REPORT : DETECTION OF 72-75-kD and 123-kD FRACTIONS OF LEISHMANIA ANTIGEN IN URINE OF PATIENTS WITH VISCERAL LEISHMANIASIS

Two polypeptide fractions of 72-75 kD were detected in the urine of 14 of 15 patients with visceral leishmaniasis (VL) and another fraction of 123 kD was found in 10 of the 15 patients by using a Western blot technique. None of these fractions was detected in the urine of 20 controls. These results suggest that antigen detection in urine could be a powerful, noninvasive method for VL diagnosis. An increasing number of cases of leishmaniasis in patients infected with the human immunodeficiency virus (HIV) and the low humoral response found in these individuals urges the development of new diagnostic methods based on anti gen detection. Noninvasive techniques may be of great value in epidemiologie studies and antigen detection in urine is a good option if the technique is sufficiently sensitive and spe cific. Detection of Leishrnania antigen in the urine of patients with visceral leishmaniasis (VL) was performed by Kohanteb and others in 1987' using agar precipitation techniques. In 1992, Dourado and others (unpublished data) presented preliminary results of Leishmania antigen detection in urine using an immunocapture technique. In neither case were the detected antigens identified. The work reported here was undertaken to develop a method to detect Leishmania antigen in urine for use as a diagnostic tool. Preliminary unpublished experiments using a dot-enzyme-linked immunosorbent assay (ELISA) failed to give satisfactory results; a small proportion of healthy controls and some patients with acquired immunodeficiency syndrome (AIDS), but no history of leishmaniasis, gave pos itive results. For this reason, we assessed the utility of a Western blot technique in the identification of the specific polypeptide fractions in urine. Urine samples from 35 individuals were studied, including 15 from clinically proven cases of VL (nine HIV-positive and six HIV-negative) and 20 from negative control patients (eight HIV-positive and 12 HIV-negative) who had no his tory of leishmaniasis. Visceral leishmaniasis was diagnosed on the basis of clinical symptoms, including intermittent fe ver, hepatosplenomegaly, pancytopenia, and anemia, and on the detection of Leishmania amastigotes in bone marrow as pirates. Antibodies in serum were detected using a dot-ELISA and an ELISA in all HIV-negative patients and in seven of nine HIV-positive patients with VL. Urine samples were collected and stored in aliquots at -40Â°C until use; 15 ml of the urine sample was concentrated using Centripep-IO Concentrators (#4304; Amicon, Beverly, MA) through three consecutive centrifugations of 40, 10, and 5 min at 2,800 X g at room temperature. The final volume obtained was 0.61 ml, depending on the amount of sediment. Leishmania hyperimmune serum (LHS) was obtained by immunization of rabbits with two subcutaneous inoculations of 2.5 X 10" promastigotes (MHOM/FR/78/LEM 75) emul sified in Freund's incomplete adjuvant (Sigma, St. Louis, MO). These were given one month apart. The rabbits were given booster injections seven, 14, and 21 days after the final immunization with an intravenous inoculation of 1 X 10", 1.6 X IO8,and 2 X IO8promastigotes in phosphate-buffered saline. Immunization was monitored by dot-ELISA and rab bits were bled one week after the final inoculation. Serum was separated and the immunoglobulin fraction was obtained by ammonium sulfate precipitation. Rabbit IgG antisera (R-3382; Sigma) was biotinylated us ing N-hydroxysuccinimidobiotin (H-1759; Sigma). Proteins were separated on a 7.5% gel by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) under re ducing conditions. A Leishmania promastigote suspension (LEM 75) was used as a positive control. Molecular weights were estimated from the relative mobility of simultaneously run standard proteins of known molecular mass (SDS-PAGE Molecular Weight Standards, low range; Bio-Rad, Rich mond, CA). Polypeptides were transblotted onto nitrocellulose sheets (0.45-u.m pore size, HAWP 304 FO; Millipore, Bedford, MA) and the sheets were blocked with 20 mM Tris, 0.13 mM NaCl, pH 7.4, (TS)-5% skim milk for 30 min. All in cubations were carried out at room temperature (approxi mately 25Â°C). The nitrocellulose sheets were then incubated with a 1:500 dilution of LHS in TS, 0.05% Tween 20 (TST)1% skim milk (TST-M) for 3 hr. The sheets were then washed and incubated with antirabbit biotinylated serum di luted 1:200 in TST-M for 1 hr, then washed and incubated with 2.5 u,g/ml of avidin-peroxidase (A-3151; Sigma) in TST for 30 min. Color was developed with (3,3'-diaminobenzidine tetrahydrochloride, D-5637; Sigma)/H2O2 and re action was stopped with tap water after 30 min. Immunoblot analysis showed that LHS recognized two polypeptide fractions of 72-75 kD in the urine of 14 of 15 patients with VL and another component of approximately 123 kD in 10 of these cases; the fractions were not identified in the urine of either normal controls or VL-negative AIDS patients (Figure 1 and Table 1). Nevertheless, LHS recog-

