Field of invention

This invention concerns antigens of Schistosoma mansoni (S. mansoni), particularly antigens potentially capable of use as a vaccine against S. mansoni.

Background to the invention

Efforts have been made for many years to develop a vaccine against the parasite S. mansoni, based on various different components of S. mansoni, but these have generally been without success.

The present invention is based on work carried out with the surface membrane of the adult S. mansoni worm (referred to herein as "mb-S" for brevity), which has produced very promising results.

Summary of the invention

According to the present invention there is provided a vaccine against S. mansoni comprising at least one antigen present in mb-S and capable of provoking an immune response.

Experiments have been carried out with a range of different antigenic compositions of S. mansoni, including mb-S, whole worm homogenate (wwh), schistosomulum surface

antigen (schla-S), cercarial antigen (cere) and soluble egg antigen (weh), and in this way mb-S has been indentified as the component containing antigen(s) responsible for immunity. Evidence suggests that these antigens are the basis of the natural development of immunity in man, so they represent a promising basis of a vaccine against S. mansoni for use in man.

The vaccine may comprise mb-S, which is conveniently prepared from worms freshly harvested from infected animals, eg hamsters. The worms may be washed and suspended in phosphate-buffered saline (PBS) and centrifuged to separate out a pellet containing the mb-S fraction.

Preferably, however, the vaccine comprises only useful antigen(s) of mb-S. Work is being carried out to identity and characterise the components of mb-S responsible for provoking an immune response: it is believed that the M

32K, 25K, 22K, 20K, 15K, 13K and 8K antigens from mb-S may be of particular interest, especially the M 25K, 22K, 15K and 13K antigens. At least partial sequences of these four antigens have been established and are given below.

Work so far indicates that mb-S is only protective when presented so that it stimulates a strong antibody response against the M 25K antigen of mb-S and cross-reactive antigens of M r 90K and 38K on the schistosomulum surface,

It is known that antigens in the parasite egg can stimulate a fatal pathological reaction in man, possibly in conjunction with an immune response, so such antigens are clearly not a suitable basis for a vaccine. It has been shown that antigens of mb-S have no cross reaction with the parasite egg, enhancing the prospect of using

such antigens as the basis of a vaccine for man.

The vaccine advantageously incorporates an adjuvant such as those commonly used in vaccines, eg BCG, alum, Freund's Complete Adjuvant (FCA) etc. Saponin is the currently favoured adjuvant as it is the easiest to administer and was found to give consistent protection in experiments with mice.

The invention will be further described by way of illustration in the following Examples and by reference co the accompanying drawings; in which :-

125 Figure 1 illustrates the results of 12% SDS-PAGE of ^" I- labelled schistosoraulum surface antigens immunoprecipitated with antibodies from mice immunized with wwh or the supernatant and pellet fractions of wwh.

Molecular weights refer to the antigens precipitated. The numbers designated to each track refer to the experiment numbers shown in tables II and IV.

125 Figure 2 illustrates the results of 12% SDS-PAGE of I- labelled schistosomulum surface antigens immunoprecipitated with antibodies from mice immunized with mb-S using a variety of adjuvants. Molecular weights refer to the antigens precipitated. The numbers designated to each track refer to the experiment numbers shown in tables V and VI.

Figure 3 is a graph illustrating the results of radio-

125 im u noassay using I-labelled protein A of antibody

CMS = serum from 12-15 week infected mice

VMS = serum from mice exposed 3 times to i rrad i at-^ c e r c a r i a e .

mb-S + sap. = serum from mice vaccinated with mb-S plus saponin; the numbers in parenthesis denote the experiment numbers shown in table VI.

NMS = normal mouse serum.

Figure 4 illustrates the results of 15% SDS-PAGE of 125 I- labelled adult worm surface membrane antigens immunoprecipitated with antibodies from mice immunized with mb-S using a variety of adjuvants. Molecular weights refer to the antigens precipitated. The numbers designated to each track refer to the experiment numbers shown in tables V and VI.

Figure 5 shows the nucleotide and predicted amino acid sequence of the cDNA insert from lambda-SM33.

Figure 6 shows the binding of A) anti-mb-S antibody, B) anti-lambda-SM33 ant-ibody, C) anti-lambda-SM12 antibody,

D) anti-lambda SM70 antibody, E) anti-lambda SM-157 antibody, and F) normal mouse serum, to a Western blot of mb-s.

Figure 7 shows the nucleotide sequence and predicted amin acid sequence of lambda-SM12 encoding part of the M 22K peripheral membrane protein of S. mansoni.

Figure 8 shows the nucleotide and predicted amino acid sequence of lambda-SM70 encoding part of the M 15K acidic, integral membrane protein of S. mansoni.

Figure 9 shows the nucleotide sequence and predicted amin acid sequence, in which the amino acids are represented b the conventionally used single letter symbols, of la bda- SM157 encoding part of the M 13K membrane protein of S.

mansoni .

Figure 10 shows the nucleotide sequence and predicted amino acid sequence, in which the amino acids are represented by the conventionally used single letter symbols, of lambda-SM196 encoding part of the M 25K membrane protein of S. mansoni.

Figure 11 shows the binding of A) normal mouse serum, B) anti-normal mb-S antibody, C) anti-mb-S/CHAPS antibody, and D) anti-mb-S/OTG antibody, to a Western blot of mb-s.

The following describes initial experiments with mice.

Materials and Methods

Parasites and animals

A Puerto Rican strain of S. mansoni was maintained in albino Bio phalaria glabrata snails and Syrian hamsters. All experiments were carried out in female CBA/Ca mice.

