sensitive and specific serodiagnosis of onchocerciasis with ...

Am. J. Trop. Med. Hyg., 66(5), 2002, pp. 566–571 Copyright © 2002 by The American Society of Tropical Medicine and Hygiene

SENSITIVE AND SPECIFIC SERODIAGNOSIS OF ONCHOCERCIASIS WITH RECOMBINANT HYBRID PROTEINS PIUS N. NDE, THOMAS POGONKA, JANETTE E. BRADLEY, VINCENT P. K. TITANJI, AND RICHARD LUCIUS Department of Molecular Parasitology, Humboldt University, Berlin, Germany; Department of Biological Sciences, Salford University, Salford, United Kingdom; Biotechnology Unit, Department of Life Sciences, University of Buea, Buea, Cameroon

Abstract. Onchocerciasis remains a major health hazard in many tropical countries. However, the existing tools for diagnosis of the disease have limitations, particularly regarding the detection of low level or early infections. To design an optimized reagent, we exploited the high antibody reactivity of patient sera against the Onchocerca volvulus proteins Ov20 and Ov33, which have been described as highly sensitive and specific immunodiagnostic reagents for producing hybrid proteins. The construct OvH2 was composed of Ov20 fused to Ov33, while OvH3 consisted of the C-terminus of Ov20 linked to Ov33. When these constructs were tested with sera from patients with onchocerciasis and control sera, OvH2 showed a sensitivity of 98.5% and a specificity of 97.7% and OvH3 showed a sensitivity of 98.5% and a specificity of 95.35%. All non-responders were from Ecuador. These results surpass those of existing single recombinant antigens, suggesting that our hybrid proteins combined the sensitivity of the two parent proteins. Tests based on OvH2 should prove suitable for monitoring onchocerciasis control programs and individual diagnosis. create a new generation of tests, which combine high specificity and sensitivity with easy handling. As a prerequisite for such tests, the present investigation re-evaluated two O. volvulus antigens, Ov33 and Ov20, which were found to be recognized by greater than 90% of the onchocerciasis sera and had a very high specificity when produced as fusion proteins.13,19,20 Previous epitope mapping studies showed that frequently recognized continuous B cell epitopes were present in several parts of Ov33,21 whereas the N-terminus of Ov20 did not appear to contribute much to the overall B cell reactivity (Bradley JE, unpublished data). Therefore, slightly different elements of Ov20 in combination with the same region of Ov33 were used in the production of two hybrid proteins, OvH2 and OvH3, and their serodiagnostic potentials were evaluated with onchocerciasis sera from different geographic regions. Our aim was to create a reagent that would emulate the properties of the diagnostic cocktail,13 while eliminating its shortcomings.

INTRODUCTION Onchocerciasis, which is caused by the filarial nematode Onchocerca volvulus, affects approximately 18 million individuals worldwide.1 The disease is usually diagnosed by the detection of microfilariae in bloodless skin biopsies, an approach that is considered to be insensitive for the detection of prepatent and low level infections.2,3 Comparable results can also be obtained by a skin reaction evoked by topical application of diethylcarbamazine, a drug that eliminates microfilariae.4 Major efforts have been undertaken by the Onchocerciasis Control Program of the World Health Organization (WHO) to interrupt the transmission of O. volvulus through control of the blackfly intermediate hosts and by transient elimination of the microfilariae with ivermectin.5 These measures have led to the elimination of the parasite or reduction of its prevalence in wide areas of West Africa and Latin America.6 The remaining low-level infections or newly upcoming infections in controlled areas are difficult to diagnose, yet such a diagnosis is important as a basis for decisions on further interventions. In such situations, tests have to be very specific to keep the risk of detecting false-positive cases in large populations at a minimum, while the sensitivity is also important. Although sensitive polymerase chain reaction (PCR) assays have been developed, they also depend strictly on the presence of parasite DNA in the skin,7 and as such have disadvantages similar to parasitologic diagnosis. Therefore, antibody detection assays have been developed using recombinant O. volbulus antigens.8–14 Some of these antigens were tested in a multicenter trial organized by the Tropical Diseases Research Program of the WHO that identified highly specific proteins,15 and high sensitivity of these tests was achieved by combining three proteins in a diagnostic cocktail.13,16,17 However, these antigens were difficult to produce and the fact that they were expressed as fusions with a carrier protein made it necessary to run parallel tests to rule out antibody reactions with the carrier. These limitations have prevented the O. volvulus antigen cocktail from becoming a widely used diagnostic tool. Meanwhile, formats of antibody tests for onchocerciasis have been produced that allow rapid serodiagnosis directly on the spot of sampling.18 Such developments make it attractive to refine recombinant antigens to

