Abstract
The standard sequence-based method for the typing of Legionella pneumophila serogroup 1 strains was extended by using the gspA and neuA alleles. The use of neuA as a seventh allele for typing significantly increased the index of discrimination calculated for a panel of unrelated strains (from 0.932 to 0.963) and subdivided some known large common complexes (e.g., 1,4,3,1,1,1). This modification to the standard method is proposed as the method of choice in the epidemiological investigation of L. pneumophila infections.
Members of the genus Legionella are gram-negative bacteria and normally occupy natural aquatic environments, where they survive as intracellular parasites of protozoa. Human infections occur as sporadic or epidemic cases of disease that may be acquired from different environmental sources, such as warm water supplies, cooling towers, and evaporative condensers. They are mainly caused by the species Legionella pneumophila and mostly by L. pneumophila serogroup (sg) 1 strains (7, 10).
In order to detect the source of the infection as soon as possible, several molecular typing techniques have been used to characterize L. pneumophila strains (4, 8, 10, 12). Multilocus sequence typing is a powerful tool that is used to discriminate clonal groups within several bacterial species (14). Recently, a scheme for the sequence-based typing (SBT) of L. pneumophila that uses the sequences of six genes was described (9). It is now available through the website of the European Working Group on Legionella Infections (EWGLI) (www.ewgli.org). An SBT profile comprises a string of numbers comprising the number of individual alleles of the genes flaA, pilE, asd, mip, mompS, and proA separated by commas. The available SBT data for L. pneumophila sg 1 suggest that some of the prevalent SBT profiles, e.g., 1,4,3,1,1,1, are heterogeneous by monoclonal antibody (MAb) subgrouping (9) and/or by pulsed-field gel electrophoresis (2). Thus, it might be speculated that some SBT profiles contain several different types that cannot be distinguished by the standard EWGLI SBT scheme. To test this hypothesis we investigated whether the use of additional genes would enhance the discriminatory power of the standard SBT scheme. The epidemiological concordance (E) and stability (S) (17) of these new alleles were also examined.
We investigated the panel of 79 unrelated L. pneumophila sg 1 strains of the European Union Legionella (EUL) culture collection (8, 9). Additional strains from the collection of one of the authors (P.C.L.) (n = 16 unrelated strains) and three reference strains (strains Philadelphia-1T, Lens, and Paris) were included to explore the potential of subdividing some of the larger six-allele profiles, such as 1,4,3,1,1,1; and strains from the EUL culture collection (n = 15 related strains, n = 6 strains in the stability panel) were used to assess E and S (Table 1) (8, 9). MAb typing was performed as described previously (10). SBT was performed according to the EWGLI standard scheme (9). In a similar fashion we amplified and analyzed selected regions of the genes encoding N-acylneuraminate cytidylyl transferase (neuA) (13) and a general stress protein (gspA) (1). Primers gspA-up (5′-CCT ATC CGG CCT ATG ACA-3′) and gspA-do (5′-CGT GGT TTC GCT TCT TCC-3′) and primers neuA-up (5′-CCG TTC AAT ATG GGG CTT CAG-3′) and neuA-do (5′-CGA TGT CGA TGG ATT CAC TAA TAC-3′) were designed by using previously published sequences (1, 13).
TABLE 1.
