L3 antigen-specific antibody isotype responses ... - Wiley Online Library

Parasite Immunology, 1999: 21: 517–526

L3 antigen-specific antibody isotype responses in human strongyloidiasis: correlations with larval output N.S.ATKINS 1 , 2 , D.J.CONWAY 2 , 3 , J.F.LINDO 4 , J.W.BAILEY 5 & D.A.P.BUNDY 1 , 2 1

Centre for the Epidemiology of Infectious Disease, Department of Zoology, University of Oxford, South Parks Road, Oxford, OX1 3PS, UK, Wellcome Research Centre for Parasitic Infections, Department of Biology, Imperial College, Prince Consort Road, London, SW7 2BB, UK, 3 Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, Keppel Street London WC1E 7HT, UK, 4 Department of Microbiology, The University of the West Indies, Mona, Kingston 7, Jamaica, West Indies and 5 Department of Tropical Medicine, Liverpool School of Tropical Medicine, Pembroke Place, Liverpool, L3 5QA, UK 2

SUMMARY

INTRODUCTION

Autoinfective strongyloidiasis is potentially fatal, yet the majority of infected individuals harbour asymptomatic and chronic infections. The role of humoral responses in modulating autoinfection was assessed by examining antibody isotype responses to filariform larval antigens amongst chronically infected ex-Far East Prisoners of War (exFEPOWs) with longstanding (> 30 years) infection. Serum immunoglobulin (Ig)G1, IgG4, IgE and IgA responses to whole Strongyloides stercoralis L3 extracts and their constituent antigenic components were characterized by ELISA and quantitative immunoblotting. Comparison of two groups of S. stercoralis infected exFEPOWs with and without detectable larvae in stool demonstrated novel trends. Significantly enhanced recognition of six immunodominant antigenic components by IgA was associated with undetectable larval output, as was enhanced IgE recognition of several components. Additionally, IgE and IgG4 exhibited parallel antigen recognition patterns. These findings are consistent with roles for IgA in modulating larval output, for IgE in regulating autoinfection, and for IgG4 in blocking IgE-mediated responses in human strongyloidiasis.

Autoinfection in human strongyloidiasis is a major determinant in the progression to clinical disease. In circumstances which have yet to be defined, larvae may develop to the tissue invasive filariform third larval stage (L3) before leaving the gut, so allowing internal autoinfection to take place (Grove 1986). Typically, this presents little problem to the immunocompetent host, and in the majority of cases there is a balanced, asymptomatic coexistence between host and parasite. However, if this equilibrium is upset, for example as a result of corticosteroid treatment or concurrent immunosuppressive disease, then hyperinfection and dissemination may occur, often with fatal consequences (Purtilo et al. 1974, Grove 1989). Immunoregulation of autoinfection is likely to involve responses that modulate worm fecundity, the rate of larval development and larval migration. Defining these host regulatory responses is a priority. Whilst host immunocompetence is likely to be the primary regulator of infection levels in autoinfective strongyloidiasis, there is little experimental evidence to support this hypothesis (Genta 1992). Several studies, which would offer further insight into the regulation of infection, have failed to demonstrate a link between host immune status and parasitological or clinical parameters (Genta et al. 1984, 1986, Sato et al. 1985, Badaro´ et al. 1987). An alternative hypothesis is that regulation of the rate of autoinfection may primarily be affected by endocrine control of the rate of larval moulting (Genta 1992). An additional consequence of the autoinfective cycle is that viable worm populations can persist indefinitely in the host despite an absence of environmental exposure to infectious parasite stages. Since dissemination does not occur in the majority of individuals, regulation of parasite numbers must be occurring, but at a level which is insufficient to clear infection. The phenomenon of chronic

Keywords Strongyloides stercoralis, ex-Far East Prisoners of War, immunoblotting, chronicity, antibody isotypes

Correspondence: J.F.Lindo Received: 18 November 1998 Accepted for publication: 19 April 1999 q 1999 Blackwell Science Ltd

517

J.F.Lindo et al.