MATERIALS AND METHODS Sera. A total of 132 sera from onchocerciasis patients with proven skin microfilariae were obtained from five countries in different geographic regions of Africa and Latin America (Table 1). In addition, we tested 28 sera from persons in Ecuador who lived in the same areas as the onchocerciasis patients and were exposed to O. volvulus transmission, but did not harbor skin microfilariae. These persons have been described as putatively immune individuals.22 Furthermore, we studied 33 sera from persons living in endemic areas in Togo, who were exposed to O. volvulus transmission, but did not harbor microfilariae in their skin (Soboslay P, unpublished data). Sera from Cameroon were obtained from characterized patients living near the Sanaga River in the Central Province of Cameroon, a forest region endemic for onchocerciasis. For the field study in Cameroon, informed consent was obtained from all human adult participants and from parents or legal guardians of minors. The study was approved by the University of Yaounde I Ethical Committee. Onchocerciasis patient sera from Ivory Coast, Togo, Ecuador, and Guatemala were obtained from the WHO Filariasis Serum Bank and the Edna McConnell Clark Foundation Serum Bank, respectively. Sera from individuals infected with Wuchereria

566

567

SERODIAGNOSIS OF ONCHOCERCIASIS

TABLE 1 Characteristics of onchocerciasis serum donors Microfilariae positive

Country

Cameroon

35

Ivory Coast*

27

Togo†

19

Ecuador†

21

Guatemala*

30

Total

132

Mean ± SD microfilariae/mg of skin (range)

46.2 ± 78.5 (0.78–410.1) 65.1 ± 57.7 (2.0–242) 51 ± 62.1 (0.4–259.4) 38.4 ± 31.9 (3.5–95.5) 41.4 ± 36.0 (08.5–130)

Mean ± SD age, years (range)

43.6 ± 15.5 (13–78) 25.2 ± 11.6 (5–50) 38.6 ± 10.4 (14–60) 41.8 ± 17.2 (18–72) 32.6 ± 10.3 (20–58)

Microfilariae negative

33 28

Mean ± SD age, years (range)

32.7 ± 14.4 (12–75) 33.1 ± 9.6

61

* Sera obtained from the World Health Organization Filariasis Serum Bank. † Sera obtained from Edna McConnell Clark Foundation Filariasis Serum Bank.

bancrofti from India and Sri Lanka were obtained from the WHO Filariasis Serum Bank (Table 2). Sera from individuals infected with Brugia malayi were kindly provided by Dr. M. Yazdanbaksh and Dr. E. Sartono (Department of Parasitology, University of Leiden, Leiden, The Netherlands). The brugian sera were collected from individuals who were amicrofilaremic or microfilaremic, and also from patients with elephantiasis. European control sera were obtained from adult German students who had never travelled to the tropics. Structure of hybrid proteins OvH2 and OvH3. The two hybrid proteins, designated OvH2 and OvH3, were composed of elements of the immunodominant O. volvulus proteins Ov20 (GenBank娂 Accession No. L27686) and Ov33 (GenBank娂 Accession No. X13313). Computer-based analysis of the amino acid sequences of each designed hybrid protein was carried out to evaluate its stability and solubility for efficient expression in Escherichia coli and purification. OvH2 was designed and constructed from amino acids 17–178 of Ov20 linked by a spacer (GPGKK) to amino acids 22–239 of Ov33. OvH3 was composed of amino acids 74–178 of Ov20 joined to amino acids 22–239 of Ov33 by a similar spacer. The spacer was introduced through the 3⬘ primer used for amplification of Ov20. Polymerase chain reaction. The nucleotide sequences of Ov33 in plasmid pGEX 2a and Ov20 in plasmid pUC 18 were amplified with primers containing specific DNA restriction endonuclease sequences at their 5⬘ and 3⬘ ends to facilitate subsequent ligation and cloning reactions. Primers for the amplification of the nucleotide sequences of interest were designed using the Oligo Program.23 For amplification of the nucleotide sequence encoding amino acids 22–239 of Ov33, the 5⬘ primer had the sequence 5⬘-GAGCTCGGTGTAGTAAAAAGGTACAATAAACGT-3⬘ and the 3⬘ primer had

