A New Approach for the Analysis of HLA Class II Polymorphism: ‘HLA Oligotyping’

J.-M. Tiercy,

M. Jeannet,

B. Mach

S UM M A R Y. Histocompatibility typing allows the matching of patients and donors in organ traqhmtation, and the accuracy of I-WA matching infiuences to a great extent the clinical outcome. Recent breakthroughs in tbe molecular biology of I-ILA class II genes have revealed that tbe degree of HLA diversity and polymorphism is in fact much greater than was expected on the basis of tbe tradithal serological HLA typing assays. In parallel, it has become possible to analyse tbis exteskaivepolymorphism directly at the level of tbe HLA class II genes aad of their DNA sequc8ces. We have described a DNA typing pro&ure referred to as ‘HLA oligotyping’ which is based on the hybridisation of allele and loci speciik oIigom&otide probes. Tbis procedure has now become operational 00 a large scale and this review describes the principles and major applications of the technique. It consists in the hybrid&&ion of DNA with informative sequeucespecilic oligoaucleotide probes, following an amplification of DNA in vitro by the polymerase chain reaction (PCR). Tbe use of this highly seusitive technique for HLA-DR, -DQ aad -DP typing is discussed, focusing on the cliuical applications in tbe field of organ tmasphhtion, partiadarly for bone marrow trau@otation with unrelated donors. It now allows tbe unambiguous identilication of all HLA &types, h&ding those that cannot be rewghed otbenvhe, pad it represedp powerful complement to current methods of HLA typing. Finally this methodology is widely used in HLAdisease association studies, aiming at the characterisation of HLA class II epitopes involved in tbe susceptibility or resistance to autoimmune diseases.

The human major histocompatibility complex (MHC) contains a series of linked genes on the short arm of chromosome 6 whose products are polymorphic cell surface molecules referred to as HLA (human leukocyte antigens) and serum factors involved in the immune responses. The MHC encodes 3 major classes of antigens: the class I antigens (HLAA, -B, -C), also known as transplantation antigens, the class II antigens (HLA-DR, -DQ, -DP) and the class III antigens which correspond to components of

J.-M. Tiercy, M. Jeannet, Transplantation Immunology Unit, H&pita1 Cantonal Universitaire de Gemhe, J.-M. They, B. Mach, Department of Microbiology, University School of Medicine, Centre MBdical Universitaire, Geneva, Switzerland. Blood Reviews (1990) 4.9-15 0 1990 Longman Group UK Ltd

the complement system. Both class I and class II molecules share the common function of HLArestricted antigen presentation, that is, both bind immunogenic peptides for presentation to specific T-cells. l-5 In general class I-restricted recognition allows activation of CD8 + cytotoxic T-cells, whereas class II-restricted recognition activates CD4 + T-cells with either helper or cytotoxic functions.‘-’ In addition MHC antigens play a crucial role in transplantation immunology,* in the susceptibility to autoimmune diseases,’ and in the acquisition of immunological tolerance through thymic education.‘O In view of the biological and medical importance of HLA class II polymorphism, it appears critical to devise simple and highly sensitive techniques for HLA oxw%ox/w/ooc4-0

s10.00

10

A NEW APPROACH FOR THE ANALYSIS OF HLA CLASS II POLYMORPHISM

tissue typing. This review describes the principle and the ongoing and potential applications of a new approach to HLA DNA genotyping, referred to as oligonucleotide typing.”

‘HLA OLIGOTYPING

variable regions of the first domains of class II molecules are local&d in the bottom and in the a-helices flanking the peptide-binding groove.17v1g MHC class II antigens are encoded by the HLA-D region encompassing 12- 15 homologous non-allelic a- and P-genes over about 1100 kb.14s2’ These genes are grouped into 3 subregions HLA-DR, -DQ and -DP (Fig. 1) and are not all functional, since each subregion contains non-expressed class II loci, most of these being obvious pseudogenes. A first level of polymorphism resides in the variable number of DRB loci expressed,14 with most haplotypes expressing two different DRB-chains. In the DR3, 5, w6 haplotypes, a second DRJ3-chain is encoded by locus DRB3,” corresponding to the DRw52 supertypic specificity,” while in DR4, 7 and 9 haplotypes a second DRPchain is encoded by locus DRB4,23 corresponding to the DRw53 supertypic specificity. In the DR2 haplotypes, the second DRP-chainz4 is encoded by locus DRB525 and in DRl, DRwX and DRwlO haplotypes, a single DRP-chain is expressed. However the major contribution to MHC class II diversity resides in its very high degree of allelic polymorphism (Fig. 1).l4 Polymorphic HLA class II molecules encoded by the individual loci carry the serological and cellular histocompatibility types originally defined by alloantisera (DR typing) and by primary mixed lymphocyte culture (Dw typing). Molecular biological analysis of HLA class II genes showed that polymorphism is indeed larger than expected from serology. For example, 34 alleles of DRBl locus have now been identified at the DNA sequence level (Fig. l), whereas alloantisera identify 14 HLA-DR allelic specificities and in most routine typing laboratories only about 10 DRBl allelic specificities can be identified by alloantisera.

