Adenosine Receptors in Wound Healing, Fibrosis and Angiogenesis

Abstract

Wound healing and tissue repair are critical processes, and adenosine, released from injured or ischemic tissues, plays an important role in promoting wound healing and tissue repair. Recent studies in genetically manipulated mice demonstrate that adenosine receptors are required for appropriate granulation tissue formation and in adequate wound healing. A_2A and A_2B adenosine receptors stimulate both of the critical functions in granulation tissue formation (i.e., new matrix production and angiogenesis), and the A₁ adenosine receptor (AR) may also contribute to new vessel formation. The effects of adenosine acting on these receptors is both direct and indirect, as AR activation suppresses antiangiogenic factor production by endothelial cells, promotes endothelial cell proliferation, and stimulates angiogenic factor production by endothelial cells and other cells present in the wound. Similarly, adenosine, acting at its receptors, stimulates collagen matrix formation directly. Like many other biological processes, AR-mediated promotion of tissue repair is critical for appropriate wound healing but may also contribute to pathogenic processes. Excessive tissue repair can lead to problems such as scarring and organ fibrosis and adenosine, and its receptors play a role in pathologic fibrosis as well. Here we review the evidence for the involvement of adenosine and its receptors in wound healing, tissue repair and fibrosis.

1 Introduction

Tissue repair is an essential homeostatic mechanism that involves a series of coordinated and overlapping phases: inflammation, neovascularization, new tissue generation, and tissue reorganization. In acute inflammation, tissue damage is followed by resolution, whereas in chronic inflammation, damage and repair continue concurrently. Inflammatory cells neutralize invading pathogens, remove waste and debris, and promote restoration of normal function, either through resolution or repair. Inflammation also promotes angiogenesis and vasculogenesis, the formation of new blood vessels, which in turn may enhance the recruitment of inflammatory cells and the subsequent laying down of extracellular matrix to repair tissue damage. Although usually beneficial to the organism, inflammation may lead to tissue damage, resulting in escalation of chronic inflammation. Furthermore, aberrant or inadequate repair can lead to excessive and poorly ordered matrix deposition and fibrosis, which affects normal tissue architecture and can ultimately disable the proper functioning of organs. Like matrix generation, overly exuberant vessel formation may lead to medical problems as well, and diabetic retinopathy and macular degeneration are examples of this phenomenon.

Extracellular accumulation of adenosine in response to tissue damage is an important event in the control of all aspects of tissue repair. The nature of adenosine’s action depends on the magnitude of changes in extracellular adenosine concentrations as well as on the identity and expression levels of each adenosine receptor subtype on individual cell types. The role of adenosine in the regulation of inflammation is extensively covered in other chapters of this book. In this chapter, we will discuss the roles of specific adenosine receptors in the regulation of neovascularization and fibrosis in different organs and tissues.

2 Role of Adenosine in Neovascularization

Accumulating evidence indicates that adenosine is an important regulator of neovascularization, including angiogenesis and vasculogenesis. Stimulation of new blood vessel formation by adenosine was demonstrated in the chick chorioallantoic membrane and embryo (Adair et al. 1989; Dusseau et al. 1986; Dusseau and Hutchins 1988), the mouse retina (Afzal et al. 2003; Mino et al. 2001), and the optical tectum of Xenopus leavis tadpoles (Jen and Rovainen 1994). Adenosine reportedly modulates a number of steps involved in angiogenesis, including endothelial cell proliferation (Dubey et al. 2002; Ethier et al. 1993; Grant et al. 1999, 2001; Meininger et al. 1988; Meininger and Granger 1990; Van Daele et al. 1992), migration (Dubey et al. 2002; Grant et al. 2001; Lutty et al. 1998; Meininger et al. 1988; Teuscher and Weidlich 1985), and tube formation (Grant et al. 2001; Lutty et al. 1998). Adenosine has been also suggested to play an important role in adult vasculogenesis by directing the homing of endothelial progenitor cells to the site of tissue injury (Montesinos et al. 2004; Ryzhov et al. 2008b).

