A note on symbiosis of Legionella pneumophila and Tatlockia micdadei with human respiratory flora

Sixteen micro-organisms, representing clinically important respiratory microflora, were tested for their ability to stimulate the growth of Legionella pneumophila and Tatlockia micdadei in nutritionally-deficient agar media. Nutritional symbiosis, indicated by the appearance of satellite colonies of L. pneumophila or T. micdadei , was observed for H. influenzae and N. meningitidis. This interaction between normal flora and pathogens of the respiratory tract may have clinical relevance in the pathogenesis of Legionnaires' disease and Pittsburgh pneumonia.     

Characterization of the beta-lactamases of six species of Legionella.

The beta-lactamases of six Legionella species were characterized by isoelectric focusing, gel filtration, and substrate profiles. Fifteen strains of L. bozemanii, L. dumoffii, L. gormanii, L. longbeachae, and L. pneumophila produced beta-lactamases active against nitrocefin. L. micdadei enzymes previously reported to be beta-lactamase negative caused a very slow pH-dependent breakdown of nitrocefin and degraded penicillin G at high substrate concentrations. The bioassay revealed predominantly penicillinase activity for all species except L. micdadei, which had no activity in this assay. The apparent molecular weights of enzymes of L. bozemanii, L. gormanii, and L. pneumophila were in the range of 15,000 to 32,000, and those of L. micdadei and L. longbeachae were greater than 250,000. The isoelectric focusing of extracts of Legionella strains in polyacrylamide gels showed beta-lactamase types specific for species (L. bozemanii, L. gormanii, and L. pneumophila) and serotype (L. pneumophila). It demonstrated four different beta-lactamase types in L. pneumophila and revealed close relationships among L. pneumophila serotypes 1, 3, and 6. L. pneumophila enzymes formed band patterns only in polyacrylamide gels containing 6 M urea, whereas L. dumoffii and L. longbeachae enzymes did not form bands in any of the gels. None of the band patterns resembled those of known plasmid-mediated beta-lactamases. These experiments suggest that isoelectric focusing of chromosomal beta-lactamases may be a valuable tool for taxonomic studies.     

Rapid detection of the gene of Legionella pneumophila using the fluorescence polarization with the asymmetric PCR.

We attempted the rapid detection method of Legionella pneumophila by the asymmetric PCR and the fluorescence polarization. Eleven extracted DNAs from L. pneumophila serogroup 1 to approximately 6, L. bozemanii, L. dumoffii, L. gormanii, L. micdadei, and Pseudomonas aeruginosa were amplified by asymmetric PCR, and the polarization of those products were measured. Only the polarization of L. pneumophila serogroup 1 to approximately 6 rose within a few minutes after the beginning of measurement. The sensitivity to L. pneumophila using this method was 10(3) cells.     

Cytolytic and phospholipase C activity in Legionella species

To examine one possible mechanism of damage to leucocytes and tissue cells in legionellosis, seven species of Legionella were examined for cytolytic activity and for elaboration of phospholipase C, an enzyme that can damage mammalian cell membranes. Cytolysis was assessed using erythrocytes in agar. Phospholipase C was assayed by release of p-nitrophenol from p-nitrophenylphosphorylcholine and of tritiated phosphorylcholine from L-alpha-dipalmitoyl-[choline-methyl-3H]phosphatidylcholine. L. pneumophila, L. bozemanii, L. micdadei, L. dumoffii, L. gormanii, L. longbeachae and L. jordanis all lysed dog red blood cells, which have a high ratio of membrane phosphatidylcholine to sphingomyelin. The same strains hydrolysed varying amounts of p-nitrophenylphosphorylcholine; L. bozemanii exhibited the greatest activity. L. pneumophila, L. bozemanii, L. dumoffii, L. longbeachae and L. jordanis, but not L. micdadei, released tritiated phosphorylcholine from labelled substrate. These results indicate that several species of Legionella possess cytolytic capability; exotoxins with activity may play a role.     

