Remember me
Research ArticleEndocrinologyMetabolism
Open Access | 
10.1172/JCI179845
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Li, X. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Hu, J. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Huang, Y. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Zhang, H. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Xu, N. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Liu, Y. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Liu, X. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Ye, Y. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Zhang, X. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Xu, X. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Fan, Y. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Zhang, Z. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by
Zhang, W.
in:
PubMed
|
Google Scholar
|
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Wang, S. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by Feng, W. in: PubMed | Google Scholar
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by
Arvan, P.
in:
PubMed
|
Google Scholar
|
1Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, Tianjin, China.
2National Key Laboratory of Immunity and Inflammation, Department of Pathophysiology, Naval Medical University, Shanghai, China.
3Human Islet Resource Center, Tianjin First Central Hospital, Tianjin, China.
4Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan, USA.
Address correspondence to: Ming Liu, Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin, 300052, China. Email: mingliu@tmu.edu.cn. Or to: Peter Arvan or Wenli Feng, Division of Metabolism, Endocrinology & Diabetes, University of Michigan Medical Center, Ann Arbor, Michigan 48109, USA. Email: parvan@umich.edu (PA). Email: fengwenlil@126.com (WF).
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Find articles by
Liu, M.
in:
PubMed
|
Google Scholar
|
Authorship note: X Li, JH, YH, and HZ contributed equally to this work.
Published May 20, 2025 - More info
Published in Volume 135, Issue 14 on July 15, 2025
AbstractDefects in the early events of insulin biosynthesis, including inefficient preproinsulin (PPI) translocation across the membrane of the ER and proinsulin (PI) misfolding in the ER, can cause diabetes. Cellular machineries involved in these events remain poorly defined. Genes encoding translocon-associated protein α (TRAPα) show linkage to glycemic control in humans, though their pathophysiological role remains unknown. Here, we found that β cell–specific TRAPα-KO mice fed a chow diet or a high-fat diet (HFD) had decreased levels of circulating insulin, with age- and diet-related glucose intolerance. Multiple independent approaches revealed that TRAPα-KO not only causes inefficient PPI translocation but also leads to PI misfolding and ER stress, selectively limiting PI ER export and β cell compensatory potential. Importantly, decreased TRAPα expression was evident in islets of wild-type mice fed the HFD and in patients with type 2 diabetes (T2D). Furthermore, TRAPα expression was positively correlated with insulin content in human islet β cells, and decreased TRAPα was associated with PI maturation defects in T2D islets. Together, these data demonstrate that TRAPα deficiency in pancreatic β cells impairs PPI translocation, PI folding, insulin production, and glucose homeostasis, contributing to its genetic linkage to T2D.
IntroductionInsulin (INS) is a master hormone regulating energy metabolism. In pancreatic cells, INS derives from its precursors, preproinsulin (PPI) and proinsulin (PI). Newly synthesized PPI, led by its N-terminal signal peptide, undergoes co- and posttranslational translocation across the membrane of the ER via the Sec61 translocon (1, 2). Upon entering the lumen of the ER, the PPI signal peptide is excised by signal peptidase, forming PI that rapidly undergoes oxidative folding to form 3 intramolecular disulfide bonds that are critical for achieving PI native structure, leading to its anterograde transport from the ER to the Golgi complex (3–6). Over the past 2 decades, increasing genetic and biological evidence has indicated that multiple factors can affect these early events of INS biosynthesis, and defects in these events can lead to cell failure and diabetes both in humans and animal models (7–11).
Insights into these processes have emerged from studies of monogenic diabetes caused by INS gene mutations. To date, more than 90 INS gene mutations have been identified to cause diabetes in humans (7). Among those mutations that have been experimentally tested, more than 90% cause diabetes by impairing either PPI translocation into the ER (2, 12), or signal peptide cleavage (3, 13), or PI folding in the ER (4, 8, 14, 15). These impairments not only directly decrease PI/INS production, they may also lead to abnormal intracellular accumulation of INS precursors (including PPI and/or PI) that may affect coexpressed WT bystander PPI and/or PI and may induce ER stress and cell toxicity, all of which can contribute to the development and progression of diabetes (1, 10, 16). Importantly, WT PPI and PI are predisposed to inefficient translocation and disulfide mispairing during oxidative folding (2, 12, 17). Moreover, defective PI folding and maturation have been found in db/db mice as well as in patients with type 2 diabetes (T2D) (18–20), suggesting that defects in early events of INS biosynthesis may contribute to the development and progression of mutant INS gene–induced diabetes and T2D.
