Evidence That Human Β-Cells Process Proinsulin with Prohormone Convertase

Diabetes Volume 69, July 2020 1451

Revisiting Proinsulin Processing: Evidence That Human b-Cells Process Proinsulin With Prohormone Convertase (PC) 1/3 but Not PC2

Adam Ramzy,1 Ali Asadi,1 and Timothy J. Kieffer1,2

Diabetes 2020;69:1451–1462 | https://doi.org/10.2337/db19-0276

Insulin is first produced in pancreatic b-cells as the B-chain and COOH-terminal A-chain with a connecting precursor prohormone proinsulin. Defective proinsulin C-peptide, many studies have attempted to clarify how processing has been implicated in the pathogenesis of b-cells excise the C-peptide to liberate mature insulin. both type 1 and type 2 diabetes. Though there is sub- Current theory posits that the B-chain–C-peptide junction stantial evidence that mouse b-cells process proinsulin is cleaved by prohormone convertase 1/3 (PC1/3; gene using prohormone convertase 1/3 (PC1/3) and then pro- PCSK1) before cleavage at the C-peptide–A-chain junction fi hormone convertase 2 (PC2), this nding has not been by prohormone convertase 2 (PC2; gene PCSK2). PC2 fi b fl veri ed in human -cells. Immuno uorescence with val- knockout mice have impaired processing at the C-A STUDIES ISLET idated antibodies revealed that there was no detectable junction, resulting in a buildup of des-31,32 proinsulin (2), PC2 immunoreactivity in human b-cells and little PCSK2 and PC1/3 knockout mice have severely impaired process- mRNA by in situ hybridization. Similarly, rat b-cells were ing at the B-C junction, resulting in a buildup of des-64,65 not immunoreactive for PC2. In all histological experi- proinsulin (3). Based on relative processing rates of intact ments, PC2 immunoreactivity in neighboring a-cells acted as a positive control. In donors with type 2 diabe- human proinsulin versus des-31,32 proinsulin or des-64,65 tes, b-cells had elevated PC2 immunoreactivity, sug- proinsulin by rat PC1/3 and PC2, processing at the B-C gesting that aberrant PC2 expression may contribute junction by PC1/3 likely occurs before processing by PC2 at to impaired proinsulin processing in b-cells of patients the C-A junction (4), but some data have countered this with diabetes. To support histological findings using hypothesis by showing more buildup of des-64,65 proinsu- a biochemical approach, human islets were used lin during the processing of rat insulin 2 in islets (5). for pulse-chase experiments. Despite inhibition of PC2 While past work supports the theory that primary function by temperature blockade, brefeldin A, chloro- mouse b-cells process proinsulin sequentially by PC1/3 quine, and multiple inhibitors that blocked production of and then PC2, there is some indication that PC2 may not mature glucagon from proglucagon, b-cells retained the be as important in human b-cells as it is in mouse b-cells. ability to produce mature insulin. Conversely, suppression Humans with mutant PCSK1 have circulating hyperproin- of PC1/3 blocked processing of proinsulin but not proglu- sulinemia (6), but there are no associations between PCSK2 cagon. By demonstrating that healthy human b-cells pro- polymorphisms and circulating proinsulin (7). Though cess proinsulin by PC1/3 but not PC2, we suggest that a human b-cell line (EndoC-bH2) has abundant PCSK1 and there is a need to revise the long-standing theory of pro- PCSK2 (8), RNA-sequencing experiments on sorted pri- insulin processing. mary a-cells and b-cells indicate higher expression of PCSK2 than PCSK1 in mouse b-cells (9), whereas human In 1967, Steiner et al. (1) demonstrated with pulse-chase b-cells expressed ;20 times more PCSK1 than PCSK2 (10) experiments that insulin is generated from a larger pre- (Supplementary Fig. 1). Additionally, Davalli et al. (11) cursor they named “proinsulin.” After the general structure reported a deficiency of immunoreactive PC2 in human islet fi b of insulin was identi ed as containing an NH2-terminal -cells transplanted into mice and some human insulinomas