Unrelated Legionella pneumophila sg 1 strains investigated by additional sequencing of neuA and gsp alleles
| Strain | MAb subgroupb | Standard EWGLI SBTc | No. of alleles
|
|
|---|---|---|---|---|
| neuA | gsp | |||
| EUL074a | Philadelphia | 1,4,1,1,14,9 | 1 | 13 |
| EUL001, EUL003, EUL037, EUL042, EUL043, EUL060, EUL067, EUL082, EUL084, EUL085, EUL088, EUL093, EUL119, L02-570, Ulm145432, Char6297/2000, W03-685, Paris, L01-127, L04-541 | Oxford (2), OLDA (10) Philadelphia (8) | 1,4,3,1,1,1 | 1 | 10 |
| EUL017, EUL112, L03-610 | OLDA (2), Philadelphia | 1,4,3,1,1,1 | 6 | 10 |
| München 1 | Benidorm | 1,4,3,1,1,1 | 9 | 10 |
| EUL013, EUL014, EUL016 | Benidorm (3) | 1,4,3,1,1,1 | 14 | 10 |
| EUL117 | Benidorm | 1,4,3,1,1,1 | 15 | 10 |
| EUL104, EUL110 | Oxford | 1,6,3,1,1,1 | 1 | 10 |
| EUL053 | OLDA | 1,6,3,1,14,1 | 1 | 13 |
| EUL038 | OLDA | 1,10,3,1,1,1 | 1 | 9 |
| EUL072 | Philadelphia | 1,10,19,1,9,4 | 1 | 7 |
| EUL029 | Knoxville | 2,2,18,15,2,1 | 6 | 10 |
| EUL036 | Knoxville | 2,3,3,15,2,6 | 6 | 1 |
| EUL026 | OLDA | 2,3,6,10,2,1 | 6 | 2 |
| EUL004, EUL008, EUL028, EUL041, L01-403 | Allentown (4), Philadelphia | 2,3,9,10,2,1 | 6 | 1 |
| EUL049 | Knoxville | 2,3,18,15,2,1 | 6 | 6 |
| EUL087 | Knoxville | 2,4,3,10,9,4 | 9 | 7 |
| EUL111 | Benidorm | 2,6,17,15,12,8 | 6 | 4 |
| EUL018 | Bellingham | 2,6,21,12,12,8 | 11 | 4 |
| EUL007 | Allentown | 2,10,9,13,2,5 | 6 | 2 |
| EUL032 | Benidorm | 2,10,18,10,2,1 | 9 | 2 |
| L92-448, L03-518 | Philadelphia, Knoxville | 3,4,1,1,1,9 | 1 | 13 |
| EUL118, Philadelphia-1T | Philadelphia | 3,4,1,1,14,9 | 1 | 13 |
| EUL069, EUL073 | Philadelphia (2) | 3,4,1,1,14,9 | 11 | 13 |
| EUL030 | France | 3,4,1,14,14,9 | 11 | 13 |
| EUL033 | France | 3,6,1,14,14,9 | 11 | 13 |
| EUL020 | Benidorm | 3,10,1,3,14,9 | 1 | 13 |
| HannP9 | Knoxville | 3,10,1,3,14,9 | 11 | 13 |
| L03-572 | Knoxville | 3,10,1,5,14,9 | 11 | 13 |
| EUL019, EUL097, EUL098 | Knoxville (3) | 3,10,1,3,14,9 | 11 | 13 |
| EUL052 | Philadelphia | 3,10,3,1,14,9 | 11 | 10 |
| EUL051 | Benidorm | 3,10,15,3,21,7 | 9 | 7 |
| EUL099 | Bellingham | 3,13,1,25,14,9 | 6 | 13 |
| EUL050 | Benidorm | 4,6,11,3,11,12 | 9 | 13 |
| EUL006 | Benidorm | 4,7,9,3,11,12 | 9 | 9 |
| EUL027, EUL039, EUL075, EUL105, EUL116, EUL120, Vie53 | Benidorm (6) | 4,7,11,3,11,12 | 9 | 9 |
| EUL025 | Allentown | 4,8,11,10,10,12 | 2 | 9 |
| EUL103 | OLDA | 5,1,22,26,6,10 | 12 | 8 |
| EUL068, EUL086 | Benidorm (2) | 5,1,22,5,6,10 | 15 | 8 |
| EUL048 | Bellingham | 5,2,22,27,6,10 | 12 | 8 |
| EUL031, EUL070 | Allentown (2) | 5,10,22,15,6,2 | 6 | 8 |
| EUL055 | OLDA | 6,4,3,1,1,1 | 1 | 10 |
| EUL063 | Knoxville | 6,4,14,12,2,3 | 6 | 10 |
| EUL066 | Knoxville | 6,10,14,10,2,1 | 6 | 7 |
| L01-472 | Knoxville | 6,10,15,28,9,14 | 6 | 7 |
| EUL083 | Benidorm | 6,10,15,24,17,14 | 6 | 3 |
| EUL002 | Knoxville | 6,10,19,3,19,4 | 9 | 13 |
| EUL081, EUL092 | Bellingham (2) | 6,10,21,12,9,4 | 11 | 3 |
| EUL101 | Benidorm | 7,6,17,3,13,11 | 9 | 4 |
| EUL100, EUL102 | Bellingham (2) | 7,6,17,3,13,11 | 11 | 4 |
| EUL054, EUL071 | France, Allentown | 8,10,3,15,18,1 | 6 | 9 |
| EUL091 | Bellingham | 9,6,3,10,22,15 | 11 | 4 |
| EUL040 | Philadelphia | 11,14,16,1,15,13 | 6 | 4 |
| Lens, L03-407 | Benidorm (2) | 12,9,26,5,26,17 | 15 | 16 |
For the 79 unrelated EUL strains, the allelic profiles of the six genes were published previously (9). For the strains isolated in Germany as well as other EUL strains, all sequences were determined anew. The allele numbers obtained by the standard protocol and the additional gene, neuA, will be available through the website of the EWGLI (www.ewgli.org) (for review purposes, see http://www.hpa-bioinfotools.org.uk/legionella/php/sbt_query1_neu.php). GspA allele 16 was taken from the genome sequence of strain Lens (5), and gspA allele 17 is from the original description (1).
