The Principal Molecules Of Allorecognition

Synopsis: Allogeneic stimulation results when specific clones of recipient T cells "see" donor major histocompatibility complex (MHC) as nonself, in conditions favorable to triggering. The molecules central to understanding allorecognition are the MHC, T cell receptor (TCR), immunoglobulin (Ig), CD4 and CD8, all of which are members of the immunoglobulin (Ig) superfamily; the adhesion molecules, and the cytokines and their receptors. Allogeneic stimulation is thus based on specific antigen recognition, plus a wide variety of permissive nonantigen specific interactions of proteins with complementary sites on other proteins.

First, a reminder. Protein structure is classified as primary, secondary, tertiary, and quarternary. The primary structure is the amino acid sequence, formed by peptide linkages between amino acids (NH-C-C(=O)-NH-C-C(=O)). Secondary structures can be either a-helices or p-pleated sheets, formed by hydrogen bonds between the NH groups in peptide linkages and the oxygens of carboxy (C=O) groups. If these bonds form internally between amino acids four residues apart, an a-helix forms. If H bonds form externally, with a remote portion of the protein, or with a different protein, the adjacent strands of amino acids (p strands) form a p-pleated sheet. Portions with no secondary structure are often termed "loops". Tertiary structure is the folding and assembly of the sheets, helices, and loops of a polypeptide into a distinct shape. For example, adjacent a-helices can form a bundle, and p-sheets can form barrels. The quarternary structure is the assembly of individual polypeptides into multimers.

Distinct regions (domains) of a protein serve distinct functions. The exons of a gene encoding a protein often echo the domain structure of the protein, with separate exons encoding each domain. Proteins that will be expressed in membranes or secreted often have leader peptides to guide their insertion into membranes. Leader peptides are encoded by leader sequences in the gene.

The Immunoglobulin (Ig) Superfamily

Ig superfamily proteins contain one or more Ig domains.17 The Ig domain is a polypeptide of about 90 amino acids (molecular weight about 12 kd) typically encoded by one exon. It contains seven p strands, designated A-G, separated by six loops, 1-6 (Fig. 1.1). The p strands align to form two antiparallel p-pleated sheets, one four-stranded (A, B, E, D) and one three-stranded (C, F, G), connected by a disulfide bond. The p strands confer the structure and the loops mediate many of the functions, especially loops 2, 3 and 6. Many Ig superfamily proteins evolved by tandem duplication of the exon for the Ig domain. They also have other domains, including membrane anchors; intracytoplasmic domains which may have signalling functions; and "sheet and helix" domains, as seen in the MHC proteins.

The Major Histocompatibility Complex

The human MHC is the human leukocyte antigen or HLA complex of genes. It spans four million base pairs (bp) on the short arm of chromosome 6 (6p).18,19 These genes encode the strong transplantation antigens, the class I and II MHC proteins. We shall examine the structure of these proteins and the organization of the genes.

THE Ig DOMAIN AND THE V VARIANT

(a) typical Ig domain

(a) typical Ig domain

(b) V or variable region type of Ig domain

Fig. 1.1. The secondary and tertiary structures of a typical immunoglobulin (Ig) domain with a variable (V) type of Ig domain. The typical Ig domain has seven p strands (a to g) separated by six loops (1 to 6). The V domain, a specialized Ig domain, is found at the N terminal of Ig light and heavy chains, all TCR chains (a, p, y , 8), CD4, CD8, and ICAM-1. In V domains, the loop 3 between strand C and strand D forms two more p strands, C' and C". The CDRs (complementarity determining regions) form the combining sites in antibodies and T-cell receptors that recognize specific antigens.

Fig. 1.1. The secondary and tertiary structures of a typical immunoglobulin (Ig) domain with a variable (V) type of Ig domain. The typical Ig domain has seven p strands (a to g) separated by six loops (1 to 6). The V domain, a specialized Ig domain, is found at the N terminal of Ig light and heavy chains, all TCR chains (a, p, y , 8), CD4, CD8, and ICAM-1. In V domains, the loop 3 between strand C and strand D forms two more p strands, C' and C". The CDRs (complementarity determining regions) form the combining sites in antibodies and T-cell receptors that recognize specific antigens.

The MHC Proteins

The MHC class I and II proteins are antigen presenting structures. They bind peptides inside cells and display them on the cell surface for T cells to "read" for signs of intracellular infection. They also play a role in the ontogeny of T cells in the thymus.

