110 1992; 166 (December)
Concise Communications
nism for syphilis, the systemic mycoses and other infectious diseases. J Theor Bioi 1972;36:617-25. 9. Fitzgerald TJ, Tomai MA, Trachte GJ, Tice T. Prostaglandins in experimental syphilis:treponemes stimulate adherent cells to secrete prostaglandin E2 and indomethacin upregulates immune functions. Infect Immun 1991;59:143-9. 10. van der Sluis JJ. Kant M. Onvlee PC, Stolz E. The inaccessibility ofthe outer membrane of adherent Treponema pallidum (Nichols strain) to anti-treponemal antibodies, a possible role of serum proteins. Genitourin Med 1990;66: 165-70.
1453
II. Radolf JD, Norgard MY, Schulz WW. Outer membrane ultrastructure explains the limited antigenicity of virulent Treponema pallidum. Proc Nat! Acad Sci USA 1989;86:2051-5. 12. Walker EM, Zampighi GA, Blanco DR. Miller IN, Lovett MA. Freezefracture analysis demonstrates rare protein in the outer membrane of Treponema pallidum subsp. pallidum. J Bacteriol1989; 171:5585-8. 13. Turner TB, Hollander DH. Biology of the treponematoses. Geneva: World Health Organization. 1957. 14. Alderete JF. Baseman JB. Surface-associated host proteins on virulent Treponema pallidum. Infect Immun 1979;26: 1048-56.
James w. Kazura, Fred Eo Hazlett, Jr., Eric Pearlman, Karen Day, Adel EI-Zeiny, Timothy W. Nilsen, and Michael P. Alpers
Departments of Medicine (Division of Geographic Medicine) and Molecular Biology and Microbiology, Case Wesern Reserve University. Cleveland. Ohio; Department of Pure and Applied Biology. Imperial College, London, United Kingdom; Al-Azhar University, Cairo. Egypt; Papua New Guinea Institute ofMedical Research. Goroka
A 92-kDa fusion protein that encodes amino acids 1-479 of a 62-kDa Brugia malayi antigen induces resistance to microfilariae in mice. The antigenicity of this recombinant protein was explored in asymptomatic residents of Wuchereria bancrofti-endemic areas of Papua New Guinea and Egypt. Unlike sera from individuals in nonendemic areas, sera from residents of endemic areas contained IgG3 antibodies (up to dilution 1:1280) reactive with the fusion protein. There was little or no recombinant antigen-specific IgGl, IgG2, IgG4, or IgE. The mean levelof IgG3 antibodies to the amino acid 1-479 construct in sera of putatively "immune" adult Papua New Guineans and children in whom microfilaremia was 1000 parasites/ml., These data indicate that the filarial antigen corresponding to the recombinant protein is highly immunogenic in naturally infected children and adults and that the isotype and magnitude of antibody reactivity with it do not correlate with the microfilaremic status of asymptomatic persons. Infection by the lymphatic-dwelling filarial parasites Wuchereria bancrofli and Brugia malayi is a major public health problem in many areas of the tropics [1]. Identification of specific parasite antigens that induce the immunopathologic consequences of infection, for example, elephantiasis, acute lymphangitis, tropical pulmonary eosinophilia, and partial resistance to various stages of the helminth, is necessary for
Received 22 April 1992; revised 7 July 1992. Presented in part: meeting of the American Society of Tropical Medicine and Hygiene, Honolulu. December 1989. The studies were approved by the Human Studies Committee. University Hospitals of Cleveland, and health authorities of the government of Papua New Guinea. Informed consent for donation of blood was obtained from all individuals and parents of children. Grant support: World Health Organization/Tropical Disease Research and National Institutes of Health (AI-15351 and contract 73262). Reprints or correspondence: Dr. James W. Kazura, Case Western Reserve University School of Medicine, Division of Geographic Medicine. 2109 Adelbert Rd., Rm. W137. Cleveland, OH 44106-4983. The Journal of Infectious Diseases 1992;166:1453-7 © 1992 by The University of Chicago. All rightsreserved. 0022-1899/92/6606-0042$01.00
development of safe and effective immunoprophylactic measures [2]. Isolation and structural characterization of such antigens have been hampered by the biochemical complexity of these multicellular organisms and the fact that large numbers of each filarial stage cannot be cultivated in vitro. Advances in this area have become possible since the cloning of several filarial proteins. We previously described a cDNA clone selected from a B. malayi Agt11 expression library that has significant nucleotide and predicted amino acid sequence homology to aspartyl tRNA synthetase of Saccharomyces cerevisiae [3]. The cDNA encodes a 62-kDa antigen that is expressed by microfilariae and adult worms. Sensitization of mice to a 92-kDa fusion protein that includes amino acids 1-479 of the predicted 548 amino acid polypeptide induces partial resistance to subsequent challenge with B. malayi microfilariae [4]. The utility of this cloned protein for study of the human immune response has not been established. The objectives of the current study were therefore to determine whether the recombinant filarial polypeptide is antigenic in humans infected with W. bancrofli and, if so, to examine whether there
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on November 15, 2015
Antigenicity of a Protective Recombinant Filarial Protein in Human Bancroftian Filariasis
1454
Concise Communications
are differences in antibody levels to it among residents in endemic areas harboring relatively low versus high microfilarialloads. Because IgG3, IgG4, and IgE antibody levels to extracts of B. malayi adult worms have been shown to correlate with the microfilaremic status of residents of W. bancrofti-endemic areas [5], IgG subclass and IgE antibody reactivities to the recombinant protein were evaluated.