For the two additional genes, the lengths of the fragments analyzed were 354 bp (nucleotide positions 229 to 583) for neuA and 225 bp for gspA (nucleotide positions 61 to 286). The numbers of silent and nonsilent mutations and the percentage of nucleotide substitutions were 20.0, 8.0, and 6.6, respectively, for neuA and 21.0, 6.0, and 10.7, respectively, for gspA. This is in the range of values for other L. pneumophila genes (3, 9, 15). Altogether, we determined 15 allele variants for neuA and 16 for gspA.
The previously described panel of 79 unrelated L. pneumophila sg 1 strains (8) was used to estimate the Hunter-Gaston diversity index (11), with precision expressed as 95% confidence intervals (CIs) by using the V-DICE tool (http://www.hpa-bioinfotools.org.uk/cgi-bin/DICI/DICI.pl). The Dice coefficient (D) values for the single genes neuA and gspA were 0.800 (95% CI, 0.783 to 0.816) and 0.836 (95% CI, 0.814 to 0.859), respectively; these values are approximately in the range of values for other genes (3, 9, 13). The use of the EWGLI standard (six-gene) SBT scheme yielded a D value of 0.932 (95% CI, 0.913 to 0.951) for the panel of 79 unrelated strains, but with the addition of the neuA gene this increased to 0.963 (95% CI, 0.952 to 0.974), which is above the value recommended for a good epidemiological typing system (17). Although gspA was heterogeneous in the strains tested, the additional use of this gene did not enhance the discrimination of the SBT scheme; i.e., the D value remained 0.932 (95% CI, 0.913 to 0.951). However, it is noteworthy that MAb subgrouping further increased the D value in all cases to 0.974 (95% CI, 0.968 to 0.980) in the standard scheme and to 0.98 (95% CI, 0.974 to 0.985) when neuA was included.
The addition of neuA to the standard scheme allowed the differentiation of the complex 1,4,3,1,1,1 into five subgroups (Table 1). The fact that the MAb subgrouping appears to correlate to some extent with the subgrouping provided by SBT might be a further indication that the strains within this standard SBT type are really dissimilar, albeit closely related. Thus, within the 1,4,3,1,1,1 complex, neuA alleles 1 and 6 were found exclusively in strains of the MAb subgroups OLDA, Oxford, and Philadelphia. The MAb subgroups Philadelphia and OLDA might be considered closely related because the loss of the lag-1 gene resulted in a switch from the Philadelphia subgroup to the OLDA subgroup (4, 18). In contrast, neuA alleles 9, 14, and 15 were found only in strains belonging to MAb subgroup Benidorm within this same complex. The neuA allele also further divided two other six-allele complexes, 7,6,17,3,13,11 and 3,10,1,3,14,9, into two further types each, each of which in turn corresponded to a different MAb subgroup (Table 1). Based on the use of the neuA allele, complex 3,4,1,1,14,9 determined by SBT could also be divided into two subgroups within the same MAb subgroup, Philadelphia (Table 1). Conversely, strains L03-518 and L92-448 were different by MAb typing but indistinguishable by SBT when up to eight genes were used. Epidemiological concordance (E = 1) was demonstrated by using six epidemiologically related sets, and stability was demonstrated (S = 1) by using the stability panel comprising Corby strain variants (Table 2).
TABLE 2.
Allelic profiles of L. pneumophila sg 1 isolates belonging to epidemiologically related sets and stability panel
| Strain set and strain | MAb subgroup | Standard EWGLI SBTc | No. of alleles
|
|
|---|---|---|---|---|
| neuA | gsp | |||
| Epidemiologically related setsa | ||||
| EUL003, EUL009 | Philadelphia | 1,4,3,1,1,1 | 1 | 10 |
| EUL037, EUL044 | Philadelphia | 1,4,3,1,1,1 | 1 | 10 |
| EUL019, EUL023 | Knoxville | 3,10,1,3,14,9 | 11 | 13 |
| EUL048, EUL056 | Bellingham | 5,2,22,27,6,10 | 12 | 8 |
| EUL073, EUL079 | Philadelphia | 3,4,1,1,14,9 | 11 | 13 |
| EUL097, EUL107 | Knoxville | 3,10,1,3,14,9 | 11 | 13 |
| EUL140, EUL141, EUL142 | Knoxville | 12,8,11,5,20,12 | 5 | 15 |
| Stability panel,b strains EUL135, EUL136, EUL137, EUL138, EUL139, EUL147 | Knoxville/Denver | 6,10,15,28,9,14 | 6 | 7 |
Related patients and environmental isolates from clusters or epidemics of legionellosis.