Adjuvants

ipopolysaccharide B (S. enteritidis) (LPS) was obtained from Difco Laboratories, Detroit, Michigan, USA and administered at a dose rate of 50 ug/mouse. MDP (N- acetylmuramyl L-L-alanyl-D-isoglutamine 2H 0) was obtained from Behring Diagnostics, La Jolla, CA 92037, USA, and administered as above. BCG vaccine was obtained om Staters Serumi nsti ut DK-2300 Copenhagen, Denmark and given at a rate of 10 organisms per mouse (reference I). Freund's Complete Adjuvant (FCA) and incomplete adjuvant (TFΛ) were obtained from Difco Laboratories and emulsified

with antigen in phosphate-buffered saline (PBS) in a ratio of 2:1 (v/v). Alum-precipitated antigen was prepared according to Dresser (reference 2). Recombinant murine Interf eron-gamma with specific activity of approximately 2 x 10 U/mg was obtained from Ernst-Boehringer-Institut; 7,000 units were used per mouse. Ribi Adjuvant Systems (RAS), a mineral oil containing monophosphoryl lipid A and trehalose dimycolate, was obtained from Ribi Immunochem Research Inc. , PO Box 1409, Hamilton, Montana, 59840, USA. Each vial was warmed to 40°C before 2 ml of antigen in PBS was added with vortexing; 0.2 ml of this mixture was administered to each mouse. Saponin was obtained in powder form from PPF International, Bromborough Port, Wirral, Merseyside, UK and administered in PBS at the rate of 50 ug/mouse.

Antigens

Whole worm homogenate (wwh) was prepared by homogenising adult S. mansoni on ice in a Teflon tissue grinder with PBS. This suspension was centrifuged at 10,000g for 2 minutes and the supernatant stored at -20°C before use. Two further fractions were obtained by centrifuging wwh at 100,000g for 1 hour at 4°C, resulting in a soluble supernatant and an insoluble pellet. The pellet was resuspended in PBS and both fractions were stored at -20°C before use.

Adult worm surface membranes (mb-S) were prepared as follows: freshly harvested worms from 6 week infected hamsters were washed several times with cold PBS, resuspended in warm PBS containing 2.4 mM CaCl„ and 2 M MgCl„ and incubated with gentle mixing at 37°C for 10 mins. The supernatant was drawn off and the worms

- 1 - vortexed in a minimum of buffer for one minute. This supernatant was added to the first and the procedure repeated. The supernatant was then centrifuged at 100,000 g for 1 hour at 4°C. The pellet containing the membrane fraction was resuspended in PBS and stored in liquid N„.

Schistosomulum surface antigen (schla-S) was prepared from mechanically transformed cercariae (reference 3). After washing with PBS, the schistosomula were incubated in 1% Triton - X100 in Krebs-Ringer saline ph 8.0, ImM PMSF and 0.5M EDTA on ice for 1 hour with occasional mixing. After centrifuging for 3 minutes at 14,000g the supernatant was collected and stored at -20°C before use.

Cercarial antigen (cere) was prepared by homogenising cercariae on ice in PBS in a Teflon tissue grinder. The suspension was centrifuged at l,000g for 5 mi ns and the supernatant stored at -20°C before use.

Soluble egg antigen (weh) was prepared in the same way as the cercarial antigen, eggs having been extracted and washed clean from hamster livers by trypsin digestion and centrifugation (reference 4).

The protein content of all antigenic preparations was estimated by the Bradford method (reference 5) using the Bio-Rad Protein Assay (Bio-Rad Laboratories Ltd., Caxton Way, Watford, WD1 8RP, UK).

Antisera

Serum from mice vaccinated with radiation-attenuated cercariae (VMS) was prepared by exposing mice on the abdomen to 500 cercariae irradiated at 20 Krad from a

cobalt source. The mice were boosted with an identical dose af ter 1 and 2 months and bled out af ter a further 14 days. Serum from i nf ected mice (CMS ) was obtai ned by exposi ng mice on the abdomen to 20 cercariae and bleedi ng them out af ter 12-15 weeks. Antisera were prepared from experimental mice by tail-bleedi ng one day before challenge.

Experimental regimes

LPS and MDP were both given with antigen as 2 subcutaneous injections one month apart, and the mice were challenged percutaneously (reference 6) one week after the second injection. BCG and antigen were given intradermally in one dose and the animals challenged after one month. FCA emulsified with antigen was injected intramuscularly followed by a second injection of IFA and antigen after 3 weeks. The challenge was made 2 weeks later. Alum- precipitated antigen was injected subcutaneously on two occasions one month apart and the mice challenged 2 weeks after the second injection. Interferon and antigen were injected subcutaneously on two occasions 3 weeks apart, and the mice were challenged 2 weeks after the second injection. RAS and antigen were given subcutaneously, followed by an intraperitoneal injection of adjuvant and antigen 3 weeks later. Animals were challenged after a further 2 weeks. Saponin and antigen were injected subcutaneously on two occasions three weeks apart and the animals challenged after a further two weeks.

Assessment of protection

The degree of protection to a live infection elicited by the various immunisation procedures was measured by

challenging the mice percutaneously with approximately 130 cercariae and perfusing the livers four weeks later (reference 7). The number of worms recovered from each immunised group of mice was compared with a non-immunised control group and protection was expressed as a percentage according to the following formula:

Mean number of - mean number of worms from the worms from the control group immunised group % immunity = X 100

Mean number of worms from the control group

The degree of significance was determined by the students - t test. Each group consisted of not less than seven mice.