infection in strongyloidiasis is best illustrated by the longstanding infections observed in ex-Prisoners of War held captive in endemic areas (Gill et al. 1977). Many of these individuals retained their infection upon repatriation and remain infected decades later (Grove 1980, Pelletier 1984, Goulston et al. 1985, Leighton & MacSween 1990). Such individuals provide an unique opportunity to study the interaction between parasite populations and host immune responses. These studies are important in the elucidation of the mechanisms involved in regulation of autoinfection and the maintenance of chronic infections. In the studies described here, we examined the levels of Strongyloides stercoralis L3 antigen-specific antibody isotype responses amongst individuals with contrasting levels of larval output in stool, and sought associations between parasitological and immunological parameters. The study subjects comprised two groups of chronically infected ex-Far East Prisoners of War (ex-FEPOWs) with (copropositive) and without (copronegative) detectable larvae in stool. If it is assumed that individuals with undetectable larvae are able to exert greater immunological control over their infections, then this differential provides an opportunity to compare the immune responses of individuals with a similar history and duration of infection, but with contrasting abilities to modulate larval output. MATERIALS AND METHODS Patients and serum samples The exFEPOWs studied here had been subjects of previous investigations (Gill & Bell 1979, Gill & Bailey 1989). Initial studies of this group showed that 15% of individuals retained Strongyloides stercoralis infections 30 years after repatriation from endemic areas (Gill & Bell 1979). Whilst larval shedding could be demonstrated in the majority of these exFEPOWs by charcoal culture (copropositive), a subgroup of ‘presumed positive’ individuals were repeatedly stool negative (copronegative), despite a history of unexplained eosinophilia and larva currens (Bailey 1989). These copronegative individuals had demonstrable serum immunoglobulin (Ig)G responses to filariform larval extracts of Strongyloides cebus (Bailey 1989), and to a S. stercoralis-specific IgG diagnostic enzyme-linked immunosorbent assay (ELISA) and confirmatory immunoblot (Conway et al. 1993a,b). The majority of these individuals seroreverted upon chemotherapy and experienced resolution of their symptoms and signs of infection (Archibald et al. 1993). Taken together these findings are strongly suggestive of the presence of occult strongyloidiasis in the copronegative individuals. Gill & Bell (1979) reported a significantly (P < 0·001) higher rate of eosinophilia amongst copronegatives (88%) 518

Parasite Immunology

compared to copropositives (53%). Therefore, to remove eosinophilia as a possible confounding factor in the current analyses, copronegative exFEPOWs were pair matched to within 2% eosinophilia with copropositives. Serum samples were collected during previous studies, and aliquots from two groups of 18 individuals were assessed here by quantitative ELISA. Assessment of antibody responses by immunoblotting was carried out on samples from two groups of 17 individuals because of depleted serum stocks. Filariform antigenic extracts A sequential antigenic extraction of filariform larvae in PBS followed by solubilization in sodium deoxycholate (DOC) was carried out as previously described (Conway et al. 1993b). Briefly, larvae were obtained from chronically infected patients by faecal culture and, after washing and sterilization, were resuspended in PBS with the following protease inhibitors: 5 mM ethylenediaminetetraacetic acid (EDTA), 5 mM EGTA [ethylene-bis (b-aminoethyl ether) N,N,N0 ,N0 -tetraacetic acid], 5 mM NEM (N-ethylmaleimide), 5 nM pepstatin, 1·7 mM PMSF (isovaleryl-Val-Val-Sta-AlaSta-phenylmethylsulphonylfluoride) and 0·5 mM TPCK (N-tosyl-L-phenylalanine chloromethyl ketone). Subsequent to disruption by repeated freeze-thawing in liquid nitrogen, the larval suspension was sonicated and left overnight at 48C. The larval suspension was then subjected to ultracentrifugation, and the protein concentration of the supernatant determined prior to storage at ¹ 708C for future use as PBSsoluble filariform extract (PFE). Following PBS extraction, detergent solubilization of the residual larval pellet was carried out. The sonicated pellet was resuspended in PBS with protease inhibitors as before, and 1% w/v DOC added. The mixture was agitated, and then left overnight at 48C. Following ultracentrifugation, the supernatant was decanted and dialysed against PBS for 24 h at 48C to remove the DOC (Furth 1980). The protein concentration of the dialysed extract was determined as before, aliquoted and stored at ¹ 708C for use as DOC-soluble filariform extract (DFE). Quantitative ELISA Serum samples were assayed for the levels of IgG1, IgG4, IgE and IgA to PFE by quantitative ELISA as described elsewhere (Atkins et al. 1997). Previous studies have demonstrated that parasite-specific IgG2 and IgG3 levels are low in human strongyloidiasis (Conway et al. 1994). A brief summary of the methods used is given here: IgG1 ELISA 96-well microtitre plates (Nunc Polysorp, Gibco BRL Ltd, UK) were coated with 2 mgml¹1 PFE overnight at 208C. q 1999 Blackwell Science Ltd, Parasite Immunology, 21, 517–526

