The neutral cysteine proteinase of Entamoeba histolytica degrades IgG and prevents its binding

Patients infected with Entamoeba histolytica generate specific IgG that does not prevent invasive amebiasis or recurrent infection. Studies investigated whether the effectiveness of the human humoral response was limited by cleavage of IgG by the extracellular neutral cysteine proteinase of E. histolytica trophozoites, one of the first amebic products to interact directly with components of host defenses. Purified proteinase cleaved polyclonal human and monoclonal murine IgG in a dose-dependent manner. Peptide sequencing of the major cleavage fragment(s), which contained the protein A binding site, suggested that cleavage occurred near the hinge region. Intact trophozoites also cleave IgG in both growth media and serum-free media. Cleaved monoclonal antibody to a 29-kDa surface antigen of E. histolytica bound to trophozoites 83.5% +/- 6.7% less than did uncleaved antibody. These results suggest that cleavage of IgG by the extracellular cysteine proteinase may limit the effectiveness of the host humoral response.

Peroxidase activity of a TSA-like antioxidant protein from a pathogenic amoeba

The 29 kDa surface protein of Entamoeba histolytica is an abundant antigenic protein expressed by pathogenic strains of this organism. The protein is a member of a widely-dispersed group of homologues which includes at least two cysteinyl peroxidases, Salmonella typhimurium alkyl hydroperoxidase C-22 protein (AhpC) and Saccharomyces cerevisiae thiol-specific antioxidant protein (TSA). Here, for the first time in a pathogenic eukaryote, we have demonstrated that the amoebic protein also possesses peroxidatic and antioxidant activities in the presence of reductants such as dithiothreitol or thioredoxin reductase plus thioredoxin. Although the S. typhimurium AhpF flavoprotein was not an effective reductant of the amoebic TSA protein, one inhibitory monoclonal antibody directed toward amoebic TSA was also partially inhibitory toward reduced but not oxidized bacterial AhpC. These antioxidant proteins are likely to be important not only in general cell protection, but also in the promotion of infection and invasion by these pathogenic organisms through protection against oxidative attack by activated host phagocytic cells.

Luteinizing hormone-releasing hormone in undifferentiated and differentiated SK-N-SH human neuroblastoma cells

The presence of luteinizing hormone-releasing hormone (LHRH) in SK-N-SH human neuroblastoma cells was investigated by immunocytochemistry and enzyme-linked immunosorbent assays of whole cell extracts and culture medium. In addition, ribonuclease protection assays were utilized to quantitate LHRH messenger RNA. The expression of LHRH mRNA and LHRH protein level was correlated with neuronal differentiation induced by retinoic acid (RA). In differentiated SK-N-SH cells the LHRH mRNA level as well as the amount of LHRH in cell extracts and cell medium were significantly lower than in differentiated cells. These results suggest that RA affects the expression of LHRH mRNA and the level of LHRH protein in SK-N-SH cells. These data show that altering the growth state of the human neuroblastoma SK-N-SH cells toward more neuronal phenotype results in a significant decrease in expression of LHRH mRNA and the protein. The ability of RA to induce changes in LHRH at the mRNA and at the peptide levels will allow further study of RA regulation of LHRH at the neuronal level.

Surface localization, regulation, and biologic properties of the 96-kDa alcohol/aldehyde dehydrogenase (EhADH2) of pathogenic Entamoeba histolytica

The 96-kDa surface antigen of Entamoeba histolytica was demonstrated through extensive immunologic evaluation with monoclonal and monospecific antibodies to be identical to or an isoform of the amebic alcohol/aldehyde dehydrogenase (EhADH2). EhADH2 was secreted, excreted, or shed into the culture medium in quantities commensurate with amebic growth when studied in a novel culture system. Of importance, using RNase protection assays, specific mRNA coding for the EhADH2 gene product(s) was up-regulated by treatment of viable trophozoites with the enzyme substrate ethanol. These data provide insight into the biology of this enzyme and its regulation by appropriate stressors.