1Laboratory of Molecular and Cellular Medicine, Department of Cellular and This article contains supplementary material online at https://doi.org/10.2337/ Physiological Sciences, Life Sciences Institute, The University of British Columbia, db20-4567/suppl.12074721. Vancouver, British Columbia, Canada © 2020 by the American Diabetes Association. Readers may use this article as 2 Department of Surgery, The University of British Columbia, Vancouver, British long as the work is properly cited, the use is educational and not for profit, and the Columbia, Canada work is not altered. More information is available at https://www.diabetesjournals Corresponding author: Timothy J. Kieffer, [email protected] .org/content/license. Received 13 March 2019 and accepted 3 April 2020 1452 Human b-Cells Process Proinsulin With PC1/3 Diabetes Volume 69, July 2020 are not immunoreactive for PC2 (12). Collectively, there is no Immunohistofluorescence definitive evidence that proinsulin is processed by both PC1/3 Immunofluorescent staining was performed as previously and PC2 in human b-cells,andthereissomeindicationthat described (21). Briefly, sections were deparaffinized in PC2, while critical for proglucagon processing in a-cells xylene (three times for 5 min) and rehydrated in graded (13), is not abundantly expressed in human b-cells. ethanol (100%, two times for 5 min, 95% for 5 min, 70% In this study, we use validated antibodies and oligonu- for 5 min, and PBS for 10 min) before heat-induced epitope cleotide probes to determine that primary human b-cells retrieval in an EZ-Retriever microwave oven (BioGenex, Fre- have no detectable PC2 and little detectable PCSK2.We mont, CA) for 15 min at 95°C in 10 mmol/L citrate buffer also performed pulse-chase experiments and suppressed (0.5% Tween 20, pH 6.0) (Thermo Fisher Scientific, Waltham, the function of PC2 by temperature blockade (14), bre- MA). Samples were blocked in Dako Protein Block, Serum Free feldin A (15,16), the weak base chloroquine (14), and (Dako Canada, Burlington, Ontario Canada), and incubated multiple inhibitors (17,18). In the absence of full PC2 overnight in primary antibody diluted in Dako Antibody function, human b-cells can produce mature insulin, but Diluent. The following day, slides were washed and incubated neighboring a-cells produce little mature glucagon. in secondary antibody (Alexa Fluor–conjugated secondary Moreover, we provide evidence that PC1/3 is responsible antibodies; Life Technologies) for 1 h at room temperature for processing human proinsulin by inhibiting the function before mounting and counterstaining with VECTASHIELD of PC1/3 using two inhibitors (18,19) to impair formation HardSet Mounting Medium with nuclear stain DAPI (Vector of mature insulin with no significant effect on proglucagon Laboratories, Burlingame, CA). All images were captured processing. These findings provide a more advanced un- and analyzed with an ImageXpress Micro XLS System derstanding of the processing of proinsulin and suggest the (Molecular Devices, LLC, San Jose, CA) with a scientific need to reconsider the widely accepted thought that human CMOS camera, a Nikon 20X Plan Apo objective (numer- proinsulin is processed sequentially by PC1/3 and then PC2. ical aperture of 0.75, 1-6300-0196; Nikon, Tokyo, Japan), Our findings suggest that in human b-cells, PC1/3 is re- and DAPI (DAPI-5060B), FITC (FITC-3540B), Cy3 (Cy3– sponsible for processing human proinsulin without PC2. 4040B), Texas Red (TXRED-4040B), and Cy5 (Cy5– 4040A) filter cubes. Image analysis was performed on RESEARCH DESIGN AND METHODS MetaXpress software (version 6.2.3.733; Molecular Devi- Experimental Models and Subject Details ces, LLC). HumanpancreastissuebiopsieswerecollectedbytheIke Barber Human Islet Transplant Laboratory (Vancouver, British Cell Sorting Columbia, Canada). Human pancreas tissue biopsies and A total of 10,000–20,000 human islet equivalents were cadaveric human islets were provided by the Alberta Diabetes dispersed in 0.05% trypsin before magnetic bead purifica- Institute IsletCore (Edmonton, Alberta, Canada) after isolation tion according to a published protocol (22). by standardized protocol (20). Sample collection and islet Western Blotting isolation were approved by the Human Research Ethics Board Groups of 250 mouse islets, 1,000 human islet equivalents, at the University of Alberta (Pro00013094). All donors’ fam- or ;1 million EndoC-bH1 cells were lysed in 200 mL lysis ilies gave informed consent for the use of pancreatic tissue in buffer (50 mmol/L Tris, pH 8.0, 150 mmol/L NaCl, 0.02% research, and all work with human tissues was approved by the Na azide, 0.1% SDS, 1% Nonidet P-40, 0.5% sodium Research Ethics Board (H14-02949), The University of British deoxycholate, 1 mmol/L phenylmethyl sulfonyl fluoride, Columbia (Vancouver, British Columbia, Canada). Basic donor and protease inhibitor cocktail [Sigma-Aldrich, St. Louis, demographics are detailed in Supplementary Tables 2 and 3. MO]) as per a published protocol (15). Levels of proteins of All experiments with animals were approved by the University interest were assessed by fluorescent Western blotting of British Columbia Animal Care Committee and carried out methods using two PC2 antibodies (1:1,000, MAB6018, in accordance with the Canadian Council on Animal Care R&D Systems; and 1:1,000, PA5-14594, Thermo Fisher Guidelines. Scientific), anti-PC1/3 antibody (1:2,500; gift from Lakshmi Devi), and an anti–a-tubulin antibody (1:1,000; Sigma- Paraffin-Embedded Samples Aldrich) and detected on a LI-COR Odyssey 9120 Imaging C57BL/6J mice (The Jackson Laboratory, Bar Harbor, ME) system (LI-COR Biosciences, Lincoln, NE). were sacrificed at 12 weeks of age. Following euthanasia, the pancreas was quickly dissected out of mice, washed in PBS, and In Situ Hybridization fixed in 4% paraformaldehyde overnight before being trans- For all in situ hybridization experiments, we used tissue ferred to 70% ethanol for storage prior to paraffin embedding samples collected and sectioned in all RNase-free solu- and sectioning (5-mm thickness) (Wax-it Histology Services tions (including paraformaldehyde, 70% ethanol, and Inc., Vancouver, British Columbia, Canada). A similar protocol water for sectioning) and cleaned all equipment with was used to collect pancreas from rats, pigs, and dogs. Human RNase AWAY (Thermo Fisher Scientific). We performed pancreas tissue biopsies were collected using the same protocol a standard protocol using a modified version of a com- (Ike Barber Human Islet Tranplant Laboratory). mercially available in situ hybridization kit (BioChain diabetes.diabetesjournals.org Ramzy, Asadi, and Kieffer 1453

Institute Inc., Newark, CA). We used three probes specific Quantification and Statistical Analysis for PCSK2 in regions highly conserved between human Statistical analyses were done in GraphPad Prism V7, and and mouse and used a primary mouse anti-digoxigenin statistical tests performed are noted in figure legends. We antibody and secondary goat anti-mouse secondary anti- used P , 0.05 as the cutoff for significance. body conjugated to alkaline phosphatase (Supplementary Data and Resource Availability Key Resources Table). Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Quantitative PCR T.J.K. The data sets generated and/or analyzed during the Quantitative PCR was performed using a standard pro- 2DCt current study are available from the corresponding author tocol. mRNA expression level is represented as 2 on reasonable request. Critical antibodies and probes used in (where Ct is threshold cycle) using the average of two the current study are also available from the corresponding internal control genes (GAPDH and ACTB). author on reasonable request.

