CHAPTER 138 HUMAN LEUKOCYTE AND PLATELET ANTIGENS
CHAPTER 138 HUMAN LEUKOCYTE AND PLATELET ANTIGENS
KAREN A. SULLIVAN
THOMAS J. KIPPS
Human Leukocyte Antigens
Function of the Human Leukocyte Antigens
Genetics of the Major Histocompatibility Complex
Inheritance of HLA Antigens
Antiplatelet Antibodies After Transfusion
The human leukocyte histocompatibility antigens, HLA, are polymorphic cell surface glycoproteins that present antigen peptide fragments to T-cell receptors. HLA antigens are encoded by multiple, closely linked genes, located in a 4-Mb region of DNA on chromosome 6, that comprise the major histocompatibility complex (MHC) and play a central role in the regulation of immune responses. In general, the MHC genes are inherited as a single unit in simple Mendelian fashion. The products of the MHC HLA-A, HLA-B, and HLA-C genes are called class I antigens. Class I antigens are expressed on essentially all tissues in the body and present small peptide fragments to CD8+ T cells. The HLA-DR, HLA-DQ, and HLA-DP genes of the MHC encode class II antigens. Class II antigens present antigen peptide fragments to CD4+ T cells and are limited in expression primarily to B cells, monocytes, and macrophages. The HLA antigens are the principal barriers to transplantation. The degree of similarity between donor and recipient HLA antigens determines the risk of allograft rejection and, in the case of stem-cell transplantation, the risk of graft-versus-host disease (GVHD). In addition to the HLA antigens, platelets also express glycoproteins that can be recognized by autoantibodies or by antibodies made by recipients of platelet transfusions. The latter are due to platelet alloantigens that reflect polymorphism in the genes encoding major platelet glycoproteins. Immune responses to platelet alloantigens are involved in the pathogenesis of several clinical syndromes, including neonatal alloimmune thrombocytopenia, posttransfusion purpura, and refractory responses to platelet transfusion. The treatment of immune thrombocytopenia is discussed in Chap. 121. This chapter instead describes the major platelet autoantigens and outlines typing strategies used when necessary for effective platelet transfusion therapy or treatment of neonatal alloimmune thrombocytopenia.
Acronyms and abbreviations that appear in this chapter include: CTL, cytotoxic T lymphocyte; ER, endoplasmic reticulum; F-SSCP, fluorescent-based single-strand conformation polymorphism; GVHD, graft-versus-host disease; HLA, human leukocyte histocompatibility antigens; HPA-3, human platelet antigen 3; HTC, homozygous typing cells; LMP, low-molecular-weight protein; MHC, major histocompatibility complex; MLC, mixed lymphocyte culture; MLR, mixed lymphocyte reaction; PCR, polymerase chain reaction; PLT, primed lymphocyte test; RFLP, restriction fragment length polymorphism; SSCP, single-stranded conformational polymorphism; SSOP, sequence-specific oligonucleotide probe hybridization; SSP, sequence-specific primer amplification; TAP, transporter-associated-with-antigen processing; TCR, T-cell receptor; TNF, tumor necrosis factor; WMDA, World Marrow Donor Association.
HUMAN LEUKOCYTE ANTIGENS
The human leukocyte antigens, HLA, are highly polymorphic glycoproteins encoded by a cluster of genes on the short arm of chromosome 6.1 The genes encoding HLA antigens comprise the major histocompatibility complex. This is because, next to the ABO system, the MHC is the principal barrier to transplantation. As products of the genes influencing the outcome of transplanted tissue or organs, the HLA molecules are called histocompatibility antigens. The biological function of HLA molecules, however, is the presentation of antigenic peptides to T cells (see Chap. 86).
There are six major groups of HLA antigens: HLA-A, HLA-B, HLA-C, HLA-DR, HLA-DQ, and HLA-DP. These groups are divided into classes of antigens designated as class I and class II, representing the two types of HLA molecules. The HLA-A, HLA-B, and HLA-C antigens are the class I antigens. The HLA-DR, HLA-DQ, and HLA-DP antigens are the class II antigens. The class I antigens were the first and most easily identified on leukocytes by antigen and antibody reactions, i.e., serological techniques. The class II antigens were originally defined by lymphocyte proliferation assays such as the mixed lymphocyte reaction (MLR) or primed lymphocyte test (PLT) and only later were identified by serologic techniques. The antigens identified by such cellular assays were designated HLA-Dw. The HLA-Dw antigens (or specificities) are now known to be an in vitro measure of T-cell responses to the cumulative effects of the allotypic HLA-DR, HLA-DQ, and HLA-DP antigens. The class I antigens are ubiquitous in tissue distribution, being found on most nucleated cells of the body. HLA class II antigens have a more restricted distribution than class I. These antigens primarily are found on B lymphocytes, monocytes, macrophages, dendritic cells, and endothelial cells. However, class II antigens can be induced on other cell types through activation.2
Class I antigens are encoded by genes at the HLA-A, HLA-B, and HLA-C loci on chromosome 6. Class II antigens are encoded by genes at the HLA-DR, HLA-DQ, and HLA-DP loci on the chromosome. Class I and class II protein molecules are highly homologous to each other, i.e., share most of their amino acid sequences. However, the amino acid residues at small segments of the molecules vary from each other. It is these areas of variability (polymorphism) in amino acid sequences that distinguish individual HLA molecules and confer antigen specificity. Each HLA locus can code for one of many HLA antigens. Thus, each HLA locus has multiple alleles (alternative or variant forms of genes). Although there are multiple possible alleles encoding different antigens, each individual carries only one of the possible alleles at each HLA locus on each chromosome. Therefore, each person expresses only two of the many HLA-A antigens, two HLA-B antigens, and so on. A complete list of recognized antigens is provided in Table 138-1.3
TABLE 138-1 COMPLETE LISTING OF RECOGNIZED HLA SPECIFICITIES
CLASS I ANTIGENS
The HLA-A, HLA-B, and HLA-C molecules are transmembrane glycoproteins with a Mr of 56,000.4 They are heterodimers consisting of noncovalently bound protein chains; an a heavy chain (Mr = 45,000) and a b light chain (Mr = 11,000). The a heavy chain is the polymorphic glycoprotein encoded by the HLA genes. It consists of an NH2-terminal extracellular hydrophilic region, a transmembrane hydrophobic region, and an intracellular hydrophilic region, the COOH-terminus. The extracellular region of the a heavy chain consists of three domains (a1, a2, and a3) based upon its folding and disulfide bonding. Each of these domains contains approximately 90 amino acids. Antigenic specificity resides in the a1 and a2 extracellular domains (see Fig. 138-1b). The b light chain is b2-microglobulin (b2M), a nonpolymorphic globular protein encoded by a gene on chromosome 15, which stabilizes the class I molecule on the cell surface.
FIGURE 138-1 SCHEMATIC REPRESENTATION OF THE HLA-A2 MOLECUL
(a) The b strands are shown as broad arrows and the a helices as helical ribbons. A “pocket” or “groove” is formed by the a helices and a b-pleated sheet. The former form the rim of the pocket and the latter forms the floor. The groove holds the processed piece of antigen for presentation to the T-cell receptor in the orientation shown. N is the amino-terminus of the chain. (b) Schematic representation of the entire molecule showing the four domains of the molecule, 1 and 3. The a helices are present in the a1 and a2 domains. C and N represent the carboxy- and amino-termini of the two chains forming the molecule. (Both diagrams from Bjorkman and colleagues.6 Reprinted with permission from Nature 329:506. Copyright 1987 Macmillan Journals Limited.)
