684 ARTHRITIS & RHEUMATISM Volume 38 Number 5, May 1995, pp 684689 0 1995, American College of Rheumatology

SUSCEPTIBILITY TO REITER’S SYNDROME IS ASSOCIATED WITH ALLELES OF TAP GENES

KARYL S. BARRON, JOHN D. REVEILLE, MARY CARRINGTON, DEAN L. MA”, and MARY ANN ROBINSON

Objective. Although HLA-B27 is strongly associ- ated with susceptibility to Reiter’s syndrome (RS), recent data suggest that an additional modifying or susceptibility gene(s) acts in concert with HLA-B27 to contribute to disease pathogenesis. The recently de- scribed TAP genes (transporters associated with antigen processing) are potential candidates because they are polymorphic and their function is to transport antigenic peptides to be loaded in HLA class I molecules.

Methods. TAPl and TAP2 alleles were deter- mined for 34 patients with RS (28 HLA-B27 positive, 6 HLA-B27 negative), and their frequencies were com- pared with those observed for 52 HLA-B27 positive and 80 random disease-free control subjects.

Results. The allele frequency of TAPlC was greater in patients with RS (8 of 62, 13%) than in random controls (5 of 160, 3%) (P = 0.009). The frequency of TAP2A was greater in RS patients (51 of 66,77%) than in random controls (88 of 160,55%) (P = 0.002); likewise, the frequency was greater in HLA-B27 positive RS patients (41 of 54, 76%) than in HLA-B27 positive disease-free controls (49 of 94, 52%) (P = 0.004). Furthermore, the TAP2A allele was present in all RS patients (loo%), whereas TAP2A was present in 79% (63 of 80) of the random controls (P = 0.003).

Conclusion. The association observed between TAP alleles and RS is independent of the presence of

Karyl S. Barron, MD. Mary Ann Robinson, PhD: National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland; John D. Reveille, MD: University of Texas Health Science Center at Houston; Mary Camngton, PhD: Biological Carcinogenesis and Development Program, Program Re- sources, IncJDynCorp, and the National Cancer Institute, Freder- ick Cancer Research and Development Center, Frederick, Mary- land; Dean L. Mann, MD: National Cancer Institute, Frederick Cancer Research and Development Center.

Address reprint requests to Karyl S. Barron, MD, Labora- tory of Immunogenetics, National Institute of Allergy and Infectious Diseases, Twinbrook I1 Facility, 12441 Parklawn Drive, Rockville, MD 20852.

Submitted for publication July 18, 1994; accepted in revised form November 14, 1994.

HLA-B27, and despite the physical proximity of TAP and HLA class I1 genes, linkage disequilibrium does not account for the observed associations between TAP and RS. Thus, TAP genes are genetically separated but functionally linked to class I genes, and both contribute to susceptibility to RS.

Reiter’s syndrome (RS) can occur as a sequela to certain enteric and venereal infections (l), and the majority of individuals affected are HLA-B27 positive (2). These observations suggest that the disease pro- cess has its basis in an immune-mediated mecha- nism(s). Antigenic mimicry is an attractive hypothesis to explain the disease mechanism; it proposes that an infectious agent(s) elicits an immune response that recognizes ‘‘self’ antigens presented by HLA-B27, leading to the destruction of tissue by cytotoxic T cells.

Cytotoxic T cells are CD8+ and recognize foreign and self antigens presented as peptides by HLA class I molecules. These peptides are generated in the cellular cytoplasm, presumably by a proteasome- like structure called the low molecular mass peptide (LMP) (3-8) and then are actively transported across the endoplasmic reticulum by a heterodimer that is composed of the products of the TAPl and TAP2 genes (transporters associated with antigen process- ing) (9). In the endoplasmic reticulum, the peptides are assembled with the HLA heavy and light chains (lO,ll), and the assembled molecule is then trans- ported to the cell surface (7,12). The TAP proteins are members of the ABC (ATP-binding cassette) super- family of transporters, which are characterized by a conserved hydrophilic cytoplasmic ATP binding do- main (9-14). This domain is thought to couple energy to a hydrophobic domain in which 5-8 transmembrane segments form part of a transport channel (1 1).