Volume 21, Number 10, October 1999

After blocking with 2% bovine serum albumin (BSA), triplicate aliquots of sera diluted 1 : 200 in 2% BSA were incubated on separate plates. Bound IgG1 was detected by an antihuman IgG1 monoclonal antibody (MAb clone NL-16, Oxoid Ltd, UK) diluted 1 : 2000 in 2% BSA followed by a rabbit antimouse Ig horseradish peroxidase (HRP)conjugate (P-260, Dako Diagnostics Ltd, UK) diluted 1 : 2000 in 2% BSA. This catalysed 0·1 mgml¹1 tetramethylbenzidine (TMB), and optical densities (OD) were measured at 450 nm. To control for interassay variability positive control sera or pools were tested concurrently in triplicate on each plate. Measurement of other isotypes utilized this procedure, with the modifications detailed below. IgG4 ELISA Serum test concentrations were increased to 1 : 100 in 2% BSA, and a MAb specific for human IgG4 (MAb clone RJ-4, Oxoid) was substituted at a dilution of 1 : 4000 in 2% BSA. IgE ELISA High binding capacity plates (Nunc Maxisorp, Gibco) were coated with 8 mgml¹1 PFE. After blocking with 10% soya milk (SM), serum samples diluted 1 : 50 in 10% SM were added. Detection of bound IgE was by a rabbit antihuman IgE polyclonal (A-094, Dako) diluted 1 : 1000 in 10% SM, followed by a swine antirabbit Ig-HRP conjugate (P-217, Dako) at a dilution of 1 : 1000 in 10% SM. IgA ELISA 96-well plates (Linbro, ICN Flow Laboratories, UK) were sensitized with 2 mgml¹1 PFE and blocked with 2% BSA. Serum samples were tested at a dilution of 1 : 800 in 5% BSA, and bound antibody detected by a rabbit antihuman IgA-HRP conjugate (Jacksons ImmunoResearch Laboratories Inc., USA) diluted 1 : 5000 in 5% BSA. Western blots for quantitative antibody detection Proteins in pooled PBS-and DOC-soluble filariform extracts were separated using SDS-PAGE (Laemmli 1970). Polyacrylamide gels (1 mm × 13 cm × 16 cm) were poured with a 8–20% acrylamide gradient (pH 8·8), and stacking gels (3% acrylamide, pH 6·8) added with a 12·5-cm solid comb for protein loading. Equal protein concentrations of PFE and DFE were incubated for 5 min at 1008C in sample buffer (10% glycerol/2% SDS/62·5 mM Tris, pH 6·8) with 2·5% b-mercaptoethanol and 0·02% bromophenol blue. Antigen was loaded at a total rate of 20 mgcm¹1 of trough, with high and low range molecular weight (MW) standards (14·4–200 kDa, Bio-Rad Labs Ltd, UK) run in an adjoining lane. Gels were electrophoresed at 25 mA for 3–4 h in a q 1999 Blackwell Science Ltd, Parasite Immunology, 21, 517–526

Antibody isotype responses in human strongyloidiasis

vertical electrophoresis system (BRL Life Technologies Inc., USA). Separated polypeptides were electrophoretically transferred to nitrocellulose according to the method of Towbin et al. (1979) utilising a semidry blotting unit (LKB Multiphor Novablot II, Pharmacia UK Ltd) run for 2 h at 194 mA. After blotting, nitrocellulose was briefly soaked in Ponceau S solution (P-7170, Sigma Chemical Co. Ltd, UK) to confirm that successful separation and transfer had taken place. A strip containing the MW standards was removed, and stored for future reference. After destaining in distilled water, the blot was processed as described below. IgG1 quantitative immunoblot Nitrocellulose was blocked overnight at 48C in Tris buffered saline (10 mM Tris, 150 mM NaCl, pH 8·3) with 0·1% Tween 20 and 2% BSA (TBS-T/2% BSA). After blocking and 3 × 10 min washes in TBS-T, the nitrocellulose was cut into 5-mm wide numbered strips, and each then incubated for 2 h with 2 ml of the specified test or control serum diluted 1 : 200 in TBS-T/2% BSA. Positive and negative control sera or pools were tested concurrently in triplicate on each blot to control for interblot variability. Strips were washed, and incubated with an antihuman IgG1 MAb (MAb clone NL-16, Oxoid) diluted 1 : 2000 in TBS-T/2% BSA overnight at 48C. After a further wash, the strips were incubated for 2 h with a rabbit antimouse Ig-HRP conjugate (P-260, Dako) diluted 1 : 1000 in TBS-T/2% BSA. Blots were visualized with 0·5 mgml¹1 3,30 -diaminobenzidine tetrahydrochloride (DAB) in 50 mM Tris (pH 7·6) with 2 mM H2O2. After washing in distilled water, wet strips were realigned in numerical order on a sheet of plastic, together with the MW standards previously excised, and allowed to dry for densitometry. All incubations were at room temperature except where indicated. A similar procedure was used for the other isotypes, with the modifications detailed below. IgG4 quantitative immunoblot Sera were diluted 1 : 150 in TBS-T/2% BSA, and bound IgG4 was detected with a mouse antihuman IgG4 subclass MAb (MAb clone RJ-4, Oxoid) at a dilution of 1 : 2000 in TBS-T/2% BSA. IgE quantitative immunoblot To maximize sensitivity serum concentrations were increased to 1 : 50 in TBS-T/2% BSA and incubations carried out for 24 h at 48C. The detection antibody, a rabbit antihuman IgE polyclonal (A-094, Dako) diluted 1 : 1000 in TBS-T/2% BSA, was also incubated with the strips for 24 h at 48C. Finally, a swine antirabbit Ig HRP-conjugate (P-217, Dako) diluted 1 : 1000 in TBS-T/2% BSA was utilized. 519

J.F.Lindo et al.

Parasite Immunology

Table 1 Summary of Mann–Whitney U comparisons of antibody isotype-specific recognition of separated Strongyloides stercoralis L3 antigenic components by copropositive and copronegative ex-Far East Prisoners of War (exFEPOWs) as measured by immunoblotting.

