A Schistosoma mansoni fatty acid-binding protein, Sml4, is the

Proc. Natl. Acad. Sci. USA Vol. 93, pp. 269-273, January 1996 Medical Sciences

A Schistosoma mansoni fatty acid-binding protein, Sml4, is the potential basis of a dual-purpose anti-helminth vaccine MIRIAM TENDLER*t, CRISTIANA A. BRITOt, MONICA MAGNO VILAR*, NICOLAU SERRA-FREIRE§, CATIA M. DIOGO*, MARILIA S. ALMEIDA*, ALEXANDRE C. B. DELBEM1, JOSEi FIGUEIREDO DA SILVAII, WILSON SAVINO*, RICHARD C. GARRATTF, NAFTALE KATZt, AND ANDREW J. G. SIMPSONt** *Instituto Oswaldo Cruz-Fundacao Oswaldo Cruz, Rio de Janeiro, RJ, Brazil; tCentro de Pesquisas Rene Rachou-Fundacao Oswaldo Cruz, Belo Horizonte, 30190-002, MG, Brazil; §Universidade Federal Rural do Rio de Janeiro, Rio de Janeiro, RJ, Brazil; lInstituto de Fisica de Sao Carlos, Universidade de Sao Paulo, Sao Carlos, SP, Brazil; and 'lUniversidade Federal de Pernambuco, Recife, PE, Brazil

Communicated by Warwick E. Kerr, Universidade Federal de Uberldndia, Uberlandia, Brazil, July 13, 1995

aimed primarily for veterinary use against an economically important disease such as fascioliasis, may represent an attractive route to preclinical development of a vaccine against human schistosomiasis.

Molecular cloning of components of protecABSTRACT tive antigenic preparations has suggested that related parasite fatty acid-binding proteins could form the basis of the protective immune crossreactivity between the parasitic trematode worms Fasciola hepatica and Schistosoma mansoni. Molecular models of the two parasite proteins showed that both molecules adopt the same basic three-dimensional structure, consisting of a barrel-shaped molecule formed by 10 antiparallel (8-pleated strands joined by short loops, and revealed the likely presence of crossreactive, discontinuous epitopes principally derived from amino acids in the C-terminal portions of the molecules. A recombinant form of the S. mansoni antigen, rSml4, protected outbred Swiss mice by up to 67% against challenge with S. mansoni cercariae in the absence of adjuvant and without provoking any observable autoimmune response. The same antigen also provided complete protection against challenge with F. hepatica metacercariae in the same animal model. The results suggest that it may be possible to produce a single vaccine that would be effective against at least two parasites, F. hepatica and S. mansoni, of veterinary and human importance, respectively.

MATERIALS AND METHODS Sequence Analysis and Molecular Modeling. The amino acid sequences of Sm14, Fhl5, and nine human homologues were obtained from the OWL nonredundant protein sequence database (7). The crystallographic structures of FABP from human muscle (8), mouse adipocyte (9), and rat intestine (10) were obtained directly from the Brookhaven Protein Data Bank (accession codes 2HMB, 1ALB, and 2IFB, respectively). Sequences of known crystal structure were aligned by leastsquares superposition of the molecules using C' coordinates alone, and the remaining sequences were subsequently incorporated into the alignment by the method of Barton and Sternberg (11) as implemented in the AMPS package. For the construction of the Sm14 model, the backbone of the 10 (3-strands and three a-helices was based on that of 1ALB. (3-Turn types were determined on the basis of the position of glycine and/or asparagine and aspartic residues within the turn, which normally dictates their conformation. Side chains were based on the conformation of identical residues in homologous positions, where available, or otherwise on preferred rotamers for the given secondary structure. The model was energy minimized by means of GROMOS within the graphics program WHATIF (12). The final model shows good stereochemistry as evaluated by the program PROCHECK (13), with 95% of the residues falling within the most favored regions of the Ramachandran plot. The model presents an Eisenberg environment quality of 47.65, which is acceptable for a protein of 133 residues in length (14), and WHATIF quality of -0.52, which is above the threshold expected for correct structures (15). An analogous procedure was adopted for the modeling of Fhl5, yielding Eisenberg and WHATIF qualities of 59.92 and -0.33, respectively. Expression of Recombinant Sm14 (rSml4). The entire open reading frame of Sm14 was excised from the construct pDSSml4 (5) with BamHI and Hindlll restriction endonucleases, cloned into pGEMEX-1 (Promega), and used to transform Escherichia coli strain BL21(DE3). Expression of a 40-kDa fusion protein consisting of the first 260 aa of the major bacteriophage T7 capsid protein together with the complete

