CHAPTER 13 THE CLUSTER OF DIFFERENTIATION ANTIGENS
CHAPTER 13 THE CLUSTER OF DIFFERENTIATION ANTIGENS
THOMAS J. KIPPS
Definition and History
General Structure of Membrane Antigens
Type I Transmembrane Proteins (I)
Type II Transmembrane Proteins (II)
Type III Transmembrane Proteins (III)
Type IV Transmembrane Proteins (IV)
Type V Glycosyl-Phosphatidylinositol-Anchored Proteins
Tissue Distribution of Membrane Antigens
Acronyms and abbreviations that appear in this chapter include: CD, cluster of differentiation; GPI, glycosyl-phosphatidylinositol; mAbs, monoclonal antibodies; MDR1, multidrug resistance transporter protein; PI-PLC, phosphatidylinositol phospholipase C; PNH, paroxysmal nocturnal hemoglobinuria; sIg, surface immunoglobulin.
DEFINITION AND HISTORY
The advent of monoclonal antibody technology revolutionized the classification of cell surface antigens. The availability of virtually unlimited quantities of monospecific typing reagents permitted the identification and study of previously unrecognized lymphoid and myeloid-specific surface proteins. However, as the number of monoclonal antibodies (mAbs) detecting cell surface differentiation antigens grew, it became apparent that international standardization was required.
Accordingly, six international workshops have been held to exchange monoclonal antibodies to compare their ability to react with human cells and/or human cell proteins. Monoclonal antibodies that have similar patterns of reactivity with various tissues or cell types are assigned to a cluster group. An antigen that is recognized by a cluster of antibodies can be assigned a cluster of differentiation number, or CD number. If only one monoclonal antibody defines a cluster, or if all monoclonal antibodies defining a cluster originate from the same laboratory, a suffix w is added to the CD designation. The last conference, held in Kobe, Japan, November 1996, compiled the data obtained from testing hundreds of different monoclonal antibodies.1 This conference culminated in the classification of over three dozen new CD antigens.
All CD antigens defined at this and previous workshops are presented in Table 13-1, along with any common names used before a CD number was assigned in the column marked Other Names. Table 13-1 summarizes what is known about each CD antigen’s: molecular size, orientation or attachment to the plasma membrane (O), tissue distribution, and known or suspected physiology. Table 13-1 also indicates the chromosomal location of the gene encoding each CD antigen and the GenBank accession number of the reference cDNA encoding the antigen in the column marked Genetics. Finally, in the column labeled Selected References, Table 13-1 cites a few key primary papers and review articles for each CD antigen.
TABLE 13-1 CLUSTER OF DIFFERENTIATION ANTIGENS DEFINED AS OF THE SIXTH INTERNATIONAL WORKSHOP ON LEUKOCYTE TYPING
Additional information regarding the CD antigens can be found on the Internet. The accession numbers provided in the column labeled Genetics can be used to obtain the primary nucleic acid and protein sequences of each CD antigen using the GenBank website on the Internet (see http://www.ncbi.nlm.nih.gov or http://www3.ncbi.nm.nih.gov/Entrez/) or by email at: firstname.lastname@example.org. Other useful websites for analyzing protein or genomic structure are the websites for the European Bioinformatics Institute (see http://www.ebi.ad.uk), the SWISSPROT protein structure database (see http://www.expasy.ch/), the central repository for genomic mapping data from the Human Genome Initiative (see http://gdbwww.gdb.org/), or the archive of three-dimensional structures from the Brookhaven National Laboratory (see http://pdbpdb.bnl.gov/). A comprehensive list of other useful servers is provided by SWISSPROT on the Internet, at http://www.expasy.ch/alinks.htm.
GENERAL STRUCTURE OF MEMBRANE ANTIGENS
Membrane antigens are classified into different groups, depending on how they orient or anchor themselves to the plasma membrane (Fig. 13-1).
