Determination of anti-Cryptosporidium coproantibodies by time-resolved immunofluorometric assay

330 TRANSACTIONS OF THE ROYALSOCIETY OF TROPICAL MEDICINE AND HYGIENE (1993) 87, 33C-332

Determination of anti-Cryptosporidium immunofluorometric assay

coproantibodies

by time-resolved

Laboratoire de Biologie Animale et N. Kapel*, D. Meillet, M. Buraud, L. Favennec, D. Magne and J. G. Gobert Parasitaire, Faculte de Pharmacie, Universite Rend Descartes, 4 avenue de l’observatoire, 75006 Paris, France Abstract

The role of the mucosal immune responseagainst Ctyptosporidiumhas been suggestedby studies on the therapeutic effects of hyperimmune colostrum. In order to study the intestinal response to this infection, we have developed a sandwich-type time-resolved immunofluorometric assay for the determination of antiCyptosporidium coproantibodies. This assay has the inherent sensitivity of an immunoassay without the problems due to background responsesfrom other biological compounds, and is thus suitable for faecal samples. The intra-assay coefficients of variation (5.1%, 4.6%, and 5.8% for immunoglobulins (Ig) A, M and G respectively), inter-assay coefficients of variation (9.4%; 10.5% and 12.2% for IgA, IgM and IgG, respectively) and specificity (100% for all 3 isotypes) were all satisfactory. Using this assayto study 12 patients with the acquired immune deficiency syndrome (AIDS) who were infected with cryptosporidiosis, we found a marked elevation of anti-Cryptosporidium IgA and IgM coproantibody titres relative to 18 healthy control values, but no correlation with the gravity of the infection in terms of oocyst shedding. These results suggest that a non-protective mucosal immune responsedevelops to Cyptosporidium in AIDS patients. Introduction

Cryptosporidiosis, a potential cause of self-limiting diarrhoea in immunocompetent individuals, is a major opportunistic pathogen in those with the acquired immune deficiency syndrome (AIDS), causing unremitting and frequently life-threatening diarrhoea (CURRENT et al., 1983; TZIPORI, 1988; CURRENT& GARCIA,1991). A serum antibody response has been reported but its protective role is questionable (UNGARet (II., 1986; KASSAet al., 1991). The role of the mucosal immune response against Cyptosporidium has been suggestedby studies on the therapeutic effects of hyperimmune colostrum (FAYERet al., 1990), but very little information is available on intestinal production of immunoglobulins during cryptosporidiosis. We have previously found a large increase in faecal polymeric immunoglobulin (Ig) A and IeM outuut in natients with AIDS and with intestinal ODpirtunisiic infections (MEILLET et al., 1992), and Cypfosporidium-specific secretory IgA has been detected in faecesfrom exnerimentallv infected lambs (HILL et al.. 1990). We theiefore examined the specificity of the im: munoglobulins secretedin the gut lumen of AIDS patients with cryptosporidiosis to evaluate the mucosal response to this infection. Faecal specimens were used, as collection is non-invasive, easily controlled and well accepted by the patient. Faecal IgA shows a high degree of resistance to cleavageby proteolytic agents, due to the unusual structure of the hinge region and to the presence of the extraenithelial moiety of the nolv-Ig recentor, i.e. the secretory component (MESTE& & M&HE& 1987; KAPEL et al.. 1990). We have develoned a sensitive assav for faecal an&Cr+&osporidium antibddies, basedon timeresolved fluorescence using a non-isotopic europium label. This time-resolved immunofluorometric assavJTRIFMA) has the sensitivity of a fluoroimmunoassay,.without the limitations resulting from background fluorescence of biological compounds (SOINI & HEMMILA, 1979; HEMMIL~~et al., 1984). We report here on the performance of the assay and the results obtained with a group of AIDS patients with cryptosporidiosis relative to those obtained with persons seronegative for human immunodeficiency virus (HIV). Concentrations of antiCyptosporidium immunoglobulins were compared with oocyst shedding, an index of diseaseseverity. Materials and Methods Faeces collection

Faeceswere collected from 18 HIV-seronegative controls (6 women and 12 men; mean age47+ 13 years) with no gastrointestinal disturbance and parasite-free faecal specimens, and from 12 AIDS patients with cryptospori*Author for correspondence.

