HLA represents an antigenic specificity comprising 23 different alleles, also called subtypes, named HLA–B*2701 to B*27020 (1). The first 10 subtypes, i.e., B*2701 to B*2710, have been found in at least 1 patient with ankylosing spondylitis (AS) or other spondylarthropathies (SpAs). Among these, however, only the relatively more common subtypes (e.g., B*2705, B*2704, B*2702, and B*2707) have been demonstrated in correct epidemiologic studies to be disease-associated (1). The remaining 13 (i.e., most recently identified) subtypes have not yet been studied for their disease association.

The B*2709 subtype, which is observed among Italians (primarily Sardinians), has been found in healthy individuals but not in AS patients in Sardinia (2). However, B*2709 positivity has been reported in 2 SpA patients living in mainland Italy (3, 4).

The aim of the current study was to evaluate the frequency and role of B*2709 in B27-positive patients in Lucania (Basilicata), a region of the southern Italian mainland. All consecutive serologically B27-positive SpA patients born and living in Lucania who were seen for the first time during a 12-month period in the new Rheumatology Department of Lucania were tested by genomic typing for B27 subtypes, according to methods described elsewhere (4). Twenty-six serologically B27-positive healthy bone marrow donors born and living in the same geographic area were similarly tested.

Forty-seven B27-positive SpA patients (33 men and 14 women mean age 44.8 years, range 21–72; mean disease duration 9.9 years, range 1–23) were seen during the 12-month period. Of these, 22 had undifferentiated SpA (uSpA), 14 primary AS, 10 psoriatic arthritis, and 1 SpA associated with ulcerative colitis. Of the 47 SpA patients, 35 (74.4%) showed B*2705, 8 (17%) B*2702, 2 (4.2%) B*2707, and 2 (4.2%) B*2709. Of the 26 healthy B27-positive bone marrow donors, 13 (50%) showed B*2705, 10 (38%) B*2702, 2 (7.2%) B*2709, and 1 (3.8%) B*2707. Both B*2709-positive patients had uSpA and lacked axial involvement. The odds ratio for each HLA–B27 subtype was 2.91 for B*2705, 1.10 for B*2707, 0.53 for B*2709, and 0.32 for B*2702.

Of the B27 subtypes, B*2705, B*2704, B*2702, and B*2707 are clearly associated with AS and the related SpAs (1). B*2706 and B*2709 subtypes are, at most, only weakly associated with SpA in Southeast Asia (5) and on the island of Sardinia in Italy (2), respectively. Recently, a report on families in which both B*2704 and B*2706 occurred suggested that B*2706 does not protect against disease in the presence of a predisposing subtype (6).

B*2709 has been found in Sardinia and in continental Italy, where the genotypic frequency of HLA–B27 in the general population is ∼2% (7). B*2709 accounts for 25% of HLA–B27 subtypes in Sardinia and 3% in continental Italy. D'Amato et al tested 35 Sardinian patients with AS and 40 Sardinian B27-positive healthy individuals by genomic typing (2). None of the AS patients were found to be B*2709-positive, in contrast to 25% of the healthy controls. The authors suggested that B*2709 is not associated with AS. B*2709 differs from B*2705 by a single His116Asp substitution in the F pocket of the peptide-binding groove area. D'Amato and colleagues (2) are of the opinion that the substitution at position 116 could exclude the acceptance of arthrogenic peptide by the B*2709 subtype. Garcia-Peydro et al (8) demonstrated that B*2705 and B*2709 share 80% of their alloantigenic T cell epitopes, implying that the putative arthritogenic peptide should belong to a limited subset of the B*2705-bound repertoire.

