Abstract
Legionella pneumophila serogroup (SG) 1 is the most frequent cause of legionellosis. This study analyzed environmental isolates of L. pneumophila SG 1 in Japan using monoclonal antibody (MAb) typing and sequence-based typing (SBT). Samples were analyzed from bathwater (BW; n = 50), cooling tower water (CT; n = 50), and soil (SO; n = 35). The distribution of MAb types varied by source, with the most prevalent types being Bellingham (42%), Oxford (72%), and OLDA (51%) in BW, CT, and SO, respectively. The ratios of MAb 3/1 positive isolates were 26, 2, and 14% from BW, CT, and SO, respectively. The environmental isolates from BW, CT, and SO were divided into 34 sequence types (STs; index of discrimination [IOD] = 0.973), 8 STs (IOD = 0.448), and 11 STs (IOD = 0.879), respectively. Genetic variation among CT isolates was smaller than seen in BW and SO. ST1 accounted for 74% of the CT isolates. The only common STs between (i) BW and CT, (ii) BW and SO, and (iii) CT and SO were ST1, ST129, and ST48, respectively, suggesting that each environment constitutes an independent habitat.
INTRODUCTION
Legionella pneumophila serogroup (SG) 1 is the most common agent causing legionellosis found in patients; however, differences in SGs have been found both in isolates from patients and from soil and various freshwater environments (9), such as cooling towers and bathing facilities. In patients, most strains (80%) belonged to SG 1 in our previous study (2). In cooling tower water isolates of Japan, L. pneumophila SGs 1 and 7 accounted for 67 and 23%, respectively, with other SGs being rarely isolated. On the other hand, the isolates from bathwater and from soil were more serotypically diverse, but SG 1 was still dominant in both environments, at 31% (1) and 26% (10), respectively. L. pneumophila SG 1 can be divided based on having or not having the virulence-associated epitope recognized by monoclonal antibody (MAb) 3/1 (13). In England and Wales, of the clinical isolates, 91.6% were MAb 3/1 positive compared to only 8.3% of the environmental isolates (12).
L. pneumophila isolates can be characterized by sequence-based typing (SBT) using the seven loci (flaA, pilE, asd, mip, mompS, proA, and neuA) proposed by the European Working Group on Legionella Infections (EWGLI; http://www.ewgli.org/ [11, 21]). This is a separate classifier from serogroup or MAb subtyping and is generally more precise due to the mutability of the latter factors. It allows for phylogenetic studies and identification of isolates that are closely related. The variation in STs of clinical and environmental isolates of L. pneumophila worldwide is very diverse. The indices of discrimination (IODs) (14) of environmental isolates and clinical isolates were determined to be 0.888 and 0.964, respectively, in Canada and 0.822 and 0.946, respectively, in the United States (15, 22). In England and Wales, however, environmental isolates are more variable than clinical ones (IODs of 0.933 and 0.901, respectively [12]), but the diversity is comparably great.
When 69 SG1 clinical isolates from Japan were subjected to SBT, they could be divided into 41 sequence types (STs). The IOD was 0.979. The ST with the most isolates (n = 7) was ST1. This is the most common ST occurring in the environment and among patients worldwide. Other major STs were ST306 (n = 6), ST120 (n = 5), and ST138 (n = 5). All ST306 and ST138 isolates, with one exception (ST306), were derived from bathwater (or suspected to be), suggesting that these strains readily adapt to bathwater habitats. The source of all ST1 and ST120 isolates remains unclear (2). In Japan, data from the National Epidemiological Surveillance of Infectious Diseases indicate that hot springs and public baths are primary sources of L. pneumophila, rather than cooling towers; however, in most cases the source of the bacteria is unknown (19).
We analyzed here environmental isolates of L. pneumophila SG 1, which is the principal cause of legionellosis in bathwater (the main source of infection in Japan), soil (a potential source of contamination for various water systems), and cooling tower water (another major source of legionellosis). Isolates were identified using MAb typing and SBT and then compared to previous clinical isolates (2) to determine relations between isolates from different environments and from patients.
MATERIALS AND METHODS
L. pneumophila strains.
A total of 135 environmental strains of L. pneumophila SG 1, which were independently isolated and unrelated to cases of infection, were analyzed, including isolates from bathwater (BW; n = 50), cooling tower water (CT; n = 50), and soil (SO; n = 35). All of the CT and BW isolates were obtained from different facilities: 66% of the CT isolates and 42% of the BW isolates originated from the Kanto region in central Japan. The SO isolates, which were independently collected from across Japan, were obtained from topsoil samples from roadsides, farmlands, gardens, etc. (10).
MAb subgrouping.
A total of 135 environmental strains of L. pneumophila SG 1 were subtyped serologically, using MAbs as described previously, into nine subgroups named Allentown/France, Bellingham, Benidorm, Camperdown, Heysham, Knoxville, OLDA, Oxford, and Philadelphia (13).
SBT.
SBT was performed according to the EWGLI SBT protocol (http://www.ewgli.org/) as described previously (11, 21). The isolates that failed amplification of neuA (whose indicated allele number was “0”) were not given ST numbers but were allocated arbitrary numbers prefixed by J (2). A minimum-spanning tree that had categorical coefficients of similarity and the priority rule of the highest number of single-locus variants as parameters was used to indicate differences in the number of loci among operational taxonomic units (OTU). The neighbor-joining method was then used to find pairs of OTU that minimized the total branch lengths by number of base substitutions on flaA, pilE, asd, mip, mompS, proA, and neuA concatenated sequences (2,501 bp) at each stage of OTU clustering. Both trees were constructed using BioNumerics software (version 6.5; Applied Maths, Sint-Martens-Latem, Belgium).
