CELLULAR IMMUNOLOGY ARTICLE NO.
190, 83– 89 (1998)
CI981395
RAPID COMMUNICATION Experimental Autoimmune Encephalomyelitis in Intercellular Adhesion Molecule-1-Deficient Mice Elena B. Samoilova, Jennifer L. Horton, and Youhai Chen1 Institute for Human Gene Therapy and Department of Molecular and Cellular Engineering, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104 Received September 2, 1998; accepted September 9, 1998
induce autoimmune diseases. For the induction of autoimmune disease, autoreactive T cells need to become activated and extravasate at the target organs. The precise mechanisms whereby myelin-specific T cells become activated and extravasate in the central nervous system (CNS) are not clear. Intercellular adhesion molecule (ICAM)-1 is a 90kDa glycosylated cell surface protein that has been implicated in both leukocyte activation and migration. ICAM-1 is a member of the Ig superfamily and is constitutively expressed at low levels by a variety of cell types that include leukocytes, dendritic cells, vascular endothelial cells, and mucosal epithelial cells (1, 2). The levels of ICAM-1 expression are dramatically upregulated during inflammation (1, 3–5). There are two types of counter-receptors for ICAM-1: lymphocyte function-associated antigen (LFA)-1 and macrophage (Mac)-1 antigen. LFA-1 is expressed by all leukocytes whereas Mac-1 is expressed primarily by monocytes/macrophages and granulocytes (1, 2). ICAM-1:LFA-1/Mac-1 interaction has been shown to be important in leukocyte extravasation during some but not all inflammatory processes. Thus, mice deficient in ICAM-1 are crippled in their ability to mount effective inflammatory responses in the peritoneum, lung, and skin to such stimuli as bacteria, viruses, and chemicals (6, 7). ICAM-1:LFA-1 interaction has also been shown to play important roles in T cell activation, presumably by delivering a costimulatory signal to T cells (8 –12). Experimental autoimmune encephalomyelitis (EAE) is an inflammatory disease of the CNS, which is often used as an animal model for human multiple sclerosis. EAE can be induced in susceptible strains of animals by coadministration of specific myelin antigens and adjuvants. The myelin antigens provide necessary peptides required for generating the antigen-specific signal, whereas the adjuvant may be crucial for generating costimulatory signals required for T cell activation. The roles of ICAM-1 in EAE and multiple sclerosis are not clear. ICAM-1 is expressed at low levels in the CNS
Intercellular adhesion molecule (ICAM)-1, or CD54, is a member of the immunoglobulin superfamily that binds to lymphocyte function-associated antigen-1 and macrophage-1 antigen. ICAM-1:LFA-1/Mac-1 interaction may be involved in both activation and extravasation of leukocytes. To determine the roles of ICAM-1 in the development of autoimmune disease, we studied experimental autoimmune encephalomyelitis (EAE) in mice deficient in ICAM-1. We found that T cell proliferation and TH1-type cytokine production in response to myelin antigen were significantly reduced in ICAM-1-deficient mice, whereas TH2-type cytokine IL-10 was increased. Unexpectedly, EAE induced by either myelin oligodendrocyte glycoprotein or myelin basic protein was significantly enhanced in mice deficient in ICAM-1. The enhancement was evidenced primarily by the increase in disease severity, mortality, and the degree of central nervous system inflammation. The cellular composition of the inflammatory infiltrates in the central nervous system is similar in control and ICAM-1-deficient mice. These results suggest that (1) ICAM-1 is involved in the activation of autoreactive TH-1, but not TH2 cells, and (2) ICAM-1 plays an important role in down-regulating autoimmune inflammation in the central nervous system. © 1998 Academic Press
Key Words: brain; EAE/MS; autoimmunity; in vivo animal models; ICAM-1.
INTRODUCTION Although autoreactive T lymphocytes recognizing self-myelin antigens are present in normal individuals, they normally remain at the precursor stage and do not 1 To whom correspondence should be addressed at BRB-1 Room 401, Institute for Human Gene Therapy and Department of Molecular and Cellular Engineering, University of Pennsylvania School of Medicine, 422 Curie Boulevard, Philadelphia, PA 19104. Fax: (215) 573-8606. E-mail:
[email protected].