Adenosine has direct mitogenic effects on vascular cells that may contribute to angiogenesis (Ethier and Dobson Jr. 1997; Meininger et al. 1988; Sexl et al. 1995; Van Daele et al. 1992). However, the main proangiogenic actions of adenosine have been attributed to its ability to regulate the production of pro- and antiangiogenic substances. Adenosine modulates the release of angiogenic factors from various cells and tissues (Feoktistov et al. 2003, 2004; Gu et al. 1999, 2000; Hashimoto et al. 1994; Leibovich et al. 2002; Olah and Roudabush 2000; Pueyo et al. 1998; Takagi et al. 1996; Wakai et al. 2001; Zeng et al. 2003), thus regulating capillary growth in a paracrine fashion. In addition, adenosine can modulate release of angiogenic factors from endothelial cells (Desai et al. 2005; Feoktistov et al. 2002; Fischer et al. 1995, 1997; Grant et al. 1999; Khoa et al. 2003; Takagi et al. 1996), which may regulate capillary growth in an autocrine fashion.

All four adenosine receptor (AR) subtypes have been implicated in the regulation of neovascularization. In a similar manner to our early observation that the stimulation of A₁ARs on neutrophils increased their adherence to vascular endothelium (Cronstein et al. 1992), we have recently demonstrated that A₁ARs located on embryonic endothelial progenitor cells promote their adhesion to cardiac microvascular endothelial cells, suggesting an important role of this receptor subtype in vasculogenesis (Ryzhov et al. 2008b). A₁ARs have been also reported to upregulate vascular endothelial growth factor (VEGF) production from monocytes, thus promoting angiogenesis in an in vitro model (Clark et al. 2007). Among all of the AR subtypes, A₁ARs have the highest affinity to adenosine (Fredholm et al. 2001). It is possible, therefore, that engagement of the high-affinity A₁ARs is especially important for circulating cells moving toward a gradient of adenosine concentrations generated by tissue injury and/or hypoxia, whereas the lower-affinity A₂ARs are more important for the regulation of cells located in the vicinity of the injured or ischemic loci, where concentrations of adenosine are the highest.

Indeed, both A₂AR subtypes, A_2A and A_2BARs, have been implicated in regulation of angiogenesis and vasculogenesis. Depending on tissue or cell studied, either one of these receptor subtypes can take the lead and play a dominant role in the regulation of angiogenic factors. For example, A_2BARs upregulate the proangiogenic factors VEGF, basic fibroblast growth factor (bFGF), insulin-like factor 1, and IL-8 in human microvascular endothelial cells (Feoktistov et al. 2002; Grant et al. 1999). Conversely, A_2AARs were reported to upregulate VEGF in macrophages (Leibovich et al. 2002; Pinhal-Enfield et al. 2003). However, A_2BARs may also contribute to regulation of VEGF in these cells, since genetic deletion of A_2BARs significantly decreased adenosine-dependent secretion of VEGF in mouse peritoneal macrophages (our unpublished observations). In addition, the stimulation of A₃ARs in mast cells and some tumors can result in the upregulation of certain proangiogenic factors, complementing the actions of adenosine mediated via A_2BARs (Feoktistov et al. 2003; Merighi et al. 2005, 2007). Thus, the contribution of adenosine to the regulation of neovascularization can be dictated by the expression profile of AR subtypes and by the intracellular machinery to which they are coupled in specific cell types. Furthermore, the expressions of AR subtypes and their functions are subject to dynamic regulation by conditions present during inflammation, such as hypoxia and cytokine exposure (Bshesh et al. 2002; Eltzschig et al. 2003; Feoktistov et al. 2004; Khoa et al. 2003). Because the A_2BAR promoter contains a functional binding site for hypoxia-inducible factor (Kong et al. 2006), the onset of hypoxia strongly induces A_2BAR expression. Hypoxia-induced upregulation of A_2BARs has been reported in human tumor cells (Zeng et al. 2003), rat hippocampus (Zhou et al. 2004), and human dermal microvascular endothelial cells (Eltzschig et al. 2003). This may have important functional implications for regulation of angiogenesis. For example, in human bronchial smooth muscle cells and human umbilical vein endothelial cells, adenosine does not stimulate VEGF secretion under normoxic conditions, but hypoxia increases expression of A_2BARs, which are then able to stimulate VEGF release (Feoktistov et al. 2004). Similarly, treatment of human dermal microvascular endothelial cells with interferon (IFN)-γ increases A_2BAR expression but decreases A_2AAR levels. In contrast, other proinflammatory cytokines, such as interleukin (IL)-1 and tumor necrosis factor alpha (TNF-α) increase both A_2A and A_2B AR expression and function (Khoa et al. 2003). Because the expression and function of adenosine receptor subtypes may differ depending on the tissue and the nature of the tissue injury, we will next examine the role of AR subtypes in specific organs and pathological states.

