Specific phospholipids enhance the activity of beta-lactam antibiotics against Pseudomonas aeruginosa.

Pseudomonas aeruginosa PAO1 became considerably more sensitive to the action of ampicillin when grown in the presence of certain phospholipids. Only phospholipids capable of forming lipid bilayers or micelles proved to be capable of enhancing ampicillin activity. Of the phospholipids tested, 1-palmitoyl-2-hydroxy-sn-glycero-3-phosphate, also called monopalmitoylphosphatidic acid (MPPA), was the best enhancer. In the absence of MPPA, the MIC and MBC of ampicillin for P. aeruginosa PAO1 were 1 and 2 g/L, respectively. In the presence of MPPA, the MIC and MBC were 20 and 40 mg/L, respectively. MPPA was shown to enhance ampicillin activity by binding both Ca(2+) and Mg(2+), suggesting that the mechanism of enhancement is similar to that previously reported for Ca(2+) and Mg(2+) chelators. Surprisingly, MPPA by itself slowed the growth of four mucoid multiply antibiotic-resistant strains of P. aeruginosa recently isolated from the sputum of cystic fibrosis patients, and enhanced their sensitivity to piperacillin. It also increased the sensitivity of two ceftazidime-resistant P. aeruginosa cystic fibrosis strains to ceftazidime.

A functional cra gene is required for Salmonella enterica serovar typhimurium virulence in BALB/c mice.

A minitransposon mutant of Salmonella enterica serovar Typhimurium SR-11, SR-11 Fad(-), is unable to utilize gluconeogenic substrates as carbon sources and is avirulent and immunogenic when administered perorally to BALB/c mice (M. J. Utley et al., FEMS Microbiol. Lett., 163:129-134, 1998). Here, evidence is presented that the mutation in SR-11 Fad(-) that renders the strain avirulent is in the cra gene, which encodes the Cra protein, a regulator of central carbon metabolism.

A Salmonella typhimurium mutant unable to utilize fatty acids and citrate is avirulent and immunogenic in mice.

Salmonella typhimurium SR-11 is extremely virulent at a dose as low as 10(5) colony forming units (cfu) when administered perorally to BALB/c mice. Utilizing mini-transposon mutagenesis, a mutant of S. typhimurium SR-11 was isolated that was unable to utilize oleate and citrate as carbon sources. This mutant, designated S. typhimurium SR-11 Fad- (Fatty acid), was found to utilize sugars under cya/crp control as sole carbon sources, suggesting that the mutation is not in either of these genes. In addition, SR-11 Fad- utilized pyruvate and succinate, but was unable to utilize either acetate or isocitrate as sole carbon source. In contrast to SR-11, SR-11 Fad- was found to be avirulent, i.e. BALB/c mice were completely healthy after oral infection with 10(9) S. typhimurium SR-11 Fad- cells. Moreover, 21 days after SR-11 Fad- infection, BALB/c mice were found to be protected against an oral challenge with 10(9) cells of S. typhimurium SR-11.

Isolation and characterization of an attenuated strain of Pseudomonas aeruginosa AC869, a 3,5-dichlorobenzoate degrader.

