TRANSACTIONS OF THE ROYAL SOCIETY OF TROPICAL MEDICINE AND HYGIENE (1991) 85, 645-647 645

Detection of serum antibodies in human Hymenolepis infection by enzyme immunoassay

A. G6mez-Priego, A. L. Godinez-Hana and M. Gutibrez-Quiroz Immunoparasitology Laboratory, Departamento de Ecologia Humana, Facultad de Medicina, Universidad National Autbnoma de M&co, CP 04510, Mdxico

Abstract Although hymenolepiasis is the commonest cestode

infection of man, there are no data available on the human immune response to this parasite. Thus, in order to determine if infection induces antibodies against Hymenolepis nana antigens, sera from 52 infected children were initially studied on Ouchter- lony plates and then by enzyme-linked immunosor- bent assay (ELISA), using a crude antigenic extract prepared from scolex and neck regions of adult worms. In addition, sera from persons with cysticer- cosis, taeniasis and other parasitoses, and normal human serf, were studied. Only one serum from the Hymenolepzs group showed precipitin antibodies against H. nana antigen, while several were positive by ELBA. The sensitivity of the ELBA was 84.62% and its specificity was 100%. Very high cross-reactivity rates were obtained with taeniasis (70.6%) and cysticercosis (75%) sera. These results show that Hymenolepis infection in man induces a low but detectable humoral immune response. Although not useful for diagnostic purposes, this may be relevant to the serodiagnosis of other tissue cestode infections of man, since antibodies detected in serological tests used for the diagnosis of cysticercosis, and probably hydatidosis, could be induced by H. nana instead of Taenia solium or Echinococcus larvae.

Introduction Hymenolepis nana commonly parasitizes children,

but whether serum antibodies against H. nana are produced and, if so, whether they have any protective function, is unknown. Since the parasite is attached to the host intestinal wall by its scolex, antigens (surface, somatic, or metabolic) derived from this region could possibly be used for antibody detection. Any humoral immune response against this parasite might also be relevant to the reliability of serodiagnosis of other tissue cestodes infecting man (G~MEz-PRIEGO & GUTI~RREZ-QUIROZ, 1991).

We report the use and evaluation of the enzyme- linked immunosorbent assay (ELBA) (VOLLER et al., 1979) for detecting antibodies against a crude anti- genie extract prepared from the neck and scolex regions of adult H. nana. We also investigated possible cross-reactivity with sera from patients with taeniasis and cysticercosis.

Materials and Methods Antigen

Complete adult H. nana worms, obtained from the

Correspondence to: Albert0 G6mez-Priego, Department of Human Ecoloev. School of Medicine. Universidad National -* I Aut6noma de Mexico, Ciudad Ukversitaria, Coyoacdn 04510, Mexico, D. F., Mexico.

gut of naturally infected mice, were washed 3 times with 150 mM saline solution (SS), incubated with SS plus antibiotics (Sodic penicillin and streptomycin), and washed afterwards thrice with sterile SS. Neck and scolex regions of the parasites were dissected under sterile conditions, and an antigenic extract was prepared according to the methods of TATO et al. (1979) and MOLINARI et al. (1983).

Protein determination Total protein content of the antigen was deter-

mined according to LOWRY et al. (195 l), using bovine serum albumin as standard.

Sera Serum samples were obtained by venous puncture

from 52 children, male and female, all infected with H. nana (diagnosed by faecal examination). Their ages ranged from 2 to 14 years and all were out-patients at the Hospital Infantil section of the Hospital General de la Ciudad de Mexico. Fifteen serum samples from children of similar age but with no H. nana eggs detected in their faeces were used as normal (control) sera. Serum samples from 12 indi- viduals serologically positive (indirect haemagglutina- tion titre 3 1: 32) for the metacestode of Tuenia solium (cysticercosis) (LARRALDE et al., 1986), 17 samples from patients infected with T. solium or T. saginata (taeniasis). and 24 samDles from individuals infected with 0th;; intestinal or-tissue parasites (10 ascariasis, 3 malaria, 8 amoebiasis, and 3 trichinellosis) were used for cross-reactivity studies. Sodium azide (10 mM final concentration) was added and samples were stored frozen at -20°C until used.