2.1 Regulation of Neovascularization in the Skin

We have previously reported (Montesinos et al. 2002) that mice with genetically disrupted A_2AARs form significantly fewer microvessels in healing wounds and in response to mechanical trauma by the formation of an air pouch (Montesinos et al. 2002). Furthermore, application of an A_2AAR agonist to wounds increases microvessel formation from both pre-existing endothelial cells and bone marrow-derived endothelial progenitors as compared to vehicle-treated mice, observations that provide the first in vivo evidence that A_2AAR occupancy promotes angiogenesis and vasculogenesis (Montesinos et al. 2002, 2004). Further studies indicate that the angiogenic effects of A_2AAR occupancy are mediated both directly on endothelial cells (increased endothelial cell migration and microvascular endothelial cell VEGF production; Khoa et al. 2003; Montesinos et al. 1997) and indirectly via promotion of VEGF production by macrophages (Leibovich et al. 2002). Desai and colleagues (Desai et al. 2005) have also reported evidence to indicate that A_2AAR occupancy suppresses the production of thrombospondin I, a potent inhibitor of angiogenesis, and this inhibition is responsible for enhanced vascular tube formation in vitro. Thus, there is growing evidence that A_2AARs play an important role in skin neovascularization, and particularly during wound healing.

2.2 Regulation of Neovascularization in the Heart and Skeletal Muscles

Many studies have demonstrated that chronic elevation of tissue adenosine concentrations induced by the adenosine reuptake blocker dipyridamole (Adolfsson et al. 1981, 1982; Adolfsson 1986a, b; Belardinelli et al. 2001; Mall et al. 1987; Mattfeldt and Mall 1983; Symons et al. 1993; Tornling et al. 1978, 1980a, b; Tornling 1982a, b; Torry et al. 1992), or long-term administration of adenosine and its analogs (Hudlicka et al. 1986; Wothe et al. 2002; Ziada et al. 1984), promotes capillary proliferation in the heart and skeletal muscles. Antagonism of ARs with caffeine abrogated VEGF upregulation in skeletal muscles induced by local injection of adenosine 5′-N-ethyluronamide (NECA) into the mouse hind limb and produced a 46% reduction in neovascularization in a mouse ischemic hind limb model (Ryzhov et al. 2007). In the isolated heart model, adenosine but not selective A_2A or A₃ AR agonists increased retention of embryonic endothelial progenitors to microvascular endothelium, suggesting that A₁ and A_2BARs may play an important role in the initial phase of vasculogenesis, promoting homing of endothelial progenitor cells to the site of ischemic injury (Ryzhov et al. 2008b). Indeed, endothelial progenitor cells and cardiac microvascular endothelial cells preferentially express functional A₁ and A_2BARs, respectively, and both subtypes are involved in the regulation of the adhesion of endothelial progenitors to microvascular endothelial cells in the heart. Moreover, the interaction between P-selectin and its ligand PSGL-1 plays an important role in these process, and stimulation of A_2BARs in cardiac microvascular endothelial cells induces rapid cell surface expression of P-selectin (Ryzhov et al. 2008b). These findings suggested a role for A₁ and A_2BARs in myocardial vasculogenesis, and provided a rationale for the potential use of adenosine to stimulate engraftment in cell-based therapies.

