Abstract
Pancreatic β-cells respond to metabolic stress by upregulating insulin secretion, however the underlying mechanisms remain unclear. Here we show, in β-cells from overweight humans without diabetes and mice fed a high-fat diet for 2 days, insulin exocytosis and secretion are enhanced without increased Ca2+ influx. RNA-seq of sorted β-cells suggests altered metabolic pathways early following high fat diet, where we find increased basal oxygen consumption and proton leak, but a more reduced cytosolic redox state. Increased β-cell exocytosis after 2-day high fat diet is dependent on this reduced intracellular redox state and requires the sentrin-specific SUMO-protease-1. Mice with either pancreas- or β-cell-specific deletion of this fail to up-regulate exocytosis and become rapidly glucose intolerant after 2-day high fat diet. Mechanistically, redox-sensing by the SUMO-protease requires a thiol group at C535 which together with Zn+-binding suppresses basal protease activity and unrestrained β-cell exocytosis, and increases enzyme sensitivity to regulation by redox signals.
Similar content being viewed by others
Introduction
Type 2 diabetes (T2D) occurs when insulin secretion from pancreatic β-cells fails to meet peripheral demand, which is increased with insulin resistance often coincident with obesity1. While obesity and insulin resistance are major risk factors for T2D, most individuals with obesity maintain normoglycemia as insulin secretion is increased adequately through the up-regulation of β-cell function and mass2,3,4, although the relative contribution of each remains unclear1. Up-regulation of β-cell insulin secretory capacity precedes increases in islet mass in mice on a high-fat diet (HFD), and at early time points β-cell functional changes likely outweigh structural changes in increased insulin responses5,6. Loss of β-cell functional up-regulation may contribute to early progression and deterioration to T2D even with intact β-cell mass7.
Upon short-term starvation, insulin secretion is attenuated with a shift from glucose metabolism to fatty acid oxidation8. On the other hand, overnutrition leads to a compensatory increase of insulin secretion with adaptations in β-cell stimulus-secretion coupling. These include enhanced intracellular [Ca2+]i responses and up-regulation of metabolic coupling factors that potentiate Ca2+ efficacy and amplify insulin secretion9,10,11. In a longitudinal study measuring parallel β-cell insulin secretion and [Ca2+]i responses, the efficacy of Ca2+-induced insulin secretion was enhanced via Epac signaling in the early stages of prediabetes5. This may be accompanied by up-regulation of reducing equivalents12,13,14 and the exocytotic machinery per se in prediabetes15, subsequently followed by a reduced expression in T2D16.
SUMOylation is the posttranslational conjugation of small ubiquitin-like modifier (SUMO) peptides to target proteins, which can be reversed by the sentrin-specific proteases (SENPs) such as SENP1 which in β-cells localizes near sites of docked insulin granules17. The activity of SENP1 is subject to redox regulation that may involve direct oxidation-reduction of cysteine residues near the enzyme catalytic site18, although the exact structure-function relationship of SENP1 redox-sensing remains unclear. Whether ‘permissive’ or ‘regulatory’ to insulin secretion, cytosolic reducing signals help to maintain a robust pool of releasable secretory granules19,20. These signals act in part via SENP1, which modifies several exocytotic and exocytosis-related proteins to facilitate insulin granule priming without significantly affecting Ca2+ entry17,21. Indeed, glucose stimulates deSUMOylation of the exocytotic proteins synaptotagmin VII and tomosyn 1 as a mechanism to regulate exocytotitc protein complex formation21,22. Several additional exocytotic proteins also appear regulated by SUMOylation23. In a chronic overnutrition model, glucose-stimulated insulin secretion (GSIS) is impaired, in part due to partial inactivation of SENP1 by oxidative stress24,25. In contrast, it remains unknown whether, at an early compensatory or prediabetic stage, altered redox state and SENP1 activity affect β-cell functional compensation and glucose homeostasis.
We therefore assessed the role for redox signaling in the up-regulation of β-cell exocytosis that occurs early upon high fat feeding in mice, and the structure-function relationship of SENP1 redox sensing. A reduced β-cell redox, commensurate with metabolic rewiring and an increased basal oxygen consumption, enhances exocytotic capacity via a mechanism that requires SENP1. In pancreas- and β-cell–specific SENP1 knockout mice, we demonstrate the requirement for SENP1 in the maintenance of glucose homeostasis within 2 days of high fat feeding via upregulated insulin secretion. Finally, we show that C535 near the catalytic site, together with an interaction with Zn2+, is required for redox-regulation of enzyme activity and control of β-cell exocytosis.
Results
Higher β-cell secretory capacity from overweight humans and short-term HFD-fed mice
Islets from non-diabetic human donors (Supplementary Data 1) of 21–45 years of age and BMI >25 exhibited significantly elevated glucose-stimulated insulin secretion (GSIS) compared to similarly aged donors with BMI <25. This greater functionality was absent in T2D (BMI >25) (Fig. 1A). Depolarization-induced β-cell exocytosis measured by patch-clamp electrophysiology is amplified by glucose17. Intriguingly, β-cells from the donors with BMI >25 did not show any difference in depolarization-induced exocytosis compared to BMI <25 at 5- or 10-mM glucose. Instead, the exocytotic response was elevated in the non-diabetic donors with BMI >25 at 1 mM glucose (Fig. 1B). This means that the β-cells of higher-BMI donors upregulate their capacity to mount an exocytotic response at low glucose, to the same upper limit observed when the cells are pre-cultured at elevated glucose. That capacity does not translate to more insulin secretion at low glucose, however, but instead to an increased secretory response upon subsequent stimulation with glucose at levels sufficient to elicit an electrical and Ca2+ response.
As with insulin secretion, this greater exocytotic capacity at 1 mM glucose was missing in β-cells of donors with T2D and BMI >25 (Fig. 1B). These differences were not as clear in islets and β-cells from female donors (Supplementary Fig. S1) or older donors (age >45; Fig. 1C, D), the latter perhaps being confounded by increasing stress responses, β-cell immaturity phenotypes, and generally reduced insulin secretion seen with age26. Notably however, insulin secretion and β-cell exocytosis were still lower in T2D (Supplementary Fig. S1, Fig. 1C, D).
To investigate β-cell functional changes that occur early under metabolic stress, we fed male C57BL/6NCrl mice either a chow diet (CD) or HFD for 2 days. Mice fed a 2-day HFD maintained normal glucose tolerance in an intraperitoneal glucose tolerance test (IPGTT) (Fig. 1E), along with increased plasma insulin (Fig. 1F). GSIS was increased from islets of 2-day HFD mice (Fig. 1G), without any effect on insulin content (Fig. 1H). Comparatively, after a 4-week HFD, mice were glucose intolerant (Supplementary Fig. S2A) with increased plasma insulin (Supplementary Fig. S2B), increased islet insulin content, and enhanced in vitro insulin secretion (Supplementary Fig. S2C–E). In β-cells from 2-day (Fig. 1I) and 4-week (Supplementary Fig. S2F) HFD-fed mice, exocytotic responses are increased at low glucose, to the same level as at elevated glucose. At 10 mM glucose there were no differences in voltage-activated Ca2+ entry (Fig. 1J) and intracellular Ca2+ responses (Fig. 1K; Supplementary Fig. S2G), consistent with previous demonstrations of ‘increased Ca2+ efficacy’ under short-term HFD5. Indeed, under conditions that ‘clamp’ intracellular Ca2+ responses to assess glucose-dependent amplification of insulin secretion27, islets from 4-week HFD fed mice exhibited enhanced KCl-stimulated insulin secretion at low glucose (Supplementary Fig. S2E).
Shifts in metabolic pathway expression after short-term HFD
To gain insight into the mechanism contributing to increased insulin secretion following short-term HFD28, we performed RNA sequencing on FACS-sorted β-cells from 2-day CD and HFD mice (Supplementary Data 2). We found 213 differentially expressed (DE) genes (Fig. 2A). Enrichment analysis identified pathways (PW) such as histone methylation29, MAPK signaling, cholesterol biosynthesis, insulin receptor signaling, and ER-associated misfolded protein response (Fig. 2B; Supplementary Data 3)29,30. Histone methylation, the most significantly enriched pathway, can interact or overlap with genes in other enriched pathways, such as MAPK signaling, consistent with the ability of methylation to adapt insulin secretion through MAPK and regulation of glucose metabolism29. All DE genes were submitted to the STRING database31 for protein-protein interaction. Gene set enrichment analysis (GSEA) showed that up-regulated genes were enriched for glycolysis (Tpi1, Pgk1, Pkm…) and oxidative phosphorylation (Uqcrq, Vcp, Atp5g3…) pathways (Fig. 2C; Supplementary Data 4). Interestingly, the cholesterol biosynthesis pathway (Fdft1, Hsd17b7, Msmo…) was significantly down-regulated (Fig. 2C, D). Cholesterol biosynthesis relies heavily on NADPH consumption32, and its inhibition may enhance insulin secretion by increasing NADPH33. Therefore, the downregulation of cholesterol biosynthesis genes could spare cytosolic NADPH, which might be built up by increased glucose metabolism, for the amplification of insulin release (Fig. 2D).