the sequence 5⬘-GTCGACCATAGATTGCGACGCAGAAATG-3⬘. For amplification of the nucleotide sequence encoding amino acids 17–178 of Ov20, the 5⬘ primer had the sequence 5⬘-GGATCCAATGTTGTTCCGTTTTCAAATGAG-3⬘ and the 3⬘ primer had the sequence 5⬘-GAGCTCTTTTTTCCCGGGACCATTCTTTTGCAGAAAACTGTTTGC-3⬘; the 5⬘ portion of this primer also codes for the amino acid sequence in the spacer region linking the proteins. The 5⬘ primer used for the amplification of the nucleotide sequence encoding amino acids 74–178 of Ov20 had the sequence 5⬘-GGATCCAAAAACAAATCGGATAAACTCTACCA-3⬘. The anti-sense primer was the same as that used for the amplification of the nucleotide sequence encoding amino acids 17–178 of Ov20. For the amplification reactions, 50 pM of each forward and reverse primers, 200 ␮M dNTP, 50 ng of plasmid DNA, one unit of Pwo polymerase enzyme (Angewandte Gentechnologie Systeme [AGS], Heidelberg, Germany), and 1× complete buffer in a total reaction volume of 50 ␮l were used. The amplification reactions were carried as follows. For amplification of the Ov33 nucleotide sequence, after an initial denaturation step of 95°C for 3 minutes, the primers were annealed at 60°C for 1 minute and nascent DNA was synthesised at 72°C for 1 minute for 30 cycles. Afterwards, there was a final synthesis step at 72°C for 10 minutes. For the amplification of the Ov20 nucleotide sequences, the cycling conditions were similar to those of Ov33 with the exception that the primers were annealed at 56°C for 1 minute. Plasmid constructions. Standard procedures used for the cloning of amplified DNA were as previously described.24 All DNA-modifying enzymes used were obtained from AGS unless otherwise stated. The PCR products were purified using a QIAquick PCR purification kit (Qiagen, Hilden, Germany)

TABLE 2 Characteristics of control serum donors* Type of serum

No. of sera

Mean ± SD age years (range)

Origin

Brugia malayi infection Wuchereria bancrofti infection European sera

21 22 12

37.5 ± 12.6 (17–70) 27.36 ± 11.8 (10–55) ND

Indonesia India and Sri Lanka Germany

Total * ND ⳱ not determined.

55

568

NDE AND OTHERS

as described by the manufacturer. The purified Ov33 amplicon was digested overnight with Sac I and Sal I and subcloned into pBluescript SK+ (Stratagene Europe, Amsterdam, The Netherlands) previously digested with the same enzymes. The purified Ov20 amplicons were digested overnight with Bam HI and Sac I and similarly subcloned. To construct the nucleotide sequence of OvH2, the expression plasmid pQE 30 was digested with Bam HI and Sal I. The previously digested Ov20 and Ov33 amplicons were ligated into the plasmid using a rapid DNA ligation kit (Boehringer Mannheim, Mannheim, Germany) as described by the manufacturer. The ligation of the Ov33 sequence and the sequence encoding amino acids 74–178 of Ov20 into a similarly digested pQE30 plasmid gave rise to the complete nucleotide sequence of OvH3. The plasmids were used to transform competent E. coli M15 already containing the repressor plasmid pREP4. Positive clones were selected against an ampicillin/kanamycin background. Expression and purification of hybrid proteins. The selected positive E. coli clones were induced with 1.0 mM isopropyl-␤-D-thioglactopyranoside (IPTG) for 3 hr at 25°C and the OvH2 and OvH3 proteins were affinity purified under denaturing conditions using the Ni-NTA matrix (Qiagen) as described by the manufacturer with some modifications, as follows. For the purification of OvH2, the pellet of the induced bacteria was lysed in a buffer containing 6 M GuHCl, 0.1 M NaH2PO4, 0.01 M Tris-Cl, pH 8.0, and 20 mM ␤-mercaptoethanol on ice. The bacterial lysate was centrifuged at 27,000 × g for 20 minutes at 4°C and the supernatant was mixed with Ni-NTA agarose previously equilibrated with lysis buffer without ␤-mercaptoethanol. For sufficient binding of the expressed 6-histidine tagged protein to the matrix, the mixture was kept shaking for 3 hr at 4°C. The slurry was loaded onto a column and washed with the lysis buffer. Subsequent washes were done with a buffer containing 8 M urea, 0.1 M NaH2PO4, 500 mM NaCl, 0.1% Triton X-100, 0.01 M Tris-Cl, pH 8.0, followed by a buffer containing 8 M urea, 0.1 M NaH2PO4, 500 mM NaCl, 0.01 M Tris-Cl, pH 8.0 and then with the same buffer without 500 mM NaCl, pH 6.3. The bound proteins were eluted using the last washing buffer at pH 4.0. The purification of OvH3 was similar, except that the third washing step was with a buffer containing 8 M urea, 1 M NaCl, 20% glycerol, 0.1 M NaH2PO4, 0.01 M Tris-Cl, pH 8.0. The purified proteins were analyzed by electrophoresis on a 13% polyacrylamide gel. The pure fractions of each hybrid protein were pooled and dialyzed against a buffer containing 1 M urea, 0.1 M NaH2PO4, 0.01 M Tris-Cl, pH 7.4 and then exhaustively dialyzed against phosphate-buffered saline (PBS), pH 7.4. Enzyme-linked immunosorbent assay (ELISA). A standard ELISA technique was used to investigate the reactivity of sera with the purified hybrid proteins. The optimal antigen concentration required for the coating of the Maxisorp ELISA plates (Nunc, Wiesbaden, Germany) was determined by a checkerboard titration using several well-characterized onchocerciasis sera. The ELISA plates were coated with 50 ␮l/well of antigen (OvH2 or OvH3, 3 ␮g/ml or OvAg, 2.5 ␮g/ml) in 0.2 M carbonate buffer, pH 9.6, and the plates were incubated overnight at 4°C. After the plates were washed with PBS, pH 7.4, they were blocked with 3% bovine serum albumin in PBS (200 ␮l/well) for 2 hr at room temperature. Serum samples diluted in blocking buffer at 1:1,500 for the determination of IgG antibody responses were added in du-