Schistosomiasis, caused principally by Schistosoma mansoni, S. haematobium, and S. japonicum, afflicts some 200 million individuals in tropical regions of the world. Fascioliasis caused by Fasciola hepatica is an economically important disease of cattle and sheep in Europe, the Americas, Australia, and New Zealand. There are no vaccines against Schistosoma or Fasciola species; however, there is evidence for protective immune crossreactivity between S. mansoni and F. hepatica. Hillyer and coworkers (1-3) have isolated a low molecular weight F. hepatica fraction that protects against both S. mansoni and F. hepatica infections. A component of this fraction is an antigen with homology to mammalian fatty acid-binding proteins that is termed Fhl5 (4). A similar antigen, Sm14, was cloned from S. mansoni following studies of a protective saline extract of adult worms, SE (5). These results suggested that the pair of similar parasite proteins could mediate immune crossreaction and represent the basis of a subunit vaccine effective against both species. We have investigated the molecular relationship of Fhl5, Sml4, and mammalian fatty acid-binding proteins (FABPs) (6) in detail and found that Fhl5 and Sml4 contain potential discontinuous, crossreactive epitopes not present in their mammalian counterparts. Furthermore, we have found that a recombinant fusion protein containing the complete Sml4 polypeptide stimulates a protective response against both S. mansoni and F. hepatica infection. The data thus suggest that Sm14 could form the basis of a single vaccine that is effective against both parasites. Such a dual-purpose vaccine,

Abbreviations: FABP, fatty acid-binding protein; rSml4, recombinant Sml4; SE, saline extract of adult Schistosoma mansoni. tTo whom reprint requests should be addressed at: Departamento de Helmintologia, Insituto Oswaldo Cruz-Fundagao Oswaldo Cruz Av. Brasil, 4365-Manguinhos, Cx. Postal 926, CEP 21045-900 Rio de Janeiro, RJ, Brazil. **Present address: Ludwig Institute for Cancer Research, Sao Paulo, SP, Brazil.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

269

270

Proc. Natl. Acad. Sci. USA 93

Medical Sciences: Tendler et al.

sequence of Sm14 (rSml4) was induced during logarithmicphase growth of transformed bacteria in the presence of ampicillin (100 ,ag/ml) by addition of 1 mM isopropyl /-Dgalactopyranoside followed by a further 3 hr of culture. To recover the expressed fusion protein, bacterial cells were collected by centrifugation at 15,000 x g, suspended in 50 mM Tris HCl/2 mM EDTA/1 mM dithiothreitol/0.2% lysozyme at pH 7.5, and incubated on ice for 15 min. The lysate was then sonicated for two 30-sec cycles and recentrifuged, and the pellet was suspended in 50 mM Tris HCl/10 mM EDTA/1 mM dithiothreitol/0.5% Triton X-100 at pH 7.5, and centrifuged again. After a further round of resuspension and centrifugation, the final pellet was suspended in water and subjected to SDS/PAGE. The major band representing the expressed fusion protein was electroeluted and shown to be homogeneous by reelectrophoresis. Immunization Experiments. New Zealand White rabbits were immunized with two intradermal/subcutaneous doses of 600 ,ug of SE [a saline extract of adult S. mansoni (16)] or 80 ,ug of rSml4 in the presence of Freund's complete adjuvant at an interval of 7 days, followed 21 days later by an intraperitoneal booster injection with the same amount of antigen in the absence of adjuvant. For outbred Swiss mice the doses were 300 ,tg and 10 ,ug respectively, in the presence or absence of adjuvant and were administered by the same route and with the same immunization regimen. For assays of protection against S. mansoni, the rabbits were challenged percutaneously with 1000 cercariae and the mice with 100 cercariae (except where indicated otherwise) 60 days after the last immunization and perfused 45 days later. Overall protection was calculated by the formula [(C - V)/C] x 100, where C is the average number of worms in control animals and V is the average number of worms in vaccinated animals. Statistical analysis was done with Student's t test. For histopathological analysis of the mice used for protection experiments, fragments of the liver, intestine, spleen, lungs, and kidneys were taken at the time of perfusion and fixed with 10% formaldehyde, embedded in paraffin, sectioned, and stained with hematoxylin/eosin. For assays of

protection against F. hepatica, Swiss mice, permissive hosts for F. hepatica, were challenged orally with 3 metacercariae 45 days after the last immunization and sacrificed 30 days later. Worms were recovered from the bile duct and the livers of all animals analyzed by using sequential histological sections stained with hematoxylin/eosin. Protection was determined on the basis of histopathology where intraparenchymal liver lesions, characteristics of normal F. hepatica infection, were taken as indicating nonprotected animals.