In addition to PPI and PI molecules themselves, previous studies showed that dysfunction of the ER folding environment due to deficiency of key ER machineries can impede oxidative folding of PI (21–23). Recently, we showed that the translocon-associated protein (TRAP; also called signal sequence receptor [SSR]) complex, composed of 4 subunits (named TRAP/SSR1, TRAP/SSR2, TRAP/SSR3, and TRAP/SSR4), is critical for INS biosynthesis in pancreatic cell lines and mouse islets (24–27). Importantly, GWAS from multiple populations show that polymorphisms in the locus of TRAPα/SSR1 genes are associated with glycemic control defects, including fasting glucose, gestational diabetes, and T2D (28–30). However, to date, no evidence directly links those risk variants with the expression or function of TRAPα/SSR1 in β cells. Therefore, further studies examining the connections between the variants and expression of TRAPα/SSR1 in β cells are needed to determine whether TRAPα is a (or the only) causal gene at this locus associated with T2D.
Furthermore, although TRAPα/SSR1 is important for INS biosynthesis in β cell lines (26), no direct evidence shows the pathophysiological role of TRAPα in vivo. Here, we examined the role of TRAP in INS biosynthesis and glycemic control in whole animals. For this purpose, we established a β cell–specific TRAP-KO mouse line and found that TRAP deficiency causes INS deficiency with age- and diet-related glucose intolerance. Functional analysis revealed that TRAPα-KO not only causes PPI translocation inefficiency but also impairs PI oxidative folding, which limits PI ER export and INS production and induces β cell ER stress and apoptosis. Interestingly, we found that TRAPα expression is positively correlated with INS content in human islet β cells, and decreased TRAPα is associated with PI maturation defects in T2D islets, as well as in mice fed an HFD. These data demonstrate that TRAPα is critical for PPI translocation, PI folding, and INS production, contributing to its role in blood glucose homeostasis.
ResultsTRAPα is highly expressed in human and mouse pancreatic cells; TRAPα-KO decreases circulating INS, resulting in age-related glucose intolerance. By immunofluorescence in both human and mouse pancreas, we found that TRAPα was highly expressed in pancreatic islets compared with exocrine pancreatic tissue. Notably, TRAPα colocalized primarily with INS rather than glucagon (GCG) or somatostatin, suggesting that TRAPα is more highly expressed in cells than islet cell types (Figure 1, A and B). To further investigate the subcellular localization of TRAPα, we performed confocal microscopy analyses using antibodies against TRAPα, KDEL (an ER marker), or TGN46 (a trans-Golgi network marker). Consistent with our previous findings in pancreatic cell lines (26), TRAPα exhibited strong colocalization with KDEL, indicating its predominant localization in the ER (Supplemental Figure 1; supplemental material available online with this article; https://doi.org/10.1172/JCI179845DS1).
Figure 1TRAPα is highly expressed in human and mouse pancreatic cells, and its expression is regulated by glucose. (A and B) The expression of TRAPα in human (A) and mouse (B) pancreases was detected via immunostaining with anti-TRAPα (green) and anti-INS (red), anti-GCG (red), or anti-somatostatin (red). (C) The islets isolated from 8- to 12-week-old WT mice were incubated in medium with either 2.8 mM or 16.7 mM glucose for 4 hours. The expression of mRNA of TRAPα, TRAPβ, TRAPγ, and TRAPδ was examined by real-time quantitative PCR. The mRNA levels in islets incubated with 2.8 mM glucose were set as 100% (n = 3). (D) The protein expression of TRAPα, TRAPβ, TRAPγ, and TRAPδ from WT islets treated with 2.8 mM and 16.7 mM for 4 hours was examined by Western blots (n = 3). (E) Quantification of TRAPα, TRAPβ, TRAPγ, and TRAPδ protein levels in D. (F) The mRNA levels of TRAPα, TRAPβ, TRAPγ, and TRAPδ were measured in islets isolated from 8- to 12-week-old control (Con) and TRAPα-βKO mice. The mRNA levels in Con islets were set as 100% (n = 4). (G) The protein expression of TRAPα, TRAPβ, TRAPγ, and TRAPδ from 8- to 12-week-old Con and TRAPα-βKO mice was examined by Western blots (n = 4–12). (H) Quantification of TRAPα, TRAPβ, TRAPγ, and TRAPδ protein levels in G. Values are reported as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001, unpaired Student’s t test and 2-way ANOVA.