plicate to the plates (50 ␮l/well) and incubated overnight at 4°C. Thereafter, the plates were washed three times with PBS containing 0.025% Tween-20. Detection of bound human IgG was performed with a monoclonal antibody against human IgG (Immunotech, Hamburg, Germany) diluted 1:2,000 in blocking buffer, which was added to the plates and incubated at 37°C for 3 hr. The plates were washed and alkaline phosphatase-conjugated anti-mouse antibody (Dianova-Jackson, Hamburg, Germany) diluted 1:3,000 in blocking buffer was added. The plates were incubated for 1 hr at 37°C and unbound antibodies were washed away. Bound antibodies were revealed by adding the substrate (5 mg of p-nitrophenyl phosphate [Sigma, Deisenhofen, Germany] dissolved in 10 ml of carbonate coating buffer containing 1 mM MgCl2). The enzyme reaction continued in the dark until the positive control sera attained an approximate absorbance of 1.1 at 405 nm. The quotient of the observed mean absorbance of the positive control sera of each plate relative to the expected absorbance was used to multiply all observed optical density values of that plate to give the absorbance units. All serum samples were analyzed in duplicate. Serologic parameters. The cut-off ELISA value for the determination of the sensitivity was calculated from the arithmetic mean plus three standard deviations of the lymphatic filariasis sera (cumulative W. bancrofti and B. malayi sera). All values below that line were considered negative. Accordingly, the sensitivity was the percentage of sera above the cut-off value. The specificity was 100% minus the percentage of lymphatic filariasis sera with absorbances above the cut-off value. The rating index (J index), the positive predictive value (PPV), and the negative predictive value (NPV) were calculated as described.25 RESULTS Expression and purification of the hybrid proteins. The two hybrid proteins OvH2 and OvH3 were expressed in E. coli after induction with IPTG. Induced bacteria produced a 6-histidine-tagged protein with a molecular mass of 44 kD (OvH2), which represented approximately 30% of the bacterial biomass, and a second protein with a molecular mass of 38 kD (OvH3), which represented approximately 30% of the total bacterial proteins. The affinity-purified hybrid proteins were soluble and stable after exhaustive stepwise dialysis against PBS. From one liter of E. coli culture, approximately 10 mg of OvH2 protein and 12 mg of OvH3 protein could be purified. Immunodiagnostic properties of OvH2 and OvH3. To establish the immunodiagnostic properties of OvH2 and OvH3, the IgG reactions against the hybrid proteins were evaluated using characterized sera as summarized in Tables 1 and 2. As a positive control, we evaluated the reaction of the sera with O. volvulus total extracts (OvAg). In four of the five serum groups tested, 100% of the onchocerciasis sera recognized the two hybrid proteins. Onchocerciasis sera from Ecuador were slightly less reactive. Less than 5% of the lymphatic filariasis sera and none of the European control sera reacted with OvH2 or OvH3. OvH2 was relatively more specific than OvH3. The ELISA values are shown in Figures 1 and 2 and the sensitivities, J-rating indices, and predictive values are summarized in Tables 3 and 4. OvH2. Of the 132 onchocerciasis sera from microfilariaepositive patients, 130 reacted positively with OvH2 (sensitiv-

569


FIGURE 1. IgG reactivity of sera from patients with onchocerciasis against OvH2. The line marked S is the cut-off line for sensitivity and specificity (mean ± 3 standard deviations of the absorbance units of the cumulative sera from patients infected with Wuchereria bancrofti and Brugia malayi). Ca ⳱ Cameroon; IC ⳱ Ivory Coast; To ⳱ Togo; Eu ⳱ Ecuador; Gu ⳱ Guatemala; W.b ⳱ W. bancrofti; B. m ⳱ B. malayi; EC ⳱ European control.

