Dot enzyme-linked immunosorbent assay (Dot-ELISA): Comparison with standard ELISA and complement fixation assays for the diagnosis of human visceral leishmaniasis

Dot enzyme-linked immunosorbent assay (Dot-ELISA): Comparison with standard ELISA and complement fixation assays for the diagnosis of human visceral leishmaniasis

Veterinary Parasitology, 14 (1984) 239--249 Elsevier Science Publishers B.V., Amsterdam -- Printed in The Netherlands 239 DOT ENZYME-LINgED IMMUNOSO...

609KB Sizes 0 Downloads 47 Views

Veterinary Parasitology, 14 (1984) 239--249 Elsevier Science Publishers B.V., Amsterdam -- Printed in The Netherlands

239

DOT ENZYME-LINgED IMMUNOSORBENT A S S A Y (DOT-ELISA): COMPARISON WITH S T A N D A R D ELISA A N D COMPLEMENT FIXATION A S S A Y S FOR THE DIAGNOSIS OF HUMAN VISCERAL LEISHMANIASIS*

M I C H A E L G. PAPPAS**, R U T A H A J K O W S K I , LOUIS T. C A N N O N W A Y N E T. H O C K M E Y E R

SR. and

Department of Immunology, Walter Reed Army Institute of Research, Washington, DC 20307 (U.S.A.)

ABSTRACT Pappas, M.G., Hajkowski, R., Cannon, L.T., St. and Hockmeyer, W.T., 1984. Dot enzymelinked immunosorbent assay (Dot-ELISA): comparison with standard ELISA and complement fixation assays for the diagnosis of human visceral leishmaniasis. Vet. Parasitol., 14: 239--249. The dot enzyme-linked immunosorbent assay (Dot-ELISA), standard ELISA and the complement fixation (CF) tests were compared in the serodiagnosis of African visceral leishmaniasis (kala-azar). Assay sensitivity was determined using sera from 44 patients with parasitologically confirmed kala-azar. Using the Dot-ELISA, 42 of 44 patients (95%) were positive at a reciprocal titer of ;* 32 (titer range 512--524 288). In the standard ELISA technique, 43 of 44 patients (98%) were positive (titer range 32--32 768). At a reciprocal titer of ~ 8 in the CF test, 35 patients (80%) were positive, 1 (2%) was negative and 8 patients (18%) showed anticomplementary (AC) activity (titer range 8--2048). Specificity, determined using 33 sera from healthy individuals not living in endemic areas, was 97% in both the Dot-ELISA and the standard ELISA (32 of 33 sera); in the CF test, all sera were negative except 1 (3%) which showed AC activity. Sera from patients with Chagas' disease cross-reacted in the Dot-ELISA up to a titer of 512. In the standard ELISA, cross-reactions occurred mainly using sera from patients with Chagas' disease, malaria and syphilis, and to a lesser extent with sera from amebiasis, schistosomiasis and trichinosis patients. Overall titer agreement in replicate experiments was highest in the Dot-ELISA (89%), followed by the standard ELISA (80%) and the CF test (72%).

INTRODUCTION

Human visceral leishmaniasis (kala-azar) is a phlebotomine sandflytransmitted protozoan disease of great morbidity and mortality in East Africa. The disease is characterized by hepatosplenomegaly, emaciation *The views of the authors do not purport to reflect the position of the Department of the Army or the Department of Defense (Para. 3--4, AR 360--5). **Author to whom correspondence should be addressed.

0304-4017/84/$03.00

© 1984 Elsevier Science Publishers B.V.

