Regulation of enteric functions by adenosine: Pathophysiological and pharmacological implications

Abstract

The wide distribution of ATP and adenosine receptors as well as enzymes for purine metabolism in different gut regions suggests a complex role for these mediators in the regulation of gastrointestinal functions. Studies in rodents have shown a significant involvement of adenosine in the control of intestinal secretion, motility and sensation, via activation of A₁, A_2A, A_2B or A₃ purinergic receptors, as well as the participation of ATP in the regulation of enteric functions, through the recruitment of P2X and P2Y receptors. Increasing interest is being focused on the involvement of ATP and adenosine in the pathophysiology of intestinal disorders, with particular regard for inflammatory bowel diseases (IBDs), intestinal ischemia, post-operative ileus and related dysfunctions, such as gut dysmotility, diarrhoea and abdominal discomfort/pain. Current knowledge suggests that adenosine contributes to the modulation of enteric immune and inflammatory responses, leading to anti-inflammatory actions. There is evidence supporting a role of adenosine in the alterations of enteric motor and secretory activity associated with bowel inflammation. In particular, several studies have highlighted the importance of adenosine in diarrhoea, since this nucleoside participates actively in the cross-talk between immune and epithelial cells in the presence of diarrhoeogenic stimuli. In addition, adenosine exerts complex regulatory actions on pain transmission at peripheral and spinal sites. The present review illustrates current information on the role played by adenosine in the regulation of enteric functions, under normal or pathological conditions, and discusses pharmacological interventions on adenosine pathways as novel therapeutic options for the management of gut disorders and related abdominal symptoms.

Introduction

The earliest experimental observation describing the biological activity of adenosine in the digestive system can be dated back to 1949, when it was demonstrated that adenine derivatives induced a significant inhibition of the spontaneous contractile activity of ileum in rabbits, hamsters and guinea-pigs (Ewing et al., 1949). This study was followed by other pioneering investigations which examined the effects of exogenous adenosine on intestinal motility (Mihich et al., 1954, Stafford, 1966), but it was in the early 1970s that more specific studies led to the identification of the purinergic system, and described its significant impact on the physiology of enteric neurotransmission. In particular, based on findings from experiments designed to assess the role of previously identified non-adrenergic, non-cholinergic (NANC) neurotransmitters (Burnstock et al., 1966, Ambache and Freeman, 1968), Burnstock et al. (1970) proposed the novel concept that adenosine triphosphate (ATP) and related nucleotides/nucleosides (adenosine diphosphate, ADP; adenosine monophosphate, AMP; adenosine) act as transmitters involved in NANC-mediated relaxing responses of smooth muscle in the gastrointestinal tract and bladder. Two years later, the term “purinergic” was coined and the purinergic neurotransmission hypothesis was put forward (Burnstock, 1972). This concept initially met considerable resistance and skepticism in the scientific community, used to regarding the purine system only as an ubiquitous biochemical source of energy, but subsequently it met wide acceptance and became a cornerstone in the physiology of gastrointestinal tract.

Implicit in the purinergic hypothesis was the existence of purinergic receptors, which were first characterized in 1976 (Burnstock, 1976, Spedding and Weetman, 1976). A step forward was taken in 1978 when, in a seminal review, Burnstock proposed a basis for distinguishing two types of purinoceptors, named P1 and P2, which were preferentially activated by adenosine and ATP, respectively (Burnstock, 1978). About at the same time, pharmacological studies allowed the distinction of two P1 receptor subtypes, based on their ability to inhibit (A₁ receptor) or stimulate (A₂ receptor) intracellular cAMP accumulation (Van Calker et al., 1979, Londos et al., 1980). In 1985, Burnstock and Kennedy proposed pharmacological criteria for discriminating two types of ATP P2 receptors (P2X and P2Y) (Burnstock and Kennedy, 1985). Subsequently, on the basis of studies on transduction mechanisms and cloning of nucleotide receptors, it was established that P2X are ligand-gated ion channel receptors, whereas P2Y belong to the G-protein-coupled receptor family (Abbracchio & Burnstock, 1994). Several ATP receptor subtypes have been then identified on myenteric and submucosal neurons, where they control the synaptic neurotransmission and contribute to the neuromodulation of gut functions (Burnstock, 2008).

