NCAM and lymphocyte adhesion in leucocyte adhesion deficiency (LAD) syndrome

kaleidoscope letters particular he places great weight on the paper by Karp and Hildemann ~ which we discussed in our resF. nse to Dr Karp. As made clear there, the Karp-Hildemann study is scientifically inadequate for lack of essential controls and the use of moribund animals and cannot be considered seriously as evidence with respect to the immunological competence of starfish. The only reliable echinoderm evidence that remains is the series of Coffaro studies "--~ that we cited. That s,arfish and echinoids differ fundamentaiiy in the nature of their competence to reco~maize allografts is on the face of it unlikely, and this view cannot be justified in any case, since there are no reliable starfish data. Among the contributions of Coffaro~4 were a detailed cellular study of the allograft rejection phenomenon in sea urchins, which establishes it firmly as a typical such response, mediated by the immune effector cells, the coelomocytes, and at this level it is in all ways comparable to other such responses. There is no basis whatsoever for the assertion in Dr Coopc"'s letter that echinoids are invali'. material for allograft rejection studies because of their calcite test, since it is the cellular components of the body wall, not the test, that are involved in the rejection response. The 'vexing issue' encountered in allografting experiments involving the urochordates is the apparent

difference among genera that have been studied. Species of two tunicate genera (Botryllus s and Haloo, ntbia 6) show acceptance when the grafted tissue and the host share one allel~ at the fusion locus. These systems must therefore function by detecting the shared allele. Furthermore recent studies confirm that immune memory cannot be demonstrated in the ailogenic response of Botryllus. This species does not display accelerated responses to secondary allografts nor does it show accelerated resorptions of one chimeric partner from repeated semi-allogeneic fusions --8. In contrast, species of another genus, Styela, show specific recognition of some third party allografts 9-1°, al-d also respond with hemocyte proliferation to injected allogeneic cellslL These animals appear to function by recognizing the unshared allele(s). Dr Cooper states in his letter that his work on tunicates with David Raftos confirms the existence of specific immunological memory in these animals. We refer Dr Cooper to the accompanying letter from his former coauthor, Dr Raftos, and Robert Raison, in which they express the view that the data that have been used to support adaptive immunity in invertebrates have traditionally been overinterpreted, and that convincing evidence for adaptive immunity does not exist below

NCAM and lymphocyte adhesion in leucocyte adhesion deficiency (LAD) syndrome

In this regard, we have reported on the involvement of a molecule homologous to transmembrane NCAM, namely CD56, in the adhesion of lymphocyte function associated 1 (LFA-1) negative cultured thymic cells to vascular endothelium in humans 1. The functional importance of members of the 132 i~regrin family is underlined by the clinical manifestations of leukocyte cell adhesion molecule (LeuCAM) deficiency in the leucocyte adhesion deficiency (LAD) syndrome-'. Nonetheless, in these patients the CDw49d/CD29 (VLA4) heterodimer, which belongs to the 13! integrin family, can substitute, at least in part, for 132 integ-

In their review on the mechanisms regulating leukocyte adhesion, R. Pardi, L. Inverardi and J.R. Bender {ImmunoL Today 1992, 13, 224--230) reported that GPl-linked and transinembrane isoforms of neural cell adhesion molecule (NCAM) are targeted to different surfaces of polarized epithelial cells, and speculated about the possible significance of such compartimentalization in leukoc~e adhesion and transmigration.

lmmunc %gy. Today

04

the phylogenetic level of bony fish. These are, of course, the positions developed in our review. One of the reasons we wrote our review was to emphasize the conceptual significance of lower deuterostome immune responses. The present discussion emphasizes the point that new and mechanistic data in this rewarding area are indeed much required. L. Courtney Smith Eric H. Davidson California Institute of Technology., Division of Biology 156-29 Pasadena, California 91 •25, USA.

References 1 Karp, R. and Hildema..m, W. (1976) Transplantation 22, 434-439 2 Coffaro, K. (1979) Ph.D. Thesis, University of California 3 Smith, L. and Davidson, E.H. (1992) MoL Biol. Cell. 3,403-414 4 Coffaro, K. and Hinegardner, R. (1977) Science 197, 1298-1390 5 Mukai, H. and Watanabe, H. (1975) Proc. Japan Acad. 51, 44-47 6 Fuke, M. and Nakamura, I. (1985) Biol. Bull. 169, 631-637 7 Rin!:evich, B. and Weissman, I. {1990) Eur. J. Immunol. 20, 1776-1779 8 Ibid. (1992)J. Exp. Zool. 263, 105-111 9 Raftos, D. et al. (1987) Dev. Comp. lmmunol. 11,343-351 10 Raftos, D. and Briscoe, D. (1990)

J.