CLASS II ANTIGENS
The class II antigens also are transmembrane glycoproteins formed by two noncovalently bound protein chains5: an a heavy chain (Mr = 34,000) and a b light chain (Mr = 29,000). Both chains are encoded by genes in the HLA region. Class II HLA molecules, like class I, also consist of an extracellular hydrophilic NH2-terminal region, a hydrophobic transmembrane region, and an intracellular COOH-terminus region. Unlike class I molecules, however, the extracellular region of each chain contains only two domains. The two domains of the a chain are designated a1 and a2, and the two domains of the b chain are designated b1 and b2. The a chain of HLA-DR is constant for all HLA-DR molecules, while the b chain is polymorphic and determines the specificity of the molecule. Both the a and b chains of HLA-DQ and HLA-DP molecules are polymorphic, although the a chain is less so than the b chain. Polymorphism of class II molecules resides predominantly in the b1 domain of the b chain. The a2 and b2 domains show significant amino acid homology with b2-microglobulin, a3 domains of HLA class I molecules, and immunoglobulin constant region domains (see Chap. 85).
The crystallization of the HLA-A2 class I antigen and, subsequently, the HLA-DR1 class II antigen, has greatly increased our knowledge of the structure and function of the HLA molecules.6,7 The first and second a-chain domains of HLA-A2 form an interactive platform composed of a single b-pleated sheet “floor” topped by two a-helical structures with a cleft or groove between the a helices. This interactive structure is distal to the cell surface and is supported by the a3 domain of the HLA a heavy chain and the b2M molecule (Fig. 138-1). The structure of HLA-DR1 is essentially similar to class I HLA molecules. In the case of class II, however, the a1 domain of the a chain and the b1 domain of the b chain form the interactive structure of two a helices and b-pleated sheet floor. The tertiary structure of the a2 and b2 domains correspond to b2M and the class I a3 domain respectively. Polymorphic residues of the HLA class I and class II molecules are predominantly at positions within the a helices and the floor formed by the b-pleated sheet. Antigenic peptide fragments are bound within the groove of the interactive structure for presentation to the T cell receptor8,9 (see Chap. 86).
HLA antigen expression is especially high on leukocytes. Because of their easy availability, lymphocytes are used to identify HLA types. However, class I, HLA-A, -B, and -C antigens are found on most normal tissues including platelets.10 Although platelets express class I antigens, they often lack expression of some HLA-B antigens and most HLA-C antigens.11 There is evidence also that some of the HLA antigens on platelets are absorbed from the plasma.12 The class II antigens, HLA-DR, -DP, and -DQ, are more restricted in their distribution, being found primarily on B lymphocytes. Class II antigens may be found to a lesser degree on dendritic cells, endothelial cells, monocytes, and macrophages, as well as activated, but not resting, T cells.5 Expression or increased expression of class II antigens can be induced by treatment with interferon gamma.
FUNCTION OF THE HUMAN LEUKOCYTE ANTIGENS
The function of the class I and class II molecules is to bind peptide fragments that have been derived by the intracellular degradation of protein antigens and present them to T cells (see Chap. 86). T-cell receptor (TCR) recognition of MHC peptide complexes is restricted by the class of HLA molecule.13 CD4+ T cells recognize antigen peptides bound by class II molecules, and CD8+ T cells recognize antigen peptides bound by class I molecules (see Chap. 86). Both classes of HLA molecules bind peptide fragments in the groove or cleft formed by the b-pleated sheet floor and the two a helices of the molecule, i.e., the interactive platform8,9 (Fig. 138-1). The class I binding groove is closed at each end with conserved amino acid residues that bind amino and carbonyl terminal peptide residues. Because the groove is closed-ended, binding peptides are restricted in length to 8 to 11 amino acid residues. The middle portions of binding peptides can bulge out toward the T-cell receptor. The class II binding groove, on the other hand, is open at each end and binds peptides of between 15 to 24 residues in an extended conformation.
Besides presenting antigen to different T-cell subsets and binding peptides of different lengths, class I and class II molecules differ in the source or type of peptide antigen that they present.14 Class I molecules present peptides derived from endogenous proteins synthesized within the cell (cytosol), e.g., virus-encoded proteins. Class II molecules present peptides derived from exogenous antigens that are ingested by the cell, e.g., bacterial pathogens. Class I molecules are assembled in the endoplasmic reticulum (ER), and peptides must be transported into the ER to bind empty class I molecules. The genes encoding the low-molecular-weight protein (LMP) and transporter-associated-with-antigen processing (TAP) complex map within the MHC complex. The products of these genes function in antigen-peptide production (LMP) and in the transport of peptides into the ER and in the assembly of peptide-class I complexes (TAP). Class I molecules are unstable without bound peptide and can be rapidly degraded. Class II molecules are also assembled in the ER as multimers of three class II molecules with invariant chain (Ii or CD74) proteins inserted in their grooves. The invariant chain is required in folding, transport, and peptide loading of class II molecules. The invariant protein is cleaved in the endosomes of the endocytic pathway leaving a portion of Ii in the groove (CLIP) that then is exchanged for peptides of digested exogenous antigens. The nonclassical HLA class II molecules encoded by HLA-DM genes facilitate dissociation of the invariant chain peptide, CLIP, and the loading of antigenic peptides in the groove of class II molecules. Class I and class II molecules, with their bound peptides, are subsequently transported to the cell surface.
The polymorphic nature of HLA antigens appears to be related to the need to present a huge array of different antigenic peptides. For both class I and class II molecules, the polymorphic residues clustered in their binding groove determine which antigen peptide fragments can bind. These polymorphic residues are positioned outward toward the TCR or along the a helices and form pockets along the groove that have preferences for certain amino acid side chains of peptides. Such binding pockets create unique environments characteristic of the different class I and class II alleles. As such, the pocket environment determines HLA allele-specific binding motifs and dictates which antigen peptide fragments can be bound.
GENETICS OF THE MAJOR HISTOCOMPATIBILITY COMPLEX
The genes of the major histocompatibility complex encompass a DNA segment of about 4000 kb on the short arm of chromosome 6. Along with genes encoding the HLA antigens are other genes, the protein products of which also play a role in immune responses. The a and b terminology used to describe HLA genes is the same as, but should not be confused with, that used to designate the a heavy-chain and b light-chain structural domains of the HLA molecules. The arrangement of genes in the human MHC is depicted in Fig. 138-2.15
FIGURE 138-2 REPRESENTATIVE DIAGRAM OF THE GENES OF THE MAJOR HISTOCOMPATIBILITY COMPLEX ON CHROMOSOME 6
The centromere is on the left side. See text section entitled “Genetics of the Major Histocompatibility Complex.” (Adapted and simplified from Campbell and Trowsdale.15)
The MHC is divided into several regions of genes: the class I genes, class II genes, and a set of intervening genes called class III genes. The HLA class II loci are the most centromeric and encompass about 1000 kb of DNA. They are ordered sequentially, beginning with the HLA-DP genes (DPb2, DPa2, DPb1, and DPa1), followed by the HLA-DQ genes (DQb2, DQa2, DQb3, DQb1, and DQa1), and the HLA-DR genes (DRb1, DRb2, DRb3, DRb4, DRb5, DRb6, DRb7, DRb8, DRb9, and DRa). Genes for complement proteins (C4A, C4B, Bf, C2), 21-hydroxylase deficiency (called 21A and 21B), heat shock protein (HsP70), and tumor necrosis factor (TNF) separate the class II region from the class I region.15 These genes are collectively termed class III and encompass about 1000 kb of DNA. The class I genes are the most telomeric covering about 2000 kb of DNA and include, in order from class III, HLA-B, HLA-C, and HLA-A genes. Interspersed within the class I and class II regions are other genes, some involved in antigen processing (LMP), some involved in peptide transport (TAP) and loading (HLA-DMA, HLA-DMB), some that are nonfunctional (i.e., pseudogenes), and some whose functions await elucidation. The currently recognized HLA genes are presented in Table 138-2.