RESULTS
MAb subgrouping.
The isolates examined here were comprised of nine MAb types in all. The BW isolates were comprised of all nine MAb types, the CT isolates were comprised only of four, and the SO isolates were comprised of six. The distributions of MAb subgroups in the environmental isolates differed from one another, and from that found in clinical isolates (Fig. 1). The most common MAb subgroup in BW isolates was the Bellingham subgroup (42%), whereas the Oxford subgroup was the most common in CT (72%) and the OLDA subgroup was the most common in SO (51%). Bellingham, Oxford, and OLDA are MAb 3/1-negative subgroups. On the other hand, the most common subgroup observed in Japanese clinical isolates was Benidorm (45%), which is MAb 3/1 positive (2). Benidorm was detected in 12% of isolates from bathwater and 3% of isolates from soil. Of the 135 environmental isolates, only 14% had the virulence-associated epitope recognized by MAb 3/1: Benidorm, Allentown/France, Philadelphia, and Knoxville (13). In BW, 26% of the isolates were MAb 3/1 positive, compared to 14% in SO and a mere 2% in CT.
Fig 1.
Distributions of MAb types. (A) Isolates from bathwater (n = 50); (B) isolates from cooling tower water (n = 50); (C) isolates from soil (n = 35); (D) isolates from patients of legionellosis (n = 69 [2]). Allentown/France, Benidorm, Knoxville, and Philadelphia are MAb 3/1-positive subgroups. Bellingham, Camperdown, Heysham, OLDA, and Oxford are MAb 3/1-negative subgroups. MAb 3/1 indicates the virulence-associated epitope.
SBT.
The 135 environmental isolates (with the exception of one SO isolate in which amplification of the neuA target failed) could be divided into 50 STs, including 33 singletons (IOD = 0.886; Tables 1 and 2). The ST with the largest number of isolates was ST1 (n = 43, 29%), followed by ST48 (n = 10, 6.7%), ST129 (n = 7, 4.7%), ST739 (n = 6, 4.0%), and ST22 (n = 5, 3.3%). Strains with indigenous STs were isolated from each environment. The only common STs across environments were ST1 (37 from CT and 6 from BW), ST48 (9 from SO and 1 from CT), and ST129 (5 from BW and 2 from SO).
Table 1.
STs of Japanese environmental isolates of L. pneumophila serogroup 1
| Source environment and ST | No. (%) of isolates | MAb(s) (no. of isolates) |
|---|---|---|
| Cooling tower water | ||
| 1 | 37 (74) | Oxford (27), OLDA (10) |
| 154 | 4 (8) | Oxford (2), OLDA (1), Philadelphia (1) |
| 598 | 3 (6) | Oxford |
| 150 | 2 (4) | Oxford |
| Others | 4 (8) | |
| Total | 50 (100) | |
| Bathwater | ||
| 1 | 6 (12) | OLDA (4), Oxford (2) |
| 129 | 5 (10) | Bellingham |
| 599 | 3 (6) | Bellingham |
| 52 | 2 (4) | OLDA (1), Oxford (1) |
| 86 | 2 (4) | Bellingham |
| 127 | 2 (4) | Bellingham |
| 136 | 2 (4) | Oxford (1), Philadelphia (1) |
| 141 | 2 (4) | Philadelphia |
| Others | 26 (52) | |
| Total | 50 (100) | |
| Soila | ||
| 48 | 9 (26) | Bellingham |
| 739 | 6 (18) | OLDA |
| 22 | 5 (15) | OLDA |
| 448 | 3 (9) | OLDA (1), Oxford (1), Benidorm (1) |
| 129 | 2 (6) | Bellingham |
| 352 | 2 (6) | Allentown/France |
| 445 | 2 (6) | OLDA |
| 593 | 2 (6) | OLDA |
| Others | 3 (9) | |
| Total | 34 (100)b |
Excluding an isolate with failed neuA amplification.
The sum of percentages is not 100% because each percentage was rounded.
Table 2.
