Stathmin-2 Mediates Glucagon Secretion From Pancreatic α-Cells

Inhibition of glucagon hypersecretion from pancreatic α-cells is an appealing strategy for the treatment of diabetes. Our hypothesis is that proteins that associate with glucagon within alpha cell secretory granules will regulate glucagon secretion, and may provide druggable targets for controlling abnormal glucagon secretion in diabetes. Recently, we identified a dynamic glucagon interactome within the secretory granules of the α cell line, αTC1-6, and showed that select proteins within the interactome could modulate glucagon secretion. In the present study, we show that one of these interactome proteins, the neuronal protein stathmin-2, is expressed in αTC1-6 cells and in mouse pancreatic alpha cells, and is a novel regulator of glucagon secretion. The secretion of both glucagon and Stmn2 was significantly enhanced in response to 55 mM K+, and immunofluorescence confocal microscopy showed co-localization of stathmin-2 with glucagon and the secretory granule markers chromogranin A and VAMP-2 in αTC1-6 cells. In mouse pancreatic islets, Stathmin-2 co-localized with glucagon, but not with insulin, and co-localized with secretory pathway markers. To show a function for stathmin-2 in regulating glucagon secretion, we showed that siRNA—mediated depletion of stathmin-2 in αTC1-6 cells caused glucagon secretion to become constitutive without any effect on proglucagon mRNA levels, while overexpression of stathmin-2 completely abolished both basal and K+-stimulated glucagon secretion. Overexpression of stathmin-2 increased the localization of glucagon into the endosomal-lysosomal compartment, while depletion of stathmin-2 reduced the endosomal localization of glucagon. Therefore, we describe stathmin-2 as having a novel role as an alpha cell secretory granule protein that modulates glucagon secretion via trafficking through the endosomal-lysosomal system. These findings describe a potential new pathway for the regulation of glucagon secretion, and may have implications for controlling glucagon hypersecretion in diabetes.


INTRODUCTION
Hyperglucagonemia is a characteristic sign of diabetes, causing fasting hyperglycemia and glycemic volatility. Clinically, glycemic variability contributes to the development of diabetes complications (1). Persistent hyperglucagonemia may exacerbate abnormal glucose metabolism in patients with type 2 diabetes and lead to metabolic disturbances in obese and prediabetic individuals (2). Further complicating glucose homeostasis in diabetes is the direct effect of glucose on the alpha cells; while glucagon secretion is maximally suppressed at plasma glucose concentrations of 5-10 mM, it increases at glucose levels above 10 mM, thus exacerbating hyperglycemia (3)(4)(5)(6). Therefore, controlling excess glucagon secretion may be a potential therapeutic strategy for diabetes (7) so that glycemia and glucose metabolism may be better regulated. Such an approach has been suggested as a priority for the treatment of diabetes (1).
Combating hyperglucagonemia could be theoretically achieved by (i) inhibition of glucagon action at target organs by blocking the glucagon receptor, or (ii) inhibition of glucagon secretion from the pancreatic α cells. While in the shortterm the former could be an effective therapeutic strategy, it can lead to α cell hyperplasia and hyperglucagonemia over a long-term period (8), along with a risk of hypoglycemia (9) and disturbances in lipid metabolism (10). Therefore, inhibiting glucagon secretion, rather than blocking the glucagon receptor, may be a more appropriate therapeutic approach for the treatment of hyperglucagonemia of diabetes (11).
It has been documented that suppression of glucagon secretion can be mediated at the systemic, paracrine or intrinsic level (12). As a systemic modulator, GLP-1 inhibits glucagon secretion; however, there are controversies as to whether GLP-1 directly inhibits glucagon secretion from α cells by signaling through the alpha cell GLP-1R (13)(14)(15), or indirectly by increasing inter-islet somatostatin or insulin secretion (13,16). Glucagon secretion is also suppressed by paracrine signaling through the insulin, somatostatin and GABA A receptors on the α cell (16)(17)(18) At an intrinsic level, glucose directly or indirectly inhibits glucagon secretion from the α cell (19)(20)(21)(22) by altering downstream activities of Ca 2+ channels, K ATP channels (23), and trafficking of secretory granules (24).
We are pursuing the hypothesis that glucagon secretion can also be controlled by proteins within the secretory granule that associate with glucagon. By conducting secretory granule proteomics in αTC1-6 cells, we have recently described a dynamic glucagon interactome, and shown that components of this interactome can play a role in modulating glucagon secretion (25). Of these components, a protein of particular interest is Stathmin-2 (Stmn2 or SCG10), a member of the stathmin family of Golgi proteins (26) that may play a role in the regulation of neuroendocrine secretion (27). In the human islet, Stmn2 expression may be unique to alpha cells, as shown by genome-wide RNA-Seq analysis (28) and single cell transcriptomics (23), and alpha cell Stmn2 mRNA expression is differentially regulated in type 2 diabetes (29). These studies, together with our proteomics findings, led us to hypothesize that Stmn2 may function in α cells to modulate glucagon secretion. In the present study, we show that Stmn2 is co-localized with glucagon in secretory granules of αTC1-6 cells and controls glucagon secretion by trafficking through the endosomal/lysosomal system.

