The increasing use of GLP-1 mimetics to treat diabetes and obesity highlights the therapeutic importance of this class of drugs (1). Indeed, weight loss recently reported using the GLP-1 mimetic semaglutide in humans with obesity is highly impressive (2). Nevertheless, the pronounced metabolic benefits of bariatric surgery, often resulting in remission of diabetes, unquestionably exceed positive effects of GLP-1 mimetics, highlighting the important interplay between gut-derived hormones to induce distinct and sustained benefits in obesity-diabetes (3). In full agreement, administration of GLP-1 mimetics in combination with related gut-derived hormones imparts metabolic advantages beyond that seen with GLP-1 alone (4, 5). Recent advances in peptide drug design have also witnessed generation of several unimolecular dual- or triple-acting hybrid peptides, that contain a GLP-1 backbone structure (6). For example, Tirzepatide, a synthetic dual-acting GIP and GLP-1 receptor agonist, demonstrates remarkable antidiabetic effectiveness in phase II clinical trials (7). With such successes in mind, a strong rational exists for generating unimolecular agents that mimic meal-induced gut hormone release by co-activating receptors for GLP-1 plus other hormones involved in conveying the metabolic and satiety signals initiated by feeding.

Cholecystokinin (CCK) is a hormone primarily secreted from gastrointestinal I-cells as well as neurons of the enteric and central nervous systems (8). The peptide circulates in various molecular forms, with the octapeptide CCK-8 being the smallest form to exhibit full biological activity (9). To impart its physiological actions, CCK binds and activates specific receptors, namely CCK1 and CCK2, that modulate various physiological processes including satiety, pancreatic endocrine, and exocrine function as well as gastric acid and bile secretion (8). In particular, CCK-based regulation of energy intake and pancreatic endocrine islet function make the hormone a potentially attractive therapeutic agent for obesity and diabetes (9). As such, CCK dose-dependently inhibits food intake consistently across species, from rodents (10) to non-human primates (11) and humans (12) mediated via CCK1 receptors (13). CCK also stimulates insulin secretion in rodents and humans (14, 15), with sustained administration of a long-acting CCK1 selective receptor agonist exerting notable metabolic benefits in rodent models of type 2 diabetes (16, 17). In agreement, transgenic mice with genetically-induced CCK ablation (18) or overexpression (19) exhibit impaired or improved beta-cell function, respectively. Moreover, CCK has recently been shown to exert important beta-cell proliferative and anti-apoptotic effects (20), which might counter beta-cell loss in diabetes.

When taken together, the biological action profile of CCK has strong parallels with those of GLP-1 (9). Both hormones initiate separate, but complementary, cell signalling pathways via phospholipase C or adenylate cyclase, respectively (21). At the level of the pancreatic beta-cell an important GLP-1/CCK intra-islet loop has also been described that helps protect against beta-cell apoptosis, with obvious benefits for diabetes (21). Thus, it is unsurprising that combined activation of CCK1 and GLP-1 receptors has been noted to induce numerous benefits on body weight reduction and blood glucose control (22–24). Preliminary studies also provide proof-of-principle evidence for this therapeutic approach, both in terms of combined administration of individual long-acting GLP-1 and CCK1 agonists (5, 25), as well as dual-acting hybrid peptides (26, 27). This is highly relevant for clinical translation since methods to augment the antidiabetic effectiveness of approved GLP-1 therapeutics are much sought after (28).

Two separate dual-acting GLP-1/CCK fusion peptides have been described to date, with both exerting remarkable benefits on body weight reduction and pancreatic islet morphology, resulting in improved metabolic control in obese high fat fed mice (26, 27). Although uncharacterised in terms of in vivo islet effects, the fusion peptide described by Hornigold and colleagues (27) appeared to have the most attractive biological action profile, with confirmed balance between GLP-1 and CCK1 receptor activation. Based on this knowledge, and to progress attractiveness of the GLP-1/CCK hybrid peptide approach, we employed acylation of this molecule to create a compound capable of sustained, simultaneous activation of GLP-1 and CCK1 receptors, namely [Lys¹²Pal]Ex-4/CCK, with a pharmacodynamic profile more suitable for clinical development. Thus, a C-16 fatty acid was conjugated to the free amino group of Lys¹² in the GLP-1/CCK hybrid peptide via a glutamyl linker ( Table 1 ).

Our primary objective was to examine the therapeutic efficacy and islet cell morphology after 28 days twice daily treatment with [Lys¹²Pal]Ex-4/CCK in high fat fed (HFF) mice with limited beta-cell compensation induced by low-dose STZ administration. Prior to this, bioactivity of [Lys¹²Pal]Ex-4/CCK was established in BRIN BD11 beta-cells, alongside assessment of acute and more prolonged in vivo effects on satiety and glucose homeostasis in mice. In addition, the positive interplay between GLP-1 and CCK receptor signalling on beta-cell growth and survival, with related impact on gene expression, was also studied. Ultimately, our data confirm that the profound beneficial pharmacological effects of combined activation GLP-1 and CCK receptor pathways in obesity-diabetes can be effectively harnessed within an appropriately designed single long-acting hybrid peptide.