Methods Preparation of B. malayi extracts and recombinant antigens. B. malayi were obtained by lavage of the peritoneal cavities of
h at 37°C, sera were removed, wells washed with PBS-T, and 90 J.LL ofa 1:500 dilution of peroxidase-conjugated murine monoclonal antibody (MAb) to human IgG I (clone HP 6070; Zymed, South San Francisco), IgG2 (HP 6014; Binding Site, San Diego), IgG3 (HP 6047; Zymed), or IgG4 (HP 6025; Zymed) was added. These clones were previously shown to produce MAbs specific for each IgG subclass [9]. After incubation for 18 h at 4°C, MAbs were removed, wells were washed six times with PBS-T, and o-phenylenediamine substrate (Sigma) was added. OD at 492 nm was read at 15 min on an automated ELISA reader. In studies involving the 92-kDa fusion protein, individual serum samples and the pools were preabsorbed with excess trpE. Results of experiments examining antibody levels to the fusion protein in pooled sera are presented as OD. Results ofstudies comparing Ig03 antibody levels in sera from individual Papua New Guinea subjects are expressed as units of antibody per milliliter, which is defined as the ratio of the OD generated by the sample relative to an equivalent dilution of the standard pool from 100 Papua New Guineans. One hundred units ofantibody reactive to the 92-kDa fusion protein indicates that the value for an individual serum sample is equivalent to that of the standard pool at the same dilution. This method of expressing antibody reactivity enables comparison of individual values using a standard curve generated with every assay [10]. IgE antibody to the fusion protein and AWE was measured by ELISA in the laboratory of E. Ottesen (National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD) as described [5]. To remove blocking antibodies, sera were preabsorbed with protein A coupled to Sepharose (Pharmacia). After overnight incubation of I: 10-1 :80 dilutions of serum in fusion protein-coated wells, MAb to human IgE (MAb 7.12, provided by A. Saxon, UCLA School of Medicine, Los Angeles) was added for 2 h at 37°C. Plates were washed and incubated for another 2 h at 37°C with biotinylated goat antimouse IgE (Jackson Immunoresearch, West Grove, PA). This antibody does not react with human IgE. Streptavidin-conjugated alkaline phosphatase (Jackson Immunoresearch) was added and incubation was continued at room temperature for I h. Color was developed using fJ"'nitrophenyl phosphate as substrate. Results are expressed as OD at 405 nm. Cross-reactivity of antibodies to 92-kDa fusion protein and A WE. To confirm that antibody directed against the 92-kDa fusion protein recognized the corresponding native filarial antigen, aliquots of serum (1:400 dilution) from 3 Papua New Guinea subjects were preincubated for 18 hat 4°C in microtiter wells coated with either 10 J.Lg of AWE or BSA (control protein). Antibody to the 92-kDa fusion protein was then measured by ELISA as described above. Statistics. The Wilcoxon rank-sum test was used to assess the significance of difference for antibody levels between groups of Papua New Guinea study subjects with high versus low intensities of microfilaremia. Student's t test was used to compare mean values for Gib 13 phosphocholine epitope levels and relative antibody reactivities. P < .05 was considered significant.
Results IgG subclass and IgE antibodies to microfilaria extract and 92-kDa fusion protein in serum pools. The serum pool es-
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on November 15, 2015
chronically infected jirds. Extracts of adult worms (AWE) and microfilariae were prepared as described [6]. Protein concentration was measured by the method of Lowry et al. [7] with bovine serum albumin (BSA; Sigma, St. Louis) as standard. To produce the recombinant filarial polypeptide, a 1400-base EcoRI-HindIIl restriction fragment of a cDNA encoding the 55-kDa amino terminal of the predicted filarial protein was subcloned in the pATH3 expression vector as described [4]. E. coli producing a 92-kDa fusion protein (55 kDa corresponding to the filarial cDNA and 37 kDa to bacterial trpE; amino acids 1-479 of the 548 amino acid parasite protein) or control bacteria producing trpE were suspended in extraction buffer containing 6 M urea and 1% SDS in 10 mM TRIS-HCI, pH 7.0. Control trpE and fusion protein were isolated by passage of bacterial Iysates over a Sepharose CL-6B column (Pharmacia Fine Chemicals, Piscataway, NJ). Human study populations. Control sera from residents of non endemic areas were obtained from 22 individuals living in North America (age range, 20-45 years) and 5 adult residents of Cairo. Sera from W. bancrofti-infected or chronically exposed persons were obtained from residents oftwo endemic areas: East Sepik Province, Papua New Guinea, and Qalibaya in the Nile Delta of Egypt. Separate serum pools from the Papua New Guinean and Egyptian populations were established by adding equal volumes of serum from 100 and 40 persons, respectively. An additional 45 East Sepik residents not included as donors to the Papua New Guinea serum pool were studied in more detail. These included sera from 20 children « I0 years old) and 25 adults (> 15 years old). The subjects were grouped into those having high intensities of microfilaremia (> 1000/mL of blood by Nuclepore [Pleasanton, CAl filtration ofa 2-mL blood sampie) and those with no or low concentrations of circulating parasites (,,;;;2/mL of blood). The level of circulating Gib 13 (phosphocholine) epitope, an additional index ofparasite burden, was determined by immunoradiometric assay in samples from adult donors [8]. No subject had lymphatic disease attributable to W. bancrofti infection. Antibody measurements. IgG subclass antibodies to microfilaria extract, AWE, and the 92-kDa fusion protein were measured by ELISA. Flat-bottom 96-well microtitration plates (3912; Falcon Oxnard, CA) were coated with I J.Lg ofmicrofilaria extract or AWE or 0.5 J.Lg of 92-kDa fusion protein per well. To coat the plates, proteins suspended in 100 J.LL of0.06 M sodium carbonate buffer, pH 9.6, were incubated in wells for 18 h at 37°C. Triplicate 80-J.LL aliquots of pooled sera were diluted serially (1:20-1:5120) in PBS containing 0.1%Tween 20 (PBST) and added to the antigen-coated wells. After incubation for I
