Role of leukocytes and leukocyte adhesion molecules in renal ischemic-reperfusion injury

Hamid Rabb, M.D.

Division of Nephrology and Hypertension, J. A. Haley VA Hospital & University of South Florida, Tampa, FL 33612

Received 12/07/95; Accepted 12/25/95; On-line 1/1/95

4. The role of leukocyte adhesion molecules in renal reperfusion injury

Over the last decade there has been an explosion of information regarding the biological basis of leukocyte interactaction with other cells as well as with matrix. Developments in this field of leukocyte adhesion have led to a resurgence of interest in studying the role of leukocytes in mediating tissue injury, particularly due to the development of specific blocking antibodies and other reagents directed to these adhesion molecules. There are many excellent reviews on the basic biology of the three main leukocyte adhesion molecule families: integrins (22,23), immunoglobulin superfamily members (24), and selectins (25). Thus, only the some of the most relevant aspects of these molecules will be reviewed here. This will be followed by a section discussing the role of these molecules in renal IRI.

4.1 Integrins: The integrins which appear most important in leukocyte-endothelial adhesion, particularly in terms of clinical relevance in the kidney, are the beta-1 integrin VLA-4 (very late antigen-4) and the beta-2 integrins, CD11/CD18 (26,27). The VLA-4 molecule is found on lymphocytes, basophils, and eosinophils, but not neutrophils. This protein binds to endothelial cells through an inducible ligand, vascular cell adhesion molecule-1 (VCAM-1) (28). CD11/CD18 integrins are expressed solely on leukocytes, and there are 3 known members of this family, all with a distinct alpha subunit and a common beta subunit. CD11/CD18 deficiency disease (leukocyte adhesion deficiency) is a rare childhood immunodeficiency that initially shed light on the key physiologic role of CD11/CD18 (29). Subsequently, the molecular basis for complete and partial CD11/CD18 deficiency has been largely worked out (30-33). The major ligand for CD11a/CD18 is intercellular adhesion molecule-1 (ICAM-1), though other ligands such as ICAM-2 and ICAM-3 exist (34-37). CD11b/CD18, in addition to binding to ICAM-1, also binds to complement fragment iC3b, factor X and fibrinogen (38). CD11c has not been studied as extensively. CD11/CD18 functions are tightly regulated, in part by key amino acid residues on the CD18 cytoplasmic domain (39).

4.2 Immunoglobulin superfamily members: Immunoglobulin (Ig) - like adhesion receptors are single-chain transmembrane proteins that contain one or more immunoglobulin domains (24). Among the many members of this family are ICAM-1, VCAM-1, CD4, CD8 and carcinoembryonic antigen (24). ICAM-1 is expressed on endothelial, epithelial and fibroblast cells as well as on leukocytes (24). Various cytokines (eg. interferon, tumor necrosis factor-alpha, and interleukin-1) can induce ICAM-1 expression. VCAM-1 is one of the major ligands for VLA-4 (28). VCAM-1 is expressed on many different tissues, including vascular endothelium, epithelial cells, bone marrow cells, and some macrophages (40). Like ICAM-1, VCAM-1 is also cytokine-inducible, but may have differential regulators (41).

4.3 Selectins: The selectin family consists of 3 different single chain transmembrane receptors: E, P, and L-selectin. These receptors have an N-terminal lectin domain, an epidermal growth factor domain, complement regulatory repeats, and a transmembrane and short cytoplasmic domain (25). E-selectin is expressed on cytokine (eg. IL-1, TNF-alpha, endotoxin) stimulated endothelium but usually not on non-activated endothelium (42). L-selectin is constitutively expressed on lymphocytes, neutrophils, monocytes and other myeloid cells (43). L-selectin on lymphocytes mediates adhesion to high endothelial venules in peripheral lymph nodes, as well as neutrophil recruitment to the inflammatory sites (25). P-selectin is expressed in platelets and activated endothelial cells, and appears to have different inducers than E-selectin (eg. hydrogen peroxide, thrombin) (25). The selectins bind to sialylated and fucosylated structures related to the Lewis-X blood group antigen (44,45). Like CD11/CD18 deficiency disease, some children have been identified that suffer from a defect in the selectin pathway (46). Children lacking SLe^x, a selectin ligand, are predisposed to infections. It is currently felt that selectins are important in initiating leukocyte rolling on vascular endothelium. This binding is a pre-requisite for firmer adherence via integrins before leukocyte emigration from blood vessels into the target tissues (47).