Glucose-stimulated INS biosynthesis and secretion are characteristic features of cells. To determine whether the expression of TRAPα responds to glucose, we treated isolated mouse islets with low (2.8 mmol/L) and high (16.7 mmol/L) glucose concentrations for 4 hours and found that TRAPα as well as other subunits of TRAP complexes were all upregulated at both transcriptional and translational levels (Figure 1, C–E). Among all 4 subunits, TRAPα was the most upregulated mRNA(Figure 1C).
To further determine the role of TRAPα in cells, we established a TRAPα-βKO mouse line (Supplemental Figure 2A) using homologous recombination mediated by Ins2-IRES-Cre, in which an internal ribosome entry site after the Ins2 stop codon but before the poly-A sequence allows for Cre expression and activity restricted to β cells without detectable expression in the hypothalamus and without altering the expression of Ins2 or Ins1, circulating INS, or blood glucose in either the fasting state or during intraperitoneal glucose tolerance tests (IPGTTs) at age 3–4 months (30). To further confirm these findings, we set up 2 additional mouse cohorts including male Ins2-IRES-Cre mice and WT control mice. One cohort of mice had been fed a chow diet for up to 4 months, and another cohort of mice had been fed a chow diet for 2 months followed by additional 2-month HFD. Again, no differences were observed in body weight and glucose tolerance between Ins2-IRES-Cre and WT mice fed the chow diet (Supplemental Figure 3, A–C) or fed the HFD (Supplemental Figure 3, D and E).
Next, we examined efficiency of TRAPα deletion in TRAPα-βKO mice. A markedly decreased expression of TRAPα was independently confirmed by qRT-PCR (Figure 1F), Western blot (Supplemental Figure 2, B and C), and immunohistochemistry (Supplemental Figure 2D). The percentage of TRAPα-positive cells decreased 79% in TRAPα-βKO islets compared with that of the control (Supplemental Figure 2, E and F). Interestingly, although 3 other subunits (TRAPβ, TRAPγ, and TRAPδ) of the TRAP complex were upregulated at the mRNA level in TRAPα-βKO islets (Figure 1F), TRAPα deficiency resulted in decreased protein levels for all 3 other subunits (Figure 1, G and H), suggesting that TRAPα is important for the stability (or translational efficiency) of other subunits of the TRAP complex.
Next, we examined the impact of β cell TRAPα deficiency on blood glucose homeostasis. TRAPα-βKO mice started to have decreased circulating INS as early as 1 month of age without affecting body weight (Figure 2, A and B). Starting from 4 months of age, TRAPα-βKO male mice progressively developed INS-deficient impaired glucose intolerance (IGT), which became more profound at 6 months of age (Figure 2, D–F, and Supplemental Figure 4, A–J). Female TRAPα-βKO mice exhibited milder phenotypes with delayed onset of IGT (Figure 2, G–J, and Supplemental Figure 4, K–M).
Figure 2TRAPα-βKO decreases circulating INS and causes glucose intolerance. (A) The body weight of male and female control (Con) and TRAPα-βKO mice (n = 10). (B) Fasting serum INS levels of the same groups of mice as in A. (C) Fasting blood glucose levels of the same groups of mice as in A. (D–F) IPGTTs were performed in 2-, 4-, and 6-month-old Con and TRAPα-βKO male mice (Con, n = 5–10; TRAPα-βKO, n = 5–8). (G–I) IPGTTs were performed in 2-, 4-, and 6-month-old Con and TRAPα-βKO female mice (Con, n = 6-7; TRAPα-βKO, n = 8-10). (J) The AUC for INS level of the same groups of mice as D–I. Values are reported as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001, unpaired Student’s t test and 2-way ANOVA.