lID 1992; 166 (December)
lID 1992;166 (December)
ConciseCommunications
IgG2
IgGI
o o
.400
.200
.200
.100
tions of I:20, I:40, and 1:80 for the serum pool, the ODs generated on microtiter plates sensitized with 10 ILg of 92kDa fusion protein/well were, respectively, 0.165, 0.096, and 0.095. Values for an equivalent amount of control trpE were 0.199,0.175, and 0.155. Sera (1:10 dilution) from 3 subjects with tropical pulmonary eosinophilia generated mean ODs of 0.208, 0.225, and 0.236 in the ELISA for the 92-kDa fusion protein. The respective values for IgE antibody to trpE were not different (ODs of 0.222, 0.257, and 0.239), indicating a lack of IgE specific for the filaria-encoded portion of the protein. Cross-reactivity ofIgG3 antibodies to 92-kDa fusion protein and B. malayi lysates. IgG3 antibodies to the fusion protein were partly depleted by preabsorption of serum with AWE. When sera from 3 Papua New Guinea residents (2 microfilaremic, I amicrofilarernic) were preabsorbed with 10 ILg of AWE, the ELISA value generated in wells coated with the 92-kDa fusion protein decreased 48%-52%, whereas preabsorption with BSA had no effect (ODs decreased from 0.675 to 0.327, 1.134 to 0.584, and 0.236 to 0.112 after absorption with AWE; no change was observed after absorption with BSA). IgG3 antibodies in subjects with high versus low intensities of microfilaremia. Because experiments using the serum pool indicated that antibody to the fusion protein was primarily IgG3, antibodies of this IgG subclass were compared in sera from Papua New Guinea adults and children with high (> 1000/mL blood) versus low levels of microfilaremia (0-4/mL). The mean values for children with high versus low intensities of microfilaremia were similar (table I). With respect to adults, 3 of 14 with relatively low intensities of microfilaremia and Gib 13 antigen indices did not have detectable IgG3 antibodies to the recombinant protein. The mean value for this group, however, was not significantly different from that of adults with high intensities of microfilaremia and Gib 13 epitope levels (54 ± 9 vs. 64 ± 8 units/ mL, respectively).
0 .300
.400
l:gG4
.200
o
o
160
2560
160
2560
SERUM DILUTION-I Figure 1. IgG subclass antibodies to 92-kDa fusion protein in serum pool established from 100 residents of Papua New Guinea. Sera were preabsorbed with trpE protein, diluted from 1:201:5120, and ELISA was done. Shaded area is mean + 2 SD of value of 22 North American control sera.
Discussion These results indicate that a recombinant protein corresponding to a 62-kDa filarial antigen shown previously to induce resistance to microfilariae in mice [4] is antigenic in human Bancroftian filariasis. Antibody reactivity to the recombinant protein was demonstrable in sera from residents from two widely separated endemic areas, Egypt and Papua New Guinea. The corresponding native antigen appears to be highly immunogenic since antibody is detectable in children and adults. The reasons for the nearly universal recognition of a filarial molecule that is unlikely to be disposed of on the surface of the filarial parasite are unclear (aspartyl tRNA synthetases are involved in synthesis of nascent polypeptides on ribosomes; unpublished data showed that MAbs to the fusion protein bind to internal and not surface compart-
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on November 15, 2015
tablished from 100 Papua New Guinea residents of the area endemic for W. bancrofli infection contained IgG I, IgG2, IgG3, and IgG4 antibodies to microfilaria extract at dilutions of I: 1280-1 :2560. IgG2 was detectable at serum dilutions of ,,;;; I:320. The pattern ofIgG subclass reactivities to the 92kDa fusion protein was more restricted (figure I). IgG2 antibodies were not detectable. Low levels of IgG I and IgG4 antibodies were present (measurable at serum dilutions of 1:20-1 :40). In contrast, IgG3 antibodies to the 92-kDa fusion protein were present at serum dilutions up to I: 1280. Sera of Egyptian W. bancrofli-infected subjects also demonstrated a restricted pattern of IgG subclass to the fusion protein such that IgG I and IgG3 antibodies were observed at serum dilutions of I:80-1: 160, whereas fusion protein-reactive IgG2 or IgG4 was detected at dilutions of'< I:40. Sera of the 5 Egyptian control subjects did not contain anti-fusion protein antibodies of any IgG subclass. As expected, Egyptian residents of'Qalibaya had IgG antibodies to microfilariae of all subclasses (dilutions of 1:320-1:640 for IgGI, -2, -3, and -4). IgE antibodies to the 92-kDa fusion protein were not detectable in any sera examined. The serum pool from the 100 Papua New Guineans showed no reactivity by ELISA when I or 10 ILg of the 92-kDa fusion protein was coated to wells. Pooled sera of W. bancrofli-infected Indian subjects with high levels ofIgE-specific antibody to AWE and sera from 3 Indians with tropical pulmonary eosinophilia were also examined for IgE reactivity to the 92-kDa fusion protein. At dilu-
1455
1ID 1992; 166 (December)
Concise Communications
1456
Table 1. IgG3 antibody levels to 92-kDa fusion protein in asymptomatic children and adults by intensity of microfilaremia.
