Recent investigations have confirmed beneficial effects of sustained activation of pancreatic beta-cell neuropeptide Y1 receptors (NPY1R’s) on the growth, survival and overall secretory function of insulin-producing cells (1–5). In this regard, enzymatically stable PYY(1–36) peptides from phylogenetically ancient fish have been demonstrated to function as long-acting, potent and specific NPY1R agonists (4). Although the NPY1R is not believed to be expressed on pancreatic alpha-cells, PYY is known to be locally secreted from these alpha-cells (6), and positive effects of the piscine-derived PYY(1-36) peptides on alpha-cell morphology and glucagon secretion have also been noted (4). Since diabetes is a metabolic disorder characterized by both aberrant alpha- and beta-cell secretory function (7, 8), it suggests NPY1R modulation may have extremely credible antidiabetic potential (9). Indeed, the most efficacious piscine-derived PYY(1-36) sequence, namely sea lamprey PYY(1-36) (SL-PYY(1-36)) (4), has already been shown to combat beta-cell loss in diabetic rodents via augmentation of beta-cell proliferation and a reduction in apoptosis (4).

These positive islet effects are unlikely to account for the full pancreatic architectural benefits of SL-PYY(1-36). As such, although changes in beta-cell growth and survival were significantly improved following sustained SL-PYY(1-36) administration, relative changes were modest in quantitative terms, with less obvious impact on alpha-cell turnover (4). In relation to this, recent elegant cell lineage tracing studies have highlighted the transitioning ability of mature islet alpha- and beta-cells, leading to alterations in islet architecture, especially in diabetes (10–13). Thus, despite their conflicting roles in the control of blood glucose, alpha- and beta-cells share similar transcriptomes (14, 15), and it seems plausible that SL-PYY(1-36) might influence transdifferentiation of both cell types, to help elicit the observed NPY1R-mediated improvements in pancreatic islet morphology (4, 5, 16). Importantly in terms of clinical relevance, islet cell transdifferentiation is not restricted to rodents and has been evidenced in human beta-cells (17–20) and islet cells from patients with diabetes (21). Therefore, the present study was conducted using transgenic mice with alpha- and beta-cell lineage tracing capabilities, to investigate the contribution of transdifferentiation of alpha- and beta-cells to SL-PYY(1-36)-induced improvements of pancreatic islet architecture in diabetes.

Fully characterized Glu^CreERT2 ;ROSA26-eYFP and Ins1 ^Cre/+;Rosa26-eYFP transgenic mouse models (22, 23) were used to directly investigate the impact of sustained NPY1R signaling on both alpha- to beta-, as well as beta- to alpha-cell, transdifferentiation, respectively. It is generally considered that islet cell transdifferentiation only occurs following severe pancreatic insult, as experienced in diabetes (11). Thus, in the current setting chemically induced beta-cell ablation was used to provoke and study transdifferentiation of individual alpha- and beta-cells in Glu^CreERT2 ;ROSA26-eYFP and Ins1 ^Cre/+;Rosa26-eYFP mice (10). Importantly, this method of diabetes induction aligns well with previous investigations on the antidiabetic and pancreatic architectural benefits of SL-PYY(1-36) treatment (4). Therefore, the impact of sub-chronic pharmacological upregulation of NPY1R pathways in the two diabetic transgenic rodent models was investigated, through 11-day twice daily administration of SL-PYY(1-36). Taken together, our datasets suggest that pancreatic islet architectural benefits of prolonged NPY1R activation in diabetes are partly linked to positive alteration of the transdifferentiation of both alpha- and beta-cells.

As expected, multiple low dose STZ administration resulted in characteristic pancreatic islet beta-cell destruction, acute insulin deficiency and elevations of blood glucose in both Glu^CreERT2 ;Rosa26-eYFP and Ins1^Cre/+ ;Rosa26-eYFP transgenic mice. Interestingly, appreciable differences were apparent between these transgenic mice in terms of STZ susceptibility and effects on body weight, food and fluid intake as well as alpha-cell derived circulating glucagon and degree of hyperglycaemia, that indicated a more severe diabetic phenotype in Ins1^Cre/+ ;Rosa26-eYFP mice.