HLA Class II Polymorphism HLA class II antigens are highly polymorphic transmembrane glycoproteins composed of non-covalently associated CL(33 kDa) and p (28 kDa) chains.” In contrast to HLA class I antigens which are expressed on the surface of almost all nucleated cells, class II antigens are expressed on B-lymphocytes, macrophages, activated T-cells and dendritic cells, which are collectively referred to as antigen-presenting cells.5*12 MHC class II molecules exert a major control on T-cell responses, by the phenomenon of MHC-restriction.4*5V’3The capacity of a given antigenic peptide to trigger an immune response varies qualitatively and quantitatively with the diversity and level of expression of the HLA class II molecules on the APC. Depending on its affinity for a particular antigenic peptide, a given MHC allele is said to be a high or a low responder. Therefore MHC polymorphism corresponds functionally to a diversity of the immune performance14 and MHC class II genes are also referred to as Ir (immune response) genes.15 Another important property of HLA class II antigens is their capacity to induce proliferation of alloreactive T-cells (alloreactivity). The recognition of foreign HLA molecules by alloreactive T-cells may also be imparted by peptide-MHC interactions. l6 Alloreactivity, which is measured in vitro by the primary mixed lymphocyte culture (MLC), plays a key role in graft rejection and graft-versus-host disease in bone marrow transplantation. Recently a theoretical 3D model has been derived” from the cristallographic analysis of the HLA-A2 molecule.‘8 The model predicts that a- and P-chains of HLA class II antigens fold to form a peptidebinding groove consisting of the external aminoterminal domains of both chains. Interestingly, the polymorphic variations which are clustered in hyper-

The Oligotyping Technique Following the initial cloning and sequencing of the polymorphic DRB gene,26 it was possible to demonstrate that HLA class II polymorphism can be analysed directly at the DNA level by RFLP-DNA

MAP OF HLA-D REGION

DN

DO

-mHH 82

A2

w

w

Number of alleles :

61

Al

Bl (D~ci)

19

4

1

(D&)

1

Al

Bl

(c!?$) (DAx2u)

2

1

82

83

84

B5

A

4

1

4

1

w

13

8

34

Fig. 1. Schematic map of the HLA-D region of the human major histocompatibility complex. The number of alleles is according to the WHO-Nomenclature Report 1 989.2S DRBl locus encodes for the DRl -DRwl8 allospecificities, DR83 and DRB4 encode for the supertypic DRw52 and DRw53 specificities, respectively. cpindicates a pseudogene. Former names of DNA, DOB, DOB2 and DQ.42 loci are indicated in parentheses.

BLOOD REVIEWS

typing (restriction fragment length polymorphism).27 However, in view of the limitations of the RFLP technique (detection of phenotypically irrelevant polymorphic sites, no possibility for direct analysis of specific amino acid differences, e.g. for DR4 subtypes, use of several restriction enzymes), a new procedure was proposed, based on the hybridisation of HLA class II genes with locus- and allele-specific oligonucleotide probes under conditions (washing temperatures) allowing the detection of a single base pair mismatch. l1 Such methodology, commonly referred to as oligotyping, thus allows the detection of phenotypically relevant differences at the level of single amino acid residues. Originally the oligotyping procedure was performed by hybridisation with HLA class II sequencespecific oligo probes on southern genomic blots.” The technique allowed for example the identification of three serologically undetected alleles of the DRw52 specificity,** or the DNA typing of all DRw6-DQwl subtypes 29 thus replacing the cumbersome Dw typing methodology. Other applications of the technique have been reported by the analysis of the DR4 subtypes,30 or for the detection of the 3.1 and 3.2 splits of DQw~.~’ The introduction of the PCR technology3* allows a rapid amplification of the relevant DNA and thus greatly facilitates the hybridisation step. The polymerase chain reaction (PCR) is the in vitro enzymatic amplification of a specific DNA fragment by using two oligonucleotide primers that are complementary to opposite strands of the target sequence. The reaction consists of repeated cycles of denaturation, primer annealing and primer extension by the thermostable Taq Polymerase,33 resulting in a 106- to lOa-fold amplification of a particular DNA segment.32*34 The technique allows the use of samples with a DNA content too low or too degraded for RFLP typing or for subcloning and sequencing. The only requirement for PCR amplification of HLA class II genes is to know the sequence of the DNA flanking the polymorphic regions which are to be amplified and this condition is fulfilled for HLA-DR, -DQ and -DP polymorphic first domain exons. Sequence-