Protection of gerbils from amebic liver abscess by immunization with recombinant Entamoeba histolytica 29-kilodalton antigen

The goal of our study was to obtain a highly conserved Entamoeba histolytica recombinant antigen for study as a subunit amebiasis vaccine. We screened a Uni-Zap cDNA library of E. histolytica (strain HM1:IMSS) with human immune sera and isolated a dominant 804-bp cDNA clone. A 33-kDa fusion protein expressed from the cDNA clone was determined by monoclonal antibody binding, DNA hybridization, and nucleotide sequence to be the complete E. histolytica 29-kDa antigen. Serum antibodies to the recombinant protein were detected by enzyme-linked immunosorbent assay in 80% of subjects from Egypt and South Africa with amebic liver abscess. Similar results were found with the native 29-kDa protein. Native and recombinant 29-kDa antigens induced proliferation of lymphocytes harvested from patients with amebic liver abscess (P < 0.01 compared with controls). Intraperitoneal immunization of gerbils with the recombinant fusion protein (10 micrograms) with Titermax adjuvant elicited an antigen-specific serum immunoglobulin G antibody response and was partially protective (54%) against intrahepatic challenge with 5 x 10(5) virulent axenic trophozoites (strain HM1:IMSS). In summary, the recombinant form of the E. histolytica 29-kDa antigen demonstrated serologic specificity for amebic liver abscess, exhibited conserved T-cell epitopes, and was effective as a subunit vaccine in an experimental animal model of amebic liver abscess.

Serologic reactivity to purified recombinant and native 29-kilodalton peripheral membrane protein of pathogenic Entamoeba histolytica

The 29-kDa peripheral membrane protein of Entamoeba histolytica has recently been demonstrated to have epitopes on pathogenic clinical isolates which were not detected by monoclonal antibodies on nonpathogenic isolates. To analyze the serological response to this protein, we tested 93 serum specimens (from 33 patients with amebic liver abscess, 7 patients with colitis, 2 patients with ameboma, 18 individuals harboring a nonpathogenic zymodeme strain, 10 healthy Mexican migrant workers, and 23 healthy controls) by enzyme-linked immunosorbent assay (ELISA) using immunoaffinity-purified native or recombinant protein. When tested by ELISA with the native antigen, 79% (26 of 33) of the serum specimens from patients with amebic liver abscess, 4 of 9 serum specimens from symptomatic patients with colitis or ameboma, and serum from one migrant worker were positive. None of the 18 subjects harboring a nonpathogenic strain or 23 control individuals were seropositive to the native antigen (sensitivity, 71%; specificity, 98%). Of 30 serum specimens from patients with amebic liver abscess tested with recombinant antigen, 27 were seropositive (90%). In addition, six patients with colitis or ameboma and two individuals who harbored a nonpathogenic strain were seropositive to the recombinant antigen. One healthy Mexican migrant worker tested positive by both ELISAs (sensitivity, 87%; specificity, 94%). Immunoblotting of 51 serum specimens to sodium dodecyl sulfate-denatured native 29-kDa protein was less sensitive (65%) than ELISA in detecting serum antibodies to the antigen. These results suggest a similar antibody response to native and recombinant antigens (r = 0.86) and support the potential utility of a quantitative assay with defined recombinant antigen for the serodiagnosis of invasive amebiasis in nonendemic areas in conjunction with other diagnostic tools.

Structural analysis and demonstration of the 29 kDa antigen of pathogenic Entamoeba histolytica as the major accessible free thiol-containing surface protein

The 29 kDa protein of pathogenic Entamoeba histolytica is a cysteine-rich surface antigen which we recently characterized by cDNA sequencing and by using monoclonal antibodies which differentiated between pathogenic and non-pathogenic clinical isolates. To determine the structure and biochemical attributes of this protein, a repertoire of immunological techniques using monoclonal antibodies, and radiolabelling were employed. We demonstrated that the 29 kDa protein forms a 60 kDa dimer and a high-molecular-mass oligomer(s) on the surface of the organism through disulphide bonds, and is the major accessible free thiol-containing surface protein of E. histolytica. The deduced amino acid sequence encoding the 29 kDa protein was found to share a common amino acid domain with sequences reported for Helicobacter pylori, Salmonella typhimurium, MER5 gene expressed in murine erythroleukemia cells, Clostridium pasteurianum, and a Bacillus spp.