4.4 Do leukocyte adhesion molecules mediate renal reperfusion injury? With evidence implicating leukocytes and oxygen free-radicals in renal reperfusion injury, there has been recent interest in delineating the role of leukocyte adhesion molecules in these processes. In animal experiments administration of monoclonal antibodies to leukocyte adhesion molecules has attenuated reperfusion damage in many organs, including heart, liver and skeletal muscle (14, 48).

As compared to control rats, administration of blocking monoclonal antibodies to the CD11a & CD11b subunits prior to 60 min of renal artery occlusion was associated with a 25% lower serum creatinine and less evidence of pathologic damage 24 hours later (9). Although the increase in renal neutrophils was slightly lower in the treatment group, it was not statistically significant (9). These same antibodies virtually abolished neutrophil migration to LPS-stimulated dermis (9), but caused little change in airway leukocytes while significantly attenuated allergic airway hyperresponsiveness to antigen (49). These studies suggested that CD11/CD18 may play a role in renal IRI, but the neutrophil's dependance on this pathway for tissue migration is stimulus- and organ- specific. ICAM-1 blockade by monoclonal antibody IA29 in rats afforded more protection against renal ischemia (50). This treatment was associated with better preservation of structure and function (by nearly 50%), and attenuated the rise in the renal tissue neutrophils after induction of ischemia (50). It is important to note that neither CD11a/CD11b, ICAM-1 or control antibodies caused leukopenia. It is likely that the renal protection from anti CD11 or ICAM-1 antibodies was due to blockade in neutrophil migration, however other protective effects of the antibodies such as alteration of signal transduction or even apoptosis may have played a role (9,50).

In a different set of studies, monoclonal antibody blockade of CD11a had a small effect and blockade of ICAM-1 had a pronounced effect in affording protection against post-ischemic renal injury which occured after 30 min of renal pedicle clamping (11). Consistent with ICAM-1's role in neutrophil migration, ICAM-1 blockade significantly attenuated the rise in postischemic renal myeloperoxidase activity, (11). The functional protection that was observed by Kelly et al with the IA29 antibody to ICAM-1 was more pronounced than our study with the same antibody (50). We suspect that this may in part be because our model consisted of 60 min of renal artery clamping, thus producing a more severe tissue injury by ICAM-1 independant mechanisms. Consistent with a role of ICAM-1 in renal IRI, Kelly et al have also demonstrated that ICAM-1 deficient mice were protected from renal IRI (51). Further evidence supporting the role of ICAM-1 in renal IRI is provided by the isolated kidney perfusion model in which ICAM-1 has been shown to play a role in neutrophil retention in post-ischemic kidney (19). Taken together, these findings support the concept that leukocyte adhesion molecules may play a role in the pathogenesis of renal IRI.

Protection of renal function by anti-CD18 or ICAM-1 antibodies however, has not been universal. In a study in a rabbit model of IRI neither renal structure nor function was protected by the monoclonal 60.3 against CD18 (21). In this study, both 38 and 50 minutes of renal artery occlusion were tested (21). Also in rabbits, neither mAb R15.7 to CD18 nor R6.5 to ICAM-1 was found to protect kidney from IRI (Neuringer J & Brady HR, unpublished observations).

Selectins are felt to be required to set the stage for the CD11/CD18 - ICAM-1 interactions (47). Recent work in skeletal muscle, ear and myocardium have demonstrated that selectin blockade affords protection against postischemic damage (52-54). We evaluated renal IRI in L-selectin deficient mice generated by homologous recombination. These mice, except for small lymph nodes and thymus, are grossly normal in appearance and survive well under sterile conditions (55). Using flow cytometry, we confirmed that L-selectin on leukocytes was deficient from the batches of mice we used. Chemical peritonitis was significantly reduced in the L-selectin deficient mice, with a 47% reduction in peritoneal neutrophils 4 hours after thioglycollate injection (10). Using 30 min of bilateral renal pedicle clamping, there was more than 10-fold increase in the number of peritubular neutrophils 24 hours after ischemia was induced in normal mice. However, as compared to normal mice after renal ischemia, the L-selectin deficiency did not lead to lower number of renal tissue neutrophils. In addition, renal function was similar in L-selectin mice and controls 24 and 48 hours postischemia. This study suggests that neutrophil migration to postischemic kidney can be L-selectin independent, and that L-selectin does not mediate renal IRI. However, one must interpret results in knockout mice with caution, as aberrant adaptive mechanisms may be occurring which may not normally take place. For example other selectins may be substituting the function which is normally exerted by L-selectin. Evaluation of the role of other selectin knockouts in renal IRI is being performed. These studies and studies that block all three selectin pathways or use of glycomimetic ligand analogs may clarify the precise role of selectins in renal IRI.