Increased β-cell exocytosis after 2-day HFD requires a reducing signal
Reducing signals ensure a robust pool of releasable insulin granules17. Islets from 2-day HFD-fed mice appeared to have higher basal O2 consumption and proton-leak measured by Seahorse assay (Fig. 3A). We confirmed an increased basal O2 consumption rate in 2-day HFD islets by fluorescence lifetime measurement (Fig. 3B). In islets of mice expressing the cytosolic redox sensor Cyto-roGFP2-Orp134, we found that the cytosol in β-cells from 2-day HFD-fed mice was more reduced compared to control mice fed CD (Fig. 3C). A reduced redox state appears required for enhanced β-cell exocytosis since this was recapitulated in β-cells of CD-fed mice upon direct intracellular dialysis of reduced glutathione (GSH), which could not increase exocytosis further in β-cells of HFD-fed mice (Fig. 3D) and was reversed by direct intracellular application of H2O2 (Fig. 3E). The redox enzyme glutaredoxin 1 (GRX1) is required for GSH- and NADPH-dependent facilitation of insulin exocytosis35,36, and we showed this likely acts via control of SENP1 activity (Fig. 3F)17. We observed an upregulation of Senp1 expression by qPCR in islets after 2-day HFD, which was lost by 8-weeks (Fig. 3G). Direct intracellular dialysis of active SENP1 catalytic domain increases exocytosis in β-cells from CD-fed mice but, like GSH, cannot increase exocytosis further in the β-cells of HFD-fed mice (Fig. 3H).
In pSENP1-KO mice24, males become rapidly glucose intolerant compared with control littermates, with significant IP and oral glucose intolerance after 2-day HFD (Fig. 4A–D). Loss of β-cell SENP1 prevents the up-regulation of exocytosis after either a 2-day (Fig. 4E) or 4-week (Fig. 4F) HFD. Similar to the pSENP1-KO line, male βSENP1-KO mice24 become rapidly intolerant of IP glucose compared with control littermates upon 2-day HFD (Fig. 4G, H), with an impaired plasma insulin response (Fig. 4I). Consistent with our observation in human islets (Supplementary Fig. S1), these responses were less obvious in female mice from both the pSENP1-KO and βSENP1-KO lines (Supplementary Figs. S3 and S4), which are more resistant to the development of insulin resistance upon HFD37. Also, after 4-week HFD, oral glucose intolerance became more prominent in the βSENP1-KO females (Supplementary Fig. S4I) and males (Fig. 4J–L, Supplementary Fig. S5), although again the females were generally more insensitive, consistent with our recent report at 8 weeks of HFD in both the pSENP1- and βSENP1-KO models24.
Redox regulation of exocytosis requires SENP1 C535
SENP1, a cysteine protease, is subject to redox regulation via a thiol side chain of the catalytic cysteine C603 or possibly other cysteines18. C603 is the most conserved cysteine across SENP isoforms and species, while C535 is the least conserved near the active site (Fig. 5A). Although redox regulation of intra-molecular disulfide bonds may control cysteine protease activity38, the key cysteines of the SENP1 catalytic domain appear too distant from the catalytic C603 to facilitate such regulation39 (Supplementary Fig. S6A). We therefore generated a series of recombinant SENP1 catalytic domain proteins with cysteine-to-serine substitutions and measured SUMO-protease activity in vitro (Supplementary Fig. S6B). While mutation of the catalytic cystine (C603S) completely abolished SENP1 activity as expected, the C535S mutations maintained a 2-fold higher activity compared to the SENP1 WT and other mutants (Fig. 5B). SENP1 activity was also abolished in H533S or D550S mutants, the latter being partially rescued by concomitant C535S mutation (Fig. 5C). Consistent with proton shuffling from C603 via the nearby H533 towards D550 as a potential mechanism of activation (Supplementary Fig. S6C)40, competing away suppressive effect of C535 interaction with H533 by excess histidine (1 mM) indeed prevents inactivation of SENP1 by 0.1 mM H2O2 (Fig. 5D).
In the original experiments shown in Fig. 5B, we believe differences in activity resulted from differential sensitivity to oxidation upon exposure to the air. Under conditions where SENP1 is fully activated (minimal oxidization and absence of divalent cations), the activities of SENP1 WT and SENP1 C535S were similar, and while the WT was strongly inhibited by H2O2, the C535S was not (Fig. 5E). Intracellular dialysis of SENP1 WT rescued the exocytotic defect (at 5 mM glucose) in β-cells from pSENP1-KO mice (Fig. 5F). While this rescue effect was reversed by co-dialysis of 10 μM H2O2, rescue of exocytosis in the pSENP1-KO by SENP1 C535S was resistant to H2O2 (Fig. 5F). Similar results were observed when SENP1 WT or SENP1 C535S were dialyzed into human β-cells (Supplementary Fig. S7A). The activity of SENP1 WT could be increased by GSH together with GRX1, which uses GSH as a cofactor to modulate protein thiols and disulfide bonds41, while SENP1 C535S appeared maximally activated (Fig. 5G). Altogether these results suggest that C535 is required for SENP1 sensitivity to redox and its function in augmenting insulin exocytosis.
We wondered whether an additional mechanism contributes to suppression of baseline SENP1 activity. Zn2+-cysteine is a critical mediator of redox-regulation and protease activity42. In silico analysis predicts that C603-C535-H533 is a Zn2+-binding site in SENP1 (Supplementary Fig. S6D), and we found that Zn2+, and to a lesser extent Ni2+ (present during re-folding in some of our earlier experiments) inhibited the activity of SENP1 WT but not SENP1 C535S (Fig. 6A). The IC50 for SENP1 inhibition by Zn2+ is right-shifted by nearly-30-fold in the C535S mutant, indicative of allosteric interaction of C535 with Zn2+ to inhibit SENP1 (Fig. 6B). This can be readily reversed by chelation with EDTA (Fig. 6C) and was not due to non-specific effect of Zn2+ on His×6 tag from recombinant SENP1 (Supplementary Fig. S6E).
While Zn2+ inhibits SENP1 WT with an IC50 (0.22 μM), similar to the well-established Zn2+-inhibition of enzymes43 including proteases such as the caspases (IC50 = 0.1–3.4 μM;44,45), this is above the free cytoplasmic Zn2+ (sub-nM) in eukaryotic cells46. Compared to the in vitro Zn2+ titration assay, regulation of Zn2+ activity in a cellular environment is more complex, yielding compartmentalized Zn2+ signaling events with defined spatiotemporal characteristics. We examined the potential involvement of metallothionein, a Zn2+-binding protein participating in Zn2+-storage and distribution to other proteins47. Metallothionein is abundantly present in the pancreas (average 264 μg/mL or 43.6 μM in human pancreas48). Because MT1X interfered with the binding of Hisx6 tagged SENP1 substrate (Hisx6-SUMO-mCherry) to Nickel-NTA agarose for the SENP1 assay, (see Methods), we developed an alternative assay that does not require Nickel-NTA agarose (Supplementary Fig. S6E). Zn2+-bound metallothionein 1X (MT1X), a subtype present in β-cells, negatively regulates insulin secretion49 and suppresses SENP1 activity similarly to free Zn2+ (Fig. 6D), while Zn2+-free MT1X; does not (Fig. 6E). Zn2+ chelation with EDTA also reverses SENP1 inhibition by MT1X (Fig. 6F). In β-cells of the βSENP-WT mice, dialysis of 100 µM ZnCl2 impaired exocytosis without decreasing Ca2+ currents, to a level similar to that in the βSENP-KO (Fig. 6G; Supplementary Fig. S7B). While 100 µM ZnCl2 did not decrease exocytosis further in the βSENP-KO, higher concentrations of ZnCl2 (1 mM) dramatically decreased exocytosis in both groups possibly by inhibiting Ca2+ currents directly50 (Fig. 6G, Supplementary Fig. S7B). Finally, in the presence of Zn2+ SENP1 was more sensitive to inactivation by H2O2 (Fig. 6H), and more robustly activated by GSH and GRX1 (5.3-fold) while the C535S mutant remained more active (Fig. 6I).