Pulse-Chase Experiments RESULTS After receipt of islets, they were incubated overnight in Human b-Cells Were Not Immunoreactive for PC2 complete media (CMRL) (Corning insulin-transferrin- Using Four Antibodies sodium selenite [Corning, NY], GlutaMAX [Thermo Fisher To minimize risk of nonspecific findings, we performed fi Scienti c], BSA [Roche Diagnostics, Laval, Quebec, Canada], antibody validation experiments.Therewereidentical and penicillin/streptomycin). The next day, islets were immunostaining patterns for PC2 using multiple antibod- preincubated for 60 min in Krebs Ringer bicarbonate ies in human and mouse pancreas (Fig. 1A). To negatively buffer, HEPES, and 16.7 mmol/L glucose (KRBH-16.7) validate with a knockout model, we used an antibody that at 37°C, pulsed for 40 min in 1 mL KRBH-16.7 with binds to the deleted NH -terminal prodomain of PC2 (26). m 3 2 800 Ci/mL [ H]Leu (PerkinElmer, Waltham, MA), and This antibody produced no immunoreactivity in pancreas chased for 120 min. Islets were split into groups of 1,000 from PC2 knockout mice (Fig. 1B). By Western blot, two and chased in KRBH-2.8 in eight conditions: 37°C, 20°C, antibodies (MAB6018, red, R&D Systems; and PA5-14594, m 37°C with 10 g/mL brefeldin A (Sigma-Aldrich), 37°C with green, Thermo Fisher Scientific) immunoreacted with m 30 mol/L chloroquine, 37°C with one of two PC1/3 inhibitors recombinant human PC2 and produced a blot with the (537076, EMD Millipore; or 166811, MedChem ShortCut expected bands for pro-PC2 (75 kDa) and mature PC2 LLC) (18,19), or 37°C with one of two PC2 inhibitors (64 kDa) from both human and mouse islet extracts (Fig. (166830, MedChem ShortCut LLC; or 5408-0471, ChemDiv) 1C). Using four antibodies, we detected robust immuno- m (17,18). Islets were lysed in 150 Llysisbuffer(15). reactivity for PC2 in mouse b-cells and neighboring islet cells but failed to detect immunoreactivity for PC2 in b Immunoprecipitation, SDS-PAGE, and Scintillation human -cells (Fig. 1D and see Supplementary Fig. 2A for Counting single channels). Notably, though not detectable in human Islet lysates were incubated overnight at 4°C with 5 mg b-cells, we observed the expected robust PC2 immunore- carrier-free anti-insulin antibody (C27C9; Cell Signaling activity in a-cells. These findings were replicated in four Technology,Danvers,MA)followedby6-hincubationat additional organ donors (Supplementary Fig. 3). We de- 4°C with 50 mL protein-A/G magnetic beads (50% slurry in termined whether there is a distinct distribution of PC2 in lysis buffer) (Thermo Fisher Scientific). After washing the head, body, and tail regions of human pancreas and whole three times, immunoprecipitated insulin was eluted by mouse pancreas by immunostaining for insulin, glucagon, heating to 95°C in Tris-tricine sample buffer (Bio-Rad and PC2 (Fig. 1E). Unlike in the mouse, in human pancreas Laboratories). We immunoprecipitated glucagon with the intensity of immunoreactivity for PC2 in b-cells was the same protocol using an anti-glucagon antibody dramatically less than in a-cells and there was almost no (EP3070; Abcam). Samples were then separated using colocalization of insulin and PC2 in all regions of human a Tris-tricine PAGE approach to separate small proteins pancreas. The rate of colocalization of PC2 and insulin was (23). After separation, gels were stained by colloidal not significantly different than the rate of colocalization of Coomassie (24), and visualized mature and prohormone insulin and glucagon in the head, body, or tail of human bands were excised manually. Gel slices were then dis- pancreas (P . 0.05; data not shown). In all regions of human solved in 0.3 mL of 30% hydrogen peroxide for 3 days at and mouse pancreas, there was high colocalization of PC2 50°C. Scintillation counting was performed on an LS6500 with glucagon. Unlike in the mouse, PC1/3 and PC2 did not scintillation system (Beckman Coulter, Brea, CA) with colocalize in human pancreas (Supplementary Fig. 2B). We quench curve correction for final disintegrations per min- immunostained for the essential PC2 cofactor neuroendocrine ute counts in Ultima Gold LLT scintillation fluid (Perkin- protein 7B2 (gene SCG5), and though there was immunore- Elmer). Counts were adjusted for background and number activity for 7B2 in human a-cells and b-cells, signal intensity of leucines in prohormones compared with processed was lower in b-cells, unlike in mice, in which there was similar mature hormones (25). immunoreactivity in both cell types (Supplementary Fig. 4). 1454 Human b-Cells Process Proinsulin With PC1/3 Diabetes Volume 69, July 2020

Figure 1—Human b-cells were not immunoreactive for PC2 using four antibodies. A: Immunostaining for PC2 using three antibodies (see Supplementary Key Resources Table for antibody details) in adult human and mouse pancreas. Representative images shown (n 5 3). Scale bar, 100 mm. B: Immunostaining for insulin (green) and PC2 (red) in wild-type and PC22/2 mouse pancreas. Representative images shown (n 5 2). Scale bar, 100 mm. C: Western blot for PC2 using two antibodies (MAB6018, red, R&D Systems; and NH2-terminal [N-term], PA5- 14595, green, Thermo Fisher Scientiﬁc) in human islet (HI) and mouse islet (MI) lysates with a recombinant human PC2 control. D: Immunostaining for insulin (green) and PC2 (red) using four PC2 antibodies. Representative images shown (n 5 5). Scale bar, 100 mm. See also Supplementary Figs. 2–4. E: Immunostaining for insulin (INS), glucagon (GCG), and PC2 COOH-terminal (C-term) (Supplementary Key Resources Table) in head, body, and tail of human pancreas and mouse pancreas (representative images of n 5 3 shown in left panel). Scale bar, 100 mm (insets [white boxes] are enlarged 34). E: Quantiﬁcation of the relative PC2 immunoreactivity intensity in a-cells versus b-cells (middle panel) and the rate of colocalization of PC2 with insulin or glucagon (right panel). Groups were compared by one-way ANOVA with Tukey post hoc test (**P , 0.01; ***P , 0.001). diabetes.diabetesjournals.org Ramzy, Asadi, and Kieffer 1455

