Identification of Chitinase as the Immunodominant ...

1 downloads 43 Views 704KB Size Report
Richard Lucius‡. From the ‡Division of Parasitology, Institute for Zoology, University of Hohenheim, Emil-Wolff-Strasse 34,. 70599 Stuttgart, Germany, the ¶Swiss ...
THE JOURNAL OF BIOLOGICAL CHEMISTRY © 1996 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 271, No. 3, Issue of January 19, pp. 1441–1447, 1996 Printed in U.S.A.

Identification of Chitinase as the Immunodominant Filarial Antigen Recognized by Sera of Vaccinated Rodents* (Received for publication, September 5, 1995, and in revised form, November 8, 1995)

Ralf Adam‡§, Brigitte Kaltmann‡, Werner Rudin¶, Thomas Friedrichi, Thomas Marti**, and Richard Lucius‡ From the ‡Division of Parasitology, Institute for Zoology, University of Hohenheim, Emil-Wolff-Strasse 34, 70599 Stuttgart, Germany, the ¶Swiss Tropical Institute, Socinstrasse 57, 4051 Basel, Switzerland, the iBASF AG, 67056 Ludwigshafen, Germany, **Bernhard-Nocht-Institute for Tropical Medicine, Bernhard-Nocht-Strasse 74, 20359 Hamburg, Germany

Filarial parasites, such as Onchocerca volvulus, Brugia malayi, and Wuchereria bancrofti, the causative agents of river blindness and lymphatic filariases, affect more than 100 million people throughout the tropics (1). Infection of the human host occurs by a bite of an infected arthropod. There is evidence that some individuals develop protective immunity naturally, as they do not acquire filarial infections despite high levels of local transmission (2, 3). The humoral immune response of these patients differentially recognizes antigens of infective third stage larvae (L3) (4), and it is conceivable that vaccination with such antigens could prevent the infection. As the study of protective immunity in human filariae is hampered by the host specificity of the parasites, we adopted the approach to identify immunorelevant antigens of the rodent filaria Acanthocheilonema viteae, a parasite of the jird (Meriones unguiculatus). For animal models, the vaccination with irradiation-attenuated infective larvae (L3) has been demonstrated to be the most effective way to induce protective immunity (5). Immunization of jirds with irradiation-attenuated L3 of A. viteae induces

* This work was supported by grants of the Deutsche Forschungsgemeinschaft and the Edna McConnell Clark Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The nucleotide sequence(s) reported in this paper has been submitted to the GenBankTM/EMBL Data Bank with accession number(s) U14638. § Received a scholarship from the “Studienstiftung des Deutschen Volkes.” To whom correspondence should be addressed: Inst. for Molecular Parasitology, Humboldt University Berlin, Invalidenstr. 43, D-10115 Berlin, Germany. Tel.: 49-30-2895255; Fax: 49-30-2895251; E-mail: [email protected].

more than 90% protection against a challenge infection (6). Sera of such vaccinated animals recognize few L3 proteins. We report here on the characterization and the molecular cloning of the immunodominant antigen recognized by the humoral immune response of challenge resistant jirds, a chitinase of A. viteae L3. Chitinases were recently described from microfilariae of B. malayi (7). It was supposed that the enzymes might have a role in casting the microfilarial sheath, a modified egg shell. This structure was believed to contain chitin on biochemical grounds (8, 9) and due to its analogy with the chitincontaining egg shell of other nematodes. Interestingly, the egg shell is the only structure of nematodes known to contain chitin. However, recent reports deny the presence of chitin in the microfilarial sheath (see Refs. 10 –12) and call the chitindegrading role of filarial chitinase into question. Our characterization of L3 chitinase (an active, chitin-degrading enzyme) in a filarial stage that is not yet known to contain chitin suggests that the enzyme has additional substrate specificities. Chitinase-like proteins of vertebrates, a class of animals not producing chitin, were described to degrade extracellular matrix under inflammatory or degenerative conditions and to play a role in the process of fertilization (13, 14). The close homology of the L3 chitinase with these molecules suggests that the described enzyme degrades N-acetylglucosamine-containing structures of the parasite during molting and might help the parasite to migrate through the host’s tissues. MATERIALS AND METHODS

Parasites—The life cycle of A. viteae was maintained in jirds (M. unguiculatus) and soft ticks (Ornithodoros moubata) as described earlier (15). Filarial stages were handled in parasite medium (PM) consisting of RPMI 1640 (Life Technologies, Inc.) supplemented with 20 mM Hepes, pH 7.3, 4 mM L-Gln, and 100 units/ml penicillin/streptomycin. Adult filariae were collected by dissection from the subcutaneous tissues of jirds. Uterine MF1 were obtained by teasing out and mincing the uterus of female worms and collecting the embryonic stages from the PM. Newborn MF were collected from female worms kept in culture for 3 days in PM. Blood MF were isolated from the blood of infected jirds by centrifugation through a Percoll gradient. To isolate vector-derived L3, ticks with 8-week-old infections were dissected in PM and after 30 min, the emerging L3 were collected. Induced L3 were produced by inoculating batches of each 1,000 L3 subcutaneously into jirds, retrieving the larvae from the muscle tissue 3 days postinfection in a Baermann funnel. L4 were obtained by dissecting the jirds 5 days postinfection and culturing the released L3 in PM, where they molted synchronously within 24 h. L4 were obtained from the rodents 13 days postinfection. L3 culture supernatant and molting supernatant were produced by

1 The abbreviations used are: MF, microfilaria(e); L3, infective stage larva(e); L4, fourth-stage larva(e); mAb, monoclonal antibody; IFAT, indirect fluorescent antibody test; PBS, phosphate-buffered saline; SBA, soybean agglutinin agarose; CTAB, cetylmethylammonium bromide; PAGE, polyacrylamide gel electrophoresis.