FIGURE 13-1 This figure depicts the major different types of surface proteins with respect to how they integrate into the membrane bilayer. The types of membrane protein are indicated at the top of the figure. The straight lines attached to the open circles represent the lipid bilayer. The dark black lines represent the polypeptide backbones, and the thin black pegs extending from the polypeptide backbone represent carbohydrates. At the far left is CD19, a type I transmembrane protein that passes through the membrane once and has the C-terminus (COOH) in the cytoplasm and N-terminus (NH2) outside the cell. To the immediate right is CD70, a type II single-pass transmembrane protein with the N-terminus inside the cell. To the right of this is CD20, a type III multispan protein that also is a tetraspan molecule in that it traverses the lipid bilayer four times. The tetraspan proteins have both the N-terminus and C-terminus in the cytoplasm. To the far right is CD52, a glycosyl-phosphatidylinositol (GPI)-anchored protein. The labels at the far right indicate the extracellular and intracellular membranes.
TYPE I TRANSMEMBRANE PROTEINS (I)
Type I transmembrane molecules have their COOH-termini in the cytoplasm and their NH2-termini outside the cell. Each of these molecules generally has a signal sequence at the NH2-terminus that is cleaved off after the molecule passes into the endoplasmic reticulum. Afterwards it may be glycosylated in the Golgi apparatus (if it contains glycosylation sites) and then expressed on the cell surface. These proteins commonly serve as cell surface receptors and/or ligands. Many belong to the immunoglobulin superfamily (see Chap. 83, Functions of B lymphocytes and Plasma Cells, and Chap. 84, Functions of T Lymphocytes).
Each type I protein generally has a transmembrane domain of approximately 25 hydrophobic amino acid residues followed by a cluster of basic amino acids that bind the protein to phospholipid head groups inside the surface membrane bilayer. The transmembrane domain does not contain any charged amino acid residues, such as Arg, Asn, Asp, Glu, Gln, His, or Lys, except when it associates with the transmembrane domain of another cell surface protein to form a multimeric complex. An example of this is the multimeric complex formed by the CD3 proteins and the two chains of the T-cell receptor for antigen (see Chap. 84, Functions of T Lymphocytes).
TYPE II TRANSMEMBRANE PROTEINS (II)
Type II transmembrane proteins have an opposite orientation to that of type I transmembrane proteins. The NH2-terminus is located inside the cell, and the COOH-terminus is extracellular. These proteins of-ten have uncleaved signal sequences for transmembrane domains, allowing for their cleavage and release from the cell surface. As such, these proteins may double as cell surface antigens and plasma proteins, each often having a physiologic effect on cells bearing the respective ligand.
TYPE III TRANSMEMBRANE PROTEINS (III)
Type III transmembrane proteins cross the plasma membrane more than once. Some pass through the bilayer as many as 12 times, such as the multidrug resistance transporter protein, MDR1. Because they cross the membrane multiple times, these molecules can form channels that often are used to transport ions or small molecules through the lipid bilayer. An important subgroup of type III transmembrane proteins that commonly are found on leukocytes is the tetra-span family. These proteins each pass through the surface bilayer 4 times and have both their COOH-termini and NH2-termini inside the cell. Most of the type III transmembrane proteins listed in Table 13-1 belong to this family. An example is CD20, a molecule that is postulated to form a calcium channel for B lymphocytes that is required for B-cell activation.
TYPE IV TRANSMEMBRANE PROTEINS (IV)
Type IV proteins can be distinguished from type III proteins by the presence of a water-filled transmembrane channel. None of the current CD antigens have such a membrane organization.
TYPE V GLYCOSYL-PHOSPHATIDYLINOSITOL-ANCHORED PROTEINS
Type V proteins use lipid to attach themselves to the plasma membrane. The most common attachment for extracellular proteins in this category is the glycosyl-phosphatidylinositol (GPI) anchor. The GPI anchor can be cleaved by the bacterial enzyme, phosphatidylinositol phospholipase C (PI-PLC). Release of an antigen from the cell surface by treatment with PI-PLC often is used to verify that the surface protein has a GPI anchor. However, this criterion is not absolute, as some GPI-anchored proteins may be resistant to PI-PLC.