diosis documented by examination of faecesor histological sections (1 woman and 11 men; mean age 39f 10 years). After addition of sodium azide (0.1% w/v, final concentration) and diisopropylfluorophosphate (final concentration 1 mM), 5 g of homogenized material was vigorously stirred for 1 h with 10 ml of 0.15 M NaCl at 4°C. After centrifugation (10 000 g, 10 min), the supernatant was immediatelv stored at -20°C. All assavswere performed using these 3-fold diluted stool samples. Cyptosporidium oocysts were sought in faecal specimens by staining with the Ziehl-Nielsen reagent according to Henriksen’s method (HENRIKSEN et al., 1981). Two slides were prepared for each sample and a minimum of 40 microscope fields (400x magnification) were examined. The mean number of oocysts per field was noted. Cryptosporidium antigens Cyptosporidium oocysts were separatedfrom the faeces of infected calves and kept at 4°C in 2.5% K2Cr207. To eliminate debris, the faeces were sieved sequentially through stainless-steel screens, with a final mesh size of 32 pm. The specimens were then concentrated by centrifugation at 150 g for 30 min over a sucrose gradient (densities of 1.04 and 1.10). The layer between the 2 densities was removed and washed ,with 0.15 M NaCl. Microbial contaminants were eliminated by exposure to 2% sodium hypochlorite solution in an ice-water bath for 10 min (BURAUDet al., 1991). The oocysts were then disrupted by 3 cycles of freezing and thawing. Non-disrupted oocystswere removed by centrifugation at 1000 g for 10 min, and the protein concentration of the supernatant was determined by means of the Coomassieblue assay. The supernatant, used as the source of soluble antigens, was divided into aliquots and stored in liquid nitrogen. Time-resolved immunojluorometric assay

Nun@ plates were coated overnight at 37°C with 200 ~1 per well of 5 ug/ml Cyptosporidium antigen solu-

tion in 0.05 M K2HP04 buffer, pH 8.5. They were then washed 3 times with 0.05 M Tris-HCI, pH 7.75, containing 0.025% Tween [email protected] binding sites were blocked by post-coating the wells for 1 h at 37°C with 250 yl of 0.05 M NaH2HP04 containing 4% bovine serum albumin (BSA) and sorbitol at 60 g/l. After washing, 200 yl of pooled reference anti-Cyptosporidium serum antibodies (diluted l/10 for IgA and IgM and 1140 for IgG) were added, with 200 1.11 of the test faecal extract (116final dilution), to triplicate reaction wells. After incubation for 2 h at room temperature with shaking, the wells were washed as described above. The mates were then incubated for 1 h at room temperature with continuous shaking with 200 ul of europium-labelled

331 antibodies (111000rabbit anti-human LXchain, 11600rabbit anti-human u chain or l/l000 rabbit anti-human y chain; Dako, Glostrup, Denmark). Immunoreactions were carried out in 0.1 M Tris-HCl buffer, nH 7.75, containing 8.85 g/l NaCl, 0.05% NaNa and O:S%BSA td reduce non-specific binding, and 12 PM diethylenetriaminepentaacetic acid to bind free europium. After incubation, the wells were washed as described above and europium fluorescence ‘of the bound antibody was enhanced by adding 200 ul of 0.1 M acetate-phthalatebuffer, pH 3.2, containing Triton x100@ (O.lO%), 2-naphtholyltrifluoroacetone (15 PM), and trioctylphosphineoxide (50 uM) (enhancement solution; Pharmacia, Uppsala, Sweden) to each well, followed by shaking for 10 min. Fluorescence was measured after a ‘rest’ period of 10 min in a fluorometer with a xenon flash lamp (1230 Arc@ Fluorometer? WALLACPharmacia). During measurement, a 1 us excitation pulse was emitted at 340 nm and single-photon emission was counted for 0.40 ms at 613 nm. The results are given as counts per second (cps). The background fluorescence of the counting solution was measured in 4 wells, and the mean value was subtracted from the values for each sample well. The results were then calculated in arbitrary units as the ratio of cps between the sample and the pooled anti-Cyptosporidium positive reference serum. Statistical analysis

As the data did not show a Gaussian distribution, the results are given as medians and were compared using the non-parametric Mann-Whitney U test. Correlations were sought by meansof Spearman’scorrelation test. Results

