TRANSACTIONS OF THE ROYAL SOCIETY OF TROPICAL MEDICINE AND HYGIENE (1993) 87, 173-176 173 lmmunodiagnosis of Strongyloides stercorak infection: a method for increasing the specificity of the indirect ELISA D. J. Conway’, N. S. Atkins’, J. E. Lillywhitel, J. W. Bailey2, R. D. Robinson3, J. F. Lindo3, D. A. P. Bundy’ and A. E. Bianco’ ‘Wellcome Trust Research Centre for Parasitic Infections, Department of Biology, Imperial College, Prince Consort Road, London, S W7 2BB, UK; ‘Liverpool School of Tropical Medicine, Pembroke Place, Liverpool, L3 SQA, UK; 3Department of Zoology, University of the West Indies, Mona, Kingston 7, Jamaica Abstract Indirect enzyme-linked immunosorbent assay(ELISA)allows sensitive detection of serum immunoglobulin (Ig) G against a soluble extract of Strongyloides stercoralis infective larvae. In this study, 40140 (100%) human strongyloidiasis serahad high levels of anti-S. stercoralis IgG, but 30140 (75%) filariasis sera, and 12140 (30%) necatoriasis seraalso had higher levels than control serafrom UK residents. In attempts to increase the assay specificity by absorption of cross-reactive IgG, the effectiveness of pre-incubation of sera with extracts of different parasitic nematodes was investigated. One hour of incubation with 20 pg/ml aqueous extract of Onchocerca gutturosa absorbed cross-reactive IgG in most filariasis and necatoriasis sera, reducing the pro- portion with IgG levels above the positivity threshold by more than one-half. Preliminary results suggest that absorption with extracts of other filarial nematodes is equally effective, and that some of the cross-reac- tive IgG is directed against phosphorylcholine. Cross-reactive IgG in most necatoriasis sera was effectively absorbed with 20 pgiml extract of Necator americanus. Cross-reactiveIgG was not effectively absorbed with an extract of Ascaris lumbricoides. Absorption of cross-reactive IgG is an effective means of increasing the specificity of the indirect ELISA,for use in the immunodiagnosis and immuno-epidemiology of S. stercoralis infection. Introduction The intestinal nematode Strongyloides stercoralis is fre- quently difficult to detect in individuals with a low in- tensity of infection (GROVE, 1989). This causes wide- spread underestimation of infection prevalences in endemic communities (PAWLOWSKI, 1989; GENTA, 1989), and increases the risk of unexpected induction of severe hyperinfection in carriers receiving immunosup- pressive therapy for other conditions (IGRA-SIEGMAN et al., 1981; DEVAULT~~ al., 1990). In attempts to increase the sensitivity of, and decrease the examination time required for, diagnosis, severalim- munodiagnostic methods have been developed (GENTA, 1986). The most widely tested of these is the indirect enzyme-linked immunosorbent assay (ELISA) for detec- tion of serum immunoglobulin (Ig) G against total aqueousantigenic extracts of infective filariform (L3) lar- vae of S. stercoralis (NEVA et al., 1981; SATO et al., 1985; GAM et al., 19871, S. ratti (TRIBOULEY-DURET et al.. 1978; CARROLL ei’al., 1981; NEVA et al., 1981), or S: cebus (BAILEY, 1989). The specificity of the indirect ELISA is difficult to assess accurately, however! due to the limited sensitivity of any confirmatory parasitological test. Several investi- gators have referred to the possibility of false positivity in the ELISA, due to cross-reactive Igc responsesinduced bv filarial or hookworm infections. although oublished A information is limited. Among ind&iduals iith filariasis, for example, GAM et al. (1987) noted that 5 of 12, and BAILEY (1989) noted that one of 3, were seropositive in the Strongyloides ELBA. Although the ELBA has been used as a sensitive meansof primary screening of individ- uals (SATO et al., 1990), seropositivity in the absenceof parasitological or clinical indications is considered to have insufficient positive predictive value for diagnosis (GENTA, 1986). In this report, an assessment is made of the frequency of seropositivity in the indirect ELISA for Strongyloides among individuals with lymphatic filariasis and hook- worm infections. A single step method of serum absorp- tion is described which increasesthe specificity of detec- tion of anti-S. stercoralis IgG. Materials and Methods Human sera Sera were derived from 40 individuals with S. sterco- ralis infections, 20 of whom were British ex-Far East prisoners of war and had acquired their infections more than 30 years before serum sampling; the remaining 20 had presented to clinics in Britain or Dubai after an un- determined period of infection. Sera were also obtained from 40 individuals with Wuchereria bancrofti infections, most of whom lived in Sri Lanka and had clinical lym- phatic filariasis, for whom information on intestinal hel- minth infections was unavailable. Sera were also pro- vided from 40 individuals with Necator americanus infections from a study village in Zimbabwe where N. americanus was the only human helminthiasis detected (BRADLEY et al., 1992). Control sera were from 20 resi- dents of the UK. Indirect ELISA The indirect ELISA for detection of serum IgG against S. stercoralis was basedon those described by NEVA et al. (1981), SATO et al. (1985), and GAM et al. (1987). For antigen preparation, S. stercoralis filariform larvae were obtained in Jamaicaby faecal culture from a patient with strongyloidiasis. Larvae were washed 5 times in phos- phate-buffered saline (PBS), suspendedin 0.25% sodium hypochlorite for 5 min, washed 5 times again, and sus- pended in PBS with protease inhibitors (5 mM EDTA, 5 mM EGTA, 5 mM NEM, 5 nM pepstatin, 1.7 mu PMSF, 0.5 mM TPCK)*. Larvae were disrupted by sonication (6x20 set pulses at 8 pm, 4”C), centrifuged at 13 000 g for 30 min at PC, and the supernatant was used as the crude soluble antigen preparation after stor- age at -70°C. This antigen preparation was used at 100 yliwell and 2 pgiml protein concentration to sen- sitize 96-well Linbro@ plates (Flow Laboratories, UK). After removal of unbound antigen by 3x3 min washes, 150 1.11 of 2% bovine serum albumin in PBS were added to each well for 1 h. Wells were washed 3 times, and serum samples were added (100 ~1 at 11200 dilution, in triplicate) and incubated at room temperature for 2 h. After washing wells 3 times, a 112000 dilution of horser- adish peroxidase-conjugated polyvalent rabbit and anti- human IgG antibody (DakopattP, Dako Ltd.) was incu- bated (100 d/well) for 2 h. Wells were washed 3 times, and 100 yl.of 0.01% o-phenylenediamine with O+O03°k Hz02 were added to each well. After 30 min. 25 ul of 5 N H2SO4were added to each well to stop thk &&me- substratereaction. Indirect ELISAS were also performed identically using plates coated with 2 kg/ml of PBS-soluble extract of *Respectively, ethylenediaminetetraacetic acid,, ethylene-bis (fi- aminoethyl ether) N,N,N’,N’-tetraacetic acid, N-ethylmalei- mide? isovaleryl-Val-Val-Sta-Ala-Sta, phenylmethylsulphonyl fluoride, and N-tosyl-L-phenylalanine chloromethyl ketone.