Group, microfilaremia intensity (n) Children Low (10) High (10) Adults Low (14) High (II)
Age (years)
Level of parasitemia'
antigen!
level!
(parasites/rnl.)
(antigen index)
(units/mL)
7 (5-10) 8 (3-10)
I (0-2) >1000 (all>1000)
Not done Notdone
73 ± 9 (47-105) 73 ± 8 (18-98)
9.0 ± 1.31 17.5±1.l
54 ± 9 (10-115) 64 ± 8 (23-102)
36(15-55) 36(17-57)
Gib 13
I (0-4) 1995 (1100-7741)
Ig03 antibody
ments of adult worms and microfilariae). Several "non-surface" proteins of filarial parasites (e.g., paramyosin) [10] and other helminths (e.g., Schistosoma mansoni triosephosphate isomerase and glutathoine-S-transferase) have been reported to be highly immunogenic in naturally infected hosts. A remarkable finding of the present study was the relatively restricted recognition of the cloned antigen by IgG3 antibodies. Filaria specific IgG3 antibodies are preferentially increased over other IgG subclasses in persons with a variety of manifestations of onchocerciasis. Subjects with sowdah, a variant of Onchocerca volvulus infection in which microfilaridermia is associated with severe dermatologic pathology, have been reported to have marked elevations of filaria-specific IgG 3 [II]. In addition, Boyer et al. [12] demonstrated that the level ofIgG3 antibodies to parasite antigens of -20 kDa correlates with asymptomatic amicrofilardermic status of Guatemalans residing in endemic areas. A major strategy to identify filarial antigens involved in acquired resistance in humans has been to use antibodies in sera from "putatively immune" adult residents of endemic areas (asymptomatic amicrofilaremic persons) to identify unique or preferentially recognized molecules contained in parasite Iysates or cDNA expression libraries [2]. For example, recent work using immunoblots to probe Iysates of B. malayi-infective larvallysates identified a 43-kDa molecule that is preferentially recognized by IgG antibodies ofamicrofilaremic W. bancrofli-exposed subjects [13]. Sera from adult residents of an endemic area of Papua New Guinea were used in an analogous fashion in the current study. Because results of prospective seroepidemiologic surveys suggested that resistance to filarial parasites in humans does not develop until -20 years of age (intensity ofmicrofilaremia and serum phosphocholine epitope levels increase up to this age and level off thereafter) [14], we evaluated the levels of IgG3 antibody to the recombinant antigen in children
with high or low intensities of microfilaremia. No significant differences in IgG3 antibody levels to the recombinant protein were noted among any of the groups. There are several possible interpretations of these results. First, although the cloned protein is protective in mice [4] and antigenic in humans, the native antigen may not be involved in acquisition ofresistance in naturally infected persons. Second, our definition ofresistance, which is based primarily on microfilaremic and clinical status, may be imprecise and potentially misleading. Prospective evaluation of immune reactivity, which monitors changes in intensities of parasitemia and circulating filarial antigen levels over a period of several years, may be more informative in this regard [14]. Third, antibody reactivity to the recombinant antigen may not be a surrogate of protective immunity against infective larvae or microfilaremia in humans. With respect to this possibility, it will be important to evaluate the capacity of this (and other) recombinant antigens to drive production of cytokines such as interleukin-4 and interferon-v by T cells of persons who differ in apparent parasite burdens. These cytokines reciprocally regulate polyclonal and filaria lysate-driven production of IgE and IgG subclasses and, on the basis of results of murine studies, are likely to be more direct and sensitive measures of acquired resistance to helminthic parasites than serum antibody levels [15].
Acknowledgments
We thank Robert Poindexter and Eric Ottesen for performing the IgE studies and Pat Amato for expert secretarial assistance.