Im unoprecipitation and SDS PAGE

Schla-S and mb-S were iodinated using the iodogen method as previously described (reference 8); mb-S was solubilized in 1% DOC in 10 mM Tris pH 8.2 prior to labelling. Electrophoresis of schla-S was carried out in 12.5% polyacrylamide gels, and mb-S in 15% gels. All gels were dried and autoradiograhed.

E. L. I.S. A.

The enzyme-linked immunosorbent assay was carried out according to Voller et al (reference 9). All antigens were used at a concen ration of 10 ug/ml. Antiser were titrated from an initial concent ion of 1:20 by doubling dilutions to 1:20,48(1. The second antibody was goat anti-

mouse conjugated to horseradish peroxidase, from ICN Biomedicals Ltd.

Radioimmunoassay

The binding of antisera to 3 hr-schistosomula was measured by a radioimmunoassay carried out according to Omer Ali et al (reference 10).

Scanning of autoradiatiographs

The intensity of bands on an autoradiograph was measured in a Joyce Loebl Chromoscan 3.

Results

1. Experiments with Whole Worm Homog e t a te ( wwh )

Initial experiments were designed to test the ability of wwh, in combination with a variety of adjuvants, to induce protective immunity in mice. In two out of three experiments, wwh administered with BCG induced significant but low levels of immunity and in one of two experiments wwh emulsified in FCA promoted a similar level of protection. No protection, however, was achieved with the adjuvants LPS or MDP (Table 1). Protection was also obtained with wwh, when saponin was used as the adjuvant. Since saponin with antigen was easy to administer, and caused little or no trauma in the host, a larger series of experiments was carried out in which various amounts of wwh were injected subcutaneously with this adjuvant. In 12 experiments, significant levels of immunity (19-37%) were induced on 8 occasions (Table II). Experiments 8, 10 and 11 - hut not experiment 9 - showed that protection ci

not induced by antigen alone but only when adminsitered with the adjuvant. The amounts of wwh injected varied from 20 ug - 1,000 ug; all were capable of stimulating a protective response, but no improvement in protection was achieved with the higher doses of antigen.

In order to examine the ability of other schistosome antigenic preparations to induce protection when administered with saponin, mice were vaccinated with schistosomular surface antigen (schla-S), cercarial homogenate (cere) or egg homogenate (weh). Immunity was induced by the schistosomular and cercarial preparations but at no higher levels than that achieved with wwh (Table III).

Two experiments were carried out in which the wwh was fractionated into supernatant and pellet by centrifugatioπ at 100,000g for 1 hour and the fractions administered separately with saponin. The pellet fraction induced a reduction in the number of challenge worms recovered compared to recoveries from non-vaccinated animals, although the immunity induced was statistically significant in only one of the experiments. In neither experiment did the supernatant fraction induce protection (Table IV).

In order to confirm that vaccination with wwh raises antibodies to schistosomulum surface antigens, a number of antisera from the wwh/saponin experiments were used to

125 immunoprecipitate I-labelled schistosomulum surface antigens. Mice immunised with wwh or the wwh pellet induced antibodies to the schistosom lum surface which recognized predominantly antigens of M 20K and M _^ V2-3HK.

No surface antigens wore precipitated by serum from mice

4

immunized with the supernatant fraction of wwh (Figure 1).

2. Expermients with Adult Surface Membranes (mb-S)

In order to further demonstrate that adult membrane associated antigens are protective, additional experiments were carried out in which 100 ug of membranes isolated from the surface of the adult worm (mb-S) were used with variety of different adjuvants for vaccination. No protection was achieved with FCA (1 exp. ) , RAS (1 exp.), alum (2 exps. ) or interferon (2 exps. ) ; in one of two experiments a significant level of protection was obtained when BCG was used as the adjuvant (Table V).

As with wwh, however, 100 ug. mb-S in combination with 50 ug saponin, given subcutaneously on 2 occasions 3 weeks apart, induced significant protection (18-33%, average 27%) in 6 of 7 consecutive experiments (Table VI).

3. Analysis of Antibody Responses following Immunization with mb-S

The specificity of the immune response stimulated by immunization with mb-S was investigated by analysis of th antibodies induced. First, antibody levels against mb-S, schla-S and weh were measured by ELISA (Table VII). Adjuvant alone induced no detectable antibody to any of the antigen preparations. Antigen alone induced titres o up to 1 in 2,560 to mb-S, but no detectable antibody to schla-S or weh. Where adjuvant and membrane were administered together, the response to mb-S was very high Immunization in the presence of saponin produced titres ranging from 1 in 5,120 to 1 in 20,480, which is far in

excess of the anti-mb-S titre in sera from either mice vaccinated with radiation-attenuated cercariae (VMS) or 15-week infected mice (CMS). However, anti-schla-S titres were comparable to that of VMS. Negligible anti-weh antibody was detectable even in those sera with very high titres of anti-mb-S antisera.

The antigens recognized on the surface of schistosomula following immunization with mb-S were determined by

. . . 125 immunoprecipitation of 1-labelled schistosomulum surface antigens (Figure 2). As with mice immunized with wwh or wwh-pellet, the major response in all cases was to the M 38-32K and 20K antigens, but in addition, a weak precipitation of M 90K and M 8K antigens was also r r apparent in some sera. Comparison of the antibody response that resulted from the different immunization protocols showed that saponin specifically increased the level of antibody against the M 38K antigen. In addition, the M 90K antigen was only precipitated by antibodies from mice where saponin, FCA or RAS were used as adjuvants.