Pseudomonas aeruginosa AC869, a 3,5-dichlorobenzoate degrader, is a mouse pathogen and has a reported 50% lethal dose (LD50) of 1.05 x 10(7) CFU when given intranasally to C3H/HeJ mice (S.E. George, M.J. Kohan, M.I. Gilmour, M.S. Taylor, H.G. Brooks, J.P. Creason, and L.D. Claxton, Appl. Environ, Microbiol. 59:3585-3591, 1993). AC869 was serotyped as O6 when grown in CD-1 mouse cecal and lung mucus but could not be assigned an O serotype when grown in Luria broth (LB). After growth in mouse cecal mucus, a less virulent mutant, AC869-11, was isolated from AC869 by using bacteriophage E79, which adsorbs to the O side chain of lipopolysaccharide (LPS). AC869-11 produced significantly less O antigen on its LPS than AC869 when grown in mouse lung and cecal mucus. The mutant also produced half the amount of exoenzyme S and 16-fold less extracellular protease than AC869 and was more sensitive than its parent to a number of antibiotics when grown either in LB or in mouse lung mucus. AC869-11 had ninefold higher LD50 than AC869 in CD-1 mice when administered intranasally. AC869-11 was found in the lungs, small intestine, cecum, and large intestine in numbers at least 100-fold below AC869, 3 h after intranasal exposure of mice to a sublethal dose of the two strains. Moreover, AC869-11 induced a decreased pulmonary inflammatory response relative to AC869. In contrast to AC869, AC869-11 did not translocate to the mesenteric lymph nodes, liver, and spleen following a sublethal dose. Despite attenuation, AC869-11 grew as well as AC869 with 3,5-dichlorobenzoate as the sole carbon and energy source. However, although AC869-11 survived in 3,5-dichlorobenzoate-contaminated soil as well as AC869 for 1 week, it failed to survive as well thereafter. These results suggest the possibility that mutations that lead to pulmonary attenuation of P. aeruginosa in mice also lead to weakness in the environment, despite such mutants maintaining the ability to degrade toxic substances under laboratory conditions.

The Escherichia coli K-12 gntP gene allows E. coli F-18 to occupy a distinct nutritional niche in the streptomycin-treated mouse large intestine.

Escherichia coli F-18 is a human fecal isolate that makes type 1 fimbriae, encoded by the fim gene cluster, and is an excellent colonizer of the streptomycin-treated mouse intestine. E. coli F-18 fimA::tet, lacking type 1 fimbriae, was constructed by bacteriophage P1 transduction of the fim region of the E. coli K-12 strain ORN151, containing the tetracycline resistance gene from Tn10 inserted in the fimA gene, into E. coli F-18. E. coli F-18 fimA::tet was found to occupy a distinct niche in the streptomycin-treated mouse intestine when fed in small numbers (10(4) CFU) to mice, along with large numbers (10(10) CFU) of E. coli F-18, as defined by the ability of the E. coli F-18 fimA::tet strain to grow and colonize only 1 order of magnitude below E. coli F-18. The same effect was observed when mice already colonized with E. coli F-18 were fed small numbers of E. coli F-18 fimA::tet. Experiments which show that the E. coli K-12 gene responsible for this effect is not fim::tet but gntP, which maps immediately downstream of the fim gene cluster, are presented. gntP encodes a high-affinity gluconate permease, suggesting that the distinct niche in the mouse large intestine is defined by the presence of gluconate. The data presented here support the idea that small numbers of an ingested microorganism can colonize the intestine as long as it can utilize an available nutrient better than any of the other resident species can.

Escherichia coli F-18 and E. coli K-12 eda mutants do not colonize the streptomycin-treated mouse large intestine.

The Escherichia coli human fecal isolates F-18 and K-12 are excellent colonizers of the streptomycin-treated mouse intestine. E. coli F-18 and E. coli K-12 eda mutants (unable to utilize glucuronate, galacturonate, and gluconate) were constructed by insertional mutagenesis. Neither the E. coli F-18 eda nor the E. coli K-12 eda mutant was able to colonize the streptomycin-treated mouse intestine, whether they were fed to mice together with their respective parental strains or alone. Complementation of the eda mutants with pTC190 (containing a functional E. coli K-12 eda gene) completely restored the colonization ability of both eda mutants. Relative to their parental strains, the E. coli F-18 eda mutant and the E. coli K-12 eda mutant grew poorly in cecal mucus isolated from mice fed either normal mouse chow or a synthetic diet containing sucrose as the sole carbon source, yet the mutants and parental strains demonstrated identical growth rates in minimal medium with glucose as the carbon source. E. coli F-18 edd eda and E. coli K-12 edd eda double mutants colonized the streptomycin-treated intestine when fed to mice alone; however, when fed simultaneously with their respective parental strains, they were poor colonizers. Since the edd gene is involved only in gluconate metabolism via the Entner-Doudoroff pathway, these results implicate the utilization of gluconate and the Entner-Doudoroff pathway as important elements in E. coli colonization of the streptomycin-treated mouse large intestine.