Variants of strain Corby that differ in their ability to multiply in amoebae and macrophages showed different reactivity patterns with MAbs due to a mutation in a lipopolysaccharide synthesis gene or that were resistant to rifampin (9).
From reference 9.
Recently, sequence typing, which identifies variations in the nucleotide sequences of internal fragments of selected genes, has been established for certain bacterial species (6). In a recently published study (9), the utility of the SBT protocol for epidemiological investigations of L. pneumophila sg 1 infections was described. In the present study, we could significantly enhance the discriminatory power of the standard scheme using neuA as the seventh target. Similar results were recently published for Streptococcus pneumoniae (16). Our data suggest that the extended seven-gene allelic profiles more accurately reflect the diversity between closely related L. pneumophila sg 1 strains than the standard scheme does. It remains an open question and it needs to be experimentally proven whether the use of additional genes, e.g., the icm and dot loci, that have been shown to be highly heterogeneous (15) can further discriminate the presently established SBT types. The seven-gene SBT scheme (including neuA) of L. pneumophila produces robust, epidemiologically concordant, and highly discriminatory data that can be easily exchanged between laboratories. The authors propose that the standard EWGLI SBT scheme be extended to include the neuA allele, and this proposal is now under review by the members of EWGLI. It is anticipated that this revised scheme with seven targets will then be available via the Internet (http://www.ewgli.org). It is not anticipated that the scheme will change again.
Nucleotide sequence accession numbers.
The sequences of the alleles of the gspA gene have been deposited in the EMBL nucleotide sequence database under accession numbers AM490070 to AM490085.
Acknowledgments
Legionella strains were kindly supplied by C. Aepinus (Würzburg), E. Dinger (Wernigerode), W. Ehret (Augsburg), E. Halle (Berlin), D. Jonas (Freiburg), R. Kämmerer (Lübeck), M. Maiwald (Heidelberg), W. Matthys (Münster), R. Pfüller (Berlin), R. Marre (Ulm), and C. Schoerner (Erlangen). We are grateful to Jutta Paasche, Kerstin Seeliger, Sigrid Gäbler, Ines Wolf, Silke Rachlitz, and Susanne Thomas for technical assistance; Steve Platt, Martin Edwards, and Anthony Underwood for bioinformatic support; and Tim Harrison for constructive comments on the manuscript.
This study was supported by the Federal Ministry of Education and Research of Germany Network of Competence in Medicine, CAPNetz.
Footnotes
Published ahead of print on 18 April 2007.
REFERENCES
- 1.Abu, K., and N. C. Engleberg. 1994. Cloning and molecular characterization of a Legionella pneumophila gene induced by intracellular infection and by various in-vitro stress conditions. Mol. Microbiol. 13:243-251. [DOI] [PubMed] [Google Scholar]
- 2.Amemura-Maekawa, J., F. Kura, B. Chang, and H. Watanabe. 2005. Legionella pneumophila serogroup 1 isolates from cooling towers in Japan form a distinct genetic cluster. Microbiol. Immunol. 49:1027-1033. [DOI] [PubMed] [Google Scholar]
- 3.Aurell, H., P. Farge, H. Meugnier, M. Gouy, F. Forey, G. Lina, F. Vandenesch, J. Etienne, and S. Jarraud. 2005. Clinical and environmental isolates of Legionella pneumophila serogroup 1 cannot be distinguished by sequence analysis of two surface protein genes and three housekeeping genes. Appl. Environ. Microbiol. 71:282-289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bernander, S., K. Jacobson, J. H. Helbig, P. C. Lück, and M. Lundholm. 2003. A hospital-associated outbreak of Legionnaires' disease caused by Legionella pneumophila serogroup 1 is characterized by stable genetic fingerprinting but variable monoclonal antibody patterns. J. Clin. Microbiol. 41:2503-2508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cazalet, C., C. Rusniok, H. Brüggemann, N. Zidane, A. Magnier, L. Ma, M. Tichit, S. Jarraud, C. Bouchier, F. Vandenesch, F. Kunst, J. Etienne, P. Glaser, and C. Buchrieser. 2004. Evidence in the Legionella pneumophila genome for exploitation of host cell functions and high genome plasticity. Nat. Genet. 36:1165-1173. [DOI] [PubMed] [Google Scholar]