TRAPα-βKO impairs PPI translocation and decreases INS production but does not affect glucose-stimulated INS secretory responsiveness. Decreased circulating INS (Figure 2, B and J) could be caused by a decrease in INS gene expression a decrease of INS secretory responsiveness, a decrease of islet β cell mass, or a decrease of INS content due to defective INS biosynthesis. We found that the mRNA levels of both Ins1 and Ins2 were not decreased; surprisingly, they were upregulated in TRAPα-βKO islets (Figure 3A). Additionally in isolated islets, we examined glucose-stimulated INS secretion (GSIS; with low [2.8 mmol/L] and high [16.7 mmol/L] glucose concentrations) and found that although the absolute amount of INS was markedly decreased, demonstrated both by immunohistochemistry with anti-INS (Figure 3B) and by ELISA for INS measurement (Figure 3C), the fold-change of stimulated INS secretion was not affected in TRAPα-βKO islets (Figure 3D), demonstrating that TRAPα deficiency does not affect glucose-stimulated secretory responsiveness. Furthermore, neither islet size (Figure 3, E and F) nor the composition and distribution of 3 major islet cell types (α, β, and δ cells) (Figure 3G) were significantly affected by TRAPα-βKO. These data suggest the reduction in circulating INS was not caused by decreased INS gene transcription or β cell mass but might result from reduced INS content that was likely caused by defective INS biosynthesis.
Figure 3TRAPα-βKO impairs PPI translocation and decreases INS production but does not affect GSIS or islet size and cell composition. (A) mRNA levels of INS genes, including Ins1 and Ins2 from 8- to 12-week-old male control (Con) and TRAPα-βKO mice islets (n = 10). (B) Immunohistochemistry staining was performed using anti-INS in pancreatic sections of 8- to 12-week-old male Con and TRAPα-βKO mice. (C) INS content from 8- to 12-week-old male Con and TRAPα-βKO mouse islets was measured using INS ELISA normalized with either islet total protein (left) or islet DNA content (right) to minimize the effects of interislet heterogeneity. The INS content from Con mouse islets was set as 100% (n = 5–10). (D) GSIS was performed using islets isolated from 8- to 12-week-old male Con and TRAPα-βKO mice (n = 6). Secreted INS from male control islets treated with 2.8 mM glucose was set as 100% (n = 6 in each group). (E) H&E staining was performed of pancreases of 8- to 12-week-old male Con and TRAPα-βKO mice. (F) The quantification of the islets size of E (n = 10 in each group). (G) Pancreatic sections of 8- to 12-week-old male Con and TRAPα-βKO mice were immunostained with anti-INS (red), anti-GCG (green, left panel), or anti-somatostatin (green, right panel) as indicated. (H) Islets isolated from 8- to 12-week-old male Con and TRAPα-βKO mice were treated with or without MG132 (10 μM) for 2 hours before being analyzed by Western blotting using anti-INS or anti-TRAPα, as indicated. (I) Quantification of TRAPα protein levels in islets treated with or without MG132 in H (n = 10). (J) Quantification of PI and INS in TRAPα-βKO islets with and without MG132 treatment shown in H (n = 6–16). Values are reported as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, unpaired Student’s t test and 2-way ANOVA.
To dissect which steps of INS biosynthesis might be affected by TRAPα, we first examined PPI translocation into the ER via the Sec61 translocon (1). The half-life of PPI is normally only 2–3 minutes, in part because untranslocated PPI is rapidly degraded by the proteasome (31). To detect untranslocated PPI, we treated isolated islets with or without MG132 (a proteasomal inhibitor that prevent untranslocated PPI from degradation; ref. 12) before lysis. We found that, without affecting TRAPα expression (Figure 3, H and I), MG132 treatment allowed PPI to become visible in TRAPα-βKO islets (Figure 3H, lane 2 vs. lane 4). TRAPα-βKO was accompanied by markedly decreased PI and mature INS (Figure 3, H and J), suggesting that TRAPα deficiency impairs PPI translocation and INS biosynthesis.