1441

Downloaded from http://www.jbc.org/ by guest on November 21, 2016

Acanthocheilonema viteae is a parasitic nematode of rodents. We identified the chitinase of A. viteae infective stage larvae (L3) as the main target of the humoral immune response of jirds, which were protected against challenge infection after vaccination with irradiation attenuated L3. The cDNA of the L3 chitinase has been sequenced, and the deduced amino acid sequence shows significant homologies to chitinases of Brugia malayi microfilariae, insects, yeast, bacteria, and Streptomyces sp. The protein has been characterized by monoclonal antibodies and substrate activity gels. The chitinase of L3 may contribute to degrading the nematode cuticle during molting and thus represents a target of protective immune responses in a phase where the parasite is highly vulnerable. In addition, it has been shown that a similar enzyme exists in uterine microfilariae, which probably has a role in casting the egg shell.

1442

Filarial Chitinases A chitinase-specific DNA probe was amplified by PCR with phage DNA from the L3 library and degenerate primers P1 (derived from the N terminus of the 205-kDa protein, sense) and the degenerate primer P2, based on the consensus region of amino acids 211–216 of B. malayi and 73–78 of W. bancrofti, respectively (7, 23). PCR was performed according to Saiki et al. (24) (1 min denaturation at 94 °C, 2 min annealing at 50 °C, and 3 min extension at 72 °C for 30 cycles) with phage DNA of the L3 library using Taq polymerase (AGS, Heidelberg). Amplification products were directly cloned into the pGEM-T vector (Promega, Madison, WI) and sequenced by the dideoxy method (25). The cDNA probe was labeled with digoxigenin (Boehringer, Mannheim) and used to screen the L3 library. Positive clones were carried through three rounds of screening. Full-length clones were determined by PCR with primer P1 and a poly(T) primer. The insert DNA was excised in vivo (22), and the purified plasmid DNA was sequenced. Sequencing of the PCR fragment and the cDNA insert was performed with vector-specific primers adjacent to the cloning site in l ZAP and the sequence-specific primers P1–P5. All DNAs were sequenced from both sides. Computer Sequence Analysis—Data base searches were performed with the HUSAR network service of the German Cancer Research Center, Heidelberg, based on the program package of the GCG Inc. (Madison, WI) (26). Nucleotide Sequence Accession Numbers—The sequence data reported in this paper have been submitted to the GenBank data base (accession number U14638). RESULTS

Characterization of L3 Chitinase of A. viteae—Sera of jirds, which had been vaccinated with irradiation-attenuated L3 of A. viteae and showed a resistance to challenge infection of .90%, reacted strongly with two antigen bands of L3 (205 and 68 kDa) in immunoblots (Fig. 1A, panel 1). Within the L4 larva, i.e. the larval stage developing from the L3 after molting, two bands of 68 and 140 kDa are recognized preferentially (Fig. 1A, panel 1), whereas sera of naive jirds did not react (Fig. 1A, panel 2). To characterize the immunoreactive L3 antigens two mAbs were produced against irradiated L3 and against molting fluid, respectively. mAb 24-4, resulting from fused spleen cells of a mouse immunized with irradiation-attenuated L3, specifically recognized the 205 kDa band of L3 and of L3 culture supernatant and reacted slightly with the 68 kDa band (Fig. 1A, panel 4). mAb 2H2, produced by immunization with molting fluid of L3, recognized in immunoblots the 205 and 68 kDa bands of L3 (Fig. 1A, panel 3). mAb 2H2 also reacted in immunoblots with a 205- and a 68-kDa antigen of L3 culture supernatant, and with a 68 kDa band of culture supernatant collected after the in vitro molting of L3 (Fig. 1C). Both mAbs did not recognize any antigen in L4, indicating that the main target antigens of the immune response of vaccinated jirds are stage-specific. N-terminal sequence analysis of the 205-kDa protein of L3 generated a single amino acid residue at each cycle. Six amino acids could be determined from the 205-kDa antigen (Fig. 2). A data base search of the EMBL and GenBank data bases revealed an identity with the amino acids of the N terminus of a endochitinase of MF of B. malayi. For the characterization of the 68-kDa protein, it has been purified from a supernatant of A. viteae L3 after extraction with PBS containing 1% CTAB by lectin affinity chromatography using SBA-agarose (see “Materials and Methods”). N-terminal sequencing of the purified glycoprotein revealed 20 amino acids showing an identity between the 68- and the 205-kDa protein (Fig. 2). To test for chitinase activity, extracts of L3 were electrophoresed under nonreducing conditions in SDS gels containing glycol chitin. Chitinase activity was localized preferentially in the 68 kDa band of L3, and weakly in the 205-kDa protein, which was recognized by both mAbs (Fig. 1D). Extensive boiling of L3 antigens using a sample buffer containing 100 mM dithiothreitol resulted in the disappearance of the 205-kDa protein (Fig. 3). Therefore, we assume that monomeric 68-kDa chitinase can form trimers, which are linked by disulfide bridges