ity ⳱ 98.5%). All sera from four geographic regions (Cameroon, Ivory Coast, Togo, and Guatemala) recognized OvH2 in the IgG ELISA (sensitivity ⳱ 100%, Table 3), while 19 of 21 sera from Ecuador were positive (sensitivity ⳱ 90.5%). The two individuals from Ecuador who had no specific IgG antibody to OvH2 had very low microfilarial densities of 7 and 8 microfilariae/mg of skin biopsy, respectively, and showed a very weak IgG response against OvAg with absorbance units of 0.06 and 0.12, respectively. One of the 43 lymphatic filariasis sera (a W. bancrofti-infected serum) had a weak positive IgG reaction (Figure 1). Therefore, the specificity of the total IgG responses against OvH2 was 97.7%. The J index, which combines the sensitivity with the specificity and describes the diagnostic capability of an antigen, was 97.7% for the sera from the various countries except for the Ecuadorian sera, where it was 88.2%. The PPV, which is the probability that a positive IgG antibody response against OvH2 is a true positive test result, was 100% for all groups of sera evaluated, except the samples from Ecuador, where it was 90.5%. The NPV, which is the probability that a negative IgG antibody response to OvH2 is a true negative test result, was 97.7%. In an attempt to determine the sensitivity of OvH2 with sera from persons exposed to O. volvulus transmission, but without detectable skin microfilariae microfilariae, we tested sera from persons in Togo (n ⳱ 33) who were likely to be exposed to O. volvulus transmission, and sera of putatively immune persons from Ecuador (n ⳱ 28). In the assay, 64.3% and 63.6% of these individuals from Ecuador and Togo, respectively, were positive. Therefore, these individuals were exposed to O. volvulus infection although the parasite could not be demonstrated in their skin snips.

FIGURE 2. IgG reactivity of sera from patients with onchocerciasis against OvH3. The line marked S is the cut-off line for sensitivity and specificity (mean ± 3 standard deviations of the absorbance units of the cumulative sera from patients infected with Wuchereria bancrofti and Brugia malayi). Ca ⳱ Cameroon; IC ⳱ Ivory Coast; To ⳱ Togo; Eu ⳱ Ecuador; Gu ⳱ Guatemala; W.b ⳱ W. bancrofti; B. m ⳱ B. malayi; EC ⳱ European control.

OvH3. The overall sensitivity of the IgG antibody response of sera from microfilaria-positive patients against OvH3 was 98.5% (130 of 132 sera were positive, Table 4). While all sera form Cameroon, Ivory Coast, Togo, and Guatemala reacted positively (sensitivity ⳱ 100%), 19 of 21 sera from Ecuador were positive (sensitivity ⳱ 90.5%). The two individuals from Ecuador who displayed no specific IgG antibody response to OvH3 were the same patients who did not react with OvH2. Two lymphatic filariasis sera (both W. bancrofti-infected sera) showed weak IgG reactions, with absorbance units of 0.21 and 0.14, respectively, compared with the cut-off value of 0.139 (Figure 2). Thus, the specificity of the IgG antibody responses against OvH3 was 95.3%. Consequently, the J rating indices were 95.3% for all countries tested, except Ecuador (85.8%). Similarly, the PPV was 100% for all countries tested, except Ecuador (90.5%) and the NPV was 95.3%. Testing of the sera from individuals who were free of skin microfilariae despite living in the same areas as the onchocerciasis patients showed that 57.1% and 66.7% of the sera from TABLE 3 Sensitivities, J rating indices, and positive predictive values of the IgG antibody responses to OvH2* Country

Sensitivity (%) (neg/pos)

J Index (%)

PPV (%)

Cameroon Ivory Coast Togo Ecuador Guatemala

100 (0/35) 100 (0/27) 100 (0/19) 90.5 (2/19) 100 (0/30)

97.7 97.7 97.7 88.2 97.7

100 100 100 90.5 100

* The sensitivity cut-off value is the mean ± 3 standard deviations of the cumulative absorbance units of sera infected with Wuchereria bancrofti and Brugia malayi sera. PPV ⳱ positive predictive value; neg ⳱ negative; pos ⳱ positive.

570

NDE AND OTHERS

TABLE 4 Sensitivities, J rating indices, and positive predictive values of the IgG antibody responses to OvH3* Country

Sensitivity (%) (neg/pos)

J Index (%)

PPV (%)

Cameroon Ivory Coast Togo Ecuador Guatemala

100 (0/35) 100 (0/27) 100 (0/19) 90.5 (2/19) 100 (0/30)

95.3 95.3 95.3 85.8 95.3

100 100 100 90.5 100

* The sensitivity cut-off value is mean ± 3 standard deviations of the cumulative absorbance units of sera infected with Wuchereria bancrofti and Brugia malayi. PPV ⳱ positive predictive value; neg ⳱ negative; pos ⳱ positive.