MATERIALS AND METHODS
Cell Culture αTC1-6 cells (a kind gift from C. Bruce Verchere, University of British Columbia, Vancouver, BC, Canada) were cultured in regular DMEM medium containing 5.6 mM glucose (Cat# 12320032, Thermo Fisher Scientific) supplemented with 15% horse serum (Cat# 26050088, Thermo Fisher Scientific), 2.5% FBS (Cat# 16000044, Thermo Fisher Scientific), L-glutamine and sodium pyruvate. For secretion experiments, cells were plated in 6 -well plates and for all experiments, a low passage number (up to P6) was used. Twenty-four hours prior to secretion experiments, media were removed and replaced with DMEM without supplements. To evaluate (25) the regulated secretion of glucagon, cells were washed twice with HBSS, preincubated for 2 h in DMEM without supplements, then incubated for 15 min with or without KCl (55 mM). These media were collected into tubes containing protease inhibitor (PMSF, 45 mM) and phosphatase inhibitors (sodium orthovanadate, 1 mM; and sodium fluoride 5 mM) while tubes were kept on ice. After collection, media were centrifuged at 13,000×g for 5 min at 4 • C, and the supernatant was collected into new microfuge tubes and immediately kept at −80 • C until analysis. After media were removed, cells were washed using ice-cold PBS (pH 7.4) and lysed in RIPA buffer (Cat# 89900, Thermo Fisher Scientific) containing abovementioned protease and phosphatase inhibitors. The lysed cells were centrifuged at 13,000×g for 5 min at 4 • C and the supernatant was kept at −80 • C for protein assays.

Gene Construct and Plasmid Preparation
To generate the expression plasmid for Stmn2, the Kozak sequence (GCCACC), signal peptide sequence and coding sequence of mouse Stmn2 (https://www.uniprot.org/uniprot/ P55821) were ligated into the NheI and ApaI restriction sites of pcDNA3.1(+) MAr. The construct was synthesized by GENEART GmbH, Life Technologies (GeneArt project 2018AAEGRC, Thermo Fisher Scientific). Then, Max Efficiency DH5α competent cells (Thermo Fisher Scientific, Cat# 18258012) were transformed according to the manufacturer's protocol. Plasmids were then extracted and purified using the PureLink HiPure Plasmid Maxiprep Kit (Thermo Fisher Scientific, Cat# K210006) for downstream experiments. Correct assembly of the final construct was verified by gene sequencing at the London Regional Genomics Facility, Western University. For Stmn2 overexpression studies, αTC1-6 cells were transiently transfected by pcDNA3.1 (+) MAr-stmn2 construct or empty vector (negative control). All transfections were done using Lipofectamine 2000 (Cat#11668-027, Invitrogen). To monitor normal cell growth and morphology, cells were checked daily by the EVOS cell imaging system (Thermo Fisher Scientific). The efficiency of transfection was determined at >70% through co-transfection with pEGFP.

