Construction and consequences of directed mutations affecting the hemin receptor in pathogenic Corynebacterium species.

Genes encoding an ATP-binding cassette transporter system involved in hemin iron utilization from Corynebacterium ulcerans were cloned and characterized. The genes are homologous to a hemin transport system previously identified in Corynebacterium diphtheriae. Disruption of the hmuT gene, which encodes the putative hemin receptor, resulted in greatly reduced ability of C. ulcerans to use hemin or hemoglobin as an iron source. Inactivation of hmuT in C. diphtheriae by site-specific recombination had no effect on hemin utilization, which suggests that C. diphtheriae has an additional system for transporting hemin.

Corynebacterium diphtheriae genes required for acquisition of iron from haemin and haemoglobin are homologous to ABC haemin transporters.

Corynebacterium diphtheriae and Corynebacterium ulcerans use haemin and haemoglobin as essential sources of iron during growth in iron-depleted medium. C. diphtheriae and C. ulcerans mutants defective in haemin iron utilization were isolated and characterized. Four clones from a C. diphtheriae genomic library complemented several of the Corynebacteria haemin utilization mutants. The complementing plasmids shared an approximately 3 kb region, and the nucleotide sequence of one of the plasmids revealed five open reading frames that appeared to be organized in a single operon. The first three genes, which we have termed hmuT, hmuU and hmuV, shared striking homology with genes that are known to be required for haemin transport in Gram-negative bacteria and are proposed to be part of an ABC (ATP-binding cassette) transport system. The hmuT gene encodes a 37 kDa lipoprotein that is associated with the cytoplasmic membrane when expressed in Escherichi coli and C. diphtheriae. HmuT binds in vitro to haemin- and haemoglobin-agarose, suggesting that it is capable of binding both haemin and haemoglobin and may function as the haemin receptor in C. diphtheriae. This study reports the first genetic characterization of a transport system that is involved in the utilization of haemin and haemoglobin as iron sources by a Gram-positive bacterium.

Analyses of the cag pathogenicity island of Helicobacter pylori.

Most strains of Helicobacter pylori from patients with peptic ulcer disease or intestinal-type gastric cancer carry cagA, a gene that encodes an immunodominant protein of unknown function, whereas many of the strains from asymptomatically infected persons lack this gene. Recent studies showed that the cagA gene lies near the right end of a approximately 37kb DNA segment (a pathogenicity island, or PAI) that is unique to cagA+ strains and that the cag PAI was split in half by a transposable element insertion in the reference strain NCTC11638. In complementary experiments reported here, we also found the same cag PAI, and sequenced a 39 kb cosmid clone containing the left 'cagII' half of this PAI. Encoded in cagII were four proteins each with homology to four components of multiprotein complexes of Bordetella pertussis ('Ptl'), Agrobacterium tumefaciens ('Vir'), and conjugative plasmids ('Tra') that help deliver pertussis toxin and T (tumour inducing) and plasmid DNA, respectively, to target eukaryotic or prokaryotic cells, and also homologues of eukaryotic proteins that are involved in cytoskeletal structure. To the left of cagII in this cosmid were genes for homologues of HsIU (heat-shock protein) and Era (essential GTPase); to the right of cagII were homologues of genes for a type I restriction endonuclease and ion transport functions. Deletion of the cag PAI had no effect on synthesis of the vacuolating cytotoxin, but this deletion and several cag insertion mutations blocked induction of synthesis of proinflammatory cytokine IL-8 in gastric epithelial cells. Comparisons among H. pylori strains indicated that cag PAI gene content and arrangement are rather well conserved. We also identified two genome rearrangements with end-points in the cag PAI. One, in reference strain NCTC11638, involved IS605, a recently described transposable element (as also found by others). Another rearrangement, in 3 of 10 strains tested (including type strain NCTC11637), separated the normally adjacent cagA and picA genes and did not involve IS605. Our results are discussed in terms of how cag-encoded proteins might help trigger the damaging inflammatory responses in the gastric epithelium and possible contributions of DNA rearrangements to genome evolution.

