Dipeptidyl peptidase IV inhibitors in diabetes: more than inhibition of glucagon-like peptide-1 metabolism?

Inhibitors of the protease dipeptidyl peptidase IV (DPP-IV) are promising new drugs for the treatment of type 2 diabetes. They are thought to act by inhibiting the breakdown of glucagon-like peptide-1 and, thereby, selectively enhancing insulin release under conditions when it is physiologically required. These drugs are selective for DPP-IV, but the enzyme itself has a broad range of substrates other than glucagon-like peptide-1. Other high affinity substrates of DPP-IV including peptide YY may also play a role in the regulation of energy homeostasis. Moreover, DPP-IV is also known as CD26 and considered to be a moonlighting protein because it has a wide range of other functions unrelated to energy homeostasis, e.g. in immunity. The potential role of DPP-IV inhibition on substrates other than glucagon-like peptide-1 in diabetes patients remains to be elucidated.

Glucagon-like peptide-1 (GLP-1) is a hormone which is released following meals and stimulates insulin release from the pancreas. Its effects are terminated by breakdown by the enzyme dipeptidyl peptidase IV (DPP-IV). Therefore, inhibition of DPP-IV increases GLP-1 levels in the circulation and, hence, insulin release under conditions when it is needed, i.e. after a meal but not during fasting. Consequently, inhibition of GLP-1 inactivation is an insulinotropic principle which is unlikely to cause hypoglycaemia between meals. The lower risk for hypoglycaemic events as compared with other insulinotropic or insulin-sensitising agents makes DPP-IV inhibitors very promising candidates for a more physiological treatment of type 2 diabetes (Combettes and Kargar 2008).

In recent years, a number of selective DPP-IV inhibitors such as vildagliptin and sitagliptin have been evaluated in clinical trials (Hermansen et al. 2007; Utzschneider et al. 2008) and may have a future role in the treatment of type 2 diabetes (Combettes and Kargar 2008). In this issue of the journal, a novel DPP-IV inhibitor, ASP8497, is being introduced, which is highly selective for DPP-IV as compared to other peptidases (Someya et al. 2008). However, DPP-IV itself is not selective for GLP-1 but has a wide range of other natural substrates (Boonacker and Van Noorden 2003). Therefore, we wish to highlight potential implications of this lack of selectivity of DPP-IV for the use of DPP-IV inhibitors in diabetes treatment.

Another high affinity substrate of DPP-IV is peptide YY (PYY; Mentlein et al. 1993). Whereas cleavage of GLP-1 by DPP-IV causes inactivation, cleavage of PYY yields the long C-terminal fragment PYY3–36, which is inactive at some but active at other subtypes of PYY receptors. Specifically, it converts the non-subtype-selective agonist PYY into a selective agonist at Y2 and Y5 receptors (Michel et al. 1998). Thus, DPP-IV does not inactivate PYY but, rather, qualitatively alters its biological activity.

Indeed it has been reported that peripherally administered PYY3–36 inhibits food intake in rats, whereas PYY is a potent central stimulator of food intake (Batterham et al. 2002). Given the role of obesity in type 2 diabetes, prevention of the formation of such an endogenous food intake inhibitor by a DPP-IV inhibitor may be undesirable. It has been proposed that the effects of peripherally administered PYY3–36 on central nervous functions such as food intake may be mediated by excitation of afferent vagal fibres (Koda et al. 2005). However, the majority of subsequent rodent studies did not confirm inhibition of food intake by PYY3–36, particularly not following chronic administration (Boggiano et al. 2005). On the other hand, recent studies in non-rodents such as pigs (Ito et al. 2006) or humans (Degen et al. 2005; Sloth et al. 2007a; Sloth et al. 2007b) have reported reduced food intake upon peripheral administration of PYY3–36 but typically, these effects were found only at relatively high concentrations. Moreover, the effect of PYY3–36 on food intake was biphasic, depending on the duration of its administration (Parkinson et al. 2008). Two additional findings deserve consideration. Firstly, PYY3–36 was reported to promote fat oxidation and ameliorate insulin resistance in mice even under conditions of chronic administration where it did not reduce food intake (van den Hoek et al. 2006). Secondly, PYY3–36 was reported to lower plasma glucose levels even in the absence of alterations in circulating insulin levels (Bischoff and Michel 1998).

Taken together, the presently available data on PYY3–36 on food intake and metabolic parameters are not yet conclusive. However, it is clear that PYY3–36 is largely formed by DPP-IV, raising the possibility that selective DPP-IV inhibitors may exert part of their effects by modulating the PYY/PYY3–36 ratios. Therefore, it remains to be explored how possible effects on PYY cleavage contribute to metabolic effects of DPP-IV inhibitors in diabetic patients.

Moreover, DPP-IV is not only a protease for substrates relevant to energy homeostasis, but it also has a range of additional functions (Boonacker and Van Noorden 2003). Therefore, it is considered to be a moonlighting protein. As a protease, it has several other substrates, and it also acts as a receptor and costimulatory protein in the immune system. In this regard, CD26 is considered to be an important regulator of T-cell function (Reinhold et al. 2008). These pleiotropic effects of DPP-IV or CD26 lead to numerous potential uses of its inhibitors other than type 2 diabetes including inflammatory diseases (Ohnuma et al. 2006; Reinhold et al. 2007; Thielitz et al. 2008; Thompson et al. 2007) and, perhaps, certain types of cancers (Kikkawa et al. 2005; Thompson et al. 2007). Some of these effects may manifest as useful secondary actions when being used for the treatment of diabetic patients whereas others may manifest as adverse events. Most of these potential additional effects may not yet have manifested in the currently published diabetes literature, but it appears prudent to keep an eye on them.