11

specific oligonucleotide probes are derived by comparison of aligned DNA sequences of the first domain exons. Although most of the HLA oligotyping studies used radiolabelled oligo probes, non-radioactive detection methods have recently been described.34*35 An alternative to this HLA oligotyping procedure has been very recently described, whereby the oligonucleotide probes are immobilised on a nylon support and the filter hybridised with the biotin-labelled amplified DNA sample. 36 This procedure, however, requires the use of unique washing conditions for all probes on a given filter, which in several cases will be difficult to achieve. HLA-DR Oligotyping As indicated in Figure 1, DRBl is the most polymorphic locus with now 34 allelic DNA sequences (for references, see ref. 25). DRBl encodes the serological determinants DRl to DRwl8, as well as a number of determinants recognised by cellular typing (e.g. DR4 subtypes). The other loci of the HLA-DR subregion are either less polymorphic with 4 allelic sequences for DRB3 (DRw52a, 52bl,52b2,52c), 4 for DRBS (DR2 subtypes) or non-polymorphic with 1 allele for DRB4 (DRw53). By using a limited number of oligo probes, mainly derived from DNA sequences of the first hypervariable region, all major serological specificities can be identified: DRl to DRwl4. On the basis of the available DNA sequence information it is possible to identify the subtypes of each serological specificity (Table 1) by using additional oligo probes. A number of alleles can be identified by strictly allele-specific oligo probes, e.g. for the DRBl*1501, 1201, 1303, 1401, 0901, 1001 or DRB3*0101 alleles. However, due to the ‘patchwork’ structure of the first domain exon polymorphic sequences of class II genes,14 the identification of some alleles require the use of two (or three) oligo probes, as illustrated in Figure 2A for the DR1.l (DRBl*OlOl) and DR1.3 (DRB1*0103) alleles or for the DRw8 and DRwll subtypes. The oligotyping procedure on PCR-amplified DNA has been shown to be a method of choice for the detection

Table 1 HLA-DR and -DQ micropolymorphism (splitsof serologicalspecificities)as currentlydetectedby oligotypinganalysis.The major DRI to DRwl4 specificitiesare also recognisedby sequence-specific oligonucleotideprobes.“M The alleles listed for DQAl and DQBl loci are those found in linkage disequilibrium (in Caucasians) with the DR specificities indicated in the first column. Nomenclature of the alleles are according to ref. 25. Three additional DR4 subtypes can now be detected by oligotyping DRBI DRI DR2 DR3 DR4 DRwll DRwl2 DRwl3 DRwl4 DR7 DRw8 DR9 DRwlO

0101,0102, 0103 1501, 1502,1601, 1602 0301, 0302 0401, 0402, 0403, 0404, 0405 1101,1102,1103 1201 1301, 1302, 1303 1401, 1402 0701, 0702 0801,0802,0803 0901 1001

DRB3

0101,0201 0101,0201 0201 0101,0201, 0301 0101,0201,0202

DQAl

DOB1

0101 0102,0103 0501,0401 0301 0103,0501 0501 0102,0103 0101 0201 0103, 0601,0401 0301 0101

0501,0301 0601,0602, 0502,030l 0201,0402 0301,0302,0401 0301,0603 0301 0603,0604,0301 0503,0301 0201,0303 0601,0301,0402 0303 0501

12

A NEW APPROACH

A)ANR,~IQ,

FOR THE ANALYSIS

-DR1.113

Qc

-#Rl.3/14

HLA-DP Oligotyping 2

B) CAP-&

WBl-

‘HLA OLIGOTYPING’

DQwl specificity which is subdivided into at least 7 alleles (Table 1, Fig. 2D).

MEY- II

CAC-

OF HLA CLASS II POLYMORPHISM

15/l

a

m

w1

11

1111

*c

-DR15.1116.1

dim

c

412 > .i

3

4

411

612

613

7

*

-DRll.fll6.t

8 -DQ6.3/3.1 - DQ6.213.2

Fig. 2. Examples of oligonucleotide typing analysis for HLADRBI and -DOB1 alleles: (A) identification of two DRI subtypes: DR1.1 (DRBl‘OlOl; DRI-Dwl) and DRl.3 (DRB1’0103; DR’BR’); (B) identification of two DR2 subtvoes: DR15.1 IDRBI ll 501: DR2-Dw2) and DRl6.1 (DRB1’1601; DR2-Dw21); (C)’ identification of two DR4 subtypes: DR4.1 (DRBl’0401; DR4-Dw4) and DR4.4 (DRB1’0404; DR4-Dwl4); (D) identification of four DQ subtypes DQ6.3 (DClBl l 0603), 006.2 (DQBl l 0602), DQ3.1 (DQBl.0301; DQw7) and DQ3.2 (DOB1 l 0302; DQw8). DNA samples from leukemic patients have been amplified by PCR and hybridised with sequence-specific oligonucleotide probes as described elsewhere.41