TABLE 138-2 NAMES FOR GENES IN THE HLA REGION
Class I antigens are encoded by a single a chain gene at each locus, i.e., HLA-A, HLA-B, and HLA-C. Class II antigens are encoded by one a chain gene and one b chain gene. Genetic polymorphism is found in both the a and b chains of HLA-DP and HLA-DQ, although to a lesser degree in the a chain. All detected serological specificities, however, appear to reside on the b chain. The HLA-DR molecules are unique in that the product of one a chain gene, which is not polymorphic, combines with any one of the products of the four functional b chain genes, i.e., DRB1, DRB3, DRB4, DRB5, to generate class II molecules that have different serological specificities. All HLA-DR antigens except DR51, DR52, and DR53 arise from polymorphism in the DRB1 gene. Most individuals express two b-chain gene products per chromosome or haplotype, one from DRB1 and another from DRB3, DRB4, or DRB5. Exceptions are individuals with the DR1, DR103, DR10, and DR8 antigens in whom only the DRB1 gene is expressed. The DR51, DR52, and DR53 serological specificities are encoded by the DRB5, DRB3, and DRB4 genes respectively. Different combinations of DRB1 alleles and DRB3, DRB4, or DRB5 are expressed depending on the haplotype. DRB3 (DR52) is found on haplotypes with the DRB1 alleles encoding DR3, DR5, DR6, DR11, DR12, DR13, DR14, DR1403, DR1404, DR17, and DR18. DRB4 (DR53) is found with the DRB1 alleles encoding DR4, DR7, and DR9. DRB5 is found with DRB1 alleles encoding DR2, DR15, and DR16.
With the presence of genes that encode complement components or molecules involved in antigen processing (LMP1 and LMP7) or peptide transport into the endoplasmic reticulum (TAP1, TAP2), the MHC is a complex of genes that encode molecules that are immunologically relevant.
It is important to distinguish polymorphic variations that are defined serologically or by cellular assay from those that are defined by molecular techniques. Serologically and cellularly defined entities are designated as specificities or antigens, while the terms gene and allele are reserved for loci defined by nucleic acid analyses.16 As new loci, genes, alleles, and antigens within the MHC are recognized, the terminology used is standardized by the World Health Organization through an HLA Nomenclature Committee. The reports of this committee describe the naming of new HLA genes, alleles, and serological specificities.3,16 The most recent report also contains complete lists of all the accepted genes and alleles, as well as the serologically and cellularly defined specificities.3 The number of MHC class I and class II gene alleles are too numerous to include here as they currently number over 900 (Table 138-3). The DNA-based, serologic-based, or cellular-based terminology is acceptable for use. Cellular terminology is used infrequently. Table 138-4 provides a comparison of serologic and DNA-based terminology.
TABLE 138-3 NUMBER OF ACCEPTED HLA ALLELES
TABLE 138-4 COMPARISON OF HLA TERMINOLOGY FOR TYPING RESOLUTION
The HLA-A and HLA-B antigens were defined, and many antigens were named before it was recognized that the MHC is a multilocus system. Instead of changing the numbers already assigned to accepted antigens, subsequent HLA-A and HLA-B specificities and alleles continue to be numbered jointly as if they were products of a single locus, e.g., A34, B35, A36, B37. For all other HLA loci, alleles and specificities are numbered consecutively within that locus, e.g., Cw1, Cw2, etc., DR1, DR2, etc., DQ1, DQ2, etc.
GENES ENCODING HUMAN LEUKOCYTE ANTIGENS
New class I region genes are designated HLA followed by a letter in alphabetical order omitting D; e.g., HLA-E, HLA-F, etc. All class II genes are designated D followed by a letter that identifies a locus that is defined by location within the class II region of the chromosome and by the similarity of its genes, e.g., HLA-DQ, HLA-DP, etc. The locus letter is followed by the letters A or B for alpha (a) or beta (b) chain genes, and the A and B are followed by a number where there are more than one a or b chain gene to a locus, e.g., DQA1, DQB1, DQA2, DQB2, etc. (see Table 138-2 and Fig. 138-2).
New alleles of HLA genes are recognized through DNA sequencing of at least several clones. Alleles are designated using the gene name, e.g. DRB1, followed by an asterik (*), followed by a four-digit number. The first two digits of the number identify any previously characterized antigen, and the latter two digits identify the allele/variant. This method was chosen to maintain as much as possible the relationship between alleles and serologic specificities. As an example, DRB1*1201 designates an allele of the protein formerly defined serologically as DR12, which itself was a serologically defined variant (split) of DR5. There are currently seven alleles that have been identified as associated with DR12. As new alleles of a gene are sequenced and accepted, they are numbered consecutively. In some cases a five-digit number is assigned to an allele. The fifth digit indicates that alleles have different DNA nucleotide sequences, but their amino acid sequences, and thus the proteins, expressed are the same. Examples of this are the DRB1*12021 and DRB1*12022 alleles.
FREQUENCIES AND RACIAL DISTRIBUTION OF HUMAN LEUKOCYTE ANTIGENS
Different races and ethnic groups can vary greatly in the frequency with which HLA antigens are found.20,21 For example, HLA-A36 and -B42 are infrequent in Caucasians as compared to other ethnic groups. Similarly, HLA-B46, although found in other populations, has a high frequency among Asians. Similar racial or ethnic differences continue to be noted using molecular techniques.21
Tissue typing for HLA antigens can be performed using a number of methods of varying degrees of sophistication and complexity, e.g., mixed lymphocyte culture (MLC), primed lymphocyte test, cytotoxic T lymphocyte (CTL) clones, 2-dimensional electrophoresis, isoelectric focusing, protein sequencing, and molecular assays. The most frequent procedures used in the clinical laboratory have been serological and cellular assays. However, with the advent of the polymerase chain reaction (PCR) (see Chap. 11), DNA-based typing is becoming common for class I and class II. It has essentially replaced cellular assays for the definition of HLA-DR, -DQ, and -DP (i.e., HLA-D) and, in some laboratories, has replaced serological testing completely.
Serological specificities (antigens) are recognized only if the serological reagents identify products encoded by accepted allelic DNA sequences. For example, the newer serological specificities of HLA-A2, corresponding to the allele sequences HLA-A*0203 and HLA-A*0210, are designated as HLA-A203 and HLA-A210. Similarly, the class II serologic specificity that is the product of the DRB1*0103 allele is designated DR103. Since new serological specificities are based on a correlation with an identified DNA sequence, the designation w for workshop or provisional characterization has been dropped. Exceptions to this rule are: HLA-C locus specificities, to avoid confusion with complement components, the Dw and DP specificities, defined by mixed lymphocyte reaction or the primed lymphocyte test, and Bw4 and Bw6, to distinguish them from other B locus specificities. Bw4 and Bw6 are public antigens (i.e., epitopes or serological specificities) defined by amino acid residues at positions 79, 80, and 83 on the class I a chain.22 Public antigens are epitopes shared by multiple HLA antigens. All HLA-B a chains carry either Bw4 or Bw6, and a few HLA-A a chains carry Bw4.