240

and anemia, and is almost always fatal without adequate chemotherapy. Presently, kala-azar is positively diagnosed by finding parasites in spleen, liver or bone marrow biopsies (Chulay and Bryceson, 1983). However, this technique is tedious and insensitive, when low concentrations of parasites are present in infected tissue. The difficulties in diagnosing visceral leishmaniasis from histopathological materials emphasizes the need for more sensitive and specific diagnostic tests. The complement fixation (CF) test has been evaluated extensively for many parasitic diseases (Kagan, 1974), is a highly standardized technique (Kent and Fife, 1963; U.S.P.H. Monograph No. 74, 1968) and has been used for diagnosing kala-azar in East Africa (Hockmeyer et al., 1984; Smith et al., 1984). However, this method is tedious, sensitive to day-to-day variation and anticomplementary (AC) reactions occur in varying percentages of human sera tested, rendering serodiagnosis difficult (Davis et al., 1944). The enzyme-linked immunosorbent assay (ELISA), a relatively new procedure first developed for determining immunoglobulin G (IgG) concentrations (Engvall and Perlmann, 1972), has also been applied to a number of parasitic diseases including visceral leishmaniasis (Hommel et al., 1978; Edrisian and Darabian, 1979), but requires overnight adsorption of solubilized parasite antigen to microtiter plate wells, a procedure shown to give inconsistent results between different brands of microtiter plates, and even between wells in the same plate (Chessum and Denmark, 1978; Kricka et al., 1980). This procedure also normally requires a photometer to determine absorbances of test results. Recently, we described a rapid, visually read micro enzyme-linked immunosorbent assay, utilizing minute volumes of parasite antigen "dotted" on nitrocellulose filter discs (Dot-ELISA), for the diagnosis of human visceral leishmaniasis (Pappas et al., 1983). In this report, we compared results from the Dot-ELISA, the standard ELISA and the CF test using sera from patients with confirmed visceral leishmaniasis, normal healthy individuals and patients with other diseases. MATERIALS AND METHODS

Sera Serum samples from 44 patients hospitalized at Kenyatta National Hospital in Nairobi, Kenya, with visceral leishmaniasis were analyzed. Positive diagnosis was based upon identifying parasites in cultures from splenic aspirations or in stained, microscopically viewed smears of biopsied material. Control sera consisted of 33 samples from healthy Americans with no known exposure to leishmaniasis. Cross-reactivity studies were performed using sera from patients with a variety of bacterial, fungal and parasitic diseases. Sera were aliquoted in 50 gl volumes and stored at --20°C. All samples were coded before testing and assayed twice in each test procedure.

241

Parasite Leishmania donovani (WR 311) promastigotes, isolated from a naturally infected Kenyan in 1981, were used to prepare antigen for all procedures. Parasite speciation was based upon geographic location, clinical manifestations and restriction endonuclease analysis of kinetoplast DNA (Jackson et al., 1984). Promastigotes grown in RPMI 1640 medium supplemented with 20% heat-inactivated fetal bovine serum (FBS) and gentamicin sulfate (Sigma, St. Louis, MO) were harvested during log-phase growth and characteristically contained/> 95% motile organisms. Antigen preparation Parasites for use in the Dot-ELISA were washed 3 times with Hank's balanced salt solution, without calcium and magnesium (HBSS), pH 7.6 (GIBCO, Grand Island, NY) and centrifuged 1500 X g for 10 min at 4°C. An aliquot of parasites were fixed in 1.5% formalin-HBSS for 1 h at 22°C, washed 3 times in triethanolamine-buffered saline (TBS), pH 7.5 (Kent and Fife, 1963), then dotted on nitrocellulose filter discs in 1 pl volumes containing 2.5 X 104 promastigotes using a 10 pl Hamilton syringe. Antigen discs were then dried for 10 min at 56°C and stored at -20°C until used. Antigen for the CF test was prepared by first washing parasites in HBSS as above, freezing and thawing promastigotes 3 times and then sonicating the preparation at 20 kHz in three 5-rain bursts while on ice. The freeze-thaw-sonicate was then aliquoted in 5 ml volumes and stored at -70°C. ELISA antigen was prepared in the same way as the CF test, except that the freeze--thaw-sonicate suspension was centrifuged at 2500 X g for 30 min at 4°C. Using this procedure, the resulting solubilized antigen suspension contained 30.2% of the total protein associated with the parasite pellet. ELISA antigen was then aliquoted in 1 ml volumes and stored at -70°C. Block titrations, against positive and negative control sera, were performed on each antigen batch to determine optimal antigen dilutions in the standard ELISA and CF assays. Protein levels in each antigen group were determined using the Lowry method.