Discussion
β-cell compensation and decompensation is a determinant of early T2D progression51, yet the underlying mechanisms remain unknown. Here we demonstrate that a reduced cytosol acting via SENP1 facilitates an upregulation of β-cell function to maintain glucose tolerance very early after high fat feeding. Redox-sensing by SENP1 relies on C535 and is tuned by an interaction with Zn2+, possibly delivered via MT1X (Supplementary Fig. S7C). Notably, the enhanced depolarization-induced exocytosis observed at low glucose soon after 2-day HFD is recapitulated by intracellular dialysis of reducing molecules or active SENP1, which exerts a similar priming effect21, and is lost upon β-cell knockout of SENP1. This is not expected to increase basal insulin secretion in the absence of an increased intracellular Ca2+ concentration. Instead, an increased insulin granule priming at low glucose will increase the size of the releasable pool of insulin granules on which subsequent glucose-dependent Ca2+ responses can act, resulting in increased glucose-stimulated insulin secretion. This is consistent with the enhanced efficacy of Ca2+ in 1-week HFD-fed mice5 and pre-diabetic db/db mice15.
Redox signals are important regulators of insulin secretion and functional compensation12,17, although there is debate as to whether reactive oxygen species52 or reducing signals53 augment insulin secretion. Possibly both are true, depending on timing, localization, and mechanisms involved, such as the modulation of excitability by reactive oxygen species via K+ channels54 or by the direct regulation of membrane fusion by reducing signals35. Reducing signals may also be ‘permissive’ through their role in maintaining or setting a robust pool of releasable granules before subsequent glucose stimulation52. We observed a remodeling of metabolic pathways in RNAseq data and increased proton leak and basal O2 consumption after 2-day HFD. Coincident with this, we observed a more reduced cytosol in the β-cell. Although the underlying mechanism remains unclear, down-regulation of cholesterol biosynthesis as seen in the RNAseq data may contribute to an increase in cytosolic NADPH33. Increased mitochondrial proton-leak may also contribute to higher insulin secretion55, and increased sub-maximal glucose-fueled mitochondrial TCA cycle flux may generate higher cytosolic reducing equivalents56. We and others have shown that NADPH, produced by either the mitochondrial export of reducing equivalents or the pentose phosphate pathway57,58, augments Ca2+-triggered insulin exocytosis17,35,36. One caveat is that the Cyto-roGFP2-Orp1 redox-sensor reports the steady state balance between oxidizing and reducing signals20,59. The reduced redox could result from either decreased H2O2 production or increased reducing equivalents. Adaptive glucose metabolism may also enhance de novo GSH biosynthesis independent of NADPH generation60. Therefore, increased de novo GSH biosynthesis could contribute to a compensatory increase in insulin secretion. Nonetheless, a reduced cytosolic redox state appears critical for the increased β-cell exocytosis seen after 2-day HFD since it could be reversed by H2O2 and replicated by GSH or NADPH.
The SUMO-protease activity of SENP1 is required for its ability to augment insulin exocytosis (Fig. S7A), consistent with our previous report that SENP1 can rescue exocytosis suppressed by SUMO1 or the SUMO-ligase Ubc9, and deSUMOylates synaptotagmin VII21. Compared to other isoforms, SENP1 is a redox-sensitive protease with an unique interaction between a free thiol on the catalytic cysteine (C603) and other cysteines18. C613 is reported to be a ‘positive’ redox sensor by forming an intermolecular disulfide bond with the C603 and protect SENP1 from irreversible oxidation18. We find however that mutation of C613 had little effect on SENP1 activity, but rather that C535 appears to restrain SUMO-protease activity. The C535S mutant had 2-fold higher activity and was resistant to inactivation by H2O2, indicating that C535 is a ‘negative’ modulator of redox sensing. Indeed, the atomic distance between thiols of C535 and C603 prevents formation of an intra-molecular disulfide bond, consistent with the absence of disulfide bonds in available SENP1 crystal structures39. C535 also does not control enzyme activity via the formation of intermolecular SENP1 disulfide bonds18. Thus, it seems most likely that C535 impacts SENP1 redox-sensing via a proton transfer pathway, and deprotonation of the thiol group on C603 to activate SENP1 (Supplementary Fig. S6C). Specifically, proton transfer from the thiol group of C603 towards the carboxyl group of D550 via H533 results in the conversion of a thiol on C603 to an active thiolate in SENP140. C535 may donate a proton to either H533 or D550 to competitively restrain proton transfer from C603, leading to inhibition of SENP1, under oxidizing conditions.
SENP1 couples reducing power to insulin secretion17,21,22,61. Interestingly, we find that Zn2+ allosterically inhibits SENP1 in a manner that is dependent upon C535. Mechanistically, the predicted Zn2+-binding site includes C603, H533 and C535, which is one of the most ubiquitous Zn2+ binding motifs62. A coordinate bond between Zn2+ and these residues likely further restricts proton transfer. Zn2+ coordination is expected to lower the pKa of a thiol by two orders of magnitude, facilitating the formation of a thiolate63. Indeed, thiol-binding Zn2+ readily dissociates in the presence of oxidant, leaving a highly reactive thiolate for an oxidant64. Therefore, interaction between C535 and Zn2+ sensitizes SENP1 to oxidative inhibition and allows robust reactivation by reducing signals, not dissimilar to the reported interaction between reactive oxygen species and Zn2+ in the regulation of protein tyrosine phosphatases65. This is consistent with the recent demonstration of Zn2+-lactate binding at C535 of SENP166 which, together with the recent finding that lactate is higher than expected in some human β-cells67, suggests an additional possible mechanism for tuning β-cell exocytosis.
Zn2+ inhibits exocytosis via SENP1 with a calculated free Zn2+ at ~0.08 µM. However, cytosolic free Zn2+ levels are not perturbed by glucose stimulation68. Zn2+ binding of SENP1 may occur at the time of protein folding in the ER, or via direct interaction with Zn2+-binding/buffering proteins such as MT1X which have been shown to control islet Zn2+ fluctuations69 and inhibit glucose-stimulated insulin secretion via a so-far unidentified interaction with the exocytotic machinery49. While free Zn2+ is in the low-nM-range in the β-cell cytosol68, the abundant Zn2+-binding MT1X likely transfers Zn2+ to SENP1, tuning its redox-sensitivity and activity. During β-cell compensation, MT1X was downregulated to increase insulin secretion49. The combined effect of less Zn2+ transferred to SENP1, upregulation of Senp1 expression, and a reduced redox state would be expected to result in a more active SENP1 and augmentation of β-cell exocytosis.
Knockout of islet and β-cell SENP1, in two separate Cre-driver models, led to IP glucose intolerance after 2-day HFD associated with a loss of insulin response. This was more obvious in males, which are known to be more sensitive to HFD37 and consistent with differences we observed in human islets, where insulin secretion and exocytosis appeared most impacted by BMI in younger males. However, oral glucose tolerance was not different following 2-day HFD in most cases and in contrast with the 4-week HFD where the SENP1-KO showed selectively impaired oral, not IP, glucose tolerance. This discrepancy highlights the adaptive changes that occur during different stages of high fat feeding70. The incretin response is paramount in limiting oral glucose intolerance after long-term HFD24,71, where incretin release from the gut24 and β-cell incretin receptors70 are upregulated. β-cell SENP1 is required for these compensational changes to maintain incretin-stimulated insulin secretion and oral glucose tolerance24,72. However, the β-cell GLP-1 receptor is not necessary for oral glucose tolerance under normal conditions73, similar to the role of SENP1 for oral glucose tolerance under the same conditions24 or after 2-day HFD. Additionally, the extra-islet action of GLP-1 could mask the effect of β-cell SENP1, such that OGTT is similar in βSENP1-KO and -WT mice73. Finally, we note that the less cell-type-selective pSENP1-KO mice, which have impaired GIP-stimulated glucagon secretion24, did show impaired oral glucose tolerance after 2-day HFD, indicating the potential importance of α-to-β cell communication under this condition and as previously reported71.
Thus, we show that insulin secretion is upregulated from islets from human donors with increased BMI, at early/mid-age, and from mouse islets shortly after high fat feeding. This is not due to an increased Ca2+ response in β-cells, but to an increase in the availability of releasable insulin granules that is promoted at low glucose (i.e., prior to the glucose-stimulated secretion of these granules). This requires a reducing environment and activity of the SUMO-protease SENP1 via a sensing mechanism involving the Zn2+-dependent redox control of enzyme activity by C535 near the enzyme catalytic site. In mice this enhanced insulin secretory capacity acts to maintain glucose tolerance in the early stages of high fat feeding.
Methods
All human islet studies were approved by the Human Research Ethics Board (Pro00013094; Pro00001754) at the University of Alberta and all families of organ donors provided written informed consent. All studies with mice were approved by the Animal Policy and Welfare Committee (AUP00000291) at the University of Alberta.