Acquisition of apparently intact and unmodified lipopolysaccharides from Escherichia coli by Bdellovibrio bacteriovorus

The ability of Bdellovibrio bacteriovorus to relocalize the OmpF major outer membrane porins from its Escherichia coli prey to its own outer membranes is diminished in prey expressing smooth lipopolysaccharide (S-LPS). Since porins exist in the membrane complexed with LPS, we examined the LPS associated with relocalized porin to determine whether it had been acquired intact, mixed or replaced with Bdellovibrio LPS, or derivatized by the bdellovibrios. The relocalized trimers were found associated with the same LPS originally bound to them in the E. coli. The bulk-phase LPS from bdellovibrios grown on various chemotypes of rough prey was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis to determine whether more than the trimer-bound LPS was acquired by the bdellovibrios. This analysis revealed bands of Bdellovibrio LPS matching the LPS chemotype of the prey. One or two other bands were identical in migration to the LPS of prey-independent mutants of B. bacteriovorus and represented bdellovibrio-synthesized LPS. The LPS of bdellovibrios grown on prey with radiolabeled lipid A showed radioactivity only in gel band positions identical with those of the prey's LPS. The amount of this prey-derived LPS was shown by enzyme-linked immunosorbent assay to reach a constant value during the purification of the bdellovibrios, and it represented approximately 25% of the total Bdellovibrio LPS. Immunoelectron microscopy confirmed the presence of prey-derived LPS on the cell surface of bdellovibrios, and no evidence could be found for bdellovibrio-induced modifications of the relocalized prey LPS.

Identification of human, mouse, and rat retinoic acid receptor alpha using monoclonal antibodies

Monoclonal antibodies that recognize the human, mouse, and rat retinoic acid receptor alpha (RAR alpha) protein have been generated using synthetic peptides. Less well-characterized monoclonal antibodies were also generated against the RAR beta and RAR gamma proteins. Monoclonal antibodies of the IgG1 (R alpha 10) and IgG2a (R alpha 13) isotypes effectively and specifically recognize both the human and mouse RAR alpha protein. Preincubation of the antibodies with the synthetic RAR alpha peptide, but not with the RAR beta or RAR gamma peptides, blocked recognition of the approximately 55 kDa RAR alpha protein on western blots. These monoclonal antibodies also detected differing levels of RAR alpha in various rat tissues. These monoclonal antibodies will serve as powerful reagents to study the structure and regulation of the retinoic acid receptor protein.

Molecular and cellular characterization of the 29-kilodalton peripheral membrane protein of Entamoeba histolytica: differentiation between pathogenic and nonpathogenic isolates

To further characterize the 29-kDa surface antigen of Entamoeba histolytica, we analyzed the complete nucleotide sequence and compared the immunoreactivity of this antigen in pathogenic and nonpathogenic strains. Five cDNA clones (one 1.0-kb full-length clone, designated p13, and four partial-length clones) encoding the antigen were analyzed for allelic variation. Comparison of the nucleotide sequences revealed several single-nucleotide substitutions in all five cDNAs, two of which resulted in amino acid differences. Localization of the antigen to the amebic surface in a previous report (B. E. Torian, B. M. Flores, V. L. Stroeher, F. S. Hagen, and W. E. Stamm, Proc. Natl. Acad. Sci. USA 87:6358-6362, 1990) was corroborated by transmission electron microscopy showing colloidal gold particles on the surface of the trophozoites. Computer analysis of the deduced amino acid sequence predicted that the protein encoded by p13 was a hydrophilic peripheral membrane protein, and these data were confirmed by a Triton X-114 membrane extraction showing the presence of the 29-kDa antigen primarily in the aqueous phase of the detergent partition. Monoclonal antibodies to a fusion peptide differentiated between pathogenic and nonpathogenic clinical strains of E. histolytica in immunoblots. Although no immunoreactive epitopes were detected on nonpathogenic strains, Northern (RNA) analysis and DNA-DNA hybridization with a 700-bp cDNA probe demonstrated that mRNA and the gene encoding the 29-kDa surface antigen were present in both pathogenic and nonpathogenic clinical isolates.

