Learning from rejection: What transplantation teaches us about (other) vascular pathologies

Journal of Autoimmunity 45 (2013) 80e89

Contents lists available at SciVerse ScienceDirect

Journal of Autoimmunity journal homepage: www.elsevier.com/locate/jautimm

Review

Learning from rejection: What transplantation teaches us about (other) vascular pathologies Richard N. Mitchell* Pathology and Health Sciences and Technology, Harvard Medical School and Brigham and Women’s Hospital, 77 Avenue Louis Pasteur/NRB 730D, Boston, MA 02115, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 May 2013 Accepted 30 May 2013

Allograft vasculopathy is an accelerated intimal hyperplastic lesion leading to progressive vascular stenosis; it represents the major long-term limitation to successful solid organ transplant. Although allograft vasculopathy is not formally an autoimmune disease, nor does it constitute a major cause of cardiovascular disease on a purely numerical basis, its pathogenesis provides an important window on the mechanisms by which immune injury can drive more common vascular pathologic entities. Thus, insights gleaned from vascularized solid organ transplants can shed new mechanistic (and therapeutic) light on: 1) the intimal vascular responses accompanying typical atherosclerosis and other inflammatory vessel diseases (e.g., scleroderma); 2) the pathogenesis of vascular stenosis versus aneurysm formation; 3) the sources of intimal smooth muscle cells in the healing of any vascular injury; and 4) the mechanisms by which smooth muscle cells are recruited into intimal lesions. Indeed, research on allograft vasculopathy has led to the understanding that interferon-g plays a similar pathogenic role in a host of vascular stenosing lesionsdand that Th2 cytokines can drive vascular remodeling and aneurysm formation. Moreover, circulating precursors (and not just medial smooth muscle cells) contribute to the intimal hyperplasia seen in atherosclerosis and in-stent restenosis. That non-vessel smooth muscle cells can be recruited to sites of vessel injury further suggests that chemokine and adhesion molecule interactions may be viable targets to limit vascular stenosis in a wide range of vascular lesions. This review will describe the pathogenesis of allograft vasculopathy, and will relate how understanding the underlying pathways informs our understanding of both human transplant-associated disease, as well as other human vascular pathologies. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Allograft vasculopathy Atherosclerosis Aneurysm Intimal hyperplasia Chemokine

1. Introduction Allograft vasculopathy (AV) is a progressive and diffuse intimal hyperplastic lesion of arteries that leads to insidious vessel narrowing, and eventually to allograft ischemia. As opposed to atherosclerosis which typically requires decades before becoming clinically significant, the pace of vessel stenosis in AV is accelerated, accruing over the course of just months to years. Even in an era of immunosuppressive therapies that can effectively block acute cellular and humoral rejection, AV remains the major cause of long-term allograft failure [1]; the “half-life” of most transplanted hearts, for example, remains stubbornly pinned at approximately a decadedlargely unchanged since the 1980’s [2].

* Tel.: þ1 617 525 4303. E-mail addresses: [email protected], [email protected]. 0896-8411/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jaut.2013.05.006

In kidneys, the disease manifests as a progressive renal insufficiency; in liver, ischemic damage is reflected by bile duct drop-out; and in the lungdin addition to the vascular compromiseda kindred process leads to airway narrowing called bronchiolitis obliterans. In cardiac allografts, AV involvement of the coronary and intramyocardial arterioles results in gradual ischemic congestive heart failure and/or lethal arrhythmias. Although AV has been called “chronic rejection”, it should not be considered the consequence of a smoldering parenchymal rejection or even an ongoing rejection of the vessel wall; indeed, AV can progress even in the absence of allospecific responses [3]. Rather, AV is probably better understood as a variation on stereotypical healing. Functionally, the repertoire of vessel responses after injury are extremely limited, and vascular repairdregardless of the underlying causedultimately funnels through a final common pathway that conscripts smooth muscle cells and extracellular matrix synthesis to buttress the “damaged” wall. Thus, the vascular wall thickening that accompanies atherosclerosis is a consequence

R.N. Mitchell / Journal of Autoimmunity 45 (2013) 80e89

of a lifetime of accumulated insults ranging from smoking to hypertension to hypercholesterolemia to diabetes. Similarly, the restenosis that occurs within months to years after venous grafting into arterial circulation or endovascular stenting results from the mechanical trauma (with associated thrombosis and inflammation) of the procedure and/or device. In the case of AV, immunemediated vascular injury clearly initiates the process, but it is also worth noting that non-immune pathways (e.g., ischemic injury, free radical damage, hypercholesterolemia, etc.) can contribute. This is by way of justifying that lessons to be gleaned from AV will undoubtedly provide pathogenic insights (and suggest potential therapeutic targets) for the more common vascular lesions that confront cardiologists, rheumatologists, and vascular surgeons. Given that this review appears in Autoimmunity, it may also be germane to point out that AV can actually be thought of as a selfdirected immune response, except that “self” in this case is a transplanted organ. Such AV-associated intimal hyperplasia is also strongly reminiscent of the vascular changes that occur in the setting of scleroderma, where the fundamental target of autoimmune injury is the vessel wall [4]. Similar intimal lesions are also seen in vasculitis, where the endothelium and media are subjected to the manifold effects of inflammatory mediators secondarily recruited in the wake of immune complex deposition. Interestingly, graft-versus-host disease after mismatched bone marrow transplantation can also manifest as a vasculopathy that entirely mimics AV [5,6]. The following will include an overview of the current thinking regarding the mechanisms that contribute to AV, highlighting the important contributions of innate and adaptive immunity, and emphasizing the role of interferon-g (IFNg). Additional insights from animal models and human disease will then be summarized as they pertain to vascular remodeling and aneurysm formation. After encapsulating the experiments that show the various sources

81

of intimal smooth muscle cells, the review will conclude with a discussion of the mechanisms by which intimal smooth muscle cells may be recruited to sites of vascular injury. Throughout, it is worth comparing AV and more common lesions like atherosclerosis (Fig. 1). As noted, they likely share pathogenic mechanisms. Moreover, the relatively accelerated onset of AV makes it more amenable to time-course analysis, and because therapeutic endpoints also occur earlier, AV may well represent a surrogate disease for developing effective interventions.

2. Overview: AV characteristics and pathogenesis [7] Although allograft veins can develop AV lesions, the most clinically relevant effects are all in the arterial circulation. Thus, even though all vessels should be subject to similar immune injury, an arterial predilection probably reflects the consequences of higher shear stresses. It may also be a consequence of the greater volume of (or functionally different) medial smooth muscle cells. Regardless, it is noteworthy that other vascular pathologies, ranging from atherosclerosis to vasculitis, also preferentially affect arteries over veins. Atherosclerosis characteristically involves only discrete areas along the artery and is not uniformly distributed; these lesions are also eccentric (the vessel wall is not uniformly concentrically affected), composed of varying proportions of a grummous, necrotic atheromatous core (plus or minuscalcification) and an overlying fibrous cap, composed of smooth muscle cells (SMC) and extracellular matrix (ECM). The underlying media is typically only secondarily affected, largely as a consequence of increased diffusion distance from the lumen, and mechanical atrophy, and the adventitia is usually unchanged. Mononuclear inflammation in atherosclerotic lesions likely drives the disease pathogenesis and is an

Fig. 1. Atherosclerosis versus allograft vasculopathy (AV). Atherosclerosis (left) is driven by a host of vascular insults including hypertension, diabetes, hypercholesterolemia, cigarette smoking, and inflammation. It tends to occur as focal, eccentric lesions with a central atheromatous core including calcification, cholesterol clefts, and necrotic debris, overlying a fibrous connective tissue cap with scattered smooth muscle cells. Atherosclerosis typically requires several years to become clinically significant, although acute plaque rupture with superimposed thrombosis can precipitate abrupt vascular occlusion. AV (right) is initiated by an alloresponse, and is characterized by a concentric intimal hyperplasia composed predominantly of smooth muscle cells and their associated extracellular matrix; it can diffusely involve the entire arterial tree within a transplanted organ, extending from the epicardial vessels into penetrating intramyocardial arterioles. Onset is typically within a few months after transplantation, and becomes clinically significant in 50% of patients within 5 years. Inflammatory infiltrates composed of T lymphocytes, macrophages, and NK cells (visible as blue cellular infiltrates) can be variably present in both, although they tend to be more extensive and diffusely distributed in AV, while more sparse and centered on plaque shoulders in typical atherosclerosis. Intimal hyperplasia accompanying venous grafting into arterial circulations, in-stent restenosis, and the changes associated with chronic vasculitis all show concentric intimal lesions similar to AV, although the initiating mechanisms of injury are presumably different. SMC, smooth muscle cells. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