siRNA-Mediated Depletion of Stathmin-2
Functional analysis of Stmn2 was done by gene silencing experiments. siRNAs targeting three regions within the Stmn2 mRNA (Cat# s73356, s73354, s73355) were chosen from pre-designed mouse siRNAs (Silencer siRNA, Thermo Fisher Scientific). The control group was treated with Mission siRNA Universal Negative Control # 1 (Cat# SIC001, Sigma-Aldrich). Gene silencing was done based on a previously published protocol (30) and as we have done previously with some modifications (25). Briefly, αTC1-6 cells were cultured in regular DMEM to 60% confluency. Media were removed and replaced with 2 mL Opti-MEM (Cat# 31985-070, Gibco) containing 50 nM pooled siRNAs with Lipofectamine 2000. After 8 h, media were changed to regular DMEM without FBS and cultured for 72 h. Then, media were refreshed and cells were cultured for 15 min in the absence or presence of 55 mM KCl as described above. Gene silencing was confirmed by analyzing mRNA expression levels of Stmn2. After removing media, cells were washed by cold PBS (pH 7.4) and total RNA was extracted (RNeasy extraction kit; Cat # 74104, Qiagen). cDNA synthesis was performed using the SuperScript III First Strand Synthesis Supermix for qRT-PCR (Cat # 11752050, Thermo Fisher Scientific), according to the supplier's protocol. Real-time PCR was performed using Quant Studio Design and Analysis Real-Time PCR Detection System in conjunction with the Maxima SYBR Green qPCR Master Mix (Cat # K0221, Thermo Fisher Scientific) using specific primers for Stmn2: forward, 5 ′ -GCAATGGCCTACAAGGAAAA-3 ′ ; reverse, 5 ′ -GGTGGCTTCAAGATCAGCTC-3 ′ ; and β-Actin; forward, 5 ′ -AGCCATGTACGTAGCCATCC-3 ′ ; reverse, 5 ′ -CTCTCAGC TGTGGTGGTGAA-3 ′ . Gene expression levels for stathmin-2 were normalized to that of β-Actin. The normalized level of transcripts in the depleted cells was shown relative to that of the non-targeting negative control. Relative expression levels were determined as percent of alterations compared to the control. Statistical analysis was performed using t-test at α = 0.05.

Mouse Pancreatic Islets
All mice were treated in accordance with the guidelines set out by the Animal Use Subcommittee of the Canadian Council on Animal Care at Western University based on the approved Animal Use Protocol AUP 2012-020. Six to eightweek old male C57BL/6 mice (n = 7) were sacrificed by cervical dislocation under anesthesia with inhalant isoflurane. Pancreata were collected and fixed in 10% buffered formalin for 3 days and treated with 70% ethanol for 1 day before paraffin embedding at the Molecular Pathology Core Facility, Robarts Research Institute, Western University. The paraffin-embedded blocks were longitudinally sectioned in 5 µm slices and fixed onto glass microscope slides. The samples were de-paraffinized by graded washes using xylene, ethanol and PBS. Background Sniper (Cat# BS966H, Biocare Medical) was used to reduce non-specific background staining. Samples were incubated with primary antibodies against glucagon (1:500), Stmn2 (1:250), insulin (Cat# ab7842, Abcam; 1:250) and TGN46 (Cat# ab16059, Abcam; 1:200) and followed by secondary antibodies of goat anti-mouse IgG Alexa Fluor 488 (1:500), donkey anti-goat IgG Alexa Fluor 555 (1:500), and goat anti-guinea pig IgG Alexa Fluor 647 (Cat# A21450, Invitrogen; 1:500). Nuclei were stained with DAPI (1:1000), and tissues were mounted in Prolong Antifade mountant (Cat# P36982, Thermo Fisher Scientific). As a background control for Stmn2, islet staining for Stmn2 was done using only the secondary antibody.

Image Acquisition
High-resolution images were acquired through a Nikon A1R Confocal microscope with a ×60 NA plan-Apochromat oil differential interference contrast objective and NIS-Elements software (Nikon, Mississauga, Canada) using a pinhole of 1 Airy unit. Images were sampled according to Nyquist criteria, and images of the Nyquist-cropped areas were captured at 1,024 × 1,024 pixel resolution, and deconvoluted by the 2D-deconvolution algorithm of the NIS-Elements software, thereby optimizing images for accurate co-localization of fluorescent signals.