Sequence of a cysteine-rich galactose-specific lectin of Entamoeba histolytica

Entamoeba histolytica trophozoites adhere to human colonic mucins and epithelial cells by a cell surface galactose-specific lectin. This lectin, which is composed of two subunits linked by disulfide bonds, has been shown to be a protective antigen in an animal model of amebiasis. We have determined the sequence of the mature form of the 170-kDa heavy subunit from cDNA clones and PCR-amplified fragments. The heavy subunit sequence consisted of a putative extracellular domain containing 1209 amino acids with 16 potential sites for N-linked glycosylation, a 26-amino acid hydrophobic region, and a 41-amino acid cytoplasmic tail. The presence of N-linked oligosaccharides was confirmed by culturing amebae with tunicamycin, which resulted in a decrease in the heavy subunit molecular mass to 160 kDa and a loss of lectin activity. The extracellular domain was remarkable for an extensive cysteine-rich domain that shared identify with similar regions of several other cell surface proteins and appeared to confer protease resistance to the subunit.

Differentiation of Naegleria fowleri from Acanthamoeba species by using monoclonal antibodies and flow cytometry

Monoclonal antibodies to Naegleria fowleri and Acanthamoeba polyphaga were analyzed by enzyme-linked immunosorbent assay, indirect immunofluorescence microscopy, and fluorescence flow cytometry to assess specificity and cross-reactivity with axenically cultured N. fowleri and Acanthamoeba spp. Four monoclonal antibodies to N. fowleri were specific for N. fowleri and had no reactivity to A. polyphaga. Similarly, four monoclonal antibodies to A. polyphaga did not react with N. fowleri. Two of the four monoclonal antibodies to A. polyphaga did not react with other Acanthamoeba spp. tested, while two of the antibodies demonstrated a high degree of cross-reactivity with a putative Acanthamoeba castellanii strain by immunofluorescence microscopy; this was confirmed by fluorescence flow cytometry for one of the antibodies. These monoclonal antibodies were used to identify Acanthamoeba trophozoites in infected brain sections of a patient who died of suspected Acanthamoeba-caused granulomatous amoebic encephalitis, demonstrating potential utility in the direct identification of N. fowleri and Acanthamoeba spp. in clinical specimens.

cDNA sequence analysis of a 29-kDa cysteine-rich surface antigen of pathogenic Entamoeba histolytica

A gamma gt11 cDNA library was constructed from poly(U)-Sepharose-selected Entamoeba histolytica trophozoite RNA in order to clone and identify surface antigens. The library was screened with rabbit polyclonal anti-E. histolytica serum. A 700-base-pair cDNA insert was isolated and the nucleotide sequence was determined. The deduced amino acid sequence of the cDNA revealed a cysteine-rich protein. DNA hybridizations showed that the gene was specific to E. histolytica since the cDNA probe reacted with DNA from four axenic strains of E. histolytica but did not react with DNA from Entamoeba invadens, Acanthamoeba castellanii, or Trichomonas vaginalis. The insert was subcloned into the expression vector pGEX-1 and the protein was expressed as a fusion with the C terminus of glutathione S-transferase. Purified fusion protein was used to generate 22 monoclonal antibodies (mAbs) and a mouse polyclonal antiserum specific for the E. histolytica portion of the fusion protein. A 29-kDa protein was identified as a surface antigen when mAbs were used to immunoprecipitate the antigen from metabolically 35S-labeled live trophozoites. The surface location of the antigen was corroborated by mAb immunoprecipitation of a 29-kDa protein from surface-125I-labeled whole trophozoites as well as by the reaction of mAbs with live trophozoites in an indirect immunofluorescence assay performed at 4 degrees C. Immunoblotting with mAbs demonstrated that the antigen was present on four axenic isolates tested. mAbs recognized epitopes on the 29-kDa native antigen on some but not all clinical isolates tested.