82

R.N. Mitchell / Journal of Autoimmunity 45 (2013) 80e89

important component of so-called vulnerable plaques, i.e., those that are prone to rupture and thrombosis (Fig. 1). In comparison (Fig. 1), AV diffusely involves the length of affected vessels (in the heart, this extends from coronary ostium into intra-myocardial arterioles), and is classically a concentric intimal-based expansion of SMC and ECM. Although medial SMC may undergo rejection-mediated apoptosis [8], the media is usually relatively spared. Likewise, adventitia is often unaffected, or only occasionally shows adventitial scarringdpresumably related to prior perivascular inflammation [9]. Calcification and atheromatous cores are uncommon until very late after transplantation or unless atherosclerotic plaque was already present in the transplanted organ. T and B lymphocytes, natural killer (NK) cells, and macrophages, as well as platelets, neutrophils, and eosinophils are all variably present in evolving lesions. Endothelial cells (EC) are activated in association with a sub-endothelial accumulation of mononuclear inflammation at early stages of disease [10,11]. In broad strokes, atherosclerosis results from a host of endothelial insults, priming the vessel wall for subsequent inflammatory cell recruitment and activation (including elements of adaptive immunity). The atherosclerotic plaque then evolves as a consequence of intimal-based SMC and ECM “healing” responses (Fig. 2A) [12,13]. The focality of these lesions reflects the contribution of altered vascular shear forces in atherogenesis, while the typically slow tempo of atherosclerosis development speaks to a relatively low (but nevertheless persistent) level of vascular wall injury and activation.

In an analogous fashion (Fig. 2B), AV pathogenesis begins with a non-specific perioperative trauma and ischemic injury, but then is critically driven by allospecific endothelial and vascular wall immune-mediated injury. Subsequent macrophage recruitment and activation, in conjunction with activated, dysfunctional EC and SMC, leads to a cytokine and chemokine milieu that attracts and activates SMC, stimulating proliferation and ECM synthesis [7,14]. 2.1. Non-immune graft injury The non-immune contributions to AV include vasopressors used to support donor blood pressure prior to harvest, or catecholamines released after closed head injury, both of which increase vascular tone and cause allograft ischemia. The duration of ischemia after harvest, and any subsequent reperfusion injury will also materially impact vascular integrity and graft function. Such non-immune injury chiefly impacts AV pathogenesis by modulating adaptive immune responses [15]; indeed, experimental models show that alloresponses supercede other forms of vessel wall injury [11,16] Such effects are partly attributable to tissue necrosis, providing “danger signals” that trigger more robust host immune responses [17]. However, ischemic EC are also pro-thrombotic, promoting coagulation cascade and platelet activation, both of which can stimulate multiple cell types. Moreover, ischemia and reperfusion can activate complement non-specifically [18]. Ischemic EC dysfunction also leads to augmented major histocompatibility complex (MHC) and adhesion molecule expression, which will increase innate and adaptive effector cell recruitment and activation [19e21]. Other non-immune contributions (e.g., diabetes, hypertension, and hyperlipidemia) are unfortunate side-effects of chronic immunosuppression (especially steroids and calcineurin inhibitors). In the same way that these risk factors drive atherogenesis, they can also contribute to vascular dysfunction and subsequent response to injury. 2.2. Immune graft injury

Fig. 2. General models of pathogenesis in atherosclerosis versus allograft vasculopathy (AV). The schematics depict three comparable stages in the pathogenesis of atherosclerosis (A) and AV (B); each stage involves the elaboration of specific cytokines, chemokines, and adhesion molecules that recruit and activate inflammatory cell populations. The acute vascular injury in atherosclerosis can occur through a variety of insults, whereas the earliest injury in AV is likely ischemic. In atherosclerosis, the initial recruited elements involve macrophages and platelets, whereas in AV, neutrophils, macrophages, and innate humoral mediators will be most significant. In both atherosclerosis and AV, the early cellular recruitment and activation sets the stage for subsequent adaptive immune responses. In atherosclerosis, subsequent T cell and macrophage elements (e.g., responding to modified lipids or denatured host proteins) drive a localized chronic inflammatory state. In AV, macrophages and T lymphocytes in the ensuing alloresponse elaborate additional cytokines; combined with antibodymediated injury, these bring about the subsequent diffuse vascular wall healing responses. In both vascular lesions, interferon-g is a central mediator. As discussed in the text, macrophages and smooth muscle cells may autonomously produce interferon-g, even after an initial adaptive response has subsided. The vascular wall cytokine milieu then generates an adhesion molecule and chemokine environment that can recruit and activate smooth muscle cells (some derived from circulating precursors), resulting in proliferation and extracellular matrix production. The eccentric (atherosclerosis) versus concentric (AV) nature of the resulting lesions is likely a consequence of flow perturbations significantly contributing to the location of atherosclerotic plaques, while immune injury will be more diffuse. oxLDL, oxidized low density lipoproteins; HTN, hypertension; DM, diabetes mellitus.

Alloimmune injury obviously plays a dominant role in AV pathogenesis since host vasculature is not affected in transplant recipients. Both cellular and humoral responses can participate [22,23], and indeeddunlike some of the animal models of AVdit is doubtful that any one cell type (or form of alloresponse) is uniquely responsible for human AV pathogenesis. While allograft vascular injury probably results from multiple different insults, delayedtype hypersensitivity responses to MHC driven by CD4þ T cells probably represent the major initial pathway. In that setting, macrophages are likely the predominant effector cells, with eicosanoids, reactive oxygen species, and degradative proteases being important sources of vascular wall injury. In addition, macrophage activation can drive thrombosis, while dysregulated nitric oxide synthesis is also associated with AV [8,24]. Finally, macrophages (as well as the alloreactive T cells) produce a plethora of cytokines that can influence SMC accumulation as well as ECM synthesis. 2.2.1. Cell-mediated injury in AV Host T cell activation requires receptor recognition of donor MHC, plus ligation of additional co-stimulatory molecules. Both signals are typically provided by specialized antigen presenting cells (APC)dmost commonly dendritic cells (but also including macrophages). Involvement of “non-professional” APC such as EC, often without the same level of co-stimulation, potentially leads to incomplete immune responses and even tolerance. AV lesions develop in animal models where T cell responses are constrained to direct allorecognition (MHC presented on donor

R.N. Mitchell / Journal of Autoimmunity 45 (2013) 80e89

APC) or when indirect presentation (processed MHC presented on host APC) is not possible [25]. Nevertheless, AV incidence correlates with the frequency of CD4þ T cells with indirect allospecificity [26], and substantial evidence now points to indirect allorecognition as the major pathway in AV pathogenesis [27]. Thus, even as T cells with direct anti-donor specificity fade with timedowing to gradual graft DC lossdAV pathogenesis can still proceed. 2.2.2. Co-stimulation in AV Besides T cell receptor-MHC ligation, full T cell activation requires second signals provided by interactions between CD40 and/ or CD80/CD86 on APC, with CD154 and CD28 on T cells, respectively. Interruption of these interactions can lead to ineffective T cell activation; such blockade forms the basis for several models of tolerance. Interestingly, although CD40-CD154 blockade induced specific allograft tolerance, the absence of this second signal did not prevent AV development [28,29]. Apparently, the initial level of activation induced by T cell receptor ligation is sufficient to recruit secondary effectors that eventually result in AV. In comparison, CTLA-4 blockade of CD28 signaling reduced both parenchymal rejection and AV [30]. Moreover, CD80 blockade (using either congenitally-deficient hosts or monoclonal antibodies) inhibited AV development, although CD86 blockade did not. Interference with CD80 did not affect cytokine profiles or intragraft infiltration of inflammatory cells, but presumably interfered with indirect activation of alloreactive T cells by host APC [31,32]. The inducible co-stimulator (ICOS, CD278) molecule on T cells can also impact AV development by ligating ICOS-ligand (CD275) on activated SMC. Since ICOSeICOS-ligand binding leads to SMC proliferation, it is not unsurprising that interruption of that interaction can attenuate AV [33]. Since CD275 is present not only on SMC but also on APC, it is also not clear whether the beneficial effects might also involve dysregulation of T cell co-stimulation. Programmed death protein-1 (PD-1; CD279) on T cells and its ligands PD-L1 (CD274) and PD-L2 ligand (CD273) represent a negative T cell co-stimulatory pathway. Absence of PD-L1 on donor hearts resulted in an augmented parenchymal rejection and vasculopathy with an accelerated course [34]. It is not clear that this co-stimulatory pathway is specifically associated with AV generation, or whether the results simply reflect a more aggressive allograft rejection overall. 2.2.3. Antibody-mediated injury in AV AV can be induced by direct administration of allospecific antibodies in animal models [35,36], and it seems likely that alloantibodies will also contribute to human AV pathogenesis [22,23]. Notably, some 20% of human transplant recipients exhibit measurable anti-MHC antibody titers [37]. EC-bound antibodies activate complement and lead to cytolysis in addition to producing vasoactive and pro-inflammatory complement fragments. Bound antibody can also drive von Willebrand factor and P-selectin expression, culminating in platelet and leukocyte recruitment [38]. Macrophage accumulation also occurs through the production of chemokines (e.g., monocyte chemoattractant protein-1) that are released by EC after antibody binding to MHC. Finally, virtually all of the activated inflammatory cells in AV lesions will also express Fc-receptors that can interact with bound antibody and trigger additional responses. Despite this theoretical foundation, the contribution of antibody-mediated pathways to AV pathogenesis remains unsettled [22,39]. Certainly, the presence of donor-specific antibodies (DSA) is associated with increased graft failure, and anti-MHC class I antibodies portend a worse outcome than MHC class II-directed antibodies; moreover, DSA correlate with increased rates of AV