Although unlikely given the mechanism of action of STZ (29), presence of the transgene within beta-cells of Ins1^Cre/+ ;Rosa26-eYFP mice could contribute to the difference in susceptibility to the toxin. However, studies in Glu^CreERT2 ;ROSA26-eYFP mice confirm that activation of the Cre recombinase enzyme does not affect normal islet cell processes (22), and although unlikely, production of fluorescent GFP protein within beta cells of Ins1^Cre/+ ;Rosa26-eYFP mice could increase metabolic demand leaving cells more susceptible to STZ. Glu^CreERT2 ;Rosa26-eYFP mice, unlike Ins1^Cre/+ ;Rosa26-eYFP, also require a single, low-dose, tamoxifen injection to induce Cre-lox recombination, raising the possibility that estrogen receptor antagonism might curb some detrimental effects of STZ, but this appears only to occur during sub-chronic dosing regimen (30). Indeed, the use of tamoxifen-dependent Cre recombinase induction is a commonly employed tool to successfully study gene function with minimal adverse effects. However, we recognize that an additional group of control Glu^CreERT2 ;Rosa26-eYFP mice would have been useful to confirm lack of effect of the single dose of tamoxifen on diabetic phenotype. A more likely explanation relates to variations in STZ susceptibility between (25), and even within (26), strains of mice, comparable to the established gender variations of STZ effects in rodents (31). However, given both transgenic mouse models displayed clear STZ-induced beta-cell ablation, with distinct diabetes-induced alterations in islet cell transitioning (23, 32), they represent valid tools to assess the impact of sustained NPY1R agonism on alpha- to beta- and beta- to alpha-cell transdifferentiation. Moreover, the difference in presenting diabetic phenotype of Glu^CreERT2 ;Rosa26-eYFP and Ins1^Cre/+ ;Rosa26-eYFP mice allows for differentiation between direct positive effects of SL-PYY(1-36) on islet cell transdifferentiation, as opposed to indirect benefits related to improved glycaemic control and metabolic status. However, because Glu^CreERT2 and Ins1^Cre/+ are not equivalent lineage tracers, labeling islet cells either during tamoxifen exposure or from the moment of Ins1 expression, respectively, direct quantitative comparisons of islet cell changes between the models should be avoided.

In keeping with earlier observations (4), SL-PYY(1-36) increased beta-cell area, plasma and pancreatic insulin in both Glu^CreERT2 ;Rosa26-eYFP and Ins1^Cre/+ ;Rosa26-eYFP mice, as well as reversing the characteristic infiltration of glucagon positively stained central alpha-like cells (4, 5). Beta-cell benefits are undoubtedly linked, in part, to increased proliferation and reduced beta-cell apoptosis, as documented previously with sustained NPY1R activation in diabetic mouse models (1, 4, 5). Interestingly, previous studies in insulin-deficient mice also demonstrated reductions in alpha-cell area following SL-PYY(1-36) treatment, which coupled with increased beta-cell mass, resulted in no obvious change in overall pancreatic islet area (4, 5). In the current setting, islet area increased with 11-day SL-PYY(1-36) treatment in both diabetic transgenic models, as a result of elevations in alpha- and beta-cell areas. We assume that the slightly different scenario compared to our previous study (4) is related to differences in the strain of mice employed and associated physiological responses to STZ (26), as well as duration of SL-PYY(1-36) treatment. Nonetheless, observed metabolic and islet benefits reaffirm the recognized beta-cell pro-survival effects of NPYR1 signaling (16). This established benefit, coupled with evidence of beta-cell loss following selective KO of PYY expressing cells (1), as well as expansion of PYY expressing alpha-cell like islet cell populations during regeneration of the adult pancreas (33), indicate the importance of NPY receptor signaling for positive islet cell adaptations in response to metabolic stress. Furthermore, upregulation of NPY receptor activation has also been linked to the improved insulin secretion and resolution of diabetes following bariatric surgery (34–36).

