Modulation of the cannabinoid receptors by hemopressin peptides

Martha G. Bomar and Amit K. Galande

Abstract

Introduction

The endocannabinoid pathways have diverse roles in signaling and are involved in a variety of biological processes, such as addiction, appetite, memory, pain sensation, cognition, immune function and in behavioral responses to reward and stress. Many pathological conditions directly or indirectly involve change in the cannabinoid system, thus making this system an attractive for therapeutic development.

Cannabinoid ligands

The cannabinoid receptors are activated by mainly lipophilic compounds including the well-characterized endocannabinoids. Endocannabinoids, such as anandamide and 2-arachidonoylglylcerol or 2AG, are synthesized from membrane phospholipids on demand in response to intracellular increases in calcium. After their synthesis and release, the lipophilic messengers act on nearby cannabinoid receptors. The cannabinoid system also responds to stresses through altered expression of the cannabinoid receptors (Miller and Devi 2011).

Because of the involvement of the cannabinoid receptors in many diseases, numerous selective agonists and antagonists have been developed, facilitated by the moderate to low sequence identity between the two cannabinoid receptors. The endogenous and synthetic cannabinoid ligands exhibit wide chemical diversity without a common pharmacophore (Figure 1). Although the ligands all interact with a hydrophobic pocket within the transmembrane part of the receptors, they do not completely share the same binding sites (Lambert and Fowler 2005). As is the case for many GPCRs, deducing structure-activity relationships for receptor/ligand interactions is difficult and exasperated by the lack of a cannabinoid receptor crystal structure.

Figure 1

CB1 ligands: (A) Δ9-Tetrahydrocannabinol (Δ9-THC); (B) Rimonabant (SR141716); (C) Anandamide; (D) Hemopressin.

Cannabinoid receptor agonists include endogenous cannabinoids, synthetic agonists, phytocannabinoids such as Δ9-tetrahydrocannabinol, the active ingredient in marijuana (Howlett et al. 2002). Cannabinoid agonists may show promise in treating and managing neuronal disorders, loss of body weight, nausea/vomiting, and pain. In addition, development of selective CB1 inverse agonists and antagonists has been of great interest for therapeutic use for addictive disorders, pain, appetite suppression, and blood pressure reduction. Typified by rimonabant (SR141716), the 1,5-diarylpyrazoles are the most extensively studied class of CB1 antagonists/inverse agonists. Although studies demonstrated the efficacy of rimonabant in treating obesity and addictive disorders (Lazary et al. 2011), severe CNS side effects prevented its approval in the United States and suspended its use elsewhere. Selective cannabinoid agonists and antagonists lacking adverse side effects while maintaining therapeutic benefits are highly desired.

Peptide modulators of the cannabinoid system

N-terminally extended hemopressin peptides

N-terminally extended peptides of hemopressin, RVD-hemopressin and VD-hemopressin, were later identified and reported to be CB1 agonists (Gomes et al. 2009). Thus, a difference of only two or three residues appears to determine whether the hemopressin peptides exhibit antagonistic or agonistic activity, eliciting opposite effects at the CB1 receptor. Moreover, the signaling pathways for the N-terminally extended peptide agonists were found to be distinct from the classic G-protein-mediated pathway of lipid-based and synthetic agonists, resulting in a robust and sustained increase in Ca2+ release (Gomes et al. 2009). It remains unclear whether hemopressin is an endogenous peptide or instead a cleavage product of the longer RVD-hemopressin peptide. The Asp-Pro peptide bond is one of the most labile bonds (Figure 2), especially under acidic conditions (Marcus 1985), which were used in the extraction of rat brains when hemopressin was originally identified. Indeed, a subsequent mass spectrometry study of mouse brain extracts under different conditions did not identify hemopressin but instead only found the N-terminally extended peptides amongst others. Although processing of neuropeptides into bioactive peptides with varying lengths is quite common, such as with hemorphin, β-endorphin and angiotensin peptides, more research is needed to elucidate the mechanism of action and signaling pathways associated with the various hemopressin peptides.