Downloaded from http://www.jbc.org/ by guest on November 21, 2016

culturing batches of 1,000 vector-derived L3 or 1,000 induced L3, respectively, in 500 ml of PM for 24 h. The supernatants were cleared by centrifugation and filtering through a 0.22-mm membrane filter. Sera and Monoclonal Antibodies (mAbs)—Sera were collected from jirds vaccinated with irradiation-attenuated L3 and challenged with vector-derived L3 as described previously (6). The worm burden of the vaccinated animals was ,10% of the worm burden of control jirds. For the production of mAb 24-4 (IgG1, k L chains) against irradiated L3, 6-week-old BALB/c mice were immunized twice in biweekly intervals by subcutaneous injection of 50 irradiated L3 of A. viteae. To produce mAb 2H2 (IgM, k L chains) against molting supernatant, animals were immunized thrice in biweekly intervals with 500 ml of molting supernatant mixed with 200 ml of STP (squalane, Tween 80, pluronic L101) adjuvant (see Ref. 6). 5 days prior to the fusion, the animals were boosted with a further dose of irradiated L3 or molting fluid, respectively, given intraperitoneally. Spleen cells of these animals were fused with the BALB/c myeloma X63/Ag8 and selected according to standard methods (16). A mAb against a genus-specific epitope of O. volvulus (IgG1) served as control antibody. Immunochemical Characterization of Chitinase—Filarial proteins were solubilized in sample buffer consisting of 1% (w/v) SDS, 5% (v/v) 2-mercaptoethanol, 10% (v/v) glycerol in 63 mM Tris-HCl, pH 6.8, to which were added traces of bromphenol blue and protease inhibitors. 1% (v/v) of a stock solution of 1 M e-amino-n-caproic acid (Serva), 0.1 M EDTA, 0.1 M benzamidine (Sigma) in water, and 1% (v/v) of a 0.1 M solution of phenylmethylsulfonyl fluoride (Serva, Heidelberg, Germany) in methanol. Alternatively, the proteins were solubilized in sample buffer containing 100 mM dithiothreitol instead of 2-mercaptoethanol as a reducing reagent. The proteins were separated by SDS-PAGE (17) and blotted onto nitrocellulose membranes (18). The sheets were reacted with undiluted hybridoma supernatant or appropriate dilutions of jird sera. Immune complexes were detected with horseradish-conjugated anti mouse IgG1IgM (Tago, Burlingame, CA). For the indirect fluorescent antibody test (IFAT), L3 were fixed overnight at 4 °C in 1% (v/v) formaldehyde in PBS. PBS washed uterine contents or newborn MF of A. viteae were dried on glass slides and fixed in acetone for 10 min at 4 °C. The preparations were sequentially reacted with undiluted hybridoma supernatants and fluorescein-conjugated anti-mouse IgG 1 IgM 1 IgA (Nordic, Tilburg, The Netherlands). Immunogold Technique—A. viteae adult females and L3 were fixed in 0.1 M PBS containing 0.5% glutaraldehyde (v/v), pH 7.4, and processed and embedded in Lowicryl K4 M or HM 20 for electron microscopy. Thin sections (70 nm) were reacted with appropriate dilutions of the antibody or an isotype-matched control mAb. The immune complexes were visualized with rabbit immunoglobulins against mouse antibodies (Dako, Copenhagen, Denmark) conjugated with 10-nm gold particles as described previously (19). Lectin Affinity Chromatography—L3 were suspended in PBS, pH 7.4, containing 1% cetylmethylammonium bromide (CTAB) and sonicated in five 30-s intervals (200 watts) with 1 min of cooling on ice in between. Parasite material, which remained insoluble was separated by centrifugation at 10,000 3 g for 30 min. The supernatant was applied to a soybean agglutinin agarose (SBA) column (Sigma) equilibrated in PBS. After extensive washing with PBS, bound antigens were eluted by 0.2 M GalNAc in PBS. The fractions were subsequently assayed by SDSPAGE and Western blot analysis with mAb 2H2. Amino Acid Sequencing—L3 antigens extracted with PBS, pH 7.4, containing 1% CTAB were separated by one-dimensional SDS-PAGE in the presence of 1% thioglycolic acid and blotted onto polyvinylidine difluoride membrane (Immobilon, Millipore, Bedford, MA). The 205 kDa band was excised and sequenced directly using an ABI 476 protein sequencer according to Matsudaira (20). The SBA-purified 68-kDa monomer of L3 chitinase has been treated in the same manner. Measurement of Chitinase Activity—Chitinase activity of parasite extracts and culture supernatants was analyzed in 12.5% polyacrylamide gels copolymerized with glycol chitin under nonreducing conditions (21). Parasites were extracted with PBS, pH 7.4, without protease inhibitors in the presence of either 1% Triton X-100 or 1% CTAB. Following electrophoresis, proteins were renatured in the gel and stained with Calcofluor white M2R according to Fuhrman et al. (7). cDNA Libraries, PCR Amplification, Library Screening, and DNA Sequencing—cDNA libraries were prepared in l ZAP (22) with RNA of 110,000 L3 or 50 female worms, respectively, of A. viteae according to the manufacturer’s instructions (Stratagene, La Jolla, CA). The L3 library contained 2 3 106 plaque forming units, consisting of 91% recombinant phage with a mean insert length of 1,200 base pairs. The female library contained 3.5 3 106 plaque forming units, consisting of 98% recombinant phage with a mean insert length of 1,000 base pairs.