In terms of islet cell lineage tracing, studies in Ins1^Cre/+ ;Rosa26-eYFP mice revealed that STZ induced notable increases in the number of pancreatic beta-cells losing their identity, as well as mature beta-cells transitioning to glucagon positive alpha-like cells (23). Interestingly, in Glu^CreERT2 ;Rosa26-eYFP mice a seemingly high intrinsic transition of glucagon positive alpha-cells to insulin positive beta-cells was decreased by STZ intervention. Triple staining of GFP, glucagon and insulin would have been useful to provide additional insight on these islet cell lineage processes, especially in Glu^CreERT2 ;Rosa26-eYFP mice, but was not possible due to methodological constraints. Furthermore, STZ susceptibility of cells transitioning from/to beta-cells and the relevance of established cytoprotective effects of SL-PYY(1-36) in such cell populations are unknown. However, when taken together, the data suggest that a likely source of the additional positively stained glucagon cells in both diabetic models in the present study is linked to transdifferentiation of beta-cells toward alpha-cells and possible lack of alpha-cell dedifferentiation, although the latter point requires further investigation. Most importantly, SL-PYY(1-36) treatment appeared to be able to fully, or partially, counter each of these STZ-induced detrimental islet cell transdifferentiation events. Intriguingly, despite increased alpha-cell area, pancreatic glucagon content was decreased by SL-PYY(1-36) treatment in both mouse models. This might suggest that a significant number of alpha-cells were currently in the process of transition, probably toward a more beta-cell like phenotype, and therefore have reduced glucagon content. In full agreement, SL-PYY(1-36) reduced the percentage of glucagon^+ve/GFP^+ve cells in Glu^CreERT2 ;Rosa26-eYFP mice, and augmented alpha- to beta-cell transdifferentiation. The location on different mouse islet cells of NPY1R, which we detected previously to be expressed at relatively low levels by alpha-, beta- and delta-cells and the role of endogenous PYY, expressed in adult mouse alpha- and delta-cells (37), will need to be determined in the future. In terms of alpha-cell secretion, non-fasted circulating glucagon concentrations were unaltered by SL-PYY(1-36), however it is envisaged that any effect on glucagon would be more apparent under fasting conditions, which unfortunately were not assessed in the current study.

Unlike Ins1^Cre/+ ;Rosa26-eYFP mice, the Glu^CreERT2 ;Rosa26-eYFP transgenic model exhibited an almost complete restoration of normoglycaemia, suggesting that exogenous peptide administration can directly influence islet cell transdifferentiation. As such, improvements in glycaemic status have been shown to independently alter pancreatic islet cell lineage (38). Related NPYR1 dependent mechanisms still need to be investigated, and although beyond the scope of the current study, consideration of levels of specific alpha- and beta-cell transcription factors such as aristaless-related homeobox (Arx) (39), paired box gene 4 (Pax4) (40), pancreatic and duodenal homeobox 1 (Pdx-1) or forkhead box O1 (FOXO1) (41) would be interesting. Indeed, in the absence of such additional investigations, some constraint is required in terms of interpretation of the full impact of SL-PYY(1-36) on transdifferentiation of alpha- to beta-cells. Findings should also be interpreted in relation to the decrease in islet PYY expression observed following STZ-induced insulin deficiency in rodents (1, 16), although islet PYY levels were not directly assessed in the current setting. In addition, observations of increased numbers of individual islets following SL-PYY(1-36) treatment might suggest islet neogenesis is involved in beneficial NPYR1 mediated islet adaptations, but this would also require further, more detailed study. Indeed, an established role for PYY during rapid islet cell development and replication within the embryo makes this scenario plausible and worthy of investigation (42). In this regard, it may be interesting to assess the impact of SL-PYY(1-36) on islet cell lineage under normal physiological conditions.

The present study demonstrates that sustained NPYR1 activation by SL-PYY(1-36) rescued beta-cell loss and augmented beta-cell function in chemically induced insulin-deficient diabetes, leading to significantly enhanced overall metabolic control. As well as increased proliferation and reduced apoptosis of beta-cells (3–5, 16), SL-PYY(1-36) also appeared to positively influence the transitioning of both islet alpha- and beta-cells. This newly discovered, beneficial islet cell lineage effect emphasizes the potential of stable long-acting NPYR1 agonists to promote prevailing, disease-modifying benefits in diabetes, linked to possible alteration of alpha cell function and transdifferentiation of alpha- to beta-cells.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by the Ulster University Animal Welfare and Ethical Review Body (AWERB).

RL and NT performed the experiments and prepared the first draft of the manuscript. RM, PF, and NI designed the experiments and reviewed and revised all subsequent versions of the manuscript. FR and FG reviewed and revised the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by a PhD studentship (awarded to RL) from the Department for the Economy (DfE) Northern Ireland and University of Ulster strategic research funding. Work in the Reimann/Gribble laboratory is currently funded by the Welcome Trust (106262/Z/14/Z and 106263/Z/14/Z) and the MRC (MRC_MC_UU_12012/3).

PF and NI are named on patents filed by the University of Ulster for exploitation of peptide therapeutics.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.