83
0008-8749/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
84
RAPID COMMUNICATION
by microvascular endothelial cells and is dramatically upregulated during inflammation (3, 4). The upregulation of ICAM-1 is restricted primarily to infiltrating inflammatory cells and microvascular endothelial cells located at the site of the inflammation (3, 4), suggesting that ICAM-1 may be involved in regulating CNS inflammation. However, studies using neutralizing antibodies to ICAM-1 generated conflicting results. While in the actively-induced EAE, treatment of Lewis rats with anti-ICAM-1 antibodies at the time of immunization inhibited T cell activation and ameliorated the disease (13–16), in the adoptive transfer model of EAE, treatment of recipient rats at the time of cell transfer either had no effect (13, 14), or, paradoxically, accelerated the disease (14). To address this ICAM-1 paradox and to circumvent the potential problems associated with the use of anti-ICAM-1 antibodies, we studied EAE in ICAM-1 deficient C57BL/6 mice. Our results revealed a previously unexpected role of ICAM-1 in the regulation of autoimmune inflammation in the CNS. MATERIALS AND METHODS Mice Four- to 6-week old C57BL/6J female mice homozygous for ICAM-1 mutation and their normal littermate controls, were purchased from Jackson Laboratory (Bar Harbor, ME). All mice were housed in the University of Pennsylvania Animal Care Facilities under pathogen-free conditions. The ICAM-1 gene mutation was created by inserting the neor cassette into the fifth exon of the ICAM-1 gene (17). Mice were screened for ICAM-1 gene mutation by reverse transcription polymerase chain reaction and Southern blot analysis (17). Induction and Clinical Evaluation of EAE For the induction of EAE, mice received a subcutaneous injection on flank of 200 mg mouse myelin basic protein (MBP) or MOG38-50 peptide in 0.1 ml PBS emulsified in an equal volume of complete Freund’s adjuvant containing 4 mg/ml of mycobacterium tuberculosis H37RA (Difco, St. Louis, MO) and an intraperitoneal injection of 200 ng pertussis toxin in 0.1 ml PBS. A second injection of pertussis toxin (200 ng per mouse) was given 24 or 48 h later. Mice were examined daily for signs of EAE and scored as follows (18): 0, no disease; 1, tail paralysis; 2, hind limb weakness; 3, hind limb paralysis; 4, hind limb plus forelimb paralysis; 5, moribund or dead. Antigens, Antibodies, Recombinant Cytokines, and ELISA MOG38-50 peptide was synthesized using Fmoc solid phase methods and purified through HPLC by Research Genetics (Huntsville, AL). MBP was purified from mouse brain as described (19). Pertussis toxin
was purchased from List Biological Laboratories (Campbell, CA). The following reagents were purchased from PharMingen (San Diego, CA): purified rat anti-mouse IL-2 (clone JES-1A12), IL-4 (clone BVD41D11), IL-10, IFN-g (clone XMG1.2) mAb; recombinant mouse IL-2, IL-4, IL-10, IFN-g. Quantitative enzymelinked immunosorbent assay (ELISA) for IL-2, IL-4, IL-10, and IFN-g was performed using paired mAbs specific for corresponding cytokines per manufacturer’s recommendations (20). Cell Culture (18) Splenocytes, 1.5 3 106 cells/well, were cultured in 0.2 ml of serum-free medium X-Vivo 20 (Biowhittacker, Walkersville, MD), containing various concentrations of MOG38-50 peptide or concanavalin (Con) A. Culture supernatants were collected 40 h later for cytokine assays. For proliferation assay, 1 mCi of [3H]thymidine was added to each culture at 72 h, and cells were harvested 16 h later. Radioactivity was counted using a flatbed beta counter (Wallac, Gaithersburg, MD). Histology Brains and spinal cords were harvested at the end of each experiment, fixed in 10% formalin and embedded in paraffin. Five-micrometer-thick paraffin