In order to determine whether antibodies from mice immunized against mb-S with saponin recognized carbohydrate or polypeptide epitopes on the surface of schistosomula, the binding of anti-sera to intact schistosomula and to schistosomula pre-treated with sodium metaperiodate was measured by radioimmunoassay (Figure 3). The binding of antibodies from mice immunized with mb-S to schistosomula was unaffected by periodate treatment of the parasite. Thus, anti-mb-S antibodies resemble those in VMS rather than CMS and binding is predominantly through polypeptide rather than periodate-sensitive ca bohydr e epitopes.

4

Sera from the immunized mice were also used to immunopreci .pi .tate 125 I-labelled mb-S (Figure 4). The antigenic consitution of mb-S is more complex than schla-

S: nevertheless, major low m.wt. antigens of M 32K, 25K,

20K, 12K and 8K can be distinguished by one-dimensional

SDS-PAGE, which are recognized by antibodies in VMS and

CMS. Of these, the M 25K antigen is recognized specifically by CMS (reference 11). Immunization with mb-

S alone resulted in a weak antibody response against the major low m.wt. antigens. The use of saponin, FCA and RAS as adjuvants greatly increased the antibody response to the M 25K antigen. This specific enhancement was absent in mice which had been vaccinated with mb-S together with the other adjuvants tested.

4. Correlation of anti-surface antibody response with protective immunity

The variation in intensity of precipitation of the 125 I- labelled antigens in both schla-S and mb-S (Figures 3 and

4), as well as the titre of anti-mb-S antibody measured by

ELIS (Table VII) was compared with the level of resistance observed in each group of mice (Tables V, VI) by analysis of correlation coefficients (Table VIII).

There was no significant correlation of total antibody to mb-S or the precipitation of the M 32K or 20K antigens from mb-S, or the M 20K antigen from schla-S. However, a high degree of correlation (p less than 0.002) was observed with the intensity of precipitation of the M

38K-32K complex in schla-S and protection. Comparison of the precipitation clearly demonstrates that it is the intensity o precipitation of the M 38K rather than M 32K antigen that is variable, although the two antigens

have to be measured together during the scanning of the autoradiographs. A lower degree of correlation with protection (0.02 greater than p greater than 0.01) was observed with the intensity of precipitation of the M 90K antigen in schla-S. Of the antigens precipitated only the precipitation of the M 25K antigen correlated with protection (0.02 greater than p greater than 0.01).

Discussion

Although protective immunity to schistosome infection has been unambiguously demonstrated in a variety of animal models, until recently only sporadic progress had been made in reproducing such immunity with antigenic preparations or defined antigens. Latterly there have been a number of reports in which protection has been obtained in the mouse model following vaccination with dead antigen (references 12, 1, 13, 14 and 15) and others in which protection has been transferred passively with monoclonal antibodies (references 16, 17). Thus a variety of schistosome molecules have been shown to stimulate, as well as to act as targets of, protective immune responses. Such antigens include schistosomulum surface glycoconjugates (reference 18), glycosolic polypeptides (reference 19) and a muscle-associated protein (reference 20). The present work demonstrates that immunization with purified adult worm surface membranes induces an antibody response against epitopes on the schistosomulum surface and protects mice against a cercarial challenge.

In preliminary investigations, wwh with BCG, FCA and saponin and mb-S with BCG and saponin induced signi icant immunity in some experiments and, since saponin was the easiest of the adjuvants to administer and caused loss

trauma to the mice than either BCG or FCA, further work was carried out with this adjuvant. In 8 of 12 experiments using wwh with saponin and 6 of 7 experiments using mb-S with saponin, significant levels of immunity ranging from 18%-37% (average 26%) were achieved. Although these levels of immunity are not high, they should be viewed against the level of protection developed by the CBA/Ca mouse after vaccination with live radiation- attenuated cercariae, where in recent experiments (reference 21) levels of only 42-44% immunity have been achieved even after multiple vaccinations. It is

therefore encouraging that the present immunization procedures with non-living antigenic preparations, given in conjunction with saponin, have induced between 41% and 84% of the level of immunity that is achieved in this model after vaccination with a live attenuated vaccine.

Analysis by ELISA of the antibody response of mice immunized with mb-S in conjunction with various adjuvants showed that saponin induced high titres of antibody against mb-S greatly exceeding the titres normally demonstrate with CMS and VMS, with negligible cross- reaction to SEA, but with similar levels of anti- schistosomulum surface antibody as VMS. CMS recognized a higher level of both schistosomulum surface and egg epitopes than either anti-mb-S antisera or VMS. Analysis of binding to intact schistosomula demonstrated that these differences in antibody specificity could be accounted for, at least in part, by the absence in VMS and anti-mb-S antisera of antibody to the dominant sodium metaperiodate- sensitive carbohydrate epitopes. Antibody against these epitopes, shared by the schistosomulum surface and the egg, accounts for more than 90% of schistosomulum surface recognition by CMS (reference 10).

The major antigens in mb-S and on schla-S precipitated by anti-mb-S antisera were a subset of those recognized by

CMS and VMS, and varied depending on the adjuvant used.

Saponin, FCA and RAS in conjunction with mb-S gave the response to the major antigens, whilst BCG, alum and interferon gave no or weaker responses, although both BCG and alum induced antibodies to low molecular weight antigens which were recognized only weakly by the other antisera. The dominant responses in mice successfully vaccinated with mb-S and saponin were to the Mr 38, 32 and

20K antigens on the schistosomulum suface and the M 32, 25 and 20K antigens from adult surface membranes.