Filarial Chitinases

1443

that can be reduced by high concentrations of sulfhydryl compounds. As mAb 24-4 was directed against an epitope stable under a variety of conditions, it served to localize the filarial chitinase in L3. Immunogold staining of ultrathin sections of L3 with mAb 24-4 revealed the presence of A. viteae chitinase in the cellular cytoplasm and in the lumen of the glandular oesophagus of vector-derived L3 (Fig. 4A). No chitinase could be detected in muscle, cuticle, or on the outermost surface of L3. Neither live L3 nor formaldehyde-fixed L3 studied under a variety of conditions by IFAT carried detectable amounts of chitinase on the surface. Characterization of L3 Chitinase cDNA—Degenerate primers P1 and P2 derived from the N-terminal sequence of the A. viteae L3 chitinase and a region conserved within the chitinases of B. malayi and W. bancrofti (7, 23) were used to amplify a 580-base pair cDNA sequence by PCR with a L3 cDNA library of A. viteae. The amplification product was sequenced, and 88% of the nucleotides were identical with the corresponding sequence of the published B. malayi MF chitinase. This indicated that a part of a chitinase gene of A. viteae had been amplified. The sequence was labeled with digoxigenin and used as a probe to isolate clones from the L3 library. Screening of 8 3 104 phage plaques yielded 13 positive clones after three rounds of screening. All inserts had the same length as shown by PCR with

vector-specific primers corresponding to the T3 and T7 promotor regions. One full-length clone was sequenced in both strands using sequence-specific primers P1–P6 and vector specific primers for the T3 and T7 promoter regions. The obtained sequence was 1,670 base pairs long and contained an open reading frame coding for 520 amino acids with a theoretical molecular mass of 58 kDa (Fig. 2). The 59-end consisted of a potential signal sequence of 17 residues (27), indicating that L3 chitinase is secreted. This sequence was followed by the N terminus of the mature protein, which corresponded exactly to the N-terminal amino acid sequence of the protein as obtained by protein sequencing (Fig. 2). The sequence showed one potential N-glycosylation site, six myristylation sites, and one attachment site for glycosaminoglycan. The noncoding 39-end showed a consensus signal for polyadenylation (28). The 59-end of the mRNA did not contain a spliced leader sequence, as confirmed by DNA sequencing and by PCR with phage DNA of the L3 library using an oligonucleotide primer corresponding to the nematode spliced leader sequence (29) and the sequence-specific primer P3. The cDNA exhibited high homology (69% nucleotide identity over the whole sequence) to the MF chitinase of B. malayi (7) and weaker similarities to chitinases of W. bancrofti (23), insects, bacteria, Streptomyces sp., fungi, and plants (30 –32). The domain structure was similar to the one of B. malayi MF

Downloaded from http://www.jbc.org/ by guest on November 21, 2016

FIG. 1. Part A, stage specificity and immunodominance of the A. viteae L3 chitinase. 15 mg of A. viteae L3 (A) and L4 (B) extracts were electrophoresed on 12.5% SDS-PAGE and transblotted. Blots were probed with sera of vaccinated jirds (panel 1), sera of naive jirds (panel 2), mAb 2H2 (panel 3), mAb 24-4 (panel 4), control mAb (IgG1, panel 5). Part B, reaction of mAb 2H2 with A. viteae male (lane 1) and female worms (lane 2), blood MF (lane 3), and infective stage larvae (lane 4). Parasites were extracted as above and electrophoresed on a 4 –15% SDS-polyacrylamide gel, and the transblotted proteins were immunostained. Part C, reaction of mAb 2H2 with extracts of vector-derived L3 (lane 1), L3 3 days postinfection (lane 2), L4 16 days postinfection (lane 3), culture supernatants of 1,000 L3 after 24 h cultivation under vertebrate conditions (lane 4), molting supernatants of 1,000 L3 (lane 5). Samples were processed as in part B. Part D, chitinase activity is demonstrated in substrate-SDS/ polyacrylamide gels containing glycol chitin for total female worm extract (lane 1) and total L3 extract (lane 2).

