Impaired cleavage of preproinsulin signal peptide linked to autosomal-dominant diabetes

Abstract

Recently, missense mutations upstream of preproinsulin's signal peptide (SP) cleavage site were reported to cause mutant INS gene-induced diabetes of youth (MIDY). Our objective was to understand the molecular pathogenesis using metabolic labeling and assays of proinsulin export and insulin and C-peptide production to examine the earliest events of insulin biosynthesis, highlighting molecular mechanisms underlying β-cell failure plus a novel strategy that might ameliorate the MIDY syndrome. We find that whereas preproinsulin-A(SP23)S is efficiently cleaved, producing authentic proinsulin and insulin, preproinsulin-A(SP24)D is inefficiently cleaved at an improper site, producing two subpopulations of molecules. Both show impaired oxidative folding and are retained in the endoplasmic reticulum (ER). Preproinsulin-A(SP24)D also blocks ER exit of coexpressed wild-type proinsulin, accounting for its dominant-negative behavior. Upon increased expression of ER-oxidoreductin-1, preproinsulin-A(SP24)D remains blocked but oxidative folding of wild-type proinsulin improves, accelerating its ER export and increasing wild-type insulin production. We conclude that the efficiency of SP cleavage is linked to the oxidation of (pre)proinsulin. In turn, impaired (pre)proinsulin oxidation affects ER export of the mutant as well as that of coexpressed wild-type proinsulin. Improving oxidative folding of wild-type proinsulin may provide a feasible way to rescue insulin production in patients with MIDY.

FIG. 1.

A(SP23)S exhibits no abnormal phenotype, whereas A(SP24)D exhibits a defect in SP cleavage and insulin production and induces ER stress in β-cells. A: Signal sequence alignment of preproinsulins of various species. The arrow indicates the predicted signal peptide cleavage site; the −1 residue is boxed and a Ser-SP23 (−2) residue in chimpanzee preproinsulin is circled. B: MIN6 cells were transiently transfected with plasmids encoding hProCpepMyc-WT, A(SP23)S, or A(SP24)D and lysed after 48 h posttransfection, and 20 μg total protein was used for anti-Myc Western blot as described in research design and methods. C: MIN6 cells were transiently transfected with human preproinsulin-WT, A(SP23)S, or A(SP24)D plasmids. At 40 h posttransfection, the cells were incubated with fresh DMEM medium containing 25.5 mmol/L glucose and 10% FBS for additional 16 h; the media were collected and cells lysed as in research design and methods. Human insulin in the lysates (black bars) and media (white bars) were quantified using human-specific insulin RIA normalized to human-specific insulin mRNA (a measure of the efficiency of insulin production from the transfected translation product). Results shown are mean values ± SD from three independent experiments. *P < 0.01 compared with preproinsulin-WT or A(SP23)S. D: Min6 cells cotransfected to express BiP promoter-firefly luciferase, CMV promoter-driven R. luciferase, and preproinsulin-WT or mutants were lysed at 48 h posttransfection, and the ratio of firefly/R. luciferase was measured. Cells expressing WT proinsulin or empty vector (EV) served as a negative control. Results are expressed as mean ± SD from three independent experiments. *P < 0.05 compared with preproinsulin-WT.

FIG. 2.

SP cleavage and ER export of A(SP23)S and A(SP24)D. A: Transfected 293T expressing human preproinsulin-WT or mutants were pulse-labeled with ³⁵S-Cys/Met for 10 min followed by 0 or 90 min chase. Cell lysates (C) and chase media (M) were immunoprecipitated with anti-insulin and analyzed using 4–12% NuPage under reducing conditions. B: From repeat experiments like that shown in A, without chase, the uncleaved and processed preproinsulin bands were quantified by scanning densitometry (± SD). P < 0.01 compared to either WT or A(SP23)S. C: Transfected 293T cells expressing WT or A(SP24)D were pulse-labeled with ³⁵S-Cys/Met for 30 min followed by 0, 1, or 4 h chase with or without 10 μmol/L MG132. The cell lysates and chase media were combined and immunoprecipitated with anti-insulin and analyzed using 4–12% NuPage under reducing conditions.

FIG. 3.