Newly synthesized proteins destined to receive a GPI anchor each contain a secretion signal sequence at the NH2-terminus and another signal sequence at the COOH-terminus. The latter directs cleavage and subsequent appendage of a GPI anchor soon after the molecule’s biosynthesis and extrusion into the endoplasmic reticulum. This biosynthetic pathway is defective in paroxysmal nocturnal hemoglobinuria (PNH) (see Chap. 36, Paroxysmal Nocturnal Hemoglobinuria).
The site of attachment for the GPI generally precedes a hydrophobic domain of 7 to 20 amino acids that sometimes may double as an actual transmembrane domain. In this case, the molecule may exist as either of two isoforms, one attached to the membrane via a GPI anchor and another as a type I transmembrane protein. Because GPI-anchored proteins associate specifically with sphingomyelin lipids, follow a different path of transport to the cell surface than type I transmembrane proteins, are excluded from coated pits, and are not able to associate directly with intracellular proteins, a GPI isoform of a given surface protein usually has a physiology that is distinct from that of its respective type I transmembrane isoform.
TISSUE DISTRIBUTION OF MEMBRANE ANTIGENS
The tissue distribution for each CD antigen listed in Table 13-1 summarizes the work of many laboratories. For most CD antigens, however, a comprehensive analysis of the full gamut of different tissues has not been performed. Therefore, failure to list a cell type in this table for a particular CD antigen does not necessarily mean that cell type does not express that antigen. A complete review of the tissue distributions of the CD antigens is reviewed in the summary books published after each workshop.1,2 and 3 References to these books are implied, but not necessarily cited, for each CD antigen listed.
Some of surface antigens are useful for delineating the cell lineage of leukocytes. Unique assignment of a surface antigen to a particular lineage is best when the antigen is related to a unique functional property of a given cell type. The CD3 surface antigens form part of the T-cell receptor complex for antigen (see Chap. 84, Functions of T Lymphocytes). As such, CD3 is expressed exclusively by mature lymphocytes of the T cell lineage. In a similar vein, surface immunoglobulin (sIg) is a B-cell lineage specific marker. The presence of sIg on a given cell may be misleading, however, owing to the expression on diffuse cell types of Fc receptors for soluble and/or aggregated Ig. Instead expression of CD79a and CD79b, two chains that associate with sIg to form part of the B-cell surface antigen receptor, may be more precise in defining B-lineage cells (see Chap. 83, Functions of B Lymphocytes and Plasma Cells). In addition, CD20 is another antigen found exclusively on lymphocytes of the B-cell lineage.
Most CD antigens, however, are expressed at varying levels by many different cell types. Rather than the exclusive expression of a single CD antigen with a particular cell type, it is the peculiar constellation of surface antigens expressed by a given cell that helps assign it to a particular lineage or sublineage of cells. Increasingly, the resolution of many important cell subpopulations requires two or more color multiparameter flow cytometric analyses.
Kishimoto T, Kikutani H, von dem Borne AE, et al (eds): Leucocyte Typing VI, White Cell Differentiation Antigens. Garland, New York & London, 1998.
Barclay AN, Brown MH, McKnight AJ, et al: The Leucocyte Antigen Facts Book, 2nd ed. Academic, San Diego, 1997.
Schlossman SF, Boumsell L, Gilks W, et al (eds): Leucocyte Typing V. White Cell Differentiation Antigens. Oxford University Press, Oxford, 1995.
Copyright © 2001 McGraw-Hill
Ernest Beutler, Marshall A. Lichtman, Barry S. Coller, Thomas J. Kipps, and Uri Seligsohn