of DRl subtypes (Fig. 2A), of DR2 subtypes3’ (Fig. 2B), of DR4 subtypesJ8 (Fig. 2C), of DRw6 subtypeszg or of DRwl 1 subtypes (Fig. 2B) (Tiercy et al, unpublished data). In addition, the combination of the enzymatic amplification by PCR with direct subcloning and sequencing allows the identification of new HLA class II alleles, thereby improving the definition of the extent of HLA polymorphism.3g-41 This approach has immediate applications in the field of organ transplantation.41 HLA-DQ Oligotyping A number of studies used PCR followed by subcloning and sequencing to characterise HLA-DQAl and -DQB 1 polymorphism. 3g*40*42So far 8 allelic sequences for DQAl and 13 allelic sequences for DQBl have been described. 25 Following the initial report on HLA-DQA oligotyping43 major applications of DQ typing have been for HLA-disease association studies,3g*40 as will be described below. Since the DQBl locus is more polymorphic than the DQAl locus, we have used a combination of 13 oligo probes derived from the polymorphic regions of DQBl first domain exons to achieve unambiguous DQ typing for 12 DQBl alleles on homozygous as well as on heterozygous individuals (Morel et al, in preparation). High resolution DQ typing is particularly relevant within

The polymorphism of the first domain exon of DPA 1 and DPBl loci has been analysed by PCR amplification, subcloning and sequencing.44 So far 4 allelic sequences of DPAl and 19 allelic sequences of DPBl have been described. 25 Oligotyping analysis on PCRamplified DNA using 14 sequence-specific oligo probes allowed reliable genotyping for DP antigens.45 Recently, an alternative, referred to as PCR-RFLP method, has been proposed for DPB typing as well as for DQA typing. The procedure consists of digesting PCR-amplified DNA with a number of restriction enzymes recognising allelic sequence variations in the DPB second exon followed by subsequent electrophoretie analysis.46 HLA Class II Oligotyping in Bone Marrow Transplantation Although HLA class II oligotyping has been largely applied for refining HLA-disease associations, very few studies report its use for organ transplantation and particularly for bone marrow transplantation. Oligotyping on PCR-amplified DNA is of value in defining the HLA-DR type of recipients and potential donors (kidney or bone marrow), or in replacing serology when its results are difficult to interpret. This occurs frequently with leukemic (chronic myeloid leukemia) patients and obviously class II serology is deficient in cases of severe combined immunodeficienties with defect of expresion of class II molecules. Bone marrow transplantation represents a major therapeutic option in the treatment of patients with leukemic and bone marrow failure syndromes. Successful bone marrow transplantation depends largely upon the degree of HLA matching betwen donor and recipient. 47 If polymorp hi c residues in the HLA antigens are mismatched, these molecules may be recognised as foreign by the immune system with such as graft-versus-host disease consequences (GVHD), graft rejection or reconstitution failure. Bone marrow transplantation with HLA-matched unrelated donors represents a suitable alternative for those patients who lack an HLA-identical sibling donor. However, a major limitation is inadequate matching due to the undetected allelic differences in the matching of such donor/recipient pairs. We have recently proposed the use of the highly sensitive HLA class II oligotyping technique as a new tool to improve matching of unrelated donors.48 We have suggested that selection of an optimal donor could first be achieved on the basis of HLA class I serology and then on the basis of class II typing for the major DR specificities (DRl to DRwl4) either by serology or by oligotyping analysis. A final selection would be achieved by the oligotyping analysis for

