Some genetic defects in adhesion mechanisms result in variable degrees ofim-munodeficiency, for example, leukocyte adhesion deficiency (LAD)-1 (lack of functional 02 integrins) (Kishimoto et al. 1987), LAD-2 (deficiency in selectin ligands) (Etzionietal. 1992), or LAD-3 (defects in integrin activation) (Kinashi et al. 2004). These paradigmatic "experiments of nature" imply that the manipulation of leukocyte adhesion/homing could be a useful immunotherapeutic tool. In fact, a mAb against the integrin LFA-1 (efalizumab) is being used in the clinic for the treatment of psoriasis (Lebwohl et al. 2003). Anti-LFA may also be effective in the treatment of refractive cutaneous GVHD (Stoppa et al. 1991) and possibly other inflammatory/immune-mediated diseases. However, blocking either LFA-1 or one of its ligand ICAM-1 has not been successful in stroke, myocardial infarction, and traumatic shock (Schreiber et al. 2001; Yacyshyn et al. 2002; Yonekawa and Harlan 2005). On the other hand, because LFA-1 and ICAM-1 are required for lymphocyte adhesion in a multitude of tissues, interfering with these molecules could cause undesired immunosup-pression. It has been postulated that modulating lymphocyte adhesion in a tissue-specific fashion could be an effective and less immunosuppressive alternative in this regard.
Recent events have prompted intense discussion on the validity of this argument. Clinical trials with Natalizumab (Tysabri), a mAb that blocks the binding of both a401 (VLA-4) to VCAM-1 and a407 to MAdCAM-1 on Th1 cells that infiltrate the brain and gut, respectively (von Andrian and Engelhardt 2003), have shown that the mAb is efficacious in the treatment of both Crohn disease (Ghosh et al. 2003) and, especially, multiple sclerosis (Miller et al. 2003). These exciting clinical results were initially hailed as validation of the approach of blocking traffic molecules to treat autoimmune inflammatory diseases. However, Natalizumab was subsequently withdrawn from the market when it was learned that a small percentage of patients treated with this drug developed progressive multifocal leukoencephalopathy (PML), a deadly opportunistic CNS infection caused by JC polyomavirus. The occurrence of PML in this setting was unanticipated because this condition is normally only seen in patients with severely impaired cell-mediated immunity, such as in AIDS, leukemia, or organ transplantation. The majority of humans have detectable antibody titers against JC virus, and many individuals test positive for viral DNA but never experience viral disease. Therefore, although Natalizumab blocked the entry of encephalitogenic CD4+ T cells and the subsequent CNS demyelination, the drug apparently also inhibited the prerequisite immunosurveillance by (presumably) a subset of T cells that normally protect the brain against JC virus infection.
Another tissue with defined homing specificity is the skin. Consistent with the role of E- and P-selectin in skin homing, mice lacking these receptors have a compromised delayed-type contact hypersensitivity response and suffer from spontaneous skin infections (Hirata et al. 2002; Staite et al. 1996). In nonhuman primates, recruitment of lymphocytes into DTH skin depended on VCAM-1 and E-selectin (Silber et al. 1994), and blocking of the E-selectin lig-and CLA prevented migration of human T cells into human skin transplanted into SCID mice (Biedermann et al. 2002). However, blocking E-selectin alone was ineffective in clinical trials for psoriasis (Bhushan et al. 2002). This failure could have been due to the redundancy of P- and E-selectin expression and function in the skin (Schon et al. 2004; Staite et al. 1996; Weninger et al. 2000). Similarly, although blocking the CCR10-ligand CCL27 effectively decreased skin inflammation in DTH and atopic dermatitis in mice (Homey et al. 2002), there is likely redundancy among the skin-associated chemokine receptors CCR10 and CCR4, at least in some settings of skin inflammation (Reiss et al. 2001). Moreover, it is possible that other inflammation-induced receptors play a role in inflamed skin. In fact, clinical trials are currently underway targeting CCR1 and CXCR3 in psoriasis (Johnson et al. 2005).