RESULTS Identification of Potential Crossreactive Epitopes on Sml4 and Fhl5. Sm14 is as closely related to several human proteins-including P2 myelin protein (-42% sequence identity), and FABP from cardiac muscle (%Z42%)-as it is to Fhl5 (-44%). The schistosome and F. hepatica molecules do not exhibit simple, coordinated alterations in size or continuous sequence when compared with their human homologues however. Sm14 shows a marked falloff in conservation with human sequences toward the C terminus (from about residue 85 onward) whereas the two parasite sequences show -47% mean identity within the same region (Fig. 1). This region is the most poorly conserved across the family as a whole (17-19), and this low conservation suggests a possible role for this region in determining Sml4/Fhl5 epitopes. Molecular models constructed for both Sm14 and Fhl5 show that the two molecules adopt the same three-dimensional topology as other members of the FABP family. This consists of a 10-stranded antiparallel ,B-barrel with short interstrand connections which generally form (3-turns and is illustrated for Sml4 in Fig. 2. The C-terminal portion of the molecule, which essentially consti-

SNKFLOTWKLVSSENFDDYHEALOGVGATRKLGNLA KPTVIISKK CDAFVOTWKLVSSENFDDYHKEVOVOFATRKVAGMA KPNHIISVH PVDFTGYWKMLVNENPEEYLRALDVNVALRRIANLL KPDKEIVQD

my-lIn P2...... aFA}DP .......... cREPI .......... cRASPI ........ cRAP2 .........

pPABP-hom ...... iFABP .......... IFASP ..........

PNFAOTWKMRSSENFDELLKALOVNAMLRKVAVAAASKPHVEIRQD

PNFSNWKIXIR9ENFEELLKVLGVNOVLRKIAVAAASKPAVEIKOE APDSTKVDRSENYDKFHEKHGVNXV RAEAAHD MSFSOKYQLOSQENrEAFMIAIQLPEELIQKOKDI VDAYLGTWALVDSKNFDDYHKSLGVQFATROVASMT h2 1 hi

KPTTIIEKN

MAFDQ5TwKVDRNENYEKFHEKHQINVVKRALQAHD

NLKLTITQE

Phl5 ...........

lFABP .......... hFAEP ..........

(MOUSE)..

Fhl5 ...........

Gt

tI I

I

P2 ......

FABP ..........

OVORW

EVQIOW

GEEK3

iqg

70

so

*

8 TEO

CTQTLLEGDGPKTYWTRELAND ELILTFGADDVVCTRIYVRE

cRABP2 .........

CEQKLLtOEOPKTSWTRELTNWDELILTNTADDVVCTRVYVRE DOKESTITRKLEDO KLVVECVMNNVTCTRIYEKVE OHOEW QKFKRTD NCNELNTVREXIID ELVQTYVYEOVEAKRIFKKD TT FKNIKSVTZLNGD IITNTHTLODIVFKRISKRI

It

DQQETTKVRELIDQ KLILTLTHGTAVSTRTYFKEA

9 6 7 10 DKSTTIKRR aD:...........LV1E. 3KQVTTRVYSR sVQ})~ OKFKRVD HOKELIAVREISON ELIQTYTYEGVERI FKKE 'S

5m14 ........... FhlS ...........

ES

EKEQRGWTQWIEEGD ELHLEHRVEQVVCKQVFKKVQ

cRABPI .........

aYABP (HOUSE).. lPABP (RAT)....

e

S

NGKETTIKRKLVDG KMVAECKMKOVVCTRIYEKV DGKSTTIKRKREDD KLVVECVMKOVTSTRVYERA

cRBPI ..........

BLQKW

***

90

go

CVQKQ

pFABP-hom ...... iFABP .......... lFABP ......... hJABP .........