Class I is expressed on most cells and samples the peptides in the cytosol, typically for virus infection. Class II has a restricted tissue distribution, confined to specialized antigen presenting cells (APCs) (macrophages and B cells). Class II samples the peptides in the endosomal compartment of antigen presenting cells, looking for proteins taken up by endocytosis, e.g., from extracellular infectious agents. Differences between class I and II are listed in Table 1.3. They share a similar organization: a pair of Ig domains adjacent to the membrane, plus a pair of "sheet and helix" domains, plus transmembrane and intracytoplasmic portions (Fig. 1.2).

The MHC "Sheet and Helix" Domain in Class I and II Structures

The sheet and helix domains form the peptide binding groove, which is central to the whole immune response. The first half of each sheet and helix domain (about

Table 1.3. Features of class I and class IIMHC

Distribution Structure

Size of peptide presented Source of peptide

Important T cell co-receptors

Important assembly factors

Class I

diffuse—all cells single a chain non-covalently bound to p2-microglobulin

cytosol

CD4 (on helper T lymphocytes)

LMPs, TAPs, chaperone proteins

Class II

specialized— macrophages and B cells a-p heterodimer

endosomes

CD8 (on cytotoxic T lymphocytes)

invariant chain

Fig. 1.2. The structure of MHC class I and II molecules are compared. The class I molecule includes the p2-microglobulin protein (labelled p2, shown in gray). The class II molecule has the same pattern, but is formed by a dimer of an a and p chain. The domain in the a (a3) chain adjacent to the membrane is similar to p2-microglobulin. C = C terminal. Arrow shows the loop 3 region, which is the site for CD8 interaction with class I and possibly CD4 with class II (Reproduced with permission from Sigurdardottir S, Borsch C, Gustafsson K, et al: J. Immunol. 1992; 148:968-973.

Fig. 1.2. The structure of MHC class I and II molecules are compared. The class I molecule includes the p2-microglobulin protein (labelled p2, shown in gray). The class II molecule has the same pattern, but is formed by a dimer of an a and p chain. The domain in the a (a3) chain adjacent to the membrane is similar to p2-microglobulin. C = C terminal. Arrow shows the loop 3 region, which is the site for CD8 interaction with class I and possibly CD4 with class II (Reproduced with permission from Sigurdardottir S, Borsch C, Gustafsson K, et al: J. Immunol. 1992; 148:968-973.

45 amino acids) has four p strands—A, B, C, D—folded antiparallel to form a p sheet. The remainder of the domain forms a long interrupted a-helix. Two sheet and helix domains pair face-to-face: the p-pleated sheets align to form a single eight-strand p-pleated sheet which serves as the floor of the groove, and the a-helices form the walls.

The class I groove accommodates short peptides of about nine amino acids, and the class II groove accommodates longer peptides—13-25 amino acids. A concerted effort is underway to solve the rules which govern the occupation of the groove by peptides.20-22

The structures of both class I23,24 and class II25 are known. Class I and II molecules are organized differently (Fig. 1.2). The class I has a long a chain, with two sheet and helix domains (a1 and a2), one typical Ig domain (a3), a membrane anchor, and an intracytoplasmic domain. The structure is completed by p2-microglobulin, a single Ig domain, which interacts with the a3 domain. An important region of class I is loop 3 of the a3 domain which interacts with CD8.26,27

The class II molecule is assembled from a pair of nonidentical class II proteins, an a chain and a p chain. Each has a sheet and helix domain, an Ig domain, a membrane anchor, and an intracytoplasmic domain. The two sheet and helix domains (a1 and p1) form the peptide binding groove. The loop 3 region of the second domain of class II b chain forms the site of interaction with CD4.24,28

The MHC Genes

The DNA of the human MHC can be divided into four regions: class II and III regions, each 106 bp; the class I region, 1.5 x 106 bp; and the class Ib region, 0.5 x 106 bp. The organization of the HLA genes is shown in Figure 1.3.

A class I gene has eight exons: a leader sequence, two exons encoding sheet and helix domains (a1 and a2), the exon for the Ig domain (a3), an exon for the transmembrane region, and three short exons for the cytoplasmic domain. Most of the polymorphism is in selected regions of exons 2 and 3. While about eight class I genes are expressed in HLA, the most important for clinical transplantation are A and B. The p2-microglobulin gene is encoded separately on chromosome 15.

A class II gene has five or six exons: a leader sequence; exon 2 encoding the sheet-and-helix domain; exon 3 encoding the Ig domain; and two or three exons encoding the membrane anchor and cytoplasmic domain, for a total of five or six exons. Most of the class II polymorphism is in selected sites in exon 2. The expressed class II genes, in order, are two DP genes (DPB1, DPA1), one DN gene (DNA), one DO gene (DOB), two DQ genes (DQB1, DQA1); a variable number (1-3) of DRB genes, depending on the haplotype; and DRA. For transplantation the important class II genes are the DRA and B.