Temperature-dependent expression of flagella in Legionella

Legionella pneumophila, the causative agent of Legionnaires' disease, was analysed by electron microscopy for production of surface structures. Crystalline surface (S-) layers and fimbriae were not detected, but monotrichous flagellation was seen. Polyclonal antibodies specific for the 47 kDa flagellin subunit of L. pneumophila Philadelphia I were used in Western blots to confirm the presence of flagella subunits in various L. pneumophila strains tested, but the antiserum also reacted with flagellin subunits of L. micdadei, L. hackelia [serogroup (SG) 1 and SG2] and L. longbeachae (SG2). Flagellation of Legionellae was shown to be temperature regulated. When the growth temperature of virulent and avirulent of strain L. pneumophila Philadelphia I was shifted from 30 degrees C to either 37 or 41 degrees C, a decrease in the percentage of flagellated bacteria within the population was observed.     

Vaccination with the major secretory protein of Legionella induces humoral and cell-mediated immune responses and protective immunity across different serogroups of Legionella pneumophila and different species of Legionella.

In a previous study, we demonstrated that immunization of guinea pigs with the major secretory protein (MSP) of Legionella pneumophila, serogroup 1 induced humoral and cell-mediated immune responses to MSP and protective immunity against lethal aerosol challenge with this serogroup of L. pneumophila. Although serogroup 1 L. pneumophila cause most cases of Legionnaires' disease, other serogroups of L. pneumophila and species of Legionella cause many cases. In this study, we have examined if immunization with MSP induces humoral and cell-mediated immune responses and protective immunity across different serogroups of L. pneumophila and species of Legionella. By immunoblot analysis, MSP from L. pneumophila serogroup 1 (Lp1 MSP), L. pneumophila serogroup 6 (Lp6 MSP), and Legionella bozemanii (Lb MSP) shared common epitopes recognized by guinea pig anti-Lp1 MSP antiserum. These MSP molecules, however, were not identical as they had different apparent m.w. Immunization of guinea pigs with MSP induced strong cell-mediated immune responses across the different serogroups and species, as indicated by splenic lymphocyte proliferation and cutaneous delayed-type hypersensitivity in response to both homologous and heterologous MSP. Immunization with MSP induced strong protective immunity across two serogroups of L. pneumophila; overall, 9 survived aerosol challenge with L. pneumophila serogroup 1 compared to 0 of 12 (0%) sham-immunized control animals (p = 3 x 10(-4), Cochran-Mantel-Haenzel chi 2 statistic for pooled data). Immunization with MSP also induced protective immunity across species of Legionella but protection was species-specific. Whereas immunization with Lb MSP induced protective immunity against L. pneumophila, neither immunization with Lp1 MSP nor immunization with Lb MSP induced protective immunity against L. bozemanii, which produces MSP. Not surprisingly, immunization with MSP did not induce protective immunity against MSP-negative Legionella micdadei. In the case of both L. bozemanii and L. micdadei, immunization with a sublethal dose did confer protective immunity to aerosol challenge indicating that these species do contain immunoprotective components. This study demonstrates that immunization with MSP induces humoral and cell-mediated immune responses across different serogroups of L. pneumophila and species of Legionella, but that the capacity of MSP immunization to induce protective immunity is species-specific. Nevertheless, an MSP vaccine has the potential to induce protective immunity against the great majority of cases of Legionnaires' disease.     

Tatlockia micdadei (Pittsburgh pneumonia agent) growth kinetics may explain its infrequent isolation from water and the low prevalence of Pittsburgh pneumonia.

Sediment and indigenous microflora taken from water distribution systems has been shown to promote the survival of Legionella pneumophila. The effect of sediment and indigenous microflora on Tatlockia micdadei (Pittsburgh pneumonia agent, PPA) was evaluated by growth curve experiments. Symbiosis between PPA and environmental bacteria was demonstrated by satellitism experiments. Unlike L. pneumophila, the concentration of PPA remained stationary in test tube suspensions containing both microflora and sediment. The difference in the ecology between the two organisms may explain the infrequent environmental recovery of PPA and, ultimately, the epidemiologic differences between Legionnaires disease and Pittsburgh pneumonia.     

