Abstract
Proglucagon is expressed in pancreatic alpha cells, intestinal L cells and brainstem neurons. Tissue-specific processing of proglucagon yields the peptide hormones glucagon in the alpha cell and glucagon-like peptide (GLP)-1 and GLP-2 in L cells. Both glucagon and GLP-1 are secreted in response to nutritional status and are critical for regulating glycaemia. The sorting of proglucagon to the dense-core secretory granules of the regulated secretory pathway is essential for the appropriate secretion of glucagon and GLP-1. We examined the roles of carboxypeptidase E (CPE), a prohormone sorting receptor, the processing enzymes PC1/3 and PC2 and putative intrinsic sorting signals in proglucagon sorting. In Neuro 2a cells that lacked CPE, PC1/3 and PC2, proglucagon co-localised with the Golgi marker p115 as determined by quantitative immunofluorescence microscopy. Expression of CPE, but not of PC1/3 or PC2, enhanced proglucagon sorting to granules. siRNA-mediated knockdown of CPE disrupted regulated secretion of glucagon from pancreatic-derived alphaTC1–6 cells, but not of GLP-1 from intestinal cell-derived GLUTag cells. Mutation of the PC cleavage site K70R71, the dibasic R17R18 site within glucagon or the alpha-helix of glucagon, all significantly affected the sub-cellular localisation of proglucagon. Protein modelling revealed that alpha helices corresponding to glucagon, GLP-1 and GLP-2, are arranged within a disordered structure, suggesting some flexibility in the sorting mechanism. We conclude that there are multiple mechanisms for sorting proglucagon to the regulated secretory pathway, including a role for CPE in pancreatic alpha cells, initial cleavage at K70R71 and multiple sorting signals.
Introduction
Proglucagon is a prohormone that is expressed in the alpha cells of the pancreas, the L cells of the intestine and in select neurons of the brainstem. In the pancreatic alpha cell, PC2 cleaves proglucagon to yield glucagon (Rouille et al. 1994, Furuta et al. 2001). By contrast, proglucagon is processed to glucagon-like peptide (GLP)-1 and GLP-2 in the intestine and brain by PC1/3 (Dhanvantari et al. 1996, Dhanvantari & Brubaker 1998, Damholt et al. 1999). Glucagon is secreted primarily in response to low blood glucose levels, and maintains euglycaemia by stimulating hepatic glycogenolysis and gluconeogenesis (Jiang & Zhang 2003). By contrast, GLP-1 is secreted in response to nutrient ingestion, and enhances glucose-stimulated insulin secretion (Drucker 2007). Therefore, both hormones are critical in maintaining glucose homoeostasis in response to nutritional status. While mechanisms of exocytosis of glucagon and GLP-1 have been extensively studied (Tolhurst et al. 2009, Gonzalez-Velez et al. 2012), and the regulation of biosynthesis and processing of proglucagon continue to be investigated (Cawley et al. 2011, Helwig et al. 2011, Whalley et al. 2011), the mechanisms that govern the sorting of proglucagon into secretory granules are yet to be identified.
Endocrine and neuroendocrine cells are characterised by the presence of a regulated secretory pathway, with specific intracellular compartments and associated proteins for the sorting, processing and storage of peptide hormones. Peptide hormones are first synthesised as larger precursors, or prohormones, and are selectively targeted to the regulated secretory pathway via the trans-Golgi network (TGN). Proteins destined for the regulated secretory pathway are separated from constitutive and lysosomal proteins and are packaged into budding immature secretory granules. The process of sorting can occur at the TGN through interaction with a membrane-bound sorting receptor, or with membrane lipids, via a sorting signal within the prohormone. Alternatively, sorting can occur by selective retention of prohormones within the immature secretory granule, while other proteins are removed. Prohormones then undergo endoproteolysis to yield their constituent peptide hormones, which are then stored in mature, dense-core secretory granules until secretion is triggered in a Ca2+-dependent manner.
Some clues to how proglucagon is sorted are provided by the proteins expressed in the granules of alpha and L cells, and by the primary and secondary structures of proglucagon. The prohormone processing enzyme carboxypeptidase E (CPE) is expressed in both alpha and L cells, while PC1/3 is found in L cells and PC2 in alpha cells. Several lines of evidence support the hypothesis that the membrane-bound forms of CPE and PC1/3 can direct prohormones to the regulated secretory pathway, thus acting as receptors or tethers. Depletion of CPE disrupts the regulated secretion of proopiomelanocortin (POMC), proenkephalin and proinsulin in cells in culture (Normant & Loh 1998, Dhanvantari et al. 2003b), and in the CPEfat mouse, an obese mouse model in which CPE is degraded in the pituitary, POMC is secreted constitutively (Cool et al. 1997, Shen & Loh 1997). The presence of full-length PC1/3 with its C-terminal region intact is required for the processing of prorenin to renin in granules in AtT-20 cells (Jutras et al. 1997). Without its lipid-binding C-terminal region (Jutras et al. 2000, Arnaoutova et al. 2003), PC1/3 is not sorted to the regulated secretory pathway, and renin is secreted constitutively. PC2 also contains an alpha-helical, lipid-binding C-terminal tail that directs it to the regulated pathway (Assadi et al. 2004), thus enabling it to act as a membrane-bound receptor/tether. However, no evidence to date exists to support a role for PC2 in the sorting of prohormones to the regulated secretory pathway.