The intensity of precipita ion of the major antigens of mb-S and schla-S by sera collected from all mice immunized with mb-S, was measured and these values were compared with the levels of protection induced in the serum donors by analysis of correlation coefficients. Interestingly, the major antigens fell into two distinct types: (1) those that had induced high titres of antibodies in mice which were protected, i.e. the M 38K and M 90K antiqens from schla-S and the M 25K antigen from mb-S, and (2) those antigens which induced levels of antibodies which showed no relationshipr to perotection, i.e. Mr 32K, Mr 20K and M 12K antigens. This result does not necessarily imply a direct relationship between certain antibodies and protective immunity, but it does discriminate a number of antigens which merit further analysis of their role in protection.

Of the antigens recognized by protected mouse sera, the strongest correlation between precipitation and protection was observed with the M 38K antigen on the schistosomulum surface. This antigen is not present in mb-S (reference 11) and thus the presence of antibody against it must be due to its cross-reaction with one of the mb-S antigens.

It has previously been demonstrated that the M 32 and 38K r antigens are closely related but that they contain distinct epitopes (references 22 and 23). In the present study, it is clear that recognition of the M 32K antigen does not vary between mice vaccinated according to the present protocols as judged by the precipitation of mb-S. Thus the basis for increased recognition of the . 38κ antigen on the schistosomulum surface remains to be

defined. Nevertheless, the lack of carbohydrate recognition on the schistosomulum demonstrates that it is polypeptide epitopes on this antigen which are being recoginzed.

The increased recognition of the M 25K antigen in mb-S in protected mice is also of interest. Although this antigen is not exposed on newly transformed cercariae, it has been demonstrated on the surface of 5-day lung stage schistosomula (reference 11). Since it is the lung stage parasites which are susceptible to immune attrition in vivo (reference 24) antigens associated with this stage are also clearly potentially protective immunogens.

It is noteworthy that the schistosomulum surface recognition following protective vaccination with mb-S in the presence of saponin is identical to that detectable in subjects chronically infected with S. mansoni in terms of the antigens precipitated and the lack of antibody to carbohydrate epitopes (reference 25). This raises the possibility that immune responses against the adult membrane may contribute to protective immunity in man. The mouse immunized with membrane antigens in the presence of the adjuvant, saponin, provides a model which should enable identification of those membrane antigens responsible for inducing protection.

Further work has identified a cDNA clone encoding part of the Mr 25K antig^en of mb-S, and has demonstrated that the cloned gene encodes the antigen described above.

Screening of a lambda-g l1 library containing aduir s. mansoni cDNA, using antibodies raised against purified adult worm tegumental surface membranes, identified a cDNA

clone lambda-SM33 expressing a polypeptide which bound antibodies from VMS, CMS and rabit anti-membrane serum (Ram). The cDNA insert of lambda-SM33 was sequenced in both directions, following subcloning into Ml3mpl0 and was also sequenced in situ in lambda-gtll to confirm the selected reading frame. Figure 5 shows the derived nucleotide sequence and predicted amino acid sequence of the 141bp insert, indicating that the cDNA, lacking a termination codon, encodes an internal portion of a polypeptide. A' search of all published schistosome DNA sequences in April 1988 revealed no homology between this cDNA and the genes of other schistosome proteins.

Antibodies raised against a lysate of a lysogen of lambda- SM33 in E. coli host strain Y1089 in mice bound weakly to whole adult worm homogenate in ELISA assays but exhibited binding comparable with that of CMS against purified membranes. These binding characteristics are consistent with the antibody recognizing an antigen present at relatively high density levels within the surface membrane but not generally distributed throughout the adult worm. This was confirmed by binding antibodies against la bda- SM33 to sections of adult S. mansoni, which demonstrated that antigens determinants recognized by anti-lambda-SM33 antibodies are confined to the tegument and the cytons underlying the muscle layer. In contrast, CMS exhibited strong binding throughout the body of the worm. In some sections, antibody to lambda-gtll exhibited an apparent binding to internal structures, which may have been due to auto luoresence. Schistosomula were also exposed to anti- lambda-SM33 antibody but no surface binding was detected.

Immunoprecipitation of cell-free translation products derived from adult S. mansoni mRNA demonstrated that anti-

lambda-SM33 antibodies recognized a nascent polypeptide of

M 22K which was also strongly precipitated by anti- membrane antibody and more weakly by VMS and CMS. When

125 1-labelled surface membranes were used for immunoprecipitation, anti-lambda-SM33 antibodies were found to immunoprecipitate the major M 25K antigen of mb-

S also recognized by CMS. It has been shown of the M 25

• r antigen corresponds to an in vitro translation product of M 22K and is not the processed product of a higher M polypeptide but an authentic low M major membrane component. In addition, a number of lower molecular weight peptides were precipitated which probably represent proteolytic fragments of the major antigen. The identity of the antigen encoded by la bda-SM 33 was substan iated by raising antibody against the recombinant fusion protein purified by preparative gel electrophoresis. The excised polypeptide, which was judged to be pure by silver staining of the protein following SDS-PAGE, also induced anti-M 25K antigen antibody when used to immunize mice.

Further immunoprecipitation of 125ι-labelled adult worm surface membranes demonstrated that the M 25K antigen of mb-S is also recognized by antibodies from infected rats as well a antibodies in the sera of Kenyan and Eyptian subjects infected with S. mansoni and that the M 25K antigen is a predominant antigenic component of the adult worm membrane precipitated by the antibodies.

In the work described above, assays were carried out using radiolabelling and immunoprecipitation. Further assays have been carried out using Western blot techniques: these assays have revealed Mr 22 and Mr 13K antig ^J ens in mb-S, and have confirmed the presence of all previously detected antigens in mb-S apart from that of M 8κ.. In

addition these analyses have suggested that the antigen mb-S previously identified as M 12K (as in Table VIII a Figure 4) is nearer to 15K in size, and this antigen wil be referred to hereafter as 15K.