STs and MAb subtypes of 135 Japanese environmental isolates of L. pneumophila serogroup 1a
| Strain | Origin | MAb subgroup | MAb 3/1 | Allele no. |
STb | Yr | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| flaA | pilE | asd | mip | mompS | proA | neuA | ||||||
| NIIB 267 | B | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2000 |
| NIIB 273 | B | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2000 |
| NIIB 277 | B | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2000 |
| NIIB 720 | B | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 766 | B | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 772 | B | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 270 | B | OLDA | Neg. | 1 | 10 | 3 | 1 | 1 | 1 | 1 | 52 | 2000 |
| NIIB 271 | B | Oxford | Neg. | 1 | 10 | 3 | 1 | 1 | 1 | 1 | 52 | 2000 |
| NIIB 275 | B | Oxford | Neg. | 2 | 3 | 18 | 5 | 5 | 1 | 2 | 92 | 2000 |
| NIIB 886 | B | Allentown/France | Pos. | 2 | 6 | 17 | 14 | 12 | 8 | 11 | 788 | 2001 |
| NIIB 1090 | B | Bellingham | Neg. | 2 | 10 | 3 | 3 | 9 | 4 | 11 | 545 | 2005 |
| NIIB 729 | B | Bellingham | Neg. | 2 | 10 | 3 | 6 | 9 | 4 | 11 | 614 | 2005 |
| NIIB 370 | B | Philadelphia | Pos. | 2 | 12 | 3 | 6 | 8 | 14 | 9 | 141 | 2002 |
| NIIB 885 | B | Philadelphia | Pos. | 2 | 12 | 3 | 6 | 8 | 14 | 9 | 141 | 2004 |
| NIIB 158 | B | Bellingham | Neg. | 3 | 13 | 1 | 10 | 14 | 9 | 11 | 127 | |
| NIIB 925 | B | Bellingham | Neg. | 3 | 13 | 1 | 10 | 14 | 9 | 11 | 127 | 2004 |
| NIIB 160 | B | Bellingham | Neg. | 6 | 6 | 15 | 28 | 4 | 14 | 11 | 129 | |
| NIIB 231 | B | Bellingham | Neg. | 6 | 6 | 15 | 28 | 4 | 14 | 11 | 129 | |
| NIIB 887 | B | Bellingham | Neg. | 6 | 6 | 15 | 28 | 4 | 14 | 11 | 129 | 2002 |
| NIIB 889 | B | Bellingham | Neg. | 6 | 6 | 15 | 28 | 4 | 14 | 11 | 129 | 2004 |
| NIIB 890 | B | Bellingham | Neg. | 6 | 6 | 15 | 28 | 4 | 14 | 11 | 129 | 2005 |
| NIIB 888 | B | Camperdown | Neg. | 6 | 6 | 15 | 28 | 4 | 4 | 11 | 164 | 2002 |
| NIIB 712 | B | Camperdown | Neg. | 6 | 6 | 15 | 28 | 9 | 14 | 11 | 601 | 2005 |
| NIIB 929 | B | Benidorm | Pos. | 6 | 7 | 15 | 3 | 4 | 14 | 2 | 165 | 2004 |
| NIIB 715 | B | Benidorm | Pos. | 6 | 10 | 11 | 28 | 4 | 14 | 9 | 602 | 2005 |
| NIIB 1073 | B | Knoxville | Pos. | 6 | 10 | 15 | 3 | 19 | 4 | 11 | 604 | 2005 |
| NIIB 1197 | B | Heysham | Neg. | 6 | 10 | 15 | 3 | 21 | 4 | 3 | 606 | 2005 |
| NIIB 707 | B | Benidorm | Pos. | 6 | 10 | 15 | 13 | 17 | 14 | 11 | 122 | 2005 |
| NIIB 128 | B | Bellingham | Neg. | 6 | 10 | 15 | 28 | 4 | 4 | 11 | 125 | |
| NIIB 743 | B | Bellingham | Neg. | 6 | 10 | 15 | 28 | 44 | 14 | 11 | 201 | 2005 |
| NIIB 1213 | B | Bellingham | Neg. | 6 | 10 | 15 | 28 | 4 | 14 | 11 | 278 | 2005 |
| NIIB 229 | B | Philadelphia | Pos. | 6 | 10 | 17 | 6 | 9 | 4 | 9 | 136 | |
| NIIB 295 | B | Oxford | Neg. | 6 | 10 | 17 | 6 | 9 | 4 | 9 | 136 | 2000 |
| NIIB 278 | B | Bellingham | Neg. | 6 | 10 | 17 | 28 | 19 | 4 | 6 | 599 | |
| NIIB 696 | B | Bellingham | Neg. | 6 | 10 | 17 | 28 | 19 | 4 | 6 | 599 | 2005 |
| NIIB 699 | B | Bellingham | Neg. | 6 | 10 | 17 | 28 | 19 | 4 | 6 | 599 | 2005 |
| NIIB 710 | B | Bellingham | Neg. | 6 | 10 | 19 | 28 | 19 | 14 | 11 | 600 | 2005 |
| NIIB 230 | B | Philadelphia | Pos. | 6 | 10 | 21 | 6 | 9 | 4 | 9 | 137 | |
| NIIB 733 | B | Benidorm | Pos. | 6 | 10 | 21 | 13 | 17 | 14 | 11 | 131 | 2005 |
| NIIB 126 | B | Oxford | Neg. | 6 | 16 | 14 | 3 | 21 | 14 | 3 | 124 | |
| NIIB 159 | B | Bellingham | Neg. | 7 | 6 | 17 | 3 | 14 | 11 | 11 | 128 | |
| NIIB 1109 | B | Bellingham | Neg. | 7 | 6 | 17 | 28 | 36 | 11 | 11 | 86 | 2005 |
| NIIB 1115 | B | Bellingham | Neg. | 7 | 6 | 17 | 28 | 36 | 11 | 11 | 86 | 2005 |
| NIIB 805 | B | Bellingham | Neg. | 7 | 8 | 17 | 3 | 14 | 11 | 11 | 603 | 2005 |
| NIIB 1044 | B | Bellingham | Neg. | 7 | 10 | 17 | 3 | 13 | 9 | 11 | 605 | 2005 |
| NIIB 268 | B | Allentown/France | Pos. | 8 | 10 | 3 | 10 | 2 | 1 | 6 | 610 | 2000 |
| NIIB 1206 | B | Benidorm | Pos. | 10 | 12 | 7 | 3 | 16 | 18 | 6 | 138 | 2005 |
| NIIB 594 | B | Benidorm | Pos. | 10 | 22 | 7 | 3 | 16 | 9 | 6 | 162 | 2001 |
| NIIB 891 | B | Oxford | Neg. | 11 | 14 | 16 | 1 | 15 | 13 | 2 | 159 | 2005 |
| NIIB 1099 | B | OLDA | Neg. | 12 | 8 | 11 | 23 | 29 | 26 | 2 | 260 | 2005 |
| NIIB 65 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 1996 |
| NIIB 121 | C | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | |
| NIIB 122 | C | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | |
| NIIB 124 | C | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | |
| NIIB 182 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 1997 |
| NIIB 217 | C | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 1986 |
| NIIB 223 | C | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 1986 |
| NIIB 224 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 1986 |
| NIIB 225 | C | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 1986 |
| NIIB 226 | C | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 1986 |
| NIIB 228 | C | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 1986 |
| NIIB 237 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 1996 |
| NIIB 239 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 1993 |
| NIIB 418 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2004 |
| NIIB 547 | C | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2001 |
| NIIB 563 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2001 |
| NIIB 568 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2001 |
| NIIB 586 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2001 |
| NIIB 597 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2001 |
| NIIB 697 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 717 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 722 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 725 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 732 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 739 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 742 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 744 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 758 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 764 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 802 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 1048 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 1050 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 1052 | C | OLDA | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 1057 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 1082 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 1201 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2005 |
| NIIB 1592 | C | Oxford | Neg. | 1 | 4 | 3 | 1 | 1 | 1 | 1 | 1 | 2006 |
| NIIB 946 | C | Bellingham | Neg. | 5 | 2 | 22 | 27 | 6 | 10 | 12 | 48 | 2005 |
| NIIB 552 | C | Oxford | Neg. | 11 | 4 | 3 | 1 | 1 | 1 | 1 | 161 | 2001 |
| NIIB 595 | C | Oxford | Neg. | 11 | 14 | 16 | 1 | 15 | 13 | 1 | 150 | 2001 |
| NIIB 1207 | C | Oxford | Neg. | 11 | 14 | 16 | 1 | 15 | 13 | 1 | 150 | 2005 |
| NIIB 694 | C | Oxford | Neg. | 11 | 14 | 16 | 10 | 15 | 13 | 11 | 598 | 2005 |
| NIIB 746 | C | Oxford | Neg. | 11 | 14 | 16 | 10 | 15 | 13 | 11 | 598 | 2005 |
| NIIB 1098 | C | Oxford | Neg. | 11 | 14 | 16 | 10 | 15 | 13 | 11 | 598 | 2005 |
| NIIB 590 | C | Oxford | Neg. | 11 | 14 | 16 | 16 | 15 | 13 | 2 | 154 | 2001 |
| NIIB 591 | C | Oxford | Neg. | 11 | 14 | 16 | 16 | 15 | 13 | 2 | 154 | 2001 |
| NIIB 778 | C | OLDA | Neg. | 11 | 14 | 16 | 16 | 15 | 13 | 2 | 154 | 2005 |
| NIIB 1077 | C | Philadelphia | Pos. | 11 | 14 | 16 | 16 | 15 | 13 | 2 | 154 | 2005 |
| NIIB 1269 | C | Oxford | Neg. | 11 | 14 | 16 | 16 | 15 | 13 | 1 | 607 | 2005 |
| NIIB 611 | C | OLDA | Neg. | 18 | 4 | 3 | 1 | 1 | 1 | 1 | 163 | 2001 |
| NIIB 2366 | S | OLDA | Neg. | 2 | 3 | 6 | 10 | 2 | 1 | 6 | 22 | 2001 |
| NIIB 2388 | S | OLDA | Neg. | 2 | 3 | 6 | 10 | 2 | 1 | 6 | 22 | 2001 |
| NIIB 2403 | S | OLDA | Neg. | 2 | 3 | 6 | 10 | 2 | 1 | 6 | 22 | 2001 |
| NIIB 2404 | S | OLDA | Neg. | 2 | 3 | 6 | 10 | 2 | 1 | 6 | 22 | 2001 |
| NIIB 2409 | S | OLDA | Neg. | 2 | 3 | 6 | 10 | 2 | 1 | 6 | 22 | 2001 |
| NIIB 2375 | S | Oxford | Neg. | 2 | 3 | 18 | 10 | 2 | 1 | 6 | 448 | 2001 |
| NIIB 2395 | S | OLDA | Neg. | 2 | 3 | 18 | 10 | 2 | 1 | 6 | 448 | 2001 |
| NIIB 2338 | S | Benidorm | Neg. | 2 | 3 | 18 | 10 | 2 | 1 | 6 | 448 | 2001 |
| NIIB 2380 | S | Allentown/France | Pos. | 2 | 3 | 18 | 10 | 25 | 5 | 6 | 740 | 2001 |
| NIIB 2363 | S | OLDA | Neg. | 2 | 3 | 18 | 13 | 2 | 1 | 6 | 445 | 2001 |
| NIIB 2392 | S | OLDA | Neg. | 2 | 3 | 18 | 13 | 2 | 1 | 6 | 445 | 2001 |
| NIIB 2339 | S | OLDA | Neg. | 2 | 3 | 40 | 13 | 2 | 1 | 6 | 593 | 2001 |