PC2 Immunoreactivity Was Not Detected in Rat b-Cells PC2 Did Not Play a Significant Role in the Processing of Using Three Validated Antibodies, and EndoC-bH1 Proinsulin in Human b-Cells Cells Have PC2 Given that PC2 is ;80–100 times more catalytically active We immunostained pancreas from pig, rat, and dog (Fig. than PC1/3 (29), we sought to validate histological find- 2A). Unexpectedly, b-cells were not immunoreactive for ings with sensitive biochemical approaches to assess PC2 in rat pancreas. b-Cells in pig and dog pancreas whether human b-cells have low yet catalytically relevant were immunoreactive for PC2 using two antibodies, but levels of PC2. Given that neighboring a-cells within human one antibody failed to produce immunoreactivity in any islets require PC2 to produce glucagon from proglucagon islet cells of these species, likely due to incompatible (13), we assessed both proinsulin processing and proglu- epitopes (Supplementary Table 1). There is an R-to-G cagon processing in pulse-chase experiments. Importantly, and N-to-H substitution in pig PC2 in the immunizing we chose to study primary human islets in order to peptide sequence used to generate the PC2 COOH-terminal specifically answer the question as to how human antibody and a D-to-N substitution in dog PC2 in the likely b-cells process proinsulin. It seems that PC2 can access epitope of the PC2 NH2-terminal antibody (Supplementary and cleave human proinsulin (30), but studying the ability Table 1). of PC2 to process human proinsulin in models like EndoC- We investigated the presence of PC2 in the human bH1 cells or other cell lines would not clarify the important b-cell line EndoC-bH1 (27). We used the PC2 knockout question of how proinsulin is processed in bona fide aTC1DPC2 mouse a-cell line, a-TC1 mouse a-cell line that human b-cells.Inhumanislets,therewasanear-total has PC2 nearly exclusively, and b-TC3 mouse b-cell line blockage of glucagon production by all PC2-inhibiting fi with both PC2 and PC1/3 as controls. EndoC-bH1 cells are chase conditions (Fig. 4A) but no signi cant difference in immunoreactive for both PC1/3 and PC2 (Fig. 2B). Tran- proinsulin processing (Fig. 4B). To determine if PC1/3 is script levels of PCSK1 and PCSK2 are comparable in predominantly responsible for proinsulin processing in human b-cells, we used two PC1/3 inhibitors. Neither EndoC-bH1 cells, similar to human islets (Fig. 2C). We inhibitor significantly impaired the processing of proglu- investigated protein levels of PC1/3 and PC2 by Western cagon, but both led to a significant blockade of proinsulin blots on EndoC-bH1 cells and human islets and confirm processing. histological findings of comparable levels of PC1/3 and PC2 in EndoC-bH1 cells (Fig. 2D). The major PC1/3 band Patients With Type 2 Diabetes Had Increased b-Cell was at the expected size of mature PC1/3 (66 kDa; labeled PC2 and Increased a-Cell PC1 with pink arrowhead in Fig. 2D), and minor bands were at Given the known defects in proinsulin processing in the expected sizes of the PC1/3-processing precursors patients with diabetes, we immunostained pancreata from (74 kDa and 87 kDa) (28). Additionally, we analyzed sorted donors with type 1 (n 5 7) and type 2 (n 5 22) diabetes human b-cells and generated four to eight times enrich- and compared them to those of a new cohort of control 5 ment of b-cells (based on Ins DCt) and .99% removal of subjects without diabetes (n 10). We again observed a-cells (based on Gcg DCt; data not shown). Using highly minimal colocalization of PC2 and insulin in control subjects, but there was some heterogeneity, with 1 out of limited sample sizes of protein extracts for Western blot, 10 donors having occasional colocalization of PC2 and there was no visible PC2 band in purified b-cells, whereas insulin. This contrasts with observations that almost all whole islets from the same donors yielded clear PC2 bands donors with type 2 diabetes (20 out of 22) had obvious (Supplementary Fig. 5). insulin and PC2 colocalization (see high-magnification fi b PCSK2 insets in blue boxes in Fig. 5A). We then quanti ed the Human -Cells Had Less Than Neighboring fl a-Cells, Unlike Mouse b-Cells uorescent intensity of PC1/3 or PC2 immunoreactivity in b b To localize PCKS2 mRNA within pancreatic islets, we per- -cells (using insulin immunoreactivity to designate -cell a formed in situ hybridization on sections of human and area) and -cells (using glucagon immunoreactivity to a mouse pancreas (Fig. 3). With overlaying immunofluores- designate -cell area). Pancreas from donors with type 2 diabetes had higher PC1/3 and PC2 immunoreactivity in cence for insulin and glucagon, intense PCSK2 signal within both b-cells and a-cells (Fig. 5B). Furthermore, the in- human islets was localized to a-cells, with a marginally tensity of PC1/3 immunoreactivity in a-cells relative to detectable signal in b-cells (Fig. 3A). This contrasts to the b-cells was significantly higher in donors with type 2 di- expected pan-islet signal in mouse islets (Fig. 3B). We abetes, and the intensity of PC2 immunoreactivity in prioritized generating robust, sensitive signal in human b-cells relative to that in a-cells was also significantly pancreas and note that this led to substantial background higher in donors with type 2 diabetes compared with signal in mouse pancreas. By assessing the pattern of donors without diabetes (Fig. 5C). There was no predictive chromogenic deposition, positive cytoplasmic signal (with value of diabetes duration, HbA1c, BMI, sex, use of met- discernable nuclear holes) in mouse pancreatic islets can be formin (n 5 15) versus solely lifestyle intervention (n 5 7), clearly separated from background signal on the edges of age, or cold ischemia time of organ collection to predict acinar lobules. outcome variables of PC immunoreactivity (data not 1456 Human b-Cells Process Proinsulin With PC1/3 Diabetes Volume 69, July 2020

Figure 2—PC2 immunoreactivity was not detected in rat b-cells using three validated antibodies, and EndoC-bH1 cells have PC2. A: Representative immunofluorescence for insulin (INS) and PC2 using three antibodies in pig, rat, and dog pancreas (n 5 1–3). One antibody produces no immunofluorescence in pig pancreas (PC2 COOH-terminal [C-term], PA1-058) or dog pancreas (PC2 NH2-terminal [N-term], PA5-14595) likely due to incompatible epitopes (see Supplementary Table 1 and Supplementary Key Resources Table for antibody details). Scale bar, 100 mm. B: Immunostaining for PC1/3 and PC2 in human EndoC-bH1, PC2 knockout mouse aTC1DPC2, PC2-expressing mouse aTC-1, and mouse bTC-3 cell lines. Scale bar, 50 mm. C: Relative expression of PCSK1 and PCSK2 in EndoC-bH1 cells and whole human islets by quantitative PCR. 22DCt presented in box-and-whisker plots with GAPDH and ACTB as reference genes. D: Western blot for PC2 (MAB6018, red, R&D Systems; and NH2-terminal, PA5-14595, green, Thermo Fisher Scientific) and PC1/3 (gift from Dr. Lakshmi Devi, AB_2665530; pink arrowhead at 66 kDa, mature PC1/3) in human islet and EndoC-bH1 cell lysates with a recombinant human PC2 (rhPC2) control and with a-tubulin (a-tub) (T9026; Sigma-Aldrich) used as a loading control. Quantification presented on the right. diabetes.diabetesjournals.org Ramzy, Asadi, and Kieffer 1457

Figure 3—Human b-cells had less PCSK2 than neighboring a-cells, unlike mouse b-cells. PCSK2 detected in human (A) and mouse (B) pancreas by in situ hybridization using digoxigenin-labeled probes and detected using alkaline phosphatase–conjugated antibody with nitro- blue tetrazolium and 5-bromo-4-chloro-39-indolyphosphate substrate. After in situ hybridization, sections were immunostained for insulin (red) and glucagon (green) and overlaid with the in situ hybridization image. Representative images shown (n 5 3). Scale bars, 100 mm. Insets are enlarged 34.

shown). We also immunostained pancreas from donors immunoreactivity in human b-cells (31) and, in one case, with type 1 diabetes (n 5 7). Regrettably, we were unable interpreted the finding as a sign of b-cell dysfunction after to clearly identify insulin-immunoreactive b-cells in tissue transplantation into immunodeficient rodents (11). Yet sections. Abundantly identifiable glucagon-immunoreactive others have reported abundant PC2 immunoreactivity in a-cells in pancreas from donors with type 1 diabetes had human b-cells (32), but that work used a polyclonal an- PC1/3 and PC2 immunoreactivity similar to that of control tibody (33) with an immunizing peptide containing sig- pancreas (data not shown). Furthermore, in sections from nificant homology between human PC1/3 (RRDELEE) and donors with type 1 diabetes, we observed frequent PC2- PC2 (KKEELEE), raising the possibility of significant cross- immunoreactive cells that were not immunoreactive for reactivity with PC1/3. Likewise, prior detection of PC2 insulin or glucagon (Fig. 5A). immunoreactivity in rat b-cells by immunohistochemistry (34) and immunogold electron microscopy (35) relied on DISCUSSION polyclonal antibodies generated with large PC2 immuno- The prevailing theory that proinsulin processing requires gens containing regions of high homology to PC1/3 (PC2: both PC1/3 and PC2 is best supported by studies of islets TNACEGKEN vs. PC1/3: TRACEGQEN [34] or identical from PC2 and PC1/3 knockout mice (2,3). To our knowl- sequences FALALEAN and LTWRDMQHL in both PC2 and edge, the roles of PC1/3 and PC2 for processing human PC1/3 [35]). By Western blot using an antisera to a large proinsulin have never been rigorously examined. We fragment of PC2 (D174-S384) containing substantial performed immunohistofluorescence with well-validated regions of homology to PC1/3, rat insulinoma granules antibodies and made the surprising observation of virtu- have immunoreactivity for PC2 (36). However, using ally absent PC2 immunoreactivity in both human and rat a specific antibody with an immunogen similar to that b-cells. Interestingly, others have observed minimal PC2 of one of the antibodies we use in the current study 1458 Human b-Cells Process Proinsulin With PC1/3 Diabetes Volume 69, July 2020