Curves obtained by assaying serial dilutions of faecal extract showed no interference with the signal by other faecal components at dilutions of 116or more. The intraassay variation of the assaywas estimated in 20 parallel determinations and the inter-assay variation was determined in 10 repeated tests. Repeatability was good, with intra-assay coefficients of variation of 5-l%, 46% and 5.8% for IgA, IgM and IgG antibodies, respectively, expressed in arbitrary units (AU). With the same faecal samples, the inter-assay coefficients of variation were 9.4%, 10.5% and 12.2% for IgA, IgM and IgG antibodies, respectively. Positivity thresholds of 16 AU, 9AU and 6AU for anti-Cyptosporidium IgA, IgM and IgG antibodies, resnectivelv, were determined as the mean values obtained with -the control group plus 3 standard deviations. The snecificitv of the TR-IFMAwas lOO%,and its sensitivity was 83%, 67% and 33% for IgA, IgM and IgG, respectively. The results for the 3 isotyies of anti-Crvatosaoridium antibodies in the AIDS oatients with cryptospoiidiosis and the HIV-seronegative kontrols are shown in the Figure. The median values in the pa-

Figure. Concentration (in arbitrary units) of faecal antibodies to Cryptosporidiumin individual AIDSpatients infected with Cyptosporidium (0) and in control individuals seronegative for human immunodeficiency virus (0).

tient group were 46 AU, 11 AU and 4 AU for IgA, IgM and IgG antibodies, respectively. The anti-Cyptosporidium IgA and IgM antibody levels were significantly higher (P
TR-IFMA is a novel immunological method which has found numerous applications, especially in parasitology (ACETI et al., 1987, 1988). The high sensitivity of the time-resolved fluorescence technique with europium labelling, together with its ability to minimize the non-specific background signal, allowed us to analyse the intestinal hum&al ir&mne response to dyptosporidium infection. The aunlication of the nrocedure to faecal snecimens appearedto be satisfactory, allowing the secretion of anti-Ctyptosporidium antibodies by the whole digestive tract to be assessed.Using this method we found elevated titres of anti-Cyptosporidium IgA and IgM in AIDS patients with crvntosooridiosis (10112and 7112natients. respectively). The difference from the results o^fHILL ei al. (1990), who found elevated levels of IgA antibodies, only, m experimentally infected lambs, is probably related to the sensitivity of the TR-IFMA. In contrast, no elevation of anti-Cyptosporidium IgG levels was detected in the faecal extracts. This may be related to proteolytic degradation of this isotype or to the absenceof a selective transport systemto the gut lumen. These findings confirm our previous report of increased immunoglobulin production in AIDS patients with opportunistic intestinal infections compared with HIV-negative controls (MEILLET et aE., 1992) and indicate that AIDS patients develop mucosal antibody responsesagainst the parasite. This response is pathogenspecific and not due only to the inflammatory response and the increased production of interleukin-6. In contrast to findings in serum (KASSA et al., 1991), faecal anti-CryptosporidiumIgA secretion appearsto accompany continuing cryptosporidiosis, confirming the markedly different regulation of the serum and mucosal immune system in AIDS. The degree of the intestinal immune response appeared to be independent of the gravity of the infection (evaluated in terms of oocyst shedding) and did not seem to be protective, as the infection did not decreasein severity in AIDS patients with high anti-cyptosporidium IgA and IgM levels. Determination of anti-Cyptosporidium antibodies, especially IgA, should help in the diagnosis of cryptosporidiosis, since oocyst shedding can be intermittent or undetectable by routine laboratory tests. Indeed,, one patient in our studv had biopsy-proven crvptosporidiosis without oocyst shedding but had a particularly-high concentration of anti-CrvatosaoridiumIgA and IaM. In conclusion, the’TR-iFMA techmque offers clear advantages for the detection of anti-Cyptosporidium antibodies, especially in faecal samples. It is sensitive, is not limited by background fluorescence in complex biological media, and is easier to standardize than enzymelinked immunofluorescent assayand indirect immunofluorescent assay, in which temperature variations can modify the enzyme reaction catalysing substrate turnover. Using such stable non-isotopic immunoreagents, we demonstrated the development of non-protective mucosal antibody responsesagainst Cyptosporidium in AIDS patients. However, this responsemight be of fmportance in the development of cryptosporidiosis as the infection might progress faster in the absenceof antibody. I.