83

[40]. However, proving causality has been confounded by the lack of reliable and broadly accepted criteria for humoral rejection diagnosis. Histologic criteria for humoral rejection (swollen EC, dilated vessels, interstitial edema, and adherent inflammatory cells) are neither sensitive nor specific [22,41], and immunoglobulin (not necessarily allospecific) can be deposited on EC after a variety of insults (e.g., ischemia or immune complexes from monoclonal antibody immunotherapy). Immunohistochemical staining for complement fragment C4dda slowly catabolized product of complement activationdhas emerged as a reasonably sensitive marker to start to address these issues. Coupled with DSA, C4d staining is perhaps the best criterion for objective documentation of humoral rejection [41]. Nevertheless, only half of transplants with C4d staining show any functional graft compromise [41], and AV occurs in only a third of long-term patients who develop C4d deposition. Notably, AV also occurs in a significant fraction of transplants without DSA or C4d staining [40]. The distinction between humoral and cellular causes of AV has therapeutic significance because most maintenance immunosuppression (e.g., calcineurin inhibitors and steroids) targets T cells and macrophages. In comparison, agents that would suppress antibodydriven pathways (plasmapheresis, intravenous immunoglobulin, and anti-CD20) are not standard long-term therapies. While some agents can potentially impact both cellular and humoral rejection (e.g., antimetabolites), tailored therapy would obviously be preferred. 3. Role of cytokines in vascular disease Cytokines are central mediators in AV pathogenesis. Although innate and adaptive immune cells are traditionally considered the major producers, EC, SMC, and epithelium are also potent sources in the intragraft milieu [42,43]. Interferong (IFNg), produced primarily by infiltrating T-helper 1 cells (but also to a lesser extent by NK cells, CD8þ T cells, and dendritic cells), is the most important. Among other activities, it increases MHC I and II expression, potently drives macrophage activation, and sets the stage for SMC and inflammatory cell recruitment (Table 1) [15]. That IFNg also dysregulates local EC nitric oxide synthase activity has been suggested to contribute to AV pathogenesis [24]. Cytokines subsequently produced by activated macrophages (e.g., interleukin-1 and tumor necrosis factor-a) can also contribute by modulating inflammatory responses, promoting coagulation and inflammatory cell recruitment, and altering SMC proliferation and ECM synthesis. Cytokines such as transforming growth factorb IFNg also underlie allograft and vascular fibrosis. Table 1 Roles of Interferon-g (IFNg) in allograft vasculopathy 

Enhance MHC and co-stimulator expression - Augmented alloresponses



Endothelial cell and smooth muscle cell activation - Adhesion molecule expression - Chemokine elaboration



Macrophage activation - Cytokine induction (interleukin-1, tumor necrosis factor-a) - Chemokine production - Mediator expression (e.g., nitric oxide)



Intimal smooth muscle cell precursor recruitment

84

R.N. Mitchell / Journal of Autoimmunity 45 (2013) 80e89

3.1. The role of IFNg in AV Experimental animal work clearly shows that IFNg is both necessary and sufficient to drive AV. Thus, IFNg blockade abrogates AV even in the setting of severe cellular rejection, and administration of IFNg to vascular grafts induces AV without involving host inflammation [44e46]. IFNg accomplishes this through many pathways, but effects on macrophages are likely the most important. Once activated, macrophages synthesize IL-12 (and IL-18) that stimulate both CD4þ and CD8þ T cells, forming a positive feed-back loop; this “IFNg-IL12 axis” is proposed as a critical element in AV pathogenesis [15]. Significantly, activated macrophages can also produce IFNg thereby forming an autocrine loop independent of T cell alloresponses [47]. Importantly, SMC can also synthesize IFNg after IL-12 and IL-18 stimulation [48]; indeed, the ability of macrophages and SMC to synthesize IFNg in an autocrine feed-back cycle may explain how AV can develop in the absence of ongoing allospecific responses [3]. IFNg also induces the secretion of a number of chemokines that recruit T cells, representing another amplification loop to increase the local inflammatory cell population. Thus, Mig (monokine induced by IFNg), IP10 (IFN-inducible protein 10), and I-TAC (IFNinducible T-cell a-chemoattractant) recruit memory T cells expressing IFNg-inducible CXCR3 receptors. Similarly, RANTES (regulated on activation normal T cell expressed and secreted), and fractalkine recruit both T cells and monocyte/macrophages. These chemokines and their ligands all not only influenced by IFNg expression, but have also been linked to AV pathogenesis in humans [49,50]. Given the plethora of IFNg-dependent contributions, the success of IFNg blockade is probably not at a single locus, but rather through a combination of effects, including inhibiting SMC proliferation in the intima [7]. 3.2. TNFa in AV While IFNg is both necessary and sufficient for AV pathogenesis, other cytokines induced by IFNg also likely participate. For example, TNFa can modulate the expression of chemokines, chemokine receptors, and adhesion molecules; TNF also boosts IL-12 production that can increase IFNg production. Consequently, in animal models where allografts were deficient in both forms of the TNF receptor, AV failed to develop despite normal IFNg production [51]. The effects of TNF blockade were not attributable to inhibition of host inflammatory cells, but rather were proposed to occur through interference with SMC-macrophage activation loops. To bring the discussion back to other forms of vascular disease, it is worth mentioning that inflammatory cytokines are increasingly viewed not only as important players in the pathogenesis of atherosclerosis [52,53], but also as potential therapeutic targets [12,54]. In this regard, it is significant that IFNg [55] and TNF [56] have both been identified as centrally important in atherogenesis; in addition, both regulate the expression of chemokines, chemokine receptors, and adhesion molecules that can influence the recruitment and activation of smooth muscle cells in atherosclerotic plaque [57]. 4. Vascular remodeling in vascular disease: stenosis or aneurysm? Luminal stenosis is the major clinical consequence of AV, leading to allograft ischemia [7,9]]. Such negative (inward) remodeling is chiefly attributable to intimal ECM and cell accumulation, but will also be influenced by vascular scarring and increased vasomotor tone. Conversely, positive (outward) remodeling can result from SMC