BLOOD REVIEWS

HLA-DR and -DQ (and possibly -DP) micropolymorphism (i.e. splits of known serological specificities). The analysis of HLA-DR and -DQ micropolymorphism by oligotyping on PCR-amplified DNA of class I and class II serologically identical unrelated donor/recipient pairs allowed us to show that this technique could in most cases predict a positive MLC;48 (Tiercy et al, submitted). More specifically, the oligotyping technique represents an attractive alternative to the Dw typing method for the detemination of the cellular splits of DRl, DR2, DR4, DRw6, DRw8 or DRwll (see also’ Fig. 2). It is particularly important for DRl, DR4, DRw8 or DRwl 1 subtypes which are not detectable by RFLP typing. A recent report describes the use of the oligotyping procedure to type a patient with HLA-deficient severe combined deficiency and to check optimal matching with a serologically identical unrelated donor.4g HLA oligotyping also represents a useful tool for the identification of HLA-B/-DR or intra-class II recombination events. MHC-restricted Antigen Presentation Foreign antigens are recognised by T-cells in the form of peptides associated with MHC molecules. A prerequisite for T-cell activation seems to be the binding of antigenic peptides to MHC molecules.4 It is now widely accepted that MHC molecules have a single binding site which can associate with a variety of different antigens One approach to identify functional sites on the HLA class II molecules which could bind antigen and/or interact with TCR is to compare the amino acid sequence of class II molecules, as deduced from DNA oligotyping analysis, and their capacity to present a given antigen to specific T-cell clones. By comparing the pattern of reactivity of antigenspecific or alloreactive T-cell clones with the pattern of allele-specific hybridisation with oligonucleotide probes, correlations could be made between the functional and the structural polymorphism of HLADRB1-50,51 and -DRB3-encoded molecules.‘g*s2 HLA Disease Associations Particular MHC antigens also contribute to the susceptibility to a large number of autoimmune disorders, including insulin-dependant diabetes (IDDM), rheumatoid arthritis, multiple sclerosis, coeliac disease, ankylosing spondylitis and pemphigus vulgaris. The majority of these diseases are HLA class II-linked, possibly because class II molecules control the B-cell response to autoantigens and/or affect the T-cell repertoire through thymic selection. Although inheritance of susceptibility appears to be polygenic in most instances class II polymorphism plays a crucial role which, however, is not yet completely elucidated. Recently, molecular analysis of class II polymorphism by DNA sequencing and oligotyping

13

Table 2 Major HLA class II associations with susceptibility to autoimmune diseases, as determined by DNA sequencing and oligotyping analyses DR and DQ alleles IDDM Caucasoids

IDDM blacks Rheumatoid arthritis Pemphigus vulgaris Coeliac disease

Multiple sclerosis

DR4-DQw3.2 DR3-DQw2 9,31 DRI-DQw5.1 DR2/Dw21-DQw5.2 1 DR7-DQA3.1 54 DR9-DQw2 1 ;;t$’

;;$,-Dw15 9,30,55,56 1 DR4-Dwlb 57 DQw5.3 1 DPB4.1, DPB3 58 DR3-DQw2 1 DR~-DQw~/DR~-DQw~~~ DQw6.2, 6.3, 6.4, 3.2, 3.3”

showed first that there is no disease-specific HLA class II allele and second that, at least for some autoimmune diseases, particular alleles of HLA-DR, -DQ and -DP contribute to susceptibility. Table 2 shows major HLA class II associations in autoimmune diseases with particular alleles involved in the susceptibility. In the initial studies of IDDM, which affects 0.5% of Caucasian populations, sequencing and oligotyping analysis pointed to the localisation of the ‘susceptibility epitope’ to residue 57 in the DQ$-chain,3g with homozygous individuals non-ASP/non-ASP showing the highest relative risk. 53 However, subsequent sequencing and oligotyping analysis of black IDDM patients showed that susceptibility was closely linked to the DQAl locus, which suggests that both DQaand DQP-chains contribute to disease susceptibility. 54 Particular haplotyp es which are characterised by ASP 57-positive DQP-chains (DR2-Dw2, DR2Dw12, DR5, DR4-DQ3.1), are either neutral or confer resistance to IDDM in a dominant manner,’ either through a mechanism of HLA-DQ-restricted T-cell suppression or by deletion of autoreactive T-cells during thymic selection. Other well characterised MHC class II alleles which are implicated in the susceptibility to autoimmune diseases are the DR4Dw4, Dw14, Dw15, DRl-Dwl, and DRwlO alleles in rheumatoid arthritis.g*55*56Close to 100% of pemphigus patients have either or both of the DR4-DwlO allele and the DQB allele normally found in linkage disequilibrium with the DRwl4-Dw9 haplotype.” Concluding Remarks The highly sensitive oligotyping methodology (i.e. DNA hybridisation with sequence-specific oligonucleotide probes) represents a powerful tool to analyse HLA class II polymorphism. It is made simpler for routine usage by the prior amplification of DNA by PCR (polymerase chain reaction). This new technology is now operational and has immediate clinical applications in the field of organ transplantation, and particularly for bone marrow transplantation. It first