| NIIB 2373 | S | OLDA | Neg. | 2 | 3 | 40 | 13 | 2 | 1 | 6 | 593 | 2001 |
| NIIB 2355 | S | OLDA | Neg. | 5 | 1 | 22 | 26 | 6 | 10 | 12 | 45 | 2001 |
| NIIB 2332 | S | Bellingham | Neg. | 5 | 2 | 22 | 27 | 6 | 10 | 12 | 48 | 2001 |
| NIIB 2335 | S | Bellingham | Neg. | 5 | 2 | 22 | 27 | 6 | 10 | 12 | 48 | 2001 |
| NIIB 2346 | S | Bellingham | Neg. | 5 | 2 | 22 | 27 | 6 | 10 | 12 | 48 | 2001 |
| NIIB 2383 | S | Bellingham | Neg. | 5 | 2 | 22 | 27 | 6 | 10 | 12 | 48 | 2001 |
| NIIB 2398 | S | Bellingham | Neg. | 5 | 2 | 22 | 27 | 6 | 10 | 12 | 48 | 2001 |
| NIIB 2399 | S | Bellingham | Neg. | 5 | 2 | 22 | 27 | 6 | 10 | 12 | 48 | 2001 |
| NIIB 2405 | S | Bellingham | Neg. | 5 | 2 | 22 | 27 | 6 | 10 | 12 | 48 | 2001 |
| NIIB 2406 | S | Bellingham | Neg. | 5 | 2 | 22 | 27 | 6 | 10 | 12 | 48 | 2001 |
| NIIB 2408 | S | Bellingham | Neg. | 5 | 2 | 22 | 27 | 6 | 10 | 12 | 48 | 2001 |
| NIIB 2371 | S | Bellingham | Neg. | 6 | 6 | 15 | 28 | 4 | 14 | 11 | 129 | 2001 |
| NIIB 2411 | S | Bellingham | Neg. | 6 | 6 | 15 | 28 | 4 | 14 | 11 | 129 | 2001 |
| NIIB 2370 | S | OLDA | Neg. | 6 | 10 | 23 | 10 | 18 | 14 | 0c | J6d | 2001 |
| NIIB 2342 | S | OLDA | Neg. | 12 | 8 | 11 | 2 | 10 | 12 | 2 | 739 | 2001 |
| NIIB 2356 | S | OLDA | Neg. | 12 | 8 | 11 | 2 | 10 | 12 | 2 | 739 | 2001 |
| NIIB 2381 | S | OLDA | Neg. | 12 | 8 | 11 | 2 | 10 | 12 | 2 | 739 | 2001 |
| NIIB 2386 | S | OLDA | Neg. | 12 | 8 | 11 | 2 | 10 | 12 | 2 | 739 | 2001 |
| NIIB 2390 | S | OLDA | Neg. | 12 | 8 | 11 | 2 | 10 | 12 | 2 | 739 | 2001 |
| NIIB 2394 | S | OLDA | Neg. | 12 | 8 | 11 | 2 | 10 | 12 | 2 | 739 | 2001 |
| NIIB 2327 | S | Allentown/France | Pos. | 12 | 8 | 11 | 13 | 10 | 12 | 2 | 352 | 2001 |
| NIIB 2353 | S | Allentown/France | Pos. | 12 | 8 | 11 | 13 | 10 | 12 | 2 | 352 | 2001 |
| NIIB 2343 | S | Philadelphia | Pos. | 21 | 14 | 29 | 30 | 15 | 29 | 6 | 741 | 2001 |
Origins: B, bathwater; C, cooling tower water; S, soil. MAb 3/1: Pos., positive; Neg., negative.
Underlining indicates an ST first reported from Japan and later reported from other countries. Bold facing indicates STs unique to Japan as of 8 March 2012.
Allele number “0” means a failed amplification.
J6, an arbitrary number allocated to unique six-allele profile without neuA.
There were no regional differences in the distribution of ST1 in either the CT or the BW isolates (76% [22/29] of CT isolates from the Kanto region and 71% [15/21] from other regions; 14% [3/21] of BW isolates from the Kanto region and 10% [3/29] from other regions). ST1 was not detected among the SO isolates.
The 50 CT isolates were divided into eight STs (IOD = 0.448). The 50 BW isolates were divided into 34 STs (IOD = 0.973). The 35 SO isolates (with one exception that failed neuA amplification) were divided into 11 STs (IOD = 0.879).
The minimum-spanning tree illustrates the distribution of the STs (Fig. 2). Thirty of the fifty STs obtained in this analysis were unique to Japan, according to data submitted to the EWGLI SBT database as of March 2012. Twenty of the fifty STs had already also been detected in clinical isolates in Japan and/or abroad, according to the same database. Most SO isolates formed three distinct groups (groups S1, S2, and S3 in Fig. 2). Group S2 had no linkage with other STs. CT isolates formed group C1 and group C2. The two groups were adjacent in the minimum-spanning tree, but even the most related STs (ST161 and ST150) that belonged to group C1 and group C2, respectively, differed in four loci. The BW isolates were dispersed, forming one major group (group B1) and two smaller groups. This finding was supported by neighbor-joining analysis based on a nucleotide sequence comparison of seven concatenated loci of SBT (2,501 bp) of the same isolates as in Fig. 2 (Fig. 3). Figure 3 shows that isolates belonging to each group found in the minimum-spanning tree were also clustered. However, the relationships observed between groups in the dendrogram were different from the minimum-spanning analysis, except for the cluster of group S1 and group C1. Two groups of isolates from cooling tower water (group C1 and group C2) were located distally (unlike Fig. 2). Groups C2, B2, and B3 shared many informative sites between groups, compared to groups C1, S1, S2, S3, and B1, as shown by the bootstrap support value of 74%.