and ECM turnover [58], maintaining a normal lumenal diameter even as the intima expands [59]. On the other hand, excessive matrix degradation or SMC loss can culminate in aneurysm formation and even rupture. 4.1. Negative remodeling In transplanted organsdand indeed, in any vascular injurydmedial scarring and decreased matrix turn-over (e.g., due to increased cystatin C or tissue inhibitors of metalloproteinase) will limit outward vascular remodeling [9,58]. Similarly, perivascular inflammation and the resulting adventitial scarring will engender a thick rind of connective tissue that markedly reduces vascular compliance. Medial SMC tonedinfluenced by vasodilators such as nitric oxide, and vasoconstrictors such as endothelindalso impact negative remodeling. Chronic mediator production can alter vascular tone and translate into permanent structural lesions through effects on wall shear stress. Intravascular ultrasound shows that the negative remodeling seen over the first year after transplantation is attributable to adventitial scarring, while intimal hyperplasia drives the bulk of late-stage luminal stenosis [60] 4.2. Positive remodeling Obviously, SMC loss (due to apoptosis) or medial ECM reduction (due to increased degradation) in the absence of vessel scarring will lead to vasodilation [39]. Immune-mediated cytotoxic injury to medial SMC, and/or increased production of lysosomal cysteine proteases and matrix metalloproteinases are associated with positive remodeling [61]. Vasodilation can also be driven through increased nitric oxide or diminished endothelin production. 4.3. Stenosis versus aneurysm The balance between positive and negative remodeling in vascular injury has clear-cut clinical ramifications. Yet, understanding how these are differentially regulated and predicting which will prevail is not obvious. Thus, clinically-significant carotid and coronary artery atherosclerosis lesions are almost uniformly stenosing. In comparison, while infrarenal aortic atherosclerosis (present in virtually all humans over the age of fifty) results primarily in a stenosing lesion, roughly 3% of the population will manifest with abdominal aortic aneurysms (AAA)dand some 15,000 US patients each year will die due to rupture. Lessons learned from animal transplant models have cast a new light on these pathogenic questions. While looking at the impact of various cytokines on AV development in an aortic transplant model, Shimizu, et al. demonstrated that although IFNg-dominated inflammation led to intimal hyperplasia and vascular stenosis, interleukin 4 (IL-4)-enriched inflammation led to the development of aneurysmal disease [62]. This occurred as a consequence of alterations in the matrix metalloproteinase (MMP) repertoire of the graft-infiltrating macrophages. Thus, IL-4 induced the macrophages to synthesize markedly elevated levels of MMP-12, the murine enzyme with the greatest elastase activity. Subsequent elastin tissue degradation led to loss of vascular wall integrity and aneurysm formation [62]. Similar associations have been demonstrated in human AAA. In typical stenosing atheromatous disease, IFNg and IFNg-associated cytokines and effectors predominate. However, in AAA, IL-4 and IL4-associated mediators are markedly elevated [63]. The results suggest that although AAA may be initiated through Th1-driven vascular wall inflammation, in the subset of individuals who eventually develop aneurysms, secondary events or genetic predilections may induce a predominant Th2 milieu [63].

R.N. Mitchell / Journal of Autoimmunity 45 (2013) 80e89

The more general implication from both the experimental and clinical literature is that vascular injury can induce both positive and negative remodeling of vessel walls depending on the local cytokine environment. IFNg-dominated lesions will tend more to negative remodeling with pathologic luminal stenosis; IL-4-driven inflammation will drive positive remodeling that can also become deleterious [9] (Fig. 3).

5. Source of smooth muscle cells in vascular disease Although long assumed to derive from the recruitment of medial SMC, the source of intimal SMC in vascular lesions is actually a bit more involved. Our understanding of these pathways owes much to the transplant literature and the understanding of AV pathogenesis. It had been known for almost two decades that intimal SMCdcomprising the bulk of the cellularity in atherosclerosis and other stenosing vascular pathologiesdare phenotypically dissimilar from medial SMC, having substantially greater proliferative and synthetic capacities [64e66]. Despite the manifold differences, the phenotypic changes had been postulated to result from a reprogramming that occurred when SMC are recruited into areas of intimal inflammation and away from the normal modulatory environment of the media. For example, relative variation in pericellular laminin (inhibitory) and fibronectin (stimulatory) levels can impact SMC proliferative capacity [67]. By their ability to mark and track host and donor cells unambiguously, transplantation researchers showed that intimal SMC can derive from the host and not the donor vessel media. Indeed, in mouse models, the vast majority of the intimal SMC are of host origin (summarized in Refs. [68,69]); significantly, bone marrow precursors accounted for as much as 15e20% of the total [70,71]. In short order, investigators showed that host cells also contributed to the intimal SMC in human heart and kidney AV [72,73]. Moreover, SMC progenitors could be demonstrated in human peripheral blood [74,75]. This turns out to be not just a curiosity associated with transplanted vessels. Thus, in gender-mismatched bone marrow transplant recipients, almost 15% of the SMC in subsequent coronary atherosclerotic lesions also derive from donor marrow [76],

Fig. 3. Vascular remodeling in stenosing versus aneurysmal disease. Intimal hyperplasiadand the resulting luminal stenosisdoccurs through recruitment of inflammatory cells and smooth muscle cells (SMC), with subsequent elaboration of extracellular matrix (ECM). Interferon-g, elaborated for example by a Th1-dominated response, is the most important cytokine in this process. Conversely, interleukin-4, as produced in a Th2-dominated environment, will lead to the production of matrix metalloproteinases that will drive aneurysm formation. Increases in adventitial fibrosis will also limit positive vessel remodeling. In addition, alterations in medial tone (secondary to vasodilator or vasoconstrictor production) can influence luminal diameter; intimal hyperplastic responses will be augmented where there is increased wall shear stress (e.g., with vasoconstriction).

85

suggesting that circulating precursors are a regular component in the vascular wall healing response. It is worth highlighting a couple caveats of these studies. The first is that the relative percentage of host SMC in intimal lesions is dependent on the severity of vascular injury [68,69]. Thus, in animal transplant modelsdwhere immunosuppression is usually minimal and donor vessels can sustain moderate-to-severe damagedthe frequency of host origin SMC routinely exceeds 90% [68]. Under these conditions, circulating host precursors are likely to me relatively more numerous since they will not be subject to immune destruction [68,69]. Conversely, in human transplantation, considerably greater immunosuppression will lead to less vascular injury, and host cells will represent a smaller percentage of the intimal cells. Indeed, in gender-mismatched cardiac allografts, Ychromosome in situ hybridization showed less than 15% hostderived intimal SMC [72,77]. A second caveat is that bone marrow-derived precursors account for only a minority of the intimal SMC. In most situations, SMCderived from adventitial myofibroblasts may contribute up to a third of the intimal cellularity, anddin the setting of adequate immunosuppressionddonor SMC may in fact dedifferentiate to the proliferative and synthetic embryonic phenotype [8,78]. What nevertheless emerges from the transplant literature is that intimal SMC derive from multiple sources, and not simply by recruitment from the underlying media; both marrow and nonmarrow precursors also participate. Thus, AV therapies and therapies targeting atherosclerosis and restenosis lesions (e.g., after stenting) will need to account for distinct cellular origins, and routes of recruitment and differentiation. 6. Recruitment SMC into intimal lesions A corollary to the observation that substantial numbers of SMC can derive from the circulation is that access to vessel walls almost certainly involves recruitment pathways akin to those used by inflammatory cells. The route of access is of more than just academic interest since these pathways become potential targets of therapy. Formal evidence exists for cellular entry either from the vessel lumen or from the adventitial face [79]. In venous bypass grafts for example, EC denudation or dysfunction is probably an important proximal step allowing luminal attachment of SMC precursors via bound platelets. Platelets can adhere even at arterial shear forces (which may otherwise limit cellular attachment), [80]; relevant to AV pathogenesis, platelet activation has been shown to correlate with development of the lesions [81]. If luminal SMC use such pathways, thrombosis and platelet adhesion become relevant clinical targets [82]. Alternatively, SMC recruitment in high-pressure arterial circulations with intact endothelium may require vessel access through ablumenal pathways. This squares with observations that mononuclear inflammatory cells coalesce around arteries prior to AV onset [44]. Moreover, shear stress (20e80 dyn/cm2) and pulsatile shear gradients in arterial beds far exceed the ability of mononuclear inflammatory cells to adhere [83]. Thus, accessing vessel intima from vaso vasorum or from post-capillary venules in the surrounding parenchyma becomes a more tractable approach. Recruitment via these pathways very likely involves SMC-EC adhesion pairs and chemokineereceptor interactions analogous to those used by inflammatory cells. 6.1. Chemokines The role of chemokines in the inflammation underlying parenchymal rejection and AV is reviewed extensively elsewhere [84,85]

