J, Bensellam M, Biden TJ, Laybutt DR. Failure of the knock-in mice for beta cell-specific gene recombination. Diabetologia 2015;58:558–565 adaptive unfolded protein response in islets of obese mice is linked with ab- 27. Jonas JC, Sharma A, Hasenkamp W, et al. Chronic hyperglycemia triggers normalities in b-cell gene expression and progression to diabetes. Diabetes loss of pancreatic beta cell differentiation in an animal model of diabetes. J Biol 2013;62:1557–1568 Chem 1999;274:14112–14121 2. Eizirik DL, Cardozo AK, Cnop M. The role for endoplasmic reticulum stress 28. Schuldiner M, Metz J, Schmid V, et al. The GET complex mediates insertion in diabetes mellitus. Endocr Rev 2008;29:42–61 of tail-anchored proteins into the ER membrane. Cell 2008;134:634–645 3. Volchuk A, Ron D. The endoplasmic reticulum stress response in the 29. Tang C, Koulajian K, Schuiki I, et al. Glucose-induced beta cell dysfunction pancreatic beta-cell. Diabetes Obes Metab 2010;12(Suppl. 2):48–57 in vivo in rats: link between oxidative stress and endoplasmic reticulum stress. 4. Auld KL, Hitchcock AL, Doherty HK, Frietze S, Huang LS, Silver PA. The Diabetologia 2012;55:1366–1379 conserved ATPase Get3/Arr4 modulates the activity of membrane-associated 30. Özcan U, Yilmaz E, Özcan L, et al. Chemical chaperones reduce ER stress proteins in Saccharomyces cerevisiae. Genetics 2006;174:215–227 and restore glucose homeostasis in a mouse model of type 2 diabetes. Science 5. Zewail A, Xie MW, Xing Y, et al. Novel functions of the phosphatidylinositol 2006;313:1137–1140 metabolic pathway discovered by a chemical genomics screen with wortmannin. 31. Girod A, Storrie B, Simpson JC, et al. Evidence for a COP-I-independent Proc Natl Acad Sci U S A 2003;100:3345–3350 transport route from the Golgi complex to the endoplasmic reticulum. Nat Cell Biol 6. Shen J, Hsu CM, Kang BK, Rosen BP, Bhattacharjee H. The Saccharomyces 1999;1:423–430 cerevisiae Arr4p is involved in metal and heat tolerance. Biometals 2003;16:369–378 32. Jiang S, Storrie B. Cisternal rab proteins regulate Golgi apparatus re- 7. Sambade M, Alba M, Smardon AM, West RW, Kane PM. A genomic screen distribution in response to hypotonic stress. Mol Biol Cell 2005;16:2586–2596 for yeast vacuolar membrane ATPase mutants. Genetics 2005;170:1539–1551 33. Sun Y, Shestakova A, Hunt L, Sehgal S, Lupashin V, Storrie B. Rab6 reg- 8. Enyenihi AH, Saunders WS. Large-scale functional genomic analysis of ulates both ZW10/RINT-1 and conserved oligomeric Golgi complex-dependent sporulation and meiosis in Saccharomyces cerevisiae. Genetics 2003;163:47–54 Golgi trafficking and homeostasis. Mol Biol Cell 2007;18:4129–4142 9. Dimmer KS, Fritz S, Fuchs F, et al. Genetic basis of mitochondrial function 34. Sandvig K, van Deurs B. Transport of protein toxins into cells: pathways and morphology in Saccharomyces cerevisiae. Mol Biol Cell 2002;13:847–853 used by ricin, cholera toxin and Shiga toxin. FEBS Lett 2002;529:49–53 10. Schuldiner M, Collins SR, Thompson NJ, et al. Exploration of the function 35. Amessou M, Fradagrada A, Falguières T, et al. Syntaxin 16 and syntaxin 5 and organization of the yeast early secretory pathway through an epistatic are required for efficient retrograde transport of several exogenous and endog- miniarray profile. Cell 2005;123:507–519 enous cargo proteins. J Cell