Additionally, the structures of the major proglucagon-derived peptides predict some known sorting signals. The structure of glucagon is that of an alpha-helix, as revealed by X-ray crystallography (Sasaki et al. 1975) and the nuclear magnetic resonance solution structure of GLP-1 also predicts an alpha-helix (Chang et al. 2002). Sequence similarity suggests that GLP-2 is also an alpha-helix, and it has been suggested that the structure of the entire proglucagon molecule may form a ‘trimer’ of alpha helices (Dey et al. 2004). Conserved alpha-helix structures within the amino-terminal end of prosomatostatin (Mouchantaf et al. 2001) and the C-terminal ends of PC1/3 (Jutras et al. 2000, Arnaoutova et al. 2003), PC2 (Assadi et al. 2004), PC5/6a (Dikeakos et al. 2007) and CPE (Dhanvantari et al. 2003a, Zhang et al. 2003) are critical in targeting these proteins to secretory granules. Additionally, the sequence of glucagon contains an internal pair of basic amino acids, Arg17Arg18, which does not serve as a PC1/3 or PC2 cleavage site (Bataille 2007). Sequences of basic residues within prorenin (Brechler et al. 1996), proneurotensin (Feliciangeli et al. 2001), proneuropeptide Y (Brakch et al. 2002) progastrin (Bundgaard et al. 2004) and VGF (Garcia et al. 2005) have been identified as sorting signals for these prohormones. Therefore, proglucagon may contain a number of sorting signals to direct it into granules. In the present study, we investigated the requirement of the processing enzymes CPE, PC1/3 and PC2, and of intrinsic sorting signals, for the sorting of proglucagon to secretory granules.
Materials and methods
Cell culture, plasmids and transfections
Wild-type Neuro 2a (N2a wt) cells and Neuro 2a cells stably transfected with mouse Cpe (N2a-CPE, clone 17) were obtained from Dr Y P Loh (Bethesda, MD, USA). Cells were maintained in DMEM containing 10% FBS and stable transfectants were maintained in media containing 400 μg/ml G418. N2a wt cells were also stably transfected with the enzymatically inactive form of CPE, E300Q (Qian et al. 1999) using Lipofectamine 2000 (Invitrogen). Stable transfectants were selected in 800 μg/ml G418, pooled and maintained in 400 μg/ml G418. To examine the role of CPE in sorting proglucagon, N2a wt, N2a-CPE and N2a-E300Q cells were transfected with hamster pre-proglucagon (in pcDNA 3.1; a kind gift from Dr D F Steiner, Chicago, IL, USA). To examine the roles of other prohormone convertases, both N2a wt and N2a-CPE cells were transiently transfected with plasmids encoding PC1/3 and PC2 (kind gifts from Dr N G Seidah, Montreal, QC, Canada). To determine possible sorting signals, the sequence of proglucagon was mutated independently at three sites: at the processing site, R70K71; at the dibasic site within glucagon, R17R18; and at two leucines that were postulated to flank the helix structure within glucagon, L14 and L26. The processing site mutant, K71Q, was a kind gift from Dr D F Steiner (Chicago, IL, USA). We generated the mutation at the dibasic site, R18Q (forward primer sequence 5′-AAATACCTGGACTCCCGCCAAGCCCAAGAT-3′), and the double leucine-to-proline mutation was done in two steps; L14P was made using forward primer 1, 5′-TACAGCAAATACCCGGACTCCCGCCGAGCC-3′; and L26P was subsequently generated using forward primer 2, 5′-CAAGATTTTGTGCAGTGGCCGATGAACACC-3′. Bold sequences in primers indicate site of mutation. All site-directed mutagenesis reactions were carried out using the QuikChange II Site-Directed Mutagenesis Kit (Agilent Technologies, Santa Clara, CA, USA) and results were confirmed by sequencing at the London Regional Genomics Facility, University of Western Ontario.
Immunocytochemistry
Cells were grown on coverslips, and processed for immunocytochemistry as previously described (McGirr et al. 2005). Slides were incubated with antibodies against glucagon (1:1000; Bachem/Peninsula Laboratories, Torrance, CA, USA) and the Golgi marker p115 (1:50; Transduction Laboratories, San Jose, CA, USA) or the secretory granule marker chromogranin A (CgA; 1:50, Abcam). The anti-glucagon antibody was raised against the entire sequence of glucagon, and thus recognises unprocessed proglucagon, glicentin, oxyntomodulin and glucagon. To examine CPE immunoreactivity, slides were incubated with the CPE 7–6 antibody (kind gift from Dr Y P Loh, Bethesda, MD, USA), which recognises the C-terminus of CPE. Alexa 488-IgG (Invitrogen) was used to visualise the glucagon or CPE antibody, and Alexa 594-IgG for the p115 or CgA antibody.
Image acquisition and analysis
Cells were visualised with an Olympus IX81 widefield fluorescence microscope. Image acquisition was carried out using In Vivo software. Ten optical sections per cell were collected in 0.2 μm steps covering the z-axis field, using a 60× oil immersion objective lens. Cell images were processed using a three-dimensional (3D) blind deconvolution algorithm provided in Image-Pro Plus software (Media Cybernetics, Inc., Rockville, MD, USA).
Image analysis using Pearson's correlation coefficient
Image analyses were conducted using FIJI version 1.46 h (Schindelin 2008), a distribution of ImageJ (NIH, Bethesda, MD, USA). Co-localisation was quantified using the Co-localisation 2 plugin within FIJI. Regions of interest were manually drawn around distinct single- or multi-cell bodies from pseudo-coloured red and green fluorescence images, excluding as much background as possible. The segmented region was defined by a binary mask and applied to both red and green images. The plugin uses an iterative Costes' threshold algorithm (Costes et al. 2004) to simultaneously determine maximum threshold of intensity for each colour, above which pixels are considered to be statistically correlated. Red and green image pixels were then used to calculate the Pearson's correlation coefficient (PCC) (r) according to Equation 1:
Briefly, i represents the i-th pixel in the region of interest, Sc1 represents signal intensity of the co-localised pixel from the first channel and Sc2 represents the second channel, S1 and S2 represent signal intensities of all pixels regardless of co-localisation, and S1av and S2av represent the average pixel intensity of each channel. Note the numerator is dependent on pixels above the Costes auto threshold, while the denominator and mean pixel values use all the pixels in the region of interest. Correlation coefficients for each experiment were treated as one experimental dataset and differences were determined using a one-way ANOVA followed by Tukey's multiple comparison test. Significance was set at P>0.05. n values ranged from 10 to 25. The PCC was validated in INS-1 832/13 cells (a kind gift from Dr C B Newgard, Raleigh-Durham, NC, USA) using markers with a known low degree of co-localisation (insulin and pdx-1) and a high degree of co-localisation (insulin and CgA) (Supplementary Figures 1 and 2, see section on supplementary data given at the end of this article).