Image Analysis
For cell image analysis, we prepared three coverslips for each group. Image analysis was performed by NIS-Elements software (Nikon, Mississauga, Canada), using the co-localization option and Pearson's correlation coefficient (PCC). Regions of interest (ROI) were manually drawn around distinct single or multicellular bodies, and merged values of glucagon and Stmn2 were taken for analysis. Colocalization of the pixels from each pseudo-colored image was used to calculate Pearson's correlation coefficient, as we described previously (25,31).
For mouse pancreatic islets, images were captured using four channels of green (glucagon), red (Stmn2), purple (insulin) and blue (nucleus; DAPI). To calculate the extent of co-localization between glucagon and stathmin-2 (glucagon + , Stmn2 + ), images of 15 islets per pancreas were captured and analyzed by Pearson's correlation coefficient (PCC). To this end, we manually drew ROIs around each islet and then defined PCC values for colocalization between Stmn2 and glucagon or insulin using the colocalization option of the NIS-Elements software. To predict expression levels of Stmn2 in α or β-cells of the pancreatic islets we have performed binary analysis using M-Thresholding algorithm of NIS-Elements software, followed by regression analysis of Stmn2 vs. glucagon or insulin using GraphPad Prism 7.

Primary Islet Culture
Islet preparation and culture was done according to the Li et al. (33) protocol with some modifications. Male C57BL/6 mice (n = 5-6) were euthanized by CO2. The abdominal cavity was opened and 3 mL of 1.87 mg/mL collagenase V (Cat# C9263, Sigma;) in Hanks' Balanced Salt Solution was injected into the common bile duct. The pancreas was then removed, placed into a Falcon tube containing 2 mL of the ice-cold collagenase V solution and incubated for 12 min at 37 • C with occasional shaking. Digestion was stopped by adding 1mM CaCl 2 and the cell suspension was washed twice in the CaCl 2 solution. Islets were collected into a sterile petri dish using a 70 µm cell strainer with RPMI1640 containing 11 mM glucose plus 20 mM glutamine, 10% FBS and penicillin (110 U/mL) and streptomycin (100 µg/mL). 180 islets were handpicked into the medium under a stereomicroscope and incubated for 2 h in the cell culture incubator. The medium was then changed to RPMI1640 containing 11 mM glucose plus 10% FBS and penicillin (110 U/mL) and streptomycin (100 µg/mL) and cultured overnight at 37 • C.

Islet Glucagon Secretion Experiments
Glucagon secretion from islets was measured based on the protocol by Suckow et al. (34). Briefly, islets were washed three times using Krebs-Ringer bicarbonate (KRB) buffer (135 mM NaCl, 3.6 mM KCl, 5 mM NaHCO 3 , 0.5 mM NaH 2 PO 4 , 0.5 mM MgCl 2 , 1.5 mM CaCl 2 , 10 mM HEPES; pH 7.4) containing 11 mM glucose, and then pre-incubated in this buffer for 1 h. Glucagon secretion was tested by incubating islets in KRB containing 1 mM glucose in the presence or absence of arginine (25 mM) for 20 min. Media were collected into microcentrifuge tubes containing enzyme inhibitors (PMSF, 45 mM; Aprotinin, 5 µg/mL and sodium orthovanadate, 1 mM). Samples were centrifuged at 14,000 × g for 5 min at 4 • C and the supernatant was collected and kept at −80 • C until analysis.

Measurement of Glucagon and Stathmin-2
Glucagon levels in the media were determined by ELISA (Cat # EHGCG, Thermo Fisher Scientific) according to the  manufacturer's instructions. Stmn2 levels in the media were measured using mouse stathmin-2 ELISA kit (Cat# MBS7223765, MyBioSource) according to the manufacturer's instruction. For each measurement, the values were compared between groups by t-test and among groups by 1-Way ANOVA and Bonferroni post-hoc test (α = 0.05). Cell protein levels were determined using BCA assay and used for normalization of the glucagon or Stmn2 levels.

Statistical Analysis
Values were compared among treatment groups by one-way ANOVA or between groups by unpaired t-test using Sigma Stat 3.5 software (α = 0.05). For image analysis, co-localization of channels in the merged images was calculated by Pearson's correlation coefficient (PCC) using NIS-Elements software (Nikon, Canada).

Stmn2 Co-localizes With Glucagon and Secretory Granule Markers in αTC1-6 Cells
Immunostaining of glucagon and Stmn2 in αTC1-6 cells revealed significant co-localization, as shown in Figure 1A and by a positive Pearson's correlation coefficient (0.74 ± 0.05) between endogenously expressed glucagon and Stmn2. Linear regression of binary intensities showed a sensitive and significant relationship between colocalization of glucagon and Stmn2 ( Figure 1B). The secretion of both glucagon and Stmn2 was significantly enhanced in response to 55 mM K + (Figure 1C), with corresponding decreases in cell contents ( Figure 1D).