Immunodiagnostic procedures All test procedures have been previously described and will be briefly outlined. The Dot-ELISA procedure was performed at room temperature (RT) and all solutions were prepared in TBS at pH 7.5 (Pappas et al., 1983). Antigen discs placed in fiat bottom microtiter plate wells were blocked for 15 rain with 75 ~1 of 5% bovine serum albumin (BSA--TBS). After aspirating off the blocking solution, 50/~1 of 4-fold increasing dilutions of patient sera, diluted in 1% BSA--TBS, starting at a reciprocal titer of 32, were added to each well and incubated for 30 min, followed by 3 washes in 0.05%

242 NP°40-TBS, a non-ionic surfactant, and a 30-min incubation with 50 ~l of a 1:100 dilution of peroxidase-conjugated affinity-purified anti-human IgG (Heavy and Light chains, Kirkegaard and Perry, Gaithersburg, MD). After aspirating off the peroxidase-conjugated antibody and washing the discs, a precipitable substrate, 4-chloro-l-naphthol (Sigma, St. Louis, MO), activated with hydrogen peroxide was added to each test-well and incubated for 30 min. Finally, antigen discs were washed 3 times with TBS and read visually. Discs showing clearly-defined, blue dots were considered positive. In the standard ELISA (modified from Voller et al., 1976), flat-bottom microtiter plate wells (Immulon II, Dynatech Labs., Alexandria, VA) were filled with 50 pl of antigen, optimally diluted in TBS, and incubated overnight at 4°C. The plates were washed 3 times in TBS and blocked for 30 min with 300 t~l 5% BSA--TBS. Wells were then incubated with 100 tzl of serially diluted patient sera for 1 h, then washed 3 times in 0.05% NP-40-TBS. Peroxidase-conjugated anti-human IgG (1:800) was then added to each testwell in 100 t~l volumes and incubated for 1 h. Plates were then washed 3 more times, and 100 ttl of the substrate ABTS ® (2,2'-azino-di-(3-ethylbenzthiazoline sulfonic acid-6)), in buffer containing cacodylic acid (Kirkegaard and Perry, Gaithersburg, MD), was added and incubated for 30 min. Absorbances were determined using a p h o t o m e t e r (Flow Multiskan, Flow Labs., Springfield, VA) at 405 nm. The absorbance cut-off for a positive reaction (0.412) was determined by averaging the absorbance values, at a reciprocal dilution of 32, from all normal control sera plus 2 standard deviations of that mean. The last serum dilution having an absorbance of 0.412 was considered the titer of that sample. Serum antibodies to visceral leishmaniasis were also detected using a micromodification of the CF test developed by Kent and Fife (1963). Sera (50 #1) were complement-inactivated prior to testing b y heating at 56°C for 30 min. To wells of round-bottom microtiter plates (Cooke, Alexandria, VA), 25 pl of patient sera were added in serial 2-fold increasing dilutions of TBS (pH 7.4), starting at a reciprocal titer of 8. Lyophilized guinea pig complement (5 hemolytic units, M.A. Bioproducts, WalkersviUe, MD) and parasite antigen were each added in 25 tLl volumes to test-wells and the plates shaken on a plate rotator for 1 min. After incubation for 18 h at 4°C, 50 pl of sheep erythrocytes (5 X 10S/ml), optimally sensitized with rabbit anti-sheep hemolysin (Flow Labs., McLean, VA), were added to each well and incubated for 45 rain at 37°C. Microtiter plates were then centrifuged at 500 X g for 5 min at 4°C and titers were determined using a 50% hemolysis endpoint. RESULTS

Protein determinations of parasite antigens Parasite antigen concentrations used in each serodiagnostic assay are shown in Table I. The total protein derived from 108 L. donovani WR 311

243

promastigotes was 708+35 pg (mean +1 S.D.). Of this amount, only 213.62 pg of protein, or 30.2%, was recovered by ultracentrifugation for use in the standard ELISA. Antigen concentration on a per-well basis was lowest in the Dot-ELISA (177 ng/well) followed b y the standard ELISA, which required 4 times more antigen (781 ng/well), and the CF test which used 200 times more antigen (37 307 ng/weU). At these antigen concentrations, 10 s parasites would yield 4000 test-wells in the Dot-ELISA, 273 tests in the standard ELISA and 19 tests using the CF test. TABLE