Isotype

19 kDa

21 kDa

26 kDa

29 kDa

30 kDa

31 kDa

32 kDa

33 kDa

40 kDa

48 kDa

IgG1 IgG4 IgE IgA

NS NS – < 0·05

NS NS – < 0·05

– – – < 0·05

NS NS < 0·05* < 0·05

< 0·01 < 0·05 < 0·01 –

NS NS < 0·05 –

– – < 0·05 –

NS < 0·05 NS NS

NS NS – < 0·05

– – – < 0·05

In all but one comparison (*), greater recognition is amongst copronegative individuals. (NS, P > 0·05; –, not detectable by densitometry).

IgA quantitative immunoblot To reduce background, 5% SM (TBS-T/5% SM) was utilized in all dilutions. Sera were tested at a dilution of 1 : 50, and bound antibody was detected with a polyclonal rabbit antihuman IgA HRP-conjugate (Jackson ImmunoResearch Laboratories Inc) diluted 1 : 2000.

positive controls to each antigenic component was calculated for the first blot, and compared to the second to give a correction factor. The mean of the correction factors for all antigenic components was then calculated to give one overall correction factor. Statistical analysis

Densitometric quantification Two SDS-PAGE gels provided sufficient nitrocellulose strips to characterize each isotype. To prevent bias, serum samples from each exFEPOW group were divided between gels. Additionally, to control for variability between gels, and therefore the resultant blots, positive control strips were included in triplicate on each blot. Isotype-specific recognition of each separated antigenic component was quantified by scanning densitometry utilising a SPARCstation IPCbased (Sun Microsystems Inc., USA) Discovery Series densitometer (Vital Scientific Ltd, UK) and Quantity One 1D gel quantification software (Protein & DNA ImageWare systems, USA). The level of antibody binding was calculated as the mean OD of the band multiplied by the band width. Band width is incorporated due to the spread that occurs with high levels of antigen-specific antibody. To standardize between blots, the mean response of the three

The Mann–Whitney U-test and Spearman rank correlation were utilized to test for differences in antibody responses between groups, and correlations between variables, respectively. Where multiple comparisons were carried out, correlations were only considered statistically significant at the 1% level to guard against type 1 errors. Statistical tests were performed with the StatView 4 software package for Macintosh (Abacus Concepts Inc., USA). RESULTS Comparison of copropositive and copronegative exFEPOWs by ELISA Levels of PFE-specific IgG1, IgG4, IgE or IgA isotype were not significantly different between copropositive and copronegative exFEPOWs as measured by ELISA.

Table 2 Spearman rank correlation coefficients (rho) between combinations of antibody isotype responses to separated Strongyloides stercoralis L3 antigenic components amongst copropositive and copronegative ex-Far East Prisoners of War (exFEPOWs) as measured by immunoblotting.

Isotype

19 kDa

21 kDa

29 kDa

30 kDa

31 kDa

33 kDa

40 kDa

44 kDa

G1/G4 G1/E G1/A G4/E G4/A E/A

0·333NS – 0·376* – 0·070NS –

0·114NS – ¹0·117NS – 0·410* –

0·489** 0·307NS 0·032NS 0·517** 0·220NS ¹0·283NS

0·428* 0·425* – 0·720*** – –

0·396* 0·293NS – 0·193NS – –

0·105NS ¹0·127NS 0·239NS. 0·187NS 0·437* 0·192NS

0·008NS – 0·128NS – 0·411* –

0·383* ¹0·226NS 0·273NS ¹0·004NS 0·567** ¹0·154NS

–, Data unavailable;

520

, P > 0·05, *P < 0·05, **P < 0·01, ***P < 0·0001.

NS

q 1999 Blackwell Science Ltd, Parasite Immunology, 21, 517–526


Figure 1 Serum IgG1, recognition of separated Strongyloides stercoralis L3 antigenic components amongst ex-Far East Prisoners of War (exFEPOWs) as characterized by immunoblotting. Blots show responses for two groups of exFEPOWs with detectable (copropositives, n ¼ 17), and undetectable (copronegatives, n ¼ 17) levels of larval ouput. (kDa – molecular weight standards). q 1999 Blackwell Science Ltd, Parasite Immunology, 21, 517–526


Figure 2 Serum IgG4 recognition of separated Strongyloides stercoralis L3 antigenic components amongst ex-Far East Prisoners of War (exFEPOWs) as characterized by immunoblotting. Blots show responses for two groups of exFEPOWs with detectable (copropositives, n ¼ 17), and undetectable (copronegatives, n ¼ 17) levels of larval ouput. (kDa – molecular weight standards).

521

J.F.Lindo et al.

Figure 3 Serum IgE recognition of separated Strongyloides stercoralis L3 antigenic components amongst ex-Far East Prisoners of War (exFEPOWs) as characterized by immunoblotting. Blots show responses for two groups of exFEPOWs with detectable (copropositives, n ¼ 17), and undetectable (copronegatives, n ¼ 17) levels of larval ouput. (kDa – molecular weight standards).