rat islets (41), suggesting that rat b-cells may not depend on the function of PC2. The processing of proinsulin in rat b-cells is worth studying in greater detail in the future, and rat islets may be an excellent model for studying human proinsulin processing. Unlike in human and rat pancreas, abundant PC2 immunoreactivity was evident in b-cells from mouse, pig, and dog pancreas. In these three species, the C-A junction of proinsulin has PC1/3-unfavorable amino acids at the P4 position (mINS2, Q; pigINS, P; and dogINS, L) (42), while both rat insulins have a PC1/3-favorable R (39). Mouse pro–islet amyloid polypeptide (pro-IAPP) requires PC2 to process its NH2-terminal site (43). Mouse, dog, and pig NH2-terminal pro-IAPP all have the amino acid M in the P4 position. By contrast, both rat and human have a V at P4. We propose that the P4 M at the NH2-terminal processing site of proIAPP is unfavorable for processing by PC1/3 and a contributor to the species differences in b-cell PC2 expression. This is supported by observations that rat INS2 with a P4 M at the B-C junction of proinsulin is unique among rodent insulins in that it is processed at the CA junction before the B-C junction (5), perhaps because rat b-cells lack PC2 and PC1/3 is slow to process the B-C 2 2 junction. Additionally, in PC2 / mice that possess a human IAPP transgene, there is no increase in the ratio of circulating NH2-proIAPP1-48 to IAPP and no increased amyloid formation (44), suggesting no defect in human pro-IAPP processing in the absence of PC2. Taken to- gether, our findings of the species differences of PC2 immunoreactivity in b-cells suggest that PC2 is essential for the processing of NH2-terminal pro-IAPP when there is aP4M. In addition to examining protein levels of PC2, we assessed mRNA levels of PCSK2 in human pancreas. Our detection of low-level PCSK2 in human b-cells by in situ hybridization aligns with single-cell RNA-sequencing experiments (45,46) reporting detectable PCSK2 in b — b human -cells, albeit at much lower levels than in Figure 4 Human -cells processed proinsulin without PC2 func- a tion but required the function of PC1/3, unlike PC2-dependent -cells,unlikeinmiceinwhichPCSK2 is abundant in both a-cells. Proportion of 3H-labeled processed proglucagon into glu- a-cells and b-cells (47). Occasional intense PCSK2 signals cagon (A) and proinsulin into insulin (B) in human islet lysates after in non–a-cells within human islets are likely d-cells, m 3 40-min pulse in 800 Ci/mL H-Leu and 120-min chase in various given the known role for PC2 in excising somatostatin-14 conditions. Specific organ donors are designated by unique sym- bols. Groups were compared by one-way ANOVA with Tukey post (26). Detectable PCSK2 mRNA in conjunction with a hoc test (*P , 0.05; ***P , 0.001). BFA, brefeldin A. lack of PC2 immunoreactivity suggests that there is post- transcriptional regulation of PCSK2 in human b-cells— a possibility warranting follow-up. We also detected immunoreactivity for the essential PC2 cofactor 7B2 (gene (PC2C-term), there seems to be little PC2 in rat b-cells at SCG5) in human b-cells, albeit at lower intensity than in ;95% purity (37). Given that rat insulin 2 is processed human a-cells. These results are not surprising, given first at the C-A junction (38), a site favorable for PC1/3 high SCG5 mRNA levels in sorted human b-cells (10). 1 1 (RQKR) (39), the theory is supported that PC1/3 could be Though all brain PCSK2 cells are also SCG5 ,thereare 1 exclusively responsible for proinsulin processing in rat, many SCG5 cells that lack PCSK2 (48). 7B2 has been though the possibility is not ruled out that M at the P4 shown to suppress aggregation of b-amyloid in cell lines position (with the system of Schechter and Berger [40] for and synuclein in cell-free experiments (49). Further, 7B2 denoting positions prior to [Px] or after [P9x] the scissile has been shown to suppress cytotoxicity and aggregation bond) is the cause for slower B-C junction processing. of IAPP in culture (50) and could have other roles in Additionally, only PC1/3 is regulated by high glucose in human b-cells. diabetes.diabetesjournals.org Ramzy, Asadi, and Kieffer 1459

Figure 5—b-Cells had increased PC2 immunoreactivity in pancreas from donors with type 2 diabetes. A: Immunostaining for PC2 and PC1/3 in adult human pancreas from donors with type 1 diabetes mellitus (T1DM) (n 5 7), type 2 diabetes mellitus (T2DM) (n 5 22), or no history of diabetes (n 5 10). Representative images shown. Scale bars, 100 mm (insets [blue boxes] are enlarged 34). B: Fluorescent intensity (arbitrary units [a.u.]) of PC1/3 or PC2 immunoreactivity in b-cells (insulin-immunoreactive area) and in a-cells (glucagon-immunoreactive area). C: Relative ﬂuorescent intensity of PC1/3 in a-cells compared with b-cells and of PC2 in b-cells compared with a-cells in pancreas donors. Groups compared by Student t test (*P , 0.05; **P , 0.01; ***P , 0.001).