Role of leuX in Escherichia coli colonization of the streptomycin-treated mouse large intestine.

Escherichia coli F-18, a normal human fecal isolate, is an excellent colonizer of the streptomycin-treated mouse large intestine. E. coli F-18 Col-, a derivative of E. coli F-18 that no longer makes the E. coli F-18 colicin, colonizes the mouse large intestine as well as E. coli F-18 when fed alone, but is eliminated when fed together with E. coli F-18. Recently, a random bank of E. coli F-18 DNA was transformed into E. coli F-18 Col-, the resultant population was fed to streptomycin-treated mice, and the intestine was used to select the best colonizer. In this fashion, a 6.5 kb E. coli F-18 DNA fragment was isolated. This fragment was shown to enhance E. coli F-18 Col- mouse large intestinal colonizing ability and survival during stationary phase in intestinal mucus in vitro, as well as stimulate the synthesis of type-1 fimbriae. Here, we present evidence that the gene responsible for the enhanced E. coli F-18 Col- colonizing ability and survival during stationary phase in vitro is leuX. This gene encodes a rare leucine tRNA specific for the UUG codon. In addition, we show that the presence of a functional leuX gene is necessary for E. coli K-12 intestinal colonization and for survival in stationary phase.

Affinity purification of Hydra glutathione binding proteins.

The association of glutathione (GSH) with putative external chemoreceptors elicits feeding behavior in Hydra. In the present study, solubilized membrane proteins were chromatographed on an affinity column of immobilized GSH in order to isolate GSH-binding proteins that may represent the Hydra GSH chemoreceptor. The most abundant of the affinity-purified proteins was a triplet of peptides ranging in molecular weight from 24.5-26 kDa. Antiserum generated against the 24.5-26 kDa triplet peptides inhibited GSH-stimulated feeding behavior by 47%, implicating a role for one or more of these peptides in Hydra chemoreception.

Stimulation of Escherichia coli F-18Col- type-1 fimbriae synthesis by leuX.

Escherichia coli F-18, a normal human fecal isolate, is an excellent colonizer of the streptomycin-treated mouse large intestine. E. coli F-18Col-, a derivative of E. coli F-18 which no longer makes the E. coli F-18 colicin, colonizes the large intestine as well as E. coli F-18 when fed to mice alone but is eliminated when fed together with E. coli F-18. Recently we randomly cloned E. coli F-18 DNA into E. coli F-18Col- and let the mouse intestine select the best colonizer. In this way, we isolated a 6.5-kb E. coli F-18 DNA sequence that simultaneously stimulated synthesis of type 1 fimbriae and enhanced E. coli F-18Col- colonizing ability. In the present investigation we show that the gene responsible for stimulation of type 1 fimbriae synthesis appears to be leuX, which encodes a tRNA specific for the rare leucine codon UUG. Moreover, it appears that expression of leuX may be regulated by two proteins (22 kDa and 26 kDa) encoded by genes immediately adjacent to leuX.

Immune recognition of human Hsp60 by Lyme disease patient sera.

Members of the Hsp60 family of microbial heat shock proteins frequently serve as immunodominant antigens and immunological responses to these highly conserved proteins have been implicated in the pathology of a number of autoimmune diseases and inflammatory processes associated with microbial infection. In the present study, sera from patients with Lyme disease were examined by Western blot analysis for the presence of IgG against Borrelia burgdorferi antigens and for autoreactive IgG against recombinant human Hsp60 (huHsp60). These results were then compared to those obtained using sera from normal healthy controls, patients with a variety of acute non-spirochete infections, and patients with rheumatoid arthritis (RA). The results indicate a high incidence of autoreactive antibodies against huHsp60 in the sera of Lyme disease patients (67.9%) and patients with RA (75%). Positive reactivity was observed at lower rates in sera from healthy controls (25%) and sera from patients with acute non-spirochete infections (38%). Together the data suggest an association between the presence of autoreactive antibodies against huHsp60 and infection with B. burgdorferi. A similar association may exist between the presence of autoreactive antibodies against huHsp60 and RA.