332

Hill, B. D., Blewett, D. A,., Dawson, A. M. & Wright, S. (1990). Analysis of the kinetics, isotype and specificity of serum and coproantibodies in lambs infected with Cryptosporidium parvum. Research in Veterinav Science, 48,76-g 1. Kapel, N., Meillet, D., Iscaki, S., Bouvet, J. I’., Gobert, J. G. & Raichvarg, D. (1990). Characterisation of the main molecular forms of human faecal IgA. Clinica Chimica Acta, 195,6776. Kassa, M., Comby, E., Lemeteil, D., Brasseur, P. & Ballet, J. I. f1991‘1.Characterization of anti-CrvotosaoridiumIpA antiLo&es i; sera from immunocompete~~individuals and HIVinfected patients.gournal of Protozoology, 38, 179S-180s. Meillet, D., Kapel, N., Favennec, L. & Gobert, J. G. (1992). Protein exudation and intestinal synthesis in opportunistic digestive tract infections associatedwith the acquired immunodeficiency syndrome.Journal of Infection, in press. Mestecky, J. & McGhee, J. R. (1987). Immunoglobulin A (IgA): molecular and cellular interactions involved in IgA biosynthesis and immune response. Advancesin Immunology, 40, 153-245. Soini, E. & HemmilB, I. (1979). Fluoroimmunoassay: present status and key problems. Clinical Chemistry, 2!, 353-361. Tzipori, S. (1988). Cryptosporidiosis in perspective. Advances in

References

Aceti, A., Pennica, A., Caferro, M., Celestino, D., Paparo, B. S. & Sebastiani, A. (1987). Application of time-resolved fluoroimmunoassay (TR-FIA) for the diagnosis of invasive amoebiasis. Transactions of the Royal Society of Tropical Medicine and Hygiene, 81, 764-767.

Aceti, A., Pennica, A., Celestino, D., Paparo, B. S., Caferro, M., Sanguigni, S., Marangi? M. & Sebastiani, A. (1988). A new serological technique, time-resolved fluoroimmunoassay (TRFIA), for the immunological diagnosis of urinary schistosomiasis. Transactions of the Rqal Society of Tropical Medicine and Hygiene, 82,445-447.

Buraud, M., Forget, E., Favennec, L., Bizet, J., Gobert, J. G. & Deluo!, A. M. (1992). Sexual stage development of cryptosporidia m Caco-2 cell line. Infection and Immunzty, 59,46104613. Current? W. L. & Garcia, L. S. (1991). Cryptosporidiosis. Clinical Mzcrobiology Reviews, 4, 325-358.

Current, W. L., Reese,N. C., Ernst, J. V., Bailey, W. J., Heyman, M. B. & Weinstein, W. M. (1983). Human cryptospondiosls in immunocompetent and immunodeficient persons: studies of an outbreak and experimental transmission. New EnglandJournal

ofMedicine, 308,1252-1257.

Fayer, R., Guidry, A. & Blagburn, B. L. (1990). Immunotherapeutic efficacy of bovine colostral immunoglobulins from hyperimmunized cow against cryptosporidiosis in neonatal mice. Infection and Immunity, 58,2962-2965. Hemmill, I., Dakubu, S., Mukkula, V. M., Siitari, H. & Lavgen, T. (1984). Europium as a label in time-resolved immunofluorometric assay.AnalyticalBiochemistry, 137,335-343. Henriksen, S. A. & Pohlenz, J. F. L. (1981). Staining of cryptosporidia by a modified Zlehl-Neelsen technique. Acta Veteri-

Parasitology, 27,63-129.

Ungar, B. L. P., Soave, R., Fayer, R. & Nash, T. E. (1986). Enzyme immunoassay detection of immunoglobulin M and G antibodies to Cryptosporidium in immunocompetent and immunocompromised persons. Journal of Infectious Diseases, 153,57C-578. Received 2 July 1992; revised 10 August 1992; accepted for publication 12 August 1992

naria Scandinavica, 22, 594-596.

IAnnouncements

1 16th International Beijing,

China;

Congress of Hydatology 12-16

October

1993

Last date for registration: 31 March 1993. Further information can be obtained from Mr Ma Shi-liang, Chinese Academy of Preventive Medicine, 10 Tiantan Xiii, Beijing 100050, China; telephone +86 21 4376308; fax +86 21 4332670.

14th International 29 August-4 Walt

Leprosy

Congress

September 1993 Orlando, Florida,

Disney World Village,

USA

Further information can be obtained from: ILA Congress, c/o ALM International, ALM Way, Greenville, SC 29601, USA. Telephone: (803) 271-7040. Fax: (803) 271-7062

Seventh International

Training

Course on Identification

International

of Helminth Parasites of Economic Importance

4-12Ju2y 1994 Institute of Parasitology,

St Albans,

UK

Further information can be obtained from Dr L. M. Gibbons, International 395A Hatfield Road, St Albans, Hertfordshire, AL4 OXU, UK.

Institute of Parasitology,