It is notable that convertase expression can be regu- immaturity of the cells, as PC2 is expressed during fetal lated. In rodent models of hyperglycemia, there can be development of early endocrine cells (58). Further inves- a compensatory upregulation of PC1/3 in a-cells (51,52), tigation into the dynamic regulation of PC1/3 and PC2 in which can increase intraislet production of GLP-1 and not only rodent models but also human b-cells under reduce rates of apoptosis in b-cells (53,54). PC2-deficient normal and pathological conditions is warranted. mouse a-cells used to generate the aTC1DPC2 cells spon- By pulse chase, human b-cells did not have significant taneously produce PC1/3 and begin excising GLP-1 from impairment of proinsulin processing in conditions capable 2 2 proglucagon (55). Intestinal L cells of PC1/3 / mice of fully blocking the PC2-dependent processing of proglu- abnormally produce mature glucagon, likely attributable to cagon to glucagon. Based on our immunostaining experi- PC2 action (56). Potentially, a similar initiation of PC2 ments, if b-cells do contain any PC2, it is far less than the expression occurs in b-cells of humans with PCSK1 muta- PC2 content of a-cells, thus making blockade of proglu- tions who present with hyperglycemia and elevated circu- cagon processing a stringent positive control of PC2 in- lating des-64,65 proinsulin (57). Additionally, expression hibition. It is unsurprising that temperature blockade of PC2 in the human EndoC-bH1 cell line could reflect at 20°C and brefeldin A had variable blockade in the 1460 Human b-Cells Process Proinsulin With PC1/3 Diabetes Volume 69, July 2020 processing of proinsulin because both block shuttling of there were defects in PC immunoreactivity as an explana- proinsulin into secretory granules and temperature block- tion for defective proinsulin processing in diabetes. In- ade have been demonstrated to partially supress proneuro- terestingly, there was an increase in both PC1/3 and PC2 peptide Y processing by PC1/3, albeit to a lesser degree immunoreactivity in b-cells of donors with type 2 diabetes. than by PC2 (14). Additionally, given that we can only Notwithstanding challenges of substantial heterogeneity confirm robust PC2 inhibition based on blockade of pro- and a lack of data on circulating proinsulin in donors glucagon processing but cannot confirm a lack of PC1/3 studied, we hypothesize that increased PC1/3 immunore- suppression, nonsignificant yet numerical decreases in activity could function as a compensatory response to proinsulin processing in PC2-inhibited conditions may increased insulin-production demands during type 2 di- be best attributed to modest reductions in PC1/3 activity. abetes. Further, in conjunction with adaptive increased Both PC2 inhibitors used in pulse-chase experiments b-cell PC1/3, aberrant upregulation of PC2 may contribute function as allosteric inhibitors and have been shown to to incomplete proinsulin processing. Additionally, we ob- inhibit PC1/3 by 10–20% when used in cell-free experi- served increased a-cell PC1/3 in donors with type 2 di- ments at lower concentrations (10 mmol/L [18] and abetes, confirming previous findings (64) and indirectly 25 mmol/L [17]). Notably, b-cells are sensitive to PC1/3 suggesting increased islet production of GLP-1 in patients inhibition, as PC1/3 heterozygous mouse b-cells with with type 2 diabetes. a likely #50% loss of PC1/3 (and abundant PC2) have Clarifying the process of proinsulin processing is impor- modestly impaired proinsulin processing (3). Additionally, tant to understand defective prohormone processing during 2 2 PC2 / mouse b-cells have a substantially worse im- diabetes. The current work challenges the prevailing as- paired proinsulin processing phenotype than human sumption that human b-cells process proinsulin by PC1/3 islets exposed to robust PC2 inhibitors. In contrast to and then PC2 sequentially and reveals that abnormal PC2 PC2-inhibited conditions, PC1/3-inhibited conditions expression in human b-cells is associated with type 2 diabetes. that did not impair proglucagon processing significantly Though evidence supports the dogma that PC1/3 and then impaired proinsulin processing. We cannot confirm the PC2 process proinsulin in mouse b-cells, our findings suggest extent of PC1/3 inhibition, and incomplete PC1/3 inhibi- PC2 has little to no role in healthy human b-cells. tion is a potential explanation for the incomplete blockade of proinsulin processing. Regardless, these findings provide validation for our experimental technique and clearly Acknowledgments. The authors express their gratitude to organ donors and their families. Human samples for research were provided by the Alberta show that PC1/3 is important to the processing of Diabetes Institute IsletCore at the University of Alberta in Edmonton (http://www proinsulin. .bcell.org/human-islets.html) with the assistance of the Human Organ Pro- The pulse-chase study design was limited by a lack of a curement and Exchange (HOPE) program, Trillium Gift of Life Network (TGLN), molecular knockdown approach for PCSK2 (e.g., lentiviral and other Canadian organ procurement organizations. The authors also thank delivery of shRNA) (59). Though the specificity of a mo- Drs. Allan Cherrington (Vanderbilt University), Gunilla Westermark (Uppsala lecular approach is appealing and may be a worthwhile University), and Lakshmi Devi (Mount Sinai School of Medicine) for sharing follow-up study, the confounding effects of islet disper- antibodies and Drs. Lucy Marzban and C. Bruce Verchere (The University of sion, cell reaggregation, and prolonged culture to allow British Columbia [UBC]) for their helpful advice. A.R. gratefully acknowledges time for mRNA knockdown and PC2 protein turnover are studentship support from the Canadian Institutes of Health Research (CIHR) considerable. Another important limitation is that as we (Vanier Canada Graduate Scholarship) and Vancouver Coastal Health (CIHR-UBC were unable to differentiate intact proinsulin from pro- MD/PhD Studentship). insulin processing intermediates using size-based proin- Funding. This work was supported by grants from the Canadian Diabetes sulin/insulin separation by SDS-PAGE, we were unable to Association and the CIHR (CIHR Foundation Scheme). Duality of Interest. No potential conflicts of interest relevant to this article rule out the potential that some single-site processing of were reported. proinsulin is occurring in PC1/3-inhibited conditions. Author Contributions. A.R. and A.A. performed experiments. A.R. Given relatively normal production of fully mature insulin analyzed data and drafted the manuscript with contributions from A.A. and (necessarily processed at both the B-C and C-A junctions T.J.K. All authors approved the final draft. T.J.K. is the guarantor of this work to liberate C-peptide and reduce the molecular mass to and, as such, had full access to all the data in the study and takes responsibility for ;6 kDa) in PC2-inhibited conditions that were sufficient the integrity of the data and the accuracy of the data analysis. for an ;100% blockade of proglucagon processing, this References appears unlikely. Impaired proinsulin processing has prognostic value for 1. Steiner DF, Cunningham D, Spigelman L, Aten B. Insulin biosynthesis: evidence for a precursor. Science 1967;157:697–700 progression from autoantibody positivity to type 1 diabetes 2. Furuta M, Carroll R, Martin S, et al. Incomplete processing of proinsulin to (60) and from impaired glucose tolerance to type 2 diabetes insulin accompanied by elevation of Des-31,32 proinsulin intermediates in islets of (61). Some patients with type 1 diabetes lose detectable mice lacking active PC2. J Biol Chem 1998;273:3431–3437 circulating C-peptide but retain detectable proinsulin (62) 3. Zhu X, Orci L, Carroll R, Norrbom C, Ravazzola M, Steiner DF. Severe block in and pro-IAPP (63). With new insight into the processing of processing of proinsulin to insulin accompanied by elevation of des-64,65 pro- prohormones in healthy b-cells, we immunostained pan- insulin intermediates in islets of mice lacking prohormone convertase 1/3. Proc creas from organ donors with diabetes to determine if Natl Acad Sci U S A 2002;99:10299–10304 diabetes.diabetesjournals.org Ramzy, Asadi, and Kieffer 1461