Utilization of the mouse large intestine to select an Escherichia coli F-18 DNA sequence that enhances colonizing ability and stimulates synthesis of type 1 fimbriae.

Escherichia coli F-18, a normal human fecal isolate, is an excellent colonizer of the streptomycin-treated mouse large intestine. E. coli F-18 Col-, a derivative of E. coli F-18 which no longer makes the E. coli F-18 colicin, colonizes the large intestine as well as E. coli F-18 when fed to mice alone but is eliminated when fed together with E. coli F-18. Random sequences of E. coli F-18 DNA were cloned into pRLB2, a par-B-stabilized derivative of pHC79. The entire gene library was transformed into E. coli F-18 Col- and fed to streptomycin-treated mice. The mouse large intestine selected a predominant clone which contained a recombinant plasmid (pRLB7) that enhanced E. coli F-18 Col- colonizing ability 100-fold but did not stimulate colicin synthesis. Moreover, pRLB7 simultaneously improved the survival of E. coli F-18 Col- in stationary phase in vitro, utilizing nutrients derived from mouse cecal mucus, and stimulated synthesis of both type 1 fimbriae and three E. coli F-18 Col- outer membrane proteins (74, 71, and 69 kDa). The 6.5-kb E. coli F-18 DNA sequence in pRLB7 does not contain either the fim operon or pilG (hns), both known to be involved in type 1 fimbrial synthesis. The sequence encodes six proteins, all smaller than the three E. coli F-18 Col- outer membrane proteins whose synthesis it stimulates. Collectively, the results suggest that the cloned E. coli F-18 DNA sequence contains one or more regulators of E. coli F-18 Col- operons expressed in the mouse large intestine in vivo and in isolated mouse cecal mucus in vitro.

Escherichia coli F-18 phase locked 'on' for expression of type 1 fimbriae is a poor colonizer of the streptomycin-treated mouse large intestine.

Escherichia coli F-18, a human fecal isolate, makes type 1 fimbriae in vitro and in the streptomycin-treated mouse large intestine in vivo, and is an excellent colonizer of the cecal mucus layer in the streptomycin-treated mouse large intestine. E. coli F-18(pPKL91) harbors an extra fimB gene on a parB stabilized pPBR322 plasmid and is therefore phase-locked 'on' such that all cells express type 1 fimbriae. E. coli F-18(pPR633) contains essentially the same plasmid minus the fimB gene and in L-broth about 30% of the cells express type 1 fimbriae. When fed alone to streptomycin-treated mice, E. coli F-18(pPKL91) colonized the large intestine at about 10(7) cfu/g of feces. However, when simultaneously fed with E. coli F-18(pPR633) at either high (10(10) cfu), or low doses (10(4) cfu), E. coli F-18(pPKL91) was a poor colonizer dropping to a level of between 10(2) and 10(3) cfu/g of feces. When given enough time to establish the state of colonization (10 days), E. coli F-18(pPKL91) persisted in feces in high numbers despite subsequent challenge by E. coli F-18(pPR633). Moreover, although both E. coli F-18(pPR633) and E. coli F-18(pPKL91) grew equally well in cecal mucus in vitro, E. coli F-18(pPR633) traveled through a layer of cecal mucus in vitro much faster than E. coli F-18(pPKL91). Together, the data suggest that type 1 fimbriated cells are at a disadvantage in initiating the colonization state because they have difficulty entering the mucus layer of the intestine as rapidly as non-fimbriated cells. The data also point to the possible biological significance of type 1 fimbrial phase-variation in the mouse large intestine.

Phosphatidylserine found in intestinal mucus serves as a sole source of carbon and nitrogen for salmonellae and Escherichia coli.