Tatlockia micdadei (Pittsburgh pneumonia agent) growth kinetics may explain its infrequent isolation from water and the low prevalence of Pittsburgh pneumonia

Sediment and indigenous microflora taken from water distribution systems has been shown to promote the survival of Legionella pneumophila. The effect of sediment and indigenous microflora on Tatlockia micdadei (Pittsburgh pneumonia agent, PPA) was evaluated by growth curve experiments. Symbiosis between PPA and environmental bacteria was demonstrated by satellitism experiments. Unlike L. pneumophila, the concentration of PPA remained stationary in test tube suspensions containing both microflora and sediment. The difference in the ecology between the two organisms may explain the infrequent environmental recovery of PPA and, ultimately, the epidemiologic differences between Legionnaires disease and Pittsburgh pneumonia.     

Monoclonal antibodies to Legionella pneumophila cytolysin

The fusion of spleen cells, taken from BALB/c mice immunized with the purified preparation of L. pneumophila cytolysin, with cells Sp2/0 and NP has been carried out. As a result, hybridoma cells producing IgG1, IgG3 and IgM antibodies to this protein have been obtained. All monoclonal antibodies (McAb) thus obtained react with L. pneumophila strain lysates in the precipitation test, while IgG3 and IgM antibodies react with erythrocyte diagnostic agents prepared from the lysate of L. pneumophila cells in the hemagglutination test. In the Western blot assay, McAb react with the 37 KD protein (cytolysin) and a number of other proteins from L. pneumophila cultures and L. pneumophila cell lysate, but do not react with the species-specific protein with a molecular weight of 29 KD, contained in the outer membrane of L. pneumophila, as well as with other species: L. bozemanii, L. dumoffii, L. longbeachae, L. micdadei. The possibility of using these McAb conjugated with FITC and peroxidase for the rapid diagnosis of Legionella infection is shown.     

Development and use of the assay for Legionella pneumophila detection based on fluorescent real-time/endpoint polymerase-chain reaction

The aim of the work was to develop a PCR-based assay for detection of L. pneumophila and L. micdadei in environmental samples as well as in clinical samples from low respiratory tract and to assess its analytic characteristics. The assay was used during investigation of the outbreak developed in July 2007 in town Verkhnyaya Pyshma (Sverdlovsk region). Polymerase-chain reaction (PCR)with fluorescent detection,sequencing and cloning of DNA fragments were used. Developed assay based on the PCR with fluorescent real-time/ endpointdetection is able to detect L. pneumophila in clinical and environmental samples and to quantify amount of bacterial DNA in water. Specificity of analysis (100%) was assessed using the panel of bacterial strains and samples from healthy individuals. Analytic sensitivity of assay and quantitation limit was 1000 GU in 1 ml. Sensitivity of the assay of artificially contaminated biological samples was 1000 bacteria in 1 ml. During outbreak investigation L. pneumophila DNAwas detected in 4 lung samples from 4 fatal cases, from 1 of 2 sputum samples, 1 of 2 bronchoalveolar lavage samples with X-ray confirmed pneumonia. Legionella's DNA was found in samples from cooling towers, central hot water supply as well as from showerheads in apartments of 3 patients. Fountain and drinking water samples were PCR-negative. Specificity of PCR-positive results was confirmed by sequencing. Use of the assay during outbreak in- vestigation allowed to confirm the diagnosis in fatal cases and quickly identify the possible source of infection.     

Abundance and distribution of Legionellaceae in Puerto Rican waters

Waters in marine and freshwater areas of Puerto Rico were analyzed for the presence of Legionella spp. by direct fluorescent antibody assay with guinea pig confirmation. Several species, including L. bozemanii, L. dumoffii, L. gormanii, L. longbeachae, L. micdadei, and L. pneumophila, were widely distributed among all sites. Legionellaceae, including L. pneumophila, were found in high densities in water collected in the rain forest from epiphytes in trees 30 ft. (about 9.25 m) above the ground. Both interspecific and intersite variations were significant. L. pneumophila was the most abundant species at all sites, with average densities of 10(4) cells ml-1, very close to the range which is potentially pathogenic for humans. Densities of L. pneumophila were highest in sewage-contaminated coastal waters. These are the highest densities of Legionella spp. ever reported for marine habitats. Densities of L. pneumophila were positively correlated with concentrations of sulfates, phosphates, and pH. A survey of 88 fatal atypical pneumonia cases at a Puerto Rico hospital showed that 15% of the patients had L. pneumophila infections. This study establishes L. pneumophila as a relatively common cause of atypical pneumonia in Puerto Rico and suggests natural aquatic habitats as possible sources or reservoirs of pathogenic Legionella spp. in the tropics.     

