| Sign In to gain access to subscriptions and/or personal tools. |
The Molecular Basis of Salivary Gland Involvement in Graft-vs.-Host Disease
1 Department of Oral and Maxillofacial Surgery, Oral Biochemistry Laboratory and Salivary Clinic, Rambam Medical Center, and Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel; and Correspondence: * corresponding author, nagler{at}tx.technion.ac.il
During the past two decades, the involvement of salivary glands in graft vs. host disease (GVHD) had been intensively researched and published. GVHD occurs in 40–70% of patients treated with bone marrow and peripheral blood stem cell transplantation (PBSCT), and improved survival rates have led to a continuously increasing number of GVHD patients suffering from induced salivary insult. Limited studies suggest that a large percentage of GVHD patients is affected and that the induced salivary dysfunction occurs rapidly following the transplantation. It affects both major and minor salivary glands and reflects the severity of the disease. Moreover, profound sialochemical alterations may be diagnostic of GVHD. An additional reason for this vast research is that GVHD, as an autoimmune-like disease, seemed to be an appropriate model for studying a much more prevalent and well-known and well-studied autoimmune disease involving salivary glands: Sjögren’s syndrome. The purpose of the current review—which is, to the best of our knowledge, the first of its kind—is to describe the GVHD-related sialometric and sialochemical data published in the past two decades for both major and minor salivary glands and to discuss the pathogenesis and molecular basis of the disease.
Key Words: graft vs. host disease • saliva • salivary gland • sialochemistry • sialometry
Chronic graft-vs.-host disease (cGVHD) is a major, late complication of allogeneic stem cell transplantation (alloSCT) and is the principal cause of morbidity and non-relapse mortality. With increasing numbers of unrelated and mismatched alloSCT, and the change from bone marrow to peripheral blood grafts, the number of patients with cGVHD is increasing (Schmitz et al., 2002). Chronic GVHD may develop in 25–45% of recipients undergoing alloSCT from matched, related siblings, and its frequency increased to 40–70% in patients receiving alloSCT from matched, unrelated donors (Storek et al., 1997). The incidence is significantly lower in patients receiving human umbilical cord blood grafts and in patients receiving haploidentical CD34+ purified grafts (Aversa et al., 1998; Rocha et al., 2000). Chronic GVHD may develop as an extension of acute GVHD (progressive onset), after resolution of acute GVHD (quiescent onset), or without preceding acute GVHD (de novo onset). Chronic GVHD is categorized as either limited (localized skin and/or hepatic involvement) or extensive (diffuse skin and/or multi-organ involvement); the latter is associated with a worse prognosis (Sullivan et al., 1981). The introduction of low-intensity conditioning (LIC) and non-myeloablative alloSCT did not result in reduction in the frequencies of cGVHD; in fact, it seems that the frequencies are even increasing (Shimoni and Nagler, 2001). cGVHD usually develops more than 100 days after transplants, with a tendency for later occurrence in patients receiving peripheral blood stem cell grafts and patients undergoing LIC alloSCT (Shimoni and Nagler, 2001; Schmitz et al., 2002). Chronic GVHD typically resembles a connective-tissue-autoimmune-like immunological disorder, characterized by lichenoid or sclerodermoid lesions of skin along with joint contractors similar to those seen in systemic sclerosis. The clinical and pathologic features resemble the overlapping of several collagen vascular diseases and immune dysregulation with eosinophilia, circulating autoantibodies, hypergammaglobulinemia, and plasmacytosis of the viscera and lymph nodes (Sullivan et al., 1981). The skin changes that are the hallmark of the disease include papulosquamous dermatitis, plaques, desquamation, dyspigmentation, and vitiligo. Chronic GVHD has histopathological features similar to those of SSc, manifested predominantly by sclerosis of the thickened reticular dermis due to the increased synthesis of collagen (Shulman et al., 1978). The histologic distinction between the papillary and reticular dermis may not be apparent because of sclerosis. Cutaneous appendages become encased in collagen and tend to disappear. There is variable perivascular and interstitial inflammatory cell infiltrate, composed predominantly of lymphocytes and occasionally plasma cells (Janin-Mercier et al., 1984). Less frequent skin findings include poikiloderma, reticulated hyperpigmentation, alopecia, dystrophic nails, leucoderma bullae, discoid lupus erythematosis, and photosensitivity. The pathophysiologic mechanism involves autoreactive lymphocytes and cytokine dysregulation (Ilan et al., 2000; Nagler et al., 2000).