The microcytotoxicity test has been the fundamental tissue typing procedure used for defining HLA antigens for over 30 years.23 In this assay a suspension of lymphocytes is incubated with human alloantisera in a microtiter tray.24 Rabbit serum is added as a source of complement. Cell death is evaluated microscopically and determined by the uptake of a vital dye or by immunofluorescence. Antibody panels generally consist of two to four sera that recognize the same specificity. This requires that patients be tested with about 150 different reagents for class I and another 80 to 150 reagents for class II. Antisera are usually obtained from multiparous women and multiply transfused individuals or can be obtained from patients who have rejected allografts. Monoclonal HLA antibodies are used occasionally along with human alloantisera. Serology for HLA-DR and -DQ requires enrichment for B lymphocytes. In addition, many antisera contain reactivity to class I antigens as well as to HLA-DR and -DQ. Before being used as anti-class II reagents, such sera must be absorbed with cells that express only class I antigens (e.g., platelets). HLA-DP antigens also may be characterized by monoclonal antibodies, although these antigens are generally not included in clinical serological typing. Serologic definition of HLA antigens is important for patients destined to receive repeated platelet transfusions. It also is important in typing patients and donors for solid organ transplantation and in the initial investigation of families of patients desiring marrow or stem cell transplantation.
The HLA-D region of the MHC, i.e., HLA-DR, -DQ, and -DP, was initially identified by the capacity to stimulate allogeneic T cells in a mixed lymphocyte reaction.25 Initially HLA-D was thought to be a separate locus. Although no molecule could be associated with HLA-D, a large number of HLA-D specificities have been recognized (Table 138-1). It is now clear that HLA-Dw specificity is the cumulative effect of allogeneic differences of multiple class II molecules.
Mixed Lymphocyte Reaction A mixed lymphocyte reaction involves coculturing for several days the stimulator cells from one individual with the responder lymphocytes from another. Stimulator cells are prevented from proliferating by irradiation or exposure to mitomycin C. The responder cells that recognize alloantigens expressed by stimulator cells are induced to proliferate. Stimulator cells are B cells and monocytes, i.e., antigen presenting cells. T lymphocytes are responding cells. A radioactive nucleotide, usually 3H-thymidine, is added during the last 6 to 18 h of culture to measure newly synthesized DNA. The amount of radioactive thymidine incorporated into the DNA of responder cells is generally proportional to the degree of HLA-D disparity between responder and stimulator cells. The average degree of stimulation between cells from family members that share one HLA-D haplotype is roughly half that found between cells from family members that differ for both HLA-D haplotypes. Cells from family members that share both HLA-D haplotypes, e.g., HLA identical siblings, ordinarily do not stimulate each other. Similarly, the cells from nonrelated individuals who share both HLA-D haplotypes generally stimulate each other minimally if at all.
HLA-Dw specificities are identified with the use of homozygous typing cells.26 Homozygous typing cells (HTC) are obtained from progeny of consanguinous marriages who have inherited identical chromosomal HLA-D regions from each parent and are homozygous for all HLA-D region loci (DR, DQ, and DP). HTC do not stimulate cells from individuals who have the same HLA-D haplotype. However, they stimulate and respond to cells from individuals who are HLA-D heterozygous or fully disparate from them (see Table 138-5). Unfortunately, MLR testing for patients with hematological malignancies are often not successful, as leukemic cells generally are poor stimulators in the MLR with responder cells from almost any donor.
TABLE 138-5 IDENTIFICATION OF HLA-D ANTIGENS BY USE OF HOMOZYGOUS TYPING CELLS
The results of an MLR are shown in Table 138-5. Data can be expressed as (1) gross counts per minute; (2) a stimulation index, i.e., the ratio of A + Bx, to A + Ax, where A is the responding cell and Ax and Bx are irradiated stimulating cells; or (3) a relative response, or the percentage of the maximum stimulation observed when a cell is tested against cells from an unrelated panel. Controls include cultures of cell A, Ax, B, Bx, each responder cell population alone, and cells from two unrelated individuals to gauge the ability of cells A and B to respond to and stimulate allogeneic cells.
MLR response does not require prior sensitization of the responding individual and is an in vitro measure of an in vivo allograft response. Thus, MLR measures the biological effect of multiple proteins that are intimately involved in immune response. Other biological in vitro correlates of an allograft response include the PLT and CTL assays. CTL are generated in an MLR and are the effectors of a cellular allograft response.
Primed Lymphocyte Test HLA-DP antigens were classified originally using the PLT.27 PLT is a secondary MLR. Lymphocytes primed to antigen during an initial MLR will respond only to the priming antigen in a secondary MLR.
Although the CTL assay can be used to identify specificities and help to assess the degree of risk of allograft rejection or, graft-versus-host disease in stem-cell transplantation, it is a time-consuming and complex assay that normally is used in research laboratories.
The development of the polymerase chain reaction has radically changed the approach to HLA typing28,29 (see Chap. 11). A number of DNA-based methods can be used in HLA typing, e.g., sequence-specific primer amplification (SSP), sequence-specific oligonucleotide probe hybridization (SSOP), restriction fragment length polymorphism (RFLP), single stranded conformational polymorphism (SSCP), heteroduplex formation, and nucleotide sequencing. All involve the amplification from genomic DNA of selected portions of HLA genes with appropriate oligonucleotide primer pairs. Generally, exons 3 and 4 of class I genes and exon 2 of class II genes are amplified. These exons are the gene fragments encoding most of the polymorphisms of the class I and class II molecules. The most common methods currently in use for HLA typing are SSP and SSOP. In typing by SSOP, genomic DNA is isolated and amplified with oligonucleotide primer pairs specific for HLA gene fragments. The amplified DNA is then analyzed using a panel of oligonucleotide probes that hybridize with specific nucleotide sequences present in the amplified gene fragment. Oligonucleotide probes are labeled either with a radioactive isotope (usually 32P) or with a nonradioactive label such as biotin-avidin. Either amplified DNA is immobilized on a solid support, i.e., nylon membrane or plastic disk, and labeled probes are added (hybridization), or unlabeled probes are linked to the solid support, and amplified DNA, which is labeled during PCR, is added (reverse hybridization). Successful hybridization results in a detectable “dot” or band (see Figure 138-3a). Typing by SSP requires multiple independent PCR reactions. Genomic DNA is isolated and added to a panel of oligonucleotide primer pairs. Each primer pair has specificity for certain nucleotide sequences within the selected exon(s) of the HLA genes. A PCR is performed, and the resulting amplified products are analyzed by gel electrophoresis. Assignment of HLA type (i.e., genes and alleles) is based on the presence or absence of amplified product (as a band) from each reaction (see Fig. 138-3b). As with serological typing, the HLA type of the test sample is determined by the pattern with which the amplified gene fragments hybridize with the panel of different probes (SSOP) or by the pattern of products amplified in SSP.
FIGURE 138-3 EXAMPLES OF SSOP AND SSP DNA-BASED TYPING
(a) SSOP: In this example, probes specific for HLA-DPB1 alleles are immobilized on nylon strips. Genomic DNA isolated from three individuals was added to oligonucleotide primer pairs that amplify all DPB1 alleles and incorporate a label into the DNA during PCR. Visible bands indicate specific hybridization of amplified DNA fragments with selected probes. The left most column identifies controls and individual probe locations on the strip. The next column is a graphical representation of the expected (typical) intensity of hybridization with each probe. Each numbered strip represents the hybridization of amplified DNA fragments from each individual with specific probes. The pattern of hybridization for the individual tested on strip 1 indicates the presence of the DPB1*0201 and DPB1*1001 alleles. The patterns of hybridization on strips 2 and 3 indicate the presence of the DPB1*2201 and DPB1*3101 alleles and the DPB1*0301 and DPB1*0401 alleles respectively. (b) SSP: This is an example of low-resolution SSP typing on one individual for class II. Isolated genomic DNA was added to a panel of oligonucleotide primer pairs specific for DRB1, DRB3, DRB4, DRB5, or DQB1 alleles. Each mixture also contains a primer pair that amplifies a non-HLA DNA nucelotide sequence that is common to all individuals as a positive control for amplification. Following PCR, each reaction mixture was analyzed by gel electrophoresis. The various bands seen at each position and moving in the direction of electrophoresis are:
Unamplified genomic DNA (>1000 bp)
Positive amplification control (~700 bp)—not always seen when there is a specific product
Specific product (~75–350 bp)—Seen only in those wells that amplify the sample’s alleles, i.e., 1G, 2H, 2D, 3A, 4G, 4C
Primer dimers (usually <75 bp)—result from primer excess, and the primers form a dimer with each other. Not always seen, but some may be very bright.