Because of their location within the polymor- phic major histocompatibility complex (MHC) class I1 region, the role of polymorphism within the genes

TAP ALLELES AND REITER’S SYNDROME 685

encoding the TAP proteins has been questioned. In the rat, polymorphism of the cim genes, the equivalent of the human TAP genes, can alter the peptides bound to a specific class I molecule and thus alter antigen presentation (15,16). Polymorphism of the human TAP genes has also been reported; however, variation in human TAP alleles is less extensive than that observed in the rat (13,17-20). Functional differences among human TAP alleles have yet to be revealed. However, considering the functional relationship between TAP gene products and HLA class I molecules, the possi- bility exists that TAP polymorphism plays a role in diseases that have an association with class I alleles (21).

In the present study, TAP alleles were charac- terized in patients with Reiter’s syndrome, a disease associated with HLA class I allele B27. The purpose of the study was to determine whether specific TAP alleles were associated with RS and, if found, whether an association was dependent or independent of the presence of HLA-B27.

PATIENTS AND METHODS

Patient population. The study population consisted of 34 patients with RS, as defined by the working definition proposed by Calin (22). Patients were followed by the Divisions of Rheumatology at the University of Texas Health Science Center at Houston or at the University of Alabama at Birmingham. Of the 34 patients, 28 were HLA- B27 positive. The control population included 80 unrelated individuals from CEPH (Centre d’Etude du Polymorphisme Humain). The HLA-B27 positive control group included 52 healthy individuals with no family history of spondylarthro- pathy who were identified as HLA-B27 positive by routine serologic typing.

HLA characterization. HLA class I characterizations were performed using standard serologic techniques. The presence of B27 in RS patients was confirmed by oligonucleo- tide methods (23). HLA-DQA1 and DQBl alleles were determined using oligonucleotide typing methods for genomic DNA, as previously described (24). The nomencla- ture used was established by the Eleventh International Histocompatibility Testing Workshop (25).

TAPl and TAP2 characterization. Polymorphism in the coding regions of the TAP genes was detected using the single-strand conformational polymorphism (SSCP) proce- dure (26). Genomic DNA was extracted from peripheral blood leukocytes as described (27). TAP genes were ampli- fied from genomic DNA by the polymerase chain reaction (PCR) procedure (28) using the following oligodeoxynucleo- tide primer pairs (5’ -+ 3’): CACCCTGAGTGATTCTCT and ACTGAGTCTGCCAAGTCT to determine polymorphism at nucleotide position 1069 of TAPl (amino acid 333); CCCTATCCAGCTACAACC and TTTCCGGATCAATGC- TCGGG to determine polymorphism at nucleotide position 1982 of TAPl (amino acid 637); GCGGAGAGACCTG-

GAACG and TCAGCATCAGCATCTGCA to determine polymorphism at nucleotide position 1231 of TAP2 (amino acid 379); and ACAGTGCTGGTGATTGCT and CACAGC- TCTAGGGAAACT to determine polymorphism at nucleo- tide positions 2089 and 2155 of TAP2 (amino acids 665 and 687).

Reaction mixtures (20 pl) containing DNA (200 ng), oligonucleotide primers (1 pM), dNTP (200 pM; Pharmacia, Piscataway, NJ), Tris HC1 (10 mM, pH 8.4), KCl (50 mM), MgCl, (2 mM), gelatin (100 pg/ml), DNA polymerase from Thermus aquaticus (OSU; Perkin-Elmer Cetus, Norwalk, CT), and 1 pCi of 32P-dCTP were subjected to 30 cycles of 95°C for 30 seconds (denaturation), 60°C for 30 seconds (annealing), and 72°C for 90 seconds (extension). An aliquot of amplified DNA was diluted 1 : 10 in 0.1% sodium dodecyl sulfate and 10 mM EDTA, then mixed 1:2 with 95% forma- mide, 20 pM EDTA, 0.05% bromphenol blue, and 0.5% xylene cyanol. Immediately before loading, samples were heated at 95°C for 2 minutes.

The different allelic forms were resolved by running 5 pl of a reaction mixture on a 5% Long Ranger acrylamide gel (AT Biochem, Malvern, PA) containing 10% glycerol and 0 . 6 ~ TBE (60 mM Tris, 50 mM boric acid, 0.6 mM EDTA) run at room temperature for 2.5-3 hours at 30W.