Double immunodiffusion test This test was performed by the microtechnique

described by OUCHTERLONY & NILSSON (1979). The protein concentration of the antigen was 1 mgiml.

Enzyme-linked immunosorbent assay (ELBA) The ELISA was performed according to the general

procedures described by VOLLER et al. (1979), with slight modifications. Flat-bottomed micro-ELISA plates (Immulon@) were absorbed with 5 ygiml of the antigen. Serum samples and goat anti-human im- munoglobulin G conjugated with horseradish perox- idase (see below) were diluted 1:200 and 1:300 respectively with phosphate-buffered saline- TweenB20. o-Phenylenediamine was used as subs- trate. Serum samples from children infected with H. nana and healthy children were used as positive and negative control samples in all plates. The reaction was stopped with 2.5 M sulphuric acid and read at 492 nrn. Results were expressed as optical density

646

quotient (ODQ) calculated from the formula ODO=ODWODN. where ODP was the outical den&y of the test serum and ODN was the mean plus 2 standard deviations of the optical density obtained with the negative control serum. For plotting pur- poses, ODQs were grouped in 0.2 unit ranges.

The mean value plus twice the standard deviation of the ODQ obtained with the negative control sera was selected as the threshold value for positivity in the ELISA.

The ELBA was evaluated as described by MBNDEZ et al. (1984) and MORROW (1981), in terms of its sensitivity, specificity, predictive value, efficiency and cross-reactivity.

Goat anti-human immunoglobulin (Ig) G conjugate Human IgG was obtained by 50% ammonium

sulphate precipitation of 25 pooled normal human sera. The protein content was adjusted to 10 mgiml, and 2 ml were emulsified in an equal volume of complete Freund’s adjuvant. The emulsion was in- jected intramuscularly to a goat on days 0, 7, and 42. Seven days after the last injection, and then every 2 weeks, the goat was bled and the serum tested by immunoelectrophoresis (OUCHTERLONY & NILSSON, 1979) against normal human serum. The IgG fraction of this antiserum was then separated by precipitation with 33% ammonium sulphate solution and conju- gated with horseradish peroxidase type VI (Sigma). The specific activity of the conjugate was determined as described by CATTY er al. (1983).

Faecal examination Three consequtive faecal samples from all children

included in this study were examined by the Faust method. The Kato technique was used to assess the intensity of infection (SALAZAR-SCHETTINO & DE HARO, 1980).

Statistical analysis General statistical data (mean, standard deviation

and standard error) of the ODQs were calculated by conventional procedures. Apparent differences be- tween control (negative) and positive (infected) ODQs were examined by the method of DICE & LERAAS, (1936). The significance of differences was deter- mined by means of Student’s t test.

Results Initial experiments to detect serum antibodies in

samples from children infected with M. nana by using the double immunodiffusion test showed that this assay was not useful for this purpose. Only one sample was positive in the test, but the precipitation band was faint and very difficult to observe, even after staining with amido black. In addition, the result was not correlated with the intensity of infection, since the patient had <lOOO eggs per gram of faeces (data not shown). Therefore, this test was discarded and sera were subsequently examined by ELBA.

Values obtained in the ELISA (ODQs) are presented in the Figure. A wide range of ODQ can be seen in both the infected (1.4-4.0) and normal (06-1.8) groups. The mean ODQ value obtained with control sera was 1.25kO.38 (standard deviation), while that of the sera from infected children was 2.81 f0.73, which

1 1 1 1

INFECTED Figure. Optical density (OD) quotients (OD of test serum divided by mean OD plus two standard deviations of negative control sera) obtained with 52 sera from patients infected with Hymenolepis and 15 normal (control) sera in an enzyme-linked immunosorbent assay using 5 pgitnl of an antigenic extract prepared from neck and scolex of adult H. nana. Individual values (0) and Dice-Leraas diagrams (DICE & LERAAS, 1936) are shown, with (inset) a key to the latter; as the areas representing twice the standard error (2 x SE) of the mean (ri) do not overlap, the differences are considered to be significant (SD=standard deviation).