/--/,or~.-/;f,,,

~ 1

0~1

f~

11 Raftos, D. and Cooper, E. (1991) ]. Exp. Zool. 260, 391--400

rins both in cell-cell adhesion and lymphocyte recirculation by binding either to fibronectin or to VCAM-1 (induced by inflammatory mediators on endothelial cells) 1,3. Interestingly, activated peripheral T cells from a LAD patient could adhere to human umbilical vein endothelial cells (HUVEC); ,his adhesion was inhibited by 30% when the lymphocytes were treated with anti-NCAM monoclonal antibodies (mAbs), implying that NCAM functions as a lymphocyte-endothelial cell adhesion molecule in LAD syndrome. This is in keeping with observations of LFA1-negative thymocytes. VLA-4 can

vo, 14 No. 2 1993

kaleidoscope letters also mediate the adhesion of LeuCAM-deficient lymphocytes to vascular endothelium but this was only evident when cytokine-activated HUVEC, expressing VCAM-1 were used. In contrast, the ability of NCAM to mediate lymphocyte-endothelial cell adhesion was prcsent in resting HUVEC, suggesting a constitutive endothelial ligand NCAM is reported to act through homophylic and heterophylic interactions in the nervous system; one ligand being heparan sulfate (HS) 4. As NCAM was absent and noninducible on HUVEC, we speculate that LeuCAM-deficient lymphocytes adhere to endothelial cells through a NCAM-HS interaction. We have been unable to detect any adhesive function for NCAM expressed by peripheral LFA-I" lymphocytes from normal donors, nor did anti-LFA-1 and anti-NCAM mAbs synergize in blocking the adhesion of lymphocytes from normal donors to vascular endothelium in vitro. This might imply that NCAM can function as a heterophylic adhesion molecule in lymphocytes when dominant adhesion structures (that is LFA-1) are missing or functionally impaired. The fact that NCAM expressed by LeuCAM-deficient lymphocytes

displayed low sialic acid content compared to that of normal peripheral lymphocytes (a pI 5.4-5.7 versus a pI 4.6-4.8) might account for the differences in the adhesive capabilities of the two cell types. The regulation of the negative charge on the cell surface, which is mainly related to sialic acid, is reported to influence cell-to-cell adhesionS; in particular, in the nervous system polysialylation of NCAM antagonizes its ability to promote adhesion s. The molecular mechanisms accounting for the differential expression of polysialylated versus low-sialylated NCAM isoforms are still undefined. One might speculate that lymphocyte expression of LFA-1 and the degree of sialylation of certain NCAM isoforms are under coordinated regulation during ontogenesis. In conclusion, we hypothesize that NCAM might provide a second mechanism, beside VLA-4/ VCAM-I interaction, for the adhesion of T cells to vascular endothelium when other adhesion systems are missing or impaired. Our findings, together with those reported by other authors in NK cells~'s, provide further evidence for a key role of NCAM in the immune system.

New meanings for an old word: adjuvanticity, cytokines and T cells

A large number of microbial components are known to be recognized by receptors which are phylogenetically ancient, and unrelated to the later-evolving antigen-specific receptors of B and T lymphocytes. These components include endotoxins, lipoarabinomannans, lipoteichoic acids, peptidoglycans, polyanions, mannans, giucans, muramic acid-derivatives, formyl peptides, double-stranded RNA, and doubtless many others that have not yet been defined. Recognition of these components rapidly provides the host with 'taxonomically' useful information aboat the organism, and standard evolutionary theory allows us to postulate that the pattern of response set in motion by a given combination of these common microbial compenems is genetically programmed to be appropriate and protective'-.

Sergio RomagnanP reviewed the evidence that development of Thelper TH2 cells is favoured when the dominant cytokine is interleukin-4 (!L-4), while development of TH1 cells is favoured when IL-4 levels are low, and IL-2 and interferon-gamma (IFN-7) are abundant. He further suggested that 'natural immunity' may play a role in this regulatory 'decision' since triggering of macrophages, natural killer (NK) cells and mast cells by microbial products leads directly or indirectly to secretion of these cytokines, and so may predetermine the TH phenotype evoked by a particular organism. This concept deserves to be expanded.

Immunology Today

95

M. RaffaeUa Zocchi Laboratory of lmmunolog3,, Scientific !nstimte San Raffaele, 1-20132 Milan, Italy.

Alessandro Poggi Laboratory of ImmunoFathology, National Institute for Cancer Research, I- 16132 Genoa, Italy.

The ;uthors are partially supported by AIRC References 1 Poggi,A. and Zocchi. M.R. !1992) Cell. Immunol. 140, 319-330 2 Arnaut, M.A. (1990)hnmunol. Rev. 114, 145-180 3 Hem!er,M.E. (!990) At:;,.,;. Rev. Immunol. 8,365-400 4 Edelman, G.M. ,lnd Crossin, K.L. (1991) Annu. Rev. Biochem. 60, 155-190 5 Springer,T. (1990) Nature 346, 425-434 6 Rutishauer, U., Grumet, M. and Edelman, G.M. (1983)]. Cell. Biol. 97, 145-152 7 Nitta, T., Yagita, H., Sato, K. and Okamura, K. (1989)J. Exp. Med. 170, 1757-1763 8 Suzuki, N., Suzuki, T. and Engleman, E.G. (1991)I. Exp. Med. 173, 1451-1461

This taxonomic information is exploited in a series of conceptually different but overlapping phases. First, most of the microbial components will trigger release of tumour necrosis factor 0~ (TNF-00 and, directly or indirectly, of IFN-7, resulting in increased expression of adhesion molecules on endothelial cells, and in extravasation and activation of phagocytes. This very rapid early phase of the response may act as a temporary 'hoiding operation' in the hours before specific T cells can be mobilized. Next, by triggering cytokine release the microbial components provide the additional signals without which T-cell-mediated immunity may not occur at all. This has been reviewed by others, and was elegantly shown in a model in which transgenic mice expressing both a foreign class I molecule in the pancreatic ~ cells,

Vol. ; 4 No. 2 1993