PCR products from DNA obtained from homozygous- typing cell lines with previously characterized TAP alleles were used as standards in the SSCP assay (18,20).

Statistical analysis. Comparisons of TAP allele fre- quencies in RS patients and control groups were performed by chi-square analysis on 2 x 2 tables with Fisher’s exact test where appropriate, using the Epi-Info statistical pro- gram. Relative risks were calculated as odds ratios. Haldane’s correction was used when necessary.

RESULTS

Four potential alleles of TAPl and 4 alleles of TAP2 genes can be differentiated by combinations of codon differences at 2 sites in TAPl and at 3 sites in TAP2 (Figure 1). To date, the substitution defining TAP2.3 has been found to occur only with those defining TAP2.5; likewise, TAP2.4 has been observed only with TAP2.6 (21). The apparent linkage of sub- stitutions at these 2 sites reduces the number of potential TAP2 alleles to the 4 combinations shown in Figure 1. Of particular note, TAP2A and 2C are characterized by a shortened cytoplasmic tail because of a substitution resulting in a stop codon.

The substitutions present in TAPl and TAP2 genes were determined in RS patients and control groups. Frequencies of individual substitutions did not vary significantly between patient and control groups, with a single exception (Table 1). A significantly higher frequency of TAP2.4/2.6 was observed in RS patients compared with the control population (81% versus 67%; P = 0.04). This difference was also evident when

686

Val (2.2)

Ala (2.31

BARRON ET AL

Ile Ile (2.11 (2.1)

Thr Ala (2 4) (2.3)

TAP 1

TAP 2

687 Gln (2.51 / stop (2.6)

POTENTIAL ALLELES

(11) (1.2) (1.2) (1.1)

Asp Gly Asp (1.31 (1.4) (1.3) (1.4)

POTENTIAL ALLELES

2A - vat (2 21

Thr (2 41

stop (2 61 __

2 8 2C 2D

Figure 1. Schematic diagram of the putative membrane-spanning segments and ATP-binding cassette of TAPl and T A R molecules in the endoplasmic reticulum (left) and potential TAP alleles (right). Allele marked * has not been observed.

B27+ RS patients were compared with B27+ controls (80% versus 61%; P = 0.01). As expected, the fre- quency of TAP2.3/2.5 was decreased in patients com- pared with either control group.

TAP alleles were assigned based on combina- tions of the individual substitutions. Comparisons of the frequencies of TAPl and TAP2 alleles in patient and control groups relative to the number of chromo- somes where alleles could be assigned are shown in Table 2. TAP2A was significantly increased in fre- quency in RS patients compared with random controls (77% versus 55%) and in B27+ RS patients compared with B27+ control subjects (76% versus 52%). RS patients were more frequently homozygous for TAP2A than were individuals in the control population (52% versus 31%, P = 0.05). The frequency of TAPlC

was significantly higher in RS patients than in random controls (13% versus 3%), while the frequency of TAPlB was significantly lower in RS patients than in controls (3% versus 13%).

The frequencies of TAPl and TAP2 alleles relative to the numbers of subjects in the study groups are shown in Table 3. The frequency of TAPlC was again greater in RS patients compared with controls (23% versus 5%), while the frequency of TAPlB was lower in RS patients (6% versus 25%). The most striking observation was that 100% of the RS patients had at least one TAP2A allele, whereas 79% of the random control group and 81% of the B27+ control group had TAP2A. Comparisons of the allele frequen- cies in B27+ RS patients and B27+ disease-free individuals demonstrated that the association of TAP2A with Reiter’s syndrome is independent of the presence of HLA-B27 (Table 3).

When random controls were compared with B27+ controls, similar frequencies for the TAP alleles were demonstrated, except for TAPl B, which was significantly decreased in frequency in B27+ controls (2%, versus 25% in random controls; P = 0.001). While there was a trend toward an increased fre- quency of TAPlC in healthy B27+ controls, the difference was not statistically significant. Thus, the increased frequency of TAP2A in RS patients is inde- pendent of the presence of HLA-B27, whereas changes in frequencies of TAPlB and 1C in RS pa- tients may be related to the presence of HLA-B27.

