1. ^*Department of Clinical and Experimental Medicine, Dermatology Section, Bologna, Italy
2. ^†Microbiology Section, University of Bologna, Bologna, Italy

Correspondence: Dr Michelangelo La Placa, Department of Clinical and Experimental Medicine, Dermatology Section, University of Bologna, Via Massarenti, 1-40138 Bologna, Italy. Email: dermolap@med.unibo.it

Received 15 January 2004; Revised 24 March 2004; Accepted 29 March 2004; Published online 6 July 2004.

To the Editor:

Several studies have demonstrated the presence of actively replicating human endogenous retroviruses associated with various autoimmune disorders (^{Perron and Seigneurin, 1999;Portis, 2002}), in particular, Sjögrens's syndrome, rheumatoid arthritis, type 1 or insulin-dependent diabetes mellitus, multiple sclerosis, systemic lupus erythematosus, and systemic sclerosis. Serum antibodies specific for human intracisternal A-type particles (HIAP), a human endogenous retrovirus recognized by monoclonal antibody against HIV-1 p24 capsid protein (^{Garry et al, 1990}), have been detected significantly in association with primary biliary cirrhosis (^{Mason et al, 1998}) and systemic sclerosis (^{La Placa et al, 2004}).

Alopecia areata (AA) is a non-scarring hair disorder of unknown etiology. Various hypotheses have been involved in the pathogenesis of AA, including environmental, and psychological factors, but all these observations remain inconsistent (^{Duvic et al, 2001}). An exogenous viral infection has also been postulated by some authors (^{Skinner et al, 1995}), but the results have not been confirmed by other studies (^{Tosti et al, 1996}). Genetic factors have been linked to disease susceptibility and severity (^{Duvic et al, 2001}). Current evidence suggests that AA is an autoimmune disease, due to an aberrant T cell response against hair follicle self-antigens (^{McElwee et al, 1999;Kalish and Gilhar, 2003;Hordinsky and Ericson, 2004}).

The aim of our study was to evaluate the presence of serum antibody reactivity to HIAP proteins in patients affected by AA.

A total of 34 serum samples from patients with AA, and 20 control serum samples from healthy subjects, were examined. All the sera were collected between October 2002 and July 2003, after obtaining informed consent, and stored at -80°C, without preservatives, until use. AA patients enrolled in the study included 14 male and 20 female subjects (age range: 10–60 y; median age 29.7). All patients fulfilled the AA diagnostic criteria as described in the literature (^{Duvic et al, 2001}). In particular, 23 were affected by alopecia universalis, with total scalp and body hair loss, and 11 patients had alopecia totalis, with total scalp hair loss without body involvement. The duration of disease was between 1 and 12 y, with a median duration of 2.8 y. All patients included in the study were HIV-1/2 negative and HTLV–I and II negative (HIV Blot 2.2 Genelabs Diagnostics, Singapore; HTLV I/II Immunoblot, Cambridge Biotech, Worchester, Massachusetts), and none was affected by any other autoimmune disorder. Records of atopy were present in the anamnesis of 11 patients. Healthy controls were represented by 8 male and 12 female blood donors (age range: 19–45 y; median age 27.5).