Islet isolation and cell culture
Human islets were isolated at the Alberta Diabetes Institute IsletCore (www.isletcore.ca) and cultured overnight in DMEM (11885, Gibco) supplemented with L-glutamine, 110 mg/l sodium pyruvate, 10% FBS (12483, Gibco), and 100 U/ml penicillin/streptomycin (15140, Gibco)74. A total of 265 non-diabetic (ND) donors and 34 T2D donors (HbA1c >6.5) (Supplementary Data 1) were examined. Separation of donors by BMI and age was based on clinical diagnostic criteria for overweight, and previous studies by us74 and others75 demonstrating declines in human islet insulin secretion between the ages of 40 and 50.
C57BL/6NCrl mice (Charles River Laboratories) of both sexes were fed standard chow diet (5L0D*, PicoLab Laboratory Rodent Diet) or 60% high-fat diet (S3282, Bio-Serv) at 12 weeks of age for 2-days or 4-weeks. Pdx1-Cre mice (B6.FVB-Tg (Pdx1-Cre)6 Tuv/J, Jackson Lab, 014647) on a C57BL/6N background and Ins1-Cre mice on a mixed C57BL/6J and SV129 background76 were crossed with Senp1-floxed mice on a C57BL/6J background to generate gut/pancreas specific (Pdx1-Cre+;SENP1fl/fl - pSENP1-KO) and β-cell-specific knockouts (Ins1-Cre+;SENP1fl/fl - βSENP1-KO)17,77. Control littermates Pdx1-Cre+;SENP1+/+ mice (pSENP1-WT) and Ins1-Cre+;SENP1+/+ mice (βSENP1-WT) were used for experiments. All mice were 12-16 weeks of age at the time of experiments and housed under a 14/10 light-dark cycle with ad libitum access to food. Genotypes and SENP1 expression were confirmed as previously24. Following euthanasia by CO2 inhalation, mouse islets were isolated by collagenase digestion and purified by histopaque density gradient centrifugation before hand-picking24. All mouse islets or single cells were cultured in RPMI-1640 (11875, Gibco) with 11.1 mM glucose, 10 % FBS (12483, Gibco), 100 units/mL penicillin, and 100 μg/mL streptomycin (15140, Gibco).
Insulin secretion
After overnight culture, static insulin secretion was measured as previously74 at the glucose concentrations indicated. Diazoxide and KCl concentration were 100 μM and 30 mM respectively. For dynamic insulin responses 25 islets were pre-perifused for 30 min at 2.8 mM glucose followed by 16.7 mM glucose and KCl (30 mM) at a sample collection rate of 100 μl/min every 2–5 min24. Insulin content was extracted with acid/ethanol. The samples were stored at −20 °C and assayed by using Insulin Detection Kit (STELLUX® Chemi Rodent Insulin ELISA kit; STELLUX® Chemi Human Insulin ELISA kit, Alpco).
Exocytosis measurements
Human and mouse islets were dispersed to single cells in Ca2+-free buffer and cultured in 5.5 mM glucose DMEM and 11 mM glucose RPMI, respectively as above. After overnight culture, dispersed cells were preincubated at 1 mM (human) or 2.8 mM (mouse) glucose for 1 h and patched in bath solution containing (in mM): 118 NaCl, 5.6 KCl, 20 TEA, 1.2 MgCl2, 2.6 CaCl2, 5 HEPES at different glucose conditions (1, 2.8, 5, 10 mM) with a pH of 7.4 (adjusted by NaOH) at 32-35 oC. Whole-cell capacitance was recorded with the sine+DC lock-in function of an EPC10 amplifier and PatchMaster software (HEKA Electronics). Exocytotic responses and inward Ca2+ currents were measured 1-2 min after obtaining the whole-cell configuration in response to ten 500 ms depolarizations to 0 mV from a holding potential of −70 mV. Changes in capacitance and integrated Ca2+ charge entry were normalized to cell size (fF/pF and pC/pF, respectively). The intracellular solution contained (in mM): 125 Cs-Glutamate, 10 CsCl, 10 NaCl, 1 MgCl2, 5 HEPES, 0.05 EGTA, 3 MgATP and 0.1 cAMP with pH = 7.15 (pH adjusted with CsOH).
In the intracellular dialysis experiments, measurements were 2-3 min after obtaining the whole-cell configuration. Our previous work showed this is sufficient for rapid dialysis of both small molecules and larger peptides such as SENP121 and given the large volume of the pipette solution with respect to the β-cell cytosol we assume a homogenous distribution of infused molecules. Compounds or recombinant enzymes were added to pipette solution as indicated. For mouse β cells, these were 10 μM H2O2 (Fisher Scientific, H325-500), 10 μM reduced glutathione (GSH, Sigma), 4 μg/ml cSENP1 (Enzo Life Technologies) or glutathione-S-transferase (GST) (Enzo Life Technologies) as a control protein. For human β-cells, these were 4 μg/ml GST, wild type SENP1, C535S and C603 mutant recombinant proteins (see below) pretreated with or without 200 μM H2O2 for 10 min. Human β-cells were identified by immunostaining for insulin (Dako anti-insulin, #IR002 at 1:5 dilution, with Invitrogen AF488 secondary, #A11073 at 1:200 dilution). Mouse β-cells were identified by size (>4 pF) and a Na+ channel half-maximal inactivation at around −90 mV78.
Glucose homeostasis
Oral glucose tolerance test (OGTT) and intraperitoneal tolerance test (IPGTT) were performed as previously described24. For OGTT and IPGTT on chow diet or 2-day HFD, glucose concentration was 1 g/kg dextrose. After 4-week HFD mice became glucose intolerant and 1 g/kg dextrose produced blood glucose values above the maximum of the glucometer, and we therefore used 0.5 g/kg dextrose in these experiments. Tail blood was collected at indicated times for insulin (STELLUX® Chemi Rodent Insulin ELISA kit, Alpco) and glucose measurement. ITT was performed with a concentration of 1 U/kg Humulin R (Eli Lilly).
Intracellular Ca2+ imaging
Mouse islet cells were cultured on 35 mm glass bottom dishes with 10 mm micro-well (Cellvis) overnight in 11 mM glucose RPMI media (11875, Thermo Fisher). Cells were pre-incubated with Fura-2AM (1 μM) for 15 min and perifused with KRB solution at indicated glucose level and imaged at 0.5 Hz. Fluorescence signal was excited at 340/380 nm (intensity ratio 20:8) and detected at an emission light of 510 nm using Life Acquisition software (Till Photonics) on an inverted microscope (Zeiss Axioobserver, Carl Zeiss Canada Ltd.) equipped with a rapid-switching light source (Oligochrome; Till Photonics, Grafelfing, Germany). β-cells were marked and identified by immunostaining and fluorescence ratios were calculated using ImageJ (NIH).
Fluorescence activated cell sorting (FACS) of β-cells and mRNA sequencing
Islet α- and β-cells were identified by ZIGIR vs Ex4-Cy5 through FACS analysis as above79. Total RNA was isolated from sorted β-cells using TRIzol reagent (Invitrogen) according to the manufacturer’s protocol and cDNA libraries were prepared using Illumina’s TruSeq Stranded mRNA Library Prep kit. The cDNA libraries were validated by TapeStation DNA 1000 High Sensitivity assay (Agilent) and quantified by Qubit dsDNA High Sensitivity assay (Invitrogen). High quality libraries were submitted to the UT Southwestern Next Generation Sequencing Core facility and RNA sequencing was performed using Illumina’s NextSeq 500 High Output instrument. Raw reads were processed to transcripts per million (TPM) and differentially expressed (DE) genes were initially determined by using the edgeR package. Genes with P-value <0.1 (not corrected for multiple testing), coefficient of variation (CV) of less than 1, same change of direction (either positive or negative) and specific gene counts range (mean gene counts in at least one experimental group should be higher than 5 RPKM) were considered differentially expressed between CD and HFD. All differentially expressed genes are used for downstream analyses and listed in Supplementary Data 2.
Differentially expressed gene function and enriched pathways were identified with Metascape (http://metascape.org/). KEGG Pathway, GO Biological Processes, Reactome Gene Sets, CORUM, TRRUST and PaGenBase were included in pathway enrichment analysis with a p-value <0.01. Terms with an enrichment factor >1.5 and a similarity of >0.3 are grouped into a cluster represented by the most significant term80. To examine protein-protein interaction (PPI), STRING database (https://string-db.org/) was used and only PPIs with interaction score higher than 0.70 were retrieved and linked in the circus plot by using R-studio (circlize package)31,81. To run an unbiased gene set enrichment analysis (GSEA)82, all Genes with CV of less than 1, same change of direction and specific gene counts range (at least one experimental group should be higher than 5 RPKM) were submitted and listed in Supplementary Data 3. H.all.v7.4.symbols.gmt[Hallmarks] gene dataset was used as reference data set. Enriched gene sets were considered significant with gene size over 50, false discovery rate less than 0.05, nominal p-value less than 0.01.