Lymphocyte proliferative responses of goats vaccinated with Brucella melitensis 16M or a delta purE201 strain.

The response to a Brucella melitensis purEK deletion mutant, delta purE201 (referred to as strain 201), was compared with the response to its parental strain, 16M, in juvenile goats. Proliferative responses to gamma-irradiated bacteria were detected earlier in strain 201-infected goats. Lymphocytes from strain 16M- or 201-infected goats proliferated in response to one-dimensional polyacrylamide gel electrophoresis-separated proteins of similar mass isolated from strain 16M or Brucella abortus RB51. Data from this study suggest that some antigens stimulating cell-mediated responses are conserved among Brucella species, as 201- and 16M-infected goats recognized similar proteins expressed by RB51 and 16M.

Bacterial persistence and immunity in goats vaccinated with a purE deletion mutant or the parental 16M strain of Brucella melitensis.

To evaluate host responses, young goats were inoculated subcutaneously with a genetic deletion mutant (deltapurE201) of Brucella melitensis (n = 6), its virulent parental strain 16M (n = 6), or saline (n = 6). No clinical evidence of brucellosis was seen in any goat. Serum antibody titers peaked at postinoculation day (PID) 14. Bacteria in lymph nodes that drained sites of vaccination reached peak numbers of >10(6) CFU/g in both infected groups at PID 7 and progressively declined to PID 84. At necropsy, bacteria were present in mammary lymph nodes or spleen of 33% of goats given virulent 16M but in none of goats given the purE mutant. Lymphadenitis, most severe in goats given 16M, involved depletion of lymphocytes and germinal centers, proliferation of lymphoblasts, and vasculitis. By PID 28, lymph node architecture was restored; there was marked germinal center formation and medullary plasmacytosis. Brucellar antigens, detected with immunoperoxidase techniques, were prominent in capsular granulomas but not in lymph node cortices. Ultrastructurally, bacteria were found in macrophages (>97%) and small lymphocytes (<3%) but not in large lymphocytes. Bacteria were intact in small lymphocytes but in macrophages were in various stages of degradation. The deltapurE phenotype of deltapurE201 was preserved during infection of goat lymph nodes. Unlike Salmonella spp. purE mutants, strain deltapurE201 may be a candidate for efficacy testing; it produced immune responses, was cleared from visceral tissues, and produced less severe pathologic changes than its wild-type parent.

Deletion of purE attenuates Brucella melitensis infection in mice.

We previously showed that a purE mutant (delta purE201) of Brucella melitensis 16M is attenuated for growth in cultured human monocytes (E. S. Drazek, H. H. Houng, R. M. Crawford, T. L. Hadfield, D. L. Hoover, and R. L. Warren, Infect. Immun. 63:3297-3301, 1995). To determine if this strain is attenuated in animals, we compared the growth of the delta purE201 mutant with that of strain 16M in BALB/c mice. The number of bacteria in the spleen and spleen weight peaked for both strains between 1 and 2 weeks postinfection (p.i.), though the number of delta purE201 cells was significantly less than the number of 16M cells recovered from the spleens of infected mice. During the next 6 weeks, delta purE201 was essentially eliminated from infected mice (three of five mice sterile; < 100 CFU in two of live mice at 8 weeks p.i.), whereas bacteria persisted at a high level in the spleens of 16M-infected mice (about 106 CFU per spleen). The number of bacteria in the livers and lungs of mice infected with either strain paralleled those in the spleen. Mice infected with 16M had a strong inflammatory response, developing dramatic and prolonged splenomegaly (five to eight times normal spleen weight) and producing serum interleukin-6. In contrast, mice infected with delta purE201 developed only mild, transient splenomegaly at 1 week p.i. and produced no interleukin-6 in their serum. We further characterized the host response to infection by measuring changes in immune spleen cell populations by flow cytometry. CD4- and CD8-positive lymphocytes declined by I week in both experimental groups, while MAC-1-positive cells increased. T-cell subpopulations remained low or declined further, and MAC-1 cells increased to three times normal levels during 8 weeks of infection with 16M but returned to normal by 4 weeks after infection with delta purE201. These results document infectivity and attenuation of delta purE201 and suggest that it should be further evaluated as a potential vaccine.