522

Parasite Immunology

Figure 4 Serum IgA recognition of separated Strongyloides stercoralis L3 antigenic components amongst ex-Far East Prisoners of War (exFEPOWs) as characterized by immunoblotting. Blots show responses for two groups of exFEPOWs with detectable (copropositives, n ¼ 17), and undetectable (copronegatives, n ¼ 17) levels of larval ouput. (kDa – molecular weight standards). q 1999 Blackwell Science Ltd, Parasite Immunology, 21, 517–526


Comparison of copropositive and copronegative exFEPOWs using immunoblotting Immunoblotting results for the two exFEPOW groups are shown in Tables 1 and 2 and Figures 1 to 4. Antigen recognition patterns for IgG1, IgG4 and IgE were qualitatively similar. The major immunodominant antigens which reacted with these three isotypes comprised components of 29, 30, 31, 33, 33·5, 36, 44 and 56 kDa. Although present, IgG1 and IgG4 reactivity to the 32 kDa component could not be differentiated by densitometry. IgA reacted with more antigens than any other isotype with major immunodominant components of 19, 21, 29, 33, 40, 44 and 62 kDa. IgA reactivity with 30, 31 and 32 kDa components was minimal and could not be quantified densitometrically. Quantitative densitometry was carried out on all detectable components for each isotype, however, only results for which a statistically significant difference in recognition between copropositives and copronegatives was observed, are reported here. Comparison of antigen-specific antibody responses between copronegative and copropositive exFEPOWs demonstrated differentials in the magnitude of antibody responses to separated antigenic components (Table 1). Amongst copronegatives, the 30 kDa component was recognized significantly more strongly by IgG1, IgG4 and IgE. IgG4 responses to the 33 kDa component were also significantly higher amongst copronegatives as were IgE responses to the 31 kDa and 32 kDa components. IgA reactivity with six immunodominant components was significantly enhanced amongst copronegatives. In contrast, IgE recognition of the 29 kDa component was significantly higher amongst copropositives. Multiple comparisons of combinations of isotype responses to each separated component were carried out (Table 2). To guard against type 1 errors, Spearman rank correlation coefficients were only regarded as significant at the 1% level. These correlations showed that IgG4 and IgE recognition of the 29 and 30 kDa components were significantly positively correlated. Additionally, IgA and IgG4 recognition of the 44 kDa component, and IgG1 and IgG4 reactivity with the 29 kDa component were significantly positively correlated.

DISCUSSION A few studies have addressed the role of S. stercoralisspecific antibody responses in regulating larval output, and these have reported no association between parasitological and immunological parameters (Genta et al. 1984, 1986, Sato et al. 1985). These studies typically describe the levels of antibody responses to crude larval extracts measured by q 1999 Blackwell Science Ltd, Parasite Immunology, 21, 517–526


ELISA. However, utilization of immunoblotting facilitates the characterization of antigen-specific responses which might otherwise be masked by such assays. Indeed, comparison of groups here by ELISA provided no evidence of a differential in humoral responses, whilst elucidation of antigen-specific antibody responses using immunoblotting highlighted a number of novel patterns. Unfortunately, in the few studies where immunoblotting has been utilized (Genta et al. 1987, Genta & Lillibridge 1989), quantification of component-specific responses was not carried out. We describe here the first study to utilize quantitative immunoblotting in the analysis of S. stercoralis L3 antigen-specific antibody isotype responses. The most striking result of this investigation was the observed association between the absence of detectable S. stercoralis larvae in stool, and IgA recognition of immunodominant antigenic components. IgA reactivity with six immunodominant components was significantly elevated in individuals with undetectable larval output by stool culture. This indicated that enhanced IgA recognition of filariform larval antigens may decrease larval production, and is consistent with a role for IgA-mediated immune effector mechanisms in modulating larval output. We have previously shown that upregulation of IgA secretion may occur in longstanding S. stercoralis infection (Atkins et al. 1997). The negative association observed here between IgA responses and larval shedding supports the hypothesis of an acquired effector role for IgA in human strongyloidiasis. The precise mechanisms by which the host modulates larval output are unknown. However, parameters such as worm fecundity, egg viability, infection intensity and the autoinfective rate are likely to be involved. An effector role for IgA has been postulated in other helminthiases including Trichuris trichiura (Needham et al. 1993, Needham & Lillywhite 1994) and Taenia taeniformis (Lloyd & Soulsby 1978). IgA has also been shown to participate in eosinophildependent cytotoxicity against Schistosoma mansoni (Dunne 1993, Grezel et al. 1993) and in contrast to IgE, which plays a prominent role in acquired resistance to reinfection, IgA appears to be a major humoral factor in reducing worm fecundity and egg viability (Grzych et al. 1993, Capron et al. 1994). Similarly, it has been suggested that IgA-mediated eosinophil effector functions may be involved in immunity to human Necator americanus infection (Pritchard 1995). This is particularly interesting in light of the observation of Gill & Bell (1979) that eosinophilia was significantly elevated amongst copronegative compared to copropositive exFEPOWs. It is unclear at this stage whether the immunodominant L3 antigenic components described here are shared with adult worms. However, if this should prove to be the case then IgA-mediated effector responses analogous to those described for schistosomiasis 523

J.F.Lindo et al.