Position 1H contains a negative control consisting of appropriate primer pairs but no added genomic DNA. The bands to the left of position 1H is a marker consisting of different size DNA fragments to help determine base pair product size. Analysis of the pattern of amplified specific products identify the class II type of this individual as HLA-DRB1*01, DRB1*04; DRB4*01; DQB1*03, DQB1*05, or, in serological terms, HLA-DR1, DR4; DR53; DQ3, DQ5.
DNA-based typing of HLA is generally performed at two levels, the first using reagents (probes or primer pairs) that detect all alleles of an HLA gene (low resolution), the second using reagents with specificity for selected alleles (high resolution). Low-resolution typing identifies the HLA gene at the serological or antigen level, e.g., HLA-A*02, HLA-DRB1*01, etc. High-resolution typing identifies specific alleles, e.g., HLA-A*0201, DRB1*0103, etc. A third level is called intermediate resolution. In intermediate resolution, more than one allele of an HLA gene could be the correct one, e.g., DRB1*0302/0303/0304/0309 (see Table 138-4). If primers or probes are unavailable, and if it is necessary to clarify the specific allele, then nucleotide sequencing of amplified product can be done. Although nucleotide sequencing is thought by many to be the definitive method, it is costly, time consuming and, as yet, a more common method in research laboratories.
Molecular typing has revealed greater polymorphism in the genes encoding HLA antigens than was previously detected.3 For example, there are multiple alleles at HLA loci each encoding a molecule with the same serologic specificity, e.g., the HLA-A2 serologic specificity is associated with more than 20 HLA-A2 alleles. Molecular typing has also elucidated the origin of the class II public antigens, HLA-DR51, -DR52, and -DR53. Unlike the Bw4 and Bw6 class I public antigens that are due to amino acid substitutions on the class I a chain, separate DRB genes encode the HLA-DR51, -DR52, and -DR53 antigens. Furthermore, DNA-based typing has helped to reveal the relationship of HLA-DR, -DQ, and -DP to HLA-Dw specificities. For example, HLA-DR4 appeared to be one antigen, even though it was associated with several different HLA-Dw specificities. To date, 32 alleles of DR4 have been identified by molecular analysis, and each of the Dw specificities identified for DR4 is associated with a specific DR4 allele.
The use of molecular testing has a number of advantages over serologic typing. First, DNA-based typing does not require the isolation of viable lymphocyte populations but can be done using any nucleated cell source. Second, DNA-based assays have increased accuracy and specificity. HLA antigens have a high degree of homogeneity, and alloantisera produced against them can be cross-reactive. This cross-reactivity can lead to inconsistent assignment of individual specificities.
INHERITANCE OF HLA ANTIGENS
The genes of the MHC demonstrate more polymorphism than any other genetic system, i.e., multiple alleles exist for each locus. Each individual, however, has one allele for each locus per chromosome and, therefore, encodes two HLA antigens per locus. Additionally, the antigens at each HLA locus are codominant, i.e., each is expressed independently of the other. The identification of each HLA antigen of an individual is called a phenotype. Two unrelated individuals who express the same HLA antigens are HLA-phenotype identical.
Since HLA genes are closely linked on chromosome 6, recombination within the MHC is rare (less than or equal to 1 percent), and a complete set of HLA genes is usually inherited from each parent as a unit. The genes inherited from one parent are referred to as a haplotype. Maternal and paternal haplotypes can be identified through family studies. Identification of both the paternal and the maternal haplotypes in an individual provides a genotype. Siblings who inherit the same haplotypes are termed HLA-identical. Those who inherit the same haplotype from one parent but a different haplotype from the other parent are called haplo-identical. Lastly, siblings who inherit different haplotypes from each parent are HLA nonidentical. In general, family studies consist of testing for the HLA-A, -B, -C, -DR, and -DQ antigens to identify haplotypes and to rule out genetic recombination within the MHC complex. Since HLA genes are inherited together on a single chromosome, there are four possible combinations of maternal and paternal haplotypes, provided that there is no meiotic recombination between HLA genes (see Fig. 138-4). Thus, there is a 1 in 4 (or 25 percent) chance that two siblings will be HLA-identical, a 2 in 4 (or 50 percent) chance that two siblings will be HLA haplo-identical and, a 1 in 4 (or 25 percent) chance that two siblings will be HLA nonidentical. All progeny are haplo-identical with their parents unless recombination has occurred.
FIGURE 138-4 Representation of the inheritance of HLA. Each of the four parental chromosomes (haplotypes) is coded by a letter: a and b represent the paternal haplotypes; c and d represent the maternal haplotypes. Each child inherits one paternal haplotype (a or b) and one maternal haplotype (c or d).
Because HLA is so highly polymorphic, the chance that two unrelated individuals would be HLA identical could be astronomical. However, the situation is somewhat alleviated because the HLA system displays a phenomenon known as linkage disequilibrium. That is, certain HLA alleles are inherited together on the same chromosome more often than would be predicted if HLA loci were at equilibrium. At equilibrium the frequency of an allele at one locus is independent of the frequencies of alleles at linked loci. For example, the gene frequency of HLA-A1 in North American Caucasians is 0.138 and that of HLA-B8 is 0.09. If there were no preferential association between HLA-A1 and HLA-B8, then the frequency of the HLA-A1, B8 haplotype, predicted by equilibrium, should be 0.0124 (0.138 × 0.09 = 0.0124). However, population studies show that the actual frequency of the HLA-A1, B8 haplotype in this particular population is greater than that predicted by equilibrium, i.e., 0.0609. The degree of linkage disequilibrium is defined as the observed frequency minus its expected frequency, 0.0485 in this example. Although the particular alleles that are found in linkage disequilibria differ for various racial groups, all racial groups display significant disequilibria.
The HLA genes and antigens of the major histocompatibility complex play a central role in transplantation, in the regulation of immune responses, and in susceptibility to a variety of diseases. The most common application of HLA, however, is in the field of transplantation. In both solid organ and stem-cell transplantation, allografts from HLA-identical sibling donors have a significantly greater chance for survival than grafts from nonmatched family, or unrelated, donors. For solid organ transplantation, a living donor is not always available or even feasible, e.g., for heart transplantation. HLA matching is restricted to the HLA-A, -B, and -DR loci, and the level of typing is at the serologic (antigen) level or low resolution for DNA-based methods. Initially, a high degree of HLA match between potential recipients and unrelated donors was sought. However, with the increasing need for more organs and the advent of newer immunosuppressive therapies, such as cyclosporine, the level of HLA matching has declined, and some question whether HLA matching is necessary. Nonetheless, even with the newer immunosuppressive drugs, the long-term survival of allografts from HLA-matched donors exceeds that of HLA-mismatched donors.30 Additionally, recipients of HLA-matched organs have fewer rejection episodes and fewer complications and may require less immunosuppression.
Blood or marrow stem-cell transplantation engenders problems other than allograft survival. Lymphocytes within the graft may recognize host cells as foreign. Without the ability to mount a response to such host reactive cells, patients receiving stem-cell allografts are prone to graft-versus-host disease31,32; see Chap. 19. With HLA-identical sibling donors, disease-free survival of 80 to 90 percent can be achieved for some malignant and nonmalignant hematological disorders.32 However, less than 30 percent of individuals have an HLA-identical sibling. Thus, alternative donors, such as phenotypically matched unrelated volunteers and partially matched family members, must be considered. However, the risks and incidence of graft failure and GVHD are higher than seen with HLA-identical siblings and increase with increased HLA disparity. Fortunately, molecular typing has improved the accuracy of matching unrelated donors, allowing for improved outcomes.