Translocation of preproinsulin-A(SP24)D across the ER membrane. A: At 40 h posttransfection, 293T cells coexpressing cytosolic GFP and preproinsulin-WT or -A(SP24)D were pulse-labeled with ³⁵S-Cys/Met for 15 min and then treated with 0.01% digitonin on ice for 10 min to permeabilize the plasma membrane, which liberates a major fraction of cytosolic GFP. Cells were then sedimented at 14,000 rpm at 4°C for 10 min and each pellet (P) and supernate (S) analyzed sequentially by anti-insulin and anti-GFP immunoprecipitation, 4–12% NuPage, under reducing conditions and autoradiography. B: The 18th residue of human preproinsulin C-peptide was mutated from Ala to Asn to create an N-linked glycosylation site. At 40 h posttransfection, 293T cells expressing human preproinsulin-A(Cpep18)N bearing or lacking the A(SP24)D mutation were pulse-labeled with either pure ³⁵S-Met or mixed ³⁵S-Cys/Met for 15 min before lysis and immunoprecipitation with anti-insulin. Immunoprecipitates were split in half and either digested (+) or mock digested (−) with PNGase F at 37°C for 1 h before analysis by 4–12% NuPage under reducing conditions. At this exposure, preproinsulin-WT is not detected with pure ³⁵S-Met, indicating that the vast majority of molecules have already undergone SP cleavage; thus, the same sample labeled with ³⁵S-Cys/Met indicates glycosylated (Glyco-Proins) or deglycosylated proinsulin (Proins). However, A(SP24)D remains labeled with pure ³⁵S-Met, indicating the positions of glycosylated (Glyco-preProins) or deglycosylated preproinsulins (preProins).

FIG. 4.

The kinetics of SP cleavage and disulfide bond formation. A: At 40 h posttransfection, 293T cells expressing human preproinsulin-WT or -A(SP24)D were pulse-labeled for 10 min with either ³⁵S-Cys/Met or pure ³⁵S-Met, as indicated. The cells were pretreated with 20 mmol/L NEM in PBS on ice for 10 min before lysis and immunoprecipitation with anti-insulin. The immunoprecipitates were analyzed using Tris-Tricine urea–SDS-PAGE under nonreducing and reducing conditions, as indicated. The positions of both oxidized and reduced forms of preproinsulin (preProins) and proinsulin (Proins) are indicated. B: At 40 h posttransfection, 293T cells expressing human preproinsulin WT or A(SP24)D were pulse-labeled with ³⁵S-Cys/Met for 10 min and chased for 0 or 10 min. Newly synthesized preproinsulin (preProins) and proinsulin (Proins) were analyzed as in A. C: 293T cells transiently transfected to express preProCpepMyc-WT or the same construct bearing A(SP24)D were lysed at 48 h posttransfection, and 20 μg total proteins was resolved by 4–12% NuPage under nonreducing and reducing conditions, as indicated, followed by anti-Myc Western blotting as described in research design and methods. D: INS1 cells transfected with preProCpepMyc-A(SP24)D were pulse-labeled with ³⁵S-Cys/Met for 30 min. The cells were lysed and immunoprecipitated with anti-Myc and analyzed as in A.

FIG. 5.

Inappropriate SP cleavage of A(SP24)D and engineered preproinsulins. A: WT or engineered human preproinsulins (with the names indicated in the column at left) show predicted SP cleavage sites are based on the SignalP 3.0 algorithm (http://www.cbs.dtu.dk/services/SignalP/). SP cleavage (black arrowheads) leaves predicted additional residues attached to the NH₂ terminus of the proinsulin B-chain. B: At 48 h posttransfection, 293T cells expressing human preproinsulin-WT (AAA-Proins in this panel) or mutants as indicated were pulse-labeled for 15 min with either pure ³⁵S-Met or ³⁵S-Met/Cys as indicated. Cell lysates were immunoprecipitated with anti-insulin and analyzed using 4–12% NuPage under reducing conditions, followed by autoradiography. C: Transfected 293T cells were labeled with ³⁵S-Met/Cys for 15 min and chased for 150 min. The cell lysates (C) and chase media (M) were immunoprecipitated and analyzed as in B. (A high-quality color representation of this figure is available in the online issue.)

FIG. 6.