and necatoriasis may play a role in modulating larval production in strongyloidiasis. Despite these findings, Mansfield & Schad (1992) observed no difference in S. stercoralis parasitological parameters amongst intact dogs infected with S. stercoralis compared to those congenitally deficient in serum and mucosal IgA. It seems likely therefore that whilst IgA may not be critical in the clearance of infection, it may be involved in the modulation of worm fecundity and egg viability in chronic S. stercoralis infections. However, it must be noted that in the IgA-deficient canine model there appears to be an immune effect on adult females in the intestine which results in them becoming barren and stunted (Grove et al. 1983, Schad et al. 1984, Mansfield et al. 1996). This precedes expulsion of the worms and parasitic females have been observed in dogs in which the infection is no longer patent. These barren females were found to be significantly shorter than fecund females and had notable intestinal and ovarian ultrastructural changes (Schad et al. 1997). The effects can be reversed by passive transplantation to parasite-naı¨ve hosts and these, in turn, quite quickly experience an autoinfective burst (Schad et al. 1997). These reversible antifecundity effects on adult worms, which may not be IgA mediated, are also likely to be central to the maintenance of chronic occult infections. Furthermore, putative antibody-induced effects on larval shedding is seen in the reduction in the L1/adult ratio in the gut following transfer of immune serum to S. stercoralis infected gerbils (Thompson et al. 1997). Although, the number of L1 was reduced, females were apparently unaffected since the number of uterine eggs, length of females and intestinal and ovarian ultrastructure were unaffected. Reduction in larval shedding can therefore occur in the absence of a reduction in fecundity of the females. On the other hand, in dogs with occult infection no larvae were recovered from the mesenteric lymph nodes or lungs which are the sites of migrating autoinfective larvae (Schad et al. 1993). This supports the observation of immune damage to the eggs and larvae of the small proportion of ovigerous females seen in these dogs. IgE responses to three immunodominant antigenic components were also significantly elevated in copronegative individuals. IgE has been implicated in acquired resistance to reinfection in schistosomiasis (Hagan et al. 1991, Demeure et al. 1993), and enhanced IgE responses targeted at migrating S. stercoralis larvae might be expected to reduce the number that successfully complete autoinfection. This would result in a decrease in infection intensity and a consequent reduction in larval output. In support of this hypothesis, we have shown that concurrent human T-cell leukaemia virus type-1 (HTLV-1) infection, which is associated with hyperinfection and disseminated disease, 524

Parasite Immunology

is also associated with profound suppression of IgE responses targeted at filariform larvae (Atkins et al. 1998). Studies into the effects of S. stercoralis infection chronicity on antibody responses suggest that IgE levels are downregulated with increasing duration of infection (Gill et al. 1979, Atkins et al. 1997), whilst IgG4 responses are upregulated early and persist in chronic infections (Atkins et al. 1997). This pattern of sustained IgG4 secretion in combination with the development of an IgE hyporesponsive state may be central to the establishment and persistence of chronic S. stercoralis infection, and also to the low levels of gut-associated morbidity that have been reported in longstanding strongyloidiasis (Gill & Bell 1979). Central to these postulated roles for IgG4 in strongyloidiasis and other helminthiases (Hagan et al. 1991, Kurniawan et al. 1993), is the ability to block IgE-mediated effector responses. The parallel antigen recognition profiles of IgE and IgG4 observed here are consistent with the parallel patterns described for filariasis (Hussain & Ottesen 1986). Furthermore, a significant positive correlation between IgE and IgG4 recognition of the 29 and 30 kDa antigenic components was observed. It is possible therefore, that these components are strongly immunogenic and elicit pronounced allergic and blocking responses. The studies reported here did not examine the role of IgM on larval shedding. However, this antibody has been shown to act in concert with eosionophils in the killing of filariform larvae in the BALB/cByJ model (Abraham et al. 1995, Brigandi et al. 1996, 1997). They should be extended to examine the role of this antibody in larval shedding and maintenance of chronic infections in humans. In summary, we have shown that chronically infected exFEPOWs with undetectable larval output have enhanced IgA and IgE recognition of S. stercoralis filariform larval antigens. We postulate that IgA-mediated responses inhibit worm fecundity and egg viability in human strongyloidiasis, whilst IgE-mediated responses are involved in regulating autoinfective larvae. Parallel upregulation of IgE and IgG4 responses to certain antigenic components suggests that IgG4 blockade of IgE-mediated allergic responses occurs and may be central to the establishment and persistence of asymptomatic chronic strongyloidiasis. These studies may also have major implications for the diagnosis of occult strongyloidiasis, which is becoming increasingly important with the continuing problem of transplantation and treatment for arthritis. ACKNOWLEDGEMENTS We thank Jane Lillywhite and Wayne Forbes for their contributions to these studies. We gratefully acknowledge q 1999 Blackwell Science Ltd, Parasite Immunology, 21, 517–526