86

R.N. Mitchell / Journal of Autoimmunity 45 (2013) 80e89

It is worth noting that besides recruitment, chemokines can also activate cells, thereby promoting their local proliferation and synthetic function. The CC chemokines (e.g., MCP-1 and RANTES) are characteristically involved in recruiting and activating mononuclear inflammatory cells, and have been causally associated in a variety of chronic inflammatory diseases, including atherosclerosis and AV [86,87]. Of the CXC chemokines, the IFNg-inducible IP10, Mig, and I-TAC are noteworthy for their strong association with AV, and their production by EC, macrophages, and/or SMC [49,50]. Given the plethora of chemokines and the promiscuity of chemokineereceptor interactions, it seems unlikely that any individual chemokine or receptor will play a critical non-redundant role. Nevertheless, deficiencies of MCP-1 or CCR2 (one of the receptors than bind MCP-1) can significantly impact monocyte recruitment and reduce the development of atherosclerosis [88,89]. In transplant models, there are a host of reports indicating partial amelioration of rejection and/or AV through blockade of various chemokine interactions [85,90,91]. 6.2. Chemokine recruitment of SMC Chemokines are classically associated with inflammatory cell recruitment and activation. However, chemokine receptors are also expressed on several non-hematopoietic lineages, and SMC in particular express functional chemokine receptors. Moreover, although activated lymphocytes and macrophages are important sources of chemokines, EC and SMC can also synthesize RANTES, MCP-1, and MIP-1 [85,92]. Thus, dysfunctional vascular wall cells could conceivably drive SMC recruitment in AV without an ongoing alloresponse. In this regard, it is significant that CCR5 is expressed on normal human aortic and coronary SMC, and that the CCR5 ligand MIP-1b (macrophage inflammatory protein-1) induces tissue factor production [93]. SMC in human atherosclerotic plaques also express CXCR3 [94], and SMC in an AV model upregulate CXCR4 expression [95]. SMC in different vascular lesions (e.g., human and murine atherosclerosis and senile intimal thickening) also variously express CCR2, CCR3, and CCR5 [96e98]. Overall, the findings suggest intimal and medial SMC may have distinct chemokine responsiveness [9,99]. Specifically relevant to transplantation, SMC from AV intimal lesions expressed CCR1 and CCR2 while normal medial SMC did not. In addition, loss of CCR1 virtually abrogated AV without

Fig. 4. Chemokine mediators in the development of allograft vasculopathy (AV). At each of the steps in AV pathogenesis shown in Fig. 2B, potential chemokine effectors are shown, along their sources (in parentheses). Note that chemokines are involved not only in recruiting and activating inflammatory cells, but also in the process of smooth muscle cell recruitment and activation. IL-8, interleukin 8; MIP-2, macrophage inflammatory protein 2 (CXCL2); MCP-1, monocyte chemoattractant protein-1 (CCL2); Mig, monokine induced by interferon-g (CXCL9); IP-10, interferon-g-induced protein10 (CXCL10); RANTES, regulated on activation, normal T cell-expressed (CCL5); EC, endothelial cells; SMC, smooth muscle cells; PMN, neutrophils.

significantly affecting parenchymal inflammation [99]. Taken together, the results suggest that CCR1 blockade could potentially antagonize intimal SMC recruitment. Moreover, because the chemokine pathways involved in inflammatory cell recruitment are redundant, CCR1 blockade may allow AV inhibition without increasing immunosuppression. Fig. 4 summarizes some of the more important chemokines that can participate in the various steps of AV pathogenesis.

6.3. Role of adhesion molecules in AV Besides chemokines, adhesion to EC is a proximate event for inflammatory cell recruitment into allografts and other vascular pathologic lesions. Correspondingly, both parenchymal rejection and AV correlate with EC expression of intracellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), and the E and P-selectins [100,101]. Notably, blockade of certain adhesion molecule interactions (interfering with either the inflammatory cell integrins or their EC cognate ligands) reduces acute rejection [102] and AV [101,103]. The role of selectins and integrins for inflammatory cell adhesion and recruitment is well-documented. Indeed, and as might be anticipated, blockade of these interactions can reduce parenchymal rejection [104], as well as AV [105,106]. It is, however, uncertain (albeit nevertheless likely) that similar molecules participate in SMC recruitment, and although SMC express b1-and b3-integrins [107,108], no evidence yet exists for recruitment through integrinEC interactions. However, other non-integrins may be involved. CD44 is a cell surface glycoprotein widely expressed on a variety of inflammatory and non-inflammatory cells, including medial SMC [109]; it binds hyaluronan, an ECM constituent that is upregulated following ischemia, thus facilitating mononuclear cell recruitment at sites of injury. CD44-hyaluronan interactions mediate SMC migration [110], including at sites of vascular injury [111]. Significantly, mice with congenital absence of CD44 develop less atherosclerosis in an ApoE-/- model [112]. Germane to AV, CD44 expression also correlates with human renal allograft vasculopathy [113].

7. Conclusion Allograft vasculopathy (AV) is an intimal proliferative lesion; the resulting concentric hyperplasia leads to a diffuse vascular stenosis that remains the most important long-term limit on the success of solid-organ transplantation. The primary inciting insult is an alloresponse, although ischemia and other forms of injury are involved. Both innate and adaptive immunity contribute, and a potentially autonomous IFNg axis is a central driving component. Host SMC precursors of various sources contribute to the cellularity of AV lesions, and are likely recruited by chemokine and adhesion pathways comparable to those used by inflammatory cells. AV also provides an experimentally tractable model for exploring other vascular pathologies. Thus, vascular injury of any kind results in a stereotypical healing response culminating in intimal smooth muscle recruitment and activation. Depending on the nature of the inflammatory milieu, lesions may be stenosingddriven by an IFNg-dominated environment, or aneurysmaldinduced by IL-4 or comparable cytokines. Because AV shares many features with more common vascular lesions, insights derived from AV pathogenesis not only impact our understanding of more “conventional” (and quantitatively important) entities like atherosclerosis or restenosis lesions, but also suggest therapeutic interventions.

R.N. Mitchell / Journal of Autoimmunity 45 (2013) 80e89

8. Final comments It is a distinct pleasure and honor to be asked to write a brief reflection on the mentorship and guidance that I receiveddlo these years gone bydworking with Abul Abbas in this special issue of the Journal of Autoimmunity. This issue is devoted entirely to recognizing the contributions of Abul Abbas and part of the Journal’s efforts to recognize unique topics, themes and particularly outstanding autoimmunologists [114e117]. I was a post-doc in Abul’s lab almost a quarter century ago, back when he was ascending the academic ladder at Harvard Med and serving as a Staff Pathologist at Brigham and Women’s Hospital (BWH). For many years before being recruited to UCSF, Abul was the animated and engaging face of immunology at BWH, enlivening the hallways and labs with his ready laugh and incisive commentary. At conferences in the Department, Abul had far too much energy to simply sit in one place; instead he would schmooze and kibbitz (perhaps a bit louder than he should) in the back of the room with whoever was within earshot. On multiple occasions, our exasperated Chair (Ramzi Cotran) would stop the proceedings, turn around, and insist that Abul join him in the front rowdlike a wayward and hyperactive childdwhere he could keep an eye on him. In the lab, Abul was equally kinetic; in an era before e-mail, he was forever on the phone talking with collaborators, colleagues, and far-flung acquaintancesdall of who were his “very good friends”. To burn off some of that energy, Abul would frequently take to the squash courts. Even there, he was working and schmoozing, and some of his best post-docs (Henry Boom, for example) came from chance court encounters. While incredibly generous with his ideasdand his opinionsdAbul could also be occasionally exasperating. He has the rare gift of being able to write amazingly erudite and concise scientific prose in long-hand, and on the first draft, with no cross-outs or insertions. Where the rest of us mere mortals struggle and labordand probably over-writedAbul can compose entire elegant treatises on cross-country plane flights, or in a spare moment on the beach. Abul also wasn’t shy about telling you that an experiment probably wasn’t worth doing, largely because the result was already obviousdat least to him. On more than one occasion, the outcome was several days of protracted effort to prove him wrong (and in retrospect, perhaps he was even smarter than I thought!). Clearly a lot of his observations and snippets of veracity stuck; almost two decades later, I find myself sharing Abul-isms with residents and fellows (e.g., “In vivo veritas”). And, finally, no one is a better teacher. Abul has a talent for taking gray-area concepts and framing them in understandable black-and-whitedwhat we came to refer to as “Abulian Logic”. Although, some of the nuances might be lost in the shuffle, seeing the forest among the trees makes him an incredible educator, and has endeared him to generations of medical students and graduate students (just check out “Abul in a Box” on the internet)!