The above observations fostered strong interest in the role played by adenosine in the modulation of enteric functions. It was initially demonstrated that adenosine contributes to the inhibitory regulation of intestinal motility through the reduction of acetylcholine release from myenteric nerves (Vizi and Knoll, 1976, Kazić and Milosavljević, 1976), as well as via activation of purinergic receptors on smooth muscle cells (Ally and Nakatsu, 1976, McKenzie et al., 1977). In the same period, the presence of enzymes involved in adenosine metabolism was documented in the intestinal mucosa (Kolassa et al., 1977, Harms and Stirling, 1977), and these findings, together with the subsequent identification of adenosine receptors in the intestinal epithelium (Dobbins et al., 1984, Barrett et al., 1989) and submucosal plexus (Barajas-Lopez et al., 1991, Barajas-López, 1993), paved the way to the characterization of the role played by adenosine in the control of bowel fluid and electrolyte transport. Other investigations, indicating an involvement of adenosine in the regulation of gut sensory functions, then suggested potential therapeutic applications of drugs acting on adenosine receptors in the management of functional digestive disorders associated with abdominal pain, such as irritable bowel syndrome (IBS) (Geiger et al., 1984, DeLander and Hopkins, 1987, Takaki et al., 1993, Bueno et al., 1997).

The concept of purinergic signalling has broadened through the years to include its regulatory actions on the immune system (Sitkovsky and Lukashev, 2005, Sitkovsky and Ohta, 2005). Initial demonstrations concerning the ability of extracellular adenosine to modulate immune processes go back to 1970s (Giblett, 1976, Seegmiller et al., 1977, Allison et al., 1977). Since then, detailed studies have indicated adenosine as a prominent player in the physiological mechanisms deputed to down-regulate activated immune cells and protect tissues from inflammatory damage via specific receptor subtypes expressed on immune/inflammatory cell populations (lymphocytes, neutrophils, monocytes, macrophages and dendritic cells) (Ohta and Sitkovsky, 2001, Fortin et al., 2006, Desrosiers et al., 2007). The involvement of adenosine pathways in the anti-inflammatory and immunomodulating effects exerted by drugs employed in the medical management of chronic inflammatory diseases (i.e. methotrexate, salicylates) has also become apparent (Amann and Peskar, 2002, Montesinos et al., 2007). These observations have stimulated the research of novel drugs suitable for treatment of intestinal inflammatory disorders through the pharmacological modulation of adenosine pathways (Siegmund et al., 2001, Odashima et al., 2005, Guzman et al., 2006, Cavalcante et al., 2006, Antonioli et al., 2007). At present, some of these compounds are being tested also in preclinical models of non-digestive diseases (asthma, chronic obstructive pulmonary disease, diabetes) with encouraging results (Fozard et al., 2002, Mustafa et al., 2007, Van den Berge et al., 2007, Németh et al., 2007), while others have already entered the phase of clinical development for treatment of rheumatoid arthritis (Silverman et al., 2008).

Thus, after a troubled start, adenosine is now a rapidly expanding field, even though several aspects of its functions at gastrointestinal level need to be clarified. The present review is intended to illustrate and discuss current information on the role played by adenosine pathways in the regulation of enteric motor, secretory, sensory and immune functions, under normal conditions as well as in the presence of functional and inflammatory gut disorders. Special attention has been paid to the implications of adenosine in the pathophysiology of gut dysfunctions (dysmotility, diarrhoea, visceral pain) and to the possibility of modulating such disturbances by pharmacological interventions on molecular targets within the adenosine network (receptors, enzymes, transporters).

The biological actions of adenosine are mediated by G-protein-coupled receptors currently distinguished into four subtypes: A₁, A_2A, A_2B and A₃ (Klotz, 2000, Fredholm et al., 2000). Since the classification of adenosine receptors has been exhaustively addressed in a number of previous papers (Linden, 1991, Fredholm et al., 1996, Fredholm et al., 1997, Ralevic and Burnstock, 1998, Olah and Stiles, 2000, Fredholm et al., 2000, Fredholm et al., 2001), readers are referred to these reviews for detailed information. The ability of adenosine to regulate several biological functions is strictly related to its extracellular concentration. The levels of adenosine at its receptors are determined by a variety of mechanisms, which include intracellular and extracellular adenosine biosynthesis, as well as cellular adenosine release, reuptake and metabolism (Noji et al., 2004). These processes are intertwined, subjected to highly dynamical regulation and linked in a complex manner to the energy balance of tissues (Deussen, 2000). There is also evidence to suggest that extracellular adenosine levels can vary significantly in response to several pathological conditions (Cronstein, 1995, Latini and Pedata, 2001, Sitkovsky and Ohta, 2005).