the support of the Wellcome Trust. N.S.A was supported by a SERC research studentship. REFERENCES Abraham D., Rotman H.L., Haberstroh H.F. et al. (1995) Strongyloides stercoralis: protective immunity to third-stage larvae in BALB/cByJ mice. Experimental Parasitology 80, 297–307 Archibald L.K., Beeching N.J., Gill G.V. et al. (1993) Albendazole is effective treatment for chronic strongyloidiasis. Quarterly Journal of Medicine 86, 191–195 Atkins N.S., Lindo J.F., Lee M.G. et al. (1997) Humoral responses in human strongyloidiasis: correlations with infection chronicity. Transactions of the Royal Society of Tropical Medicine and Hygiene 91, 609–613 Atkins N.S., Lindo J.F., Lee M. et al. (1998) Strongyloidiasis and human T-cell leukaemia virus type-1: immunomodulatory effects of concurrent HTLV-1 infection. Journal of Acquired Immunodeficiency Syndrome and Human Retrovirology 18, 188–190 Badaro´ R., Carvalho E.M., Santos R.B. et al. (1987) Parasite-specific humoral immune responses in different clinical forms of strongyloidiasis. Transactions of the Royal Society of Tropical Medicine and Hygiene 81, 149–150 Bailey J.W. (1989) A serological test for the diagnosis of Strongyloides antibodies in ex Far East prisoners of war. Annals of Tropical Medicine and Parasitology 83, 241–247 Brigandi R.A., Rotman H.L., Nolan T.J. et al. (1997) Chronicity in Strongyloides stercoralis infections: dichotomy of the protective immune response to infective and autoinfective larvae in a mouse model. American Journal of Tropical Medicine and Hygiene 56, 640–646 Brigandi R.A., Rotman H.L., Yutanawiboonchai W. et al. (1996) Strongyloides stercoralis: role of antibody and complement in immunity to third-stage larvae in BALB/cByJ mice. Experimental Parasitology 82, 279–289 Capron A., Riveau G., Grzych J.M. et al. (1994) Development of a vaccine strategy against human and bovine schistosomiasis: Background and update. Tropical and Geographical Medicine 46, 242–246 Conway D.J., Atkins N.S., Lillywhite et al. (1993a) Immunodiagnosis of Strongyloides stercoralis infection: a method for increasing the specificity of the indirect ELISA. Transactions of the Royal Society of Tropical Medicine and Hygiene 87, 173–176 Conway D.J., Bailey J.W., Lindo J.F. et al. (1993b) Serum IgG reactivity with 41-, 31-, and 28-kDa larval proteins of Strongyloides stercoralis in individuals with strongyloidiasis. Journal of Infectious Diseases 168, 784–787 Conway D.J., Lindo J.F., Robinson R.D. et al. (1994) Strongyloides stercoralis: charactersiation of immunodiagnostic larval antigens. Experimental Parasitology 79, 99–105 Demeure C.E., Rihet P., Abel L. et al. (1993) Resistance to Schistosoma mansoni in humans: influence of the IgE/IgG4 balance and IgG2 in immunity to reinfection after chemotherapy. Journal of Infectious Diseases 168, 1000–1008 Dunne D.W. (1993) The use of mouse/human chimaeric antibodies to investigate the roles of different antibody isotypes, including IgA2, in the killing of Schistosoma mansoni by eosinophils. Parasite Immunology 15, 181–185 Furth A. (1980) Removing unbound detergent from hydrophobic proteins. Analytical Biochemistry 109, 207–215 q 1999 Blackwell Science Ltd, Parasite Immunology, 21, 517–526