Inhibition of Trichomonas vaginalis motility by monoclonal antibodies is associated with reduced adherence to HeLa cell monolayers

Adherence of trichomonads to host epithelial cells appears to be a critical step in the pathogenesis of trichomoniasis. We evaluated the effect of a panel of 10 monoclonal antibodies on attachment of [35S]methionine-radiolabeled Trichomonas vaginalis strains to HeLa cell monolayers. Of 10 monoclonal antibodies, 3 totally eliminated motility of PHS2J strain trichomonads and reduced their adherence to 48 to 60% of control values (P less than 0.001). However, none of the monoclonal antibodies affected motility or adherence of STD13 strain trichomonads. Although the antibodies all reacted with PHS2J trichomonads by immunofluorescence, there was no correlation between inhibition of adherence and findings on either immunofluorescence or radioimmunoprecipitation. Direct microscopic observations showed that incubation with the monoclonal antibodies did not cause cytolysis of T. vaginalis. In quantitative cultures there was no difference in the number of colonies produced by parasites that had been incubated with antibodies that inhibited or had no effect on adherence. We conclude that our monoclonal antibodies reduced adherence not by cytotoxic effects or by competing for specific sites mediating adherence of the protozoa, but by inhibiting motility of T. vaginalis.

The 96-kilodalton antigen as an integral membrane protein in pathogenic Entamoeba histolytica: potential differences in pathogenic and nonpathogenic isolates

A surface antigen (EH-96) of Entamoeba histolytica was demonstrated to be a plasma membrane antigen by immunoprecipitation of metabolically 35S-labeled antigen from live trophozoites, Triton X-114 detergent extracts, and plasma membrane-enriched fractions prepared by concanavalin A membrane stabilization and differential centrifugation. In addition, the antigen was localized to the plasma membrane by electron microscopy with colloidal gold. Antigen from E. histolytica strains immunoprecipitated with specific immunoglobulin M (IgM) or IgG2b monoclonal antibody was identical by one-dimensional peptide mapping with N-chlorosuccinimide. Additionally, antigen from different axenically cultivated amebae was demonstrated to be identical by N-chlorosuccinimide peptide mapping, as were peptide maps of IgG and IgM monoclonal antibody-purified antigen. The 96-kilodalton (kDa) surface antigen was identified on four axenically cultivated pathogenic isolates and on three polyxenically cultivated pathogenic isolates (zymodeme II) of E. histolytica but was absent or present in lesser quantity on six nonpathogenic polyxenically cultivated isolates. The 96-kDa antigen was detected in liver abscess fluid from four patients with amebic abscesses by enzyme-linked immunosorbent assay (ELISA) and immunoprecipitation. Two-dimensional gel electrophoresis profiles of the 96-kDa antigen purified from abscess material or from polyxenically cultivated trophozoites demonstrated that the antigens were related to the 96-kDa antigen found in axenically cultivated organisms.

Antigenic heterogeneity in the 115,000 Mr major surface antigen of Trichomonas vaginalis

The 115,000-molecular-weight antigen of Trichomonas vaginalis was characterized using monoclonal antibodies developed to three different strains of T. vaginalis and one strain of Tritrichomonas foetus. The antigen was found to be present on all strains or isolates of T. vaginalis examined and was demonstrated to be located on the external surface plasma membrane by agglutination assays and complement-mediated lysis assays. Characteristics of the antigen were assessed with a proteolytic enzyme and periodate oxidation. Periodate treatment of whole T. vaginalis abrogated binding for eight antibodies while use of pronase-treated antigen resulted in loss of antibody binding for two different antibodies. Screening of 19 axenized clinical isolates of T. vaginalis and one strain each of T. foetus and Giardia lamblia with type-specific antibodies delineated three major groups of T. vaginalis based on antigenic specificities (epitope distributions) within the 115,000-molecular-weight antigen. In addition, one epitope of the 115,000-molecular-weight antigen was found only on the immunizing strain. Two epitopes were present on all T. vaginalis isolates as well as T. foetus and G. lamblia. One epitope was common to all T. vaginalis except one. A minimum of six different epitopes of the 115,000-molecular-weight antigen were identified. Antigens purified with type-specific or "common" monoclonal antibodies shared the same partial peptide maps demonstrating relatedness.