G)

ADGTELTGTWTME GNKLV %

L

7

6

ADDV.±±flJ

*

MLToTEF 4lmv so

myliln

40

5

4

UtdZRStTFK*'lT Lad fflf ONKFTVkSiSNrRNIDVVFELCVDFAYSL *%

Sm14 ...........

" FLL

GDIITIRTESTFKNTEISFKLQQEFEETT ADkRKTKSIVTLQ RCSLu QDVITIKSESTFKNTEISFIL OEFDEVT ADDRKVKSTITLD OGVLV ODHMIIRTLSTFRNYIMDFQVOKEFEEDLTOIDDRKCMTTVSWD GDKLQ ODQrYIKTSTTVRTTEINFKVGEOFEEET VDGRKCRSLATWENENKIH ODTFYIKTSTTVRTTEINFKVOEEFEEQT VDGRPCKSLVRWESENKMV GKNLTIKTESTLKTTQFSCTLaEKFEETT ADGRKTOTVCNFT DGALV ONKFTVKESSAFRNIVVFELOVTFNYNL ADGTELRGTWSLE ONKLI GKIHFKPTITAGSKVIQNEFTVOEECELET MTOEKV1TVVQLEOIDNKLV ODILTLKTHSTFENTEISFRLaVEFDETT ADDRKVKSIVTLD GGKLV 3

aFADP

iFABP (RAT)....

FT4D

T

30

20

10

cRABP2 ......... pFABP-hom ...... IFABP ..........

KN)IUM

ATRQIGNTV VMS M DK ILNA FYL 0 VSE DtYF

Smi4 ...........

FAPBP .......... cRBPI.......... cRJkBPI .........

2

CDAFV§fVtMi

aFABP (MOUSE).. iPABP (PRAT) ....

P2 ......

KPDCIITCD NLKLTITQE KGVSEIVQN

HATVQQLEORWRLVDSKGFDEYMRELGVOIALRKHGAMA

hFABP ..........

my-lin

(1996)

'S

*

(QVD

N

**

iY

K

PKG &C3luEq 100

110

RNYKRLS

TTLL 120

130

FIG. 1. Sequence alignment of members of the FABP family considered in this study. The first nine sequences are all of proteins derived from human tissues: myelin P2 from peripheral nerve; adipocyte FABP, aFABP; cellular retinol-binding protein I, cRBP1; cellular retinoic acid-binding proteins I and II, cRABP1 and cRABP2; psoriasis-related FABP homologue, pFABP-hom; intestinal FABP, iFABP; liver FABP, lFABP; and heart FABP, hFABP. FABPs from mouse adipocyte and rat intestine are also shown, as they were used together with hFABP for the construction of molecular modes. The 13-sheet strands of aFABP are indicated by the shaded blocks and numbered consecutively; the two a-helices (hl and h2) are marked by the solid bars. Identical residues in the two parasite sequences (Sml4 and Fhl5) are boxed. The subset of these residues which are conserved in no more than three of the human sequences are indicated either by stars (for exposed residues) or by the percent symbol (for solventinaccessible residues).

tutes strands 7-10 of the barrel, is shown in blue. The interloop connections were ruled out as candidates for continuous epitopes because they are in general extremely small, unlike the flexible projecting loops which often correspond to antigenic determinants in other proteins (21, 22). Possible discontinuous crossreactive epitopes were thus sought by identifying residues that are identical in Sm14 and Fhl5 but that occur in no more than three of nine human sequences. Twenty-eight such residues were found, 22 of which were exposed on the surface and thus potentially contribute to antigenicity (indicated by a ball-and-stick representation in Fig. 2). Of these 22 residues, 13 were derived from the C-terminal portion of the protein. The internal variant residues represent conservative substitutions or are at highly variant positions and are expected to have a minimum effect on overall structure. The external variant residues, on the other hand, cluster at the upper and lower ends of the barrel, potentially constituting functional

Medical Sciences: Tendler et al.