Current therapeutic options are limited. Less than 20% of patients with untreated extensive cGVHD survive with Karnofsky performance scores
The pathogenesis of GVHD is based on donor graft T-lymphocytes, which recognize antigenic disparities between donor and recipient. Consequently, various tissues and organs are affected by the cytotoxicity caused by the infiltrating T-cells, including the mucosa of the oral and gastrointestinal tract (Izutsu et al., 1983a; Slavin and Nagler, 1991). The mucosal insult is further enhanced by the reduced quantity and the altered quality of the saliva, which ordinarily contributes significantly to the preservation of mucosal integrity, because the salivary glands are also a major target of GVHD (Chaushu et al., 1994a,b). The consequences of the GVHD-induced salivary injury include suffering on the part of the patients and a reduction in other related salivary functions, such as anti-infection activity, protection against mechanical and chemical epithelial injuries and against periodontal disease and caries, contribution to verbal communication, to nutrition, to soft tissue repair, etc. (Fox et al., 1985).
In a recently published long-term study, it was found that salivary gland involvement in GVHD occurs very rapidly, is severe, and that recovery does not occur during the following year, during which patients are monitored (Nagler and Nagler, 2003). In this study, mean whole salivary flow rates were obtained at 0 time (prior to allogeneic PBSCT), at 2 mos, and at 12 mos. The mean flow rate of the GVHD patients (n = 12) prior to the allogeneic PBSCT was 0.59 ± 0.11 mL/min. The values for allogeneic PBSCT/disease-free (n = 6) and healthy (n = 8) control groups were similar, 0.42 ± 0.15 mL/min and 0.51 ± 0.13 mL/min, respectively. In contrast to the control groups, in which the mean flow rates were not significantly altered during the year of follow-up, the flow rates of the GVHD group decreased gradually and were reduced by 39% at 2 mos and by 70% at 12 mos (p < 0.01). Recognition of the salivary glands as a very sensitive target for GVHD is not surprising in light of the study published by Nakhleh et al.(1989), in which it was demonstrated that, in 50% of the GVHD patients in whom minimal or no oral mucosal involvement could be demonstrated (a well-established major target of GVHD), the salivary glands presented definite GVHD involvement. Reduction in whole saliva flow rate in GVHD patients has also been reported in other studies as well as in various salivary biochemical and immunological compositional alterations (Izutsu et al., 1985; Hiroki et al., 1994; Singhal et al., 1995; Nagler et al., 1996a). In conclusion, the available reports of both human and animal models reveal a mean reduction of 55–90% in the salivary flow rate of GVHD patients (Norhagen et al., 1994; Nagler et al., 1996a,b, 1998; Nagler and Nagler, 2001).
In a recently published sialochemical study (Nagler and Nagler, 2001), it was reported that GVHD patients had significantly higher salivary concentrations of Na by 395–454% (p < 0.01), of Mg by 109–113% (p < 0.05), and of epidermal growth factor (EGF), total protein, albumin, and IgG by up to 190% (p < 0.05) as compared with controls. These altered concentrations may result in a compromised functional capacity of the saliva, since this capacity is directly related to normal salivary composition (Nagler et al., 1996b; Nagler and Nagler, 2001). In that event, the saliva may not be able to play its pivotal role in nutritional function as well as its roles in controlling oral micro-organisms, maintaining mucosal integrity and protection against foreign proteins and infections, reducing the formation of periodontal calculus, and sustaining anticariogenic function (Fox et al., 1985). Such reduced salivary capacity in controlling micro-organisms in GVHD patients is supported by Norhagen et al.(1994), who reported that patients with GVHD had less salivary IgM one year after bone marrow transplantation. It is also supported by a study published by Izutsu et al.(1985), who examined a large group of GVHD patients (n = 42 at 90 days) for up to 2 yrs following BMT. Labial gland IgA, which reflects local production of IgA from the mucosal immune system, was studied, rather than whole saliva. A decreased concentration of IgA was found 90 days after transplantation, which appeared to be associated with systemic immunosuppression. The decrease in labial IgA, the predominant immunoglobulin of the mucosal immune system, strongly suggests that mucosal immune systems are impaired in patients with GVHD. A host with impaired mucosal immune functions would be less likely to respond to local antigen presentation. In addition, patients with impaired mucosal immunity secondary to other conditions can have an increased incidence of sino-pulmonary infections. Thus, defining changes in the mucosal vs. systemic immune system is important when one is considering GVHD. In any case, the altered salivary composition may act synergistically with the compromised salivary flow rate, as previously discussed. The reported three- to four-fold increase in Na concentration (Nagler and Nagler, 2001) is in agreement with reports from two other fundamental studies in which Izutsu et al. (1983a,b) demonstrated an increase in Na concentrations in the secretions of both minor salivary glands and whole saliva of GVHD patients. This increase in Na concentration could be accounted for by GVHD-induced and lymphocyte-infiltration-mediated damage to the Na-re-absorbing salivary ductal system. A similar salivary Na increase was demonstrated in Sjögren’s syndrome patients as well, and the etiology suggested in this autoimmune disease was similar, being also based on the role played by infiltrating autoreactive lymphocytes (Chisholm and Mason, 1968). Concentrations of the three other electrolytes—K, Ca, and P—were found to be similar in patients and in controls (Nagler and Nagler, 2001). This result is also supported by Izutsu et al.