I

Parasite antigen concentrations used in each diagnostic procedure Assay

Antigen concentration (ng p e r well)

N u m b e r o f tests per 10 8 L. donovani WR 311 p r o m a s t i g o t e s a

Dot-ELISA ELISA CF test

177 781 37 307

4000 273 b 19

a T o t a l p r o t e i n p e r 108 L. donovani WR 311 p r o m a s t i g o t e s was 708+ 35 pg. b p r o t e i n r e c o v e r e d a f t e r u l t r a c e n t r i f u g a t i o n was 2 1 3 . 6 2 pg o r 30.2%.

Assay sensitivity and specificity Forty-two of 44 patients (95%) were positive in the Dot-ELISA, at a diagnostic titer of i> 32, and their titers ranged from 512 to 524 288 (Fig. 1). In the standard ELISA procedure, 43 positive reactions {97.5%) occurred, and these titers ranged from 32 to 32 768. At a diagnostic titer of >t 8, 35 patients {80%) were positive, 1 was negative (2%) and 8 patients (18%) showed AC activity in the CF test. Of the 8 patients who had AC sera, in the CF test 3 had titers of 131 072, 4 had titers of 32 768, and 1 had a titer of 2048 in the Dot-ELISA. When these sera were tested in the standard ELISA procedure, 1 had a titer of 32 768, 1 had a titer of 8192, 3 had titers of 2048, 2 had titers of 512 and 1 had a titer of 128. The geometric mean titer (GMT) of leishmaniasis sera was significantly greater in the DotELISA (GMT = 37 122} than the standard ELISA (GMT = 1663, P < 0.001, paired t-test) and the CF test (GMT = 269, P ~ 0.001). The GMT of the standard ELISA was also significantly greater than that of the CF test (P 0.001). Sera from 33 healthy individuals with no previous exposure to the disease were tested to determine assay specificity (Fig. 1). Thirty-two of these sera (97%) were negative and 1 serum (3%) showed a titer of 32 in the Dot-ELISA and standard ELISA. When these samples were tested in the CF test, none were positive and only 1 serum (3%), which was negative in both enzyme immunoassays, showed AC activity.

244

524,288

eo

131,072

°o°.o~°°°o° oooe

32,768 8,192

ooeo~e°oOOeee

oooe

oeoeoooeo

,.=, ado oeeooeoooae oooo

2,048 512

ooooooooo

oooo

2,048

oololooi@o

1,024 512 256 128 84 32 16

Qooooao

,o..

o@ooo

128

oBo ol

32

0

oo oo

-=32

0ooooooo o O0OOOeOO~OO 00ooo0oooooo

o0oo000000

oo

O000eOO o00 oogooooo%o

oooooooooo

ooooooeoooooo ooooooooooo

ooooooooo

DOT-ELISA

ELISA

b--

r~

a.

8 >8 AC

CF TEST

SERODIAGNOSTIC PROCEDURE Fig. 1. Sensitivity and specificity of the Dot-ELISA, standard ELISA and CF tests in serodiagnosis o f human visceral leishmaniasis. Sensitivity was determined in each assay using 44 sera from parasitologically confirmed cases o f visceral leishmaniasis ( . ) and specificity using 33 sera from apparently healthy individuals (o). In the standard ELISA, the last serum dilution having an absorbance value above 0.412 was considered the titer o f that serum.

