Leukocyte Adhesion : Structure and Function of Human Leukocyte β2-Integrins and their Cellular Ligands

  • View

  • Download

Embed Size (px)


  • ELK J . Bioclieni. 24.5. 715-237 (1007) 0 FEBS 1997

    Leukocyte adhesion Structure and function of human leukocyte P,-integrins and their cellular ligands

    Carl G. GAHMBERG. Martti TOLVANEN and Pekkn KOTOVUORI Department of Bioscicnces. Division of Biochemistry. University of Helsinki. Finland

    (Received 31 October l996/1S January 1997) ~ EJB 96 1613/0

    Leukocyte adhesion is of pivotal functional importance and this has resulted in extensive research and rapid develonient in the field. Leukocyte adhesion involves members of three molecular families: integrins, members of the immunoglobulin superfatnily and carbohydrate binding selectins and sialoadhe- sins. Recently, considerable structural information on leukocyte integrins and members of the immuno- globulin superfrtinily of adhesion molecules has been obtained. This fact. combined with the identification of several novel adhesion molecules. has increased our understanding o f how they function at the molecu- lar level. Furthermore, the important issue of how integrins are activated to become adhesive is rapidly advancing,.. I t i s clearly evident that the knowledge accuniulated from basic research will increasingly be applied in clinical medicine. In this review we focus on two iinportant families of adhesion molecules. the leukocyte-specific /l,-integrins and their ligands. the intercellular adhesion inolecules. Emphasis i s put on their striictLii-al/functional relationships, their mode of regulation and on novel adhesion molecules recent I y di scovered.

    K i y i w d c : integrin; CDI 1/CD18: intercellular adhesion inctlecule: adhesion: leukocyte; glycoprotein: phosphorylation.

    Le ti koc y tes continuously c i rcnlate in the body screening for the presence of altered cells. infecting microbes and foreign anti- gens. Lymphocytes enter lymphoid organs through the high en- dothelial venules (Butcher, I991 : Shiinizu et al.. 1992: Carlos and Harlan. 1994: Girard ;ind Springer. 1995: Salmi and Jalka- nen. 1997). remain there for some time. reappear i n lymphatic vessels and return t o the blood. Specific iniinime recognition is ehsential for iinmune responses. and here class I and class I1 transplantation antigens play a central role by interacting with the T cell receptor complex. But the specific immune recogni- tion is physically weak and. therefore. other nioleculnr systems have developed in order to strengthen and regulate such intel-ac- tions. Thih task is priinarily performed by the leukocyte adhesion molecules. But these molecules are also utilized for leukocyte accurnulation i n various tissues and for several other non-im- inune leukocyte- target-cell interactions.

    C ' / , / . , . c , . l " ~ / i ~ / c / i c , ~ , / ( I C . G. Gahmherg. Department of Bioscicnces, Di- v i \ i on o f Biocheiiii5tl-y. P.O. B o x 56. Viikinkaari 5. FIN-00014 Univer- sity ot Helsinki. Finland

    F((.\: +358 9 70XS9068. Ahh,.c,i,ictfio/r\. CDI la/CDI8 = (I, = LFA-I. leukocyte function-

    a\sociated antigen: CDI Ib/CDIX = u,, /I1 = Mol = Mac-l : CDI Ic/ CD18 = pISO/YS = (1, /f2: CDI Id/CDIX = u, , / f? : ICAM-I = intercellu- Iar ndhesion i i iolecule- 1 = CDS4: ICAM-2 = intercellular adhesion molecule-7 = CD102: ICAM-3 = intercellular adhesion molecule-3 = CDSO: ICAM-4 = intercellular adhesion inolrcule-4 = LW protein (Landsteiner-Wiener): CD = cluster of differentiation: VCAM-I = vas- cular cell-adhesion molecule-l = CDI 06: mAb = iiionocloniil an t i - body: LAD = leukocyre adhesion deficiency: MIDAS = metal-ion-de- pendent adhesion site: I-domain = intervening domain.

    Norc. This Review will be reprinted in EJR Kei?cw.\ 1997 which will be av;iil;ible in April I99X.

    A number of different leukocytes are known and most proba- bly more subgroups will be found concomitantly with the devel- opment i n typing procedures. 111 addition, the cells rnay represent different stages of differentiation. An individual leukocyte m 11 s t be able to interact with a number of fundamentally different cells. often modified in various ways. and sometimes with the extracellular matrix or foreign microbes. These facts may par- tially be the reason that many different leukocyte adhesion sys- tems and molecules exist. In addition i t is certainly useful for the organism to have a certain redundancy i n adhcsion molecule expression and funct ion. which could seciire vital functions cven if some adhesion functions are lost or weakened.