References [1] Dhaliwal A, Thohan V. Cardiac allograft vasculopathy: the Achilles’ heel of long-term survival after cardiac transplantation. Curr Atheroscler Rep 2006;8:119e30. [2] Stehlik J, Edwards L, Kucheryavaya A, Benden C, Christie J, Dipchand A, et al. The registry of the international society for heart and lung transplantation: 29th official adult heart transplant reportd2012. J Heart Lung Transpl 2012;31:1052e64. [3] Nagano H, Libby P, Taylor M, Hasegawa S, Stinn J, Becker G, et al. Coronary arteriosclerosis after T-cell-mediated injury in transplanted mouse hearts. Role of interferon-g. Am J Pathol 1998;152:1187e97. [4] Pattanaik D, Monica M, Postlethwaite A. Vascular involvement in systemic sclerosis (scleroderma). J Inflamm Res 2011;4:105e25.

87

[5] Prevost D, Taylor G, Sanatani S, Schultz K. Coronary vessel involvement by chronic graft-versus-host disease presenting as sudden cardiac death. Bone Marrow Transplant 2004;34:655e6. [6] Ghobrial I, Bunch T, Caplice N, Edwards W, Miller D, Litzow M. Fatal coronary artery disease after unrelated donor bone marrow transplantation. Mayo Clin Proc 2004;79:403e6. [7] Mitchell R. Graft vascular disease: immune response meets the vessel wall. Annu Rev Pathol 2009;4:19e47. [8] Rahmani M, Cruz R, Granville DJ, Mcmanus B. Allograft vasculopathy versus atherosclerosis. Circ Res 2006;99:801e15. [9] Mitchell R, Libby P. Vascular remodeling in transplant vasculopathy. Circ Res 2007;100:967e78. [10] Salomon R, Hughes C, Schoen F, Payne D, Pober J, Libby P. Human coronary transplantation-associated arteriosclerosis: evidence for a chronic immune reaction to activated graft endothelial cells. Am J Pathol 1991;138:791e8. [11] Furukawa Y, Libby P, Stinn J, Becker G, Mitchell R. Cold ischemia enhances cytokine/cell adhesion molecule expression and induces graft arterial disease in isografts, but does not accentuate alloimmune responses of nonimmunosuppressed hosts. Am J Pathol 2002;160:1077e87. [12] Libby P, Ridker P, Hansson G. Atherothrombosis. Ltno. Inflammation in atherosclerosis: from pathophysiology to practice. J Am Coll Cardiol 2009;54: 2129e38. [13] Hansson G, Hermansson A. The immune system in atherosclerosis. Nat Immunol 2011;12:204e12. [14] Libby P, Pober J. Chronic rejection. Immunity 2001;14:387e97. [15] Tellides G, Pober J. Interferon-gamma axis in graft arteriosclerosis. Circ Res 2007;100:622e32. [16] Ramzy D, Rao V, Brahm J, Miriuka S, Delgado D, Ross H. Cardiac allograft vasculopathy: a review. Can J Surg 2005;48:319e27. [17] Gallucci S, Matzinger P. Danger signals: SOS to the immune system. Curr Opin Immunol 2001;13:114e9. [18] Thurman J, Holers V. The central role of the alternative complement pathway in human disease. J Immunol 2006;176:1305e10. [19] El-Sawy T, Fahmy N, Fairchild R. Chemokines: directing leukocyte infiltration into allografts. Curr Opin Immunol 2002;14:562e8. [20] Valantine H. Cardiac allograft vasculopathy after heart transplantation: risk factors and management. J Heart Lung Transplant 2004;23(5 Suppl.):S187e93. [21] Caforio A, Tona F, Fortina A, Angelini A, Piaserico S, Gambino A, et al. Immune and nonimmune predictors of cardiac allograft vasculopathy onset and severity: multivariate risk factor analysis and role of immunosuppression. Am J Transplant 2004;4:962e70. [22] Michaels P, Espejo M, Kobashigawa J, Alejos J, Burch C, Takemoto S, et al. Humoral rejection in cardiac transplantation: risk factors, hemodynamic consequences and relationship to transplant coronary artery disease. J Heart Lung Transplant 2003;22:58e69. [23] Wehner J, Morrell C, Reynolds T, Rodriguez E, Baldwin W. Antibody and complement in transplant vasculopathy. Circ Res 2007;100:191e203. [24] Koh K, Wang Y, Yi T, Shiao S, Lorber M, Sessa W, et al. T cell-mediated vascular dysfunction of human allografts results from IFN-gamma dysregulation of NO synthase. J Clin Invest 2004;114:846e56. [25] Ardehali A, Fischbein M, Yun J, Irie Y, Fishbein M, Laks H. Indirect alloreactivity and chronic rejection. Transplantation 2002;73:1805e7. [26] Hornick P, Mason P, Baker R, Hernandez-Fuentes M, Frasca L, Lombardi G, et al. Significant frequencies of T cells with indirect anti-donor specificity in heart graft recipients with chronic rejection. Circulation 2000;101:2405e10. [27] Jiang S, Herrera O, Lechler R. New spectrum of allorecognition pathways: implications for graft rejection and transplantation tolerance. Curr Opin Immunol 2004;16:550e7. [28] Shimizu K, Schönbeck U, Mach F, Libby P, Mitchell R. Host CD40 ligand deficiency induces long-term allograft survival and donor-specific tolerance in mouse cardiac transplantation but does not prevent graft arteriosclerosis. J Immunol 2000;165:3506e18. [29] Raisky O, Spriewald B, Morrison K, Ensminger S, Mohieddine T, Obadia J, et al. CD8(þ) T cells induce graft vascular occlusion in a CD40 knockout donor/recipient combination. J Heart Lung Transplant 2003;22:177e83. [30] Glysing-Jensen T, Raisanen-Sokolowski A, Sayegh M, Russell M. Chronic blockade of CD28-B7-mediated T-cell costimulation by CTLA4Ig reduces intimal thickening in MHC class I and II incompatible mouse heart allografts. Transplantation 1997;64:1641e5. [31] Furukawa Y, Mandelbrot D, Libby P, Sharpe A, Mitchell R. Association of B7-1 co-stimulation with the development of graft arterial disease: studies using mice lacking B7-1, B7-2, or B7-1/B7-2. Am J Pathol 2000;157:473e84. [32] Kim K, Denton M, Chandraker A, Knoflach A, Milord R, Waaga A, et al. CD28B7-mediated T cell costimulation in chronic cardiac allograft rejection: differential role of B7-1 in initiation versus progression of graft arteriosclerosis. Am J Pathol 2001;158:977e86. [33] Kosuge H, Suzuki J, Haraguchi G, Koga N, Maejima Y, Inobe M, et al. Critical role of inducible costimulator signaling in the development of arteriosclerosis. Arterioscl Thromb Vasc Biol 2006;26:2660e5. [34] Yang J, Popoola J, Khandwala S, Vadivel N, Vanguri V, Yuan X, et al. Critical role of donor tissue expression of programmed death ligand-1 in regulating cardiac allograft rejection and vasculopathy. Circulation 2008;117:660e9. [35] Russell P, Chase C, Winn H, Colvin R. Coronary atherosclerosis in transplanted mouse hearts. II. Importance of humoral immunity. J Immunol 1994;152:5135e41.