The gut mucosa is arguably the most paradigmatic example of tissue-specific homing. Blockade of a407 in a spontaneous model of colitis in nonhuman primates rapidly reverts the disease (Hesterberg et al. 1996). Similarly, in a chronic colitis model caused by transfer of CD45RBHi CD4 T cells into SCID mice, blocking 07 integrins and/or MAdCAM-1 inhibited the development of colitis (Picarella et al. 1997). Consistent with these data, Natalizumab has been efficacious in clinical trials of Crohn disease (Ghosh et al. 2003). However, as discussed above, there are important safety concerns associated with global inhibition of a4 integrins (Steinman 2005). Nevertheless, a recent trial in ulcerative colitis selectively targeting the gut-homing receptor a407 has successfully reached its therapeutic end point (Feagan et al. 2005). Because this approach does not interfere with a401 pathway in the CNS, it seems less likely to elicit susceptibility to PML or other opportunistic infections.
Another pathology involving the gut is GVHD, and interfering with the function of a4 or a407 has been effective in preventing gut and liver (but not cutaneous) GVDH in mice (Murai et al. 2003; Petrovic et al. 2004; Tanaka et al. 1995). However, it is possible that other molecules induced on inflammation are also involved in this pathology, for example, the chemokine receptors CXCR3 (Duffner et al. 2003) and CCR5 (Murai et al. 2003).
Interestingly, in some studies of acute experimental colitis, blocking tissue-specific homing receptors has been only partially or not at all effective. For instance, blocking VCAM-1, but not the a407 ligand MAdCAM-1, was shown to attenuate DSS colitis (Soriano et al. 2000). This could be due in part to the fact that DSS-induced colitis does not depend on T and B lymphocytes (Axelsson et al. 1996). In addition, it is possible that other non-tissue-specific adhesion molecules induced under inflammatory conditions could (at least in part) override the role of some tissue-specific homing receptors and ad-dressins. Consistent with this possibility, blocking the CCR6-ligand CCL20 (in addition to MAdCAM-1) was necessary to decrease the accumulation of T and B cells in DSS-induced colitis (Teramoto et al. 2005). Additionally, blocking a 1 p 1, a collagen-binding integrin upregulated in many inflammatory settings, decreases TNBS-induced acute colitis in mice (Fiorucci et al. 2002). Moreover, it has been found in humans that CCR2 is expressed by most CD4 T cells infiltrating the lamina propria in Crohn's ileitis (Connor et al. 2004), suggesting that this receptor (with or without CCR9) might play a role in this intestinal pathology. Furthermore, in SAMP-1/Yit mice, a strain that develops spontaneous chronic enteritis similar to Crohn's disease, blocking ICAM-1 plus VCAM-1 reduced inflammation (Burns et al. 2001). Interestingly, blocking p7, a4p7, or MAdCAM-1 alone did not prevent ileitis, but blocking L-selectin in combination with MAdCAM-1 or a4 ameliorated the disease, suggesting that under chronic intestinal inflammation, non-gut-specific molecules such as a4p1 and L-selectin might play important roles in the pathogenesis of these diseases (Rivera-Nieves et al. 2005). However, using the same mouse model, another group has reported recently that blocking MAdCAM-1 alone was not only effective but even better than VCAM-1 blockade (Matsuzaki et al. 2005). Whether the discrepancies among these reports could have been due to differences in experimental treatment schedules or some other factors remains to be determined.
Was this article helpful?
Do You Suffer From the Itching and Scaling of Psoriasis? Or the Chronic Agony of Psoriatic Arthritis? If so you are not ALONE! A whopping three percent of the world’s populations suffer from either condition! An incredible 56 million working hours are lost every year by psoriasis sufferers according to the National Psoriasis Foundation.