    Cellular immunology used to be largely phenomenal and no t inolecular. But with the development of monoclonal antibody and inolecular cloning techniques, combined with the rapid ad- vance in structural elucidation. present day immunology is cer- tainly ii molecular science, and in inany respects in the forefiont of modern biomedical research. In fact, a number of dcvelop- inents in other fields of present day biomedical research origi- nate from breakthroughs i n iinmunology. and an excellent exani- ple is leukocyte adhesion. Although much remains poorly under- stood, adhesion receptor-ligand interactions are now appreci- ated to such an extent that much can be applied t o several other areas of research.

    But research on leiikocytc adhesion is not rewarding only from a basic point of view. Undoubtedly. the large interest i n this field steins from iiuinerous potential clinical applications. Many of the iniljor diseases affecting mankind. like cardiovnscu- lar disease. stroke, chronic and iicute inflammations. cancer. ma- laria, and inany bacterial and viral infections directly involve leukocyte adhesion molecules. Therefore. i t comes as no surprise

  • 216 Gahmberg et al. (ELM J. Biochem. 245)

    Fig. 1. The integrin subfamilies. The pairs marked with thick lines have been found in leukocytes. The P,-integrin family is leukocyte-specific and its member polypeptides have not been found to associate with non- &polypeptides. The figure is adapted from Stewart et al. (1995).

    that the pharmaceutical and biotechnology industry is heavily involved in the field, and today much of the research on adhe- sion is performed outside traditional university departments.

    A number of reviews dealing with leukocyte adhesion have been written, many of them relatively recently (Hogg, 1989; Arnaout, 1990a; Patarroyo et al., 1990; Springer, 1990; Butcher, 199 1 ; Hogg et al., 1991 ; Hynes, 1992 ; Bevilacqua, 1993 ; Carlos and Harlan, 1994; Diamond and Springer, 1994; Gahmberg et al., 1995 ; Stewart et al., 1995 ; van de Stolpe and van der Saag, 1996). But the field is developing rapidly resulting in almost exponential growth of knowledge. Therefore we thought that it would be useful to collect present-day knowledge in an up-to- date review, primarily intended for biochemists and molecular immunologists. We will not try to cover all that is currently known on leukocyte adhesion. Instead, we focus on two rela- tively limited molecular families of adhesion molecules, both of which play pivotal roles: the leukocyte-specific /I,-integrins and their ligands, the intercellular adhesion molecules (ICAMs). We will only briefly deal with the sugar-binding selectins, and the VLA-4/vascular cell-adhesion molecule (VCAM) adhesion sys- tem in their relation to /I,-integrins and ICAMs. Particular em- phasis is put on molecular aspects of leukocyte adhesion includ- ing structure and signal transduction.


    Integrins are large non-covalently linked polypeptide dimers, presently the integrin family comprises more than 20 members (Hynes, 1987; Hogg, 1989; Ruoslahti, 1991). All integrins are composed of a- and /I-chains, which are type 1 membrane glyco- proteins. In some integrins the a-chains are post-translationally cleaved. The integrins are commonly divided into subfamilies according to their /$chains (Fig. 1). Some a-chains may associ- ate with different jl-chains. The leukocyte-specific integrins are named /I,-integrins or CD1 I/CD18 according to the cluster of differentiation antigen nomenclature. These molecules have a common &chain (CD18), and presently four different a-chains have been described (CDIla,b,c,d). CDI 1a/CD18 (LFA-1) is enriched in lymphocytes, CD11 b/CD18 in neutrophils, and CDI Ic/CD18 in monocytes and macrophages. CDlld/CD18 has been found very recently and therefore is less studied, but i t seems to be confined mainly to macrophages (Danilenko et al., 1995; van der Vieren et al., 1995) (Table 1). Importantly, a large proportion of the CDllb/CD18 and CD1 lc/CD18 molecules is not located at the cell surface in non-activated neutrophils, mo- nocytes or natural killer cells, but in intracellular granules, from which they can be translocated to the cell surface upon stimula-

    Fig. 2. Schematic structure of a leukocyte &integrin. The n-chains (upper) are formed by similar repeats (I-VII) including three divalent cation-binding sequences (V-VII). The =200-amino-acid I-domain is essential for ligand binding, and contains a Mg-binding site. The (I- chains are non-covalently linked to a cotnmon /{-chain. which is tinusu- ally rich in Cys residues.

    tion (Buyon et al., 1988; Vedder and Harlan, 1988; Sengelgv et al., 1993). All integrins need divalent cations for binding activ- ity. Mg is probably the most important, but peculiarly Mn has been shown to strongly promote adhesion (Gailit and Ruos- lahti, 1988; Altieri, 1991 ; Dransfield et al., 1992b). The role of Ca+ is rather elusive (see below).

    The name integrin was given by Hynes (1987) emphasizing the ability to form a link between the extracellular matrix and the cytoskeleton. The name is actually most appropriate, because it has become increasingly apparent that integrins do not only have a strictly adhesive role, but are unusually versatile and able to act in cellular signalling in both directions across the mem