Salmonella choleraesuis (a pig pathogen), Salmonella typhimurium (a virulent strain in mice), and three strains of Escherichia coli (including a human enterohemorrhagic strain, a human urinary tract isolate, and a human fecal isolate) grew as well in vitro utilizing the lipids derived from mouse cecal mucus as the sole source of carbon and nitrogen as they did in mouse crude cecal mucus. Further analysis of the total lipid extracts of mucus dialysates showed that the acidic lipid fraction supported growth nearly as well as the total lipid fraction. Interestingly, among the many purified acidic lipids from mucus which were tested and analyzed, including several phospholipids, only phosphatidylserine was found to support the growth of all of these enteric bacteria, including Salmonella milwaukee, a human pathogen. The possible role of growth on pure phosphatidylserine in the pathogenesis of salmonellae is discussed.

Expression of Escherichia coli F-18 type 1 fimbriae in the streptomycin-treated mouse large intestine.

Escherichia coli F-18, isolated from the feces of a healthy human, makes type 1 fimbriae and is an excellent colonizer of the streptomycin-treated mouse large intestine. Recently, it was shown that the inability to produce type 1 fimbriae had no effect on the ability of E. coli F-18 to colonize the streptomycin-treated mouse large intestine, suggesting the possibility that E. coli F-18 does not express type 1 fimbriae in vivo. However, we show here that E. coli F-18 does express type 1 fimbriae in mouse cecal mucus in vivo and, in fact, appears to express substantially more type 1 fimbriae in cecal mucus in vivo than in L broth in vitro.

Characterization and identification of a porcine small intestine mucus receptor for the K88ab fimbrial adhesin.

The ability of Escherichia coli K-12(K88ab) to adhere to immobilized porcine small intestine mucus was examined. E. coli K-12(K88ab) but not the isogenic E. coli K-12 strain was found to adhere readily to immobilized crude mucus but not to bovine serum albumin. The adhesion of E. coli K-12(K88ab) was inhibited in a specific fashion by anti-K88 antiserum. Adhesion was also inhibited by pretreatment of receptor-containing crude mucus preparations with sodium metaperiodate or proteolytic enzymes. Removal of glycolipids from crude mucus by chloroform-methanol extraction did not affect the ability of E. coli K-12(K88ab) to bind to mucus preparations. Adsorption of crude mucus preparations with K88ab fimbriae but not type 1 fimbriae resulted in the removal of K88-specific receptors. Analysis of the pelleted fimbriae-receptor complex by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, together with gel filtration chromatography of crude mucus preparations, suggest that the K88-specific receptor present in porcine small intestine mucus is a 40- to 42-kDa glycoprotein.

Neither motility nor chemotaxis plays a role in the ability of Escherichia coli F-18 to colonize the streptomycin-treated mouse large intestine.

Escherichia coli F-18, isolated from the feces of a healthy human in 1977, is an excellent colonizer of the streptomycin-treated mouse large intestine and displays normal motility and chemotaxis ability. A chemotaxis-defective derivative of E. coli F-18, E, coli F-18 CheA-, and a nonflagellated derivative, E. coli F-18 Fla-, were constructed. These strains were found to colonize the streptomycin-treated mouse large intestine as well as E. coli F-18 when mice were fed both E. coli F-18 and either the CheA- or Fla- derivative at high levels (10(10) CFU of each strain per mouse) or low levels (10(4) CFU of each strain per mouse). Furthermore, E. coli F-18 lost motility and chemotaxis ability when grown in colonic or cecal mucus in vitro despite retaining the ability to synthesize flagella. Thus, it appears that neither motility nor chemotaxis plays a role in the ability of E. coli F-18 to colonize because this strain becomes functionally nonmotile upon growth in the streptomycin-treated mouse large intestine.

Characterization of the heat shock response and identification of heat shock protein antigens of Borrelia burgdorferi.