Under physiological conditions, adenosine is formed mainly at the intracellular level from S-adenosylhomocysteine by S-adenosylhomocysteine hydrolase, and transported across cell membranes by nucleoside transporters, which play a key role in the control of extracellular adenosine concentrations (Cabrita et al., 2002). These transporters are classified into two categories according to their molecular and functional characteristics (Noji et al., 2004): 1) equilibrative nucleoside transporters (ENTs), which carry nucleosides across cell membranes in either direction, depending on concentration gradients; they include four subtypes, designated as ENT1, ENT2, ENT3 and ENT4 (Baldwin et al., 2004); 2) concentrative nucleoside transporters (CNTs), subdivided in CNT1, CNT2 and CNT3, which promote the intracellular influx of nucleosides against their concentration gradient, using the sodium ion gradient across cellular membranes as a source of energy (Gray et al., 2004). After intracellular reuptake, adenosine undergoes rapid phosphorylation to AMP by adenosine kinase, or deamination to inosine by adenosine deaminase. These pathways ensure the maintenance of low intracellular adenosine concentrations through a strict enzymatic control (Noji et al., 2004) (Fig. 1A). Another relevant source of extracellular adenosine is represented by ATP physiologically released from nerve endings, immune cells and smooth muscle cells (Haskó et al., 2005, Burnstock, 2007).

Adverse conditions, including hypoxia or inflammation, are associated with increased intracellular and extracellular dephosphorylation of ATP to adenosine through ecto-apyrase (also named CD39) and 5′-nucleotidase enzymes (endo-5′-nucleotidase and ecto-5′-nucleotidase) in parallel with the suppression of adenosine kinase activity (Deussen, 2000). Recent evidence indicates that during inflammatory insults, ecto-5′-nucleotidase, designated also as CD73, represents a critical check point for the control of adenosine production, deputed to preserve tissue integrity (Narravula et al., 2000, Niemelä et al., 2004, Colgan et al., 2006). Indeed, increments of adenosine levels have been shown to result in readjustments of the energy supply-to-demand ratio via augmentation of organ blood flow by vasodilatation, and protection of inflamed tissues by down-regulation of the immune response (Niemelä et al., 2004, Sitkovsky and Ohta, 2005) (Fig. 1B). Under pathological conditions associated with impairment of energy supply to cells, such as hypoxia or ischemia, a relevant contribution to adenosine production is given also by adenylate kinase, since this enzyme acts on ADP to generate ATP plus AMP, the latter being a substrate for adenosine formation through dephosphorylation (Hardie, 2003, Yegutkin, 2008).

Adenosine concentrations in the extracellular compartment are controlled by ecto-enzymes catalysing its conversion into inosine (Haskò & Cronstein, 2004), and ultimately to the stable end product uric acid (Haskò et al., 2004). Currently known ecto-enzymes include adenosine deaminase and adenosine kinase. Adenosine deaminase, regarded mainly as a cytosolic enzyme (endo-adenosine deaminase), can be expressed also on the external membrane surface of several immune and non-immune cells (ecto-adenosine deaminase) (Cristalli et al., 2001, Latini and Pedata, 2001).

Two studies have examined the expression and localization of adenosine receptor subtypes in the human gastrointestinal tract (Puffinbarger et al., 1995, Christofi et al., 2001), through reverse transcription-polymerase chain reaction (RT-PCR) and immunohistochemical analysis. These investigations have demonstrated a wide distribution of adenosine receptors in the neuromuscular compartment and mucosa/submucosal layer of both small and large intestine. In addition, among the enzymes involved in adenosine metabolism, adenosine deaminase has been found in various compartments of the human intestinal wall, whereas the expression of CD73, ENTs and CNTs has been investigated only at the mucosal level. Current information on the expression and compartmental localization of adenosine receptors, enzymes and transporters in the human intestine are summarized in Table 1. With regard to rodents, most of data on the distribution of adenosine receptors in rat gastrointestinal tract are based on studies designed to identify mRNA without further characterization of cellular localization, and they support the expression of the four adenosine receptors in both small and large intestine, as summarized in Table 2. Such information is lacking for mice and guinea-pigs, but several pharmacological studies have demonstrated that adenosine receptors and enzymes are functionally active in the digestive system of these rodent species (Table 2). Functional studies which have identified the presence of adenosine receptors and enzymes in various parts of the gut, both in humans and rodents, are discussed in Section 2.1.

The expression of adenosine deaminase in both small and large intestine of rat and mouse has been demonstrated by immunohistochemistry, which revealed a predominant localization in the mucosal layer (Dinjens et al., 1989). The enzyme CD73 has been found in the small intestine of rat and guinea-pig as well as in rat and mouse colon (Nitahara et al., 1995, Karhausen et al., 2004, Giron et al., 2008). A faint expression of equilibrative nucleoside transporters (ENT1, ENT2 and ENT3) has been shown in the small and large intestine of rat and mouse, while a marked expression of CNT1 and CNT2 has been detected in the small intestine, and to a lesser extent in the colon, of both species. In the same study, a scarce or null presence of CNT3 has been found both in the small and large intestine (Lu et al., 2004) (Table 2).