References
I. Sasa M.Human filariasis: a global survey ofepidemiology and control. Baltimore: University Park Press, 1976.
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on November 15, 2015
NOTE. Dataare mean (range). , Geometric mean intensity oflevel ofmicrofilaremia + I SOwith parasitemia measured byNuclepore filtration of l-rnl, blood sample. t Gib 13 phosphocholine epitope level measured bytwo-site immunometric assay as described [8]. Mean + 2 SOantigen index forcontrols from nonendemic areas is 2.0. t Values are mean ± I SEofgroup. Mean ± 2 SE for controls from nonendemic areas is
Concise Communications
nism for syphilis, the systemic mycoses and other infectious diseases. J Theor Bioi 1972;36:617-25. 9. Fitzgerald TJ, Tomai MA, Trachte GJ, Tice T. Prostaglandins in experimental syphilis:treponemes stimulate adherent cells to secrete prostaglandin E2 and indomethacin upregulates immune functions. Infect Immun 1991;59:143-9. 10. van der Sluis JJ. Kant M. Onvlee PC, Stolz E. The inaccessibility ofthe outer membrane of adherent Treponema pallidum (Nichols strain) to anti-treponemal antibodies, a possible role of serum proteins. Genitourin Med 1990;66: 165-70.
1453
II. Radolf JD, Norgard MY, Schulz WW. Outer membrane ultrastructure explains the limited antigenicity of virulent Treponema pallidum. Proc Nat! Acad Sci USA 1989;86:2051-5. 12. Walker EM, Zampighi GA, Blanco DR. Miller IN, Lovett MA. Freezefracture analysis demonstrates rare protein in the outer membrane of Treponema pallidum subsp. pallidum. J Bacteriol1989; 171:5585-8. 13. Turner TB, Hollander DH. Biology of the treponematoses. Geneva: World Health Organization. 1957. 14. Alderete JF. Baseman JB. Surface-associated host proteins on virulent Treponema pallidum. Infect Immun 1979;26: 1048-56.
James w. Kazura, Fred Eo Hazlett, Jr., Eric Pearlman, Karen Day, Adel EI-Zeiny, Timothy W. Nilsen, and Michael P. Alpers
Departments of Medicine (Division of Geographic Medicine) and Molecular Biology and Microbiology, Case Wesern Reserve University. Cleveland. Ohio; Department of Pure and Applied Biology. Imperial College, London, United Kingdom; Al-Azhar University, Cairo. Egypt; Papua New Guinea Institute ofMedical Research. Goroka
A 92-kDa fusion protein that encodes amino acids 1-479 of a 62-kDa Brugia malayi antigen induces resistance to microfilariae in mice. The antigenicity of this recombinant protein was explored in asymptomatic residents of Wuchereria bancrofti-endemic areas of Papua New Guinea and Egypt. Unlike sera from individuals in nonendemic areas, sera from residents of endemic areas contained IgG3 antibodies (up to dilution 1:1280) reactive with the fusion protein. There was little or no recombinant antigen-specific IgGl, IgG2, IgG4, or IgE. The mean levelof IgG3 antibodies to the amino acid 1-479 construct in sera of putatively "immune" adult Papua New Guineans and children in whom microfilaremia was 1000 parasites/ml., These data indicate that the filarial antigen corresponding to the recombinant protein is highly immunogenic in naturally infected children and adults and that the isotype and magnitude of antibody reactivity with it do not correlate with the microfilaremic status of asymptomatic persons. Infection by the lymphatic-dwelling filarial parasites Wuchereria bancrofli and Brugia malayi is a major public health problem in many areas of the tropics [1]. Identification of specific parasite antigens that induce the immunopathologic consequences of infection, for example, elephantiasis, acute lymphangitis, tropical pulmonary eosinophilia, and partial resistance to various stages of the helminth, is necessary for
Received 22 April 1992; revised 7 July 1992. Presented in part: meeting of the American Society of Tropical Medicine and Hygiene, Honolulu. December 1989. The studies were approved by the Human Studies Committee. University Hospitals of Cleveland, and health authorities of the government of Papua New Guinea. Informed consent for donation of blood was obtained from all individuals and parents of children. Grant support: World Health Organization/Tropical Disease Research and National Institutes of Health (AI-15351 and contract 73262). Reprints or correspondence: Dr. James W. Kazura, Case Western Reserve University School of Medicine, Division of Geographic Medicine. 2109 Adelbert Rd., Rm. W137. Cleveland, OH 44106-4983. The Journal of Infectious Diseases 1992;166:1453-7 © 1992 by The University of Chicago. All rightsreserved. 0022-1899/92/6606-0042$01.00
development of safe and effective immunoprophylactic measures [2]. Isolation and structural characterization of such antigens have been hampered by the biochemical complexity of these multicellular organisms and the fact that large numbers of each filarial stage cannot be cultivated in vitro. Advances in this area have become possible since the cloning of several filarial proteins. We previously described a cDNA clone selected from a B. malayi Agt11 expression library that has significant nucleotide and predicted amino acid sequence homology to aspartyl tRNA synthetase of Saccharomyces cerevisiae [3]. The cDNA encodes a 62-kDa antigen that is expressed by microfilariae and adult worms. Sensitization of mice to a 92-kDa fusion protein that includes amino acids 1-479 of the predicted 548 amino acid polypeptide induces partial resistance to subsequent challenge with B. malayi microfilariae [4]. The utility of this cloned protein for study of the human immune response has not been established. The objectives of the current study were therefore to determine whether the recombinant filarial polypeptide is antigenic in humans infected with W. bancrofli and, if so, to examine whether there
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on November 15, 2015
Antigenicity of a Protective Recombinant Filarial Protein in Human Bancroftian Filariasis
1454
Concise Communications
are differences in antibody levels to it among residents in endemic areas harboring relatively low versus high microfilarialloads. Because IgG3, IgG4, and IgE antibody levels to extracts of B. malayi adult worms have been shown to correlate with the microfilaremic status of residents of W. bancrofti-endemic areas [5], IgG subclass and IgE antibody reactivities to the recombinant protein were evaluated.