sections were stained with hematoxylin and eosin as described (21). Statistical Analysis Cytokine concentration and day of onset were analyzed by analysis of variance (ANOVA). Disease score and mortality were analyzed by Mann–Whitney test. RESULTS Exacerbation of EAE in ICAM-1-Deficient Mice To study the roles of ICAM-1 in the development of EAE, we immunized ICAM-1-deficient (ICAM-12/2) C57BL/6 mice and their control littermates (ICAM11/1) with either mouse MOG38-50 peptide or MBP and monitored the disease by both physical examination and histochemistry. Figures 1A and 1B illustrate the typical disease courses of EAE induced by MOG and MBP, respectively. MOG-induced EAE developed in 90% of control C57BL/6 mice, starting approximately 9 days after immunization. The disease follows a chronic progressive course, maintaining a mean clinical score of 1.8 (Fig. 1A). By contrast, immunization with MBP induced very mild EAE in about 80% of C57BL/6 mice (Fig. 1B); the disease course was less remarkable than that induced by MOG, although the maximal disease score also reached 1.8. Unexpectedly, in ICAM-1-deficient animals, EAE induced by either MOG or MBP was dramatically enhanced. The maximal disease scores of
85
RAPID COMMUNICATION
17% in control animals (Table 1). Interestingly, the day of onset was only slightly reduced in ICAM-1deficient mice, and the rates of recovery were similar in normal and ICAM-1-deficient animals (Table 1). Histopathological Profiles of EAE in Normal and ICAM-1-Deficient Mice
FIG. 1. ICAM-1-deficient mice developed more severe EAE. Control (open square/circle, ICAM-11/1) and ICAM-1-deficient (filled square/circle, ICAM-12/2) mice, five to six mice per group, were immunized for EAE with MOG38-50 peptide (A) or MBP (B) as described under Materials and Methods. Data presented are means 6 SEM of EAE scores (healthy mice were not included in the calculation). The differences in disease scores between control and ICAM-1-deficient mice are statistically significant for both MOG- and MBP-induced EAE as determined by Mann–Whitney test (P , 0.001).
MOG- and MBP-induced EAE were 3.5 and 4.0, respectively (Fig. 1). The mortality rate was 67% in MBP-treated ICAM-1-deficient mice compared to
Consistent with clinical manifestations described above, histological examination of the CNS tissues 10 and 33 days after immunization revealed dramatic differences in the two groups. In ICAM-1-deficient mice, multiple inflammatory foci were observed in the white matter of the CNS including cerebrum, cerebellum, brain stem, and cervical, thoracic, lumbar, and sacral spinal cords. Hematoxylin and eosin staining revealed various cell types in the infiltrates. These include cells with morphological characteristics of lymphocytes, granulocytes, macrophages, microglial cells, and astrocytes. By contrast, although similar infiltrates were observed in control C57BL/6 mice, the degree of infiltration and the number of lesions in the CNS appear to be significantly reduced. Specifically, for mice tested in Fig. 1A, the average number of infiltrates per brain crosssection at day 33 was 9 in ICAM-1-deficient group, and this was reduced to 7 in the control group. Figures 2A and 2B are representative sections of the brains of ICAM-1-deficient and control C57BL/6 mice, respectively. When mice with same clinical scores were compared, no significant differences in the degree of infiltration or the cellular composition of the infiltrates were observed between control and ICAM-1-deficient mice (Samoilova and Chen, unpublished). Luxol Fast staining was also performed to visualize demyelination in the CNS (21); severe demyelination was noted in both control and ICAM-1deficient mice with disease scores of 2–5 (Samoilova and Chen, unpublished).