Use of monoclonal antibodies to identify, characterize, and purify a 96,000-dalton surface antigen of pathogenic Entamoeba histolytica

We identified and partially characterized a surface antigen of Entamoeba histolytica by using seven monoclonal antibodies obtained after injecting mice with a pathogenic strain of amoeba. An intrinsically radiolabeled 96,000-dalton antigen was immunoprecipitated by five of the seven monoclonal antibodies; this antigen was present in three strains of E. histolytica. The antigen was situated on the external surface of E. histolytica, as demonstrated by agglutination and immunofluorescence staining of live amoeba and by immunoprecipitation of iodinated trophozoite antigen. All seven monoclonal antibodies were specific for E. histolytica and failed to react in an ELISA with Trichomonas vaginalis, Tritrichomonas foetus, Giardia lamblia, Acanthamoeba castellonii, and Entamoeba invadens. Two monoclonal antibodies were used to purify the antigen: the purified antigen was identical when antibody binding to live organisms or antibody reactive with nonviable organisms was used for purification.

A new metabolic labelling medium for Trichomonas vaginalis and Tritrichomonas foetus using 35S methionine

A metabolic labelling medium was devised for Trichomonas vaginalis and Tritrichomonas foetus utilizing 35S methionine. T. vaginalis cultured for 24h in the medium took up approximately 27% of the available label and increased greater than two fold in number. Counts per microgram of protein were 32,555 +/- 10% between different strains or identical strains in different labelling runs. T. foetus took up approximately 5% of the available label and increased greater than two fold in 24h. This resulted in specific labelling of 12,704 cpm/ug protein +/- 10% between different runs with the same strain.

Identification of a surface antigen of Trichomonas vaginalis

A major surface antigen of Trichomonas vaginalis was purified by using three independently derived monoclonal antibodies (two immunoglobulin M and one immunoglobulin G1) prepared against T. vaginalis PHS-2J. A 115,000-molecular-weight antigen and one or more components with a molecular weight of 58,000 to 64,000 were recovered when any of the three antibodies was used as an immunoadsorbent. The purified antigen reacted with all three monoclonal antibodies in an enzyme-linked immunosorbent assay, indicating that the antibodies recognized the same antigen but not necessarily the same determinant. The purified antigen was sensitive to both pronase digestion and periodate oxidation. The antigen was shown to be on the external surface of some but not all T. vaginalis isolates by agglutination of live organisms with the monoclonal antibodies.

Tubulin and high-molecular-weight polypeptides as Giardia lamblia antigens

We determined the antigenic specificities of nine murine monoclonal antibodies to Giardia lamblia. Four distinct antibody reaction patterns were detected by immunofluorescence with G. lamblia trophozoites. Four monoclonal antibodies which reacted with the body but not the flagella of the whole trophozoites recognized two polypeptides of 170,000 and 155,000 molecular weights by immunoblotting. Two antibodies reacting with both the flagella and body also reacted with 170,000- and 155,000-molecular-weight antigens. An antibody specific for the G. lamblia attachment disk and two specific for flagella by immunofluorescence were found to react with 53,000- and 55,000-molecular-weight polypeptides by immunoblotting. These antigens comigrated with purified bovine brain tubulin, and their respective antibodies reacted in immunoblots with purified bovine brain tubulin. The antigens identified in this report were found to be present in Giardia cytoskeletal preparations. All antigens were pronase labile and heat stable. Four strains of G. lamblia reacted similarly in immunofluorescence assays as well as in immunoblotting assays.

Specific and common antigens of Trichomonas vaginalis detected by monoclonal antibodies

Monoclonal antibodies to Trichomonas vaginalis were prepared by immunizing mice with a cloned isolate of T. vaginalis. Eight antibodies reacted with the same four isolates or strains but did not react with the other T. vaginalis strains or isolates tested. All eight antibodies reacted uniformly with both the body and flagella of T. vaginalis in the immunofluorescence assay but were unreactive by immunoblotting. The antigen(s) recognized by these antibodies was determined to be present on the surface membrane by indirect immunofluorescence assay of live organisms. The antigen(s) was found to be sensitive to periodate oxidation but resistant to pronase digestion. In addition, one monoclonal antibody was derived which reacted with all T. vaginalis isolates or strains tested, as well as with Trichomonas gallinae, Tritrichomonas foetus, and Giardia lamblia. This antibody reacted with the body but not the flagella of Formalin-fixed protozoa in the immunofluorescence assay but failed to react with live organisms. The antigen was found to be periodate resistant but pronase labile. In the immunoblot assay, this antibody detected a single T. vaginalis polypeptide with a molecular weight of 62,000.