1444

Filarial Chitinases

Downloaded from http://www.jbc.org/ by guest on November 21, 2016

FIG. 2. Nucleotide sequence and deduced amino acid sequence of the A. viteae L3 chitinase. Amino acid numbering begins at the initiating methionine, with the mature N terminus beginning at amino acid 18. Lower case letters represent the signal peptide on the amino acid line and untranslated regions on the nucleic acids line. Primers used for PCR amplifications and sequencing are marked as arrows above the nucleic acids line. The consensus polyadenylation signal is underlined. The N-terminal amino acid sequence derived from the purified 205- and 68-kDa proteins is boxed and shadowed, and the amino acids resulting from the 205-kDa protein are in italics. Potential myristilation sites are indicated by an inversed Gly residue. The attachment site for glycosaminoglycan (amino acids 234 –237) is shaded. One potential N-glycosilation site is doubly underlined. Conserved Cys residues within the carboxyl-terminal domain are indicated in boldface. The four repeated amino acid sequences in the C-terminal domain are boxed and shadowed.

chitinase. The N-terminal signal sequence is followed by the well conserved catalytic domain, spanning amino acids 18 –370 (31–33). The third domain is less conserved and contains 35% Ser and Thr residues. It could be a target of extensive Oglycosilation (7, 31). This region comprises four imperfect repeats of 14 amino acids in length between positions 370 and 440 of the mature protein. The carboxyl-terminal end of the protein is closely related in structure to the B. malayi and insect chitinases as the 6 Cys residues are perfectly matched, suggesting that they could be involved in disulfide bridges. However, these regions are different from the chitin-binding domains described for yeast and class III plant chitinases. To study the presence of chitinase mRNA, cDNA from male worms, female worms, blood MF, L3, and L4 was amplified with degenerate primers P1 and P2. In all stages, except male

worms, specific amplification products were obtained, indicating that chitinase mRNA is present in most filarial stages (not shown). Characterization of Uterine MF Chitinase—Studies with the mAbs against L3 chitinase revealed that uterine MF bear epitopes common to L3 chitinases. mAb 2H2 (binding to the 68and 205-kDa chitinase bands of L3) cross-reacted with a 220 kDa band and a 26 kDa band of female worms (Fig. 1B), whereas mAb 24-4 (recognizing the 205-kDa antigen of L3) reacted with a 26 kDa band of female worms only (not shown). None of the mAbs recognized antigens of male worms, blood MF or L4 (Figs. 1, B and C). Chitinase activity was present in the 220 kDa band of female worms, but not in the 26 kDa band, neither in other filarial stages as shown by degradation of glycol chitin in substrate gels (Fig. 1D). The active 220 kDa

Filarial Chitinases band of female worms co-migrated exactly with the band recognized by mAb 2H2. Immunogold staining of ultrathin sections of female worms with mAb 24-4 revealed that chitinase was localized in the epidermis and in the laminated layer of the cuticle of most MF present in the uteri (Fig. 4B). Only a small proportion of the MF carried the antigen on the outermost surface. The egg shell itself and other compartments of the MF did not contain detectable amounts of chitinase. To study the presence of chitinase on the surface of uterine MF, we performed IFAT with mAbs 24-4 and 2H2 using cryostat sections and preparations of uterine content of female A. viteae, where both mAbs immunostained exactly the same structures. IFAT analysis revealed that the target epitopes were accessible to the mAbs on nearly mature uterine MF, where chitinase was evenly distributed on the surface. Most of these Ag-bearing MF were still within the egg shell (Fig. 5, A and B). In contrast, no more chitinase was present on the surface of fully mature, hatched uterine MF, which are more slender than the nearly mature forms and have

FIG. 4. Immunogold staining of ultrathin sections of A. viteae L3 (A) and A. viteae MF in utero (B) with mAb 24-4. c, cuticle; e, epidermis; es, egg shell; go, glandular oesophagus; mc, muscle cell; mf, microfilaria; ul, uterine lumen. (L3 3 32,000; MF 3 48,000; bars represent 0.5 mm).

pointed heads. However, MF, which had degenerated within the female worms around the time of molting and were stumpy, immotile, and sometimes broken, carried detectable amounts of chitinase on their surface. An IFAT study of newborn MF revealed that such degenerated MF bearing chitinase on their surface represented around 10% of the MF production released by in vitro cultured female A. viteae (Fig. 5, C and D). DISCUSSION

The most effective way to induce protective immunity against filarial infections in experimental animals is the immunization with irradiation-attenuated L3 (see Refs. 5 and 6). Our study reveals that L3 chitinase is the immunodominant antigen recognized by the humoral immune response of jirds vaccinated against A. viteae infection. This protein has several properties of a vaccine candidate antigen; in particular it is exported during the initial phase of the infection and during molting of L3, which are phases of key importance for the attrition of filarial larvae (34, 35). Furthermore, L3 culture supernatants and molting fluid, which both contain chitinase, induce protection against a challenge infection with A. viteae in jirds (6).2 The findings of Freedman et al. (3), describing a 43-kDa chitinase-like filarial protein to be recognized by the sera of persons resistant against infection with the lymphatic filaria W. bancrofti (23), support the notion of chitinase being a protein with protective properties. Immunization studies in our animal model will evaluate whether immune responses against L3 chitinase can prevent an infection with A. viteae. Chitinases are enzymes that hydrolyze chitin (poly-b-(1– 4)-linked GlcNAc). Well known examples of chitinases are the enzymes of insects (Manduca), which have a role in degrading the chitinous exoskeleton during the molt and the chitindegrading enzymes of fungi and streptomycetes (30 –32).