14 A NEW APPROACH FOR THE ANALYSIS OF HLA CLASS II POLYMORPHISM ‘HLA OLIGOTYPING represents a rapid and efficient way for HLA class II typing when a serological analysis is impossible or gives uninterpretable results, which is often the case for leukemic patients. Second it allows the unequivocal identification of multiple hidden alleles not detectable by serology (micropolymorphism) and can therefore replace Dw typing. It does not require the maintenance of performant (and rare) alloantisera or of a large battery of homozygous typing cell lines. Oligotyping analysis of HLA class II micropolymorphism in bone marrow transplantation with unrelated donors not only contributes to improved selection of optimally matched donors but also, through ongoing and retrospective analysis, helps to define the exact contribution of class II polymorphism to graft failure, graft rejection and GVHD. The analysis of such micropolymorphic differences, which is not feasible by current routine typing, together with clinical data and functional in vitro assays (MLC, minor antigens, cytotoxic lymphocyte-precursor frequencies) might help in selecting bone marrow donors with acceptable mismatches at class II loci. One of the consequences of our ability to type for a large number of micropolymorphic differences is that the task of identifying a perfectly matched unrelated donor for bone marrow transplantation will become more difficult, especially for rare haplotypes. If it ultimately becomes clear that optimal HLA class II matching down to single amino acid differences is clinically beneficial, this difficulty will not be considered as an obstacle. In fact, the simplicity of the procedure we have presented will greatly contribute to increase the chance of finding an appropriate match in potentially very large pools of volunteer donors.

10. 11.

12.

13. 14.

15. 16.

17.

18.

19.

20.

21.

22.

Acknowledgements We thank Barbara Battistolo, Andrea Morrison and Patricia Roux-Chabbey for their excellent technical assistance. This work has been supported by grants from the Swiss National Science Foundation and the Carlos and Elsie de Reuter Foundation.

References 1. Zinkemagel R M, Doherty P C 1979 MHC-restricted cytotoxic T-cells: studies on the biological role of polymorphic major transplantation antigens determining T-cell restriction specificity, function, and responsiveness. Advances in Immunology 27: 51-177 2. Morrison L A, Lukacher A I, Braciale V L, Fan D P, Braciale T G 1986 Differences in antigen presentation to MHC class I and class II restricted influenza virus-snecific cytolytic T lymphocyte clones. Journal of Experimental Medicine 163: 903-921 3. Townsend A R M, Rothbard J, Gotch F M et al 1986 The epitopes of influenza nucleoprotein recognized by cytotoxic T lymphocytes can be defined with short synthetic peptides. Cell 44: 959-968 4. Butts S, Sette A, Grey H M 1987 The interaction between protein-derived immunogenic peptides and Ia. Immunological Reviews 98: 115-141 5. Schwartz R H 1985 T-lymphocyte. recognition of antigen in association with gene products of the major histocompatibility complex. Annual Review of Immunology 3: 237-261

23.

24.

25.

26.

27. 28.

29.