(1983a), who reported that resting salivary K concentrations in GVHD patients are not different from those in controls. With respect to the anti-infection activity and the maintenance of mucosal integrity in the oral cavity, it should be emphasized that salivary IgA and EGF are of major importance (Deville de Périère and Aranciba, 1988–89; Marti et al., 1989). The significant salivary IgG increase in GVHD patients is also in accord with the previously mentioned study by Izutsu et al.(1983a). The significant elevation of EGF, total protein, albumin, and IgG could result from direct GVHD-induced damage to the salivary parenchyma, as previously suggested. However, it could also result from a transudation of serum components across the damaged and inflamed oral epithelium and gingiva, which are sites affected by the disease (Nagler et al., 1996a). Thus, it may be concluded that a multi-site mechanism may be differentially responsible for the observed increase of various components which either are being secreted from the salivary glands as EGF or leak from the serum into the saliva as albumin. The salivary EGF is of special importance, and more so at resting conditions, since resting salivary secretion is the dominant condition during most of the day and night, and the dominant secreting salivary gland in this condition is the submandibular gland (Nagler and Nagler, 1999). In any event, a mere "concentrating effect" of reduced secreted volume due to a decrease of the watery component of the saliva (related to a specific insult of the muscarinic signal transduction pathway, for example) is excluded, since such a mechanism might explain an identical increase of the EGF, total protein, and immunoglobulins but not a differentiated one, as observed.
It is now realized that molecules having a putative mechanistic role in GVHD are adhesion molecules as up-regulators of the intracellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM), and E-selectin, which have been demonstrated in GVHD (Matsuda et al., 2001), indicating that the activation and injury of endothelium are commonly involved in the pathogenesis of GVHD. A cascade of signaling events directs and regulates the trafficking, homing, and activating of T-lymphocytes in GVHD. These adhesion receptors include selectins, integrins, and adhesion molecules of the immunoglobulin superfamily. In addition, tissue-specific homing receptors direct the tissue-specific trafficking of T-lymphocytes. It would be of interest to define and isolate the salivary-specific homing receptors that direct the cytotoxic T-lymphocytes mediating the salivary GVHD-induced injury. Accordingly, treatment with antibodies against CCAM-1 or adhesion molecule VLA-4 blocked the adhesion-mediated tissue cell infiltration and improved liver injury in a GVHD mouse model (Itoh et al., 2000). Other adhesion molecules—such as CD31, CD49b, and CD62L—have been identified as minor histocompatibility antigens that are proposed as being the key factors in GVHD following matched sibling transplantation (Maruya et al., 1998). Another new approach illustrating and emphasizing the role of cytokines in GVHD is cytokine gene polymorphism, which is associated with functional differences in cytokine regulation. A correlation was found between the severity and frequency of GVHD and IFN-alpha, IL-6, IL-10, and TNF-alpha gene polymorphism. Specifically, patients homozygous for the IFN-alpha Intron allele 3 had more severe (Grade III-IV) acute GVHD. Similarly, patients possessing the IL-6 (–174) G allele had a trend toward higher grades of acute GVHD, and those homozygous for the IL-6 (–174) G allele were more likely to develop chronic GVHD (Cavet et al., 2001). Recently, the rather new cytokine IL-18 has been implicated in GVHD (Panoskaltsis-Mortari et al., 2000; Itoi et al., 2001). Increased serum levels of IL-18 were found in patients who developed GVHD post-allogeneic PBSCT. IL-18 is a cytokine with wide-ranging biological functions, including not only innate but also acquired immunity, including both Th responses, particularly in collaboration with IL-12, and Th2 responses. IL-18 is involved in the development of cytotoxic T-lymphocytes and natural killer cells, which may mediate the salivary injury observed in GVHD. Regarding the mechanism, in murine experiments, it was demonstrated that recipient mice transplanted with H-2 disparate donor GLD/GLD spleen cells, which lack functional Fas ligand (FasL), developed GVHD, but no elevation of IL-18 was observed (Panoskaltsis-Mortari et al., 2000), indicating that FasL mediates IL-18 release in GVHD. Furthermore, IL-18 elevation was found to be derived from host cells in a caspase-1-dependent manner (Panoskaltsis-Mortari et al., 2000). Salivary-mediated lesions in GVHD show histological features of cell death with lymphocyte infiltration. It was recently demonstrated that perforin and granzyme B are involved in the process