2 R. Adam, B. Kaltmann, W. Rudin, T. Friedrich, T. Marti, and R. Lucius, unpublished results.

Downloaded from http://www.jbc.org/ by guest on November 21, 2016

FIG. 3. Reaction of mAb 2H2 with extracts of vector derived A. viteae L3. Antigens were electrophoresed on a 4 –15% SDS-polyacrylamide gel under reducing conditions using a sample buffer containing 5% (v/v) 2-mercaptoethanol (lane 1) and 100 mM dithiothreitol (lane 2), respectively.

1445

1446

Filarial Chitinases

FIG. 5. Localization of A. viteae chitinase in intrauterine MF by IFAT with mAb 24-4 (left panels) and corresponding light microscope photographs (right panels). Uterine contents of gravid female A. viteae with nearly mature MF inside the egg shell, younger embryonic stages and a mature, hatched MF (A and B). Newborn MF (C and D). Note that fluorescent microfilariae are swollen and stumpy. es, embryonic stages; nmf, nearly mature microfilariae; hmf, hatched microfilariae. Bars represent 50 mm.

and are released around the time of molting by L3 of O. volvulus (45). Molting products of nematodes are so far poorly described and the only pharyngeal gland component described on the molecular level is a metalloprotease of H. contortus (46). Recent data2 indicate that the production of L3 chitinase is regulated by steroid hormones, which is an analogy to the tighly regulated expression of insect chitinases during molting (30). Our study suggests that the chitinase of uterine MF has a role in cleaving chitinous structures of the eggshell of A. viteae. Chitin was detected in the eggs of intestinal helminths Ascaris suum and Heligmosomoides polygyrus (41, 47) and it was described to be a component of the egg shell of the filariae Onchocerca gibsoni and O. volvulus (42). Chitinase is considered as a relevant factor for the casting of the egg shell of A. suum (47). The strict coincidence of the occurrence of chitinase on the surface of uterine MF of A. viteae and the time point of molting suggests that A. viteae chitinase contributes to hatching. It is conceivable that chitinase detected within the cuticle of immature uterine MF represents a storage form, which is transported to the MF surface prior to the hatching. The lack of chitinase in blood MF of A. viteae suggests a tightly regulated expression of the protein at the required timepoint. The contrasting presence of chitinase in blood MF of B. malayi could be explained by the fact that MF of this filarial species remain surrounded by the modified egg shell, which is not cast until the blood MF have entered the arthropod host (48). The cDNA of the A. viteae L3 chitinase shows 69% nucleotide identity with the cDNA of B. malayi MF chitinase (7). The chitinase of the tobacco hornworm Manduca sexta (30) is also closely related to the filarial chitinases as shown by the similarity of the enzyme domain structure and by the amino acid composition (28% of the amino acids are identical, and about 70% are similar), whereas the described chitinases of streptomycetes, bacteria, and fungi are less related to our molecule.

Downloaded from http://www.jbc.org/ by guest on November 21, 2016

However, chitinases can cleave substrates other than chitin and can have activities as trans-glycosidases (36). The role of chitinase in L3 is unclear, and the fact that the enzyme is released by the L3 during different phases suggests that it might have a broad substrate specificity. First, L3 chitinase is exported into the culture medium immediately after the larvae are cultured under vertebrate conditions. This time point corresponds to the early infection where L3 migrate through host tissues (37). An enzyme exported during this phase could help to degrade host tissues, provided that it has an appropriate substrate specificity. Interestingly, chitinaselike molecules were recently described from the cartilago and the oviduct of vertebrates (13, 14, 38 – 40), which are supposed to contribute to the degradation of extracellular matrix under inflammatory or degenerative conditions or to fertilization of vertebrates. Second, L3 chitinase is produced during molting, a phase where the nematodes’ cuticle is reorganized and finally cast. It is not clear whether chitin is a target substrate during this phase, as the only stage of nematodes that has been demonstrated to contain chitin is the egg (41– 42). However, recent studies described chitotriosides in the cuticle of adult Haemonchus contortus, which is indicative for chitin being a structural component of the nematode cuticle (43). Therefore, further studies have to show whether the target substrate of L3 chitinase is chitin or whether other substrates are converted as b-(1– 4)-linked N-acetylglucosamine oligomers, which were shown to occur in the filarial cuticle by lectin binding studies or enzymatic studies (19, 43). Irrespective of the substrate, the localization of the enzyme and the timing of release suggest that L3 chitinase is involved in the process of molting. The source of released L3 chitinase are the pharyngeal glands, where the enzyme was localized by immunoelectron microscopy. Pharyngeal gland products, which are exported through the buccal cavity, were described to have a role in molting in other nematodes (44)

Filarial Chitinases

Acknowledgments—We thank Kirsten Berg, Andrea Kern, Silvia Seidinger, and Werner Kutritz for excellent technical assistance and Dr.