Methods Preparation of B. malayi extracts and recombinant antigens. B. malayi were obtained by lavage of the peritoneal cavities of
h at 37°C, sera were removed, wells washed with PBS-T, and 90 J.LL ofa 1:500 dilution of peroxidase-conjugated murine monoclonal antibody (MAb) to human IgG I (clone HP 6070; Zymed, South San Francisco), IgG2 (HP 6014; Binding Site, San Diego), IgG3 (HP 6047; Zymed), or IgG4 (HP 6025; Zymed) was added. These clones were previously shown to produce MAbs specific for each IgG subclass [9]. After incubation for 18 h at 4°C, MAbs were removed, wells were washed six times with PBS-T, and o-phenylenediamine substrate (Sigma) was added. OD at 492 nm was read at 15 min on an automated ELISA reader. In studies involving the 92-kDa fusion protein, individual serum samples and the pools were preabsorbed with excess trpE. Results of experiments examining antibody levels to the fusion protein in pooled sera are presented as OD. Results ofstudies comparing Ig03 antibody levels in sera from individual Papua New Guinea subjects are expressed as units of antibody per milliliter, which is defined as the ratio of the OD generated by the sample relative to an equivalent dilution of the standard pool from 100 Papua New Guineans. One hundred units ofantibody reactive to the 92-kDa fusion protein indicates that the value for an individual serum sample is equivalent to that of the standard pool at the same dilution. This method of expressing antibody reactivity enables comparison of individual values using a standard curve generated with every assay [10]. IgE antibody to the fusion protein and AWE was measured by ELISA in the laboratory of E. Ottesen (National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD) as described [5]. To remove blocking antibodies, sera were preabsorbed with protein A coupled to Sepharose (Pharmacia). After overnight incubation of I: 10-1 :80 dilutions of serum in fusion protein-coated wells, MAb to human IgE (MAb 7.12, provided by A. Saxon, UCLA School of Medicine, Los Angeles) was added for 2 h at 37°C. Plates were washed and incubated for another 2 h at 37°C with biotinylated goat antimouse IgE (Jackson Immunoresearch, West Grove, PA). This antibody does not react with human IgE. Streptavidin-conjugated alkaline phosphatase (Jackson Immunoresearch) was added and incubation was continued at room temperature for I h. Color was developed using fJ"'nitrophenyl phosphate as substrate. Results are expressed as OD at 405 nm. Cross-reactivity of antibodies to 92-kDa fusion protein and A WE. To confirm that antibody directed against the 92-kDa fusion protein recognized the corresponding native filarial antigen, aliquots of serum (1:400 dilution) from 3 Papua New Guinea subjects were preincubated for 18 hat 4°C in microtiter wells coated with either 10 J.Lg of AWE or BSA (control protein). Antibody to the 92-kDa fusion protein was then measured by ELISA as described above. Statistics. The Wilcoxon rank-sum test was used to assess the significance of difference for antibody levels between groups of Papua New Guinea study subjects with high versus low intensities of microfilaremia. Student's t test was used to compare mean values for Gib 13 phosphocholine epitope levels and relative antibody reactivities. P < .05 was considered significant.
Results IgG subclass and IgE antibodies to microfilaria extract and 92-kDa fusion protein in serum pools. The serum pool es-
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on November 15, 2015
chronically infected jirds. Extracts of adult worms (AWE) and microfilariae were prepared as described [6]. Protein concentration was measured by the method of Lowry et al. [7] with bovine serum albumin (BSA; Sigma, St. Louis) as standard. To produce the recombinant filarial polypeptide, a 1400-base EcoRI-HindIIl restriction fragment of a cDNA encoding the 55-kDa amino terminal of the predicted filarial protein was subcloned in the pATH3 expression vector as described [4]. E. coli producing a 92-kDa fusion protein (55 kDa corresponding to the filarial cDNA and 37 kDa to bacterial trpE; amino acids 1-479 of the 548 amino acid parasite protein) or control bacteria producing trpE were suspended in extraction buffer containing 6 M urea and 1% SDS in 10 mM TRIS-HCI, pH 7.0. Control trpE and fusion protein were isolated by passage of bacterial Iysates over a Sepharose CL-6B column (Pharmacia Fine Chemicals, Piscataway, NJ). Human study populations. Control sera from residents of non endemic areas were obtained from 22 individuals living in North America (age range, 20-45 years) and 5 adult residents of Cairo. Sera from W. bancrofti-infected or chronically exposed persons were obtained from residents oftwo endemic areas: East Sepik Province, Papua New Guinea, and Qalibaya in the Nile Delta of Egypt. Separate serum pools from the Papua New Guinean and Egyptian populations were established by adding equal volumes of serum from 100 and 40 persons, respectively. An additional 45 East Sepik residents not included as donors to the Papua New Guinea serum pool were studied in more detail. These included sera from 20 children « I0 years old) and 25 adults (> 15 years old). The subjects were grouped into those having high intensities of microfilaremia (> 1000/mL of blood by Nuclepore [Pleasanton, CAl filtration ofa 2-mL blood sampie) and those with no or low concentrations of circulating parasites (,,;;;2/mL of blood). The level of circulating Gib 13 (phosphocholine) epitope, an additional index ofparasite burden, was determined by immunoradiometric assay in samples from adult donors [8]. No subject had lymphatic disease attributable to W. bancrofti infection. Antibody measurements. IgG subclass antibodies to microfilaria extract, AWE, and the 92-kDa fusion protein were measured by ELISA. Flat-bottom 96-well microtitration plates (3912; Falcon Oxnard, CA) were coated with I J.Lg ofmicrofilaria extract or AWE or 0.5 J.Lg of 92-kDa fusion protein per well. To coat the plates, proteins suspended in 100 J.LL of0.06 M sodium carbonate buffer, pH 9.6, were incubated in wells for 18 h at 37°C. Triplicate 80-J.LL aliquots of pooled sera were diluted serially (1:20-1:5120) in PBS containing 0.1%Tween 20 (PBST) and added to the antigen-coated wells. After incubation for I