Assay cross-reactivity Cross-reactions, using sera from patients with other diseases, are presented in Table II. Of 40 sera from patients with a variety of other diseases, crossreactions with parasite antigen were observed in the Dot-ELISA only with Chagas' disease sera (5/5, titers of up to 512}. In the standard ELISA, cross-reactions were observed with sera from patients with Chagas' disease (3/3} and malaria (4/5, 1 Plasmodium falciparum, 2 P. malariae, 1 P. vivax), the highest cross-reactive titers being 128. Cross-reactions occurred to a lesser extent when sera from patients with amebiasis (1/5), schistosomiasis (1 Schistosoma haematobium), syphilis (3/5) and trichinosis (1/5) were tested (all positive titers of 32). Cross-reactions were not observed in the CF test. However, sera from patients with malaria (2/5, 1 P. falciparum, 1 P. malariae) and sera from schistosomiasis patients (3/5, 1 S. rnansoni, 2 S. japonicum) showed AC activity.

245 TABLE

II

Cross-reactivity using sera from patients with diseases other than visceral leishmaniasis Disease

Reciprocal titers of:

Amebiasis Chagas' disease Coccidioidomycosis Malaria Schistosomiasis Syphilis Toxoplasmosis Trichinosis

32 Dot-ELISA

~ 32 ELISA

~ 8 CF test

0/5 5/5 b 0/5 0/5 0/5 b 0/5 0/5 0/5

1/5 a 3/3 c 0/5 4/5 a 1/5 a 3/5 a 0/5 I/5 a

0/5 0/5 0/5 0/3 d 0/2 d 0/5 0/5 0/5

aAll cross-reactive titers were 32. bCross-reactive titers ranged up to 512. cOnly 3 sera were tested. Cross-reactive titers ranged up to 128. dAC reactions occurred with 2 malaria sera and 3 schistosomiasis sera. See Results section for serum categorization.

TABLE III Reproducibility of the Dot-ELISA, standard ELISA and CF procedures Titer change

Number of samples (%) Leishmaniasis sera

Control sera

Other diseases

Combined

Dot-ELISA 0 1 2

34(77) 10(23) 0(0)

32(97) 1(3) 0(0)

38(95) 2(5) 0(0)

104(89) 13(11) 0(0)

ELISA 0 1 2

34(77) 10(23) 0(0)

31(94) 2(6) 0(0)

27(71) 10(26) 1(3)

92(80) 22(19) 1(1)

CF test 0 1

17(39) 19(43)

32(97) 0(0)

35(88) 0(0)

84(72) 19(16)

2

0(0)

0(0)

0(0)

0(0)

AC a

8(18)

1(3)

5(12)

14(12)

aAntieomplementary reactions rendering titer determinations impossible.

246

Reproducibility When test data were analyzed for titer fluctuation between experiments, overall sample reproducibility varied the least in the Dot-ELISA (Table III). Of 117 sera tested, 104 (89%) did not differ in titer and 13 varied in 1 dilution (11%). None of the sera varied more than in 1 dilution. Of 115 sera assayed, using the standard ELISA, 92 sera (80%) did not vary in titer, 22 (19%) varied in 1 dilution and 1 serum (1%) varied in 2 titer dilutions. The CF test displayed the greatest titer variability, with 19 sera (16%) varying in 1 dilution and 14 sera (12%) showing AC activity. When samples were broken down into their 3 respective groups, percent variability was greatest in all assays when testing sera from leishmaniasis patients and the lowest in normal control sera. DISCUSSION In this study, the Dot-ELISA and standard ELISA were equivalent with respect to sensitivity and specificity in the diagnosis of kala-azar. The DotELISA and standard ELISA, however, were more sensitive than the CF procedure in diagnosing patients with parasitologically-confirmed visceral leishmaniasis, due to the large number of AC sera in the CF test. Serum AC activity is commonly observed in small percentages of human samples, and has been linked to diseases associated with hyperglobulinemia (Davis et al., 1944) and lipemia, and is seen in individuals with high levels of complement activity (Hook and Muschel, 1964). Sample manipulation also influences AC activity. Sera subjected to multiple freeze--thawings show increased AC activity over fresh samples, and hemolyzed sera compromise the accuracy of 50% hemolysis endpoints. In contrast, AC reactions cannot occur in the Dot-ELISA, and lipemic and hemolyzed sera, which may also affect absorbance readings i n the standard ELISA, do not affect Dot-ELISA test results (Pappas et al., 1983). Of the 3 procedures, the CF test is the most labor-intensive, tedious and sensitive to day-to-day variation. Endogenous complement must be heatinactivated before testing sera. Reagents such as antigen, guinea pig complement, hemolysin-sensitized red cells and buffer solutions must be carefully prepared and titrated before use (Fife and Muschel, 1961). Absolute sensitivity of the CF procedure is also affected by the technique 's inability to utilize weakly complement-fixing or noncomplement-fixing parasite-specific antibodies (i.e., IgG2, IgG4) in determining serum titers. The CF and standard ELISA require soluble antigen derived from repeated freeze--thaw-sonication of great numbers of parasites. This procedure is tedious and wasteful; at the protein concentrations used in this study, 108 L. donovani WR 311 promastigotes (~708 ~g protein) would yield antigen for 273 test-wells in the ELISA, and only 19 test-wells in the CF test. In contrast, the same number of parasites would yield approximately 4000