Last year we reported the case of a B*2709-positive woman who had uSpA without axial involvement but had erosive and disabling peripheral arthritis (4). We suggested that the negative association of B*2709 with AS found in Sardinia should be confirmed in other studies that include the full spectrum of disease and are not limited to AS. In the current study, we subtyped all 47 B27-positive SpA patients and 26 healthy B27-positive controls. Two patients and 2 healthy controls were found to have the B*2709 subtype. Both patients had uSpA without axial involvement. Thus, we have observed a total of 3 B27-positive SpA patients from mainland Italy who are B*2709-positive, but none of them had sacroiliitis or axial involvement. Although the odds ratio for developing SpA is only 0.53, our results suggest that the B*2709 subtype confers susceptibility to SpA. It is possible that B*2709 is protective against axial involvement, although a B*2709-positive patient with SpA was recently reported by another Italian group (3), and this patient has sacroiliitis and oligoarthritis. This protection should be confirmed in future studies.

Another aspect of our study that is noteworthy is the high frequency (38%) of the B*2702 subtype in healthy B27-positive individuals living in Lucania. The frequency of this subtype is 4–10% in B27-positive northern Europeans, 20% in persons who are Spanish and Portuguese, and approaches 55% in Semitic populations (Arabs and Jews) (1). A study from Marseilles, France, found B*2702 in 74% of AS patients who had Spanish and North African ancestors and in 21% of healthy B27-positive controls with similar ancestry (9). In contrast, the frequency of the B*2705 subtype was 28% in patients and 79% in controls. The authors concluded that in southern France, B*2702 is a susceptibility allele, and B*2705 is a protective allele for AS in patients of Spanish and North African descent. However, our results in Lucania, where the B*2702 subtype is found quite frequently, do not confirm that B*2702 confers a higher degree of disease risk than does B*2705.

In conclusion, our study of B27-positive SpA patients and B27-positive healthy individuals born and living in Lucania suggests that the B*2709 subtype confers susceptibility to SpA, and that B*2702 does not seem to confer a higher disease risk than does B*2705.

  • 1
    Ball EJ, Khan MA. HLA–B27 polymorphism. Joint Bone Spine 2001; 68: 37882.
  • 2
    D'Amato M, Fiorillo MT, Carcassi A, Mathieu A, Zuccarelli A, Bitti PP, et al. Relevance of residue 116 of HLA-B27 in determining susceptibility to ankylosing spondylitis. Eur J Immunol 1995; 25: 3199201.
  • 3
    Marchionni L, Modena V, Roggero R, Curtoni ES. The polymorphism of HLA-B27 and the seronegative spondyloarthropathies in the Italian population [abstract]. In: 13th European Histocompatibility Conference; Crete 1999. European Federation for Immunogenetics: 36.
  • 4
    Olivieri I, Padula A, Ciancio G, Moro L, Durante B, Gaudiano C, et al. The HLA-B*2709 subtype in a patient with undifferentiated spondarthritis [letter]. Ann Rheum Dis 2000; 59: 6545.
  • 5
    Nasution AR, Mardjuadi A, Kunmartini S, Suryadhana NG, Setyohadi B, Sudarsono D, et al. HLA-B27 subtypes positively and negatively associated with spondyloarthropathy. J Rheumatol 1997; 24: 11114.
  • 6
    Sudarsono D, Hadi S, Mardjuadi A, Nasution AR, Dekker-Sayes A, Breur-Vrisendorp BS, et al. Evidence that HLA-B*2706 is not protective against spondyloarthopathy. J Rheumatol 1999; 26: 19346.
  • 7
    Rendine S, Borelli I, Barbanti M, Sacchi M, Roggero R, Curtoni ES. HLA polymorphisms in Italian bone marrow donors: a regional analysis. Tissue Antigens 1998; 52: 13546.
  • 8
    Garcia-Peydro M, Marti M, Lopez de Castro JA. High T cell epitope sharing between two HLA-B27 subtypes (B*2505 and B*2709) differentially associated to ankylosing spondylitis. J Immunol 1999; 163: 2299305.
  • 9
    Guis-Sabatier S, Nielsen W, Boetsch G, Dutour O, Mercier P, Reviron D, et al. In Southern France, HLA-B*2702 carries susceptibility and HLA-B*2705 carries resistance to ankylosing spondylitis in patients of Spanish and North African descent [abstract]. Arthritis Rheum 2000; 43( Suppl): S266.