Table 1. Evaluation of enzyme-linked immunosor- bent assay for detecting antibodies against somatic antigens (neck and scolex) of Hymenolepis nana

Sensitivity (n=52) Specificity (n= 15) Predictive value

Positive Negative

Efficiency

84.6% 100%

100% 65.2% 88.1%

Table 2. Cross-reactivity of enzyme-linked immunosor- bent assay using Hymenolepis nana neck and scolex antigens

Serum donors Infection No. tested No. positive

Taeniasis Cysticercosis f Z

12 (70.6%)

Both cestodes 9 (75.0%)

21 (72.4%) $F;f parasites :: 11 (45.8%)

53 32 (60.4%)

is more than twice the value observed in the negative control group.

The area representing the mean plus twice the standard error of the ODOs for the infected aroun does not overlao the cor;esnondina area foi the control group in {he Dice-Leraas diagrams, indicating that the differences between them were significant (Figure). This was confirmed by Student’s t test (P<O.OOl at 95% confidence level). Therefore, ODQ values >2*0 were considered to be positive. With this criterion, 44152 children with H. nana were seroloai- tally positive, while all sera from healthy child& were negative.

Serological evaluation of the test gave acceptable values, especially for specificity and the predictive value for a positive result (both 100%) (Table 1). The general cross-reactivity rate was high (60*4%).Both the cysticercosis and taeniasis groups showed high individual cross-reactivity rates (75% and 70.59% respectively) (Table 2). Considering both groups of cestode sera together, the seropositivity rate was 72.41%. The cross-reactivity rate obtained with sera from people infected with other intestinal or tissue parasites was 45.8%. Discussion

From the diagnostic point of view, considering the efficiency and simplicity of faecal examination (SALA- ZAR-SCHETTINO & DE HARO, 1980), detection of serum antibodies for serodiagnosis of human Hymen- olepis infection is not a practiTa1 procedure. However, from the immunoloaical ooint of view. the finding: of antibody in infected patiknts may be relevant to”the host-parasite relationship and to the serology of other cestode infections of man.

Previous experimental studies in mice have shown that serum antibodies against H. nana can be passive- ly transferred and also agglutinate oncospheres (DI CONZA, 1969; ITO, 1975). Antibodies against eggs and adult H. nana antigens have also been demonstrated (IT0 et al., 1986). Our results showed that children infected with Hymenolepis produce serum antibodies against antigens derived from the neck and scolex region of H. nana. Considering that 78% of the 52 children studied had very mild infections, with fewer than 1000 eggs/g of faeces (data not shown), the degree of systemic antigenic stimulation could be very slight. However, the possibility of local antibody synthesis cannot be excluded.

The most significant aspect of the detection of a humoral immune response against H. nana may be the high cross-reactivity obtained with sera from patients with taeniasis and cysticercosis, suggesting the presence of shared antigens between T. solium cysticerci and adult worms and H. namz. It is also possible, however, that the antibodies detected could have been the result of an undiagnosed Hymenolepis infection, past or present (G~MEz-PRIEGO & GUTIRR- REZ-QUIROZ, 1991).

In summary, Hymenolepis infection in man induces a humoral, serologically detectable immune response against H. nana somatic antigens, which is of low magnitude and, since reinfection is frequent, such antibodies apparently are non-protective. It is likely that these antibodies could have relevance to the specificity of serodiagnosis of cysticercosis or hydatid disease in man. Acknowledgements

We thank Dr R. Romero Cabello and Dr J. T. Sanchez

647

Vega for their valuable help in clinical studies and serum collection, and Dr R. Bernal for providing some serum samoles. The technical assistance of P. Tatn and F. I. Mendez is also grately appreciated. A. L. G.-H. received”a fellowship from CONACYT, Mexico.