The HLA criteria for the selection of an appropriate stem-cell donor varies from transplant center to transplant center and, within a center, also will depend on the transplant protocol. Within families, typing for HLA-A, -B, -DR, and -DQ at the serologic or DNA-based low-resolution level is generally sufficient to identify all haplotypes, potential recombinants, and HLA-identical pairs. In those circumstances where the haplotypes are unclear, DNA-based, allele-level typing is recommended. For unrelated volunteer donors, allele-level molecular typing of HLA-A, -B, -C, -DR, and -DQ provides the best opportunity for a successful outcome and reduced GVHD. The World Marrow Donor Association (WMDA) published guidelines for the extent of HLA typing recommended for transplant centers and donor registries that participate in the exchange of stem cells for allogeneic transplantation.33 Briefly, the WMDA recommends typing the HLA-A, -B, and -DR loci using DNA-based testing at the low-resolution level at a minimum.
Platelets express antigens that can be recognized by autoantibodies or by antibodies made by recipients of platelet transfusions. HLA antigens are present on platelets, as judged by absorption, fluorescence, and complement fixation. In addition, platelets also possess platelet-specific antigens that are unrelated to erythrocyte or leukocyte isoantigens (see Chap. 14).
Platelet antigens can be targeted by autoantibodies, resulting in immune thrombocytopenia. A dominant platelet antigen recognized by the autoantibodies by many patients with autoimmune thrombocytopenia is the platelet glycoprotein GIIb/IIIa (otherwise called aIIbb3 or CD41/CD61), although other platelet glycoproteins also may be targeted by autoantibodies.34,35 This condition is discussed in Chap. 119.
Platelet alloantigens, also referred to as platelet isoantigens, are substances that induce the production of alloantibodies when platelets bearing such antigens are infused into patients who lack the specific alloantigen.36 Immune responses to platelet alloantigens are involved in the pathogenesis of several clinical syndromes including neonatal alloimmune thrombocytopenia, posttransfusion purpura, and refractory responses to platelet transfusion.37 In addition, immune thrombocytopenia can be an unusual complication of a type of graft-versus-host disease in which donor lymphocytes make alloantibodies specific for the platelets produced by the recipient of an organ allograft.38
Patients can lack a particular platelet-associated antigen altogether because they have defective alleles of the gene encoding the antigen (see Chap. 121). Such patients can make antibodies against platelets of virtually all donors that bear the platelet-associated antigen. For example, patients with Bernard-Soulier syndrome, who lack platelet GPIb-V-IX, or patients with Glanzmann thrombasthenia, who lack expression of GPIIb (CD41) and GPIlla (CD61), can be induced to make broadly reactive antiplatelet antibodies39,40 (see Chap. 121). Also, several percent of Japanese and approximately 0.3 percent of Caucasians are deficient in CD36, one of the major platelet glycoproteins of platelets that also is known as GPIV.41 Because these patients lack a platelet antigen, they can develop antiplatelet antibodies specific for the deficient platelet protein after receiving transfusions of platelets from normal donors or after pregnancy.
More commonly, platelet-specific alloantigens result from genetic polymorphism in genes encoding functional platelet proteins.36,42 These alloantigens first were defined by antiplatelet antibodies discovered in the sera of multiparous females who gave birth to infants with neonatal thrombocytopenia. Many of these subsequently were found to recognize allotypic determinants of platelet-associated membrane glycoproteins, such as GPIIb/IIIa (CD41/CD61). Each of these allotypic determinants may be generated by only a single amino acid substitution in a major platelet-associated glycoprotein (Table 138-6).43 However, it is possible that glycosylation may contribute to or influence the expression of certain HPA epitopes, such as those associated with human platelet antigen 3 (HPA-3).44 In any case, these amino acid substitutions generally do not appear to affect the function of platelets in vitro. However, it is conceivable that the genetic polymorphism in platelet glycoproteins may be associated with more subtle differences in platelet physiology that can contribute to the relative risk for thrombosis and/or atherosclerosis.45,46 and 47
TABLE 138-6 HUMAN PLATELET ANTIGENS
A nomenclature for the human platelet alloantigens was adopted to replace the old complex “classical” nomenclatures that previously were developed independently in laboratories throughout the world (see Table 138-6). There are at least 10 HPA that have been defined at the molecular level. Each has two alleles, designated by the suffix a or b. These alleles are expressed by platelets codominantly (see Table 138-6). The a allele generally is the more prevalent allele. Allelic frequencies, however, vary between different racial groups48,49,50,51 and 52 (Table 138-6). HPA-lb, for example, is expressed on platelets of approximately 15 percent of persons of European ancestry but of less than 1 percent of those of Asian ancestry.
In addition to the recognized HPA, there are several other platelet alloantigens that can be recognized by antiplatelet alloantibodies and account for neonatal alloimmune thrombocytopenia. Two, designated Groa and Oea, are associated with GPIIIa (CD61).53,54 Another potential alloantigen, designated Vaa, is associated with the GPIIb/IIIa complex (CD41/CD61).55 Two others, designated Laa and Lya, are localized to the GPIb-IX-V complex.56,57 Another alloantigen results from genetic polymorphism in the gene encoding CD109 (see Chap. 14), accounting for the Gov allantigen system.58,59 The less common allele encoding any one of these alloantigens may present at very low gene frequencies.
HPA alleles HPA-4b, -6b, -7b, -8b, -9b, and -10b also are present at gene frequencies of less than 0.1 percent, and thus are designated private alleles (Table 138-6). Some of these alleles, such as HPA-7b or HPA-8b,53 are extremely rare. For this reason, the alloantigens encoded by such alleles are not likely to account for most cases of post-transfusion purpura but can be found on isolated cases of neonatal alloimmune thrombocytopenia in selected families. Alleles present at gene frequencies of greater than 2 percent within the population, on the other hand, are designated as public alleles. These alleles are more likely to encode the alloantigens that are targeted by antiplatelet alloantibodies.
The frequency of patients that express the more common HPA allele is greater than the gene frequency for that allele in the population. Using the classical Hardy-Weinberg equation (a2 +2(ab) + b2 = 1), the proportion of individuals that are homozygous for the a allele (a/a) is the squared product of the gene frequency for a in the population (or a2). Similarly, the proportion of cases that are homozygous for b is the squared product of the gene frequency for b (or b2). The proportion of cases that are heterozygous (either a/b or b/a) is two times the product of the gene frequency for a times that for b. Because there are two alleles, the sum of all these products should equal 100 percent (or 1). Using such considerations for example, it is evident that persons of European ancestry will have a (0.85)2, or 72 percent, chance of being homozygous for HPA-1a, a 26 percent chance of being heterozygous for HPA-1a/HPA-1b, but only a (0.15)2, or 2 percent, chance of being homozygous for HPA-1b (see Table 138-6). As such, even though the gene frequency of HPA-1a is 85 percent, 98 percent of the population have at least one HPA-1a allele and will not make alloantibodies against the HPA-1a epitope. Similarly, it should be evident that it is highly unusual for anyone to be homozygous for HPA private antigens. As such, it is highly improbable that alloantibodies against HPA-4a, -6a, -7a, -8a, -9a, and -10a will account for immune thrombocytopenia. Conversely, most persons will lack the b allele of HPA private antigens and thus can potentially develop alloantibodies to HPA-4b, -6b, -7b, -8b, -9b, and -10b alloantigens.