Proc. Natl. Acad. Sci. USA 93 (1996)

molecule. It was thus reasoned that vaccination with the intact protein was more likely to be successful than vaccination with discrete peptides. Immunization of Experimental Animals with rSml4. Sm14 was expressed as a fusion protein (rSml4) with the bacteriophage T7 major capsid protein. Western blotting studies showed that rSml4 was specifically and strongly recognized by antibodies from mice and rabbits infected with S. mansoni or humans from endemic areas. Furthermore, anti-Sm14 antibodies produced by immunizing New Zealand White rabbits with rSml4 in the presence of Freund's complete adjuvant strongly labeled sections of adult S. mansoni worms and reacted with a 14-kDa molecule in Western blots. However, these antibodies did not label any cell of normal rabbit tissues-including brain, heart, skeletal muscle, small intestine, pancreas, kidney, liver, spleen, thymus, and testis-as assessed immunocytochemically (data not shown). Preliminary assessment of the protective activity of rSml4, using four immunized rabbits challenged with S. mansoni cercariae (Table 1, Exp. 1), showed that rSml4 stimulated an extremely high level of protective immunity. Then the same rSml4/adjuvant preparation was used to vaccinate Swiss mice for experiments varying the magnitude and number of cercarial challenges (Table 1, Exps. 2-6). Again a high level of protection was achieved in all experiments. Four large, independent experiments (Table 2, Exps. 7-10) it showed that rSml4 conferred protection at a high level, with or without Freund's complete adjuvant. Vaccination with the T7 capsid protein alone did not stimulate any measurable protection. Analysis of the distribution of worm burdens among the animals in the four experiments demonstrated that vaccination with rSml4 resulted in a significant skewing with up to 66% of the animals harboring slO worms, in contrast to control groups where the majority of animals had from 21 to 30 worms and none had less than 11 worms (data admitted but not shown). In all unvaccinated mice typical immunopathological reactions were found, including periportal infiltrate, large periovular granuloma, and areas of focal necrosis. In vaccinated animals the extent of periportal infiltrate was greatly reduced and was limited to mononuclear cells. In addition, the number of periovular granulomas was reduced and areas of focal necrosis were absent. Swiss mice immunized with rSml4 were also challenged orally with F. hepatica metacercariae in two independent experiments (Table 3, Exps. 11 and 12). Detailed histopathoto the overall structure of the

FIG. 2. Ribbon diagram of the molecular model for Sm14 constructed on the basis of the three-dimensional structures of the homologous molecules from mouse adipocyte, rat intestine, and human muscle. The C-terminal region of the molecule (from residue 85 onward) is shown in blue. Residues shown in ball-and-stick representation are identical in both Sm14 and FhlS but present in no more than three of the nine human sequences. Residues which fulfill this criterion but which are buried within the interior of the molecule and therefore not expected to participate in epitopes have been excluded. The figure was produced with the program RIBBONS (20).

discontinuous epitopes with significant variation from human proteins. Indeed, the four interstrand connections which are included within the C-terminal part of the structure show pronounced spikes in the main-chain accessibility (19), often correlated with antigenic determinants (22). Thus, the antigenicity of Sm14 appears to depend on dispersed C-terminal amino acids brought together in three-dimensional space due

271

Table 1. Protection of rabbits and mice against experimental infection with S. mansoni cercariae by vaccination with Sm14 or SE in the presence of Freund's complete adjuvant Worm Challenge, % burden no. of No. per protection SEM Group cercariae group Exp. Animals 109.5 ± 1.66 PBS 1000 4 1 Rabbits 89.0* 12.0 ± 1.81 rSml4 93.2* 7.4 ± 0.98 SE 170.0 ± 3.23 PBS 1000 20 2 Mice 65.9* 58.0 ± 2.86 rSml4 62.6 ± 0.86 PBS 500 20 3 Mice 49.7* 31.5 ± 0.9 rSml4 27.2 ± 0.84 PBS 100 20 4 Mice 58.8* 11.2 ± 0.6 rSm14 52.6 ± 0.7 PBS 2 x lOOt 20 5 Mice 37.3* 33.0 ± 0.72 rSml4 ± 0.86 69.6 lOOt PBS x 3 20 Mice 6 39.2* 42.3 ± 0.83 rSml4 PBS, phosphate-buffered saline. *p < 0.05. tChallenges were at 1-week intervals and perfusion was 45 days after final challenge.

272

Medical Sciences: Tendler et al.