Figure 2

Possible cleavage of RVD-Hemopressin at the Asp-Pro bond to create Hemopressin.

Generation and regulation of hemopressin peptides

The hemopressin peptides are different from classical neuropeptides; they are not produced by selective proteases at well-defined sites, stored within vesicles or secreted upon cell stimulation. In fact, little is known about how, when and where they are generated and transported to their targets and how their production is regulated. The hemopressin peptides are likely generated through proteasomal degradation of hemoglobin, although the hemoglobin source and location remain unknown. Hemoglobin α chain mRNA and protein have been found in non-erythrocyte cells, including in neurons (Gomes et al. 2010). Other bioactive peptides derived from cellular proteins have also been reported, including other hemoglobin-derived bioactive peptides (hemorphins and neokyotrophin) that interact with opiate and angiotensin receptors (Gomes et al. 2010). Evidence suggests that global ischemia may contribute to the regulation of hemopressin peptide generation from hemoglobin (Gomes et al. 2009).

Hemopressin-based drug development

As reported in the original discovery of hemopressin, the peptide is a substrate of three proteases (ep24.15, 24.16 and ACE), but their roles have not been fully elucidated. Proteolysis of the peptides may serve to inactivate/degrade the peptides or to create shorter physiologically relevant peptides. Hemopressin was shown to be cleaved into three shorter fragments (PVNF, PVNFK, and PVNFKF) by the ep24.15 and ep24.16 peptidases (Rioli et al. 2003). In vitro incubation of hemopressin with the promiscuous ACE resulted in efficient generation of the PVNFKFL fragment; however, this shorter fragment failed to decrease blood pressure in rabbits and rats (Blais et al. 2005). ACE inhibitors did not affect the hypotensive effect of hemopressin, suggesting that ACE does not inactivate hemopressin (Blais et al. 2005). These results are distinct from those of Dale et al., who showed that shorter hemopressin fragments (PVNFKF and PVNFKFL) exhibited similar antihyperalgesic activity to hemopressin (the shorter PVNFK and PVNF fragments were inactive) (Dale et al. 2005). The first six amino acids of the nonapeptide (PVNFKF) were found to be required for binding to the CB1 receptor and deletion of the C-terminal three residues did not affect receptor recognition (Heimann et al. 2007), even though the shorter fragment was found to adopt a different conformation than that of the nonapeptide in a micellular NMR study (Scrima et al. 2010). The physiological relevance of the shorter fragments is not clear. However, the ability of these shorter fragments to recognize the receptor and maintain bioactivity may be related to the reported oral activity of hemopressin (Heimann et al. 2007), as the peptide may be able to withstand some proteolysis before adsorption and still exhibit activity at the receptor.

The potential for developing peptidomimetic drugs based off of hemopressin or RVD-hemopressin is exciting. Although the number of peptide-based drugs has increased rapidly over the last two decades, their development requires strategies to overcome their susceptibility to proteolysis and their limited bioavailability and stability, which may prove to be useful for using hemopressin as a scaffold for drug development. Although most peptides are orally inactive, hemopressin has fortunately been shown to be orally active and capable of crossing the blood-brain barrier (Heimann et al. 2007). Because peptides, including many neuropeptides, are desirable as drugs for various indications, great efforts have been made to improve peptide stability and drug delivery to the brain (Witt and Davis 2006). For example, pepducins are cell-penetrating lipidated peptides that target the intracellular loops of a wide variety of GPCRs (Covic et al. 2002), with therapeutic value against many disease models. Similarly, numerous techniques are available for increasing the stability of peptides for use as therapeutics, including peptide drugs targeting GPCRs (Bellmann-Sickert and Beck-Sickinger 2010).