The heat shock response of Borrelia burgdorferi B31 cells was characterized with regard to the heat shock proteins (Hsps) produced. Five to seven Hsps were detected by sodium dodecyl sulfate-gel electrophoresis and fluorography of proteins from cells labeled with [35S]methionine after shifts from 33 degrees C to 37 or 40 degrees C or from 20 degrees C to 33, 37, or 40 degrees C. Analysis of [35S]methionine-labeled Hsps by two-dimensional electrophoresis and autoradiography revealed 12 Hsps. Western immunoblot analysis with antisera to highly conserved Escherichia coli and Mycobacterium tuberculosis Hsps revealed a single 72-kilodalton (kDa) protein band that reacted with antibodies to E. coli DnaK and with antibodies to the M. tuberculosis 71-kDa Hsp homolog of E. coli DnaK. Two proteins with apparent molecular masses of 66 and 60 kDa reacted with antibodies against the M. tuberculosis 65-kDa Hsp homolog of E. coli GroEL. Human immune sera collected from patients with Lyme disease reacted with both the 66-kDa Hsp and the 60-kDa Hsp but failed to react with the 72-kDa Hsp. These data are discussed with regard to the possibility that host recognition of highly conserved epitopes of GroEL homologs of B. burgdorferi may result in autoimmune reactions causing arthritis and other pathologies.

Construction of stable cloning vectors that do not segregate from a human fecal Escherichia coli strain in the streptomycin-treated mouse large intestine.

Escherichia coli F-18 Col- was previously shown to be a poor colonizer of the streptomycin-treated mouse large intestine, relative to its parent, E. coli F-18. Prior to attempting to clone genes responsible for the colonization phenotype of E. coli F-18 into E. coli F-18 Col-, a suitable cloning vector had to be found. In this investigation, we report that the commonly used cloning vectors pBR322, pHC79, and pBR329 all segregate from E. coli F-18 Col- both when grown in L broth under conditions of nonselection (i.e., in vitro) and when fed to streptomycin-treated mice (i.e., in vivo). Insertion of the cer region (which promotes resolution of replicating plasmids into monomeric forms) into pHC79 stabilized this plasmid in E. coli F-18 Col- in vitro and in vivo. In contrast, two independent cer insertions into pBR329 did not stabilize the plasmid completely in E. coli F-18 Col- in vitro, and feeding the strain to streptomycin-treated mice resulted in rapid segregation of the plasmids in vivo. Also, stability of all three plasmids in E. coli F-18 Col- in vitro was achieved by insertion of the parB region of plasmid R1, which encodes a cell-killing protein, Hok, that is active only postsegregationally. However, as with cer, complete in vitro and in vivo stabilization was achieved only in parB constructs of pBR322 and pHC79.

Escherichia coli F-18 makes a streptomycin-treated mouse large intestine colonization factor when grown in nutrient broth containing glucose.

Escherichia coli F-18 FimA-, a type 1 fimbria-less derivative of a normal human fecal isolate, E. coli F-18, has previously been shown to be as good a colonizer of streptomycin-treated mouse large intestine as its parent, suggesting that type 1 fimbriae are not necessary in this process. In this study it was found that when E. coli F-18 FimA- was grown standing overnight at 37 degrees C in nutrient broth, it remained uniformly suspended; however, when grown in nutrient broth containing 1% (wt/wt) D-glucose, it settled to the bottom of culture tubes. Settling was associated with the formation of clumps (microcolonies) of more than 10 cells each. The effect of glucose could be partially reversed by growing E. coli F-18 FimA- in nutrient broth containing 1% D-glucose supplemented with cyclic AMP (greater than or equal to 1 mM). A reduced-settling mutant of E. coli F-18 FimA-, E. coli F-18 FimA- Set-, selected after Tn5 mutagenesis, was found to be a poor colonizer of streptomycin-treated mouse large intestine when fed to mice simultaneously with the parent strain. These results suggest that glucose-induced settling is, at least in part, regulated in a way related to catabolite repression and that the ability of E. coli F-18 FimA- to form microcolonies plays an important role in its ability to colonize streptomycin-treated mouse large intestine.