TRAPα is important for β cell ER function, and its deficiency disturbs PI oxidative folding in the ER, inducing ER stress. The topology of the TRAPα protein is such that most of the protein is present on the luminal side of the ER membrane (32), suggesting that, in addition to its role in PPI translocation, TRAPα has a function within the ER lumen. Interestingly, transcriptome analyses revealed that the expression of 449 genes was upregulated and that of 237 genes was downregulated in TRAPα-βKO islets (Figure 4A). The upregulated ER genes were related to unfolded or misfolded protein binding, responses, and degradation, as well as protein disulfide formation and isomerization (Figure 4B). qRT-PCR confirmed that the master ER stress regulator binding-immunoglobulin protein (Bip), the major ER oxidoreductases Ero1lb and protein disulfide isomerase, and the key ER chaperone involved in ER-associated protein degradation, Sel1l, were all upregulated, whereas the mRNA encoding Hrd1 (an E3 ubiquitin ligase) was decreased (Figure 4C). Furthermore, Western blotting confirmed that both BiP and phosphorylated eukaryotic translation initiation factor 2 subunit 1 (p-eIF2α) were increased in TRAPα-βKO islets (Figure 4D, quantified in Figure 4E). These multiple independent approaches indicate that TRAPα-βKO induces ER stress and activates the unfolded protein response (UPR).
Figure 4TRAPα is important for β cell ER function, and its deficiency causes PI misfolding and ER stress. (A) Transcriptome analyses using islets isolated from 8-week-old control (Con) and TRAPα-βKO male mice (n = 3 in each group). (B) Gene Ontology analyses of RNA-Seq data showed the pathways associated with ER genes that were significantly upregulated in islets of 8-week-old TRAPα-βKO male mice (as in A). (C) mRNA levels of indicated genes were measured by real-time quantitative PCR (qRT-PCR) in islets from 8- to 12-week-old Con and TRAPα-βKO male mice islets (n = 3-6). (D) The protein expression of TRAPα, BiP, and p-eIF2α from 12-week-old Con and TRAPα-βKO male mice was examined by Western blots. (E) Quantification of TRAPα, BiP, and p-eIF2α protein levels in D (n = 3 in each group). (F) The oxidative folding of PI in islets from 8- to 12-week-old Con and TRAPα-βKO male mice was analyzed by Western blots under both nonreducing and reducing conditions. (G) The amounts of PI dimers plus trimers under nonreducing conditions compared with total PI under reducing conditions were quantified and calculated. The ratios of dimers plus trimers to total PI in control islets were set to 100% (n = 4). (H) INS1 control cells (WT) and INS1 with both Ins1 and Ins2 gene-deleted cells (Mut) were transfected with either TRAPα siRNA or scrambled siRNA. At 72 hours after transfection, the cells were lysed, and ER stress responses were analyzed by Western blots, as indicated. (I) Quantification of H from 5–8 independent experiments. (J) Isolated islets were incubated in RPMI 1640 without FBS for 6 hours; the secretion of PI and CPE was analyzed by Western blots. (K) PI and CPE in the islets and media were quantified from J in 3 independent experiments. The secretion efficiency of PI and CPE was calculated, and secretion efficiency of PI and CPE in control islets was set to 100%. Values are reported as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001, unpaired Student’s t test and 2-way ANOVA.
Next, we asked what factor(s) mediate the ER stress induced by TRAPα deficiency. PI is the most abundant protein in β cell ER, and its oxidative folding is sensitive to impairment of ER function (17, 21). To examine the possible role of TRAPα in PI folding in the ER, we examined oxidative folding of PI under both nonreducing and reducing conditions, using established methods (14, 20). When normalized for comparable levels of total PI (detected under reducing conditions, as seen in Figure 4F, for which total islet protein was intentionally increased), the number of higher molecular weight, disulfide-linked, PI-containing protein complexes was dramatically increased, and expression of monomeric PI was decreased in TRAPα-βKO islets (Figure 4F, quantified in Figure 4G). These data suggest TRAPα-βKO impairs PI oxidative folding and promotes PI misfolding in the ER.
To determine whether misfolded PI is the direct cause of ER stress in TRAPα deficiency cells, we established a rat insulinoma INS1 cell line in which Ins1 plus Ins2 were knocked out. As shown in Figure 4, H and I, downregulating expression of TRAPα with siRNA in control INS1 cells increased expression of BiP and p-eIF2α that was consistent with the results found in TRAPα-βKO islets (Figure 4, D and E). However, in INS1 cells lacking PI expression, TRAPα deficiency did not activate UPR (Figure 4, H and I), indicating that ER stress caused by TRAPα-βKO is likely mediated by an increase of misfolded PI. Interestingly, although the PI secretion rate was decreased
Comments (0)