Ecuador and Togo, respectively, reacted positively with OvH3. DISCUSSION The serodiagnosis of onchocerciasis has several shortcomings, in spite of the fact that several highly sensitive and specific immunodiagnostic antigens have been described.3 None of these antigens has a sensitivity and specificity of 100%, and only a cocktail of antigens has been reported to have the qualities required for a reliable diagnosis.13,16,17 This cocktail, consisting of three fusion proteins, is difficult to produce and requires parallel tests for controlling seroreactivity with the carrier protein.13 To improve this situation, we attempted to emulate the potential of the two well-described immunodiagnostic proteins (Ov33 and Ov20). Ov33 was reported to have a sensitivity of 93.3% and a specificity of 96%,19 while Ov20 was reported to have a sensitivity between 54% and 95% and a specificity of 99%.13 Theoretically, the combination of these antigens into one molecule should yield an immunodiagnostic reagent with an increased sensitivity and a similar specificity. To test this hypothesis, we constructed two hybrid proteins (OvH2 and OvH3) and evaluated their immunodiagnostic potential in comparison with a crude O. volbulus soluble antigen. Previous studies have shown that the recognition of Ov33 by antibodies from patients with onchocerciasis was best when the protein was expressed with an N-terminal GST fusion partner protein.19 The N-terminally fused partner protein probably provided an optimal conformation to Ov33, making important epitopes accessible to antibodies. Thus, it was conceivable that a fusion protein combining parts of the highly sensitive and specific proteins Ov20 and Ov33 in an optimal sequence, i.e., Ov20 being the N-terminal fusion partner of Ov33, would yield hybrid antigens with optimal properties. The results presented in Figures 1 and 2 confirm this hypothesis. The 100% sensitivity of our test recorded in four of the five geographic regions tested is an improvement of previous results,9,11–13,19,26,27 and shows that the combination of both immunodiagnostic antigens into one hybrid protein had a cumulative effect. Currently, we do not have an explanation for the lack of seroreactivity of two Ecuadorian onchocerciasis sera with OvH2 or OvH3. Because this lack of reactivity was restricted to the group from Ecuador, we believe that it could be due to a particular situation of the onchocerciasis focus in Ecuador, e.g., relatively low transmission. Both patients with no antibody responses had low microfilarial densities in their skin and it is conceivable that they were clearing their infec-

tions while being not exposed to transmission of infective larvae. A similar low seroreactivity against Ov33, paralleled by weak antibody responses against other O. volbulus proteins, has been observed in microfilariae-positive patients in an area where transmission of onchocerciasis had been interrupted by vector control measures.28 The specificity of 97.7% obtained herein for OvH2 in the IgG ELISA was comparable with values reported elsewhere.11–13,19 A cocktail composed of three recombinant O. volvulus antigens was previously reported to have a specificity of 100%.13 However, another study reported that Ov20, the major component of the cocktail, displayed a low degree of cross-reactivity with sera from individuals infected with W. bancrofti.29 Thus, we assume that the specificity of OvH2 is comparable to this cocktail. Our attempts to increase the specificity of the test by detection of IgG4 antibodies against OvH2 and OvH3 did not yield better results, since the sensitivity and specificity were slightly lower than for IgG detection. This is surprising since IgG4 antibodies in filarial infections were repeatedly shown to be more specific than the total IgG response.9,19,30 The presence of a 6-histidine tag on the hybrid protein did not interfere significantly with the specificity of the test, confirming our results. We also evaluated the antibody responses of microfilariaenegative individuals who were exposed to the transmission of O. volvulus to the hybrid proteins. The serum donors lived in endemic areas together with infected individuals. The sera from Ecuador were collected from putatively immunes, i.e., well-characterized persons who were probably immune to O. volvulus infection, while the sera of microfilariae-negative exposed individuals from Togo might include sera of persons with low level and prepatent infections, as well as putatively immunes. The positive reactions of many sera from these exposed persons suggests that the early stages of the parasite induce diagnostic antibody responses against Ov33 and Ov20. This is supported by previous data that demonstrated that Ov33 is recognized by sera from patients with onchocerciasis before the onset of patency.26 Early seroconversion is conceivable, since both Ov20 and Ov33 are expressed by infective larvae of O. volvulus,8,14 which makes the hybrid protein a sensitive tool to monitor exposure to O. volvulus. The potential of OvH2 for diagnosing exposure to O. volvulus transmission as well as early and light infections, might be particularly valuable, i.e., in situations where it is difficult to diagnose onchocerciasis by detection of microfilariae or microfilarial DNA, e.g., in situations after interruption of transmission and/or following prolonged treatment with ivermectin. Further studies are needed to address this question. Consequently, our approach in constructing hybrid proteins for the diagnosis of onchocerciasis has led not only led to an improvement of diagnosis but has also yielded a reagent that is easy to produce. Similar reagents might be useful in cases where diagnosis with a single protein or with cocktails of antigens is unsatisfactory. Acknowledgments: We are grateful to the WHO Filariasis Serum Bank, the Edna McConnell Clark Foundation Serum Bank, and Drs. M. Yazdanbaksh and E. Sartono for kindly donating sera. We gratefully acknowledge the help of Drs. F. Fru, B. Drabner, and C. Foba, and the Cameroonian staff, especially J. Mishimbo, for their assistance during the field study. Financial support: This study was supported by a training grant from the Deutscher Akademischer Austauschdienst (DAAD) to Pius N. Nde, and by grants from the European Union to to Janette E.