TABLE 1 Clinical Features of EAE in Control and ICAM-1-Deficient Mice MOG-induced EAE
MBP-induced EAE
Disease parameters
ICAM-11/1 (n 5 10)
ICAM-12/2 (n 5 9)
ICAM-11/1 (n 5 6)
ICAM-12/2 (n 5 5)
Incidence Day of onset Mortality Maximal clinical score (mean 6 SEM) Recovery rate
90% 14.6 17% 1.8 6 0.4 67%
88% 11.0 25% 3.5 6 1.0 70%
83% 14.2 17% 1.8 6 0.5 50%
80% 10.7 67% 4.0 6 0.9 50%
Note. Control (ICAM-11/1) and ICAM-1-deficient (ICAM-12/2) C57BL/6 mice were immunized for EAE with either MOG38-50 peptide or mouse MBP as described under Materials and Methods. Mice were monitored daily for signs of EAE for a total of 5 weeks. Mice that did not develop EAE were not included for calculating the day of onset, mortality, maximal clinical score, and recovery rate. The differences in maximal disease scores between normal and ICAM-12/2 mice are statistically significant for both MOG- and MBP-induced EAE as determined by ANOVA (P , 0.05). The difference in mortality between normal and ICAM-1-deficient mice is statistically significant for MBP-induced EAE as determined by Mann–Whitney test (P , 0.01).
86
RAPID COMMUNICATION
FIG. 2. Histopathological profiles of CNS. Control and ICAM-1-deficient C57BL/6 mice were immunized for EAE as in Fig. 1 and sacrificed 33 days after immunization. Brains were harvested, fixed in 10% formalin, and embedded in paraffin. Five-micrometer sections were stained with hematoxylin and eosin. (A) Cerebrum of an ICAM-1-deficient mouse with a disease score of 3. (B) Cerebrum of a control mouse with a disease score of 2. Original magnifications, 3200.
Activation and Differentiation of Autoreactive TH-1, but Not TH2 Type Cells Are Hindered in ICAM-1Deficient Animals Exacerbation of EAE in ICAM-1-deficient mice can be either due to the enhancement of myelin-specific precursor T cells to differentiate into effector T cells in the periphery or due to the enhancement of inflammatory cell function in the CNS or both. To address this issue, we first examined whether activation and differentiation of myelin-specific T cells were normal in ICAM-1-deficient animals. Splenocytes were, there-
fore, collected from both control and ICAM-1-deficient mice 10 or 33 days after immunization and tested in vitro for their cytokine production and proliferation in response to MOG38-50 peptide. As shown in Fig. 3, splenocytes of control animals proliferated vigorously in response to MOG peptide; this was significantly reduced in ICAM-1-deficient mice. To determine the cytokine profile of activated splenocytes, culture supernatants were tested by cytokine-specific ELISA. As shown in Fig. 4, splenocytes of control animals produced significant amount of TH1 cytokine IL-2 and
87
RAPID COMMUNICATION
of IL-10 produced in the ICAM-11/1 cultures was 27 6 6 pg/ml; this was increased to 125 6 19 pg/ml for ICAM-12/2 cells (Fig. 4). The amount of IL-4 in the culture was below the detectable level of the ELISA for both groups (Samoilova and Chen, unpublished). Taken together, these results strongly suggest that although activation of MOG-specific TH1 cells may be hindered in ICAM-1-deficient mice, activation of IL-10producing TH2-type cells may be enhanced in these animals. FIG. 3. MOG-specific proliferation of control and ICAM-1-deficient cells. Control and ICAM-1-deficient C57BL/6 mice were treated as in Fig. 1 and sacrificed 33 days after immunization. Splenocytes were cultured in serum-free medium with various concentrations of MOG38-50 peptide and pulsed with [3H]thymidine as described under Materials and Methods. Results are shown as means 6 SD of four mice. At MOG concentration of 5 and 25 mg/ml, the differences between control and ICAM-12/2 groups are statistically significant as determined by ANOVA (P , 0.01). The experiments were repeated three times with similar results.