The most prominent antigens recognized by antibodies fro mice protectively vaccinated with mb-S as analysed by th Western blot technique are of M 25,22,15 and 13K (Figure 6) . The recognition of these proteins by antibodies fro the protectively vaccinated mice is consistent with thei possible contribution to the protective immunity stimulated and indicates their possible value as components of a vaccine. Two dimensional Western blot analyses demonstrate that the M 25K antigen is electrophoretically identical to the antigen of the same M identified by immunoprecipitation and encoded, in par by lambda-SM33. In addition the M 15K antigen co- migratess the M 15-12K antigen identified by precipitation.

Further screening of lambda-gtll expression libraries w antibody against mb-s has identified cDNA clones lambda SM12, lambda-SM70 and lambda-SM157 which encode all or part of the M 22, 15 and 13K antigens respectively. Th identity of the cDNA clones has been established by reacting antibody purified with the fusion proteins expressed by the cDNA clones against Western blots of m (Figure 6). The Figure also confirms that antibody against the peptide encoded by lambda-SM33 recognises the M 25K antigen.

The nucleotide sequence of Iambda-SM12, lambda-SM70 and Iambda-SM157 as well as the deduced amino acid sequence

the proteins they encode are shown in Figures 7, 8 and 9, respectively. The reading frame of Iambda-SM12 is identical to that of a cDNA clone reported by Stein and David (Molecular and Biochemical Parasitology, 20, 253- 264, 1986). The open reading frames of lambda-SM70 and Iambda-SM157 do not cover the entire proteins and further clones encoding the remaining portions of the polypeptides are being sought. In addition lambda-SM33 has been used to rescreen a lambda-gtll adult cDNA library and a larger length clone, lambda-SM196, identified which contains a larger part of the gene encoding the M 25K antigen. The nucleotide sequence and deduced amino acid sequence is shown in Figure 10.

In order to define further, and eventually purify, protective antigens from mb-S it was necessary to develop a procedure to solubilize the protective antigens. Two detergents were used, the non-ionic detergent octyl-beta- D-thioglucopyranoside (OTG) and the non-denaturing zwitterionic detergent 3-( [ 3-cholamidopropyl]- dimethylammonio)-l-propanesulphonate (CHAPS) . Isolated mb-s was solubilized in 1% detergent for 60 mins at 4°C and then centrifuged at 100,000xg for one hour. The supernatants were then adjusted to approximately 250 ug/ml saponin and then dialyzed against 250 ug/ml saponin in PBS for 48h to remove the solubilizing detergent. Mice were then vaccinated with a total of 50 ug of solubilized mb-S in 2-3 doses. As a control mice were also vaccinated with the same amount of normal mb-S and the residual mb-S not solubilized in the detergent in the same concentration of saponin.

It was also investigated whether the protective antigens are specifically associated with mb-S or whether they e

generally distributed throughout the parasite on non- surface membranes. Thus membranes separated from adult parasites from which the surface membrane had previously been removed by exposure to PBS (mb-A) were also solubilized in both CHAPS and OTG and used to vaccinate mice. The degree of protection against a challange cercarial infection induced by the different vaccination protocols is shown in Table IX.

The results demonstrates that both CHAPS and OTG solubilized protective antigens from mb-S and indeed that in both cases the solubilized membrane induced a greater degree of protection than that induced by normal mb-S or by material not solubilized by the detergents used. It was thus demonstrated that the protective antigens associated with mb-S are soluble and retain their protective activity in either OTG or CHAPS. Either detergent may be used for further purification of the protective antigens. The protection induced by mb-A solubilized in either of the two detergents was lower than that resulting from vaccination with mb-S and was only statistically significant in one case. The experiment thus supports the view that protective antigens are specifically associated with the surface membrane of the adult parasite. This has been confirmed by reacting antibody to mb-S to sections of adult worms, visualised with a fluorescein conjugated second antibody. Reactivity was confirmed to the tegument and was not generally distributed throughout the adult worm.

The antibody responses of the mice vaccinated with mb-S solubilized in OTG and CHAPS are shown in Figure 11. In both cases antibody against the M 25, 22, 15 and 13κ antigens was evident. This demonstrates that these

particular antigens had been solubilized and is consistent with their stimulating the protective immunity observed.

Further immunopreci •pi.tati.on experi.ments using 125i-mb-S and antibody specific to lambda-SM70 demonstrated the precipitation of the M 8K antigen as well as the M 15K antigen illustrating that these two antigens are related.

A monoclonal antibody, NIMP.M80, has been produced from a mouse vaccinated with mb-S which exhibited identical specificity following immunoprecipitation of 125i-mb-S.

It was shown that the target antigen could be isolated from total adult worm membranes solubilized in Triton-xiOO using NIMP.M80 conjugated to protein-A sepharose. The antigen was eluted in 1.0% OTG in 100 mM glycine pH 2.5.

The pH of the eluted antigen was returned to 8.0 with 1.0

M phosphate and 250 ug/ml saponin added. The solubilized antigen was then dialysed against 250 ug/ml saponin. The antigen could be visualized as a single band of M 8K following silv-er staining of a SDS-polyacrylamide gel.

The M 15K component may also be present but is known to stain poorly with silver. This procedure provides an alternative source of antigen for vaccination to that produced by expressing the corresponding gene in ______ Coli.

Analysis of antibody from mice vaccinated with the purified antigen revealed detectable antibody only to the

M 8K antigen.