lID 1992; 166 (December)
lID 1992;166 (December)
ConciseCommunications
IgG2
IgGI
o o
.400
.200
.200
.100
tions of I:20, I:40, and 1:80 for the serum pool, the ODs generated on microtiter plates sensitized with 10 ILg of 92kDa fusion protein/well were, respectively, 0.165, 0.096, and 0.095. Values for an equivalent amount of control trpE were 0.199,0.175, and 0.155. Sera (1:10 dilution) from 3 subjects with tropical pulmonary eosinophilia generated mean ODs of 0.208, 0.225, and 0.236 in the ELISA for the 92-kDa fusion protein. The respective values for IgE antibody to trpE were not different (ODs of 0.222, 0.257, and 0.239), indicating a lack of IgE specific for the filaria-encoded portion of the protein. Cross-reactivity ofIgG3 antibodies to 92-kDa fusion protein and B. malayi lysates. IgG3 antibodies to the fusion protein were partly depleted by preabsorption of serum with AWE. When sera from 3 Papua New Guinea residents (2 microfilaremic, I amicrofilarernic) were preabsorbed with 10 ILg of AWE, the ELISA value generated in wells coated with the 92-kDa fusion protein decreased 48%-52%, whereas preabsorption with BSA had no effect (ODs decreased from 0.675 to 0.327, 1.134 to 0.584, and 0.236 to 0.112 after absorption with AWE; no change was observed after absorption with BSA). IgG3 antibodies in subjects with high versus low intensities of microfilaremia. Because experiments using the serum pool indicated that antibody to the fusion protein was primarily IgG3, antibodies of this IgG subclass were compared in sera from Papua New Guinea adults and children with high (> 1000/mL blood) versus low levels of microfilaremia (0-4/mL). The mean values for children with high versus low intensities of microfilaremia were similar (table I). With respect to adults, 3 of 14 with relatively low intensities of microfilaremia and Gib 13 antigen indices did not have detectable IgG3 antibodies to the recombinant protein. The mean value for this group, however, was not significantly different from that of adults with high intensities of microfilaremia and Gib 13 epitope levels (54 ± 9 vs. 64 ± 8 units/ mL, respectively).
0 .300
.400
l:gG4
.200
o
o
160
2560
160
2560
SERUM DILUTION-I Figure 1. IgG subclass antibodies to 92-kDa fusion protein in serum pool established from 100 residents of Papua New Guinea. Sera were preabsorbed with trpE protein, diluted from 1:201:5120, and ELISA was done. Shaded area is mean + 2 SD of value of 22 North American control sera.
Discussion These results indicate that a recombinant protein corresponding to a 62-kDa filarial antigen shown previously to induce resistance to microfilariae in mice [4] is antigenic in human Bancroftian filariasis. Antibody reactivity to the recombinant protein was demonstrable in sera from residents from two widely separated endemic areas, Egypt and Papua New Guinea. The corresponding native antigen appears to be highly immunogenic since antibody is detectable in children and adults. The reasons for the nearly universal recognition of a filarial molecule that is unlikely to be disposed of on the surface of the filarial parasite are unclear (aspartyl tRNA synthetases are involved in synthesis of nascent polypeptides on ribosomes; unpublished data showed that MAbs to the fusion protein bind to internal and not surface compart-
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on November 15, 2015
tablished from 100 Papua New Guinea residents of the area endemic for W. bancrofli infection contained IgG I, IgG2, IgG3, and IgG4 antibodies to microfilaria extract at dilutions of I: 1280-1 :2560. IgG2 was detectable at serum dilutions of ,,;;; I:320. The pattern ofIgG subclass reactivities to the 92kDa fusion protein was more restricted (figure I). IgG2 antibodies were not detectable. Low levels of IgG I and IgG4 antibodies were present (measurable at serum dilutions of 1:20-1 :40). In contrast, IgG3 antibodies to the 92-kDa fusion protein were present at serum dilutions up to I: 1280. Sera of Egyptian W. bancrofli-infected subjects also demonstrated a restricted pattern of IgG subclass to the fusion protein such that IgG I and IgG3 antibodies were observed at serum dilutions of I:80-1: 160, whereas fusion protein-reactive IgG2 or IgG4 was detected at dilutions of'< I:40. Sera of the 5 Egyptian control subjects did not contain anti-fusion protein antibodies of any IgG subclass. As expected, Egyptian residents of'Qalibaya had IgG antibodies to microfilariae of all subclasses (dilutions of 1:320-1:640 for IgGI, -2, -3, and -4). IgE antibodies to the 92-kDa fusion protein were not detectable in any sera examined. The serum pool from the 100 Papua New Guineans showed no reactivity by ELISA when I or 10 ILg of the 92-kDa fusion protein was coated to wells. Pooled sera of W. bancrofli-infected Indian subjects with high levels ofIgE-specific antibody to AWE and sera from 3 Indians with tropical pulmonary eosinophilia were also examined for IgE reactivity to the 92-kDa fusion protein. At dilu-
1455
1ID 1992; 166 (December)
Concise Communications
1456
Table 1. IgG3 antibody levels to 92-kDa fusion protein in asymptomatic children and adults by intensity of microfilaremia.