Western blot
For western blot analysis, all cells were grown in six-well plates. Cells were lysed, and protein was extracted, quantified and separated on 12% SDS–PAGE gels as described previously (McGirr et al. 2005). Proteins were transferred onto nitrocellulose membranes, which were probed with the CPE 4–5 antibody (kind gift from Dr Y P Loh, Bethesda, MD, USA), or antibodies against PC1/3 or PC2 (kind gifts from Dr N G Seidah, Montreal, QC, Canada) and bands were visualised with SuperSignal Chemiluminescent system as described previously (McGirr et al. 2005). To detect secreted CPE, media were collected and protein was concentrated through evaporation. The resulting residue was resuspended in loading buffer and the entire sample was loaded into one well. These samples migrated more slowly through the gel, resulting in an apparent Mr of 60 kDa.
Secretion experiments
AlphaTC1–6 cells (a kind gift from C B Verchere, Vancouver, BC, Canada) were maintained in DMEM containing 15% horse serum and 2.5% FBS. GLUTag cells (a kind gift from D J Drucker, Toronto, ON, Canada) were grown in low-glucose DMEM containing 10% FBS. For stimulated secretion experiments, αTC1–6 cells were seeded in 24-well plates in replicates of six and pre-incubated in HBSS containing 25 mM glucose for 1 h. Cells were then incubated in HBSS containing 1 mM glucose and 10 μM each of forskolin and IBMX for 1 h. GLUTag cells were incubated in low-glucose (5.5 mM) DMEM +0.5% FBS without (basal) or with (stimulated) 10 μM each of forskolin and IBMX for 2 h. After all secretion experiments, media were collected and trifluoroacetic acid (TFA) was added to 0.1%. Cells were rinsed twice in HBSS, and scraped in 1 ml homogenisation buffer (1 M HCl, 1 M formic acid, 1% (v/v) TFA, 1% (w/v) NaCl). The cells were sonicated in one 12 s burst, centrifuged and the supernatant was collected, and both media and cell extracts were passed through a Sep-Pak C18 reverse-phase cartridge to elute proglucagon-derived peptides as previously described (Dhanvantari et al. 1996, Dhanvantari & Brubaker 1998). Glucagon and GLP-1 content were assessed by RIA using the glucagon and GLP-1 RIA kits (Linco/EMD Millipore, Bedford, MA, USA).
Design of siRNA
To deplete CPE in αTC1–6 and GLUTag cells, siRNAs were synthesised using the Silencer siRNA Construction Kit (Ambion/Life Technologies, Carlsbad, CA, USA). Target sequences within the rat Cpe mRNA sequence were selected using an algorithm on Ambion's website, and a search of the mouse genome with BLAST showed that seven of the selected targets were not homologous with any other known mouse mRNAs. siRNAs were synthesised against these seven targets and transfected into both cell lines, which were then screened for the extent of CPE depletion by western blot analysis of cell lysates using the CPE 4–5 antibody (a gift from Dr Y P Loh, Bethesda, MD, USA) (Supplementary Figure 3, see section on supplementary data given at the end of this article). One siRNA, spanning nucleotides 1408–1428 (CPE73), significantly and reproducibly decreased CPE levels in both cell lines, and was used in subsequent secretion experiments. The negative control was Ambion Silencer Negative Control #1, with no known targets in the mouse genome. In all subsequent experiments, 200 nM of control siRNA or CPE73 siRNA were transfected using Oligofectamine (Invitrogen) for 72 h prior to secretion experiments or fixation for immunofluorescence microscopy.
Protein modelling
The full-length proglucagon polypeptide sequence was submitted for de novo 3D protein structure prediction using the Robetta 3D modelling web server (Kim et al. 2004). The 3D models developed by Robetta used the incretin hormone, gastric inhibitory peptide (GIP; PDB:2B4N) as a reference parent. The Ginzu module (Chivian et al. 2005) parsed proglucagon into two domains, each using GIP as a reference parent. The N-terminal domain (aa 1–77) (PDB:2B4N) had a confidence of 5.40 by position-specific iterated BLAST (PSI-BLAST; Altschul et al. 1990); the C-terminal domain (aa 78–158) (PDB:2OBU) had a confidence of 2.33 by HH-Search (Soding 2005).
Results
The role of CPE in proglucagon sorting
Since CPE has been identified as a sorting receptor for several other prohormones, we investigated a possible role for CPE in sorting proglucagon to secretory granules. There was no CPE immunoreactivity in N2a-wt cells; it was detected only after transfection of CPE (Fig. 1A). In N2a-wt cells, proglucagon immunoreactivity associated strongly with that of the Golgi marker, p115 (Fig. 1B). By contrast, overexpression of CPE resulted in more punctate staining of proglucagon in the processes of the cells, and much weaker co-localisation with p115 (Fig. 1B). PCC analysis showed a significant (P<0.01) decrease in the co-localisation of proglucagon with p115 in the presence of CPE (Fig. 1C), suggesting that CPE expression was associated with sorting proglucagon to granules. To confirm that proglucagon was actually in the secretory granule compartment, N2a-CPE cells were transfected with proglucagon-EGFP (enhanced green fluorescent protein) and co-immunostained with antibodies against CgA and EGFP (Supplementary Figure 4, see section on supplementary data given at the end of this article). Proglucagon immunoreactivity co-localised with that of CgA in punctate structures in the cell processes, and the PCC was significant (r=0.84±0.01, n=6).