Serum antibody reactivity to HIAP proteins was studied by western blot (immunoblot) analysis performed as previously described (^{La Placa et al, 2004}). Briefly, uninfected and HIAP-infected RH9/MC lymphoblastoid cells were obtained from the Cell Culture Collection Center of the Istituto Zooprofilattico Sperimentale di Brescia (Brescia, Italy) and grown in RPMI medium (GIBCO, Grand Island, New York) plus 10% fetal calf serum. The microsomal fraction was obtained, from both types of cell cultures, by disruption of cells in a hypotonic buffer followed by sucrose density gradient ultracentrifugation. Microsomal fraction proteins were resolved by sodium dodecyl sulfate/12% polyacrylamide gel electrophoresis and blotted onto nitrocellulose strips. Nitrocellulose strips were incubated overnight, at room temperature, in the presence of 2 mL of blotting buffer (Genelabs Diagnostics) containing 20 L of serum sample. Nitrocellulose strips were then washed and further incubated for 1 h at room temperature with peroxidase-conjugated anti-human IgG rabbit serum (Sigma, Milan, Italy) in PBS containing 1% of bovine serum albumin, and specific results were revealed by employing enhanced chemiluminescent western blotting detection reagents (Amersham, Arlington Heights, Illinois). All the sera were tested with immunoblots from both HIAP-infected and uninfected cells, to assess specific reactivity to HIAP proteins, and all the tests were performed in duplicate, to check reproducibility. HIAP immunoblots were considered positive when serum samples reacted with one or more proteins with electrophoretic mobility corresponding to that of known HIAP proteins (p17, p24, p30, p46, p60, p80, p84, p97). In addition, all sera were screened for anti-thyroid antibodies and antinuclear antibodies (ANA) including anti-centromere IgG antibodies (evaluated by microscopy with an image analyzer, after indirect immunofluorescence on commercially available tissue slides—Kallenstad/Sanofi Diagnostic Pasteur, Chaska, Minnesota), and various extractable nuclear antigens (ENAs) including Scl-70, Ro(SS-A), La(SS-B), Sm, U1-nRNP, and Jo-1, using standard procedures with a commercially available reagent kit (ProfilePlus, Euroimmun GmbH, Lubeck, Germany). Statistical analysis of the data was performed utilizing the Fisher Exact Test.

None of the sera examined showed the presence of autoantibodies reacting with thyroid antigens, ANA or ENAs. Western blot reactivity to one or more HIAP proteins was repeatedly detected in 25 of 34 (73.5%) AA patients, and only in two of 20 (10%) blood donors (p=0.0009). In particular, most of the HIAP-positive AA serum samples reacted against p30 (14 samples), p46 (23 samples), and p60 (11 samples). None of the sera examined showed the presence of antibody reactivity against proteins resolved from the microsomal fraction of HIAP-uninfected RH9 cells. Representative immunoblot results are illustrated in Figure 1. The data obtained in the entire population examined are summarized in Table I and Table II.

Figure 1.

Representative results of western blot analysis of the presence of antibodies against HIAP proteins in serum samples from AA patients and healthy controls (blood donors).Lanes 1–16: HIAP-positive AA patients. Lanes 17–19: HIAP-negative control subjects (blood donors). Lane 20: serum sample reacting with only one HIAP protein.

Full figure and legend (32K)

Table I - Western blot antibody reactivity to human intracisternal A-type particles (HIAP) proteins in serum samples from alopecia areata (AA) patients and control healthy blood donor (BD) subjects.

Full table

Table II - Western blot antibody reactivity against individual human intracisternal A-type particles (HIAP) proteins in serum samples from alopecia areata (AA) patients and healthy controls.

Full table

The results of our study, obtained from reproducible and specific immunoblot analyses, demonstrated the presence of serum antibody against HIAP endogenous retrovirus proteins in a significantly (p=0.0009) high percentage (73.5%) of AA patients. None of the HIAP-positive samples had immunoblot serum reactivity against microsomal fraction proteins from the uninfected cell line, thus demonstrating the absolute specificity of immunoblot serum reactivity against HIAP-infected cell microsomal proteins.

It is not easy, however, to assess definitely the etiopathogenetic significance of the findings of our study, and more especially, to assess if the activation of an endogenous retrovirus such as HIAP may be somehow related to the outcome of AA. Exactly which antigens are actually involved in inducing an HIAP-reacting antibody response remains an open question. The presence of IgG antibody reactivity to HIAP proteins may, in fact, be either attributable to the presence of an autoimmune response to antigenically related cellular proteins, or to the production of largely cross-reacting antibodies due to the immune dysregulation associated with autoimmune disease. Alternatively, the expression of HIAP proteins, and the consequent induction of a specific immune response, might be merely the result of incidental HIAP activation induced by attendant pathological processes.

In conclusion, our study has shown that the presence of serum antibodies reacting with human intracisternal A-type (HIAP) retrovirus proteins is significantly associated with AA. Whether these results indicate the possible involvement of HIAP endogenous retrovirus in the origin and development of AA is still an open question. There is, however, a growing body of evidence suggesting that serum antibodies to endogenous retrovirus proteins might be considered a hallmark of a large array of autoimmune diseases and our findings indicate that AA should also be included in this group of disorders. Increased and concerted research efforts are required to clarify definitively the meaning of anti-endogenous retrovirus protein antibody response in the etiopathogenesis of AA and other autoimmune pathologies.