Dominant-negative blockade of proinsulin-WT caused by preproinsulin-A(SP24)D, miscleaved proinsulin, and uncleaved preproinsulin. A: 293T cells were cotransfected to express human preproins-WT with either mouse preproins-WT, A(SP24)D, or A(SP24)D-DelCys [in which all cysteines of A(SP24)D were mutated (21)] at a DNA ratio of 1:2. Beginning at 30 h posttransfection, cells were incubated with DMEM containing 25.5 mmol/L glucose plus 10% FBS for additional 16 h. Media were collected, and a human proinsulin-specific RIA was used to measure secretion of coexpressed human proinsulin-WT. The secretion of human proinsulin coexpressed with mouse preproins-WT served as a positive control (i.e., set to 100%). Results are shown as mean ± SD from three independent experiments. B: INS1 cells cotransfected with preProCpepGFP-WT with either preProCpepMyc-WT or A(SP24)D (at a DNA molar ratio of 1:2) were labeled for either 0.5 or 20 h as indicated. The cell lysates (C) and media (M) were immunoprecipitated with anti-GFP and analyzed under reducing conditions. C: INS1 cells transfected to express preProCpepMyc-WT, A(SP24)D, or PFD were lysed at 48 h posttransfection. The cell lysates were resolved by 4–12% NuPage under reducing conditions. The proteins were electrotransferred to nitrocellulose and immunoblotted with anti-Myc antibody. D: 293T cells cotransfected with preProCpepMyc-WT and untagged AAAAD-Proins (processed to AD-Proins; see Fig. 5A) at a range of DNA ratios (indicated below) were pulse-labeled with ³⁵S-Cys/Met for 30 min and chased for 0 or 2 h. Cell lysates (C) and collected chase media (M) were immunoprecipitated with anti-insulin and analyzed under reducing conditions. E: 293T cells cotransfected with untagged preproinsulin-WT (processed to Proins) and preProCpepMyc-WT, A(SP24)D, or PFD were pulse-labeled with ³⁵S-Cys/Met for 30 min and chased for 0 or 3 h. Cell lysates (C) and collected chase media (M) were analyzed as in D.

FIG. 7.

Dominant-negative blockade of human proinsulin by A(SP24)D is partially rescued by increased expression of ERO1. A: At 48 h posttransfection, media bathing 293T cells coexpressing human preproinsulin-WT, one of the mouse preproins-[WT, C(A7)Y, or A(SP24)D], and empty vector (EV), PDI, or ERO1 (at a DNA molar ratio of 1:2:3) were collected for 16 h. Secreted human proinsulin was measured using a human-specific proinsulin RIA. Secretion of human proinsulin in the presence of mouse proinsulin-WT and empty vector was set to 100%. B: INS1 cells coexpressing human preproinsulin and mouse preproinsulin-A(SP24)D plus ERO1α or ERO1β (at a DNA molar ratio of 1:3:4) were lysed with acid/ethanol to measure human insulin content using human-specific insulin RIA. The human insulin content of cells cotransfected with WT human and mouse preproinsulins was set to 100%. C: At 48 h posttransfection, 293T cells coexpressing preproinsulin-WT and either preProCpepMyc-WT (processed to Myc-tagged Proins) or preProCpepMyc-A(SP24)D (shown as Myc-tagged preProins) plus empty vector (Control) or ERO1α (at a DNA molar ratio of 1:2:3) were pulse-labeled with ³⁵S-Cys/Met for 30 min and chased for 0 or 3 h. Newly synthesized (pre)proinsulins from cell lysates (C) and chase media (M) were immunoprecipitated with anti-insulin, resolved by Tris-Tricine urea–SDS-PAGE under reducing conditions, and analyzed by autoradiography.

FIG. 8.

Increased expression of ERO1β improves oxidative folding of proinsulin-WT in presence of A(SP24)D. A: At 48 h posttransfection, 293T cells coexpressing preproins-WT with Myc-tagged preproins-WT or -A(SP24)D, plus ERO1β (+) or empty vector (−) were pulse-labeled with ³⁵S-Cys/Met for 30 min without chase. The cells were lysed as in Fig. 4A, and coexpressed proins-WT was analyzed by immunoprecipitation and Tris-Tricine urea–SDS-PAGE under both nonreducing and reducing conditions, with quantitation by densitometry after autoradiography. The fractional recovery of the native disulfide isomer of untagged proins-WT under nonreducing conditions was compared against total proins-WT recovered under reducing conditions. The relative recovery of the native disulfide isomer of untagged proins-WT in the presence of Myc-tagged preproins-WT was set to 100% for purposes of comparison. B: Total newly synthesized proins-WT (from cell lysates plus chase media) recovered from immunoprecipitates analyzed by reducing Tris-Tricine urea–SDS-PAGE is shown over 4-h time course in the presence of coexpressed Myc-tagged preproins-A(SP24)D. Note that there was increased final recovery of total newly synthesized proins-WT in cells with increased expression of ERO1β.