Proglucagon sorting is dependent on properly sorted CPE
In order to further investigate the role of CPE in sorting proglucagon, N2a-wt cells were transfected with the inactive E300Q mutant of Cpe (Fig. 1A). In these cells, proglucagon was co-localised with p115 and the PCC was actually significantly (P<0.01) greater than that of N2a-wt cells (Fig. 1C). We then examined the trafficking of the Cpe-E300Q mutant, as we hypothesised that the sorting of the mutant itself may be impaired, and this could explain its effect on proglucagon sorting. When compared with wt CPE, which had a weak Golgi signal and showed strong immunofluorescence in the tips of the cell processes, the E300Q mutant instead yielded a robust signal in the Golgi (Fig. 2A). Additionally, wt CPE was secreted, while E300Q could only be detected in the cells and not in the media (Fig. 2B), therefore suggesting that E300Q CPE is not properly sorted in N2a cells.
The role of PC1/3 and PC2 in proglucagon sorting
We explored the roles of the other neuroendocrine-specific prohormone convertases, PC1/3 and PC2, in the sorting of proglucagon, since these two enzymes are expressed in the L cell and alpha cell respectively along with proglucagon. We investigated if these other processing enzymes could sort proglucagon alone, or if they could enhance CPE-mediated sorting. Neither PC1/3 nor PC2 were endogenously expressed in N2a-wt cells (Fig. 3A). Cells were transfected with PC1/3 or PC2 alone or in combination with CPE. Figure 3A shows that when PC1/3 was expressed in the absence of CPE, one 84 kDa form was present; upon CPE expression, two forms of PC1/3, an 84 kDa form and a 66 kDa form, were detected. The expression and processing of PC2 were not changed in the presence of CPE. When either PC1/3 or PC2 were expressed alone, proglucagon appeared mainly in the Golgi, and the PCC was not significantly different from that of N2a-wt cells (Fig. 3C). Interestingly, proglucagon immunoreactivity appeared in the cell processes, indicative of granule localisation, in the presence of both CPE and PC1/3 (Fig. 3B), and the PCC significantly decreased from the value of cells expressing PC1/3 alone (Fig. 3C). However, the correlation between proglucagon and p115 localisation remained unchanged from N2a wt cells when both CPE and PC2 were expressed (Fig. 3C).
CPE affects proglucagon sorting in alpha cells, but not L cells
In order to examine the role of CPE in proglucagon sorting in a physiological context, we used two proglucagon-expressing cell lines: alphaTC1–6 cells, derived from mouse glucagonoma, and GLUTag cells, derived from mouse enteroendocrine tumours. In alphaTC1–6 cells, when CPE expression was knocked down by 61±3% (n=3, P<0.05) using CPE73 siRNA, normal stimulated secretion of glucagon was inhibited, and basal secretion increased significantly (Fig. 4A and B). Cell content decreased slightly, indicating that the absence of CPE may have affected glucagon synthesis as well as secretion (Fig. 4C). The transfection efficiency, as assessed by EGFP fluorescence, was 36.6±3.7% (calculated from ten images, each with ∼50 cells). By contrast, regulated secretion of GLP-1 in GLUTag cells remained intact after knockdown of CPE by 74±7% (n=3, P<0.05) using CPE73 siRNA (Fig. 5).
Initial processing of proglucagon at K70R71 enhances sorting to granules
Although neither PC1/3 nor PC2 are expressed in N2a-wt cells, another processing enzyme, furin, is expressed in all cells (Thomas 2002). Furin cleaves proglucagon at K70R71, which is the initial processing event that precedes further processing in secretory granules. In order to determine if processing at this site is required for sorting, we transfected N2a-CPE cells with proglucagon that harboured the R71Q mutation, which has previously been shown to inhibit processing (Dey et al. 2004). Localisation of the R71Q mutant was largely in the Golgi, although there was punctate immunofluorescence in the processes of some cells, and the PCC with p115 was significantly higher than that of wt proglucagon (Fig. 6), suggesting that processing at K70R71 may enhance sorting to granules.
Possible sorting signals are contained within the glucagon sequence
The sequence of glucagon contains two putative sorting signals, based on those described in other prohormones; a dibasic site that is not cleaved by PCs, and significant alpha-helical content. To determine the role of these structures in proglucagon sorting, the dibasic site was mutated to R18Q, and the alpha-helical content of glucagon was decreased by mutating the flanking leucines to prolines (L14P/L26P). Each mutation showed significant co-localisation with p115 (Fig. 6A) and significantly higher PCC with p115 compared with wt proglucagon (Fig. 6B), indicating that the dibasic site and the alpha-helix may each play a role in sorting proglucagon to secretory granules.
Modelling of the structure of proglucagon
We used the entire sequence of proglucagon to generate a 3D model so that clues to the mechanisms of putative sorting signals may be identified. Using the Robetta software suite (Seattle, WA, USA. http://robetta.bakerlab.org/), five structures were predicted, one of which is shown in Fig. 7. Three alpha helices, each corresponding to regions within the sequences of glucagon (proglucagon 58–78), GLP-1 (proglucagon 104–124) and GLP-2 (proglucagon 129–152), are randomly arranged within a disordered structure. The prohormone processing site K70R71 is indicated at the N-terminal end of GLP-1. Unlike a previously reported model that predicted a rigid trimeric arrangement of the alpha helices (Dey et al. 2004), our model suggests that the spatial orientation of the alpha helices is not fixed, and that the location of the K70R71 site is within a flexible, unstructured region.
Discussion
While many studies have described the tissue-specific, post-translational processing of proglucagon, and elucidated the mechanisms underlying the regulation of glucagon and GLP-1 secretion, the mechanisms by which proglucagon is sorted into secretory granules remain unidentified. Sorting is a critical step in the proglucagon biosynthetic pathway, for two reasons. First, proper sorting ensures final cleavages by PC1/3 and PC2, which must occur in the low-pH, high-[Ca2+] environment of the granules, to yield mature glucagon, GLP-1 and GLP-2. Second, appropriate compartmentalisation within secretory granules is an absolute requirement for stimulated secretion in response to nutritional status. Therefore, we sought to determine some possible mechanisms by which proglucagon is sorted into granules.