Fig 2.
Minimum-spanning tree showing how the L. pneumophila isolates, with seven determined alleles, are distributed in terms of their STs. The ST number is shown beside the circle. An underlined ST number indicates that the ST also has been reported abroad, and a boxed ST number indicates that clinical isolates with the same STs were detected. The size of the circle indicates the number of isolates. The white parts of the circles (pie charts) indicate isolates derived from cooling tower water (n = 50), the gray indicates isolates derived from bathwater (n = 50), and the black indicates isolates derived from soil (n = 34). Short thick branches connect single-locus variants, thin branches connect double- or triple-locus variants, broken branches connect four-locus variants, and thinner branches connect five- or six-locus variants. The numbers of locus variants are proportional to the length of branches. Groups that were generated with single-, double-, and triple-locus variants are indicated by differently shaded backgrounds. Groups C1 and C2 had their major isolates derived from cooling tower water, groups B1, B2, and B3 had their major isolates derived from bathwater, and groups S1, S2, and S3 had their major isolates derived from soil.
Fig 3.
Phylogenetic tree of flaA, pilE, asd, mip, mompS, proA, and neuA concatenated sequences from L. pneumophila serogroup 1 isolates determined by the neighbor-joining method. Bootstrap support values for nodes outside groups higher than 50% are shown. The scale bar indicates genetic distances between sequences. The groups correspond to those in Fig. 2. STs of isolates that are MAb 3/1 positive are denoted by asterisks (*), and STs of a part of isolates that are MAb 3/1 positive are denoted by plus symbols (+).
Combining the sequence typing and the MAb subgrouping.
Some STs were composed of isolates belonging to different MAb subgroups (and vice versa). Thus, ST1 (n = 43) was composed of isolates belonging to the Oxford (n = 29) and OLDA (n = 14) subgroups. ST154 (n = 4) contained the Oxford (n = 2), OLDA (n = 1), and Philadelphia (n = 1) subgroups. ST448 (n = 3) consisted of OLDA, Oxford, and Benidorm isolates. In contrast, all ST48 (n = 9) and all ST129 (n = 5) isolates were Bellingham. By combining the data of SBT and MAb subgrouping, we could divide the 135 isolates into 58 types (IOD = 0.933; Tables 1 and 2).
DISCUSSION
We analyzed L. pneumophila SG 1 isolates from three distinct environments using MAb typing and SBT in Japan: cooling tower water, bathwater, and soil. The distributions of MAbs and STs of isolates differed both between the environments and from previous clinical isolates (2).
Of the SG 1 clinical isolates from Japan, 80% had the virulence-associated epitope recognized by MAb 3/1 (2). As for the analyzed 135 environmental isolates, MAb 3/1-positive isolates accounted for only 14%. Similar observations have also been made in studies conducted in other countries (i.e., Germany [3], England and Wales [12], and the United States [15]). Although these data indicated MAb 3/1 as the virulence-associated epitope, our study's MAb 3/1-positive isolates dispersed on the dendrogram by SBT (Fig. 3) in the three kinds of analyzed environments, suggesting the MAb 3/1 epitope is easily lost or gained during adaptation to environments when there is no pressure to retain human pathogenicity. Loss of the MAb 3/1 epitope may bring some advantage for fitness, as MAb 3/1-negative isolates dominated in each environment.
Although 30 of the 50 STs obtained in this analysis were unique to Japan, the EWGLI SBT database indicated that the majority of unique STs have single-locus variants abroad. Among the unique STs, only ST138 and ST162 in group B3, and ST141 have neither single-locus variants nor double-locus variants abroad. ST138 of the Benidorm subgroup is the primary clinical isolate associated with bathwater in Japan (2; unpublished results). Thus, a few STs might be unique to Japan, which is isolated by water.
All of the Japanese ST1 strains were of the MAb 3/1-negative OLDA or Oxford subgroups, whereas the ST1 strains in the EWGLI database are divided into nine MAb types. This distribution of MAb types within ST1 may be a regional difference. On the other hand, a regional difference did not always apply. All nine ST48 from our results were of the Bellingham subgroup, and according to the EWGLI SBT database prior to May 2011, all of the MAb-typed ST48 strains submitted thus far were Bellingham. Since May 2011, however, Camperdown and OLDA strains containing ST48 have been deposited. If more strains could be analyzed, we predict that different MAb types within many ST groups would be detected.
The isolates from soil were divided into three groups (Fig. 2) by the spanning tree analysis and only had two common STs that were detected in different environments: ST48 with an isolate from cooling tower water and ST129 with an isolate from bathwater. These findings indicate that these bacteria generally inhabit the soil but are able to contaminate water sources. Further investigations of more isolates from soil may identify STs that link the three groups or that have more corresponding STs with isolates from water environments. Nine of the 11 STs of soil isolates were also detected in clinical isolates, in contrast to only 11 of 34 in bathwater and 3 of 8 from cooling tower water. These findings support the possibility that soil is one of the infectious sources of legionellosis.
In Canada, the distribution of STs in strains from natural water sources was noted as significantly different compared to strains from a manufactured environment (22). We note a similar finding in the present study. The water of Japanese public baths is often derived from hot springs. The characteristics of hot spring water, namely, chemical features such as pH and temperature, are highly variable, whereas the water from hot or cold water systems and cooling towers tend to have rather similar characteristics due to similar water treatment procedures. In our results, STs and MAb types of isolates from bathwater both differed from and were more varied than those of cooling tower water. These features might be related to the kind of host amoebae, which adapt to and inhabit various environments (20, 23). It has been shown that the growth of L. pneumophila in some species of host amoebae depends on bacterial genetic background (4, 8). Some isolates with particular STs adapted for amoebae that live in bathwater may be infectious to humans.