Laboratori de ParasitologÃ-a,Departament de Microbiologia i ParasitologÃ-aSanitaries, Faculta! de Farmacia, UniversitÃ¤tde Barcelona, Barcelona, Spain; Sen'ei de Microbiologia, Hospital de la Santa Creu i Sani Pau, Barcelona, Spain Abstract.Two polypeptide fractions of 72-75 kD were detected in the urine of 14 of 15 patients with visceral leishmaniasis (VL) and another fraction of 123 kD was found in 10 of the 15 patients by using a Western blot technique.None of these fractions was detected in the urine of 20 controls.These results suggest that antigen detection in urine could be a powerful, noninvasive method for VL diagnosis.
An increasing number of cases of leishmaniasis in patients infected with the human immunodeficiency virus (HIV) and the low humoral response found in these individuals urges the development of new diagnostic methods based on anti gen detection.Noninvasive techniques may be of great value in epidemiologie studies and antigen detection in urine is a good option if the technique is sufficiently sensitive and spe cific.
Detection of Leishrnania antigen in the urine of patients with visceral leishmaniasis (VL) was performed by Kohanteb and others in 1987' using agar precipitation techniques.In 1992, Dourado and others (unpublished data) presented preliminary results of Leishmania antigen detection in urine using an immunocapture technique.In neither case were the detected antigens identified.
The work reported here was undertaken to develop a method to detect Leishmania antigen in urine for use as a diagnostic tool.Preliminary unpublished experiments using a dot-enzyme-linked immunosorbent assay (ELISA) failed to give satisfactory results; a small proportion of healthy controls and some patients with acquired immunodeficiency syndrome (AIDS), but no history of leishmaniasis, gave pos itive results.For this reason, we assessed the utility of a Western blot technique in the identification of the specific polypeptide fractions in urine.
Urine samples from 35 individuals were studied, including 15 from clinically proven cases of VL (nine HIV-positive and six HIV-negative) and 20 from negative control patients (eight HIV-positive and 12 HIV-negative) who had no his tory of leishmaniasis.Visceral leishmaniasis was diagnosed on the basis of clinical symptoms, including intermittent fe ver, hepatosplenomegaly, pancytopenia, and anemia, and on the detection of Leishmania amastigotes in bone marrow as pirates.Antibodies in serum were detected using a dot-ELI-SA and an ELISA in all HIV-negative patients and in seven of nine HIV-positive patients with VL.Urine samples were collected and stored in aliquots at -40Â°C until use; 15 ml of the urine sample was concentrated using Centripep-IO Concentrators (#4304; Amicon, Beverly, MA) through three consecutive centrifugations of 40, 10, and 5 min at 2,800 X g at room temperature.The final volume obtained was 0.6-1 ml, depending on the amount of sediment.
Leishmania hyperimmune serum (LHS) was obtained by immunization of rabbits with two subcutaneous inoculations of 2.5 X 10" promastigotes (MHOM/FR/78/LEM 75) emul sified in Freund's incomplete adjuvant (Sigma, St. Louis, MO).These were given one month apart.The rabbits were given booster injections seven, 14, and 21 days after the final immunization with an intravenous inoculation of 1 X 10", 1.6 X IO8, and 2 X IO8promastigotes in phosphate-buffered saline.Immunization was monitored by dot-ELISA and rab bits were bled one week after the final inoculation.Serum was separated and the immunoglobulin fraction was obtained by ammonium sulfate precipitation.
Rabbit IgG antisera (R-3382; Sigma) was biotinylated us ing N-hydroxysuccinimidobiotin (H-1759; Sigma).Proteins were separated on a 7.5% gel by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) under re ducing conditions.A Leishmania promastigote suspension (LEM 75) was used as a positive control.Molecular weights were estimated from the relative mobility of simultaneously run standard proteins of known molecular mass (SDS-PAGE Molecular Weight Standards, low range; Bio-Rad, Rich mond, CA).
Immunoblot analysis showed that LHS recognized two polypeptide fractions of 72-75 kD in the urine of 14 of 15 patients with VL and another component of approximately 123 kD in 10 of these cases; the fractions were not identified in the urine of either normal controls or VL-negative AIDS patients (Figure 1 and Table 1).Nevertheless, LHS recog- nized a 50-kD fraction and other fractions with high molec ular weights (> 150 kD) in almost all urine samples, which may responsible for the false-positive results detected in pre liminary studies with a dot-ELISA.
In four of the five patients in which the 123-kD compo nent was not detected (no. 6, 7, 13, and 14), the samples were collected during treatment with GlucantimeÂ®(N-metilglucamine antimoniale; RhÃ´ne-Poulenc Rorer S.A., Alcorcon, Madrid, Spain).Furthermore, the one individual (a child, no.15) in whom the double band of 72-75 kD was not detected was also currently undergoing treatment for leishmaniasis.Consequently, the lack of antigen may coin cide with observations of Kohanteb and others,1 who did not find antigen in the urine of two kala-azar patients after 10 and 13 days of treatment with GlucantimeÂ®.Jaffe and others2 described two monoclonal antibodies that recognize a double band of 69-73 kD in the membrane of the different species TABLE1

Polypeptide fractions in the urine of patients with visceral leishnuiniasis
Patient no.
Antigens detected in urine"'' of the L. donovani complex, but not in other species of the genus.3 The results of this preliminary study suggest that the de tection of polypeptide fractions of 72-75 kD and 123 kDa of the Leishmania antigen in urine may be of value in the development of a direct antigen diagnostic test.