247 tests in the Dot-ELISA. Another drawback of freeze--thaw-sonication is that not all serodiagnostically important antigen species may solubilize, and heat-labile antigens may be destroyed during sonication. Variation in the ability of microtiter plates to physically adsorb soluble antigens or antibodies to well surfaces is another disadvantage of the standard ELISA. Different batches of plates, treated identically with ELISA reagents, display uneven antigen binding to plate wells along the outer edges (edge effect), possibly due to inherent thermal characteristics of the plate (Butt et al., 1979) and even between wells on the same plate (Chessum and Denmark, 1978; Kricka et al., 1980). Conversely, the Dot-ELISA technique is very antigen-conservative and rapid. As little as 2.5 × 104 parasites/test-disc give highly sensitive, specific and reproducible results (Pappas et al., 1983). The diagnostic procedure takes approximately 2 h to perform at RT, compared to the CF test which is optimally sensitive only when patient sera are incubated overnight (18 h) at 4°C, and the standard ELISA which, after overnight adsorption of parasite antigen, takes about twice as long to complete. The Dot-ELISA utilizes easily-prepared, whole, formalin-fixed parasites as antigen and also utilizes noncomplement-fixing as well as complement-fixing antibodies in determining titer endpoints against parasite antigen, features which enhance overall test versatility. The Dot-ELISA has also been utilized in rapidly determining Leishmania-specific monoclonal antibody titers (unpublished data). The phenomenon of variation in antigen-binding to plate wells is not encountered when antigen is "dotted" on nitrocellulose filter discs, a solid support which avidly binds antigen species. The enzyme immunoassay is more reagent-conservative and inexpensive (< 5 cents/test) and does not require expensive photometers; important considerations for use in developing countries. The Dot-ELISA also has great potential for field use, as this easily portable procedure is visually read, all steps are conducted at ambient temperature and the reagents are stable for long periods of time and easily disposed of. We believe the Dot-ELISA may prove very practical in laboratories where rapid and inexpensive serologies or monoclonal antibody screening to parasite antigens are desired, and for scientists conducting field surveys in remote areas. CONCLUSIONS

The Dot-ELISA, standard ELISA and CF tests were compared using sera from patients with confirmed visceral leishmaniasis, healthy individuals and patients with other diseases. Assay sensitivity and specificity were equivalent in the Dot-ELISA and standard ELISA, while sensitivity in the CF test was significantly compromised by AC reactions. In the Dot-ELISA, cross-reactions were observed only with sera from patients with Chagas' disease, whereas the standard ELISA cross-reacted with sera from a number of diseases. No sera cross-reacted in the CF test, but 5 sera showed AC activity. Overall

248 t i t e r a g r e e m e n t was highest in t h e D o t - E L I S A , f o l l o w e d b y t h e s t a n d a r d E L I S A a n d t h e CF test. O f the 3 p r o c e d u r e s , t h e D o t - E L I S A is t h e m o s t rapid and antigen-conservative, does not require expensive p h o t o m e t e r s and has g r e a t p o t e n t i a l f o r field use. ACKNOWLEDGMENTS T h e a u t h o r s are g r a t e f u l t o D.B. Tang, Ph.D., f o r assistance w i t h statistical analysis, a n d t o J.W. M c M a h o n f o r t e c h n i c a l assistance.