Additive, and frequently synergistic, benefits on satiety and pancreatic endocrine function by combined GLP-1 and CCK1 receptor activation have been documented previously (23, 36). In the current study, we have exploited this in terms of realising a new avenue for the treatment of T2DM. First, we examined the potential benefits of upregulated GLP-1 and CCK1 receptor activation on pancreatic beta-cells, given the recent observations that CCK may be less influential in terms of augmenting GLP-1 induced beta-cell insulin secretion (37). This is an important aspect to understand because GLP-1 and CCK are not only synthesised in the intestine but also released locally by islet cells to exert autocrine or paracrine actions (19, 20).

Independent benefits on beta-cell growth and survival following upregulation of GLP-1 and CCK receptor activation were verified (18, 34). Although additive effects were not apparent in terms of protection against apoptosis, there were clear benefits of combined receptor upregulation on beta-cell proliferation. Interestingly, interactions between islet-derived GLP-1 and CCK have been suggested to be important for beta-cell survival, rather than growth (21). Our contrasting observations may be linked to the experimental system employed, as well as differences in duration and extent of receptor activation. However, proliferative benefits of GLP-1 and CCK in the current setting are fully supported by changes in gene expression. As such, beta-cell regeneration is believed to be controlled by glucose metabolism and metabolic demand (38), with combination GLP-1 and CCK-8 treatment resulting in distinct increases of glucose sensor glucokinase gene expression (39). Combined treatment also resulted in a reduction in pro-apoptotic NFκB and BAX gene expression in line with known downstream actions of CCK (40). When viewed in conjunction with previous observations (5, 25), these data provide a compelling case to further explore the therapeutic potential of GLP-1/CCK dual-acting hybrid peptides (26, 27).

To this end, [Lys¹²Pal]Ex-4/CCK was engineered as a long-acting GLP-1 and CCK1 receptor agonist ( Table 1 ), based on earlier positive observations of pronounced bioactivity of a GLP-1/CCK-8 fusion peptide and the pharmacodynamic benefits of peptide acylation (27, 41). As such, [Lys¹²Pal]Ex-4/CCK combines the key complementary glucose-lowering and satiety actions of exendin-4 and CCK-8, possessing an exendin-4 like N-terminus fused to a C-terminal CCK1 receptor agonist through use of a polyether linking compound ( Table 1 ). Consistent with previous studies, activation of GLP-1 and CCK1 receptors by [Lys¹²Pal]Ex-4/CCK stimulated insulin secretion from BRIN BD11 cells (5, 26). Given that [Lys¹²Pal]Ex-4/CCK is acylated at Lys¹² to promote albumin binding and prolong half-life, an intervention which often perturbs bioactivity (41), equipotency with exendin-4 is encouraging. To further probe insulinotropic effects, [Lys¹²Pal]Ex-4/CCK was co-incubated with known modulators of beta-cell function and shown to potentiate the insulin releasing actions of glucose, alanine, elevated Ca²⁺ or the phorbol ester PMA. Further to these in vitro effects, acute administration of [Lys¹²Pal]Ex-4/CCK to mice inhibited feeding activity, improved glucose tolerance, and potentiated the plasma insulin response to glucose. Such actions were protracted, with some effects even persisting when the peptide was administered up to 24 hours previously, contrasting with actions of native exendin-4 and CCK-8 which are relatively short-lived under similar experimental conditions (16, 35). Experiments employing pharmacological GLP-1 and CCK1 receptor antagonists perhaps suggested a very slight bias towards GLP-1 receptor signalling in mediating the biological actions of [Lys¹²Pal]Ex-4/CCK, in agreement with previous reports (27), but genuine prudence is required in terms of this data interpretation. Thus, the recognised synergism between GLP-1 and CCK receptor pathways (21–23) is not quantifiable when employing knockout of individual receptor pathways. Although informative, these antagonist studies are unable to provide a complete picture of the undoubted importance of synergy between GLP-1 and CCK1 receptor pathways for [Lys¹²Pal]Ex-4/CCK mediated benefits (26, 27).

Encouraged by these observations, we embarked on a longer-term in vivo study in mice that were administered low dose injections of STZ daily over 14 days to compromise beta-cell function in association with ad libitum access to high fat diet to induce insulin resistance and beta-cell stress (33). The resulting phenotype was characterised by failure of classical islet hypertrophy and beta-cell compensation in HFF-STZ mice, culminating in severe hyperglycaemia with blood glucose >12 mmol/l. Twice daily injection of these mice with [Lys¹²Pal]Ex-4/CCK for 28 days led to significant reductions in energy intake, body weight and blood glucose levels, in keeping with upregulated GLP-1 and CCK1 receptor activity (9). Importantly, these beneficial effects were apparent early in the treatment regimen and persisted throughout, with no indication of desensitisation (42). This highlights the long-acting nature and efficacy of [Lys¹²Pal]Ex-4/CCK. Development of a specific assay to measure [Lys¹²Pal]Ex-4/CCK in plasma, as well as tissue distribution, would be useful to provide more precise details of in vivo bioavailability.