Cloning and characterization of hemolytic genes from Helicobacter pylori.

Strains of Helicobacter pylori, the bacterium associated with gastritis, peptic ulcer disease, and gastric cancer in humans, express different degrees of hemolysis on agar containing erythrocytes (RBC). Here we report the isolation and characterization of six recombinant clones from a genomic library of H. pylori ATCC 49503 that confer on Escherichia coli the ability to lyse sheep RBC. DNA hybridizations indicated no sequence homology among these hemolytic clones. Hybridization mapping of them to an ordered H. pylori cosmid library identified their separate chromosomal locations. One clone hybridized to two regions separated by approximately 200 kb. The specificities of the hemolytic activities of these clones were tested with RBC from humans, monkeys, cattle, horses, guinea pigs, rabbits, and chickens as well as with RBC from sheep. One clone conferred the ability to lyse RBC from five species, a second clone allowed the lysis of RBC from four of these species, three other clones allowed the lysis of RBC from three of these species, and the sixth clone allowed the lysis of RBC from just two species. We propose that some or all of the genes that confer these various hemolytic activities contribute to pathogen-host tissue interactions and that the different specificities seen here are important for H. pylori infections of humans of different genotypes or disease states.

Deletion of purE attenuates Brucella melitensis 16M for growth in human monocyte-derived macrophages.

We constructed a defined purine-auxotrophic mutant of Brucella melitensis 16M by chromosomal gene replacement. We electroporated B. melitensis 16M with suicide plasmids containing a kanamycin resistance cassette that replaced 226 bp at the carboxyl end of purE, the intergenic region, and 18 bases of the purK open reading frame. Recombinant B. melitensis delta purE201 required exogenous purines for growth on minimal media. Purine auxotrophy was complemented by electroporation of B. melitensis delta purE201 failed to grow in human monocyte-derived macrophages, while the growth of wild-type 16M and the complemented strain, delta purE201 (pSD5), increased by nearly two logs. These results suggest that B. melitensis delta purE201 will be attenuated in animals and humans and thus may be useful as a live attenuated vaccine.

A mutation in the Neisseria gonorrhoeae rfaD homolog results in altered lipooligosaccharide expression.

The gonococcal lsi-6 locus was cloned and shown by DNA sequence analysis to have homology with the E. coli rfaD gene, which encodes ADP-L-glycero-D-mannoheptose epimerase. This enzyme is involved in the biosynthesis of the lipopolysaccharide precursor ADP-L-glycero-D-mannoheptose. A site-directed frameshift mutation in lsi-6 was constructed by PCR amplification and introduced into the chromosome of Neisseria gonorrhoeae MS11 P+ by transformation. The lipooligosaccharides (LOS) of mutant and parental strains were characterized by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The lsi-6 mutant produced LOS components with apparent molecular masses of 2.6 and 3.6 kDa as compared with a 3.6-kDa band of the MS11 P+ strain. The parental LOS phenotype was expressed when a revertant was constructed by transformation of the cloned wild-type gene into the lsi-6 mutant. The immunoreactivity of LOS from parental and constructed strains was examined by SDS-PAGE and Western blotting. Only the parental and reconstructed wild-type strains produced a 3.6-kDa LOS component that reacted with monoclonal antibody 2-1-L8. These results suggest that the lsi-6 locus is involved in gonococcal LOS biosynthesis and that the nonreactive mutant 3.6-kDa LOS component contains a conformational change or altered saccharide composition that interferes with immunoreactivity.

Characterization and presumptive identification of Helicobacter pylori isolates from rhesus monkeys.

