Human adrenomedullin (AM), a 52-residue peptide with an amidated C-terminus and an intramolecular disulfide bond, was initially identified as a vasodilatory peptide derived from a human pheochromocytoma [1
]. AM is expressed in many organs and tissues, including the gastrointestinal tract, vascular endothelium, adipose tissue, blood, heart, lung, kidney, pancreas, and the adrenal medulla [2
]. The peptide has multiple functions in anti-inflammation, vasodilation, angiogenesis, wound healing, and protection of the heart, blood vessels, and brain [4
]. AM showed therapeutic effects in experimental models of ischemic heart disease, stroke, retinochoroidal disease, and insulin intolerance [4
AM also showed therapeutic effects in an experimental model of dextran sulfate sodium (DSS)-induced colitis. Continuous or daily subcutaneous injection of AM reduced colitis severity [7
]. Moreover, AM showed a therapeutic effect in patients with ulcerative colitis and Crohn’s disease following continuous intravenous injection [10
]. However, these preclinical and clinical applications of AM required continuous or frequent administration because the plasma half-life of AM is only 4.87 ± 0.68 min (mean ± standard error of the mean, S.E.M.) [12
]. The N-terminus of AM was conjugated to 60 kDa polyethylene glycol (PEG) to elongate its plasma half-life (60kPEG-AM). It was found that 60kPEG-AM reduced the severity of DSS-induced colitis following a single subcutaneous injection in mice [13
]. While 60kPEG-AM represents a potential peptide therapeutic, clinically approved biopharmaceuticals contain PEG polymers only up to 50 kDa in size [14
]. With the goal of developing AM therapeutics for clinical applications, we prepared a 20 kDa PEGylated form of AM (20kPEG-AM) and evaluated its physiological, pharmacokinetic, and anti-colitis properties. We also evaluated the anti-colitis effect of repetitive administration of PEGylated AM to simulate the anticipated dosing of human patients in a clinical setting.
Although many new drugs for inflammatory bowel diseases have been developed, there are notable cases of treatment resistance, and the development of new drugs is urgently needed. In this study, we developed 20kPEG-AM and demonstrated its anti-colitis effect following both single and double subcutaneous injection in mice. Our findings suggest that 20kPEG-AM represents a new candidate therapeutic for inflammatory bowel disease.
PEGylated biologics have pharmacological advantages over the unmodified parental molecules, including extended circulating half-life, reduced dosing frequency, lower toxicity, decreased kidney clearance, decreased immunogenicity, maintenance of biological activity, increased stability, and protection from proteolytic degradation [18
]. We sought to use PEGylation to overcome the short plasma half-life of AM because the extended half-life of 20kPEG-AM would make continuous intravenous injection unnecessary in a clinical setting.
We successfully prepared 20kPEG-AM by native chemical ligation of an N-terminal (PEG20000-AM(1–15)-SCH2
) and a C-terminal segment (AM(16–52))20 kDa PEG [15
]. Successful conjugation was confirmed using HPLC following CNBr digestion and amino acid analysis. The efficiency of ligation between the N- and C- terminal segments was sufficient to enable the industrial-scale synthesis of 20kPEG-AM. In addition, purity was high enough for clinical applications. 20kPEG-AM increased cAMP production in treated HEK-293 cells with a pEC50 value of 7.27 ± 0.08, slightly lower than that of native AM. However, the plasma half-life of 20kPEG-AM was 7.4 h (95% confidence interval: 4.4–12.9 h), an extension of more than 100-fold compared with native AM [12
An anti-colitis effect was exerted following the subcutaneous injection of 20kPEG-AM in a mouse model of DSS-induced colitis. Our findings are consistent with those of previous studies showing that PEGylated AM has an anti-colitis effect [13
]. However, the PEG used in our study was smaller than 50 kDa. The maximum molecular weight of PEGs used in currently approved PEGylated biotherapeutics [14
]. As the approval process for new drugs with very different molecular sizes can be challenging, our data offer a clear path for clinical development of PEGylated AM as a biotherapeutic for inflammatory bowel diseases.
Repetitive subcutaneous injection of 20kPEG-AM exerted an anti-colitis effect in mice. For the treatment of chronic diseases such as inflammatory bowel disease, repetitive administration of PEGylated peptide biotherapeutics is typically required [21
]. Our report is the first to demonstrate that repetitive administration of PEGylated AM has an anti-colitis effect. A limitation of this finding is that a direct comparison between single and double injection protocols was not performed. Nonetheless, our data provide a foundation for future studies investigating the long-term anti-colitis effect of 20kPEG-AM.
Our study provides evidence on potential target trough plasma AM concentrations capable of exerting anti-colitis effects following administration of PEGylated AM. In this study, 20kPEG-AM exerted an anti-colitis effect when the trough AM concentrations were 18.73 ± 1.57 pM and 16.23 ± 0.69 pM on days 4 and 8, respectively. However, a previous study of 60kPEG-AM showed that much higher trough AM concentrations were required to exert an anti-colitis effect. Doses of 5 and 25 nmol/kg showed an anti-colitis effect, and a dose of 10 nmol/kg yielded a trough AM concentration of 865 pM on day 7. In clinical applications, the frequency of administration and dose of PEGylated AM are designed to achieve the targeted trough AM concentration while avoiding potential toxicity from overdosing. Our data serve as an essential foundation for future clinical efficacy and safety studies of PEGylated AM.
We found that 20kPEG-AM exerted anti-colitis effects at considerably lower doses, in terms of the human equivalent dose (HED) [22
], compared with two studies of native AM administration. While native AM required the administration of approximately 43 nmol/kg every day in humans [10
] and 0.65–6.5 nmol/kg HED every day in mice to achieve an anti-colitis effect [7
], 20kPEG-AM only required a dose of 0.81 nmol/kg HED administered every 4 days. This result was unexpected because our study also showed that 20kPEG-AM had relatively low in vitro efficacy compared with native AM. Other than the difference of administration route, one explanation of this result may be that PEGylated molecules can have improved functional pharmacodynamic properties for their natural receptors despite low efficacy in vitro [23
]. This improvement may also explain the fact that a single injection was sufficient to exert an anti-colitis effect, despite relatively low plasma AM concentrations on day 4. The potential to minimize dosing is an advantage for clinical applications because reducing the dosage is likely to minimize side effects, including hypotension [25
The mechanism underlying the anti-colitis effect of AM has yet to be discovered. Many studies have shown that increased cAMP concentrations can improve epithelial barrier function [27
]. It was recently reported that intravenous administration of AM resulted in increased cAMP levels in the colonic mucosa of mice showing ameliorated DSS-induced colitis [7
]. These findings imply that AM-induced cAMP in the mucosa may contribute to epithelial barrier protection and anti-colitis effects, downregulating proinflammatory cytokines such as phosphorylated NF-kB, tumor necrosis factor-α, interleukin (IL)-1β, IL-6, interferon-γ, IL-4, IL-12, and keratinocyte chemoattractant [7
]. Another mechanism could involve the upregulation of epithelium regeneration markers, such as Lgr5, Wnt5a, Egfr, or Erbb2 [33
], and regulation of gut bacterial composition through downregulation of TLR4 [34
]. Additional studies are needed to define the molecular mechanism of AM’s anti-colitis action.
One of the limitations of our study was the potential for confirmation bias. Although we observed a statistically significant difference in mice treated with 20kPEG-AM in one objective measure (bodyweight reduction score), other measures such as stool consistency and bloody stool scores are more subjective and susceptible to bias.