Antigenic specificity of chlorpromazine-induced antinuclear antibodies

A few patients have been reported who developed systemic lupus erythematosus (SLE) in the course of prolonged treatment with chlorpromazine. Patients on this drug have also been found to have antinuclear antibodies (ANAs) although they do not develop lupus.

We have studied the antigenic specificity of ANAs in fifty-four patients on longterm chlorpromazine treatment and compared our findings with those on 175 patients on anticonvulsants, 215 patients on isoniazid, 109 SLE patients and fifty-four healthy subjects, sex and age matched to the chlorpromazine patients.

Thirty-nine per cent of patients on chlorpromazine had ANAs which were most frequently directed to single stranded (s)DNA. In contrast, patients on anticonvulsants as well as those on isoniazid had ANA directed to soluble nucleoprotein (sNP) most frequently and none of the patients on isoniazid had ANA to sDNA.

The mechanisms by which chlorpromazine, isoniazid or anticonvulsant intake results in ANAs probably differ. Our findings suggest that development of ANAs in patients on chlorpromazine may be initiated by interaction of the drug with denatured DNA.

Antinuclear antibodies in patients on anticonvulsant therapy

Antinuclear antibodies to calf thymus nuclei, NP, DNA, sDNA, sNP and Sm antigen were investigated in sera from 170 patients on various programmes of prolonged anticonvulsant treatment. Findings were compared to those on 214 tuberculous patients on isoniazid, 109 SLE patients and 66 healthy subjects.

Patients on anticonvulsants had a significantly higher incidence of ANA to DNA, sDNA, sNP and Sm antigen than the controls but had a lower incidence of ANA to all antigens, except sNP, than the SLE patients. Patients on isoniazid did not have DNA antibodies, but had antibodies to whole nuclei and to NP which were practically absent in the anticonvulsant group.

Of all patients on anticonvulsants only those receiving hydantoins had ANA to Sm antigen, while those receiving only primidone had antibodies to sNP but no antibodies to DNA.

Alteration of sNP with isoniazid did not result in an increased incidence of ANA in the anticonvulsant group as it does in isoniazid treated subjects.

It is concluded that the SLE-activating properties of diverse anticonvulsants probably resides in their potential to induce ANA. Although all anticonvulsants elicit ANA directed primarily to sNP, each may do so by different mechanisms or by altering different sites in the sNP molecule. The mechanisms by which anticonvulsant and isoniazid intake results in ANA probably differ.

Presence of DNA antibodies in some patients on anticonvulsants may indicate that their convulsions were due to SLE.

Presence of hepatitis-associated antigen in systemic lupus erythematosus

Presence of hepatitis-associated antigen (HAA) was investigated in 504 sera from 116 patients with SLE and was found in 41% of them.

HAA was present in at least one serum in 75% of the patients but there were variations in presence and titres in the same patient at different times.

Except for a tendency of HAA to appear or rise in titre with lupusi nactivation following corticosteroid or immunosuppresive therapy, there was no correlation between its presence and disease activity, specific organ involvement, antinuclear antibodies or immunoglobulin levels.

All but one of twelve lupus patients with recurrent bacterial infections had HAA at high titres. HAA appeared in the serum of a patient upon development of IgA deficiency.

HAA antigenaemia in systemic lupus erythematosus seems a consequence rather than a cause of the immunological derangement in this disease.

Serum immunoglobulins in systemic lupus erythematosus

1. Activities of the three major classes of immunoglobulins were quantitatively determined in 481 sera from 106 patients with systemic lupus erythematosus residing in Mexico City and compared with activities in 112 healthy Mexico City residents.

2. To allow comparison with studies from other laboratories, results were related to a standard reference preparation from the World Health Organization and expressed in international units of immunoglobulin activity.

3. The mean activities of IgG and IgM, but not those of IgA, were significantly higher in the lupus sera than in controls. Abnormal elevation of IgG activity was related to the presence of antibodies to whole calf thymus nuclei, but not to those directed to nucleoprotein or DNA. Sera with rheumatoid factor activity had a higher mean IgA activity than those without it.

4. Patients with active lupus had significantly higher IgM activity than those with inactive disease. High IgM activities also correlated with presence of antibodies to calf thymus nuclei, nucleoprotein and native and heat-denatured DNA.

5. Abnormal elevation of the IgM activity without concurrent elevation of IgG or IgA activity was the most frequent pattern observed but was not predominant enough to be characteristic of systemic lupus.

Demography of serum immunoglobulins: differences in IgG and IgM levels in two normal Mexican adult populations

1. The three major immunoglobulins were quantitatively determined in the serum of 112 healthy Mexico City residents by the single radial immunodiffusion method.

2. Mean levels of IgG and particularly of IgM were unusually low as compared to (a) mean levels reported from elsewhere, (b) a group of thirty-eight healthy residents of Acapulco, Mexico of similar racial background and income and (c) a ‘British’ standard reference serum.

3. It is suggested that low IgG and IgM levels in Mexico City residents may be related to its high altitude.

4. Diminished immunoglobulin levels, particularly those of IgM, might be a compensatory mechanism for the increase in blood viscosity caused by high altitude erythrocytosis.