Group, microfilaremia intensity (n) Children Low (10) High (10) Adults Low (14) High (II)
Age (years)
Level of parasitemia'
antigen!
level!
(parasites/rnl.)
(antigen index)
(units/mL)
7 (5-10) 8 (3-10)
I (0-2) >1000 (all>1000)
Not done Notdone
73 ± 9 (47-105) 73 ± 8 (18-98)
9.0 ± 1.31 17.5±1.l
54 ± 9 (10-115) 64 ± 8 (23-102)
36(15-55) 36(17-57)
Gib 13
I (0-4) 1995 (1100-7741)
Ig03 antibody
ments of adult worms and microfilariae). Several "non-surface" proteins of filarial parasites (e.g., paramyosin) [10] and other helminths (e.g., Schistosoma mansoni triosephosphate isomerase and glutathoine-S-transferase) have been reported to be highly immunogenic in naturally infected hosts. A remarkable finding of the present study was the relatively restricted recognition of the cloned antigen by IgG3 antibodies. Filaria specific IgG3 antibodies are preferentially increased over other IgG subclasses in persons with a variety of manifestations of onchocerciasis. Subjects with sowdah, a variant of Onchocerca volvulus infection in which microfilaridermia is associated with severe dermatologic pathology, have been reported to have marked elevations of filaria-specific IgG 3 [II]. In addition, Boyer et al. [12] demonstrated that the level ofIgG3 antibodies to parasite antigens of -20 kDa correlates with asymptomatic amicrofilardermic status of Guatemalans residing in endemic areas. A major strategy to identify filarial antigens involved in acquired resistance in humans has been to use antibodies in sera from "putatively immune" adult residents of endemic areas (asymptomatic amicrofilaremic persons) to identify unique or preferentially recognized molecules contained in parasite Iysates or cDNA expression libraries [2]. For example, recent work using immunoblots to probe Iysates of B. malayi-infective larvallysates identified a 43-kDa molecule that is preferentially recognized by IgG antibodies ofamicrofilaremic W. bancrofli-exposed subjects [13]. Sera from adult residents of an endemic area of Papua New Guinea were used in an analogous fashion in the current study. Because results of prospective seroepidemiologic surveys suggested that resistance to filarial parasites in humans does not develop until -20 years of age (intensity ofmicrofilaremia and serum phosphocholine epitope levels increase up to this age and level off thereafter) [14], we evaluated the levels of IgG3 antibody to the recombinant antigen in children
with high or low intensities of microfilaremia. No significant differences in IgG3 antibody levels to the recombinant protein were noted among any of the groups. There are several possible interpretations of these results. First, although the cloned protein is protective in mice [4] and antigenic in humans, the native antigen may not be involved in acquisition ofresistance in naturally infected persons. Second, our definition ofresistance, which is based primarily on microfilaremic and clinical status, may be imprecise and potentially misleading. Prospective evaluation of immune reactivity, which monitors changes in intensities of parasitemia and circulating filarial antigen levels over a period of several years, may be more informative in this regard [14]. Third, antibody reactivity to the recombinant antigen may not be a surrogate of protective immunity against infective larvae or microfilaremia in humans. With respect to this possibility, it will be important to evaluate the capacity of this (and other) recombinant antigens to drive production of cytokines such as interleukin-4 and interferon-v by T cells of persons who differ in apparent parasite burdens. These cytokines reciprocally regulate polyclonal and filaria lysate-driven production of IgE and IgG subclasses and, on the basis of results of murine studies, are likely to be more direct and sensitive measures of acquired resistance to helminthic parasites than serum antibody levels [15].
Acknowledgments
We thank Robert Poindexter and Eric Ottesen for performing the IgE studies and Pat Amato for expert secretarial assistance.
References
I. Sasa M.Human filariasis: a global survey ofepidemiology and control. Baltimore: University Park Press, 1976.
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on November 15, 2015
NOTE. Dataare mean (range). , Geometric mean intensity oflevel ofmicrofilaremia + I SOwith parasitemia measured byNuclepore filtration of l-rnl, blood sample. t Gib 13 phosphocholine epitope level measured bytwo-site immunometric assay as described [8]. Mean + 2 SOantigen index forcontrols from nonendemic areas is 2.0. t Values are mean ± I SEofgroup. Mean ± 2 SE for controls from nonendemic areas is