Juliet Fuhrman for helpful discussions. We also thank Uwe Oberla¨nder for help with the L4 production and Dr. Juliet Fuhrman for kindly reading the manuscript. REFERENCES 1. World Health Organization Expert Committee on Filariasis (1992) Technical Report Series No. 821, World Health Organization, Geneva 2. Elson, L. H., Guderian, R. H., Araujo, E., Bradley, J. E., Days, A. & Nutman, T. B. (1994) J. Infect. Dis. 169, 588 –594 3. Freedman, D. O., Nutman, T. B. & Ottesen, E. A. (1989) J. Clin. Invest. 83, 14 –22 4. Nutman, T. B., Steel, C., Ward, D. J., Zea-Flores, G. & Ottesen, E. A. (1991) J. Infect. Dis. 163, 1128 –1133 5. Philipp, M., Davis, T. B., Storey, N. & Carlow, C. K. S. (1988) Annu. Rev. Microbiol. 42, 1–53 6. Lucius, R., Textor, G., Kern, A. & Kirsten, C. (1991) Exp. Parasitol. 73, 184 –196 7. Fuhrman, J. A., Lane, W. S., Smith, R. F., Piessens, W. F. & Perler, F. B. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 1548 –1552 8. Fuhrman, J. A. & Piessens, W. F. (1985) Mol. Biochem. Parasitol. 17, 93–104 9. Devaney, E. (1985) Trop. Med. Parasitol. 36, 25–28 10. Fuhrman, J. A. (1995) Parasitol. Today 11, 259 –261 11. Schraermeyer, U., Peters, W. & Zahner, H. (1987) Parasitol. Res. 73, 550 –556 12. Klonisch, T., Bardehle, G., Linder, D., Boschek, B., Schott, H. H., Zahner, H. & Stirm, S. (1991) Parasitol. Res. 77, 448 – 451 13. Renkema, G. H., Boot, R. G., Muijsers, A. O., Donker-Koopman, W. E. & Aerts, J. M. F. G. (1995) J. Biol. Chem. 270, 2198 –2202 14. DeSouza, M. M. & Murray, M. K. (1995) Endocrinology 136, 2485–2496 15. Lucius, R. & Textor, G. (1995) Appl. Parasitol. 36, 22–33 16. Galfre, C. & Milstein, C. (1981) Methods Enzymol. 73, 3– 46 17. Laemmli, U. K. (1970) Nature 227, 680 – 685 18. Burnette, W. N. (1981) Anal. Biochem. 112, 195–203 19. Kiefer, E., Rudin, W. & Hecker, H. (1989) Acta Trop. 46, 3–15 20. Matsudaira, P. (1987) J. Biol. Chem. 262, 10035–10038 21. Trudel, J. & Asselin, A. (1989) Anal. Biochem. 178, 362–366 22. Short, J. M., Fernandez, J. M., Sorge, J. A. & Huse, W. D. (1988) Nucleic Acids Res. 16, 7583–7600 23. Raghavan, N., Freedman, D. O., Fitzgerald, P. C., Unnasch, T. R., Ottesen, E. A. & Nutman, T. B. (1994) Infect. Immun. 62, 1901–1908 24. Saiki, R. K., Gelfand, D. H., Stoffel, S., Scharf, S. J., Higushi, R., Horn, G. T., Mullis, K. B. & Erlich, H. A. (1988) Science 239, 487– 491 25. Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Proc. Natl. Acad. Sci. U. S. A. 74, 5463–5467 26. Devereux, J. (1984) Nucleic Acids Res. 12, 387–395 27. von Heijne, G. (1983) Eur. J. Biochem. 133, 17–21 28. Wickens, M. & Stephenson, P. (1984) Science 226, 1045–1051 29. Zeng, W. & Donelson, J. E. (1992) Mol. Biochem. Parasitol. 55, 207–216 30. Kramer, K. J., Corpuz, L., Choi, H. K. & Muthukrishnan, S. (1993) Insect Biochem. Mol. Biol. 23, 691–701 31. Kuranda, M. J. & Robbins, P. W. (1991) J. Biol. Chem. 266, 19758 –19767 32. Blaak, H., Schnellmann, J., Walter, S., Henrissat, B. & Schrempf, H. (1993) Eur. J. Biochem. 214, 659 – 669 33. Watanabe, T., Kobori, K., Miyashita, K., Fujii, T., Sakai, H., Uchida, M. & Tanaka, H. (1993) J. Biol. Chem. 268, 18567–18572 34. Abraham, D., Grieve, R. B., Holy, J. M. & Christensen, B. M. (1989) Am. J. Trop. Med. Hyg. 40, 598 – 604 35. Eisenbeiss, W. F., Apfel, H. & Meyer, T. F. (1994) J. Immunol. 152, 735–742 36. Flach, J., Pilet, P. E. & Jolles, P. (1992) Experientia 48, 701–716 37. Richer, J. K., Sakanari, J. A., Frank, G. R. & Grieve, R. B. (1992) Exp. Parasitol. 75, 213–222 38. Hakala, B. E., White, C. & Recklies, A. D. (1993) J. Biol. Chem. 268, 25803–25810 39. Arias, E. B., Verhage, H. G. & Jaffe, R. C. (1994) Biol. Reprod. 51, 685– 694 40. Sendai, Y., Abe, H., Kikuchi, M., Satoh, T. & Hoshi, H. (1994) Biol. Reprod. 50, 927–934 41. Arnold, K., Brydon, L. J., Chappell, L. H. & Gooday, G. W. (1993) Mol. Biochem. Parasitol. 58, 317–323 42. Brydon, L. J., Gooday, G. W., Chappell, L. H. & King, T. P. (1987) Mol. Biochem. Parasitol. 25, 267–272 43. Rhoads, M. L. & Fetterer, R. H. (1994) J. Parasitol. 80, 756 –763 44. Bird, A. F. & Bird, J. (1991) The Structure of Nematodes, Academic Press, San Diego, CA 45. Strote, G. & Bonow, I. (1991) Parasitol. Res. 77, 526 –535 46. Gamble, H. R., Purcell, J. P. & Fetterer, R. H. (1989) Mol. Biochem. Parasitol. 33, 49 –58 47. Justus, D. E. & Ivey, M. H. (1969) J. Parasitol. 55, 472– 476 48. Rogers, R., Ellis, D. S. & Denham, D. A. (1976) J. Helminthol. 50, 251–257