We first examined the role of CPE in proglucagon sorting, since it has been shown that CPE can act as a sorting receptor for some prohormones, such as POMC and proinsulin (Cool et al. 1997, Dhanvantari et al. 2003b). We used Neuro 2a cells in our study because we obtained a clone of N2a cells in which CPE expression had been lost (Zhang et al. 2003), which provided an ideal control. Sorting efficiency was determined using immunofluorescence microscopy quantified with PCC to determine the extent of co-localisation of proglucagon with the Golgi marker p115. We used a Golgi marker instead of the granule marker, CgA, since we found that CgA gave strong Golgi and granule signals that masked changes in the localisation of proglucagon. PCC has been previously used to determine the localisation of palmitoyltransferases in the endoplasmic reticulum (Gorleku et al. 2011) and is thought to be an accurate measure of co-localisation since it is a correlation between fluorescence intensities within the same pixels (Adler & Parmryd 2010). In the absence of CPE, the increased presence of proglucagon in the Golgi and the reduced punctate immunofluorescence pattern indicate a reduction in sorting efficiency, as with POMC in these same wt N2a cells (Zhang et al. 2003). To confirm that CPE plays a role in sorting proglucagon, we used pancreatic-derived alphaTC1–6 cells, which we have used previously to characterise glucagon secretion under conditions of high glucose (McGirr et al. 2005). A 60% decrease in CPE expression resulted in a loss of regulated secretion of glucagon, the extent to which is likely underestimated as CPE levels were not decreased more dramatically. Interestingly, regulated secretion of GLP-1 in GLUTag remained intact after siRNA-mediated reductions in CPE levels. Therefore, CPE plays a role in proglucagon sorting in a cell-specific manner, and different mechanisms of sorting operate in the cell types that express proglucagon.
We found that proglucagon was not targeted to granules in the presence of the E300Q Cpe mutant, which itself was retained in the Golgi and not secreted. Therefore, impaired sorting of CPE may prevent the proper targeting of proglucagon to granules. Our finding of impaired sorting of the E300Q mutant is at variance with a previous report that showed that it bound to its substrate and was sorted to granules normally in NIT3 cells (Qian et al. 1999). It is not clear how CPE is missorted in N2a cells, since its C-terminal sorting domain (Fricker et al. 1990, Zhang et al. 2003) is intact in the E300Q mutant, as evidenced by our western blot results showing strong immunoreactivity in cell extracts using a C-terminal antibody. The strong Golgi localisation also demonstrates that E300Q is not degraded, unlike another enzymatically inactive mutant, S202P (Naggert et al. 1995); rather, it is the post-Golgi trafficking of the E300Q mutant that is impaired in N2a cells. Although the missorting of E300Q CPE may be unique to N2a cells, nonetheless, these results mirror those showing that improperly sorted PC1/3 prevents the targeting of prorenin to the regulated secretory pathway (Brechler et al. 1996). Therefore, there may be a direct interaction between CPE and proglucagon that results in co-targeting to granules. The nature of the interaction between CPE and some other prohormones has been well documented; studies on the structures of POMC, proenkephalin, proinsulin and brain-derived neurotrophic factor showed that exposed acidic amino acid residues on a disulphide-bonded loop confer the ability to bind to two basic residues, K255 and R260, on CPE (Cool et al. 1995, 1997, Zhang et al. 1999, Lou et al. 2005). Our results on modelling the structure of proglucagon predict distinct alpha-helical domains within a largely disordered structure, and no disulphide-bonded loop. However, a CPE binding site may lie in the disordered regions of GRPP (proglucagon 1–30) or IP2 (proglucagon 111–133), which contain acidic amino acids in their sequences. It has been postulated that disordered structures within proteins may mediate protein–protein interactions due to their structural plasticity (Chen 2012). Further characterisation of the cognate domains within CPE that may bind to specific structural sorting determinants of proglucagon will be required to elucidate a potentially distinct molecular mechanism of CPE-mediated proglucagon sorting.
Since other prohormone processing enzymes are also co-targeted to secretory granules with their substrates, we investigated the roles of the proglucagon-processing enzymes, PC1/3 and PC2, in sorting proglucagon, in the absence or the presence of CPE. A possible role of PC1/3 in proglucagon sorting was of interest to us, since: i) PC1/3 may function in the sorting of another prohormone, prorenin; ii) the presence of PC1/3 up-regulates the secretion of the granule markers CgA and SgII, suggesting that PC1/3 may be part of the sorting machinery (Lankat-Buttgereit et al. 2008); and iii) GLUTag cells were not responsive to CPE knockdown, but express PC1/3, which would have been a candidate sorting receptor in these cells. However, our data do not support a role for PC1/3 in proglucagon sorting. Indeed, while there is impaired proglucagon processing in the intestine in both mice (Ugleholdt et al. 2004) and a human subject (Jackson et al. 2003) lacking PC1/3, GLP-1 was secreted normally in response to feeding, so it may be unlikely that PC1/3 plays a role in proglucagon sorting in the L cell. Therefore, alternative mechanisms operate in the sorting of proglucagon in L cells. This finding lends another level of complexity to the sorting mechanism of proglucagon.
The impaired sorting of the K70Q71 cleavage site mutant of proglucagon suggests that processing at this site enhances sorting to granules. The post-translational processing of proglucagon follows a temporal pattern, with cleavage at K70R71 occurring in the Golgi/immature secretory granules, and further processing by PC1/3 or PC2 into bioactive peptides occurring in mature secretory granules (Dhanvantari et al. 1996, Dey et al. 2004). Processing at this site is largely through the actions of the enzyme furin, since it occurs in all cell types in which proglucagon has been expressed, regardless of other PCs that may be present, and since proglucagon is processed to glicentin in baby hamster kidney (BHK) cells, a fibroblast cell line lacking secretory granules (Dhanvantari et al. 1996). An increase in the sorting efficiency of proglucagon after cleavage at K70R71 in the Golgi would be similar to the observations with prothyrotropin-releasing hormone (pro-TRH), where processing by PC1/3 enhances the sorting of pro-TRH-derived peptides into granules (Mulcahy et al. 2005). Future experiments will investigate more closely the role of processing in proglucagon sorting, and the possibility that the products of this initial processing, glicentin and the major proglucagon fragment (MPGF) are sorted into different granule populations, again in a pattern similar to that of pro-TRH sorting (Perello et al. 2008).