L. pneumophila SG 1 isolates in cooling tower water in Japan were divided into two genetic groups (group C1 and group C2; Fig. 3). Recombination events may have occurred between members of group C1 and group C2. For example, ST161 (flaA11, pilE4, asd-3, mip-1, mompS1, proA1, and neuA1) was a recombinant between ST1 (flaA1, pilE4, asd-3, mip-1, mompS1, proA1, and neuA1) and ST154 (flaA11, pilE14, asd-16, mip-16, mompS15, proA13, and neuA2), which was a predicted primary founder, and ST150 (flaA11, pilE14, asd-16, mip-1, mompS15, proA13, and neuA1) was also a recombinant between ST1 (with adjacent alleles, mip-1 and neuA1) and ST154, shortening the distance between the two groups on the minimum-spanning tree (Fig. 2).
The IOD (0.886) of the 135 environmental isolates was lower than described in our previous report based on clinical isolates (0.979 [2]). These findings were similar to those reported in Canada and the United States (15, 22). The lower diversity observed among environmental isolates compared to clinical isolates may be due to the high prevalence of ST1 (22). ST1 is the most prevalent ST in the world (3, 6–7, 12, 15–16, 22, 26). We have also shown that the majority of environmental isolates, especially from cooling tower waters in Japan (37/50, or 74%), are ST1. Similar results were shown in South Korea (46/68 [67.6%] of SG 1 isolates from cooling tower water were ST1), which is adjacent to Japan (16). In a Canadian study, 34.2% of L. pneumophila strains from manufactured environments and 7.7% of L. pneumophila isolates from natural water sources (lakes and hot springs) were SG 1 and ST1. Among the Canadian strains, five of six SG 1 isolates from cooling tower waters were ST1 (22). In a U.S. study, ST1 accounted for 40% of the L. pneumophila SG 1 environmental isolates; however, the types of environments were not indicated. In Singapore in tropical southeast Asia, the IOD of environmental L. pneumophila isolates was found to be 0.970, and three (two from a cooling tower and one from a water tank) of 16 SG 1 isolates were ST1 (17). In a study conducted in England and Wales, 154 of 276 L. pneumophila isolates, including 29 isolates derived from cooling tower water, were SG 1, and 54/154 (35%) SG 1 environmental isolates were ST1 (12). In our study, ST1 accounted for 74 and 12% of the environmental SG 1 isolates from cooling tower water and bathwater, respectively, whereas no ST1 was found in isolates from soil. Isolates with ST1 have adapted to water environments, especially in manufactured water systems such as cooling towers, and have been detected around the world. The ability of ST1 isolates to adapt to natural water sources such as lakes and hot springs might be rather low. Moreover, they might be unfit to survive well in soil environments. The predominant ST, ST1, of isolates from cooling tower water induced an insufficient IOD, 0.448, whereas the discrimination powers for isolates from bathwater and soil were sufficiently significant (0.973 and 0.879, respectively).
Handling of potting soil could be considered a risk factor for legionellosis. Surveys in several countries have detailed various Legionella species, including L. pneumophila SG1, that were isolated from potting soil or composted materials (5, 18, 25). SBT analysis on composted material isolates in United Kingdom revealed that their L. pneumophila SG1 isolates belonged to ST84 (18). Seven alleles belonging to ST84 were unshared by soil isolates in our study (except for one allele, flaA12, which was), although ST84 has been detected in clinical isolates in Japan (2) and other countries, according to the EWGLI database. Groups S1, S2, and S3, mainly formed by isolates derived from soil, were distant phylogenetically from groups of isolates derived from water environments. Only ST129 soil isolates shared the B1 group with isolates from a water environment. Although some isolates from cooling tower water and bathwater were included in groups S1, S2, and S3, this might imply that some part of these L. pneumophila subpopulations primarily inhabits soil, occasionally mutating and becoming fit to contaminate water environments. Recently, indigenous soil samples were collected in Thailand, and 115 Legionella isolates, including 2 L. pneumophila SG1 isolates, were identified (24; EWLGI database). One ST identified from the L. pneumophila SG1 soil sample isolates related to group S1 and the other to group S2, supporting the idea that most soil isolates belong to particular groups. It is also interesting that the most prevalent ST1 isolates from water samples were not isolated from soil in our study, suggesting the possibility of habitat segregation of L. pneumophila. To elucidate this possibility, we need to investigate more environmental isolates from both soil and water.
ACKNOWLEDGMENTS
We thank Norman K. Fry (Respiratory and Systemic Infection Laboratory, Health Protection Agency) for assigning the newly identified alleles.
This study was supported by the Health and Labor Sciences research grant H22-kenki-014 (to F.K.) and partially supported by Ministry of Education, Culture, Sports, Science, and Technology grant KAKENHI 23590530 (to J.A.-M).
The Working Group for Legionella in Japan included Mie Sasaki, Miyagi Prefectural Institute of Public Health and Environment; Mikako Hosoya, Niigata Prefectural Institute of Public Health and Environmental Sciences; Yuko Watanabe, Toshiro Kuroki, Kanagawa Prefectural Institute of Public Health; Masamichi Wada, Nagano Environmental Conservation Research Institute; Hitoshi Doi, Osaka Prefectural Institute of Public Health; and Koichi Murakami, Fukuoka Institute of Health and Environmental Sciences.
