IgM anti-phenolic glycolipid-1 antibody measurements from skin-smear sites: Correlation with venous antibody levels and the bacterial index

International Journal of Leprosy and Other Mycobacterial Diseases, Dec 1997 by Butlin, C Ruth, Soares, Des, Neupane, Kapil Dev, Failbus, Sarah S, Roche, Paul W

IgM Anti-Phenolic Glycolipid-I Antibody Measurements from Skin-Smear Sites: Correlation with Venous Antibody Levels and the Bacterial Index1

C. Ruth Butlin, Des Soares, Kapil Dev Neupane, Sarah S. Failbus, and Paul W. Roche2

Measurements of IgM anti-phenolic glycolipid-I (PGL-I) antibodies in leprosy are of value for assessing the extent of the disease (4,5) and the response to therapy (6). To assess the levels of antibodies in lesions and to minimize venipunctures we have assessed the accuracy of antibody measurements made from capillary blood collected on filter paper from the skin-smear site. We have assessed whether such measurements correlate with serum antibody levels and with venous blood collected onto filter paper and investigated whether local antibody levels also reflect the bacterial load at the skin-smear site.

MATERIALS AND METHODS

Subjects. After giving informed consent, 192 consecutive patients attending Anandaban Leprosy Hospital for routine skin smears were included in the study. These were 129 males and 63 females aged between 12 and 81 years. Patients were classified according to the Ridley-Jopling classification on clinical and bacteriological criteria (3). The patient groups consisted of 7 suspect cases, 1 indeterminate case, 19 tuberculoid, 47 borderline tuberculoid, 14 borderline, 56 borderline lepromatous, 40 lepromatous and 8 primary neuritic cases. Eight patients were tested on two separate occasions; 106 patients were under treatment at the time of testing. There were four patients on paucibacillary multidrug therapy (PB-MDT) and 102 on multibacillary multidrug therapy (MB-MDT). Patients had been treated for 0 to 79 months. The remaining 79 patients had completed treatment 1 to 10 years previously.

Samples. Venous blood (2 ml) was drawn from an arm vein and a single drop placed on filter paper (Whatman Scientific, Ltd., Kent, U.K.). The remaining blood was placed in a plain sterile tube and allowed to clot. Skin smears were collected from four standard sites from each patient: one each from both earlobes, the right arm and the right thigh. After collection of serous fluid for smears, a drop of capillary blood from the right earlobe was collected onto filter paper.

Assays. The slit-skin smears were stained by the Ziehl-Neelsen technique and the number of acid-fast bacilli (AFB) at each site was recorded on a logarithmic scale. An average bacterial index (BI) was calculated as the arithmetic mean of the indices at the four sites.

Filter-paper samples were allowed to dry, transported to the laboratory, and were stored for up to a month without significant loss of antibody. A 6-mm circle of bloodsoaked filter paper was punched out and soaked overnight at room temperature in 0.5 ml phosphate buffered saline containing 0.05% Tween 20 (PBST), pH 7.4. This was calculated to be equivalent to a 1 in 50 dilution of blood and used directly in the antiPGL-I enzyme-linked immunosorbant assay (ELISA).

IgM anti-PGL-I antibodies were measured as described previously (4). Wells of a flat-bottom, 96-well ELISA tray (Dynatech, Alexandria, Virginia, U.S.A.) were coated overnight with 250 ng/ml of the glycoconjugate disaccharide-bovine serum albumin (dBSA) in 0.01 M carbonate buffer, pH 9.6. Wells were washed in PBST and blocked with 200 pl of 1% (w/v) bovine serum albumin (BSA) in PBS for 1 hr at 37C; 100 MI of sera diluted 1:300 in 10% normal goat serum (NGS) or undiluted eluate from filter-paper specimens was added to wells for 1 hr. Wells were washed four times with PBS, and 100 mu l of goat anti-human IgMhorseradish peroxidase conjugate diluted 1: 4000 in 10% NGS was added for an additional 1 hr. Plates were again washed four times before the addition of 100 mu l 0.4 g/L o-phenylenediamine (OPD; Sigma Chemical Co., St. Louis, Missouri, U.S.A.) in 0.05 M citrate-phosphate buffer, pH 5.0, containing 0.006% hydrogen peroxide for 10 min at room temperature. The color reaction was stopped with the addition of 100 mu l 2.5 N sulfuric acid, and the plates were read at 492 nm in a Multiskan ELISA plate reader (Flow, Irvine, Scotland). Samples with an absorbance greater than 0.199 (being the mean plus three standard deviations of samples from 91 healthy Nepali subjects) were considered positive.

Statistics. The significance of differences in the proportions seropositive in the different groups was tested with the chi-squared statistic using the Yates' correction. Differences in antibody levels detected in different sites was tested by the Wilcoxon matched pairs test. Correlations between antibody levels from different samples and antibody levels and the BI were tested by the Spearman rank correlation test.

RESULTS

Of the 200 samples tested in this study, positive antibody levels to PGL-I were found in the capillary blood from the earlobe skin-smear site in 67 samples compared with 62 sera samples and 60 samples of venous blood collected onto filter paper (Table 1); 52 patients had positive antibody levels in all three samples, 8 in two of three samples and 16 in one sample only. In total, 76 patients were positive in at least one of the three samples collected. The proportion of seropositives increased from tuberculoid to lepromatous across the leprosy spectrum. There was no significant difference in the proportion of seropositives between different samples in any leprosy class. In paucibacillary (TT/BT) patients, 10/65 (15%) were positive in sera and capillary blood and 6/65 (9%) were positive in venous blood on filter paper. In lepromatous (BL/LL), 45/96 (47%) were positive in sera, 50/96 (52%) in capillary blood samples, and 47/96 (49%) in venous blood collected on filter paper.