Finally, we have for the first time identified putative sorting signals within proglucagon. We postulated that an internal dibasic site and the alpha-helical content in the sequence of glucagon could target proglucagon to granules, since both motifs have been identified as sorting signals for a number of regulated secretory proteins (Dikeakos & Reudelhuber 2007). Our results indicate that both motifs may constitute proglucagon sorting signals, and that they may act independently; mutating the dibasic site does not affect alpha-helical content, and altering the helical content leaves the dibasic site intact (Guizzetti and Dhanvantari 2012, unpublished observations). It has been shown that multiple types of sorting signals increase the efficiency of sorting to granules (Lacombe et al. 2005), which may then translate into a greater degree of physiological regulation of secretion. In this regard, the sorting efficiency of proglucagon may be similar to that of proinsulin, which also contains a number of different sorting signals (hexamerisation, CPE binding sites and dibasic sites). When these signals are disrupted, the resulting loss of regulated secretion results in the clinical syndrome of hyperproinsulinaemia (Dhanvantari et al. 2003b). A high degree of sorting efficiency would ensure that the pancreatic alpha and beta cells and intestinal L cells are in a nutrient-responsive state, and would account for the exquisite control of glycaemia exerted by glucagon, insulin and GLP-1.
It is also remarkable that altering the alpha-helical content in glucagon affects sorting, given that there are two other alpha helices, contained within GLP-1 and GLP-2, that could compensate. Either the two other alpha helices do not act as sorting signals, or these results could be consistent with the hypothesis that proglucagon is first processed in the Golgi before entering granules. If the alpha-helix within glucagon is disrupted, while the alpha helices of GLP-1 and GLP-2 stayed intact, the sorting of glicentin would be impaired, while the sorting of MPGF would not be affected. Experiments are currently underway to more precisely define the role of the alpha helices within glucagon, GLP-1 and GLP-2, as well as of other possible intrinsic signals, in proglucagon sorting.
Supplementary data
This is linked to the online version of the paper at http://dx.doi.org/10.1530/JOE-12-0468.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This study was supported by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada, the Lawson Health Research Institute Internal Research Fund, and an Early Researcher Award from the Ontario Ministry of Research and Innovation.
Author contribution statement
S D designed the experiments, R M completed the experiments and statistical analysis, L G carried out the fluorescence correlation data analysis and protein modelling, and all three authors wrote the manuscript.
Acknowledgements
The authors thank Dr J D Dikeakos, University of Western Ontario, for critical reading of the manuscript.
References
Adler J & Parmryd I 2010 Quantifying colocalization by correlation: the Pearson's correlation coefficient is superior to the Mander's overlap coefficient. Cytometry. Part A 77 733–742. (doi:10.1002/cyto.a.20896)
Altschul SF, Gish W, Miller W, Myers EW & Lipman DJ 1990 Basic local alignment search tool. Journal of Molecular Biology 215 403–410.
Arnaoutova I, Smith AM, Coates LC, Sharpe JC, Dhanvantari S, Snell CR, Birch NP & Loh YP 2003 The prohormone processing enzyme PC3 is a lipid raft-associated transmembrane protein. Biochemistry 42 10445–10455. (doi:10.1021/bi034277y)
Assadi M, Sharpe JC, Snell C & Loh YP 2004 The C-terminus of prohormone convertase 2 is sufficient and necessary for Raft association and sorting to the regulated secretory pathway. Biochemistry 43 7798–7807. (doi:10.1021/bi036331g)
Bataille D 2007 Pro-protein convertases in intermediary metabolism: islet hormones, brain/gut hormones and integrated physiology. Journal of Molecular Medicine 85 673–684. (doi:10.1007/s00109-007-0167-4)
Brakch N, Allemandou F, Cavadas C, Grouzmann E & Brunner HR 2002 Dibasic cleavage site is required for sorting to the regulated secretory pathway for both pro- and neuropeptide Y. Journal of Neurochemistry 81 1166–1175. (doi:10.1046/j.1471-4159.2002.00919.x)
Brechler V, Chu WN, Baxter JD, Thibault G & Reudelhuber TL 1996 A protease processing site is essential for prorenin sorting to the regulated secretory pathway. Journal of Biological Chemistry 271 20636–20640. (doi:10.1074/jbc.271.34.20636)
Bundgaard JR, Birkedal H & Rehfeld JF 2004 Progastrin is directed to the regulated secretory pathway by synergistically acting basic and acidic motifs. Journal of Biological Chemistry 279 5488–5493. (doi:10.1074/jbc.M310547200)
Cawley NX, Portela-Gomes G, Lou H & Loh YP 2011 Yapsin 1 immunoreactivity in α-cells of human pancreatic islets: implications for the processing of human proglucagon by mammalian aspartic proteases. Journal of Endocrinology 210 181–187. (doi:10.1530/JOE-11-0121)
Chang X, Keller D, O'Donoghue SI & Led JJ 2002 NMR studies of the aggregation of glucagon-like peptide-1: formation of a symmetric helical dimer. FEBS Letters 515 165–170. (doi:10.1016/S0014-5793(02)02466-3)