REFERENCES Butt, S.M., Carter, T.J.N. and Kricka, L.J., 1979. Thermal characteristics of microtiter plates used in immunological assays. J. Immunol. Methods, 31: 231--236. Chessum, B.S. and Denmark, J.R., 1978. Inconstant ELISA. Lancet, 1 : 161. Chulay, J.D. and Bryceson, A.D.M., 1983. Quantitation of amastigotes of Leishmania donovani in smears of splenic aspirates from patients with visceral leishmaniasis. Am. J. Trop. Med. Hyg., 32: 475--479. Davis, B.D., Kabat, E.A., Harris, A. and Moore, D.H., 1944. The anticomplementary activity of serum gamma globulin. J. Immunol., 49: 223--233. Edrisian, G.H. and Darabian, P., 1979. A comparison of enzyme-linked immunosorbent assay and indirect fluorescent antibody test in the sero-diagnosis of cutaneous and visceral leishmaniasis in Iran. Trans. R. Soc. Trop. Med. Hyg., 73: 289--293. Engvall, E. and Perlmann, P., 1972. Enzyme-linked immunosorbent assay (ELISA). III. Quantitation of specific antibodies by enzyme-labeled anti-immunoglobulin in antigen coated tubes. J. Immunol., 109: 129--135. Fife, E.H., Jr. and Muschel, L.H., 1961. Influence of magnesium and calcium ions on the hemolytic activity and stability of guinea pig complement. J. Immunol., 87: 688-695. Hockmeyer, W.T., Wellde, B.T., Sabwa, C.L., Smith, D.H., Rees, P.H. and Kager, P.A., 1984. A complement fixation test for visceral leishmaniasis using homologous antigen. Ann. Trop. Med. Parasitol., in press. Hommel, M., Peters, W., Ranque, J., Quilici, M. and Lanotte, G., 1978. The micro-ELISA technique in the serodiagnosis of visceral leishmaniasis. Ann. Trop. Med. Parasitol., 72: 213--218. Hook, W.A. and Muschel, L.H., 1964. Anticomplementary effects and complement activity of human sera. Proc. Soc. Exp. Biol. Med,, 117: 292--297. Jackson, P.R., Wolheiter, J.A., Jackson, J.E., Diggs, C.A. and Hockmeyer, W.T., 1983. Restriction endonuclease analysis of Leishmania kinetoplast DNA characterizes parasites responsible for visceral and cutaneous disease. Am. J. Trop. Med. Hyg., in press. Kagan, I.G., 1974. Advances in the immunodiagnosis of parasitic infections. Z. Parasitenk., 45: 163--195. Kent, J.F. and Fife, E.H., Jr., 1963. Precise standardization of reagents for complement fixation. Am. J. Trop. Med. Hyg., 12: 103--116. Kricka, L.J., Carter, T.J.N., Burr, S.M., Kennedy, J.H., Holder, R.L., Holliday, M.I., Telford, M.E. and Wisdom, G.B., 1980. Variability in the adsorption properties of microtiter plates used as solid supports in enzyme immunoassay. Clin. Chem., 26: 741--744.

249

Pappas, M.G., Hajkowski, R. and Hockmeyer, W.T., 1983. Dot-enzyme-linked immunosorbent assay (Dot-ELISA): A micro technique for the rapid diagnosis of visceral leishmaniasis. J. Immunol. Methods, 64: 205--214. Smith, D.H., Wellde, B.T., Sabwa, C.L., Reardon, M.J. and Hockmeyer, W.T., 1984. A complement fixation test for visceral leishmaniasis using homologous parasite antigen. If. Results from an endemic area. Ann. Trop. Med. Parasitol.,in press. U.S. Department of Public Health, Education and Welfare, Public Health Service, 1968. Standardized Diagnostic Complement Fixation Method and Adaptation to Micro-test. Public Health Monograph No. 74. Voller, A., Bartlett, A. and BidweU, D.E., 1976. Enzyme immunoassays for parasitic diseases. Trans. R. Soc. Trop. Med. Hyg., 70 : 98--106.