Redox measurement
Measurements of the intracellular H2O2 levels were made using a redox histology approach, as previously described34,83,84. Briefly, batches of 30 islets were collected and immersed in 50 mM N-ethyl-maleimide (NEM) dissolved in PBS for 20 min for sensor chemical fixation. The NEM was then removed, and islets fixed in 4% paraformaldehyde for 30 min. After paraformaldehyde removal, the islets were incubated in 100-µL HepatoQuick (Roth, Karlsruhe, Germany) mixed with human citrate plasma (1:2 v/v) and 1 % CaCl2 for 1 h at 37 °C. Clots were then incubated in 95 % ethanol at 4 °C overnight and subsequently dehydrated in ethanol prior to paraffin embedding. The paraffin embedded clots were cut in 3 µm thick sections with a semi-automated rotary microtome (Leica Biosystems, Wetzlar, Germany) and placed on silanized glass slides. Images of all the islets in each slide were obtained by the Axio Observer 7 fluorescence microscope (Zeiss, Oberkochen, Germany) with a 20× objective using excitation 405 nm and 488 nm, emission 500-530 nm. The images were analyzed with ImageJ Fiji Software, and the ratio (405/488 nm) was used to compare different groups.
Oxygen consumption
Agilent’s seahorse XFE24 Analyzer with the islet capture microplates was used for oxygen consumption measurements85. Briefly, 70 islets per well were assayed. Islets were sequentially treated with 2.8 mM glucose, 16.7 mM glucose, 5 µM oligomycin, 5 µM FCCP and 5 µM rotenone/antimycin A. Each experiment was run with a 3-min mix, 2-min wait and 3-min measurement period for all data points. Each biological replicate had three technical replicates and oxygen consumption rate was normalized to islet protein.
For oxygen consumption measured with the Fluorescence Lifetime Micro Oxygen Monitoring System (FOL/C3T175P, Instech Laboratories Inc.) around 200 islets incubated overnight were washed 3x with serum-free RPMI media at 2.8 mM glucose, added to the chamber for data collection for 20 min, and then collected for normalization of oxygen consumption to DNA content measured using a Quant-iT™ PicoGreen™ dsDNA Assay Kits (Invitrogen).
Preparation of recombinant wild-type and mutant SENP1
Human SENP1 catalytic domain cDNA (NM_001267594/NP_001254523.1, 419-644 residues) was cloned in the downstream of a His×6 coding sequence, between Xba-I and BamH-I sites of the pT7JLH plasmid to add a His×6 N-terminal tag to SENP1. Primers are listed in Supplementary Data 5. Site directed mutagenesis by primer extension was used to make SENP1 mutants, C535S, C560S, C603S, C608S, C613S, C535S/C608S, C603S/C613S, C535S/H533, C535S/D550, C603S/H533, C603S/D550, C608S/H533 and C608S/D550. In brief, two sets of overwrapping mutagenesis primers were used to amplify SENP1 catalytic domain using Phusion DNA polymerase. The SENP1 cDNA was cloned between Xba-I and BamH-I sites using Gibson assembly.
To produce wild type SENP1 without His×6 tag, human SENP1 catalytic domain cDNA was also cloned into immediately downstream of T7 promoter, between Nde-I and BamH-I of the pT7JLH plasmid, to avoid addition of a His×6 tag coding sequence.
For expression, Rosetta strain of E. coli (Novagen) was transformed with each expression vector and grown in 50 ml LB broth containing 100 μg/ml ampicillin at 37 °C with shaking until A600 reached between 0.6 and 0.8. Each SENP1 protein was overexpressed by addition of 1 mM IPTG to the culture. After 2 h induction, the E. coli pellet was harvested by centrifugation at 10,000 × g for 10 min. Inclusion bodies of SENP1 proteins were prepared from the pellet86.
Each SENP1 protein with a His×6 tag was solubilized in 6 M guanidine HCl, 20 mM Tris, 0.1% Tween20, 1 mM 2-mercaptoethanol, pH = 8, and captured with a HisPur Nickel-NTA resin column (1.5 × 3 cm). The column was washed with 4 M guanidine HCl, 20 mM Tris, 0.1% Tween20, 1 mM 2-mercaptoethanol, pH = 8 and then SENP1 protein was eluted with 4 M guanidine HCl, 50 mM sodium acetate, 0.05% Tween20, pH = 4. To remove Ni2+ leached out from nickel-NTA resin, the concentration of each SENP1 protein was adjusted to 4 mg/ml with 4 M guanidine HCl, 20 mM Tris, 0.1% Tween20, 1 mM DTT, 50 mM EDTA and 0.2 M Tris, pH = 8. Ten volumes of water were added to dilute the 4 M guanidine HCl to 0.4 M, which causes the precipitation of SENP1 protein. The precipitated SENP1 protein without nickel was recovered by centrifugation at 8000 × g for 20 min.
The inclusion bodies of wild type SENP1 without a His×6 tag, was further washed twice with 1 M guanidine HCl, 2% Triton X-100, 0.1 M Tris, pH = 8. The inclusion bodies were dissolved in 4 M guanidine HCl, 20 mM Tris, 0.1 % Tween20, 1 mM DTT, 50 mM EDTA and 0.2 M Tris, pH = 8. The SENP1 without a His×6 tag was precipitated by addition of ten volumes of water. The SENP1 without a His×6 tag was recovered by centrifugation at 8000 × g for 20 min.
The precipitated SENP1 was dissolved in 6 M guanidine HCl, 20 mM Tris, 0.2 M Arginine, 1 mM DTT, pH = 7.2 and the absorbance at 280 nm was measured to estimate protein concentration. SENP1 proteins were stored at −80 °C until in vitro refolding. Refolding of SENP1 protein was carried out in vitro by gradual decrease of guanidine HCl concentration with dialysis87 at 4 °C. The concentration of SENP1 protein was adjusted to 1 mg/ml with 6 M guanidine HCl, 20 mM Tris, 0.1% Tween20, 0.2 M arginine, pH = 7.2, and for each refolding 4-5 mg of SENP1 was used. The cysteines of SENP1 proteins were reduced to have sulfhydryl side chains with 10 mM DTT for 1 h at room temperature and dialyzed against 0.5 M guanidine HCl, 0.2 M arginine, 0.1 M Tris, 2 mM cysteine, 0.2 mM cystine, 0.05% Tween20, pH = 7.2 for 24 h, and then against the same solution without guanidine HCl for 24 h. SENP1 proteins were refolded with/without 0.1 mM ZnCl2, and then ZnCl2 was removed by dialysis. Alternatively, SENP1 and mutants were prepared without ZnCl2 in the refolding buffer then incubated with 1 µM ZnCl2 for 1 h to suppress SENP1 activities prior to reactivation by 0.1 mM GSH + 10 μg/ml GRX1. SENP1 proteins were finally dialyzed against SENP1 assay buffer, 20 mM Tris, 0.1 M sodium chloride, 0.05 % Tween20, pH = 7.2 for 24 h. SENP1 proteins, used for the intracellular dialysis experiments, were prepared without 0.05% Tween 20. The final dialysis buffer was changed to fresh buffer twice. The precipitated SENP1 proteins were removed by centrifugation at 8000 × g for 10 min. Absorbance at 280 nm of the supernatant were measured to estimate the concentration of SENP1 proteins.
For the experiments to study the effect of divalent transient metals on SENP1 activity, 0.1 mM of CaCl2, CoCl2, NiCl2 and ZnCl2 was added in the refolding solution, and then removed by extensive dialysis against SENP1 assay buffer, 20 mM Tris, 0.1 M sodium chloride, 0.05% Tween20, pH = 7.2
Preparation of recombinant glutaredoxin 1 and metallothinein 1X
The coding sequence of human GRX1 cDNA(NM_002064.3/NP_002055.1, 1-106 residues) was cloned in the downstream of a His×6 tag coding sequence, between Xba-I and BamH-I sites of the pT7JLH plasmid88. Full size human MT1X cDNA (NM_005952.4/ NP_005943.2, 1-61 residues) was cloned into immediate downstream of T7 promoter, between Nde-I and BamH-I sites of the pT7JLH plasmid88, to avoid addition of a His×6 tag coding sequence. Recombinant GRX1 was produced in E. coli, purified and carried out in vitro refolding for recombinant SENP1 with a His×6 tag.