Improvements in glucose tolerance in HFF mice with STZ-induced compromised beta-cells, although likely to be partially dependent on weight loss (43), were associated with significantly increased glucose-induced insulin concentrations. Indeed, circulating and pancreatic insulin were substantially increased in [Lys¹²Pal]Ex-4/CCK mice at the end of the 28-day study. It would also have been useful to examine metabolic effects following an oral nutrient challenge, but unfortunately that was not possible in the current study. Interestingly, the characteristic insulin resistance evoked by sustained high fat feeding (44) did not appear to be appreciably improved by [Lys¹²Pal]Ex-4/CCK intervention. Evidence for a small but significant enhancement of AAC glucose was noted, but this modest effect could be related to the fact that blood glucose levels in [Lys¹²Pal]Ex-4/CCK treated mice were well below 5 mmol/l during the experiment. This presumably resulted in activation of adaptive responses to counter the hypoglycaemia induced by bolus injection of exogenous insulin. In contrast to the small magnitude of this effect, [Lys¹²Pal]Ex-4/CCK evoked particularly prominent improvements in islet architecture in the HFF-STZ mice (19). Such morphological benefits which were closely linked to elevations in beta-cell proliferation (20, 34), in direct agreement with our in vitro observations. We also observed reductions in beta-cell apoptosis with sustained [Lys¹²Pal]Ex-4/CCK therapy that would also contribute to improved islet architecture (21). Furthermore, observed reductions in alpha-cell area and overall alpha:beta cell ratios were paralleled by decreases in pancreatic and plasma glucagon in the peptide treated mice. This observation fits well with reported glucagonostatic effects of GLP-1 receptor activation (45). However, it does contrast somewhat with the reported direct glucagonotropic actions of CCK (8), and detailed study of the overall impact of activation of GLP-1 and CCK1 receptors by [Lys¹²Pal]Ex-4/CCK in this regard is required.

Despite the obvious therapeutic attractiveness of [Lys¹²Pal]Ex-4/CCK, there are a number of significant matters that need to be considered with regards possible translation towards clinical development. Although initial fears of increased pancreatitis risk with GLP-1 mimetic therapy in humans has now been fully allayed (46), extremely high doses of CCK are employed to create experimental rodent models of pancreatic inflammation (47). However, it should be noted that the CCK1 receptor is expressed at much higher levels in pancreatic acinar cells of rodents than humans (48). Moreover, 28 days treatment with [Lys¹²Pal]Ex-4/CCK did not lead to any obvious detrimental pancreatic histopathological findings in our study. As noted above, alterations in endocrine islet morphology induced by [Lys¹²Pal]Ex-4/CCK were consistently linked to correction of the diabetes phenotype. In addition, previous studies employing sustained co-activation of GLP-1 and CCK1 receptors in rodent models reported no effect, or even reductions, in amylase and lipase secretion (25, 26). The possible impact of combined GLP-1 and CCK1 receptor activation on gastric emptying would also need to be considered (49, 50). However, we observed no obvious behavioural changes in [Lys¹²Pal]Ex-4/CCK treated mice. Finally, there is a suggestion that GLP-1 mimetics suppress CCK secretion in humans, leading to adverse gallbladder events (51). Thus, dual activation of both receptor pathways by [Lys¹²Pal]Ex-4/CCK could help to alleviate this potential problem.

In conclusion, the present study has reaffirmed clear benefits of combined GLP-1 and CCK1 receptor signalling. [Lys¹²Pal]Ex-4/CCK, created as a long-acting, dual GLP-1 and CCK1 receptor activating hybrid peptide, displayed prominent antidiabetic actions in HFF-STZ mice that were directly linked to improved islet morphology and beta-cell function. Such benefits were also coupled to induction of satiety and sustained reductions in body weight. These exciting preclinical observations necessitate the further consideration of combined GLP-1 and CCK1 receptor activation as a potential treatment option for the increasing numbers of people living with obesity and related diabetes.

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).

NI and PF contributed to the overall concept and experimental design, and reviewed the manuscript. AE, NT, and RL researched data, contributed to data interpretation, and edited the manuscript. NI, NT, and PF wrote the manuscript. All authors contributed to the article and approved the submitted version.

These studies were supported by European Foundation for the Study of Diabetes (Lilly European Diabetes Research Programme) and Department for the Economy, Northern Ireland.

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.