The increased frequency of TAP2A in RS pa- tients was likewise found to be independent of HLA- DQ. DQAl and DQBl alleles were characterized in 28 HLA-B27+ RS patients, 80 controls, and 51 HLA- B27+ controls (data not shown). There were no sig- nificant differences in the frequencies of DQAl or DQBl in controls and RS patients. The only significant differences observed between B27+ patients and B27+

Table 1. patients and control subjects with and without HLA-B27*

Comparisons of frequencies of substitution in TAP2 genes in Reiter’s syndrome (RS)

All study subjects B27 positive subjects

Controls RS patients Controls RS patients TAP2 (n = 160) (n = 68) (n = 102) (n = 56)

2.1 22 (14) 4 (6) 11 (11) 4 (7) 2.2 138 (86) 64 (94) 91 (89) 52 (93) 2.312.5 52 (33) 13 (19) 40 (39) 11 (20) 2.412.6 108 (67) 55 (81)t 62 (61) 45 (80)$

* Values are the number (%) positive; n values are the number of chromosomes. t P = 0.04, odds ratio (OR) = 2.0, 95% confidence interval (95% CI) = 1.22-3.42. $ P = 0.01, OR = 2.6, 95% CI = 1.16-6.3.

TAP ALLELES AND REITER’S SYNDROME 687

Table 2. control subjects with and without HLA-B27*

Comparisons of TAPl and TAP2 allele frequencies in Reiter’s syndrome (RS) patients and

All study subjects B27 positive subjects

Allele Controls RS patients Controls RS patients

TAP I No. of chromosomes 160 62 86 52 TAP 1 A 134 (84) 52 (84) 78 (91) 42 (81) TAPlB 21 (13) 2 (3)t 1(1) 2 (4) TAPlC 5 (3) 8 (13)$ 7 (8) 8 (15)

No. of chromosomes 160 66 94 54 TAP2A 88 (55) 51 (77)§ 49 (52) 41 (7617 TAP2B 51 (32) 12 (18) 37 (39) 10 (19)# TAP2C 20 (13) 3 (5) 8 (9) 3 (6) TAP2D 1 (1) 0 (0) 0 (0) 0 (0)

TAP2

* It was not possible to assign alleles to those who were heterozygous for both substitutions in a TAP gene. TAPl alleles were not assigned in 2 of 28 B27+ RS patients, 1 of 6 B27- patients, and 5 of 48 B27+ controls. TAP2 alleles were not assigned in 1 of 28 B27+ patients and 3 of 50 B27+ controls. Values are the number (%) positive. t P = 0.03, odds ratio (OR) = 0.22, 95% confidence interval (95% CI) = 0.02-0.95. f P = 0.009, OR = 4.6, 95% CI = 1.25-18.5. 5 P = 0.002, OR = 2.8, 95% CI = 1.39-5.8. T P = 0.004, OR = 2.9, 95% CI = 1.3-6.6. # P = 0.009, OR = 0.35, 95% CI = 0.14-0.82.

controls was a higher frequency of DQA1*0501 in B27+ patients (36%) compared with B27+ controls (22%) (P = 0.05, relative risk = 2.0). This could be explained by linkage disequilibrium between DQAl*O501 and TAP2A (29).

DISCUSSION

It has long been recognized that the majority of patients with Reiter’s syndrome, as well as patients

with a number of other spondylarthropathies, are HLA-B27 positive (1,2). The present report shows that alleles of other genes mapping to the MHC, namely TAP2A and TAPlC, are also higher in fre- quency in this syndrome. TAP2A was present in 100% of the RS patients, and its higher frequency was independent of the presence of HLA-B27. Thus, spe- cific TAP alleles may contribute to the pathogenic disease process by having functional complementarity with HLA class I molecules.

Table 3. Comparison of phenotype frequency for TAP alleles in Reiter’s syndrome (RS) patients and control subjects with and without HLA-B27*

All study subjects B27 positive subjects

Controls RS patients Controls RS patients

TAPl No. of subjects TAPlA TAPlB TAPlC

No. of subjects TAP2A TAP2B TAP2C TAP2D

TAP2

80 78 (98) 20 (25) 5 (5)

80 63 (79) 43 (54) 18 (23) 1 (1)

* Values are the number (%) positive. It was not possible to assign alleles to all subjects. t P = 0.03, odds ratio (OR) = 0.21, 95% confidence interval (95% CI) = 0.02-0.96.