MT1X was expressed in Rosetta strain of E. coli (Novagen) as above. The inclusion bodies of MT1X were washed twice with 2% Triton X-100, 0.1 M Tris, pH = 8, and then the inclusion bodies were dissolved in 6 M guanidine HCl, 10 mM DTT, 10 mM EDTA and 0.2 M Tris, pH = 8. The protein concentration was determined using BioRad Bradford protein assay (BioRad), and then adjusted to 1 mg/ml. For preparation of Zn2+-bound or Zn2+-free MT1X, two milliliter of the solution (2 mg MT1X) was used for each refolding with or without ZnCl2.
To prepare Zn2+-bound MT1X, the MT1X solution was exchanged to 6 M guanidine HCl, 0.2 M arginine, 0.5 mM zinc chloride, 0.2 M Tris, pH = 7.2 using a centrifugal molecular cut off filter unit, Amicon Ultra-4 3 kDa MWCO (Millipore), and then the final volume was adjusted to 4 ml with the same solution. The MT1X solution was transferred into a Snakeskin 3.5 kDa MWCO Dialysis tubing (Thermo Fisher), and then the dialysis was carried out at 4 °C against three different solutions subsequently. The MT1X was dialyzed against 0.2 M arginine, 0.1 mM ZnCl2, 0.2 M Tris, pH = 7.2 for 24 h, against 0.1 mM ZnCl2, 0.1 M Tris, pH = 7.2 for 24 h, and then 50 mM Tris, pH = 7.2 for 48 h. The final dialysis buffer was changed to fresh buffer three times.
To prepare Zn2+-free MT1X, the volume of MT1X solution was adjusted to 4 mL with 6 M guanidine HCl, 10 mM DTT, 10 mM EDTA and 0.2 M Tris, pH = 8, and then was transferred into a Snakeskin 3.5 kDa MWCO Dialysis tubing. The refolding of Zn2+-free MT1X was carried out at 4 °C against four different solutions subsequently; against 0.2 M Arginine, 1 mM EDTA, 1 mM DTT, 0.2 M Tris, pH = 7.2 for 24 h, against 1 mM EDTA, 1 mM DTT, 0.1 M Tris, pH = 7.2 for 24 h, against 0.1 mM DTT, 20 mM Tris, pH = 7.2 for 24 h, and against 20 mM Tris, pH = 7.2 for 24 h. The final dialysis buffer was changed to fresh buffer three times.
After refolding, MT1X was concentrated using Amicon Ultra-4 3 kDa MWCO to 200–300 μg/mL, and the filtrates of Amicon Ultra-4 were kept for using as negative buffer control for MT1X assays, particularly to confirm the removal of free zinc in MT1X solution.
In vitro SENP1-protease assays
Two assays were developed to measure SENP1 activity. Recombinant His×6-SUMO1-mCherry was used as a substrate for both assays. The cDNA of full size human SUMO1, including the cleavage site, and mCherry were amplified with PCR using Phusion DNA polymerase and the PCR fragments were cloned into a pT7JLH plasmid at Xba-I site using Gibson assembly to make pT7JLH-SUMO1-mCherry plasmid. Transformed Rosetta strain of E. coli was grown in 50 ml LB broth containing 100 µg/ml Ampicillin at 37 °C with shaking until A600 reached between 0.6 and 0.8. His×6-SUMO1-mCherry protein was expressed by slow induction with 0.1 mM IPTG at 20 °C overnight and the E. coli pellet was harvested by centrifugation at 10,000 × g for 10 min. Pink color of E. coli pellet indicated the proper folding of His×6-SUMO1-mCherry. The pellet was incubated with 500 µg/ml lysozyme, 1 mM PMSF, 5 mM EDTA, 50 mM Tris, 0.2 M sodium chloride, pH = 8 for 3 h at 4 °C, and then sonicated five times in short burst of 30 s on ice. The cell lysate was centrifuged at 10,000 × g for 10 min at 4 °C, and the supernatant was dialyzed against 20 mM Tris, 0.3 M sodium chloride, pH = 8. The soluble His×6-SUMO1-mCherry was purified with a HisPur Nickel-NTA resin column (1 ×5 cm). The purified His×6-SUMO1-mCherry was dialyzed against 5 mM EDTA, pH = 8 for 6 h and then SENP1 assay buffer (20 mM Tris, 0.1 M sodium chloride, 0.05 % Tween20, pH = 7.2) for 24 h. The final dialysis buffer was changed to fresh buffer twice. The concentration of His×6-SUMO1-mCherry protein was estimated by the absorbance at 280 nm.
For the first assay, HisPur Nickel-NTA resin was used to capture and precipitate the undigested substrate, and then the mCherry signal in the supernatant containing the digested substrate was measured. 100 nM of recombinant SENP1 protein was mixed with 1 µM of His×6-SUMO1-mCherry, and incubated at 37 °C for 2 h to cleave mCherry from His×6 tag-SUMO1. The undigested substrate was captured and removed by the addition of 25 µl of HisPur Nickel-NTA resin suspension (1:1, v/v). After vortexing for 10 s, HisPur Nickel-NTA resin was precipitated by centrifugation at 10,000 × g for 30 s at 4 °C. Each supernatant was transferred into a 96 well clear bottom black plate and then fluorescent intensity, Ex544 nm/Em615 nm, was measured using Envision Multilabel plate reader (PerkinElmer) or Synergy HTX plate readers (Bio Tek Instruments).
The second assay was used to examine the effect of MT1X on SENP1 activity. Because MT1X interfered the binding of Hisx6-SUMO1-mCherry to HisPur Nickel-NTA resin, native-PAGE (non-denaturing Tris/Glycine gel electrophoresis) was used to separate the digested substrate from undigested substrate. Recombinant SENP1 protein without His×6 tag (200 nM) was mixed with 10 or 20 μg/mL (1.65 or 3.3 µM) MT1X, and 2.5 μM His×6-SUMO1-mCherry in a 20 μl reaction, incubated at room temperature for 45 min, and then separated by electrophoresis on non-denaturing Tris/Glycine gel. The mCherry signals of the undigested and digested substrate were detected using BioRad Gel Doc XR Imaging system (BioRad).
Because SENP1 activity changes during storage due to oxidation, the enzyme was fully activated prior to assays with 10 mM DTT at 4 °C overnight. Amicon Ultra-4 10 kDa MWCO was used to remove DTT and then replaced the buffer to SENP1 assay buffer, 20 mM Tris, 0.1 M sodium chloride, 0.05% Tween20, pH = 7.2. Final concentration of SENP1 and mutants were estimated by the absorbance at 280 nm. To preserve the activity of SENP1 25% glycerol (w/w) was added and stored under nitrogen gas at −20 °C.
Analysis and statistics
Analysis was performed using FitMaster (HEKA Electronik) and GraphPad Prism (v10.1.1). Statistical outliers were identified and removed by an unbiased ROUT (robust regression followed by outlier identification) test. All data are shown as the mean ± SEM. Single-cell and single-islet data are shown as SuperPlots89, with individual cells/islets in gray, and data averaged by animal as larger dark points. Statistical analysis was performed on data both with cells/islets as replicates, and on data averaged by animal/donor. Data were analyzed by the 2-tailed Student’s t test (for two groups), or ANOVA and post-test as indicated in figure legends. A p-value less than 0.05 was considered significant. Zn2+Bind (https://zincbind.net/) was used to predict the Zn2+-binding site in SENP1 (2IYC)90. Free Zn2+ for patch-clamp experiments was estimated with MaxChelator at https://somapp.ucdmc.ucdavis.edu/pharmacology/bers/maxchelator/.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
RNA-seq data generated in this study have been deposited in the GEO database under accession code GSE249790. Other data generated in this study and used to generate figures are provided in the Source Data file. Source data are provided with this paper.
References
Chen, C., Cohrs, C. M., Stertmann, J., Bozsak, R. & Speier, S. Human beta cell mass and function in diabetes: Recent advances in knowledge and technologies to understand disease pathogenesis. Mol. Metab. 6, 943–957 (2017).
Perley, M. J. & Kipnis, D. M. Plasma insulin responses to oral and intravenous glucose: studies in normal and diabetic subjects. J. Clin. Invest. 46, 1954–1962 (1967).
Polonsky, K. S., Given, B. D. & Van, E. Cauter, Twenty-four-hour profiles and pulsatile patterns of insulin secretion in normal and obese subjects. J. Clin. Invest. 81, 442–448 (1988).
Elbein, S. C., Wegner, K. & Kahn, S. E. Reduced beta-cell compensation to the insulin resistance associated with obesity in members of Caucasian familial type 2 diabetic kindreds. Diabetes Care. 23, 221–227 (2000).
Chen, C. et al. Alterations in β-cell calcium dynamics and efficacy outweigh islet mass adaptation in compensation of insulin resistance and prediabetes onset. Diabetes 65, 2676–2685 (2016).
Hull, R. L. et al. Dietary-fat-induced obesity in mice results in beta cell hyperplasia but not increased insulin release: evidence for specificity of impaired beta cell adaptation. Diabetologia 48, 1350–1358 (2005).