Bradley, Vincent P. K. Titanji, and Richard Lucius (No. IC18-CT95– 0017). Authors’ addresses: Pius N. Nde, Thomas Pogonka, and Richard Lucius, Department of Molecular Parasitology, Humboldt University Berlin, Philippstrasse 13, 10115 Berlin, Germany, Telephone: 49-302093-6053, Fax: 49-30-2093-6051, E-mail: richard.lucius@rz. huberlin.de. Janette E. Bradley, Department of Biological Sciences, University of Salford, Salford M5 4WT, United Kingdom, Telephone: 44-161-295-5997, Fax: 44-161-295-5210. Vincent P. K. Titanji, Department of Life Sciences, University of Buea, PO Box 63, Buea, Cameroon, Telephone: 237-32-22-72, Fax: 237-32-27-69. Reprint requests: Richard Lucius, Department of Molecular Parasitology, Humboldt University Berlin, Philippstrasse 13, 10115 Berlin, Germany.

REFERENCES 1. World Health Organization, 1995. Onchocerciasis and its control. Report of a WHO Expert Committee on Onchocerciasis Control. World Health Organ Tech Rep Ser 852. 2. Taylor HR, Munoz B, Keyvan-Larijani E, Greene B, 1989. Reliability of detection of microfilariae in skin snips in the diagnosis of onchocerciasis. Am J Trop Med Hyg 41: 467–471. 3. Bradley JE, Unnasch TR, 1996. Molecular approaches to the diagnosis of onchocerciasis. Adv Parasitol 37: 57–106. 4. Toe L, Adjami AG, Boatin BA, Back C, Alley ES, Dembele N, Brika PG, Pearlman E, Unnasch TR, 2000. Topical application of diethylcarbamazine to detect onchocerciasis recrudescence in west Africa. Trans R Soc Trop Med Hyg 94: 519–525. 5. Molyneux DH, Morel C, 1998. Onchocerciasis and Chagas’ disease control: the evolution of control via applied research through changing development scenarios. Br Med Bull 54: 327–339. 6. Ciment J, 1999. WHO celebrates triumph over river blindness. BMJ 23: 1090. 7. Zimmerman PA, Guderian RH, Aruajo E, Elson L, Phadke P, Kubofcik J, Nutman TB, 1994. Polymerase chain reactionbased diagnosis of Onchocerca volvulus infection: improved detection of patients with onchocerciasis. J Infect Dis 169: 686– 689. 8. Lucius R, Erondu N, Kern A, Donelson J, 1988. Molecular cloning of an immunodominant antigen of Onchocerca volvulus. J Exp Med 168: 1199–1204. 9. Chandrashekar R, Masood K, Alvarez RM, Ogunrinade AF, Lujan R, Richards FO, Weil GJ, 1991. Molecular cloning and characterization of recombinant parasite antigens for immunodiagnosis of onchocerciasis. J Clin Invest 88: 1460–1466. 10. Chandrashekar R, Ogunrinade AF, Weil GJ, 1996. Use of recombinant Onchocerca volvulus antigens for diagnosis and surveillance of human onchocerciasis. Trop Med Int Health 1: 575– 580. 11. Lobos E, Weiss N, Karam M, Taylor HR, Ottesen EA, Nutman TB, 1991. An immunogenic Onchocerca volvulus antigen: a specific and early marker of infection. Science 251: 1603–1605. 12. Lustigman S, Brotman B, Huima T, Prince AM, 1991. Characterization of an Onchocerca volvulus cDNA clone encoding a genus specific antigen present in infective larvae and adult worms. Mol Biochem Parasitol 45: 65–76. 13. Bradley JE, Trenholme KR, Gillespie AJ, Guderian R, Titanji V, Hong Y, McReynolds L, 1993. A sensitive serodiagnostic test for onchocerciasis using a cocktail of recombinant antigens. Am J Trop Med Hyg 48: 198–204. 14. Tree TI, Gillespie AJ, Shepley KJ, Blaxter ML, Tuan RS, Brad-

15. 16.