of apoptosis, induced by cytotoxic T-lymphocytes, which leads to epidermal injury in GVDH (Higaki et al., 2001). There may be a similar mechanism at work in the salivary injury in GVHD. Interleukin-13 is a new Th2 cytokine that has recently been shown to suppress alloreactivity and to be of protective value in GVDH. Moreover, the keratinocyte growth factor (KGF) that has been demonstrated in both mice and humans as ameliorating GVHD may operate through IL-13/KGF (Panoskaltsis-Mortari et al., 2000), and IL-13 may therefore be of therapeutic value in ameliorating GVHD-mediated salivary injuries. One of the mechanisms that may be critical to the attraction and recruitment of cytotoxic T-cells to the salivary glands, thus mediating the GVHD-mediated injury, is the production of macrophage inflammatory protein 1 alpha (MIP-1 alpha), since it was recently shown that production of MIP-1 alpha by donor T-cells is important in the occurrence of GVHD and functions in a tissue-dependent fashion (Serody et al., 2000). Another factor that may be implicated in salivary injury is HSP 70. It has been shown that increased levels of HSP 70 and antibodies reactive with HSP 70 parallel the onset and severity of GVHD. Moreover, deoxyspergualin, which ameliorates GVHD, has been shown to reduce hsp 70 levels, resulting in diminished serum levels of IL-2, IFN-gamma, and TNF-alpha, which have been implicated in GVHD-mediated end-organ injury (Goral et al., 2000). A molecule that has been implicated in the tissue destruction of GVHD is nitric oxide. Increased levels of nitric oxide have been shown in GVHD, and blockage of nitric oxide production and pathways may have a therapeutic role. The onset of GVHD is accompanied by macrophage (M phi) priming, which results in expression of nitric oxide synthase and the production of nitric oxide in response to LPS. Continuous exposure to IFN-gamma is required to maintain a primed state of M phi during GVHD. Indeed, increased IFN-gamma in RNA has been demonstrated in salivary gland tissue in mice with GVHD (Kichian et al., 1996).
Two studies published by Hiroki et al.(1994, 1996) added significantly to our understanding of the cellular and molecular basis of salivary involvement in GVHD. In these studies, it was shown that this involvement is characterized by lymphocyte (mostly T-lymphocytes) infiltration into the glandular parenchyma and especially around the secretory ducts. Most of these infiltrating lymphocytes were T-cells with a predominance of CD8+ over CD4+. Ductal epithelial cell-associated lymphocytic infiltration expressed HLA-DR, while the expression of adhesion molecules and especially of vascular cell adhesion molecule 1 (VCAM-1) was found mostly in endothelial cells but also in epithelial cells. Such periductal infiltration, accompanied by salivary parenchymal atrophy and destruction, had been noted in both human and animal models in various other papers (Nagler et al., 1996b, 1998). Moreover, the levels of these pathohistological alterations were found to correlate significantly with the level of induced salivary hypofunction.
Accordingly, a suggested pathogenesis mechanism for salivary gland involvement in GVHD is composed of the following: HLA up-regulation, mononuclear infiltration, and cytokine dysregulation (Izutsu et al., 1983a; Chaushu et al., 1994a). Indeed, various cytokines—including IL-2, IL-6, IFN- In summary, the suggested multisite mechanism gains support from the following:
As for the presumed resemblance of salivary gland involvement in GVHD and in Sjögren’s syndrome, it is important to note the finding of Hiroki et al.(1996), who found two examined parameters to be different in these diseases:
Although it is tempting to assume that both entities are very similar in light of the immunological background and the mutual clinical symptoms of xerostomia and xerophthalmia, one should question this assumption.
Currently, our knowledge is still somewhat deficient in understanding the underlying mechanism of salivary involvement in GVHD. Is it really an immunological disease only, or are other molecules playing a role in the cascade of events which result in such rapid and devastating effects (almost total xerostomia in acute GVHD and very severe in chronic GVHD with no recovery [Nagler et al., 1996b; Nagler and Nagler, 2003])? It would be warranted to elucidate, on the one hand, the specific molecules involved (cytokines, immunoglobulins, free radicals, etc.) and, on the other, the targets attacked in the salivary parenchymal cells (shut-down or up-regulated genes?). Such an elucidation may pave the way for the development of means to protect against this disturbing consequence of bone marrow transplantation.
The authors’ work was performed under the auspices of the Department of Oral and Maxillofacial Surgery, Oral Biochemistry Laboratory, and Salivary Clinic at Rambam Medical Center and the Faculty of Medicine of Technion-Israel Institute of Technology in Haifa, and the Department of Bone Marrow Transplantation of Chaim Sheba Medical Center in Tel Hashomer, Israel. Received for publication December 27, 2001. Revision received September 15, 2003. Accepted for publication November 14, 2003.
Journal of Dental Research, Vol. 83, No. 2,
98-103 (2004)
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
70% (
, TNF-
, and IGF—have recently been shown to influence salivary cell function and survival and the production of salivary immunoglobulin and/or saliva (