MHC Polymorphism

MHC class I and II genes are highly polymorphic, in selected sites, namely the bases that encode amino acids which determine the shape of the peptide binding groove. These sites create pockets and reactive groups which interact with the amino acid side chains of peptides. Polymorphism of these sites may be generated by exchange of short DNA sequences between closely related genes ("interallelic

Shape Dqa And Dqb Chains

Fig. 1.3. The organization of the major histocompatibility complex genes on chromosome 6. The new genes of particular interest include the LMP (large multifunctional protease), which encodes the components of the proteasomes and the TAP (transporters associated with antigen processing). Other genes shown include the class IIA genes (DPA, DNA, DQA, DRA), the class IIB genes (DPB, DOB, DQB, DRB), the tumor necrosis factor genes (TNFa and TNFp) and the class I genes (A, B, C). The complete DNA sequence for the human HLA complex has now been published (Nature 1999(401):921-923.)

Fig. 1.3. The organization of the major histocompatibility complex genes on chromosome 6. The new genes of particular interest include the LMP (large multifunctional protease), which encodes the components of the proteasomes and the TAP (transporters associated with antigen processing). Other genes shown include the class IIA genes (DPA, DNA, DQA, DRA), the class IIB genes (DPB, DOB, DQB, DRB), the tumor necrosis factor genes (TNFa and TNFp) and the class I genes (A, B, C). The complete DNA sequence for the human HLA complex has now been published (Nature 1999(401):921-923.)

segmental exchange").29-31 Segmental DNA exchange preserves a "cassette" of amino acids which work together to create a binding site. The MHC polymorphisms have been developed over tens of millions of years.32

Control of Groove Occupancy: Antigen Processing and Presentation12

Class I MHC molecules present peptides from endogenous proteins and class II MHC molecules present peptides from exogenous proteins,33 with exceptions.34 This difference stems from the routes of intracellular trafficking for class I and II after they are synthesized in the endoplasmic reticulum (ER).35

Newly synthesized class I heavy chains fold and assemble noncovalently with p2-microglobulin and peptide in the ER.33,36 The binding of peptides stabilizes the heavy chain—p2-microglobulin complex for transport via the Golgi apparatus to the cell surface, guided by chaperone proteins.37,38

Newly synthesized class II molecules in the ER cannot bind peptide because a portion of the invariant chain occupies the peptide groove.39 Invariant chain guides the class II from the ER through the Golgi apparatus to an acidic compartment of endosomes.40-42 Proteins taken into the cell by endocytosis enter the acidic endosome and are broken down by proteases. Invariant chain protecting the class II groove is also degraded in the endosome,43 freeing the groove to bind peptide. Peptides 13-25 amino acids in length occupy the grooves of class II molecules.43 Class II molecules may "select" peptides by protecting fragments of larger proteins from degradation.44 A larger peptide bound in the class II groove could hang out the ends and the exposed ends may be "trimmed".45 After peptide binding, class II is stable and is transported to the cell surface.

Endocytic vesicles from the cell surface sample the external environment and also receive self membrane-bound molecules. Thus DR1 molecules often contain peptides from self MHC class I and II.43-46

In B cells antigen binds to the B cell receptor and is internalized into the endosome. Such antigenic proteins are broken down into peptides, bound by class II, and exported to the cell surface to permit T cells to help the B cell to make an antibody response (see below). In addition, the endosome may receive cytosol-derived peptides transported via chaperones of the heat shock protein 70 (hsp70) family.47,48 This enables class II to present some endogenously derived peptides.34

Proteasomes and Peptide Transporters

To permit cytosolic peptides to be displayed by class I molecules, proteins from the cytosol must be broken down to short peptides, and the peptides must have access to class I grooves in the ER. This requires mechanisms to degrade proteins and to transport the peptides into the ER. Peptides are generated by proteasomes, large cytoplasmic complexes containing protease activities. Genes for two proteasome components are located in the class II region, although their function is to assist class I products.49-51 The proteasome genes are termed LMP2 and LMP7 (large multifunctional protease genes). They are polymorphic subunits of the proteasome complex which lyses cytoplasmic proteins.52-54 The transporters (called TAPs or transporters associated with antigen processing) are TAP1 and TAP2.51,55-64 The transporters are located in the membrane of the ER. Polymorphisms occur in the TAP genes but the importance of these is unknown.