IFN-g. By contrast, splenocytes from ICAM-1-deficient animals produced significantly less amount of these cytokines. Unexpectedly, ICAM-1-deficient cells produced markedly increased levels of IL-10 compared to control splenocytes. Specifically, the maximum amount
DISCUSSIONS Roles of ICAM-1 in T Cell Costimulation Activation of T cells may require a minimum of two signals: the first signal delivered by MHC–peptide complex and a second signal delivered by cell surface molecules such as B7 and ICAM-1. Our demonstration that ICAM-1-deficient mice is defective in the activation of autoreactive TH1-type cells suggests that ICAM-1 may play important roles in the inductive phase of EAE. This is consistent with earlier reports that anti-ICAM-1 antibodies blocked MBP-specific T cell activation in the Lewis rats (13, 14). The stringent dependency on ICAM-1 for the activation of autoreac-
FIG. 4. MOG-specific cytokine production of control and ICAM-1-deficient cells in vitro. Control and ICAM-1-deficient C57BL/6 mice were treated and sacrificed as in Fig. 3. Splenocytes were cultured in serum-free medium with various concentrations of MOG38-50 peptide or 2.5 mg/ml of Con A. Culture supernatants were collected 40 h later and tested for cytokines by ELISA. Results are shown as means 6 SD of four mice. The differences in MOG-induced IL-2, IL-10, and IFN-g production between control and ICAM-12/2 mice are all statistically significant as determined by ANOVA (P , 0.05). The difference in Con A-induced IL-10 production between control and ICAM-12/2 mice is also statistically significant (P , 0.01). The experiments were repeated three times with similar results.
88
RAPID COMMUNICATION
tive T cells is surprising, since the complete Freund’s adjuvant used for immunization induces severe local inflammation and may potentially upregulate a number of costimulatory molecules including B7 and IL-6. Data reported here suggest that no other costimulatory molecules may fully compensate for the ICAM-1 function during autoreactive T cell activation. Our observation that IL-10 is upregulated in ICAM1-deficient mice suggests that activation of IL-10-producing TH2 type cells may not require ICAM-1. Thus, with reduced activation of TH1 cells, TH2 cell activation may be enhanced, presumably as a result of lack of TH1 cytokine-mediated suppression. However, as Con A-induced IL-10 production was also enhanced in ICAM-12/2 mice in the absence of decreased IL-2/ IFN-g production, mechanisms other than TH1/TH2 cross-regulation may be responsible for the enhanced IL-10 production. Whether this enhanced IL-10 production plays any role in regulating EAE in ICAM12/2 mice is not clear. Although TH2 cells have been shown to induce a unique type of CNS inflammation in RAG-12/2 mice (22), this type of inflammation was not observed in our ICAM-12/2 animals (Fig. 2). Roles of ICAM-1 in Leukocyte Extravasation Extravasation of leukocytes in nonlymphoid tissues is a hallmark of inflammation. Extravasation of leukocytes is mediated by a multistep adhesion cascade consisting of primary adhesion (rolling/tethering), triggering, and strong adhesion (secondary adhesion) (23, 24). A large number of membrane and secretory molecules produced by leukocytes and endothelial cells of the vascular wall are involved in leukocyte extravasation. These include selectins (E-, P-, L-selectins) (24 –27), integrins (28 –30), and chemokines (31–33). LFA-1 and Mac-1 are both members of b2 integrin family that, through interaction with ICAM-1, promote leukocyte adhesion and transendothelial migration (2). Thus, in mice deficient in ICAM-1, inflammation induced by chemicals, bacteria, and viruses was significantly reduced in the peritoneum, lung, and skin (6, 7). However, ICAM-1 may not be required for leukocyte extravasation in all inflammatory models. For instance, Streptococcus pneumoniae-induced pulmonary inflammation is not hindered in mice deficient in ICAM-1 (7), and deficiency in ICAM-1 in MRL/MpJ lpr mice does not prevent the development of glomerulonephritis (34), although development of pulmonary inflammation is abrogated in these animals. Similarly, leukocyte extravasation in CNS during EAE is not blocked by systemic administration of anti-ICAM-1 antibodies (13, 14), which is consistent with our observation that autoimmune inflammation in the CNS is not inhibited in ICAM-1-deficient mice. Thus, ICAM-1 may not be essential for leukocyte extravasation in EAE.