Germain R N 1987 The ins and outs of antigen processing. Nature 332: 687-689 Long E 0 1989 Intracellular traffic and antigen processing. Immunology Today 10: 232-234 Bach F H, Sachs D H 1987 Transplantation Immunology. New England Journal of Medicine 317: 489-492 Todd J A, Acha-Orbea H, Bell J I et al 1988 A molecular basis for MHC class II-associated autoimmunity. Science 240: 1003-1009 Schwartz, R H 1989 Acquisition of immunologic selftolerance. Cell 57: 1073-1081 Angelini G, de P&al C, Gorski J, Mach B 1986 High resolution analysis of the human HLA-DR polymorphism by hybridization with sequence-specific oligonucleotide probes. Proceedings of the National Academy of Sciences of the USA 83: 4489-4493 Kaufmann J F, Auffray C, Korman A J, Shackelford D A, Strominger J 1984 The class II molecules of the human and murine major histocompatibility complex. Cell 36: I- 13 Benacerraf B 1981 Role of MHC gene products in immune regulation. Science 212: 1229-1238 Mach B, Reith W, Berte C, Tiercy J-M 1989 Diversity and regulation of HLA class II genes. In: Silver J (ed) Molecular Biology of HLA Class II Antigens. CRC Press, Boca Raton, Florida (in press) Benacerraf B, McDevitt H 0 1972 Histocompatibility-linked immune response genes. Science 175: 273-279 Marrack P, Kappler J 1988 T cells can distinguish between allogeneic major histocompatibility complex products on different cell tvnes. Nature 232: 840-843 Brown J H, Jardetzky T, Saper M A et al 1988 A hypothetical model of the foreign antigen binding site of class II histocompatibility molecules. Nature 332: 845-850 Bjorkman P J, Saper M A, Samraoui B et al 1987 Structure of the human class I histocompatibility antigen, HLA-AZ. Nature 329: 506-512 Gorski J, IrlC C, Mickelson E M et al 1989 Correlation of structure with T cell responses of three members of the HLA-DRw52 allelic series. Journal of Experimental Medicine 170: 1027-1032 Hardy D A, Bell J I, Long E 0, Lindsten L, McDevitt H 0 1986 Mapping of the class II region of the human histocompatibility complex by pulse-field gel electrophoresis. Nature 323 453-455 Rollini P, Mach B, Gorski J 1985 Linkage map of three HLA-DR B-chain genes: evidence for a recent duplication event. Proceedings of the National Academy of Sciences of the USA 82: 7197-7201 Gorski J, Tosi R, Strubin M, Rabourdin-Cornbe C, Mach B 1985 Serological and immunological analysis of the products of a single HLA-DR-a and DR-8 chain gene expressed in a mouse cell line after DNA-mediated cotransformation reveals that the 8 chain carries a known supertypic specificity. Journal of Experimental Medicine 162: 105-l 16 Gorski J, Rollini P, Mach B 1087 Structural comparison of the genes of two HLA-DR supertypic groups: the loci encoding DRw52 and DRw53 are not truly allelic. Immunogenetics 25: 397-402 Kawai J, Ando A, Sato T et al 1989 Analysis of gene structure and antigen determinants of DR2 antigens using DR gene transfer into mouse L cells. Journal of Immunology 142: 3 12-3 17 Bodmer J G, Marsh S G E, Albert E 1990 Nomenclature for factors of the HLA system 1989. Immunology Today 11: 3-10 Long E 0, Wake C T, Gorski J, Mach B 1983 Complete sequence of an HLA-DRP chain deduced from a cDNA clone and identification of multiple non-allelic DRB chain genes. EMBO Journal 2: 389-394 Wake C T, Long E 0, Mach B 1982 Allelic polymorphism and complexity of the genes for HLA-DR8 chains-Direct analysis by DNA-DNA hybridization. Nature 300: 372-374 Tiercy J-M, Gorski J, Jeannet M, Mach B 1988 Identification and distribution of three serologically undetected HLA-DR alleles by oligonucleotide. DNA typing analysis. Proceedings of the National Academy of Sciences of the USA 85: 198-202 Tiercy J-M, Gorski J, BCtuel H et al 1989 DNA tynina of DRw6 subtypes: correlation with DRBl and DRB? afielic sequences by hybridization with oligonucleotide probes. Human Immunology 24: l-14

BLOOD REVIEWS 30. Nepom G T, Seyfried C E, Holbeck S L, Wild&e K R, Nepom B S 1% IdentiScation of HLA-Lhvl4 genes in DR4+ rimmidd.astthritis. Lancfst Zk 1002-1005 31. Markussen G, Ronningen K S, Paulsen G 1988 HLADQw3.1 and DQw3.2 associated exon polymorphism detected by oligonucleotide probes. Tissue Antigens 31: 204-210 32. Mullis K B, Faloona F A 1987 Specific synthesis of DNA in vitro via a polymerase catalysed chain reaction. Methods in Enzymology 155: 335-350 33. Saiki R K, Gelfand D H, Stoffel S et al 1988 Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239: 487-49 1 34. White T J, Amheim N, Erlich H A 1989 The polymerase chain reaction. Trends in Genetics 5: 185-189 35. Eliaou J F, Humbert M, Balaguer P, 1989 A method of HLA class II typing using non-radioactive labelled oligonucleotides. Tissue Antigens 33: 475-485 36. Saiki R K, Walsh P S, Levenson C H, Erlich H A 1989 Genetic analysis of amplified DNA with immobilized sequence-specific oligonucleotide probes. Proceedings of the National Academy of Sciences 86: 6230-6234 31. Obata T, Ito I, Kaneko T et al 1989 Oligonucleotidegenotyping as a method of detecting the HLA-DR2 (DRwlS)-Dw2, -DR2 (DRwlS)-Dwl2, -DR4-Dwl5, and -DR4D‘KT2’ haplotypes in the Japanese population. Tissue Antigens 33: 550-558 38. Erlich H A, Scharf S, Long C M, Horn G T 1989 Analysis of isotypic and allotypic sequence variation in the HLADR8 region using the in vitro enzymatic amplification of specific DNA segments. In: B. DuPont (ed) Immunobiology of HLA, Vol. 2, Springer Verlag, New York, pp 181-185 39. Todd J A, Bell J I, McDevitt H 0 1987 HLA-DQS gene contributes to susceptibility and resistance to insulindependant diabetes mellitus. Nature 329: 599-604 40. Horn G T, Bugawan T L, Long C M, Erlich H A 1988 Allelic sequence variation of the HLA-DQ loci: relationship to serology and to insulin-dependant diabetes susceptibility. Proceedings of the National Academy of Sciences of the USA 85: 6012-6016 41. Tiercy J-M, Jeannet M, Mach B 1989 A new HLA-DRBI allele within the DRw52 supertypic specificity (DRwl3DwHAG): sequencing and direct identification by oligonucleotide typing. European Journal of Immunology, in press 42. Gyllenstein, U B, Erlich H 1988 Generation of singlestranded DNA by the polymerase chain reaction and its application to direct sequencing of the DQA locus. Proceedings of the National Academy of Sciences of the USA 85: 7652-7656 43. Saiki R K, Bugawan T L, Horn G T, Mullis K B, Erlich H A 1986 Analysis of enzymatically amplified S-globin and HLA-DQu DNA with allele-specific oligonucleotide probes. Nature 324: 163-166 44. Bugawan T L, Horn G T, Long C M et al 1988 Analysis of HLA-DP allelic sequence polymorphism using the in vitro enzymatic DNA amplification of DP-a and DP-8 loci. Journal of Imunology 141: 4024-4030 45. Angelini G, Bugawan T L, Delfino L, Erlich H A, Ferrara G B 1989 HLA-DP typing by DNA amplification and hybridization with secific oligonucleotides. Human Immunology 26: 169- 177