4. Rhodes CJ, Lincoln B, Shoelson SE. Preferential cleavage of des-31,32- 25. Bukowski TR, Moffett TC, Revkin JH, Ploger JD, Bassingthwaighte JB. Triple- proinsulin over intact proinsulin by the insulin secretory granule type II endo- label beta liquid scintillation counting. Anal Biochem 1992;204:171–180 peptidase. Implication of a favored route for prohormone processing. J Biol Chem 26. Furuta M, Yano H, Zhou A, et al. Defective prohormone processing and 1992;267:22719–22727 altered pancreatic islet morphology in mice lacking active SPC2. Proc Natl Acad Sci 5. Halban PA. Proinsulin processing in the regulated and the constitutive se- U S A 1997;94:6646–6651 cretory pathway. Diabetologia 1994;37 (Suppl. 2):S65–S72 27. Ravassard P, Hazhouz Y, Pechberty S, et al. A genetically engineered human 6. Stijnen P, Ramos-Molina B, O’Rahilly S, Creemers JW. PCSK1 mutations and pancreatic b cell line exhibiting glucose-inducible insulin secretion. J Clin Invest human endocrinopathies: from obesity to gastrointestinal disorders. Endocr Rev 2011;121:3589–3597 2016;37:347–371 28. Muller L, Lindberg I. The cell biology of the prohormone convertases PC1 and 7. Zheng X, Ren W, Zhang S, et al. Association of type 2 diabetes susceptibility PC2. Prog Nucleic Acid Res Mol Biol 1999;63:69–108 genes (TCF7L2, SLC30A8, PCSK1 and PCSK2) and proinsulin conversion in 29. Lamango NS, Zhu X, Lindberg I. Purification and enzymatic characterization a Chinese population. Mol Biol Rep 2012;39:17–23 of recombinant prohormone convertase 2: stabilization of activity by 21 kDa 7B2. 8. Scharfmann R, Pechberty S, Hazhouz Y, et al. Development of a condi- Arch Biochem Biophys 1996;330:238–250 tionally immortalized human pancreatic b cell line. J Clin Invest 2014;124: 30. Kaufmann JE, Irminger JC, Mungall J, Halban PA. Proinsulin conversion in 2087–2098 GH3 cells after coexpression of human proinsulin with the endoproteases PC2 and/ 9. Benner C, van der Meulen T, Caceres E, Tigyi K, Donaldson CJ, Huising MO. or PC3. Diabetes 1997;46:978–982 The transcriptional landscape of mouse beta cells compared to human beta cells 31. Lam CJ, Chatterjee A, Shen E, Cox AR, Kushner JA. Low-level insulin content reveals notable species differences in long non-coding RNA and protein-coding within abundant non-b islet endocrine cells in long-standing type 1 diabetes. gene expression. BMC Genomics 2014;15:620 Diabetes 2019;68:598–608 10. Bramswig NC, Everett LJ, Schug J, et al. Epigenomic plasticity enables 32. Teitelman G. Heterogeneous expression of proinsulin processing enzymes in human pancreatic a to b cell reprogramming. J Clin Invest 2013;123:1275–1284 beta cells of non-diabetic and type 2 diabetic humans. J Histochem Cytochem 11. Davalli AM, Perego L, Bertuzzi F, et al. Disproportionate hyperproinsulinemia, 2019;67:385–400 beta-cell restricted prohormone convertase 2 deficiency, and cell cycle inhibitors 33. Scopsi L, Gullo M, Rilke F, Martin S, Steiner DF. Proprotein convertases (PC1/ expression by human islets transplanted into athymic nude mice: insights into PC3 and PC2) in normal and neoplastic human tissues: their use as markers of nonimmune-mediated mechanisms of delayed islet graft failure. Cell Transplant neuroendocrine differentiation. J Clin Endocrinol Metab 1995;80:294–301 2008;17:1323–1336 34. Nie Y, Nakashima M, Brubaker PL, et al. Regulation of pancreatic PC1 and 12. Itoh Y, Tanaka S, Takekoshi S, Itoh J, Osamura RY. Prohormone convertases PC2 associated with increased glucagon-like peptide 1 in diabetic rats. J Clin (PC1/3 and PC2) in rat and human pancreas and islet cell tumors: subcellular Invest 2000;105:955–965 immunohistochemical analysis. Pathol Int 1996;46:726–737 35. Tanaka S, Kurabuchi S, Mochida H, et al. Immunocytochemical localization of 13. Rouillé Y, Martin S, Steiner DF. Differential processing of proglucagon by prohormone convertases PC1/PC3 and PC2 in rat pancreatic islets. Arch Histol the subtilisin-like prohormone convertases PC2 and PC3 to generate either Cytol 1996;59:261–271 glucagon or glucagon-like peptide. J Biol Chem 1995;270:26488–26496 36. Bennett DL, Bailyes EM, Nielsen E, et al. Identification of the type 2 proinsulin 14. Paquet L, Zhou A, Chang EY, Mains RE. Peptide biosynthetic processing: processing endopeptidase as PC2, a member of the eukaryote subtilisin family. J distinguishing prohormone convertases PC1 and PC2. Mol Cell Endocrinol 1996; Biol Chem 1992;267:15229–15236 120:161–168 37. Neerman-Arbez M, Cirulli V, Halban PA. Levels of the conversion endo- 15. Marzban L, Trigo-Gonzalez G, Verchere CB. Processing of pro-islet amyloid proteases PC1 (PC3) and PC2 distinguish between insulin-producing pancreatic polypeptide in the constitutive and regulated secretory pathways of beta cells. Mol islet beta cells and non-beta cells. Biochem J 1994;300:57–61 Endocrinol 2005;19:2154–2163 38. Sizonenko SV, Halban PA. Differential rates of conversion of rat proinsulins I 16. Helms JB, Rothman JE. Inhibition by brefeldin A of a Golgi membrane enzyme and II. Evidence for slow cleavage at the B-chain/C-peptide junction of proinsulin II. that catalyses exchange of guanine nucleotide bound to ARF. Nature 1992;360: Biochem J 1991;278:621–625 352–354 39. Jean F, Boudreault A, Basak A, Seidah NG, Lazure C. Fluorescent peptidyl 17. Yongye AB, Vivoli M, Lindberg I, Appel JR, Houghten RA, Martinez-Mayorga substrates as an aid in studying the substrate specificity of human prohormone K. Identification of a small molecule that selectively inhibits mouse PC2 over convertase PC1 and human furin and designing a potent irreversible inhibitor. J mouse PC1/3: a computational and experimental study. PLoS One 2013;8: Biol Chem 1995;270:19225–19231 e56957 40. Schechter I, Berger A. On the size of the active site in proteases. I. Papain. 18. Vivoli M, Caulfield TR, Martínez-Mayorga K, Johnson AT, Jiao GS, Lindberg I. Biochem Biophys Res Commun 1967;27:157–162 Inhibition of prohormone convertases PC1/3 and PC2 by 2,5-dideoxystreptamine 41. Alarcón C, Lincoln B, Rhodes CJ. The biosynthesis of the subtilisin-related derivatives. Mol Pharmacol 2012;81:440–454 proprotein convertase PC3, but no that of the PC2 convertase, is regulated by 19. Becker GL, Lu Y, Hardes K, et al. Highly potent inhibitors of proprotein glucose in parallel to proinsulin biosynthesis in rat pancreatic islets. J Biol Chem convertase furin as potential drugs for treatment of infectious diseases. J Biol 1993;268:4276–4280 Chem 2012;287:21992–22003 42. Ledgerwood EC, Brennan SO, Birch NP, George PM. The specificity of the 20. Shapiro AM, Ricordi C, Hering BJ, et al. International trial of the Edmonton neuroendocrine convertase PC3 is determined by residues NH2- and COOH- protocol for islet transplantation. N Engl J Med 2006;355:1318–1330 terminal to the cleavage site. Biochem Mol Biol Int 1996;39:1167–1176 21. Asadi A, Bruin JE, Kieffer TJ. Characterization of antibodies to products of 43. Wang J, Xu J, Finnerty J, Furuta M, Steiner DF, Verchere CB. The prohormone proinsulin processing using immunofluorescence staining of pancreas in multiple convertase enzyme 2 (PC2) is essential for processing pro-islet amyloid poly- species. J Histochem Cytochem 2015;63:646–662 peptide at the NH2-terminal cleavage site. Diabetes 2001;50:534–539 22. Banerjee M, Otonkoski T. A simple two-step protocol for the purification of 44. Courtade JA, Wang EY, Yen P, et al. Loss of prohormone convertase human pancreatic beta cells. Diabetologia 2009;52:621–625 2 promotes beta cell dysfunction in a rodent transplant model expressing human 23. Schägger H. Tricine-SDS-PAGE. Nat Protoc 2006;1:16–22 pro-islet amyloid polypeptide. Diabetologia 2017;60:453–463 24. Dyballa N, Metzger S. Fast and sensitive colloidal Coomassie G-250 staining 45. Muraro MJ, Dharmadhikari G, Grun D, et al. A single-cell transcriptome atlas for proteins in polyacrylamide gels. J Vis Exp 2009;30:e1431 of the human pancreas. Cell Syst 2016;3:385–394.e3 1462 Human b-Cells Process Proinsulin With PC1/3 Diabetes Volume 69, July 2020