We characterized 38 Helicobacter isolates, including 22 from gastric biopsy samples obtained from 14 rhesus monkeys and single isolates from 16 monkeys in a different colony. Biochemical profiles of these isolates were nearly identical to that of Helicobacter pylori ATCC 43504. Restriction fragment length polymorphism (RFLP) analysis indicated that each infected monkey harbored one to four strains. The 17 RFLP types found among these 22 isolates differed from all seven RFLPs found among the other 16 isolates. Thus, monkeys within a given colony are more likely to be infected by Helicobacter isolates with the same or a similar RFLP than are monkeys from different colonies. A 16S rRNA gene was amplified by PCR and cloned from the Helicobacter isolate from rhesus monkey 85D08. Ribotyping with this probe demonstrated less diversity among isolates from rhesus monkeys than was reported among isolates of H. pylori from humans, as did RFLP analysis of a PCR fragment of the ureA-ureB gene cluster. The DNA sequence of the cloned 16S rRNA gene was determined and compared with sequences reported for H. pylori and other Helicobacter species. Our analysis of 127 nucleotides (corresponding with residues 1240 to 1366 of the Escherichia coli 16S rRNA gene) indicated that the Helicobacter isolate from monkey 85D08 was 99.2 to 100% homologous to isolates of H. pylori from humans but only 83.5 to 96.9% homologous with other Helicobacter species in this region of the 16S rRNA gene. These data provide strong support for the presumptive identification of these isolates as H. pylori.

Natural gastric infection with Helicobacter pylori in monkeys: a model for spiral bacteria infection in humans.

BACKGROUND/AIMS: There is no generally accepted model for Helicobacter pylori infection in humans. The aim of this study was to examine the natural history and effect of treatment in rhesus monkeys and sequentially define the immune response to H. pylori in relation to treatment. METHODS: Infection and gastritis were graded blindly by histological analysis and culture of biopsy specimens harvested during gastroduodenoscopies in 26 anesthetized colony-bred monkeys. Plasma H. pylori-specific immunoglobulin (Ig) G levels were determined by enzyme-linked immunosorbent assay. RESULTS: H. pylori and Gastrospirilum hominis-like organisms were present in 13 and 9 monkeys, respectively; 3 animals harbored both organisms, whereas 4 monkeys were not infected. Gastritis score was < or = 1.5 in animals uninfected or infected only with G. hominis-like organisms and > or = 2.0 in all H. pylori-infected animals. IgG ratios were > or = 0.5 in 12 of 13 H. pylori-infected animals and in 2 of 13 H. pylori-negative animals (P < 0.001). One monkey became infected with H. pylori during the observation period, with concurrent increase of gastritis and plasma IgG levels. In untreated animals, infection, gastritis, and plasma IgG levels remained unchanged over 7-15 months. Triple therapy eradicated H. pylori at 6 months in 4 of 6 animals while suppressing gastritis and plasma IgG levels. CONCLUSIONS: Rhesus monkeys harboring H. pylori are persistently infected and have gastritis and elevated specific IgG levels, all of which may respond to appropriate therapy, whereas G. hominis infection is associated with little inflammation.

Low Incidence of Aeromonas sp. in Livestock Feces.

Pig, beef, sheep and turkey fecal specimens were assayed for recovery of inoculated Aeromonas sp. by directly plating the samples on five different agar media. Of these, starch-ampicillin was optimal with respect to selectivity and ability to differentiate from other resident microflora. Generally, the numbers of inoculated Aeromonas sp. recovered on starch-ampicillin agar were similar to those recovered on brain heart infusion and blood ampicillin agar media, and were 101 to 103 greater than the recovery rate on either MacConkey-ampicillin or cefsulodinirgasan-novobiocin agars. The sensitivity for the direct recovery of Aeromonas sp. from inoculated beef feces with naturally contaminating microflora, using streaked starch-ampicillin agar medium, was between 102 and 103 cells per gram. Using starch-ampicillin agar, the incidence of Aeromonas detected from feces of beef, pig, sheep and turkey held at the Beltsville Agricultural Research Center was one of 32, none of 22, none of 24 and three of 21, respectively. Based upon current taxonomic criteria, the isolate from the beef feces had characteristics consistent with both Aeromonas sobria and Aeromonas caviae , whereas three isolates from turkey feces were identified as A. caviae or Aeromonas hydrophila . The organism was isolated from five of five packages of ground beef from retail sources. The discrepancy in the consistent presence of the organism in retail meat suggests that many of the food isolates are probably not of fecal origin.