15

46. Maeda M, Uryu N, Murayama N et al 1990 A simple and

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

rapid method for HLA-DP of PCRI+y amoli&ed DNA with allele-spec& rest&ion endonucIeases. Human Immunology, in p&s Anasetti C, Amos D, Beatty P G et al 1989 Effect of HLA compatibility on engraftment of bone marrow transplants in patients with leukemia or lymphoma. New England Journal of Medicine 3u): 197-204 Tiercy J-M, B&uel H, Zwahlen F, Jeannet M, Mach B 1988 Oligonucleotide typing as a tool for improved HLA class II matching in bone marrow transplantation with unrelated donors. Bone marrow transplantation 3 (supplement 1): _^ 114-115 Baxter-Lowe L A, Hunter J, Casper J T, Gorski J 1989 HLA nene amolification and hvbridization analvsis of polym&phismI Journal of Clinical Investigations 84: 613-618 Sterkers G, Tiercy J-M, Zeliszewski D, Levy J-P, Mach B 1989 Characterization of three functional sites in a81 DR of DRwl3. All three sites are potentially involved in MHCpeptide interaction. European Journal of Immunology 19: 1585-1590 Urlacher A, Tiercy J-M, Schlesier M et al 1989 A T cell clone recognizing an MLC stimulating epitope located on the DRwl lb1 chain. Human Immunology 26: 321-332 Irle C, Jaques D, Tiercy J-M et al 1988 Functional polymorphism of each of two HLA-DRS chain loci demonstrated with antigen-specific DR3- and DRw52restricted T cell clones. Journal of Experimental Medicine 167: 853-872 Morel P A, Dorman J S, Todd J A, McDevitt H 0, Trucco M 1988 Aspartic acid at position 57 of the HLA-DQS chain protects against type I diabetes: a family study. Proceedings of the National Academy of Sciences of the USA 85: 8111-8115 Todd J A, Mijovic C, Fletcher J et al 1989 Identification of susceptibility loci for insulin-dependant diabetes mellitus by trans-racial gene mapping. Nature 338: 587-589 Nepom G T, Byers P, Seyfried C 1989 HLA genes associated with rheumatoid arthritis. Arthritis and Rheumatism 32: 15-21 Watanabe Y, Tokunaga K, Matsuki K et al 1989 Putative amino acid sequence of HLA-DRB chain contributing to rheumatoid arthritis susceptibility. Journal of Experimental Medicine 169: 2263-2268 Scharf S J, Freidmann A, Steinman L, Brautbar C, Erlich H A 1989 Specific HLA-DQBI alleles confer susceptibility to pemphigus vulgaris. Proceedings of the National Academy of Sciences of the USA 86: 6215-6219 Bugawan T L, Angelini G, Larrick J et al 1989 A combination of a particular HLA-DPB allele and an HLADQ heterodimer confers susceptibility to coeliac disease. Nature 339: 470-413 Sollid L M, Markussen G, Ek J et al 1989 Evidence for a primary association of coeliac disease to a particular HLADQ a/8 heterodimer. Journal of Experimental Medicine 169: 345-350 Vartdal F, Sollid L M, Vandvik B, Markussen G, Thorsby E 1989 Patients with multiple sclerosis carry DQBl genes which encode shared polymorphic amino acid sequences. Human Immunology 25: 103- 110