Thus cytosolic proteins are digested into peptides by proteosomes, access the ER via transporters, and engage the groove. The LMP and TAP genes, like the class I heavy chain and the class II genes, are upregulated by IFN-y.49

Antigen Recognition Molecules

A specialized Ig domain—the variable or V domain—is found at the N terminal of Ig light (L) and heavy (H) chains, all TCR chains, and CD4, CD8, and ICAM-1 molecules. In V domains, loop 3 between strand C and strand D forms two more p strands C' and C'', and joins p strands C, F, and G to form a five-stranded p-sheet (C'', C', C, F, G) (Fig. 1.1).

In antigen recognition receptors (TCR, immunoglobulins), the V domains are highly variable or "hypervariable" to permit specific recognition of many different antigens. The variability is confined to loops 2, 3, and 6 (Fig. 1.1). These loops form the "complementarity determining regions" or CDRs: loop 2 forms CDR1, loop 3 CDR2 and loop 6 CDR3. The CDRs form the combining sites in antibodies and T cell receptors that recognize specific antigens. The six CDRs determine the antigenic specificity.

Immunoglobulin, B-Cell Receptors and Antibody

An antibody molecule is formed by two L chains and two H chains. Each L or H chain has a variable region, VH or VL, which is a single V domain, and a constant region, CH or CL. The L chain constant region is one Ig domain. The CH region consists of three or four Ig domains. The V regions of the L and H chains pair to form the antigen binding site: the three CDRs of VH, plus the three CDRs of VL. H chains are of five types, designated by the Greek letter for the Ig class in which they are found: a, IgA; y, IgG; |i, IgM; 6, IgD, and e, IgE. In transplantation the most relevant Ig classes are IgM and IgG.

B lymphocytes and their progeny, plasma cells, make immunoglobulin. Immunoglobulin can serve as the antigen receptor or can be released into the circulation. Each clone of B cells expresses only one type of L chain (lambda or kappa) with one type of VL region. It can make only VH, but can associate this with different CH and thus switch the class of Ig which it is making. Switching CH while retaining the same VH and the same L chain is called Ig class switching. Since the clone makes only one VL and one VH, it can make only one antigen specificity.

The B cell antigen receptor on naive B cells, i.e., never exposed to antigen, is monomeric IgM. Some B cells also have IgD receptors. After antigen exposure they undergo class switching and express IgG, IgA, or IgE on their membranes. Stimulation of the cell results in massive production of soluble antibody.

T-Cell Receptor (TCR)

The TCR (Fig. 1.4) is a dimer of nonidentical a and p chains. There is a second TCR, which is a dimer of y and 6 chains, but most allorecognition can be attributed to ap receptors. Each TCR a or p chain resembles an Ig light chain, having V and C regions, with the addition of a membrane anchor and intracytoplasmic region. The TCR V region is believed to be similar to the Ig V region.65 The V domain is hypervariable in loops 2, 3, and 6, forming CDR1, CDR2, and CDR3 in each V region of the dimer. The Va and Vp regions dimerize face to face with their CDR3s adjacent in the center and their CDR1 and 2 on the outsides. Despite the fact that the TCR structure is not solved, inferential evidence confirms this model.66 The y6 receptor may be similar.

How the TCR Engages MHC

It is likely that all six CDRs of the TCR engage the upper surface of the MHC.67,68 The outer regions of the TCR (CDRs 1 and 2) engage the a-helices of the MHC, and the central region (the CDR3s) engages the peptide. One model is that the TCR-a chain CDRs engage the a-helix of the a1 domain of class I or class II.66 The

Fig. 1.4. A schematic model for the T-cell receptor (TCR), a dimer of a, p, or y6 chains. Each chain resembles an Ig light chain, having variable (V) and constant (C) domains. The Va and Vp domains dimerize face-to-face with the CDRs 3 adjacent in the center and their CDR1 and 2 on the outsides. The outer region (CDRs 1 and 2) engages the a-helices of the MHC and the central region (the CDRs 3) engages the peptide.

Fig. 1.4. A schematic model for the T-cell receptor (TCR), a dimer of a, p, or y6 chains. Each chain resembles an Ig light chain, having variable (V) and constant (C) domains. The Va and Vp domains dimerize face-to-face with the CDRs 3 adjacent in the center and their CDR1 and 2 on the outsides. The outer region (CDRs 1 and 2) engages the a-helices of the MHC and the central region (the CDRs 3) engages the peptide.

TCR-p chain engages the other a-helix—either the a2 domain of class I or the pi domain of class II. The CDR3 region is the most variable of the CDRs in the T-cell receptor. This fits well with the notion that the CDR3s have to bind to the anti-genic peptide, whereas the CDR1 and 2 must engage the a-helices, which are much less variable. The affinity with which soluble TCR binds MHC in solution is surprisingly weak, much less than the affinity of antibody for antigen.69 This puzzle is not explained.

0 0

Post a comment