Proc. Natl. Acad. Sci. USA 93

(1996)

Table 2. Protection of Swiss mice against experimental infection with 100 S. mansoni cercariae by vaccination with Sm14 or SE in the presence or absence of Freund's complete adjuvant (FCA) No. of Worm burden ± SEM % protection mice Group Exp. 24.5 ± 0.71 12 PBS 7 59.6* 9.9 ± 0.59 19 rSml4 50.6* 12.1 ± 0.57 20 rSml4+FCA 0 27.7 ± 0.53 22 CP 0 28.4 ± 0.53 22 CP+FCA 28.0 ± 0.66 8 PBS 8 51.4* 13.6 ± 0.48 14 rSml4 65.7* 9.6 ± 0.48 10 rSml4+FCA 53.2* 13.1 ± 0.47 20 SE 72.1* 7.8 ± 0.46 21 SE+FCA 26.8 ± 0.76 8 9 PBS 67.9* 8.6 ± 0.6 12 rSml4 62.3* 10.1 ± 0.53 11 rSml4+FCA 56.7* 11.6 ± 0.58 11 SE+FCA 3.4 25.9 ± 0.88 10 FCA ± 7 0.77 28.0 PBS 10 55.3* 12.5 ± 0.66 9 rSml4 9 64.0* 10.1 ± 0.64 rSml4+FCA 10 17.8 23.0 ± 0.57 FCA T7 protein. capsid CP, bacteriophage *p < 0.05.

logical examination revealed 100% protection against parasite maturation and liver damage. This was also observed in a group of mice vaccinated with SE. The livers of all the animals showed that in nonvaccinated animals complete parasite development and migration had occurred, as indicated by extensive damage to the hepatic parenchyma and destruction of the hepatic lobes. In addition, in four and three of the unvaccinated animals in Exps. 11 and 12, respectively, adult worms were found in the bile ducts. In none of the vaccinated animals were parasites of any developmental stage observed, nor were any intraparenchymal hepatic lesions indicative of successful parasite maturation found. Nevertheless, capsular cicatricial lesions were seen in all vaccinated mice, indicating that immature parasites had successfully migrated to the liver but had not developed further. Representative cross-sections of the livers of a vaccinated and unvaccinated mouse are shown in Fig. 3.

DISCUSSION Before embarking on vaccination experiments with rSml4, we considered the structure of Sml4 and Fhl5 in detail. The analysis revealed that the parasite molecules adopt threedimensional structures which are identical in basic form to homologous host proteins, preserving both the overwhelming majority of residues which contribute to the "hydrophobic backbone" and one of the two previously identified alternative recognition motifs for the fatty acid ligand (6). The few variant Table 3. Protection of Swiss mice against experimental infection with three F. hepatica metacercariae by vaccination with Sm14 No. with No. of % protection mice hepatic lesions* Group Exp. 15 15 PBS 11 100 0 15 rSml4 100 0 15 SE 14 14 PBS 12 100 0 15 rSml4 *Intraparenchymal lesions were taken as indicating nonprotected animals.

FIG. 3. Vaccination of mice with rSml4 prevents intraparenchymal hepatic lesions following infection with F. hepatica. (a) Histopathological profile of the liver from an unvaccinated mouse where there is general disruption of the hepatocytes. (b) In contrast, the pattern from a rSml4-vaccinated mouse is normal. Material was fixed in 10% formaldehyde, embedded in paraffin, and stained with hematoxylin/ eosin. CV, centrolobular vein (X770).

amino acids in the interior of the molecule are either in highly variable positions not essential for functions or represent conservative substitutions. The putative epitope-forming amino acids are on the exterior of the molecules and are clustered in the tertiary, but not the primary, structure of the protein. The vaccine trials provided evidence that Sm14 is potently immunogenic and capable of stimulating protective immunity against both S. mansoni and F. hepatica infection, confirming that Sm14 and Fhl5 may represent the molecular basis for the previously observed cross-protection between these two parasites. All the protection data were obtained with outbred animals, thus showing that although the numbers of epitopes on rSml4 may be limited, due to its similarity to host proteins, this did not result in detectable genetic restriction of the protective responses that it stimulates. The levels of protection that we have achieved with rSml4 vaccination against schistosome infection are as high as those