Self-assembly of hemopressin

Unfortunately, the enthusiasm surrounding the identification of hemopressin as a peptide inverse agonist of the CB1 receptor has been hampered by its inexplicable invariability in pharmacological assays (Gomes et al. 2010). We recently reported that hemopressin (1 mM peptide in 25 mM phosphate, 50 mM NaCl, pH 7.4) self-assembles to form nanostructured fibrils (Bomar et al. 2012), which may contribute to its inconsistent activity. Notably, no fibrillization was observed in RVD-hemopressin or in angiotensin II at the same concentration, underscoring the unique ability of hemopressin to adopt fibrils (Figure 3). Our results corroborated previously described data from Gomes et al., who found that higher amounts of hemopressin are retained on 2 kDa dialysis cassettes following 24 h of dialysis against phosphate-buffered saline compared to those of a non aggregating control peptide angiotensin II (Gomes et al. 2010). Although the higher peptide concentrations in these experiments may not be considered physiologically relevant, they are certainly pharmacologically relevant; indeed, assays rely on more concentrated peptide stocks, which are generally made within this concentration range. Moreover, high concentrations of peptide are also generated during peptide synthesis, purification, and are relevant in peptide formulations.

Figure 3

TEM images of 1 mM hemopressin (A) and RVD-hemopressin (B) in 25 mM phosphate, 50 mM NaCl, pH 7.4.

Self-assembly and fibrillization of peptides and proteins are complex processes. Although many short peptides have been shown to self-assemble, they are mostly laboratory-derived fragments of larger proteins that exhibit self-assembly. Because hemopressin is a naturally occurring peptide found within the brain, a better understanding of the physiological role of its self-assembly is needed. It is intriguing to speculate about the mechanism of hemopressin self-assembly. For example, stacking interactions of aromatic residues are often driving forces for peptide fibrillization and thus the two phenylalanine residues may contribute to hemopressin self-assembly (Gazit 2007). However, RVD-hemopressin, which contains the hemopressin nonapeptide sequence and also has two phenylalanine residues, does not form fibrils under the same conditions. The distinct differences in fibrillization between these peptides suggest that the three additional residues, RVD, cause structural differences or ionic interactions in RVD-hemopressin that prevent fibril formation.

Peptide amyloid fibril formation is implicated in numerous diseases, including Alzheimer’s, Parkinson’s, prion diseases and Type II diabetes (Chiti and Dobson 2006). In Alzheimer’s, the self-assembly of the amyloid Aβ peptide leads to the formation of neurotoxic amyolid plaques. Inhibiting peptide self-assembly with small molecules and peptides has been a therapeutic strategy for treating Alzheimer’s (Rafii and Aisen 2009). If hemopressin self-assembly is physiologically relevant and has pathogenic properties, it may also be of therapeutic value to target the self-assembly of this bioactive peptide.

Conclusions

Many of the mechanisms underlying CB1-mediated pathways and their integration with other pathways remain unclear, including in terms of motivation, reward, and satiety. Although there remains excitement over the finding of peptide modulators of the cannabinoid system, additional research is necessary to better understand the generation and regulation of the peptide ligands, the reason for hemopressin’s inconsistent activity and how the hemopressin-related peptides modulate the cannabinoid receptors. It is possible that RVD-hemopressin and hemopressin differ in only three amino acids but elicit completely opposite responses at the CB1 receptor, namely agonist and inverse agonist effects. But to validate this concept, systematic structure-activity studies are needed which are likely to elucidate molecular basis for the action of hemopressin peptides at the CB1 receptors. The finding that the peptide agonists are involved in signaling pathways distinct from lipid-based and synthetic agonists suggests additional complexity of CB1 receptor-mediated signaling pathways. If the hemopressin peptides are indeed CB1-selective inverse agonists and agonists, is it possible to use them to develop therapeutics for treating cannabinoid-related disorders? It is critical to investigate the physiological relevance of self-assembly and fibril formation of hemopressin since finding answers to this question may facilitate additional biological studies and drug development of hemopressin.

Acknowledgments

Footnotes

References