Chen J 2012 Towards the physical basis of how intrinsic disorder mediates protein function. Archives of Biochemistry and Biophysics 524 123–131. (doi:10.1016/j.abb.2012.04.024)
Chivian D, Kim DE, Malmstrom L, Schonbrun J, Rohl CA & Baker D 2005 Prediction of CASP6 structures using automated Robetta protocols. Proteins 61 (Suppl 7) 157–166. (doi:10.1002/prot.20733)
Cool DR, Fenger M, Snell CR & Loh YP 1995 Identification of the sorting signal motif within pro-opiomelanocortin for the regulated secretory pathway. Journal of Biological Chemistry 270 8723–8729. (doi:10.1074/jbc.270.15.8723)
Cool DR, Normant E, Shen FS, Chen HC, Pannell L, Zhang Y & Loh YP 1997 Carboxypeptidase E is a regulated secretory pathway sorting receptor: genetic obliteration leads to endocrine disorders in Cpefat mice. Cell 88 73–83. (doi:10.1016/S0092-8674(00)81860-7)
Costes SV, Daelemans D, Cho EH, Dobbin Z, Pavlakis G & Lockett S 2004 Automatic and quantitative measurement of protein–protein colocalization in live cells. Biophysical Journal 86 3993–4003. (doi:10.1529/biophysj.103.038422)
Damholt AB, Buchan AM, Holst JJ & Kofod H 1999 Proglucagon processing profile in canine L cells expressing endogenous prohormone convertase 1/3 and prohormone convertase 2. Endocrinology 140 4800–4808. (doi:10.1210/en.140.10.4800)
Dey A, Lipkind GM, Rouille Y, Norrbom C, Stein J, Zhang C, Carroll R & Steiner DF 2004 Significance of prohormone convertase 2, PC2, mediated initial cleavage at the proglucagon interdomain site, Lys70-Arg71, to generate glucagon. Endocrinology 146 713–727. (doi:10.1210/en.2004-1118)
Dhanvantari S & Brubaker PL 1998 Proglucagon processing in an islet cell line: effects of PC1 overexpression and PC2 depletion. Endocrinology 139 1630–1637. (doi:10.1210/en.139.4.1630)
Dhanvantari S, Seidah NG & Brubaker PL 1996 Role of prohormone convertases in the tissue-specific processing of proglucagon. Molecular Endocrinology 10 342–355. (doi:10.1210/me.10.4.342)
Dhanvantari S, Arnaoutova I, Snell CR, Steinbach PJ, Hammond K, Caputo GA, London E & Loh YP 2003a Carboxypeptidase E, a prohormone sorting receptor, is anchored to secretory granules via a C-terminal transmembrane insertion. Biochemistry 41 52–60. (doi:10.1021/bi015698n)
Dhanvantari S, Shen FS, Adams T, Snell CR, Zhang C, Mackin RB, Morris SJ & Loh YP 2003b Disruption of a receptor-mediated mechanism for intracellular sorting of proinsulin in familial hyperproinsulinemia. Molecular Endocrinology 17 1856–1867. (doi:10.1210/me.2002-0380)
Dikeakos JD & Reudelhuber TL 2007 Sending proteins to dense core secretory granules: still a lot to sort out. Journal of Cell Biology 177 191–196. (doi:10.1083/jcb.200701024)
Dikeakos JD, Mercure C, Lacombe MJ, Seidah NG & Reudelhuber TL 2007 PC1/3, PC2 and PC5/6A are targeted to dense core secretory granules by a common mechanism. FEBS Journal 274 4094–4102. (doi:10.1111/j.1742-4658.2007.05937.x)
Drucker DJ 2007 The role of gut hormones in glucose homeostasis. Journal of Clinical Investigation 117 24–32. (doi:10.1172/JCI30076)
Feliciangeli S, Kitabgi P & Bidard JN 2001 The role of dibasic residues in prohormone sorting to the regulated secretory pathway. A study with proneurotensin. Journal of Biological Chemistry 276 6140–6150. (doi:10.1074/jbc.M009613200)
Fricker LD, Das B & Angeletti RH 1990 Identification of the pH-dependent membrane anchor of carboxypeptidase E (EC 3.4.17.10). Journal of Biological Chemistry 265 2476–2482.
Furuta M, Zhou A, Webb G, Carroll R, Ravazzola M, Orci L & Steiner DF 2001 Severe defect in proglucagon processing in islet A-cells of prohormone convertase 2 null mice. Journal of Biological Chemistry 276 27197–27202. (doi:10.1074/jbc.M103362200)
Garcia AL, Han SK, Janssen WG, Khaing ZZ, Ito T, Glucksman MJ, Benson DL & Salton SR 2005 A prohormone convertase cleavage site within a predicted α-helix mediates sorting of the neuronal and endocrine polypeptide VGF into the regulated secretory pathway. Journal of Biological Chemistry 280 41595–41608. (doi:10.1074/jbc.M509122200)
Gonzalez-Velez V, Dupont G, Gil A, Gonzalez A & Quesada I 2012 Model for glucagon secretion by pancreatic α-cells. PLoS ONE 7 e32282. (doi:10.1371/journal.pone.0032282)
Gorleku OA, Barns AM, Prescott GR, Greaves J & Chamberlain LH 2011 Endoplasmic reticulum localization of DHHC palmitoyltransferases mediated by lysine-based sorting signals. Journal of Biological Chemistry 286 39573–39584. (doi:10.1074/jbc.M111.272369)
Helwig M, Lee SN, Hwang JR, Ozawa A, Medrano JF & Lindberg I 2011 Dynamic modulation of prohormone convertase 2 (PC2)-mediated precursor processing by 7B2 protein: preferential effect on glucagon synthesis. Journal of Biological Chemistry 286 42504–42513. (doi:10.1074/jbc.M111.281980)
Jackson RS, Creemers JW, Farooqi IS, Raffin-Sanson ML, Varro A, Dockray GJ, Holst JJ, Brubaker PL, Corvol P & Polonsky KS et al. 2003 Small-intestinal dysfunction accompanies the complex endocrinopathy of human proprotein convertase 1 deficiency. Journal of Clinical Investigation 112 1550–1560. (doi:10.1172/JCI18784)
Jiang G & Zhang BB 2003 Glucagon and regulation of glucose metabolism. American Journal of Physiology. Endocrinology and Metabolism 284 E671–E678. (doi:10.1152/ajpendo.00492.2002)