Cohrs, C. M. et al. Dysfunction of persisting β cells is a key feature of early type 2 diabetes pathogenesis. Cell Rep. 31, 107469 (2020).
Tamarit-Rodríguez, J., Vara, E. & Tamarit, J. Starvation-induced changes of palmitate metabolism and insulin secretion in isolated rat islets stimulated by glucose. Biochem J. 221, 317–324 (1984).
Khaldi, M. Z., Guiot, Y., Gilon, P., Henquin, J. C. & Jonas, J. C. Increased glucose sensitivity of both triggering and amplifying pathways of insulin secretion in rat islets cultured for 1 wk in high glucose. Am. J. Physiol. Endocrinol. Metab. 287, E207–E217 (2004).
Wortham, M. & Sander, M. Mechanisms of β-cell functional adaptation to changes in workload. Diabetes Obes. Metab. 18, 78–86 (2016).
Fex, M. et al. Enhanced mitochondrial metabolism may account for the adaptation to insulin resistance in islets from C57BL/6J mice fed a high-fat diet. Diabetologia 50, 74–83 (2007).
Irles, E. et al. Enhanced glucose-induced intracellular signaling promotes insulin hypersecretion: pancreatic beta-cell functional adaptations in a model of genetic obesity and prediabetes. Mol. Cell. Endocrinol. 404, 46–55 (2015).
Gonzalez, A. et al. Insulin hypersecretion in islets from diet-induced hyperinsulinemic obese female mice is associated with several functional adaptations in individual β-cells. Endocrinology 154, 3515–3524 (2013).
Valle, M. M. R. et al. Alterations of NADPH oxidase activity in rat pancreatic islets induced by a high-fat diet. Pancreas 40, 390–395 (2011).
Do, O. H., Gunton, J. E., Gaisano, H. Y. & Thorn, P. Changes in beta cell function occur in prediabetes and early disease in the Lepr db mouse model of diabetes. Diabetologia 59, 1222–1230 (2016).
Gandasi, N. R. et al. Glucose-dependent granule docking limits insulin secretion and is decreased in human type 2 diabetes. Cell Metab. 27, 470–478.e4 (2018).
Ferdaoussi, M. et al. Isocitrate-to-SENP1 signaling amplifies insulin secretion and rescues dysfunctional β cells. J. Clin. Invest. 125, 3847–3860 (2015).
Xu, Z. et al. Molecular basis of the redox regulation of SUMO proteases: a protective mechanism of intermolecular disulfide linkage against irreversible sulfhydryl oxidation. FASEB J. 22, 127–137 (2008).
Cameron, W. D. et al. Apollo-NADP+: a spectrally tunable family of genetically encoded sensors for NADP+. Nat. Methods 13, 352–358 (2016).
Deglasse, J.-P. et al. Glucose acutely reduces cytosolic and mitochondrial H2O2 in rat pancreatic beta cells. Antioxid. Redox Signal. 30, 297–313 (2019).
Dai, X.-Q. et al. SUMOylation regulates insulin exocytosis downstream of secretory granule docking in rodents and humans. Diabetes 60, 838–847 (2011).
Ferdaoussi, M. et al. SUMOylation and calcium control syntaxin-1A and secretagogin sequestration by tomosyn to regulate insulin exocytosis in human ß cells. Sci. Rep. 7, 248 (2017).
Ferdaoussi, M. & MacDonald, P. E. Toward connecting metabolism to the exocytotic site. Trends Cell Biol. 27, 163–171 (2017).
Lin, H. et al. β-Cell knockout of SENP1 reduces responses to incretins and worsens oral glucose tolerance in high-fat diet-fed mice. Diabetes 70, 2626–2638 (2021).
Auberval, N. et al. Metabolic and oxidative stress markers in Wistar rats after 2 months on a high-fat diet. Diabetol. Metab. Syndr. 6, 130 (2014).
Shrestha, S. et al. Aging compromises human islet beta cell function and identity by decreasing transcription factor activity and inducing ER stress. Sci. Adv. 8, eabo3932 (2022).
Gembal, M., Detimary, P., Gilon, P., Gao, Z. Y. & Henquin, J. C. Mechanisms by which glucose can control insulin release independently from its action on adenosine triphosphate-sensitive K+ channels in mouse β cells. J. Clin. Invest. 91, 871–880 (1993).
Piñeros, A. R. et al. Single-cell transcriptional profiling of mouse islets following short-term obesogenic dietary intervention. Metabolites 10, 513 (2020).
Wortham, M. et al. Nutrient regulation of the islet epigenome controls adaptive insulin secretion. J. Clin. Invest. 133, e165208 (2023).
Kang, T. et al. Characterization of signaling pathways associated with pancreatic β-cell adaptive flexibility in compensation of obesity-linked diabetes in db/db mice. Mol. Cell Proteom. 19, 971–993 (2020).
Szklarczyk, D. et al. STRING v11: protein–protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 47, D607–D613 (2019).
Cerqueira, N. M. et al. Cholesterol biosynthesis: a mechanistic overview. Biochemistry 55, 5483–5506 (2016).
Hao, M., Head, W. S., Gunawardana, S. C., Hasty, A. H. & Piston, D. W. Direct effect of cholesterol on insulin secretion: a novel mechanism for pancreatic beta-cell dysfunction. Diabetes 56, 2328–2338 (2007).
Fujikawa, Y. et al. Mouse redox histology using genetically encoded probes. Sci. Signal. 9, rs1 (2016).
Ivarsson, R. et al. Redox control of exocytosis: regulatory role of NADPH, thioredoxin, and glutaredoxin. Diabetes 54, 2132–2142 (2005).
Reinbothe, T. M. et al. Glutaredoxin-1 mediates NADPH-dependent stimulation of calcium-dependent insulin secretion. Mol. Endocrinol. 23, 893–900 (2009).
Pettersson, U. S., Waldén, T. B., Carlsson, P.-O., Jansson, L. & Phillipson, M. Female mice are protected against high-fat diet induced metabolic syndrome and increase the regulatory T cell population in adipose tissue. PLoS One. 7, e46057 (2012).
Bechtel, T. J. & Weerapana, E. From structure to redox: the diverse functional roles of disulfides and implications in disease. Proteomics 17, 1000–1002 (2017).
Sun, M. et al. Prediction of reversible disulfide based on features from local structural signatures. BMC Genomics. 18, 279 (2017).
Shi, T. et al. Exploring the desumoylation process of SENP1: a study combined MD simulations with QM/MM calculations on SENP1-SUMO1-RanGAP1. J. Chem. Inf. Model. 53, 2360–2368 (2013).
López-Grueso, M. J. et al. Thioredoxin and glutaredoxin regulate metabolism through different multiplex thiol switches. Redox Biol. 21, 101049 (2019).
Giles, N. M. et al. Metal and Redox Modulation of Cysteine Protein Function. Chem. Biol. 10, 677–693 (2003).
Maret, W. Inhibitory zinc sites in enzymes. Biometals 26, 197–204 (2013).
Perry, D. K. et al. Zinc is a potent inhibitor of the apoptotic protease, caspase-3. A novel target for zinc in the inhibition of apoptosis. J. Biol. Chem. 272, 18530–18533 (1997).
Velázquez-Delgado, E. M. & Hardy, J. A. Zinc-mediated allosteric inhibition of caspase-6. J. Biol. Chem. 287, 36000–36011 (2012).
Thompson, M. W. Regulation of zinc-dependent enzymes by metal carrier proteins. Biometals 35, 187–213 (2022).
Krężel, A. & Maret, W. The Bioinorganic Chemistry of Mammalian Metallothioneins. Chem. Rev. 121, 14594–14648 (2021).
Onosaka, S. et al. Concentration of Metallothionein in Human Pancreas. J. Hygienic Chem. 34, 92–96 (1988).
Bensellam, M. et al. Metallothionein 1 negatively regulates glucose-stimulated insulin secretion and is differentially expressed in conditions of beta cell compensation and failure in mice and humans. Diabetologia 62, 2273–2286 (2019).
Büsselberg, D., Platt, B., Michael, D., Carpenter, D. O. & Haas, H. L. Mammalian voltage-activated calcium channel currents are blocked by Pb2+, Zn2+, and Al3+. J. Neurophysiol. 71, 1491–1497 (1994).
Weyer, C., Bogardus, C., Mott, D. M. & Pratley, R. E. The natural history of insulin secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus. J. Clin. Invest. 104, 787–794 (1999).
Merrins, M. J., Corkey, B. E., Kibbey, R. G. & Prentki, M. Metabolic cycles and signals for insulin secretion. Cell Metab. 34, 947–968 (2022).