2.3 Regulation of Neovascularization in the Lung

Angiogenesis is a feature of chronic lung diseases such as asthma and pulmonary fibrosis. Studies in adenosine deaminase (ADA)-deficient mice, characterized by elevated lung tissue levels of adenosine, strongly suggest a causal association between adenosine and an inflammatory phenotype (Blackburn et al. 2000; Blackburn 2003). These mice exhibit a lung phenotype with features of lung inflammation, bronchial hyperresponsiveness, enhanced mucus secretion, increased IgE synthesis, and elevated levels of proinflammatory cytokines and angiogenic factors that could be reversed by lowering adenosine levels with exogenous ADA (Blackburn et al. 2000). In particular, levels of the angiogenic chemokine CXCL1 (mouse functional homolog of human IL-8) are significantly elevated in an adenosine-dependent manner in the lungs of ADA-deficient mice, leading to substantial angiogenesis in the tracheas (Mohsenin et al. 2007a). The A_2BAR subtype appears to play an important role in this model, because pharmacological inhibition of A_2BARs significantly reduced elevations in proinflammatory cytokines as well as mediators of airway remodeling induced by high adenosine levels in the lungs of ADA-deficient mice (Sun et al. 2006). In contrast, genetic removal of the A_2AAR enhances pulmonary inflammation, mucin production, and angiogenesis in ADA-deficient mice (Mohsenin et al. 2007b).

2.4 Regulation of Neovascularization in Tumors

Metabolically active solid tumors grow rapidly and routinely experience severe hypoxia and necrosis, which causes adenine nucleotide degradation and adenosine release. Expression of A_2BARs was documented in various cancerous cells (Feoktistov and Biaggioni 1993, 1995; Panjehpour et al. 2005; Phelps et al. 2006; Rodrigues et al. 2007; Zeng et al. 2003), and analysis of gene expression in primary human tumors uncovered overexpression of A_2BARs, suggesting their potential role in cancer biology (Li et al. 2005). Studies from different laboratories demonstrate that stimulation of A_2BARs in cancer cell lines upregulates the production of angiogenic factors, suggesting that tumor A_2BARs may promote neovascularization (Feoktistov et al. 2003; Merighi et al. 2007; Zeng et al. 2003). A₃ARs expressed in some tumor cell lines may also complement these A_2BAR-mediated effects by up-regulating other proangiogenic factors (Feoktistov et al. 2003; Merighi et al. 2005, 2007). In addition, host tumor-infiltrating immune cells can also play an important role in tumor angiogenesis, since Lewis lung carcinoma isografts in A_2BAR knockout mice contained lower VEGF levels and exhibited lower vessel density compared to tumors grafted in wild-type mice (Ryzhov et al. 2008a). Furthermore, treatment with A_2A/A_2BAR antagonists inhibited neovascularization of CL8-1 melanoma in mice (Ohta et al. 2006). Thus, there is growing evidence that adenosine acting via A_2B and possibly A₃ or A_2AARs can promote tumor neovascularization. Involvement of different AR subtypes in the regulation of neovascularization is not surprising due to the multifaceted mechanism of blood vessel development.