Downloaded from http://www.jbc.org/ by guest on November 21, 2016

The sequence of the chitinase-like 43-kDa molecule of W. bancrofti (23), which is derived from genomic DNA, has two small regions of homology in common with A. viteae L3 chitinase and B. malayi MF-chitinase but is otherwise relatively distinct. The homology between these filarial chitinases is highest (86% nucleotide identity) in the N-terminal signal sequence and the adjacent 370-amino acid region, the catalytic domain. The catalytic domain of A.viteae L3 chitinase is identical to the one of uterine MF of A. viteae and very closely related to the one of uterine MF of O. volvulus.2 Downstream follows a Ser/Thr-rich domain, which encompasses three nearly perfect repeats of 14 amino acids in the B. malayi gene, whereas this region shows four imperfect repeats of each 14 amino acids in the A. viteae gene. The Ser/Thr-rich domain is 21 amino acids longer in the A. viteae cDNA. The carboxyl-terminal Cys-rich domain shows similarities between the filarial chitinases and the enzyme of M. sexta. The 6 Cys residues are perfectly matched, indicating a conserved function of this domain, which is potentially important for intra- and intermolecular bridging. The fact that the open reading frame of the L3 chitinase cDNA codes for a protein with a theoretical molecular mass of 58 kDa suggests that the protein backbone is posttranslationally modified. The dicrepancy between theoretical and actual molecular mass is probably due to extensive O-glycosylation of the Ser/Thr-rich domain and the very acidic composition of this region, which could be responsible for an anomaly in the migration in SDS gels (7, 31). Deglycosilation of native A. viteae L3 chitinase using O-glycanase resulted in a shift of molecular weight. Expression of the cDNA in Escherichia coli yielded a 58-kDa protein, which hydrolized chitin, indicating that the activity of the enzyme is not dependent on glycosylation.2 Our experiments show that the 205 kDa band of L3 is a multimeric form of chitinase, derived from 68-kDa monomers by disulfide bridging. Forming of multimers was associated with an alteration of the antigenicity and of the chitinase activity, since mAb 24-4 did not bind to the 68 kDa band and the 205 kDa protein was shown to be less active in substrate gels. The presence of chitinase in two distinct stages of the parasite’s life cycle and the stage specific localization of the enzyme suggest that the expression of chitinases of L3 and MF is specifically regulated. Furthermore, the differences of molecular weight between L3 chitinase and MF chitinase show a specific structure of the molecules. Sequencing of the C terminus of the cDNAs of both stages revealed sequence variation,2 being suggestive for the presence of several chitinase genes. However, it is also possible that stage-specific differential splicing or differential posttranslational modifications contribute to the observed differences between L3 chitinase and MF chitinase of A. viteae. It will be interesting to study which specific properties are related to the timing of expression, the localization of the enzyme within the parasites, and the substrate specificity of chitinases of various stages and species of filarial parasites.

1447

Identification of Chitinase as the Immunodominant Filarial Antigen Recognized by Sera of Vaccinated Rodents Ralf Adam, Brigitte Kaltmann, Werner Rudin, Thomas Friedrich, Thomas Marti and Richard Lucius J. Biol. Chem. 1996, 271:1441-1447. doi: 10.1074/jbc.271.3.1441

Access the most updated version of this article at http://www.jbc.org/content/271/3/1441 Alerts: • When this article is cited • When a correction for this article is posted

This article cites 46 references, 18 of which can be accessed free at http://www.jbc.org/content/271/3/1441.full.html#ref-list-1

Downloaded from http://www.jbc.org/ by guest on November 21, 2016

Click here to choose from all of JBC's e-mail alerts