Multiplication of Legionella spp. in tap water containing Hartmannella vermiformis

A model was developed to study the multiplication of various Legionella spp. in tap water containing Hartmannella vermiformis. Tap water cultures prepared with the following components were suitable for the multiplication studies: Legionella spp., 10(3) CFU/ml; H. vermiformis, 10(4.4) cysts per ml; and killed Pseudomonas paucimobilis, 10(9) cells per ml. Cocultures were incubated at 37 degrees C for at least 1 week. The following legionellae multiplied in tap water cocultures in each replicate experiment: L. bozemanii (WIGA strain), L. dumoffii (NY-23 and TX-KL strains), L. micdadei (two environmental strains), and L. pneumophila (six environmental strains and one clinical isolate). Growth yield values for these strains were 0.6 to 3.5 log CFU/ml. Legionellae which did not multiply in replicate cocultures included L. anisa (one strain), L. bozemanii (MI-15 strain), L. micdadei (a clinical isolate), L. longbeachae, (one strain), and L. pneumophila (Philadelphia 1 strain). L. gormanii and an environmental isolate of L. pneumophila multiplied in only one of three experiments. None of the legionellae multiplied in tap water containing only killed P. paucimobilis. The mean growth yield (+/- standard deviation) of H. vermiformis in the cocultures was 1.2 +/- 0.1 log units/ml. H. vermiformis supports multiplication of only particular strains of legionellae, some of which are from diverse origins.     

Infection of Tetrahymena pyriformis by Legionella longbeachae and other Legionella species found in potting mixes.

All Legionella longbeachae strains, both serogroups of L. bozemanii, and three strains of L. anisa reproducibly infected washed Tetrahymena pyriformis at 30 degrees C. L. pneumophila serogroup 1 strains infected T. pyriformis less reproducibly than did L. longbeachae. Low-level concentrations of nutrients in cocultures inhibited infection. Four L. micdadei strains and L. anisa ATCC 35292 failed to infect T. pyriformis.     

Abundance and distribution of Legionellaceae in Puerto Rican waters.

Waters in marine and freshwater areas of Puerto Rico were analyzed for the presence of Legionella spp. by direct fluorescent antibody assay with guinea pig confirmation. Several species, including L. bozemanii, L. dumoffii, L. gormanii, L. longbeachae, L. micdadei, and L. pneumophila, were widely distributed among all sites. Legionellaceae, including L. pneumophila, were found in high densities in water collected in the rain forest from epiphytes in trees 30 ft. (about 9.25 m) above the ground. Both interspecific and intersite variations were significant. L. pneumophila was the most abundant species at all sites, with average densities of 10(4) cells ml-1, very close to the range which is potentially pathogenic for humans. Densities of L. pneumophila were highest in sewage-contaminated coastal waters. These are the highest densities of Legionella spp. ever reported for marine habitats. Densities of L. pneumophila were positively correlated with concentrations of sulfates, phosphates, and pH. A survey of 88 fatal atypical pneumonia cases at a Puerto Rico hospital showed that 15% of the patients had L. pneumophila infections. This study establishes L. pneumophila as a relatively common cause of atypical pneumonia in Puerto Rico and suggests natural aquatic habitats as possible sources or reservoirs of pathogenic Legionella spp. in the tropics.     

Multiplication of Legionella spp. in tap water containing Hartmannella vermiformis.