3 Role of Adenosine in Fibrosis

3.1 A_2A Adenosine Receptor Agonists Promote Wound Healing

Recent reports indicate that topical application of an A_2AAR agonist increases the rate at which wounds close (Montesinos et al. 1997). That A_2AARs were involved in this pharmacologic effect was demonstrated by the observation that a specific A_2AAR antagonist, but not antagonists at other ARs, reversed the effect of the selective A_2AAR agonist CGS21680 on wound healing. Treatment of wounds with this AR agonist promoted fibroblast migration in vitro, and in the AR agonist-treated mice there was an increase in matrix and fibroblast infiltration into the wounds (Montesinos et al. 1997). More recent studies demonstrate that a more highly selective A_2AAR agonist, sonedenoson, is a more potent promoter of wound healing than recombinant platelet derived growth factor (becaplermin) (Victor-Vega et al. 2002). The role of A_2AARs in the promotion of wound healing was more fully confirmed by the observation that a selective A_2AAR agonist promotes wound healing in wild-type but not A_2AAR knockout mice (Montesinos et al. 2002; Victor-Vega et al. 2002). In these studies, there was a marked increase in the number of blood vessels in the healing wounds of wild-type mice treated with the A_2AAR agonist as compared to untreated controls. Absence of A_2AARs was associated with disorganized granulation tissue although re-epithelialization was not delayed in the knockout mice. In contrast to this study, Sun and colleagues observed that N₆-cyclopentyladenosine, a relatively selective A₁AR agonist, promotes wound healing (Sun et al. 1999). In this study, there was no confirmation that the high concentrations of the agonist used were indeed selective for A₁ARs or whether the phenomenon could be mediated by A_2AARs. These findings indicate that A_2AARs stimulate wound healing by modulating inflammatory cell, endothelial cell and fibroblast functions that promote wound healing. A topical A_2AAR agonist, sonedenoson, is currently undergoing testing in Phase II clinical trials for the treatment of diabetic foot ulcers.

3.2 A_2A Adenosine Receptor Occupancy Stimulates Fibroblast Matrix Production

Replacement of the collagenous matrix of the skin and other tissues is an integral part of wound healing. Once the debris and destroyed matrix at the site of injury are eliminated, fibroblasts lay down a new matrix. This matrix may be remodeled over a longer period of time and the wound develops the characteristic appearance of a scar. A_2AAR occupancy stimulates fibroblasts to synthesize type I and III collagen at an increased level, similar to that induced by the growth factor transforming growth factor (TGF)-β, and downregulates matrix metalloproteinase (MMP) 9 but not MMP2 (Chan et al. 2006a).

The observation that adenosine, acting at A_2AARs, stimulates the formation of matrix suggests the possibility that adenosine A_2AARs play a role in fibrosing conditions and scarring, a hypothesis confirmed by in vivo experiments. Animals lacking A_2AARs or treated with an A_2AAR antagonist were protected from developing diffuse dermal fibrosis in response to bleomycin (Chan et al. 2006a). The role of A_2AARs in fibrosis in tissues outside of the skin is less clear. Prior studies have demonstrated that A_2BARs regulate production of collagen in pulmonary and cardiac fibroblasts (Chen et al. 2004; Dubey et al. 2000, 2001), but other studies have demonstrated that A_2AARs regulate collagen I and III production by hepatic stellate cells (Che et al. 2007), the fibroblasts of the liver, and A_2AAR knockout mice are protected from developing hepatic fibrosis following treatment with either CCl₄ or thioacetamide (Chan et al. 2006b). These observations help to explain the protection against death from liver disease provided by coffee drinking (Corrao et al. 1994, 2001; Gallus et al. 2002; Klatsky et al. 1993, 2006; Klatsky and Armstrong 1992; Ruhl et al. 2005; Sharp et al. 1999; Tverdal and Skurtveit 2003), since caffeine is a relatively weak and nonselective AR antagonist which offers some protection (although not complete) from the development of hepatic fibrosis in murine models (Chan et al. 2006b).

In a murine model of diffuse dermal fibrosis resembling scleroderma, we have also found that A_2AARs play a central role in the development of fibrosis. A_2AARs are present on human dermal fibroblasts and, when occupied, regulate collagen production by these cells (Chan et al. 2006a). Mice treated with subcutaneous bleomycin develop diffuse dermal fibrosis and we found that both A_2AAR knockout mice and mice treated with a selective A_2AAR antagonist were protected from the development of bleomycin-induced dermal fibrosis (Chan et al. 2006a). These results are consistent with the hypothesis that A_2AARs play a role in organ and tissue fibrosis and that blockade or elimination of these receptors can prevent fibrosis.