References Cattv. D.. Ravkundalia. C. & Houba. V. li983). Bench

manual of techniques for the preparation of ‘immu’nological and immunodiagnostic reagents. Part II. Preparation of enzyme conjugates and antibodies for human immunoglobu~ lins IgG, IgA and IgM and their quality control. Perform- ance of indirect ELISA tests. General considerations for a standard protocol. Geneva: World Health Organization, mimeographed document IMMPIRI83.1.

Dice, L. & Leraas, H. (1936). A graphic method for comparing several sets of measurements. Contributions to Laboratory Vertebrate Genetics, 3, 1-3.

Di Conza, J. (1969). Protective action of passively transfer- red immune serum and immunoglobulin fractions against tissue invasive stages of the dwarf tapeworm y;9ylepis nana. Experimental Parasitology, 43, 169-

Gomez-Priego, A. & Gutierrez-Quiroz, M. (1991). DIG-ELBA en la cisticercosis humana: detection de anticueroos con el fluid0 vesicular de1 metacestodo de Taenia solium y evaluacidn de la prueba. Revista Mexicana de Parasitolo- gia, 3, in press.

Ito, A. (1975). In vitro oncospheral agglutination given by immune sera from mice infected and rabbits iniected with eggs of Hymenolepis nana. Parasitology, 71,‘ 465 473.

Ito, A., Kano, S., Hioki, A., Kasuya, S. & Ohtomo, H. (1986). Correlation between antibody response and patent infection with Hymenolepis nana-in mice. Interna- tional Journal of Parasitology, 16, 197-203.

Larralde, C., Laclette, J. I’., Owen, C. S., Madrazo, I., Sandoval, M., Bolalil, R., Sciutto, E., Contreras, L., Arzate, J., Dtaz, M. L., Govezensky, T., Montoya, R. M. & Goodsaid, F. (1986). Reliable serology of Taenia solium cysticercosis with antigens from cyst vesicular fluid: ELBA and hemagglutination tests. American Jour- nal of Tropical Medicine and Hygiene, 35, 965-973.

Lowry, 0. H., Rosebrough, J. J., Farr, A. L. & Randall, R. J. (1951). Protein measurement with the folin phenol reagent. Journal of Biological Chemistry, 193, 265-275.

Mendez, R. I., Namihira, G. D., Moreno, A. L. & de Sosa, M. C. (1984). El Protocolo de Investigacibn. Lineamientos para su Elaboraci6n y Analisis. Mexico: Editorial Trillas.

Molinari, J. L., Meza, R., Suarez, B., Palacios, S., Tato, I’. & Retana, A. (1983). Taenia solium: immunity in hogs to the cysticercus. Experimental Parasitology, 56, 327-338.

Morrow, R. H. (1981). Diagnostic measures in tropical diseases. Epidemiological perspectives. In: Report of an informal consultation on further development of diagnosis methods for tropical diseases, Morrow, R. H. (editor). Geneva: UNDPWorld Bank/WHO Special Programme for Research and Training in Tropical Diseases.

Ouchterlony, 0. & Nilsson, L. A. (1979). Immunodiffusion and immunoelectrophoresis. In: Handbook of Ex- perimental Immunology, 3rd edition, vol 1, Weir, D. M. (editor). London: Blackwell Scientific Publications, pp. 19.1-19.44.

Salazar-Schettino, P. M. & De Haro, I. (1980). Manual de Tecnicas para el Diagnostico morfolc5gico de Eas Parasitosis. Mexico: Francisco Mendez Cervantes, pp. 93-96.

Tato, I’., Flisser, A., Gavilanes, M. & Molinari, J. L. (1979). Immunogenic complexes obtained from Sal- monella typhimurium and S. typhi Ty2 by the bacterial acetone powder method. Annales de Microbiologic (Institut Pasteur), 130, 47-60.

Voller, A., Bidwell, D. E. & Bartlett, A. (1979). The enzyme-linked immunosorbent assay (ELISA). A guide with abstracts of microplate application. Alexandria, Virginia, USA: Dynatech Laboratories.

Received 30 July 1990; revised 21 February 1991; accepted for publication I3 March 1991