I P = 0.003, OR = 18.5, 95% CI = 2.94-760.87. 7 P = 0.02, OR = 13.6, 95% CI = 2.01-573.94.

f P = 0.03, OR = 4.4, 95% CI = 1.07-18.92.

# P = 0.04, OR = 0.37, 95% CI = 0.12-1.08.

688 BARRON ET AL

The importance of the TAP gene products and their functional relationship to HLA class I gene products was recognized when cell lines depleted of genes in the MHC class I1 region that mapped between HLA-DQB2 and HLA-DPA1 genes failed to present antigens to cytotoxic T cells (9,lO). Subsequent stud- ies demonstrated that the TAPl and TAP2 gene prod- ucts belonged to the family of ATP-binding membrane transporters and that their function as peptide trans- porters was important in the stable expression of HLA class I molecules at the cell surface (30).

Both TAPl and TAP2 are polymorphic (13,17- 20); however, the extent to which this polymorphism relates to functional differences remains to be com- pletely clarified. Homologs of the human TAPl and TAP2 genes have been characterized in mice and rats (15,16,31,32). In rats, differences in the spectrum of peptides were found in class I molecules when differ- ent antigen transport alleles were present in the cell (15,16). Comparable studies have not been reported in humans.

The results presented in this report suggest that functional differences may be associated with different TAP alleles. The mechanism for the association of the TAP2A polymorphism and Reiter’s syndrome may be due to the favored transport of particular peptides that have increased affinity for the HLA-B27 molecule. Cell lines that express identical HLA-B27 molecules have been shown to differ in their ability to present certain viral epitopes in the context of B27 molecules (33). This observation led to the speculation that genetic differences in transporter genes might be re- sponsible for this observed difference in HLA-B27 recognition (20). While direct evidence for a restricted peptide-transport function of the TAP alleles associ- ated with RS has not been presented, it seems likely that a degree of specificity is present and effects disease when both HLA-B27 and TAP2A are present in a host that has been challenged with certain envi- ronmental agents.

This is the first report of an association between TAP alleles and an HLA class I-associated disease. The potential association of TAP alleles with other diseases has been reported, but those studies con- cerned HLA class 11-associated diseases (17). A TAP2 association with insulin-dependent diabetes was found in 2 studies (20,34) but not in another (35). In addition, no independent association of TAP alleles was found with celiac disease (36) and juvenile rheumatoid arthri- tis (37). Although no association of HLA class I1 alleles with Reiter’s syndrome has been reported, the

possibility was considered that the observed associa- tion between RS and TAP2A may be due to linkage disequilibrium between specific alleles of HLA-DQB 1 and TAP alleles (29). The patient and control groups were typed for DQAl and DQB 1 alleles. The frequen- cies of these alleles in the disease and disease-free groups were similar, with the exception of DQA1*0501, which was more frequent in the B27+ RS patients than in the B27+ healthy controls. Thus, linkage disequili- brium between HLA class I1 and TAP2 alleles did not account for the observed increase in frequency of the TAP alleles in Reiter’s syndrome patients. However, the possibility that linked genes other than those currently recognized contribute to the disease process cannot be excluded.

Although TAP genes are encoded within the MHC, the association observed between TAP alleles and RS is independent of the presence of HLA-B27. Furthermore, linkage disequilibrium between TAP and HLA class I1 alleles does not account for the observed associations between TAP and RS. Thus, TAP is a genetically separate, but a functionally linked, genetic system that contributes to susceptibility to RS. The results of the present study further the concept that individual genes or clusters of genes within the MHC functionally interact even though they are encoded within separate subregions. The extensive polymor- phism and evolutionary selection of certain combina- tions of genes in the HLA gene complex may have biologic significance in disease processes such as Reiter’s syndrome.

ACKNOWLEDGMENTS

The authors thank Drs. Frank Arnett, Noranna Warner, and Paul Katzenstein, Division of Rheumatology and Clinical Immunogenetics (University of Texas, Hous- ton) for permitting us to study their patients, and Dr. Thomas Kindt, Laboratory of Immunogenetics (NIAID, NIH) for his critical review of the manuscript. Genomic DNA from HLA-B27 positive controls was kindly provided by Drs. Joseph Ahearn and Wilma Bias at the Johns Hopkins University School of Medicine and from control populations from the Laboratory of Viral Carcinogenesis (Frederick, MD). We also thank Dr. Ed Giannini for his assistance with the statistical analysis.