Recently published indirect evidence provides further support for a role for adenosine and its receptors in dermal fibrosis. Imiquimod is an immune modulator that promotes a shift from Th2- to Th1-type immune responses (reviewed in Schon and Schon 2007) by mechanisms that have not been fully evaluated. Studies in inflammatory cells indicate that imiquimod, at pharmacologically relevant concentrations, is an A_2AAR antagonist, and that this may account for its immunological effects (Schon et al. 2006). Imiquimod, applied topically, has been used to treat morphea, a skin disease characterized by localized fibrosis, and its use has been advocated for the treatment of Dupuytren’s contracture, another fibrosing disease (Dytoc et al. 2005; Man and Dytoc 2004; Namazi 2006; Schon et al. 2006). While intriguing (and supporting the clinical relevance of this work), we do realize the anecdotal nature of these reports.

3.3 A_2B Adenosine Receptor Occupancy Regulates Fibroblast Collagen Production and Fibrosis

As described above, a number of recent studies have demonstrated that cardiac and pulmonary fibroblasts express A_2BARs that regulate their production of collagen (Chen et al. 2004; Dubey et al. 1997, 1998; Zhong et al. 2005). Stimulation of A_2BARs in cardiac fibroblasts inhibited their proliferation, protein synthesis and collagen production (Chen et al. 2004; Dubey et al. 1997, 1998). Furthermore, it has been demonstrated in vivo that long-term stimulation of A_2BARs after myocardial infarction prevents cardiac remodeling (Wakeno et al. 2006). In contrast, studies in ADA-deficient mice indicate that these animals develop pulmonary inflammation and pulmonary fibrosis that appear to be mediated by A_2BARs (Sun et al. 2006), thus suggesting a role for A_2BARs in pulmonary fibrosis. Based on these studies and the results described above, it is reasonable to conclude that adenosine can either inhibit (heart) or stimulate (skin, liver, lungs) fibrosis, and that adenosine-regulated fibrosis is mediated by different receptors depending on which organ is studied (skin and liver vs. heart and lungs).

3.4 A₁ Adenosine Receptors Play a Role in Cardiac and Vascular Fibrosis

Recently, Kalk and coworkers reported that SLV320 (Solvay Pharmaceuticals), a highly selective A₁AR antagonist, reduced myocardial fibrosis in a model of uremic cardiomyopathy (Kalk et al. 2007). In this model, partially (5/6) nephrectomized rats were treated with SLV320 or vehicle and myocardial fibrosis was markedly reduced, as was albuminuria, without any change in blood pressure or other factors that might have accounted for the change. Another problem associated with fibrosis and abnormal “wound” healing that may be mediated by A₁ARs is intimal hyperplasia and stenosis following stent placement, and recent studies suggest that an A₁AR antagonist diminishes both intimal hyperplasia and smooth muscle proliferation in a model of stent stenosis (Edwards et al. 2008) Thus, A₁ARs may also play a role in fibrosis, although their role seems to be confined to the cardiovascular system.

4 Conclusion

Adenosine and its receptors play important roles in both matrix production and neovascularization, processes that are critical for wound healing and tissue repair. Moreover, adenosine and its receptors play a direct role in stimulating fibrosis in the skin, lungs and liver, but inhibiting fibrosis in the heart. Adenosine and its receptors may also play an important role in physiologic and pathologic angiogenesis. Targeting of ARs to promote wound healing and neovascularization of ischemic tissues or to diminish pathologic fibrosis and angiogenesis is currently underway.

Abbreviations

ADA

Adenosine deaminase

AR

Adenosine receptor

bFGF

Basic fibroblast growth factor

IFN

Interferon

IL

Interleukin

MMP

Matrix metalloproteinase

NECA

Adenosine 5′-N-ethyluronamide

TGF

Transforming growth factor

TNF-α

Tumor necrosis factor alpha

VEGF

Vascular endothelial growth factor