Campbell, J. E. & Newgard, C. B. Mechanisms controlling pancreatic islet cell function in insulin secretion. Nat. Rev. Mol. Cell Biol. 22, 142–158 (2021).
Plecitá-Hlavatá, L. et al. Glucose-stimulated insulin secretion fundamentally requires H2O2 signaling by NADPH Oxidase 4. Diabetes 69, 1341–1354 (2020).
Taddeo, E. P. et al. Mitochondrial Proton Leak Regulated by Cyclophilin D Elevates Insulin Secretion in Islets at Nonstimulatory Glucose Levels. Diabetes 69, 131–145 (2020).
Chang, H. H. et al. Targeting Apollo-NADP+ to image NADPH generation in pancreatic beta-cell organelles. ACS Sens. 7, 3308–3317 (2022).
Prentki, M., Matschinsky, F. M. & Madiraju, S. R. M. Metabolic signaling in fuel-induced insulin secretion. Cell Metab. 18, 162–185 (2013).
Plecitá-Hlavatá, L. et al. Mitochondrial superoxide production decreases on glucose-stimulated insulin secretion in pancreatic β Cells due to decreasing mitochondrial matrix NADH/NAD+ ratio. Antioxid. Redox Signal. 33, 789–815 (2020).
Gutscher, M. et al. Proximity-based protein thiol oxidation by H2O2-scavenging peroxidases. J. Biol. Chem. 284, 31532–31540 (2009).
Fu, A. et al. Glucose metabolism and pyruvate carboxylase enhance glutathione synthesis and restrict oxidative stress in pancreatic islets. Cell Rep. 37, 110037 (2021).
Zhang, G.-F. et al. Reductive TCA cycle metabolism fuels glutamine- and glucose-stimulated insulin secretion. Cell Metab. 33, 804–817.e5 (2021).
Ebert, J. C. & Altman, R. B. Robust recognition of zinc binding sites in proteins. Protein Sci. 17, 54–65 (2008).
Maret, W. & Li, Y. Coordination dynamics of zinc in proteins. Chem. Rev. 109, 4682–4707 (2009).
Kluska, K., Adamczyk, J. & Krężel, A. Metal binding properties, stability and reactivity of zinc fingers. Coord. Chem. Rev. 367, 18–64 (2018).
Hübner, C. & Haase, H. Interactions of zinc- and redox-signaling pathways. Redox Biol. 41, 101916 (2021).
Liu, W. et al. Lactate regulates cell cycle by remodelling the anaphase promoting complex. Nature 616, 790–797 (2023).
Cuozzo, F. et al. High-resolution glucose fate-mapping reveals LDHB-dependent lactate production by human pancreatic β cells. bioRxiv, https://doi.org/10.1101/2022.12.21.521364 (2022).
Vinkenborg, J. L. et al. Imaging of intracellular free Zn2+ in real time using genetically-encoded FRET sensors. Nat. Methods 6, 737–740 (2009).
Zimny, S., Gogolin, F., Abel, J. & Gleichmann, H. Metallothionein in isolated pancreatic islets of mice: induction by zinc and streptozotocin, a naturally occurring diabetogen. Arch. Toxicol. 67, 61–65 (1993).
Gupta, D. et al. Temporal characterization of β cell-adaptive and -maladaptive mechanisms during chronic high-fat feeding in C57BL/6NTac mice. J. Biol. Chem. 292, 12449–12459 (2017).
El, K. et al. GIP mediates the incretin effect and glucose tolerance by dual actions on α cells and β cells. Sci. Adv. 7, eabf1948 (2021).
Ferdaoussi, M. et al. Improved glucose tolerance with DPPIV inhibition requires β‐cell SENP1 amplification of glucose‐stimulated insulin secretion. Physiol. Rep. 8, e14420 (2020).
Smith, E. P. et al. The role of β cell glucagon-like peptide-1 signaling in glucose regulation and response to diabetes drugs. Cell Metab. 19, 1050–1057 (2014).
Lyon, J. et al. Research-Focused Isolation of Human Islets From Donors With and Without Diabetes at the Alberta Diabetes Institute IsletCore. Endocrinology 157, 560–569 (2016).
Westacott, M. J. et al. Age-Dependent Decline in the Coordinated [Ca2+] and Insulin Secretory Dynamics in Human Pancreatic Islets. Diabetes 66, 2436–2445 (2017).
Thorens, B. et al. Ins1(Cre) knock-in mice for beta cell-specific gene recombination. Diabetologia 58, 558–565 (2015).
Ferdaoussi, M. et al. Improved glucose tolerance with DPPIV inhibition requires β-cell SENP1 amplification of glucose-stimulated insulin secretion. Physiological Rep. 8, e14420 (2020).
Zhang, Q. et al. Na+ current properties in islet α- and β-cells reflect cell-specific Scn3a and Scn9a expression. J. Physiol. 592, 4677–4696 (2014).
Ghazvini Zadeh, E. H. et al. ZIGIR, a granule-specific Zn2+ indicator, reveals human islet α cell heterogeneity. Cell Rep. 32, 107904 (2020).
Zhou, Y. et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat. Commun. 10, 1523 (2019).
Gu, Z., Gu, L., Eils, R., Schlesner, M. & Brors, B. circlize implements and enhances circular visualization in R. Bioinformatics 30, 2811–2812 (2014).
Subramanian, A. et al. Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. PNAS 102, 15545–15550 (2005).
Vilas-Boas, E. A. et al. Transient NADPH oxidase 2-dependent H2O2 production drives early palmitate-induced lipotoxicity in pancreatic islets. Free Radic. Biol. Med. 162, 1–13 (2021).
Nalbach, L. et al. Improvement of islet transplantation by the fusion of islet cells with functional blood vessels. EMBO Mol. Med. 13, e12616 (2021).
Smith N. Seahorse protocol for islets using Xfe24 Analyzer. protocols.io, https://doi.org/10.17504/protocols.io.n2bvj63p5lk5/v1). (2024)
Haneskog, L. Preparation of Denatured Protein Samples from Inclusion Bodies. Cold Spring Harb. Protoc. 2006, 4265 (2006).
Umetsu, M. et al. How additives influence the refolding of immunoglobulin-folded proteins in a stepwise dialysis system. Spectroscopic evidence for highly efficient refolding of a single-chain Fv fragment. J. Biol. Chem. 278, 8979–8987 (2003).
Chen, W. et al. Evidence that a peptide spanning the B-C junction of proinsulin is an early Autoantigen epitope in the pathogenesis of type 1 diabetes. J. Immunol. 167, 4926–4935 (2001).
Lord, S. J., Velle, K. B., Mullins, R. D. & Fritz-Laylin, L. K. SuperPlots: Communicating reproducibility and variability in cell biology. J. Cell Biol. 219, e202001064 (2020).
Ireland, S. M. & Martin, A. C. R. ZincBind-the database of zinc binding sites. Database 2019, baz006 (2019).
Acknowledgements
The University of Alberta is situated on Treaty 6 territory, traditional lands of First Nations and Métis people. We thank the Human Organ Procurement and Exchange (HOPE) program and Trillium Gift of Life Network (TGLN) for their work in procuring human donor pancreas for research, and James Lyon (Alberta) for his efforts in human islet isolation. We especially thank the organ donors and their families for their kind gift in support of diabetes research. Work was funded by grants from the National Institutes of Health NIH R01 GM132610 (W.L.), Canadian Institutes of Health Research Foundation Grant 148451 (P.E.M.), and German Research Foundation TRR219 project M04, Project ID 322900939 (L.P.R.). H.L. was supported by a Sino-Canadian Studentship from Shantou University, A.R.P. holds the Canada Research Chair in Cell Therapies for Diabetes, and P.E.M. holds the Canada Research Chair in Islet Biology.
Author information
Authors and Affiliations
Contributions
Conceptualization: H.L., K.S., A.R.P., E.A., W.L., P.E.M. Methodology: H.L., K.S., L.N., L.P.R., A.R.P., E.A., W.L., P.E.M. Investigation: H.L., K.S., N.S., X.L., L.N., S.F., A.F.S., X.D., A.B., M.F., S.A., L.P.R. Visualization: H.L., P.E.M. Supervision: A.R.P., L.P.R., E.A., W.L., P.E.M. Writing—original draft: H.L., K.S., P.E.M. Writing—review & editing: H.L., K.S., N.S., X.L., L.N., S.F., A.F.S., X.D., A.B., M.F., S.A., A.R.P., L.P.R., E.A., W.L., P.E.M.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Source data
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Lin, H., Suzuki, K., Smith, N. et al. A role and mechanism for redox sensing by SENP1 in β-cell responses to high fat feeding. Nat Commun 15, 334 (2024). https://doi.org/10.1038/s41467-023-44589-x
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-023-44589-x
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.