TABLE 1

Ability of whole warm homogenate (wwh) , in combination with a variety of adjuvants, to induce protective immunity in mice.

Exp Adjuvant Antigen Mean worm Percentage No. 200 ug recovery + SD immunity

26a -

TABLE I (continued)

FCA/IFA wwh 52.7 + 10.0 20 p 0.05

FCA/IFA - 66.3 + 10.4

FCA/IFA wwh 45.8 + 6.3 n.s. FCA/IFA 47.3 + 6.9

- 27 -

TABLE_ I_I

Immunisation of mice with whole warm homogenate and saponin

Exp. Adjuvant Antigen Mean worm Percentage No. recovery _+ SD immunity

8. saponin

27a -

TABLE II (continued)

13. saponin wwh (200 ug) 39.4 + 9.5 12 n . s

44.8 + 10.8

14. saponin wwh (200 ug) 44.7 + 7.0 19 p<0.05 55.0 + 8.8

15 saponin wwh (200 ug) 45.4 + 6.2 19 p<0.05 saponin 56.3 + 9.0

16. saponin wwh (200 ug) ^'' 43.8 + 17.9 22 n.s. saponin - 56.4 + 4.8

17. saponin wwh (200 ug) 38.1 + 10.7 16 n.s. saponin - 45.3 + 6.7

18. saponin wwh (200 ug) 36.3 + 3.7 20 p< 0.01 saponin - 45.5 + 4.0

19. saponin wwh (200 ug) 34.7 + 7.6 28 p<0.01 saponin - 48.3 + 5.3

- 28 - TABLE III

Immunisation of mice with schistosomulum suface (schla-S), cercarial homogenate (cere) or egg antigen (weh) with saponin

Exp. Adjuvant Antigen Mean worm Percentage No. recovery + SD immunity

20. saponin schla-S schla-S

21. saponin cere (200 ug). 41.0 + 9.8 25 p<0.02 cere (200 ug) 51.0 + 11.0 7 n.s saponin - 55.0 _+ 8.5

22. saponin weh (200 ug) weh (200 ug) saponin

- 29 -

TABLE IV

Immunisation of mice with fractions of whole worm homogenate (wwh) ^' and saponin

Exp. Adjuvant Antigen Mean worm Percentage No. recovery + SD immunity

23. saponin wwh super 44.2 + 8.4 n.s

(200 ug) saponin wwh pellet 35.6 + 9.9 20 n.s

(200 ug)

44.8 + 10.8

24. saponin wwh super 54.9 + 9.6 0 n.s.

(200 ug) saponin wwh pellet 46.3 + 5.5 16 p<0.05

(200 ug) saponin - 55.0 + 8.8

- 30 - TABLE V

Ability of adult worm surface membrane (mb-S) in combination with a' variety of adjuvants to induce protective immunity in mice.

Exp. Adjuvant Antigen Mean worm Percentage No. 100 ^" ug recovery _+ SD immunity

25. alum

alum

26. alum mb-S 50.8 + 5.6 n.s. alum 54.8 + 8.8

27. interferon mb-S mb-S interferon

28. interferon mb-S mb-S interferon -

29 FCA/IFA mb-S 40.8 + 13.8 10 n.s. FCA/IFA 45.6 + 7.8

- 30a -

TABLE V (continued)

30. BCG mb-S 47.3 + 7.7 0 n.s. BCG - 46.6 + 9.5

31. BCG mb-S 41.8 + 6.5 20 p< 0.02 BCG 52.8 + 6.5

32. RAS mb-S 26.8 + 5.2 20 n.s RAS - 33.8 + 12.8

- 31 - TABLE VI

Immunisation of mice with adult worm surface membranes (mb-S) and saponin.

Exp. Adjuvant Antigen Mean worm Percentage No. 50 ug 100 ug recovery + SD immunity

33. saponin mb-S 40.85 + 11.3 27 p<0.02 56.30 + 9.8

34. saponin .mb-S 40.42 + 4.5 18 p<0.05 49.0 + 6.6

35 saponin mb-S 35.71 + 6.6 26 p<0.001 48.14 + 4.0

- 31a -

TABLE VI continued

- 32 -

TABLE VII

ELISA values of mouse antisera versus 3 different antigens

- 33 - TABLE VIII

Correlation of Protective Immunity with antibody levels

Correlation Signifi coefficient

Titre of antibody against mb-S (ELISA) 0.406 n.s

antibody against M 20K (mb-S) antibody against M _r 25K (mb-S) antibody against M _r 32K (mb-S)

antibody against _r 12K (schla-S) antibody against M _r 20K (schla-S) antibody against _r 32-38K ^' -(schla-S) antibody against M _r 90K (schla-S)

34

Table IX

EXP DETERGENT IMMUNOGEN PROTECTION SIGNIFICANCE

OTG Total mb-S 18% N.S. Soluble 38% p less than 0.01

N.S.

N.S.

2. CHAPS p less than 0.05 p less than 0.02

Insoluble 27% p less than 0.02

Soluble mb-A 19% p less than 0.05

Note: mb-S = isolated tegumetal surface membranes mb-A = non-surface membranes.

-35-

References

1. James, S. L. 1985. Induction of protective immunity against Schistosoma mansoni by a non-living vaccine is dependent on the method of antigen presentation. J. Immunol. 134:1956.

2. Dresser, D. W. 1986. Immunization of experimental animals in Handbook of Experimental Immunology, vol 1, 4th edition ed. D. M. Wier and others. Oxford, Blackwell Scientific, pp.8.1-8.21.