88

R.N. Mitchell / Journal of Autoimmunity 45 (2013) 80e89

[36] Hancock W, Buelow R, Sayegh M, Turka L. Antibody-induced transplant arteriosclerosis is prevented by graft expression of anti-oxidant and antiapoptotic genes. Nat Med 1998;4:1392e6. [37] Terasaki P, Ozawa M. Predicting kidney graft failure by HLA antibodies: a prospective trial. Am J Transplant 2004;4:438e43. [38] Yamakuchi M, Kirkiles-Smith N, Ferlito M, Cameron S, Bao C, Fox-Talbot K, et al. Antibody to human leukocyte antigen triggers endothelial exocytosis. Proc Natl Acad Sci U S A 2007;104:1301e6. [39] Thaunat O, Louedec L, Dai J, Bellier F, Groyer E, Delignat S, et al. Direct and indirect effects of alloantibodies link neointimal and medial remodeling in graft arteriosclerosis. Arterioscler Thromb Vasc Biol 2006;26:2359e65. [40] Frank R, Molina M, Wald J, Goldberg L, Kamoun M, Lal P. Correlation of circulating donor-specific anti-HLA antibodies and presence of C4d in endomyocardial biopsy with heart allograft outcomes: a single-center, retrospective study. J Heart Lung Transpl 2013;32:410e7. [41] Michaels P, Fishbein M, Colvin R. Humoral rejection of human organ transplants. Springer Semin Immunopathol 2003;25:119e40. [42] Rossi D, Zlotnick A. The biology of chemokines and their receptors. Ann Rev Immunol 2000;18:217e42. [43] Zlotnik A, Yoshie O. The chemokine superfamily revisited. Immunity 2012;36:705e16. [44] Nagano H, Mitchell R, Taylor M, Hasegawa S, Tilney N, Libby P. Interferon-g deficiency prevents coronary arteriosclerosis but not myocardial rejection in transplanted mouse hearts. J Clin Invest 1997;100:550e7. [45] Tellides G, Tereb D, Kirkiles-Smith N, Kim R, Wilson J, Schechner J, et al. Interferon-gamma elicits arteriosclerosis in the absence of leukocytes. Nature 2000;403:207e11. [46] Wang Y, Burns W, Tang P, Yi T, Schechner J, Zerwes H, et al. Interferon-gamma plays a nonredundant role in mediating T cell-dependent outward vascular remodeling of allogeneic human coronary arteries. FASEB J 2004;18:606e8. [47] Munder M, Mallo M, Eichmann K, Modolell M. Murine macrophages secrete interferon-g upon combined stimulation with interleukin (IL)-12 and IL-18: a novel pathway of autocrine macrophage activation. J Exp Med 1998;187: 2103e8. [48] Gerdes N, Sukhova G, Libby P, Reynolds R, Young J, Schönbeck U. Expression of interleukin 18 and functional IL-18 receptor on vascular endothelial cells, smooth muscle cells, and macrophages: implication for atherogenesis. J Exp Med 2002;195:245e57. [49] Zhao D, Hu Y, Miller G, Luster A, Mitchell R, Libby P. Differential expression of the IFN-gamma-inducible CXCR3-binding chemokines, IFN-inducible protein 10, monokine induced by IFN, and IFN-inducible T cell alpha chemoattractant in human cardiac allografts: association with cardiac allograft vasculopathy and acute rejection. J Immunol 2002;169:1556e60. [50] Vanloosdregt J, Vanoosterhout M, Bruggink A, Vanwichen D, Vankuik J, Dekoning E, et al. The chemokine and chemokine receptor profile of infiltrating cells in the wall of arteries with cardiac allograft vasculopathy is indicative of a memory T-helper 1 response. Circulation 2006;114:1599e607. [51] Suzuki J, Cole S, Batirel S, Kosuge H, Shimizu K, Isobe M, et al. Tumor necrosis factor receptor -1 and -2 double deficiency reduces graft arterial disease in murine cardiac allografts. Am J Transplant 2003;3:968e76. [52] Andersson J, Libby P, Hansson G. Adaptive immunity and atherosclerosis. Clin Immunol 2010;134:33e46. [53] Libby P. Inflammation in atherosclerosis. Nature 2002;420:868e74. [54] Libby P, Okamoto Y, Rocha V, Folco E. Inflammation in atherosclerosis: transition from theory to practice. Circ J 2010;74:213e20. [55] Mclaren J, Ramji D. Interferon gamma: a master regulator of atherosclerosis. Cytokine Growth Factor Rev 2009;20:125e35. [56] Kleinbongard P, Heusch G, Schulz R. TNFalpha in atherosclerosis, myocardial ischemia/reperfusion and heart failure. Pharmacol Ther 2010;127:295e314. [57] Plutzky J. Inflammatory pathways in atherosclerosis and acute coronary syndromes. Am J Cardiol 2001;88:10Ke5K. [58] Cummins P, Vonoffenberg-Sweeney N, Killeen M, Birney Y, Redmond E, Cahill P. Cyclic strain-mediated matrix metralloproteinase regulation within the vascular endothelium: a force to be reckoned with. Am J Physiol Heart Circ Physiol 2007;292:H28e42. [59] Glagov S, Weisenberg E, Zarins C, Stankunavicius R, Kolettis G. Compensatory enlargement of human atherosclerotic coronary arteries. N Eng J Med 1987;316:1371e5. [60] Wong C, Ganz P, Miller L, Kobashigawa J, Schwarzkopf A, Valantine H, et al. Role of vascular remodeling in the pathogenesis of early transplant coronary artery disease: a multicenter prospective intravascular ultrasound study. J Heart Lung Transpl 2001;20:385e92. [61] Liu J, Sukhova G, Sun J, Xu W, Libby P, Shi G. Lysosomal cysteine proteases in atherosclerosis. Arterioscler Thromb Vasc Biol 2004;24:1359e66. [62] Shimizu K, Shichiri M, Libby P, Lee R, Mitchell R. Th2-predominant inflammation and blockade of IFN-gamma signaling induce aneurysms in allografted aortas. J Clin Invest 2004;114:300e8. [63] Schonbeck U, Sukhova G, Gerdes N, Libby P. T(H)2 predominant immune responses prevail in human abdominal aortic aneurysm. Am J Pathol 2002;161:499e506. [64] Owens G. Regulation of differentiation of vascular smooth muscle cells. Physiol Rev 1995;75:487e517. [65] Nagai R, Suzuki T, Aizawa K, Miyamoto S, Amaki T, Kawai-Kowase K, et al. Phenotypic modulation of vascular smooth muscle cells: dissection of transcriptional regulatory mechanisms. Ann NY Acad Sci 2001;947:56e66.