Sci 2007;120:1457–1468 11. Costanzo M, Baryshnikova A, Bellay J, et al. The genetic landscape of a cell. 36. Mallard F, Tang BL, Galli T, et al. Early/recycling endosomes-to-TGN Science 2010;327:425–431 transport involves two SNARE complexes and a Rab6 isoform. J Cell Biol 2002; 12. Jonikas MC, Collins SR, Denic V, et al. Comprehensive characterization of 156:653–664 genes required for protein folding in the endoplasmic reticulum. Science 2009; 37. Hui N, Nakamura N, Sönnichsen B, Shima DT, Nilsson T, Warren G. An 323:1693–1697 isoform of the Golgi t-SNARE, syntaxin 5, with an endoplasmic reticulum retrieval 13. Stefanovic S, Hegde RS. Identification of a targeting factor for post- signal. Mol Biol Cell 1997;8:1777–1787 translational membrane protein insertion into the ER. Cell 2007;128:1147–1159 38. Stechmann B, Bai SK, Gobbo E, et al. Inhibition of retrograde transport 14. Wang F, Brown EC, Mak G, Zhuang J, Denic V. A chaperone cascade sorts protects mice from lethal ricin challenge. Cell 2010;141:231–242 proteins for posttranslational membrane insertion into the endoplasmic reticulum. 39. Johnson N, Vilardi F, Lang S, Leznicki P, Zimmermann R, High S. TRC40 Mol Cell 2010;40:159–171 can deliver short secretory proteins to the Sec61 translocon. J Cell Sci 2012;125: 15. Shao S, Hegde RS. Membrane protein insertion at the endoplasmic re- 3612–3620 ticulum. Annu Rev Cell Dev Biol 2011;27:25–56 40. Arvan P, Halban PA. Sorting ourselves out: seeking consensus on trafficking 16. Yamamoto Y, Sakisaka T. Molecular machinery for insertion of tail-anchored in the beta-cell. Traffic 2004;5:53–61 membrane proteins into the endoplasmic reticulum membrane in mammalian 41. Kim MK, Kim HS, Lee IK, Park KG. Endoplasmic reticulum stress and insulin cells. Mol Cell 2012;48:387–397 biosynthesis: a review. Exp Diabetes Res 2012;2012:509437 17. Borgese N, Fasana E. Targeting pathways of C-tail-anchored proteins. Bi- 42. Jang YY, Kim NK, Kim MK, et al. The effect of tribbles-related protein 3 on ochim Biophys Acta 2010;1808:937–946 ER stress-suppressed insulin gene expression in INS-1 cells. Korean Diabetes J 18. Voth W, Schick M, Gates S, et al. The protein targeting factor Get3 functions as 2010;34:312–319 ATP-independent chaperone under oxidative conditions. Mol Cell 2014;56:116–127 43. Lipson KL, Fonseca SG, Ishigaki S, et al. Regulation of insulin biosynthesis 19. Hemmingsson O, Kao G, Still M, Naredi P. ASNA-1 activity modulates in pancreatic beta cells by an endoplasmic reticulum-resident protein kinase sensitivity to cisplatin. Cancer Res 2010;70:10321–10328 IRE1. Cell Metab 2006;4:245–254 20. Kao G, Nordenson C, Still M, Rönnlund A, Tuck S, Naredi P. ASNA-1 pos- 44. Han D, Lerner AG, Vande Walle L, et al. IRE1a kinase activation modes itively regulates insulin secretion in C. elegans and mammalian cells. Cell 2007; control alternate endoribonuclease outputs to determine divergent cell fates. Cell 128:577–587 2009;138:562–575 21. Mukhopadhyay R, Ho YS, Swiatek PJ, Rosen BP, Bhattacharjee H. Targeted 45. Pirot P, Naamane N, Libert F, et al. Global profiling of genes modified by disruption of the mouse Asna1 gene results in embryonic lethality. FEBS Lett endoplasmic reticulum stress in pancreatic beta cells reveals the early degra- 2006;580:3889–3894 dation of insulin mRNAs. Diabetologia 2007;50:1006–1014