Jutras I, Seidah NG, Reudelhuber TL & Brechler V 1997 Two activation states of the prohormone convertase PC1 in the secretory pathway. Journal of Biological Chemistry 272 15184–15188. (doi:10.1074/jbc.272.24.15184)
Jutras I, Seidah NG & Reudelhuber TL 2000 A predicted α-helix mediates targeting of the proprotein convertase PC1 to the regulated secretory pathway. Journal of Biological Chemistry 275 40337–40343. (doi:10.1074/jbc.M004757200)
Kim DE, Chivian D & Baker D 2004 Protein structure prediction and analysis using the Robetta server. Nucleic Acids Research 32 W526–W531. (doi:10.1093/nar/gkh468)
Lacombe MJ, Mercure C, Dikeakos JD & Reudelhuber TL 2005 Modulation of secretory granule-targeting efficiency by cis and trans compounding of sorting signals. Journal of Biological Chemistry 280 4803–4807. (doi:10.1074/jbc.M408658200)
Lankat-Buttgereit B, Muller S, Schmidt H, Parhofer KG, Gress TM & Goke R 2008 Knockdown of Pdcd4 results in induction of proprotein convertase 1/3 and potent secretion of chromogranin A and secretogranin II in a neuroendocrine cell line. Biology of the Cell 100 703–715. (doi:10.1042/BC20080052)
Lou H, Kim SK, Zaitsev E, Snell CR, Lu B & Loh YP 2005 Sorting and activity-dependent secretion of BDNF require interaction of a specific motif with the sorting receptor carboxypeptidase E. Neuron 45 245–255. (doi:10.1016/j.neuron.2004.12.037)
McGirr R, Ejbick CE, Carter DE, Andrews JD, Nie Y, Friedman TC & Dhanvantari S 2005 Glucose dependence of the regulated secretory pathway in αTC1-6 cells. Endocrinology 146 4514–4523. (doi:10.1210/en.2005-0402)
Mouchantaf R, Kumar U, Sulea T & Patel YC 2001 A conserved α-helix at the amino terminus of prosomatostatin serves as a sorting signal for the regulated secretory pathway. Journal of Biological Chemistry 276 26308–26316. (doi:10.1074/jbc.M102514200)
Mulcahy LR, Vaslet CA & Nillni EA 2005 Prohormone-convertase 1 processing enhances post-Golgi sorting of prothyrotropin-releasing hormone-derived peptides. Journal of Biological Chemistry 280 39818–39826. (doi:10.1074/jbc.M507193200)
Naggert JK, Fricker LD, Varlamov O, Nishina PM, Rouille Y, Steiner DF, Carroll RJ, Paigen BJ & Leiter EH 1995 Hyperproinsulinemia in obese fat/fat mice associated with a carboxypeptidase E mutation which reduces enzyme activity. Nature Genetics 10 135–142. (doi:10.1038/ng0695-135)
Normant E & Loh YP 1998 Depletion of carboxypeptidase E, a regulated secretory pathway sorting receptor, causes misrouting and constitutive secretion of proinsulin and proenkephalin, but not chromogranin A. Endocrinology 139 2137–2145. (doi:10.1210/en.139.4.2137)
Perello M, Stuart R & Nillni EA 2008 Prothyrotropin-releasing hormone targets its processing products to different vesicles of the secretory pathway. Journal of Biological Chemistry 283 19936–19947. (doi:10.1074/jbc.M800732200)
Qian Y, Varlamov O & Fricker LD 1999 Glu300 of rat carboxypeptidase E is essential for enzymatic activity but not substrate binding or routing to the regulated secretory pathway. Journal of Biological Chemistry 274 11582–11586. (doi:10.1074/jbc.274.17.11582)
Rouille Y, Westermark G, Martin SK & Steiner DF 1994 Proglucagon is processed to glucagon by prohormone convertase PC2 in αTC1–6 cells. PNAS 91 3242–3246. (doi:10.1073/pnas.91.8.3242)
Sasaki K, Dockerill S, Adamiak DA, Tickle IJ & Blundell T 1975 X-ray analysis of glucagon and its relationship to receptor binding. Nature 257 751–757. (doi:10.1038/257751a0)
Schindelin J 2008 Fiji is just ImageJ (batteries included). 2nd ImageJ User and Developer Conference, Luxemburghttp://imagejconf.tudor.lu/archive/imagej-user-and-developer-conference-2008.
Shen FS & Loh YP 1997 Intracellular misrouting of pituitary hormones and endocrinological abnormalities in the Cpefat mouse associated with a carboxypeptidase E mutation. PNAS 94 5314–5319. (doi:10.1073/pnas.94.10.5314)
Soding J 2005 Protein homology detection by HMM-HMM comparison. Bioinformatics 21 951–960. (doi:10.1093/bioinformatics/bti125)
Thomas G 2002 Furin at the cutting edge: from protein traffic to embryogenesis and disease. Nature Reviews. Molecular Cell Biology 3 753–766. (doi:10.1038/nrm934)
Tolhurst G, Reimann F & Gribble FM 2009 Nutritional regulation of glucagon-like peptide-1 secretion. Journal of Physiology 587 27–32. (doi:10.1113/jphysiol.2008.164012)
Ugleholdt R, Zhu X, Deacon CF, Orskov C, Steiner DF & Holst JJ 2004 Impaired intestinal proglucagon processing in mice lacking prohormone convertase 1. Endocrinology 145 1349–1355. (doi:10.1210/en.2003-0801)
Whalley NM, Pritchard LE, Smith DM & White A 2011 Processing of proglucagon to GLP-1 in pancreatic α-cells: is this a paracrine mechanism enabling GLP-1 to act on β-cells? Journal of Endocrinology 211 99–106. (doi:10.1530/JOE-11-0094)
Zhang C-F, Snell CR & Loh YP 1999 Identification of a novel prohormone sorting signal-binding site on carboxypeptidase E, a regulated secretory pathway-sorting receptor. Molecular Endocrinology 13 527–536. (doi:10.1210/me.13.4.527)
Zhang CF, Dhanvantari S, Lou H & Loh YP 2003 Sorting of carboxypeptidase E to the regulated secretory pathway requires interaction of its transmembrane domain with lipid rafts. Biochemical Journal 369 453–460. (doi:10.1042/BJ20020827)
(R McGirr and L Guizzetti contributed equally to this work.)