17.

18. 19.

20.

21.

22.

23. 24. 25. 26.

27.

28. 29.

30.

571

ley JE, 1995. Characterization of an immunodominant glycoprotein antigen of Onchocerca volvulus with homologues in other filarial nematodes and Caenorhabditis elegans. Mol Biochem Parasitol 69: 185–195. Ramachandran CP, 1993. Improved immunodiagnostic tests to monitor onchocerciasis control programs—a multicenter effort. Parasitol Today 9: 76–79. Rodriguez-Perez MA, Danis-Lozano R, Rodriguez MH, Bradley JE, 1999. Comparison of serological and parasitological assessments of Onchocerca volvulus transmission after 7 years of mass ivermectin treatment in Mexico. Trop Med Int Health 4: 98–104. Maia-Herzog M, Shelley AJ, Bradley JE, Luna Dias AP, Calvao RH, Lowry C, Camargo M, Rubio JM, Post RJ, Coelho GE, 1999. Discovery of a new focus of human onchocerciasis in central Brazil. Trans R Soc Trop Med Hyg 93: 235–239. Weil GJ, Steel C, Liftis F, Li BW, Mearns G, Lobos E, Nutman TB, 2000. A rapid antibody card test for diagnosis of onchocerciasis. J Infect Dis 182: 1796–1799. Lucius R, Kern A, Seeber F, Pogonka T, Willenbu¨ cher J, Taylor HR, Pinder M, Ghalib HW, Schulz-Key H, Soboslay P, 1992. Specific and sensitive immunodiagnosis of onchocerciasis with a recombinant 33kD Onchocerca volvulus protein (Ov33). Trop Med Parasitol 43: 139–145. Trenholme KR, Tree TIM, Gillespie AJ, Guderian R, Maizels R, Bradley JE, 1994. Heterogeneity of IgG antibody responses to cloned Onchocerca volvulus antigens in microfiladermia positive individuals from Esmeraldas province, Ecuador. Parasite Immunol 16: 201–209. Pogonka T, Brattig N, Nde P, Adam R, Seeber F, Zipfel PF, Titanji VPK, Lucius R, 1997. Localization of B and T cell stimulating domains of the immunodominant 33-kDa protein of Onchocerca volvulus (Ov33). Clin Immunol Immunopathol 85: 56–66. Elson LH, Guderian RH, Araujo E, Bradley JE, Days A, Nutman TB, 1994. Immunity to onchocerciasis: identification of a putatively immune population in a hyperendemic area of Ecuador. J Infect Dis 169: 588–594. Rychlik W, 1992. Oligo Version 4.0 for Macintosh. Reference Manual. Plymouth, MN: National Biosciences, Inc. Sambrook J, Fritsch EF, Maniatis F, 1989. Molecular Cloning. A Laboratory Manual. Second Edition. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Fisher LD, van Belle G, 1993. Biostatistics, A Methodology for the Health Sciences. New York: John Wiley and Sons, Inc. Tawill SA, Kipp W, Lucius R, Gallin M, Erttmann KD, Bu¨ ttner DW, 1995. Immunodiagnostic studies on Onchocerca volvulus and Mansonella perstans infections using a recombinant 33 kDa Onchocerca volvulus protein (Ov33). Trans R Soc Trop Med Hyg 89: 51–54. Tume CB, Ngu JL, Mckerrow JL, Seigel J, Sun E, Barr PJ, Bathurst I, Morgan G, Nkenfou C, Asonganyi T, Lando G, 1997. Characterization of a recombinant Onchocerca volvulus antigen (Ov33) produced in yeast. Am J Trop Med Hyg 57: 626–633. Lucius R, Prod’Hon J, Kern A, Diesfeld HJ, 1987. Antibody responses in forest and savanna onchocerciasis in Ivory Coast. Trop Med Parasitol 38: 194–202. Bloch P, Simonsen PE, Weiss N, Nutman T, 1998. The significance of guinea worm infection in the immunological diagnosis of onchocerciasis and bancroftian filariasis. Trans R Soc Trop Med Hyg 92: 518–521. Weil GJ, Ogunrinade AF, Chandrashekar R, Kale OO, 1990. IgG4 subclass antibody serology for onchocerciasis. J Infect Dis 161: 549–554.