[66] Shah R, Mitchell R. The role of stem cells in the response to myocardial and vascular wall injury. Cardiovasc Pathol 2005:225e31. [67] Hedin U, Roy J, Tran P. Control of smooth muscle cell proliferation in vascular disease. Curr Opin Lipidol 2004;15:559e65. [68] Shimizu K, Mitchell R. Stem cell origins of intimal cells in graft arterial disease. Curr Athero Rep 2003;5:230e7. [69] Hillebrands J, Klatter F, Rozing J. Origin of vascular smooth muscle cells and the role of circulating stem cells in transplant arteriosclerosis. Arterioscl Thromb Vasc Biol 2003;23:380e7. [70] Hillebrands J, Klatter F, Hurk Bvd, Popa E, Nieuwenhuis P, Rozing J. Origin of neointimal endothelium and alpha-actin-positive smooth muscle cells in transplant arteriosclerosis. J Clin Invest 2001;107:1411e22. [71] Shimizu K, Sugiyama S, Aikawa M, Fukumoto Y, Rabkin E, Libby P, et al. Host bone-marrow cells are a source of donor intimal smooth muscle-like cells in murine aortic transplant arteriopathy. Nat Med 2001;7:738e41. [72] Glaser R, Lu M, Narula N, Epstein J. Smooth muscle cells, but not myocytes, of host origin in transplanted human hearts. Circulation 2002;106:17e9. [73] Grimm P, Nickerson P, Jeffery J, Savani R, Gough J, Mckenna R, et al. Neointimal and tubulointerstitial infiltration by recipient mesencymal cells in chronic renal-allograft rejection. N Eng J Med 2001;345:93e7. [74] Simper D, Stalboerger P, Panetta C, Wang S, Caplice N. Smooth muscle progenitor cells in human blood. Circulation 2002;106:1199e204. [75] Caplice N, Doyle B. Vascular progenitor cells: origin and mechanisms of mobilization, differentiation, integration, and vasculogenesis. Stem Cells Dev 2005;14:122e39. [76] Caplice N, Bunch T, Stalboerger P, Wang S, Simper D, Miller D, et al. Smooth muscle cells in human coronary atherosclerosis can originate from cells administered at marrow transplantation. Proc Natl Acad Sci U S A 2003;100: 4754e9. [77] Minami E, Laflamme M, Saffitz J, Murry C. Extracardiac progenitor cells repopulate most major cell types in the transplanted human heart. Circulation 2005;112:2951e8. [78] George J, Pinderski L, Litovsky S, Kirklin J. Of mice and men: mouse models and the molecular mechanisms of post-transplant coronary artery disease. J Heart Lung Transplant 2005;24:2003e14. [79] Fogelstrand P, Osterberg K, Mattsson E. Reduced neointima in vein grafts following a blockage of cell recruitment from the vein and the surrounding tissue. Cardiovasc Res 2005;67:326e32. [80] Feng S, Lu X, Resendiz J, Kroll M. Pathological shear stress directly regulates platelet alphaII betaIII signaling. Am J Physiol Cell Physiol 2006;291:1346e54. [81] Fateh-Moghadam S, Bocksch W, Ruf A, Dickfeld T, Schartl M, PogatsaMurray G, et al. Changes in surface expression of platelet membrane glycoproteins and progression of heart transplant vasculopathy. Circulation 2000;102:890e7. [82] Abele S, Weyand M, Wollin M, Hiemann N, Harig F, Fischlein T, et al. Clopidogrel reduces the development of transplant arteriosclerosis. J Thorac Cardiovasc Surg 2006;131:1161e6. [83] Weston M, Laborde O, Yoganathan A. Estimation of the shear stress on the surface of an aortic valve leaflet. Ann Biomed Eng 1999;27:572e9. [84] Nelson P, Krensky A. Chemokines, chemokine receptors, and allograft rejection. Immunity 2001;14:377e86. [85] Shimizu K, Mitchell R. The role of chemokines in transplant graft arterial disease. Arterioscl Thromb Vasc Biol 2008;28:1937e49. [86] Luster A. Chemokines: chemotactic cytokines that mediate inflammation. N Eng J Med 1998;338:436e45. [87] Vonhundelshausen P, Weber K, Huo Y, Proudfoot A, Nelson P, Ley K, et al. RANTES deposition by platelets triggers monocyte arrest on inflamed and atherosclerotic endothelium. Circulation 2001;103:1772e7. [88] Boring L, Gosling J, Chensue S, Kunkel S, Farese R, Broxmeyer H, et al. Impaired monocyte migration and reduced type 1 (Th1) cytokine responses in CCR2 knockout mice. J Clin Invest 1997;100:2552e61. [89] Gosling J, Slaymaker S, Gu L, Tseng S, Zlot C, Young S, et al. MCP-1 deficiency reduces susceptibility to atherosclerosis in mice that overexpress human apolipoprotein B. J Clin Invest 1999;103:773e8. [90] Robinson L, Nataraj C, Thomas D, Howell D, Griffiths R, Bautch V, et al. A role for fractalkine and its receptor (CX(3)CR1) in cardiac allograft rejection. J Immunol 2000;165:6067e72. [91] Gao W, Topham P, King J, Smiley S, Csizmadia V, Lu B, et al. Targeting of the chemokine receptor CCR1 suppresses development of acute and chronic cardiac allograft rejection. J Clin Invest 2000;105:35e44. [92] Shimizu K, Aikawa M, Takayama K, Libby P, Mitchell R. Cerivastatin has immunomodulatory effects on donor endothelial cells and host inflammatory cells and attenuates allograft arteriopathy. Circulation 2001;104:II212e3. [93] Schecter A, Calderon T, Berman A, Mcmanus C, Fallon J, Rossikhina M, et al. Human smooth muscle cells possess functional CCR5. J Biol Chem 2000;275: 5466e71. [94] Goldberg S, van der Meer P, Hesselgesser J, Jaffer S, Kolson D, Albright A, et al. CXCR3 expression in human central nervous system diseases. Neuropathol Appl Neurobiol 2001;27:127e38. [95] Sakihama H, Masunaga T, Yamashita K, Hashimoto T, Inobe M, Todo S, et al. Stromal cell-derived factor-1 and CXCR4 interaction is critical for development of transplant arteriosclerosis. Circulation 2004;110:2924e30. [96] Veillard N, Steffens S, Pelli G, Lu B, Kwak B, Gerard C, et al. Differential influence of chemokine receptors CCR2 and CXCR3 in development of atherosclerosis in vivo. Circulation 2005;112:870e8.

R.N. Mitchell / Journal of Autoimmunity 45 (2013) 80e89 [97] Schecter A, Berman A, Taubman M. Chemokine receptors in vascular smooth muscle. Microcirculation 2003;10:265e72. [98] Kodali R, Kim W, Galaria I, Miller C, Schecter A, Lira S, et al. CCL11 (Eotaxin) induces CCR3-dependent smooth muscle cell migration. Arterioscler Thromb Vasc Biol 2004;24:1211e6. [99] Shimizu K, Mitchell R. Chemokine-mediated recruitment of inflammatory and smooth muscle cells in transplant-associated arteriosclerosis. Curr Opin Organ Transpl 2003;8:55e63. [100] Denton M, Davis S, Baum M, Melter M, Reinders M, Exeni A, et al. The role of the graft endothelium in transplant rejection: evidence that endothelial activation may serve as a clinical marker for the development of chronic rejection. Pediatr Transplant 2000;4:252e60. [101] Dietrich H, Hu Y, Zou Y, Dirnhofer S, Kleindienst R, Wick G, et al. Mouse model of transplant arteriosclerosis: role of intercellular adhesion molecule1. Arterioscler Thromb Vasc Biol 2000;20:343e52. [102] Pribila J, Quale A, Mueller K, Shimizu Y. Integrins and T cell-mediated immunity. Annu Rev Immunol 2004;22:157e80. [103] Lacy-Hulbert A, Ueno T, Ito T, Jurewicz M, Izawa A, Smith R, et al. beta3 integrins regulate lymphocyte migration and cytokine responses in heart transplant rejection. Am J Transplant 2007;7:1080e90. [104] Shimizu K, Libby P, Shubiki R, Sakuma M, Wang Y, Asano K, et al. Leukocyte integrin Mac-1 promotes acute cardiac allograft rejection. Circulation 2008;117:1997e2008. [105] Richter M, Zahn S, Richter H, Mohr F, Olbrich H. Reduction of ICAM-1 and LFA-1-positive leukocytes in the perivascular space of arteries under mycophenolate mofetil therapy reduces rat heart transplant vasculopathy. J Heart Lung Transpl 2004;23:1405e13. [106] Soleimani B, Wieczorek G, Katopodis A, Zenke G, George A, Hornick P, et al. Anti-LFA-1 monotherapy prevents neointimal formation in a murine model of transplant intimal hyperplasia. J Heart Lung Transpl 2007;26: 724e31.

89

[107] Moiseeva E. Adhesion receptors of vascular smooth muscle cells and their functions. Cardiovasc Res 2001;52:372e86. [108] Willis A, Pierre-Paul D, Sumpio B, Gahtan V. Vascular smooth muscle cell migration: current research and clinical implications. Vasc Endovascular Surg 2004;38:11e23. [109] Pure E, Cuff C. A crucial role for CD44 in inflammation. Trends Mol Med 2001;7:213e20. [110] Goueffic Y, Guilluy C, Guerin P, Patra P, Pacaud P, Loirand G. Hyaluronan induces vascular smooth muscle cell migration through RHAMM-mediated PI3K-dependent Rac activation. Cardiovasc Res 2006;72:339e48. [111] Jain M, He Q, Lee W-S, Kashiki S, Foster L, Tsai J, et al. Role of CD44 in the reaction of vascular smooth muscle cells to arterial wall injury. J Clin Invest 1996;97:596e603. [112] Cuff C, Kothapalli D, Azonobi I, Chun S, Zhang Y, Belkin R, et al. The adhesion receptor CD44 promotes atherosclerosis by mediating inflammatory cell recruitment and vascular cell activation. J Clin Invest 2001;108:1031e40. [113] Routledge M, Rush D, Mckenna R, Gough J, Jeffery J, Stem E, et al. The receptor for hyaluronan-mediated motility is expressed in human renal allografts and is correlated with Banff chronic rejection scores. Transpl Proc 1997;29:2603e4. [114] Gershwin ME, Shoenfeld Y. Chella David: a lifetime contribution in translational immunology. J Autoimmun 2011;37:59e62. [115] Jamin C, Renaudineau Y, Pers JO. Pierre Youinou: when intuition and determination meet autoimmunity. J Autoimun 2012;39:117e20. [116] Selmi C, Leung PS, Sherr DH, Diaz M, Nyland JF, Monestier M, et al. Mechanisms of environmental influence on human autoimmunity: a national institute of environmental health sciences expert panel workshop. J Autoimmun 2012;39:272e84. [117] Tzioufas AG, Kapsogeorgou EK, Moutsopoulos HM. Pathogenesis of Sjogren’s syndrome: what we know and what we should learn. J Autoimmun 2012;39:4e8.