3. Ramalho-Pinto, F. J., G. Gazzinelli, R. E. Howells, T. A. Mota-Santos. E. A. Figueiredo and j. Pellegrino. 1974. Schistosoma mansoni: defined system for step-wise transformation of cercariae to schistosomula in vitro. Exp. Parasitol. 36:360

4. Hackett, F., A. J. G. Simpson, M. Knight, P. Ali, G. Payares, and S. R. Smithers. 1985. Schistosoma mansoni: with killed schistosomula without adjuvant. Proc. Soc. Exp. Biol. Med. 181:454.

5. Bradford, M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Bioche . 72:248

6. Smithers, S. R. and R. J. Terry. 1965. The infection of laboratory hosts with cercariae of Schistosoma mansoni and recovery of the adult worms. Parasitology 55:695.

7. Smithers, S. R. and K. Gammage. 1980. Recovery of

-36- Schistosoma mansoni from the skin, lungs and hepatic portal system of naive mice and mice previously exposed to S. mansoni: evidence for two phases of parasite attrition in immune mice. Parasitoloqy 80:189.

8. Simpson, A. J. G., F. Hackett, T. Walker, R. de Rossi, and S. R. Smithers. 1985. Antibody response against schistosomulum surface antigens and protective immunity following immunization with highly irradiated cercariae of Schistosoma mansonit Parasite Immunology. 7:133.

9. Voller, A., D. E. Bidwell, and A. Bartlett. 1979. The enzyme-linked immunosorbent assay (ELISA) Dyunatech, Europe, Guernsey.

10. Omer Ali, P., A. I. Magee, C. Kelly, and A. J. G. Simpson. 1986. A major role for carbohydrate epitopes preferentially recognized by chronically infected mice in the determination of Schistosoma mansoni schistosomulum surface antigenicit . J. Immunol. 137:3601.

11. Payares, G., D. J. McLaren, W. H. Evans, and s. R. Smithers. 1985. Antigenicity and immunogenicity of the tegumental outer membrane of adult Schistosoma mansoni. Parasite Immunol. 7:45.

12. Tarrab-Hazdai, R., F. Levi-Schaffer, V. Brenner, S. Horowitz, Z. Ξshhar, and R. Arnon. 1985. Protective monoclonal antibody against Schistosoma mansoni: antigen isolation, characterization and suitability for active immunization. J. Immunol. 135:2772.

13. Smith, M. A. and J. A. Clegg. 1985. Vaccination against Schistosoma mansoni with purified surface

-37- antigens. Science 227:535.

14. Hsu, S. Y. Li, H. F. Hsu, K. W. Svestka, and w. Clarke. 1986. Vaccination against schistosomiasis in mice with killed schistosomula without adjuvant. proc . soc . Exp. Biol. Med. 181:454.

15. Balloul, J-M., J-M. Grzych, R. J. Pierce, and A. Capron. 1987. A purified 28,000 Dalton protein from Schistosoma mansoni adult worms protects rats and mice against experimental schistosomiassis. J. Immunol . 138:3448.

16. Smith, M. A., J. A. Clegg, D. Snary, and A. J. Trejdosiewicz. 1982. Passive immunization of mice against Schistosoma mansoni with an IgM monoclonal antibody. Parasitoloqy 84:83.

17. King, C. H., R. R. Lett, J. Nanduri, S. El Ibiary,.P. A. S. Peters, G. R. Olds, and A. A. F. Mahmoud. 1987. Isolation and characterization of a protective antigen for adjuvant-free immunization against Schistosoma mansoni.

J. Immunol. 139:4218.

18. Kelly, C. , A. J. G. Simpson, E. Fox, S. M. Phillips, and S. R. Smithers. 1986. The identification of Schistosoma mansoni surface antigenss recognized by protective monoclonal antibodies. Parasite Immunol. 8:193.

19.- Taylor, J. B., A. Vidal, G. Torpier, D. J. Meyer, C. Roitsh, J-M, Balloul, C. Southan, P. Sondermeyer, S. Pemble, J-P, Lecocq, A. Capron, and B. Ketterer. 1988. The glutathione transferase activity and tissue

-38 - distribution of a cloned M 28K protective antigen of Schistosoma mansoni. EMBO J. 7:465.

20. Sher, A., E. Pearce, S. Hieny, and S. James. 1986. Induction of protective immunity against Schistosoma mansoni by a non-living vaccine. IV. Fractionation and antigenic properties of a soluble adult worm immunoprophylactic activity. J. -Immunol. 136:3878.

21. McLaren, D. J. and S. R. Smithers. 1988. Serum from CBA/Ca mice vaccinated with irradiated cercariae of Schistosoma mansoni protects naive recipients through the recruitment of cutaneous effector cells. Parasitoloqy. 97: 287-302.

22. Dissous, C, J-M. Grzych, and A. Capron. 1985. Biochemical studies on the 30-40 kDa Schistosoma mansoni surface antigens. Mol. Biol . parasitol. 16:227.

23. payares, G., C. Kelly, S. R. Smithers, and W. H. Evans. 1985. Evidence that the 32, 38 and 20 Kilodalton surface antigens of schistosomula and schistosomes are a family of conserved proteins. Mol. Biol. Parasitol. 17:115.

24. McLaren, D. J. and S. R. Smithers. 1987. The immune response to schistosomes in experimental hosts. in The Biology of Schistosomes ed Rollinson D. & Simpson A. J. G. Academic Press London New York pp. 233-263.

25. Simpson, A. J. G., P. Omer Ali, X. Yi , and S. R.

Smithers. 1988. Schistosome glycoconjugates as antigens and their relevance in experimental and human schistosomiasis . Biochemical Society Transactions. 16 : 264-265.