A model was developed to study the multiplication of various Legionella spp. in tap water containing Hartmannella vermiformis. Tap water cultures prepared with the following components were suitable for the multiplication studies: Legionella spp., 10(3) CFU/ml; H. vermiformis, 10(4.4) cysts per ml; and killed Pseudomonas paucimobilis, 10(9) cells per ml. Cocultures were incubated at 37 degrees C for at least 1 week. The following legionellae multiplied in tap water cocultures in each replicate experiment: L. bozemanii (WIGA strain), L. dumoffii (NY-23 and TX-KL strains), L. micdadei (two environmental strains), and L. pneumophila (six environmental strains and one clinical isolate). Growth yield values for these strains were 0.6 to 3.5 log CFU/ml. Legionellae which did not multiply in replicate cocultures included L. anisa (one strain), L. bozemanii (MI-15 strain), L. micdadei (a clinical isolate), L. longbeachae, (one strain), and L. pneumophila (Philadelphia 1 strain). L. gormanii and an environmental isolate of L. pneumophila multiplied in only one of three experiments. None of the legionellae multiplied in tap water containing only killed P. paucimobilis. The mean growth yield (+/- standard deviation) of H. vermiformis in the cocultures was 1.2 +/- 0.1 log units/ml. H. vermiformis supports multiplication of only particular strains of legionellae, some of which are from diverse origins.     

Immunoblot cross-reactions among Rickettsia, Proteus spp. and Legionella spp. in patients with Mediterranean spotted fever

Abstract Sera from patients suffering from Mediterranean spotted fever (i.e. an infection due to Rickettsia conorii ) were studied by immunoblot to investigate cross-reactivity. A prevalence of IgM antibodies to Proteus OX 19, Proteus 0X 2, to the Rickettsia typhus group, to Legionella pneumophila serovars 4 and 5, to L. bozemanii Wiga and to L. micdadei Tatlock was found. Western blot confirmed that the antibodies were directed against the lipopolysaccharide as demonstrated by proteinase K digestion of the antigens. Cross-adsorptions showed that there is a common cross-reacting epitope among L. bozemanii Wiga, R. typhi and Proteus OX 19 but cross-reacting antibodies to L. micdadei and OX 2 were distinct and independent. This IgM cross-reaction could lead to a misdiagnosis.     

Glycine-containing selective medium for isolation of Legionellaceae from environmental specimens.

Glycine, at a final concentration of 0.3%, has been shown to be an excellent selective agent for the isolation of Legionellaceae. Stock cultures of Legionella pneumophila were not inhibited on buffered charcoal-yeast extract agar containing the amino acid. Among the other Legionellaceae tested, only one of two strains of L. dumoffii and two of six strains of L. micdadei were appreciably inhibited. This medium permitted the isolation of L. pneumophila from environmental specimens with marked inhibition of many non-Legionellaceae bacteria. The selectivity of the medium was subsequently improved by the incorporation of vancomycin (5 microgram/ml) and polymyxin B (100 U/ml). This selective medium, glycine-vancomycin-polymyxin B agar, should facilitate the recovery of Legionellaceae from environmental sources.     

Biofilms as major sources of Legionella spp. in hydrothermal areas and their dispersion into stream water

Abstract Water and biofilms from two hydrothermal areas in central Portugal, and one hydrothermal area in New Mexico, USA, were examined for Legionella spp. In general, Legionella spp were isolated in higher numbers from biofilms than from water, although one biofilm with a temperature of 50°C, did not yield isolates of these organisms. In one area L. pneumophila serogroup (sg) 3 constituted the major population in the thermal discharge by the stream and the biofilm below it; however, L. pneumophila sg 1 was predominant in the sediments of the stream bed with minor thermal springs below the main discharge and in the water downstream. No Legionellae were isolated from water upstream of the hydrothermal area indicating that the thermal area was the source of the organisms in the stream water. In the other two hydrothermal areas, L. pneumophila sg 1 constituted the major population isolated, whereas L. pneumophila sg 3 was absent or isolated in low numbers. Isolates of L. micdadei were also recovered from one hydrothermal area, while ‘ L. londoniensis ’ was isolated from another.     

A note on the temperature tolerance of Legionella

A strain of Legionella pneumophila serogroup 1 isolated from the environment had a decimal reduction time in water at 50 degrees C (D50) of 111 min, a D54 of 27 min and a D58 of 6 min. There was little loss of viability at 46 degrees C. Other environmental organisms, a Pseudomonas sp., a Micrococcus sp. and a coliform survived less well at these temperatures. A species of Sarcina had a survival time greater than the L. pneumophila at all the temperatures tested. Other strains of legionellas were tested at 50 degrees C and decimal reduction times calculated. These ranged from 80 min for another strain of L. pneumophia serogroup 1 to 216 min for L. bozemannii . Legionella micdadei did not survive well at 50 degrees C.