Stathmin-2 Localizes to the α Cell Secretory Pathway in Mouse Pancreatic Islets
Immunostaining of mouse pancreatic islets showed a pattern of Stmn2 immunofluorescence similar to that of glucagon, and not insulin (Figure 2A). Analysis by Pearson correlation showed a strong colocalization between glucagon and Stmn2 in the islets (PCC = 0.77 ± 0.02), but not between insulin and Stmn2 ( Figure 2B). Linear regression analysis of the binary intensities revealed a very strong and significant relationship between Stmn2 and glucagon immunofluorescence (Figure 2C), while there was no significant relationship between Stmn2 and insulin ( Figure 2D). There was also a strong relationship between Stmn2 and the trans Golgi marker, TGN46 (PCC = 0.72 ± 0.09) in both islet clusters ( Figure 2E) and mature islets ( Figure 2F). In addition, double immunogold-labeling TEM revealed the presence of both glucagon and Stmn2 within secretory granules of pancreatic α-cells. The co-localization of glucagon (10 nm particles; white arrows) and Stmn2 (18 nm particles; black arrows) was mostly within the core area of the secretory granules ( Figure 3A). Finally, 25 mM Arg significantly enhanced the secretion of both glucagon and Stmn2 from isolated islets ( Figure 3B). These results indicate that Stmn2 is localized to the secretory pathway of α cells in mouse pancreatic islets.

Effects of Depletion and Overexpression of Stmn2 on Glucagon Secretion
In order to determine if Stmn2 had any functional effects in α-cells, we manipulated levels of Stmn2 and measured K + -stimulated glucagon secretion. Following siRNA-mediated knockdown of Stmn2 in αTC1-6 cells, basal secretion of glucagon was increased ∼4.5-fold, and was not significantly different from K + -stimulated secretion (Figure 4A), indicating increased constitutive secretion. Efficacy of siRNA-mediated depletion of Stmn2 was shown by a significant reduction (p < 0.01) in Stmn2 mRNA levels ( Figure 4B) and the Stmn2 immunoreactive band by western blot (Figure 4C). As well, silencing of Stmn2 did not affect proglucagon gene expression levels ( Figure 4D), indicating that the effects of Stmn2 depletion were on glucagon secretion alone. Conversely, overexpression of Stmn2 dramatically reduced both basal and stimulated glucagon secretion compared to the corresponding control groups (Figure 4E; p < 0.001). These findings suggest that Stmn2 levels may control the regulated secretion of glucagon from αTC1-6 cells.

Stmn2 Directs Glucagon Into Early Endosomes
In wild type αTC1-6 cells, there was weak co-localization between glucagon and the early endosome marker EEA1 (Figure 5A) (PCC = 0.15 ± 0.02). When Stmn2 was overexpressed (Figure 5B), the extent of colocalization between glucagon and EEA1 increased markedly (PCC = 0.53±0.08). Depletion of Stmn2 (Figure 5C) drastically reduced the extent of colocalization (PCC = 0.05) between glucagon and EEA1. Pearson's correlation coefficient of co-localization between glucagon and Stmn2 showed a significant increase when Stmn2 was overexpressed (p < 0.001) and a significant reduction when Stmn2 was knocked down (p < 0.05) compared to the control ( Figure 5D). These findings suggest that Stmn2 plays a role in directing glucagon toward early endosomes.

Stmn2 Overexpression Increases Glucagon Presence in the Late Endosome/Lysosome Compartment
Similar to our findings in early endosomes, there was a weak correlation between glucagon and the late endosomelysosome marker, Lamp2A (PCC = 0.2 ± 0.02) in wild type αTC1-6 cells (Figures 6A,D). Following overexpression of Stmn2, the levels of colocalization between glucagon and Lamp2A were significantly increased (PCC = 0.89 ± 0.05, p < 0.001) (Figures 6B,D). Depletion of Stmn2 significantly reduced the extent of colocalization between glucagon and Lamp2A compared to wild type cells (PCC = 0.001, p < 0.01)  ( Figures 6C,D). Interestingly, the signal intensity of the endolysosomal marker, Lamp2A, was significantly increased (p < 0.01) upon overexpression of Stmn2, but did not change upon depletion of Stmn2 ( Figure 6E).