Roles of ICAM-1 in Regulating Autoimmune Inflammation in the CNS Our unexpected observation that ICAM-1-deficient mice developed more severe EAE suggests that ICAM-1 may play important roles in downregulating EAE, independent of its roles in leukocyte migration or costimulation. This is consistent with report by Willenborg et al. that anti-ICAM-1 antibodies, when used at high doses, accelerate EAE in Lewis rats (14). Thus, ICAM-1 may act as a negative regulator for downregulating autoimmune inflammation in the CNS. These results are reminiscent of report by Lenschow et al. in CD28-deficient mice in that autoimmune diabetes is exacerbated in the absence of CD28-mediated costimulation (35). The precise mechanisms whereby ICAM-1 downregulates EAE need now to be investigated. Although LFA-1 and Mac-1 can both bind to ICAM-1, they are expressed by various cell types and may transmit different signals. Indeed, recent studies suggest that LFA-1 may initiate negative signals to T cells (36). Studies are under way to determine whether LFA-1, Mac-1, or any other potential ICAM-1 counterreceptor in the CNS is responsible for the observed suppressive effect of ICAM-1. It is tempting to speculate that ICAM-1 may act in a similar manner as costimulatory molecule B7 in that they may both up- and downregulate T cell functions depending on the types of counterreceptors they interact with or the state of T cell activation. In summary, we have discovered an unexpected role of ICAM-1 in the regulation of autoimmune disease in the CNS. This finding may not only be important for our understanding the basic mechanisms of autoimmunity but also aid in designing novel therapeutic strategies for the treatment of autoimmune diseases such as multiple sclerosis. ACKNOWLEDGMENTS This work was supported by grants from the National Multiple Sclerosis Society and the National Institutes of Health.
REFERENCES 1. 2. 3. 4. 5.
6. 7.
Malik, A. B., and Lo, S. K., Pharmacol. Rev. 48, 213, 1996. van de Stolpe, A., and van der Saag, P. T., J. Mol. Med. 74, 13, 1996. Dopp, J. M., Breneman, S. M., and Olschowka, J. A., J. Neuroimmunol. 54, 129, 1994. Matsuda, M., Tsukada, N., Koh, C. S., Iwahashi, T., Shimada, K., and Yanagisawa, N., Autoimmunity 19, 15, 1994. Henninger, D. D., Panes, J., Eppihimer, M., Russell, J., Gerritsen, M., Anderson, D. C., and Granger, D. N., J. Immunol. 158, 1825, 1997. Sarman, G., Shappell, S. B., Mason, E. O., Jr., Smith, C. W., and Kaplan, S. L., J. Infect. Dis. 172, 1001, 1995. Bullard, D. C., Qin, L., Lorenzo, I., Quinlin, W. M., Doyle, N. A., Bosse, R., Vestweber, D., Doerschuk, C. M., and Beaudet, A. L., J. Clin. Invest. 95, 1782, 1995.
RAPID COMMUNICATION 8.
9. 10. 11. 12. 13.
14. 15.
16. 17.
18. 19. 20. 21. 22.