46. Segerstolpe Å, Palasantza A, Eliasson P, et al. Single-cell transcriptome accompanied by differential processing of proglucagon and improved glucose profiling of human pancreatic islets in health and type 2 diabetes. Cell Metab 2016; homeostasis in mice. Diabetes 2007;56:2744–2752 24:593–607 56. Ugleholdt R, Zhu X, Deacon CF, Ørskov C, Steiner DF, Holst JJ. Impaired 47. Xin Y, Kim J, Ni M, et al. Use of the Fluidigm C1 platform for RNA sequencing of intestinal proglucagon processing in mice lacking prohormone convertase 1. single mouse pancreatic islet cells. Proc Natl Acad Sci U S A 2016;113:3293–3298 Endocrinology 2004;145:1349–1355 48. Seidel B, Dong W, Savaria D, Zheng M, Pintar JE, Day R. Neuroendocrine 57. Jackson RS, Creemers JW, Ohagi S, et al. Obesity and impaired prohormone protein 7B2 is essential for proteolytic conversion and activation of proprotein processing associated with mutations in the human prohormone convertase convertase 2 in vivo. DNA Cell Biol 1998;17:1017–1029 1 gene. Nat Genet 1997;16:303–306 49. Helwig M, Hoshino A, Berridge C, et al. The neuroendocrine protein 7B2 58. Ramond C, Beydag-Tasöz BS, Azad A, et al. Understanding human fetal suppresses the aggregation of neurodegenerative disease-related proteins. J Biol pancreas development using subpopulation sorting, RNA sequencing and single- Chem 2013;288:1114–1124 cell profiling. Development 2018;145:1–15 50. Peinado JR, Sami F, Rajpurohit N, Lindberg I. Blockade of islet amyloid 59. Harata M, Liu S, Promes JA, Burand AJ, Ankrum JA, Imai Y. Delivery of shRNA polypeptide fibrillation and cytotoxicity by the secretory chaperones 7B2 and via lentivirus in human pseudoislets provides a model to test dynamic regulation of proSAAS. FEBS Lett 2013;587:3406–3411 insulin secretion and gene function in human islets. Physiol Rep 2018;6:e13907 51. Kilimnik G, Kim A, Steiner DF, Friedman TC, Hara M. Intraislet production of 60. Truyen I, De Pauw P, Jørgensen PN, et al.; Belgian Diabetes Registry. GLP-1 by activation of prohormone convertase 1/3 in pancreatic a-cells in mouse Proinsulin levels and the proinsulin:C-peptide ratio complement autoantibody models of ß-cell regeneration. Islets 2010;2:149–155 measurement for predicting type 1 diabetes. Diabetologia 2005;48:2322–2329 52. O’Malley TJ, Fava GE, Zhang Y, Fonseca VA, Wu H. Progressive change of 61. Kahn SE, Leonetti DL, Prigeon RL, Boyko EJ, Bergstrom RW, Fujimoto WY. intra-islet GLP-1 production during diabetes development. Diabetes Metab Res Proinsulin as a marker for the development of NIDDM in Japanese-American men. Rev 2014;30:661–668 Diabetes 1995;44:173–179 53. Zummo FP, Cullen KS, Honkanen-Scott M, Shaw JAM, Lovat PE, Arden C. 62. Steenkamp DW, Cacicedo JM, Sahin-Efe A, Sullivan C, Sternthal E. Glucagon-like peptide 1 protects pancreatic b-cells from death by increasing Preserved proinsulin secretion in long-standing type 1 diabetes. Endocr Pract 2017;23: autophagic flux and restoring lysosomal function. Diabetes 2017;66:1272–1285 1387–1393 54. Wideman RD, Yu IL, Webber TD, et al. Improving function and survival of 63. Courtade JA, Klimek-Abercrombie AM, Chen YC, et al. Measurement of pro- pancreatic islets by endogenous production of glucagon-like peptide 1 (GLP-1). islet amyloid polypeptide (1-48) in diabetes and islet transplants. J Clin Endocrinol Proc Natl Acad Sci U S A 2006;103:13468–13473 Metab 2017;102:2595–2603 55. Wideman RD, Covey SD, Webb GC, Drucker DJ, Kieffer TJ. A switch from 64. Marchetti P, Lupi R, Bugliani M, et al. A local glucagon-like peptide 1 (GLP-1) prohormone convertase (PC)-2 to PC1/3 expression in transplanted alpha-cells is system in human pancreatic islets. Diabetologia 2012;55:3262–3272