DISCUSSION
Glucagon secretion is governed by systemic, paracrine and intrinsic factors. Our work has focused on the regulation of glucagon secretion by proteins that associate with glucagon within alpha cell secretory granules. To this end, we have shown that a neuronal protein, Stmn2, which we have previously identified as part of the glucagon interactome (25), can be localized to the secretory granules of αTC1-6 cells. We validated this association in mouse pancreatic islets, and through silencing and overexpression experiments, we showed that Stmn2 can play a role in glucagon secretion by trafficking through the endosomal-lysosomal pathway.
We have previously identified Stmn2 as part of a network of proteins that associate with glucagon within the secretory granules of αTC1-6 cells (25). Data in the current study show that Stmn2 is localized to the secretory granule and Golgi compartments in αTC1-6 cells and mouse pancreatic alpha cells, respectively. Stathmin 2 is part of a family of neuronal phosphoproteins that associates with intracellular membranes, notably the Golgi and vesicle transporters, in neurons (36). Although Stmn2 has been identified as a neuron-specific protein that functions in differentiation and development, its presence in pancreatic alpha cells is not surprising, as several types of neuronal proteins, such as SNARE proteins, neurotransmitters and granins, are also expressed in endocrine cells (37)(38)(39). The expression of neuronal proteins such as Stmn2 in alpha cells may be due to the absence of the transcriptional repressor RE-1 silencing transcription factor (REST) in mature endocrine cells (40,41). The absence of Stmn2 in mouse pancreatic beta cells suggests that transcriptional silencing programs may operate in a cell-specific manner.
Our results indicate that Stmn2 is localized largely to punctate structures within αTC1-6 cells, notably at the plasma membrane. Its colocalization with glucagon, ChGA and VAMP2 at the plasma membrane suggests that it is efficiently sorted from the Golgi to plasma membrane-associated secretory granules. Double immunogold labeling TEM confirmed the presence of glucagon and Stmn2 within secretory granules of α-cells in mouse pancreatic islets. The presence of Stmn2 within the dense core of the granule suggests that it is part of the soluble granule cargo along with glucagon. These results are consistent with our secretory granule proteomics, which predicted the presence of both glucagon and Stmn2 in secretory granules of αTC1-6 cells (25). Our model also predicted that the complement of proteins within alpha cell secretory granules differs in response to microenvironmental inputs, so it possible that Stmn2 is present only in a subpopulation of secretory granules. The trafficking of Stmn2 to secretory granules in αTC1-6 cells may occur through specific molecular domains within its sequence. The subcellular trafficking pattern of Stmn2 in neurons and neuroendocrine cells is determined by its N-terminal extension, which contains a Golgi localization domain and a membrane anchoring domain that contains two conserved Cys residues as sites for palmitoylation (36). This lipid modification occurs in the Golgi (42) and is  sufficient and necessary for the association of Stmn2 with Golgi membranes, and its sorting to post-Golgi vesicles (42,43). While we have not directly shown that palmitoylation of these Cys residues is required for localization in secretory granules in αTC1-6 cells, it is likely that this is a conserved sorting mechanism in neuroendocrine cells. Consistent with its localization within secretory granules, the secretion of both glucagon and Stmn2 was significantly enhanced in response to 55 mM K + . Although there was a statistically significant response to KCl, αTC1-6 cells as a rule do not show robust secretory responses to KCl, glucose or other secretagogues. In general, αTC1-6 cells differ in their complement of transcriptional and epigenetic factors from mouse primary alpha cells, which may explain the relatively blunted secretory response seen in this cell line (44). The reduced response to glucose in particular could be due to a low efficiency in coupling between the glycolytic and TCA pathways, as has been shown in pancreatic islets (45). Nonetheless, we were able to show similar secretory behavior of glucagon and Stmn2 in isolated mouse islets, thus confirming their localization within the releasable pool of secretory granules in pancreatic α-cells. While Stmn2 is a membrane-bound protein, it is also present in normal human blood as determined by serum ELISA (46) and plasma proteomics (47) and therefore can be also secreted. There is some evidence of a lower molecular weight form of Stmn2 that may correspond to a cleaved, soluble form (43,48). In our model, it may be that a portion of Stmn2 becomes cleaved within the secretory granule, thus becoming part of the soluble cargo that is released.