ANTIPLATELET ANTIBODIES AFTER TRANSFUSION
Patients who are homozygous for a given allele can develop antiplatelet alloantibodies after receiving transfusions of platelets or cells that express the other allele.60 Generally, a few transfusions of unmatched platelets can be given to patients without adverse effects. However, the risk for developing antiplatelet antibodies increases with each successive platelet transfusion. Patients receiving multiple transfusions of red blood cells also may develop antiplatelet or anti-HLA antibodies against leukocytes contaminating the red blood cell preparation.
After repeated platelet transfusions, the patient may become sensitized, possibly developing antibodies that shorten the lifespan of the transfused cells. The number of unmatched transfusions required to elicit such antibodies is variable. Some patients never make antibodies, while others become refractory after receiving only a few platelet transfusions. Others, often multiparous women, may have clinically significant antiplatelet antibodies prior to receiving any platelet transfusions. About half of all patients receiving repetitive platelet transfusions eventually produce antibodies against antigens on the transfused platelets. These antibodies may be directed against platelet-specific alloantigens, HLA antigens, or both.
Antibodies against some HPA-allelic determinants can inhibit platelet function. Anti-HPA-1 alloantibodies, for example, can inhibit clot retraction and platelet aggregation, presumably because they block the binding of GIIb/IIIa (aIIbb3) to fibrinogen. Moreover, anti-HPA-4 alloantibodies can completely inhibit aggregation of HPA-4 platelets that are homozygous for allele recognized by the alloantibodies. This is because the epitope recognized by anti-HPA-4 alloantibodies is found on 3 at amino acid position 143 that is in close proximity to the RGD-binding domain of the aIIbb3 integrin. Other anti-HPA-alloantibodies, such as alloantibodies specific for HPA-3 on the other hand, may not significantly interfere with platelet function but nonetheless can cause Fc-mediated platelet destruction and immune thrombocytopenia.
If the patient does develop anti-HLA antibodies that shorten the survival of unmatched transfused platelets, HLA-matched platelets may be more effective for subsequent transfusions. The patient and prospective platelet donors generally are matched for serological identity at the HLA-A and -B loci. Platelets derived from such HLA-matched donors generally survive longer in highly immunized patients than those received from random donors.
In some cases, it is necessary to type the platelet alloantigens of donors to provide effective platelet-transfusion therapy.61 Different techniques for phenotyping are well established and easy to perform, but they rely on the availability of antisera.62 The molecular genetic basis for the clinically most relevant alloantigens has been elucidated, thus facilitating widespread platelet alloantigen typing.54
Typing for HPA status can be achieved using the polymerase chain reaction (see Chap. 11), with subsequent analysis of the amplified gene-fragment using restriction enzymes, sequence-specific primers, dot-blot hybridization, or a fluorescent-based single-strand conformation polymorphism (F-SSCP) technique.63,64,65,66 and 67 These techniques have proved to be highly useful in identifying the platelet genotype of fetuses at risk for neonatal alloimmune thrombocytopenia,68 establishing the diagnosis of posttransfusion purpura, or in identifying causes of refractory responses to platelet transfusions. Molecular methods for detecting HPA status also should facilitate clinical studies on the association of certain HPA alleles with relative risk for thrombosis46 or response to therapy for immune thrombocytopenia.69
Breuning MH, Vanden Berg-Loonen EM, Bernini LF: Localization of HLA on the short arm of chromosome 6. Hum Genet 37:131, 1977.
Berrih S, ArenzanaSeidedos F, Cohen S, et al: Interferon-gamma modulates HLA class II antigen expression on cultured human thymic epithelial cells. J Immunol 135:1165, 1985.
Bodmer JB, Marsh SGE, Albert ED, et al: Nomenclature for factors of the HLA system, 1998. Tissue Antigens 53:407, 1999.
Thorsby E: Structure and function of HLA molecules. Transplant Proc 19:29, 1987.
Trowsdale J: Genetics and polymorphism: class II antigens. Br Med Bull 43:15, 1987.
Bjorkman PJ, Saper MA, Samraoui B, et al: Structure of the human class I histocompatibility antigen, HLA-A2. Nature 329:506, 1987.
Brown JH, Jardetzky TS, Gorga JC, et al: Three-dimensional structure of the human class II histocompatibility antigen HLA-DR1. Nature 364:33, 1993.
Borkman PJ, Saper MA, Samraoui B, Bennet WS, Strominger JL, Wiley DC: The foreign antigen binding site and T cell recognition regions of class I histocompatibility antigens. Nature 329: 512, 1987.
Stern LJ, Brown JH, Jardetzky TS, et al: Crystal structure of the human class II MHC protein HLA-DR-1 complexed with an influenza virus peptide. Nature 368:215, 1994.
Kostyu DD, Amos DB: The HLA complex; Genetic polymorphism and disease susceptibility, in The Metabolic Basis of Inherited Disease, 6th ed, edited by CR Scriver, AL Beaudet, WS Sly, D Valle, chap 4, p 225. McGraw-Hill, New York, 1989.
Mueller-Eckhardt B, Hauck M, Kayser W: HLA-C antigens on platelets. Tissue Antigens 16:91, 1980.
Lalezari P, Driscoll AM: Ability of thrombocytes to acquire HLA specificity from plasma. Blood 59:167, 1982.
Zinkernagel, RM, Doherty, PC: Restriction of in vitro T cell-mediated cytotoxicity in lymphocytic choriomeningitis within a syngeneic or semiallogeneic system. Nature, 248:701, 1974.
Mak, TW, Simard JJL: Handbook of Immune Response Genes. Plenum, New York, 1998.
Campbell RD, Trowsdale J: A map of the human major histocompatibility complex. Immunol Today 18:43, 1997.
Bodmer JB, Marsh SGE, Albert ED, et al: Nomenclature for factors of the HLA system. Tissue Antigens 39:161, 1992.
Marsh SGE: Nomenclature for factors of the HLA system, update January, 1999. Tissue Antigens 54:106, 1999.
Marsh SGE: Nomenclature for factors of the HLA system, update February, 1999. Tissue Antigens 54:108, 1999.
Marsh SGE: Nomenclature for factors of the HLA system, update March, 1999. Tissue Antigens 54:110, 1999.
Baur MP, Danilors JA: Population analysis of HLA-A, B, C, DR and other genetic markers, in Histocompatibility Testing 1980, edited by PA Terasaki, p 955. UCLA Tissue Typing Laboratory, Los Angeles, 1981.
Gjertson DW, Su-Hui, L: HLA-A/B and -DRB1/DQB1 allele level haplotype frequencies, in HLA 1998, edited by DW Gjertson, PI Teresaki, chap 8, p 365. American Society for Histocompatibility and Immunogenetics, Lenexa, 1998.
Parham P: Histocompatibility typing-Mac is back in town. Immunol Today 9:127, 1988.
Sullivan K and Amos DB: The HLA system and its detection, in Manual of Clinical Laboratory Immunology, 3rd ed, edited by NR Rose, H Friedman, JL Fahey, chap 130, p 835. American Society for Microbiology, Washington, DC, 1986.
Terasaki P, McClelland JD: Microdroplet assay of human serum cytokines. Nature (London) 204:998, 1964.
Dupont B, Hansen JA, Yunis EJ: Human mixed lymphocyte culture reaction: genetics, specificity and biological implications. Adv Immunol 23:108, 1976.
Jorgensen F, Lamm LU, Kismeyer-Nielsen F: Mixed lymphocyte cultures with inbred individuals: an approach to MLR typing. Tissue Antigens 3:323, 1973.
Shaw SA, Johnson AH, Shearer GM: Evidence for a new segregant series of B cell antigens that are encoded in the HLA-D region and that stimulate secondary allogeneic proliferative and cytotoxic responses. J Exper Med 152:565, 1980.
Saiki RK, Gelfand DH, Stoffel S: Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239:487, 1988.