REFERENCES

1 . Reveille JD: The interplay of nature versus nurture in predis- position to the rheumatic diseases. Rheum Dis Clin North Am

2. Ivanyi P: Immunogenetics of the spondyloarthropathies. Curr 19~15-27, 1993

Opin Rheumatol 5:436445, 1993

TAP ALLELES AND REITER’S SYNDROME 689

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

Monaco JJ, McDevitt HO: Identification of a fourth class of proteins linked to the murine histocompatibility complex. Proc Natl Acad Sci U S A 79:3001-3005, 1982 Arrigo A-P, Tanaka K, Goldberg AL, Welch WJ: Identity of the 19s ‘prosome’ particle with the large multifunctional protease complex of mammalian cells (the proteasome). Nature 331: 192- 194, 1988 Robertson M: Proteasomes in the pathway. Nature 353:300- 301, 1991 Brown MG, Driscoll J, Monaco JJ: Structural and serological similarity of MHC-linked LMP and proteasome (multicatalytic proteinase) complexes. Nature 353:355-357, 1991 Glynne R, Powis SH, Beck S, Kelly A, Kerr L-A, Trowsdale J: A proteasome-related gene between the two ABC transporter loci in the class I1 region of the human MHC. Nature 353:357- 360, 1991 Kelly A, Powis SH, Glynne R, Radley E, Beck S, Trowsdale J: Second proteasome-related gene in the human MHC class I1 region. Nature 353:667468, 1991 Spies T, Bresnahan M, Bahram S, Arnold D, Blanck G, Mellins E, Pious D, DeMars R: A gene in the human major histocom- patibility class I1 region controlling the class I antigen presen- tation pathway. Nature 348:744747, 1990 Trowsdale J , Hanson I, Mockridge I, Beck S, Townsend A, Kelly A: Sequences encoded in the class I1 region of the MHC related to the ‘ABC’ superfamily of transporters. Nature 348:

Bahram S, Arnold D, Bresnahan M, Strominger JL, Spies T: Two putative subunits of a peptide pump encoded in the human major histocompatibility complex class I1 region. Proc Natl Acad Sci U S A 88:10094-10098, 1991 Kleijmeer MJ, Kelly A, Geuze HJ, Slot JW, Townsend A, Trowsdale J: Location of MHC-encoded transporters in the endoplasmic reticulum and cis-Golgi. Nature 357:342-344, 1992 Powis SH, Mockridge I, Kelly A, Kerr L-A, Glynne R, Gileadi U, Beck S, Trowsdale J: Polymorphism in a second ABC transporter gene located within the class I1 region of the human major histocompatibility complex. Proc Natl Acad Sci U S A 89:1463-1467, 1992 Deverson EV, Gow IR, Coadwell WJ, Monaco JJ, Butcher GW, Howard JC: MHC class I1 region encoding proteins related to the multidrug resistance family of transmembrane transporters. Nature 348:738-741, 1990 Powis SJ, Howard JC, Butcher GW: The major histocompati- bility complex class I1 linked cim loci controls the kinetics of

741-744, 1990

intracellular transport of a classical class I molecule. J Exp Med 173:913-921, 1991 Powis SJ, Deverson EV, Coadwell WJ, Ciruela A, Huskisson NS, Smith H, Butcher GW, Howard JC: Effect of polymor- phism of an MHC-linked transporter on peptides assembled in a class I molecule. Nature 357:211-215, 1992 Colonna M, Bresnahan M, Bahram S, Strominger JL: Allelic variants of the human putative peptide transporter involved in antigen processing. Proc Natl Acad Sci U S A 89:3932-3936, 1992 Powis SH, Tonks S, Mockridge I, Kelly AP, Bodmer JG, Trowsdale J: Alleles and haplotypes of the MHC-encoded ABC transporters TAPl and TAP2. Immunogenetics 37:373-380, 1993 Aoki Y, Isselbacher KJ, Pillai S: Polymorphisms involving the transmembrane domains of human TAP2. Immunogenetics 38: 382, 1993 Jackson DG, Capra JD: TAP1 alleles in insulin-dependent diabetes mellitus: a newly defined centromeric boundary of disease susceptibility. Proc Natl Acad Sci U S A 90:11079- 11083, 1993 Carrington M, Colonna M, Spies T, Stephens JC, Mann DL:

Haplotype variation of the transporter associated with antigen processing (TAP) genes and their extension of HLA class I1 region haplotypes. Immunogenetics 37:266-273, 1993

22. Calin A: Reiter’s syndrome. In, Spondyloarthropathies. Edited by A Calin. London, Grune and Stratton, 1984

23. Appleyard J, Ahearn JM, Taurog JD, Choo SY, Barron KS, Amett FC, Reveille JD: HLA-B27 subtypes in patients with spondyloarthropathy (abstract). Arthritis Rheum 35 (suppl 9): S58, 1992

24. Barron KS, Silverman ED, Gonzales J, Reveille JD: Clinical, serologic, and immunogenetic studies in childhood-onset sys- temic lupus erythematosus. Arthritis Rheum 36:348-354, 1993

25. Bodmer JG, Marsh SG, Alben ED, Bodmer WF, Dupont B, Erlich HA, Mach B, Mayr WR, Parham P, Sasazuki T, Schreuder GMT, Strominger JL, Svejgaard A, Terasaki PJ: Nomenclature for factors of the HLA system. Hum Immunol 34:4-18, 1992

26. Orita M, Suzuki Y, Sekiya T, Hayashi K: Rapid and sensitive detection of point mutations and DNA polymorphisms using the polymerase chain reaction. Genomics 5:874-879, 1989

27. Blin N, Stafford DW: A general method of isolation of high molecular weight DNA from eukaryotes. Nucleic Acids Res

28. Saiki RK, Gelfand DH, Stoffel S, Scharf J, Higuchi G, Horn T, Mullis KB, Erlich HA: Primer-directed enzymatic amplification of DNA with thermostable DN,4 polymerase. Science 239:487- 491, 1988

29. Carrington M, Stephens C, Klitz W, Begovich A, Erlich H, Mann D: MHC class I1 haplotypes and linkage disequilibrium values observed in the CEPH families. Hum Imrnunol41:23& 240, 1994

30. Kelly A, Powis SH, Kerr L-A, Mockridge I, Elliott T, Bastin J, Uchanska-Ziegler B, Ziegler A, Trowsdale J, Townsend A: Assembly and function of the 2 ABC transporter proteins encoded in the human major histocompatibility complex. Na- ture 355:641-644, 1992

31. Monaco JJ, Cho S, Attaya M: Transport protein genes in the murine MHC: possible implications for antigen processing. Science 250:1723-1726, 1990

32. Powis SJ, Townsend ARM, Deverson EV, Bastin J, Butcher GW, Howard JC: Restoration of antigen presentation to the mutant cell line RMA-S by an MHC-linked transporter. Nature 354528-531, 1991

33. Pazmany L, Rowland-Jones S, Huet S, Hill A, Sutton J, Murray R, Brooks J, McMichael A: Genetic modulation of antigen presentation by HLA-B27 molecules. J Exp Med 175:361-369, 1992

34. Caillat-Zucmann S, Bertin E, Timsit J , Boitard C, Assan R, Bach J - F Protection from insulin-dependent diabetes mellitus is linked to a peptide transporter gene. Eur J Immunol 23:1784 1788, 1993

35. RGnningen KS, Undlien DE, Ploski R, Maouni N, Konrad RJ, Jensen E, Hornes E, Reijonen H, Colonna M, Monos DS, Strominger JL, Thorsby E: Linkage disequilibrium between TAP2 variants and HLA class 11 alleles: no primary association between TAP2 variants and insulin dependent diabetes mellitus. Eur J Immunol 23:1050-1056, 1993

36. Powis SH, Rosenberg WMC, Hall M, Mockridge I, Tonks S, Ivinson A, Ciclitira PJ, Jewel1 DP, Lanchbury JS, Bell J1, Trowsdale J: TAPl and TAP2 polymorphism in coeliac disease. Immunogenetics 38:345-350, 1993

37. Ploski R, Undlien DE, Vinje 0, FGrre @, Thorsby E, Rgnningen KS: Polymorphism of human major histocompatibility complex- encoded transporter associated with antigen processing (TAP) genes and susceptibility to juvenile rheumatoid arthritis. Hum Immunol39:54-60, 1994

312303-2308, 1978