Genta R.M. (1992) Dysregulation of strongyloidiasis: a new hypothesis. Clinical Microbiology Reviews 5, 345–355 Genta R.M., Frei D.F. & Linke M.J. (1987) Demonstration and partial characterization of parasite-specific immunoglobulin A responses in human strongyloidiasis. Journal of Clinical Microbiology 25, 1505–1510 Genta R.M., Harper J.S., Gam A.A. et al. (1984) Experimental disseminated strongyloidiasis in Erythrocebus patas. Part II. Immunology. American Journal of Tropical Medicine and Hygiene 33, 444–450 Genta R.M. & Lillibridge J.P. (1989) Prominence of IgG4 antibodies in the human responses to Strongyloides stercoralis infection. Journal of Infectious Diseases 160, 692–699 Genta R.M., Schad G.A. & Hellman M.E. (1986) Strongyloides stercoralis: parasitological, immunological and pathological observations in immunosuppressed dogs. Transactions of the Royal Society of Tropical Medicine and Hygiene 80, 34–41 Gill G.V. & Bailey J.W. (1989) Eosinophilia as a marker for chronic strongyloidiasis-use of a serum ELISA test to detect asymptomatic cases. Annals of Tropical Medicine and Parasitology 83, 249–252 Gill G.V. & Bell D.R. (1979) Strongyloides stercoralis infection in former Far East Prisoners of War. British Medical Journal 2, 572–574 Gill G.V., Bell D.R. & Fifield R. (1979) Lack of immunoglobulin E response to longstanding strongyloidiasis. Clinical and Experimental Immunology 37, 292–294 Gill G.V., Bell D.R. & Reid H.A. (1977) Strongyloidiasis in ex-Far East prisoners of war. British Medical Journal 1, 1007 Goulston K.J., Dent O.F. & Chapuis P.H. (1985) Gastrointestinal morbidity among World War II prisoners of war: 40 years on. Medical Journal of Australia 143, 6–10 Grezel D., Capron M., Grzych J.M. et al. (1993) Protective immunity induced in rat schistosomiasis by a single dose of the Sm28GST recombinant antigen: effector mechanisms involving IgE and IgA. European Journal of Immunology 23, 454–460 Grove D.I. (1980) Strongyloidiasis in Allied ex-prisoners of war in south-east Asia. British Medical Journal 3, 598–601 Grove D.I. (1986) Replicating helminth parasites of man. Parasitology Today 2, 107–111 Grove D.I. (1989) Clinical manifestations. In Strongyloidiasis: a Major Roundworm Infection of Man, ed. D.I. Grove, p. 155–173, Taylor & Francis, London Grove D.I., Heenan P.J. & Northern (1983) Persistent and disseminated infections of Stronglyoides stercoralis in immunocompromised dogs. International Journal of Parasitology 13, 483–490 Grzych J.-M., Grezel D., Xu C.B. et al. (1993) IgA antibodies to a protective antigen in human Schistosomiasis mansoni. Journal of Immunology 150, 527–535 Hagan P., Blumenthal U.J., Dunne D.W. et al. (1991) Human IgE, IgG4 and resistance to reinfection with Schistosoma haematobium. Nature 349, 243–245 Hussain R. & Ottesen E.A. (1986) IgE responses in human filariasis. IV. Parallel antigen recognition by IgG and IgG4 subclass antibodies. Journal of Immunology 136, 1859 Kurniawan A., Yazdanbakhsh M., van Ree R. et al. (1993) Differential expression of IgE and IgG4 specific antibody responses in asymptomatic and chronic human filariasis. Journal of Immunology 150, 3941–3950 Laemmli U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685

525

J.F.Lindo et al.

Leighton P.M. & MacSween H.M. (1990) Strongyloides stercoralis: the cause of an urticarial-like illness of 65 years duration. Archives of Internal Medicine 150, 1747–1748 Lloyd S. & Soulsby E.J.L. (1978) The role of IgA immunoglobulins in the passive transfer of protection to Taenia taeniaformis in the mouse. Immunology 36, 939–945 Mansfield L.S., Niamatali S., Bhopale V. et al. (1996) Strongyloides stercoralis: maintenance of exceedingly chronic infections. American Journal of Tropical Medicine and Hygiene 55, 617–624 Mansfield L.S. & Schad G.A. (1992) Strongyloides stercoralis infection in IgA-deficient dogs. American Journal of Tropical Medicine and Hygiene 47, 830–836 Needham C.S. & Lillywhite J.E. (1994) Immunoepidemiology of intestinal helminthic infections. 2. Immunological correlates with patterns of Trichuris infection. Transactions of the Royal Society of Tropical Medicine and Hygiene 88, 262–264 Needham C.S., Lillywhite J.E., Didier J.M. et al. (1993) Agedependency of serum isotype responses and antigen recognition in human whipworm (Trichuris trichiura) infection. Parasite Immunology 15, 683–692 Pelletier L.L. (1984) Chronic strongyloidiasis in World War II Far East ex-prisoners of war. American Journal of Tropical Medicine and Hygiene 33, 55–61 Pritchard D.I. (1995) The survival strategies of hookworms. Parasitology Today 11, 255–259

526

Parasite Immunology

Purtilo D.T., Meyers W.M. & Connor D.H. (1974) Fatal strongyloidiasis in immunosuppressed patients. American Journal of Medicine 56, 488–493 Sato Y., Takara M. & Otsuru M. (1985) Detection of antibodies in strongyloidiasis by enzyme-linked immunosorbent assay (ELISA) Transactions of the Royal Society of Tropical Medicine and Hygiene 79, 51–55 Schad G.A., Hellman M.E. & Muncey D.W. (1984) Strongyloides stercoralis: hyperinfection in immunosuppressed dogs. Experimental Parasitology 57, 287–296 Schad G.A., Thompson F., Talham G. et al. (1997) Barren female Strongyloides stercoralis from occult chronic infections are rejuvenated by passive transfer to parasite-naı¨ve recipient hosts and give rise to an autoinfective burst. The Journal of Parasitology 83, 785–791 Schad G.A., Smith G., Megyeri Z. et al. (1993) Strongyloides stercoralis: an initial autoinfective burst amplifies primary infection. American Journal of Tropical Medicine and Hygiene 48, 716–725 Thompson F., Nolan T.J., Bhopale V.M. et al. (1997) Passive transfer of immune serum reduces the L1/adult ratio of worms recovered from the intestines of Strongyloides stercoralis infected gerbils. The Journal of Parasitology 83, 170–172 Towbin H., Staehelin T. & Gordon J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: proceedure and some applications. Proceedings of the National Academy of Sciences USA 76, 4350–4354

q 1999 Blackwell Science Ltd, Parasite Immunology, 21, 517–526