By manipulating the expression of Stmn2, we were able to demonstrate a role in the control of glucagon secretion. There is precedence for a role for Stmn2 in the regulation of neuroendocrine secretion that has some interesting parallels with our results. While we showed that silencing of Stmn2 increased the constitutive secretion of glucagon, one study showed that silencing Stmn2 in PC12 cells decreased both basal and stimulated secretion of chromogranin A (27). In that study, it was shown that Stmn2 interacted directly with ChgA and its depletion reduced the buoyant density of chromaffin granules, suggesting that Stmn2 may participate in secretory granule formation, perhaps in partnership with ChgA (42) and thus promote regulated secretion. Although we could not demonstrate a direct interaction between Stmn2 and glucagon (data not shown), our results align with the idea that Stmn2 may be a sorting partner for glucagon, perhaps by interacting with other granule proteins such as ChgA or carboxypeptidase E (31), in the regulated secretory pathway of alpha cells. Blue arrows indicate the normal trafficking of glucagon, together with Stmn2, to secretory granules, where they are stored until their release is triggered by a stimulus. Stmn2 overexpression (red arrows) reduces the amount of glucagon available for secretion by diverting secretory granules to lysosomes. Stmn2 depletion (black arrows) reduces the trafficking of glucagon into secretory granules and promotes the constitutive release of glucagon.
Overexpression of Stmn2 resulted in an almost complete shutdown of glucagon secretion and an increase in the localization of glucagon in early endosomes and lysosomes. These results may suggest induction of ER stress. However, overexpression of Stmn2 in Neuro2A cells, another cell line commonly used to examine regulated secretion, had the opposite effect and enhanced trafficking of post-Golgi carriers to the plasma membrane (49). There are only a few reports on how ER stress in alpha cells is manifest; in one report, palmitateinduced ER stress in isolated rat alpha cells resulted in an increase in glucagon secretion, not inhibition, even though the traditional markers of ER stress [Ddit3 (Chop), Xbp1s, Hsp5a (BiP)] were elevated (50). Another study showed that in vivo depletion of Xbp1, which plays a crucial role in the unfolded protein response, induced dysfunctional glucagon secretion that was not fully suppressed by insulin (51). Therefore, the suppression of glucagon secretion by Stmn2 overexpression is not consistent with the phenotype of ER stress in the alpha cell. Instead, these results, together with the increased the presence of glucagon in the lysosomal compartment, suggest that Stmn2 overexpression causes alterations in glucagon trafficking independent of ER stress.
We believe that the overexpression experiments suggest a mechanism whereby an increase in Stmn2 in the alpha cell inhibits glucagon secretion by targeting glucagon secretory granules for degradation in the endosome-lysosome pathway, perhaps in a manner similar to that of insulin secretory granules in Type 2 diabetes (52). Very recently, it has been shown that insulin secretion can be inhibited by the targeting of proinsulin for lysosomal degradation by Rab7-interacting lysosomal protein (RILP) (53). The proposed mechanism of action was through interactions between RILP and an insulin secretory granule membrane protein, Rab26. It is tempting to speculate that Stmn2 could have a similar mechanism of action in alpha cells. The increase in LAMP2 fluorescence intensity upon Stmn2 overexpression further suggests a role for Stmn2 in modulating glucagon secretion through increased lysosome function or biogenesis. We are currently investigating this mechanism of control of glucagon secretion in mouse islets.
In conclusion, we propose that Stmn2, a protein that is associated with glucagon in secretory granules, modulates glucagon secretion in alpha cells by playing a role in its intracellular trafficking. Under conditions that decrease Stmn2 levels, constitutive secretion of glucagon is increased; and under conditions that increase levels of Stmn2, glucagon is targeted to the endolysosomal system, presumably for degradation (Figure 7). Our findings, which are mainly based on the clonal α-cell line αTC1-6, represent a potentially novel intracellular pathway for the control of glucagon secretion, and may lead to new mechanistic insights in the dysregulation of glucagon secretion in diabetes.

DATA AVAILABILITY STATEMENT
All datasets generated for this study are included in the article/supplementary material.

ETHICS STATEMENT
The animal study was reviewed and approved by Animal Use Subcommittee of the Canadian Council on Animal Care at Western University, Animal Use Protocol AUP 2012-020.