Wingren, A. G., Parra, E., Varga, M., Kalland, T., Sjogren, H. O., Hedlund, G., and Dohlsten, M., Crit. Rev. Immunol. 15, 235, 1995. Lamphear, J. G., Stevens, K. R., and Rich, R. R., J. Immunol. 160, 615, 1998. Hagerty, D. T., J. Immunol. 156, 3652, 1996. Dubey, C., Croft, M., and Swain, S. L., J. Immunol. 155, 45, 1995. Gonzalo, J. A., Martinez, C., Springer, T. A., and GutierrezRamos, J. C., Int. Immunol. 7, 1691, 1995. Archelos, J. J., Jung, S., Maurer, M., Schmied, M., Lassmann, H., Tamatani, T., Miyasaka, M., Toyka, K. V., and Hartung, H. P., Ann. Neurol. 34, 145, 1993. Willenborg, D. O., Simmons, R. D., Tamatani, T., and Miyasaka, M., J. Neuroimmunol. 45, 147, 1993. Kobayashi, Y., Kawai, K., Honda, H., Tomida, S., Niimi, N., Tamatani, T., Miyasaka, M., and Yoshikai, Y., Cell. Immunol. 164, 295, 1995. Kawai, K., Kobayashi, Y., Shiratori, M., Sobue, G., Tamatani, T., Miyasaka, M., and Yoshikai, Y., Cell. Immunol. 171, 262, 1996. Sligh, J. E., Jr., Ballantyne, C. M., Rich, S. S., Hawkins, H. K., Smith, C. W., Bradley, A., and Beaudet, A. L., Proc. Natl. Acad. Sci. USA 90, 8529, 1993. Chen, Y., Kuchroo, V. K., Inobe, J.-I., Hafler, D. A., and Weiner, H. L., Science 265, 1237, 1994. Higgins, P., and Weiner, H. L., J. Immunol. 140, 440, 1988. Chen, Y., Inobe, J., Marks, R., Gonnella, P., Kuchroo, V. K., and Weiner, H. L., Nature 376, 177, 1995. Moore, G. R., Traugott, U., Farooq, M., Norton, W. T., and Raine, C. S., Lab. Invest. 51, 416, 1984. Lafaille, J. J., Keere, F. V., Hsu, A. L., Baron, J. L., Haas, W., Raine, C. S., and Tonegawa, S., J. Exp. Med. 186, 307, 1997.
23. 24.
25. 26. 27. 28. 29.
30. 31. 32. 33.
34. 35.
36.
89
Hogg, N., and Berlin, C., Immunol. Today 16, 327, 1995. Rossiter, H., van Reijsen, F., Mudde, G. C., Kalthoff, F., Bruijnzeel-Koomen, C. A., Picker, L. J., and Kupper, T. S., Eur. J. Immunol. 24, 205, 1994. Crockett-Torabi, E., and Fantone, J. C., Immunol. Res. 14, 237, 1995. Jutila, M. A., and Kurk, S., J. Immunol. 156, 289, 1996. Alon, R., Rossiter, H., Wang, X., Springer, T. A., and Kupper, T. S., J. Cell Biol. 127, 1485, 1994. Alon, R., Kassner, P. D., Carr, M. W., Finger, E. B., Hemler, M. E., and Springer, T. A., J. Cell Biol. 128, 1243, 1995. Berlin, C., Bargatze, R. F., Campbell, J. J., von Andrian, U. H., Szabo, M. C., Hasslen, S. R., Nelson, R. D., Berg, E. L., Erlandsen, S. L., and Butcher, E. C., Cell 80, 413, 1995. Hauzenberger, D., Klominek, J., Holgersson, J., Bergstrom, S. E., and Sundqvist, K. G., J. Immunol. 158, 76, 1997. Weber, C., Alon, R., Moser, B., and Springer, T. A., J. Cell Biol. 134, 1063, 1996. Dunon, D., Piali, L., and Imhof, B. A., Curr. Opin. Cell Biol. 8, 714, 1996. Murphy, W. J., Tian, Z. G., Asai, O., Funakoshi, S., Rotter, P., Henry, M., Strieter, R. M., Kunkel, S. L., Longo, D. L., and Taub, D. D., J. Immunol. 156, 2104, 1996. Lloyd, C. M., Gonzalo, J. A., Salant, D. J., Just, J., and Gutierrez-Ramos, J. C., J. Clin. Invest. 100, 963, 1997. Lenschow, D. J., Herold, K. C., Rhee, L., Patel, B., Koons, A., Qin, H. Y., Fuchs, E., Singh, B., Thompson, C. B., and Bluestone, J. A., Immunity 5, 285, 1996. Green, J. M., and Thompson, C. B., Cell. Immunol. 171, 126, 1996.