Erlich HA, Bugawan TL: HLA class II gene polymorphism: DNA typing, evolution and relationship to disease susceptibility, in PCR Technology: Principles and Applications for DNA Amplification, edited by HA Erlich, p 193. Stockton, New York, 1989.
Dyer, PA, Taylor C, Martin S: HLA and clinical solid organ transplantation: recent publications and novel approaches, in HLA 1998, edited by DW Gjertson, PI Terasaki, chap 2, p 13. American Society for Histocompatibility and Immunogenetics, Lenexa, 1998.
Martin PJ, Gooley T, Anasetti C, Petersdorf EW, Hanson J: HLAs and risk of acute graft-vs-host disease after marrow transplantation from an HLA-identical sibling. Biol Blood Marrow Transplant 4:128, 1998.
Petersdorf EW, Gooley TA, Anasetti C, et al: Optimizing outcome after unrelated marrow transplantation by comprehensive matching of HLA class I and class II alleles in the donor and recipient. Blood 92:3515, 1998.
Hurley CK, Wade JA, Oudshoorn M, et al: A special report: histocompatibility testing guidelines for hematopoietic stem cell transplantation using volunteer donors. Tissue Antigens 53:394, 1999.
Wadenvik H, Stockelberg D, Hou M: Platelet proteins as autoantibody targets in idiopathic thrombocytopenic purpura. Acta Paediatr Suppl 424:26, 1998.
Beardsley DS, Ertem M: Platelet autoantibodies in immune thrombocytopenic purpura. Transfus Sci 19:237, 1998.
Davis GL: Platelet specific alloantigens. Clin Lab Sci 11:356, 1998.
Warkentin TE, Smith JW: The alloimmune thrombocytopenic syndromes. Transfus Med Rev 11:296, 1997.
West KA, Anderson DR, McAlister VC, et al: Alloimmune thrombocytopenia after organ transplantation. N Engl J Med 341:1504, 1999.
Bierling P, Fromont P, Elbez A, Duedari N, Kieffer N: Early immunization against platelet glycoprotein IIIa in a newborn Glanzmann type I patient. Vox Sang 55:109, 1988.
Li C, Pasquale DN, Roth GJ: Bernard-Soulier syndrome with severe bleeding: absent platelet glycoprotein Ib alpha due to a homozygous one-base deletion. Thromb Haemost 76:670, 1996.
Ikeda H: Platelet membrane protein CD36. Hokkaido Igaku Zasshi 74:99, 1999.
Newman PJ, Valentin N: Human platelet alloantigens: recent findings, new perspectives. Thromb Haemost 74:234, 1995.
Santoso S, Kiefel V: Human platelet-specific alloantigens: update. Vox Sang 74(suppl 2): 249, 1998.
Lyman S, Aster RH, Visentin GP, Newman PJ: Polymorphism of human platelet membrane glycoprotein IIb associated with the Baka/Bakb alloantigen system. Blood 75:2343, 1990.
Nurden AT: Polymorphisms of human platelet membrane glycoproteins: structure and clinical significance. Thromb Haemost 74:345, 1995.
Bray PF: Integrin polymorphisms as risk factors for thrombosis. Thromb Haemost 82:337, 1999.
Goldschmidt-Clermont PJ, Roos CM, Cooke GE: Platelet PlA2 polymorphism and thromboembolic events: from inherited risk to pharmacogenetics. J Thromb Thrombolysis 8:89, 1999.
Holensteiner A, Walchshofer S, Adler A, Kittl EM, Mayr WR, Panzer S: Human platelet antigen gene frequencies in the Austrian population. Haemostasis 25:133, 1995.
Tanaka S, Ohnoki S, Shibata H, Okubo Y, Yamaguchi H, Shibata Y: Gene frequencies of human platelet antigens on glycoprotein IIIa in Japanese. Transfusion 36:813, 1996.
Tanaka S, Taniue A, Nagao N, et al: Genotype frequencies of the human platelet antigen, Ca/Tu, in Japanese, determined by a PCR-RFLP method. Vox Sang 70:40, 1996.
Kim HO, Jin Y, Kickler TS, Blakemore K, Kwon OH, Bray PF: Gene frequencies of the five major human platelet antigens in African American, white, and Korean populations. Transfusion 35:863, 1995.
Kekomäki S, Partanen J, Kekomäki R, Platelet alloantigens HPA-1, -2, -3, -5 and -6b in Finns. Transfus Med 5:193, 1995.
Simsek S, Vlekke AB, Kuijpers RW, Goldschmeding R, von dem Borne AE: A new private platelet antigen, Groa, localized on glycoprotein IIIa, involved in neonatal alloimmune thrombocytopenia. Vox Sang 67:302, 1994.
Kroll H, Kiefel V, Santoso S: Clinical aspects and typing of platelet alloantigens. Vox Sang 74(suppl 2): 345, 1998.
Kekomäki R, Raivio P, Kero P: A new low-frequency platelet alloantigen, Vaa, on glycoprotein IIbIIIa associated with neonatal alloimmune thrombocytopenia. Transfus Med 2:27, 1992.
Peyruchaud O, Bourre F, Morel-Kopp MC, et al: HPA-10w(b) (La(a)): genetic determination of a new platelet-specific alloantigen on glycoprotein IIIa and its expression in COS-7 cells. Blood 89:2422, 1997.
Kiefel V, Vicariot M, Giovangrandi Y, et al: Alloimmunization against Iy, a low-frequency antigen on platelet glycoprotein Ib/IX as a cause of severe neonatal alloimmune thrombocytopenic purpura. Vox Sang 69:250, 1995.
Kelton JG, Smith JW, Horsewood P, Humbert JR, Hayward CP, Warkentin TE: Gova/b alloantigen system on human platelets. Blood 75:2172, 1990.
Bordin JO, Kelton JG, Warner MN, et al: Maternal immunization to Gov system alloantigens on human platelets. Transfusion 37:823, 1997.
McFarland JG: Alloimmunization and platelet transfusion. Semin Hematol 33:315, 1996.
Kekomäki R: Use of HLA- and HPA-matched platelets in alloimmunized patients. Vox Sang 74(suppl 2): 359, 1998.
Kaplan C: Evaluation of serological platelet antibody assays. Vox Sang 74(suppl 2): 355, 1998.
Unkelbach K, Kalb R, Santoso S, Kroll H, Mueller-Eckhardt C, Kiefel V: Genomic RFLP typing of human platelet alloantigens Zw(PlA), Ko, Bak and Br (HPA-1, 2, 3, 5). Br J Haematol 89:169, 1995.
Panzer S, Kaplan C: Report on the 1997 International Society of Blood Transfusion Workshop for Genotyping of Platelet Alloantigens. Platelet and granulocyte workshop ISBT. Transfus Med 8:125, 1998.
McFarland JG: Platelet and neutrophil alloantigen genotyping in clinical practice. Transfus Clin Biol 5:13, 1998.
Quintanar A, Jallu V, Legros Y, Kaplan C: Human platelet antigen genotyping using a fluorescent SSCP technique with an automatic sequencer. Br J Haematol 103:437, 1998.
Metcalfe P, Cavanagh G, Hurd C, Ouwehand WH: HPA genotyping by PCR-SSP: report of 4 exercises. Vox Sang 77:40, 1999.
Skupski DW, Bussel JB: Alloimmune thrombocytopenia. Clin Obstet Gynecol 42:335, 1999.
Nomura S, Matsuzaki T, Yamaoka M, et al: Genetic analysis of HLA- and HPA-typing in idiopathic (autoimmune) thrombocytopenic purpura patients treated with cepharanthin. Autoimmunity 30:99, 1999.
Copyright © 2001 McGraw-Hill
Ernest Beutler, Marshall A. Lichtman, Barry S. Coller, Thomas J. Kipps, and Uri Seligsohn