- 6.Cooper, J. E., and E. J. Feil. 2004. Multilocus sequence typing—what is resolved. Trends Microbiol. 12:373-377. [DOI] [PubMed] [Google Scholar]
- 7.Fields, B. S., R. F. Benson, and R. E. Besser. 2002. Legionella and Legionnaires' disease: 25 years of investigation. Clin. Microbiol. Rev. 15:506-526. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Fry, N. K., S. Alexiou-Daniel, J. M. Bangsborg, S. Bernander, M. Castellani-Pastoris, J. Etienne, B. Forsblom, V. Gaia, J. H. Helbig, D. Lindsay, P. C. Lück, C. Pelaz, S. A. Uldum, and T. G. Harrison. 1999. A multicenter evaluation of genotyping methods for the epidemiologic typing of Legionella pneumophila serogroup 1: results from a pan-European study. Clin. Microbiol. Infect. 5:462-477. [DOI] [PubMed] [Google Scholar]
- 9.Gaia, V., N. K. Fry, B. Afshar, P. C. Lück, H. Meugnier, J. Etienne, R. Peduzzi, and T. G. Harrison. 2005. Consensus sequence-based scheme for epidemiological typing of clinical and environmental isolates of Legionella pneumophila. J. Clin. Microbiol. 43:2047-2052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Helbig, J. H., S. Bernander, P. Castellani, J. Etienne, V. Gaia, S. Lauwers, D. Lindsay, P. C. Lück, T. Marques, S. Mentula, M. F. Peeters, C. Pelaz, M. Struelens, S. A. Uldum, G. Wewalka, and T. G. Harrison. 2002. Pan-European study on culture-proven Legionnaires' disease: distribution of Legionella pneumophila serogroups and monoclonal subgroups. Eur. J. Clin. Microbiol. Infect. Dis. 21:710-716. [DOI] [PubMed] [Google Scholar]
- 11.Hunter, P. R., and M. A. Gaston. 1988. Numerical index of discriminatory ability of typing systems: an application of Simpson's index of diversity. J. Clin. Microbiol. 26:2465-2466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lück, P. C., H. M. Wenchel, and J. H. Helbig. 1998. Nosocomial pneumonia caused by three genetically different strains of Legionella pneumophila and detection of these strains in the hospital water supply. J. Clin. Microbiol. 36:1160-1163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lüneberg, E., N. Zetzmann, D. Alber, Y. A. Knirel, O. Kooistra, U. Zähringer, and M. Frosch. 2000. Cloning and functional characterization of a 30 kb gene locus required for lipopolysaccharide biosynthesis in Legionella pneumophila. Int. J. Med. Microbiol. 290:37-49. [DOI] [PubMed] [Google Scholar]
- 14.Maiden, M. C. J., J. A. Bygraves, E. Feil, G. Morelli, J. E. Russell, R. Urwin, Q. Zhang, J. Zhou, K. Zurth, D. A. Caugant, I. M. Feavers, M. Achtman, and B. C. Spratt. 1998. Multilocus sequence typing: a portable approach to the identification of clones within populations of pathogenic microorganisms. Proc. Natl. Acad. Sci. USA 95:3140-3145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Morozova, I., X. Qu, S. Shi, G. Asamani, J. E. Greenberg, H. A. Shuman, and J. J. Russo. 2004. Comparative sequence analysis of the icm/dot genes in Legionella. Plasmid 51:127-147. [DOI] [PubMed] [Google Scholar]
- 16.Pettigrew, M., K. P. Fennie, M. P. York, J. Daniels, and F. Ghaffar. 2006. Variation in the presence of neuraminidase genes among Streptococcus pneumoniae isolates with identical sequence types. Infect. Immun. 74:3360-3365. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Struelens, M. J., and members of the European Study Group on Epidemiological Markers (ESGEM) of the European Society for Clinical Microbiology and Infectious Diseases. 1996. Consensus guidelines for appropriate use and evaluation of microbial epidemiologic typing systems. Clin. Microbiol. Infect. 2:2-11. [DOI] [PubMed] [Google Scholar]
- 18.Zou, C. H., Y. A. Knirel, J. H. Helbig, U. Zähringer, and C. S. Mintz. 1999. Molecular cloning and characterization of a locus responsible for O-acetylation of the O-polysaccharide of Legionella pneumophila serogroup 1 lipopolysaccharide. J. Bacteriol. 181:4137-4141. [DOI] [PMC free article] [PubMed] [Google Scholar]