Medical Sciences: Tendler et al. achieved with other candidates for a subunit anti-schistosome vaccine (23-26). It will be of interest to determine whether Sml4 also stimulates protective immunity against other schistosome species, particularly since the homologous protein in S. japonicum contains almost all the amino acids associated with predicted epitope formation (27). In our evaluation, the possibility of a multipurpose vaccine for both veterinary and human use is of fundamental significance in terms of real progress toward effective immunoprophylaxis against schistosomiasis. We acknowledge the Daresbury Laboratory (U.K.) for access to the SEQNET facility. We thank Erika Recone Borges for technical support. This investigation received financial support from the United Nations Development Program/World Bank/World Health Organization Special Program for Research and Training in Tropical Diseases and from the Brazilian agencies Conselho Nacional de Pesquisas, Fundasao de Amparo a Pesquisa do Estado de Sao Paolo, Financidora de Estudos e Projetos, and Fundficao Oswaldo Cruz. 1. Hillyer, G. V. (1984) Vet. Parasitol. 14, 263-283. 2. Hillyer, G. V. (1985) Am. J. Trop. Med. Hyg. 34, 1127-1131. 3. Hillyer, G. V., Haroun, E. T. M., Hernandez, A. & Soler de Galanes, M. (1987) Am. J. Trop. Med. Hyg. 37, 363-369. 4. Rodriguez-Perez, J., Rodrigues-Medina, J. R., Garcia-Blanco, M. A. & Hillyer, G. V. (1992) J. Exp. Parasitol. 74, 400-407. 5. Moser, D., Tendler, M., Griffiths, G. & Klinkert, M.-Q. (1991) J. Biol. Chem. 266, 8447-8454. 6. Banaszak, L., Winter, N., Xu, Z., Bernlohr, D. A., Cowan, S. & Jones, T. A. (1994) Adv. Protein Chem. 45, 89-151. 7. Bleasby, A. & Wootton, J. (1990) Protein Eng. 3, 153-159. 8. Zanotti, G., Scapin, G., Spadon, P., Vaerkamp, J. H. & Sacchettini, J. C. (1992) J. Biol. Chem. 267, 18541-18550. 9. Xu, Z., Bernlohr, D. A. & Banaszak, L. (1992) J. Biochem. 31, 3484-3492.

Proc. Natl. Acad. Sci. USA 93 (1996)

273

10. Sacchettini, J. C., Gordon, J. I. & Banaszak, L. J. (1989) J. Mol. Biol. 208, 327-339. 11. Barton, G. J. & Sternberg, M. J. E. (1987) J. Mol. Biol. 198, 327-337. 12. Vriend, G. (1990) J. Mo. Graphics 8, 52-56. 13. Laskowski, R. A., MacArthur, M. W., Moss, D. S. & Thornton, J. M. (1993) J. Appl. Crystallogr. 26, 283-291. 14. Luthy, R., Bowie, J. U. & Eisenberg, D. (1992) Nature (London) 356, 83-85. 15. Vriend, G. & Sander, C. (1993) J. Appl. Crystallogr. 26, 47-60. 16. Tendler, M., Pinto, R. M., Lima, A. O., Gebara, G. & Katz, N. (1986) Int. J. Parasitol. 16, 347-352. 17. Jones, T. A., Bergfors, T., Sedzik, J. & Yunge, T (1988) EMBO J. 7, 1597-1604. 18. Sacchettini, J. C., Gordon, J. I. & Banaszak, L. J. (1988) J. Biol. Chem. 263, 5815-5819. 19. Muller-Fahrow, A., Egner, U., Jones, T. A., Rudel, H., Spener, F. & Saenger, W. (1991) Eur. J. Biochem. 199, 271-276. 20. Carson, M. (1987) J. Mo. Graphics 5, 103-106. 21. Westhof, E., Altschuh, D., Moras, D., Bloomer, A. C., Mondragon, A., Klug, A. & Van Regenmortal, M. H. V. (1984) Nature (London) 311, 123-126. 22. Thornton, J. M., Edwards, M. S., Taylor, W. R. & Barlow, D. T. (1986) EMBO J. 5, 409-413. 23. Balloul, J. M., Grzych, J. M., Pierce, R. J. & Capron, A. J. (1987) J. Immunol. 138, 3448-3453. 24. Lanar, D., Pearce, E. J., James, S. L. & Sher, A. (1986) Science 234, 593-596. 25. Soisson, L. A., Reid, G. D., Farah, I. O., Nyinfo, M. & Strand, M. (1993) J. Immunol. 151, 4781-4789. 26. Wright, M., Henkle, K. & Mitchell, G. (1990) J. Immunol. 144, 3195-3200. 27. Becker, M. M., Kalinna, B. H., Waine, G. J. & McManus, D. P. (1994) Gene 148, 321-325.