Footnotes
Published ahead of print 6 April 2012
REFERENCES
- 1. Amemura-Maekawa J, et al. 2008. Distinct difference of flaA genotypes of Legionella pneumophila between isolates from bath water and cooling tower water. Microbiol. Immunol. 52:460–464 [DOI] [PubMed] [Google Scholar]
- 2. Amemura-Maekawa J, et al. 2010. Characterization of Legionella pneumophila isolates from patients in Japan according to serogroups, monoclonal antibody subgroups, and sequence types. J. Med. Microbiol. 59:653–659 [DOI] [PubMed] [Google Scholar]
- 3. Borchardt J, Helbig JH, Lück PC. 2008. Occurrence and distribution of sequence types among Legionella pneumophila strains isolated from patients in Germany: common features and differences to other regions of the world. Eur. J. Clin. Microbiol. Infect. Dis. 27:29–36 [DOI] [PubMed] [Google Scholar]
- 4. Buse HY, Ashbolt NJ. 2011. Differential growth of Legionella pneumophila strains within a range of amoebae at various temperatures associated with in-premise plumbing. Lett. Appl. Microbiol. 53:217–224 [DOI] [PubMed] [Google Scholar]
- 5. Casati S, Conza L, Bruin J, Gaia V. 2010. Compost facilities as a reservoir of Legionella pneumophila and other Legionella species. Clin. Microbiol. Infect. 16:945–947 [DOI] [PubMed] [Google Scholar]
- 6. Cazalet C, et al. 2008. Multigenome analysis identifies a worldwide distributed epidemic Legionella pneumophila clone that emerged within a highly diverse species. Genome Res. 18:431–441 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Chasqueira MJ, Rodrigues L, Nascimento M, Marques T. 2009. Sequence-based and monoclonal antibody typing of Legionella pneumophila isolated from patients in Portugal during 1987–2008. Euro. Surveill. 14:409–412 [DOI] [PubMed] [Google Scholar]
- 8. Dey R, Bodennec J, Mameri MO, Pernin P. 2009. Free-living freshwater amoebae differ in their susceptibility to the pathogenic bacterium Legionella pneumophila. FEMS Microbiol. Lett. 290:10–17 [DOI] [PubMed] [Google Scholar]
- 9. Fields BS. 1996. The molecular ecology of legionellae. Trends Microbiol. 4:286–290 [DOI] [PubMed] [Google Scholar]
- 10. Furuhata K, Okabe Y, Dogasaki C, Hara M, Fukuyama M. 2002. Isolation of Legionella spp. from soils in Japan. Bokin Bobai 30:555–561 (In Japanese.) [Google Scholar]
- 11. Gaia V, et al. 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]
- 12. Harrison TG, Afshar B, Doshi N, Fry NK, Lee JV. 2009. Distribution of Legionella pneumophila serogroups, monoclonal antibody subgroups and DNA sequence types in recent clinical and environmental isolates from England and Wales (2000–2008). Eur. J. Clin. Microbiol. Infect. Dis. 28:781–791 [DOI] [PubMed] [Google Scholar]
- 13. Helbig JH, et al. 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]
- 14. Hunter PR, Gaston MA. 1988. Numerical index of the 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]
- 15. Kozak NA, et al. 2009. Distribution of lag-1 alleles and sequence-based types among Legionella pneumophila serogroup 1 clinical and environmental isolates in the United States. J. Clin. Microbiol. 47:2525–2535 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Lee HK, Shim JI, Kim HE, Yu JY, Kang YH. 2010. Distribution of Legionella species from environmental water sources of public facilities and genetic diversity of L. pneumophila serogroup 1 in South Korea. Appl. Environ. Microbiol. 76:6547–6554 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Lim YH, et al. 2011. Environmental surveillance and molecular characterization of Legionella in tropical Singapore. Trop. Biomed. 28:149–159 [PubMed] [Google Scholar]
- 18. McCabe S, Brown A, Edwards GF, Lindsay D. 2011. Enhanced isolation of Legionella species from composted material. Clin. Microbiol. Infect. 10:1517–1520 [DOI] [PubMed] [Google Scholar]
- 19. National Institute of Infectious Diseases and Tuberculosis and Infectious Diseases Control Division, Ministry of Health, Labour, and Welfare 2000. Legionellosis, April 1999-July 2000. IASR (Infectious Agents Surveillance Report) 21:186–187 [Google Scholar]
- 20. Patterson WJ, Hay J, Seal DV, McLuckie JC. 1997. Colonization of transplant unit water supplies with Legionella and protozoa: precautions required to reduce the risk of legionellosis. J. Hosp. Infect. 37:7–17 [DOI] [PubMed] [Google Scholar]
- 21. Ratzow S, Gaia V, Helbig JH, Fry NK, Lück PC. 2007. Addition of neuA, the gene encoding N-acylneuraminate cytidylyl transferase, increases the discriminatory ability of the consensus sequence-based scheme for typing Legionella pneumophila serogroup 1 strains. J. Clin. Microbiol. 45:1965–1968 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Reimer AR, Au S, Schindle S, Bernard KA. 2010. Legionella pneumophila monoclonal antibody subgroups and DNA sequence types isolated in Canada between 1981 and 2009: laboratory component of national surveillance. Eur. J. Clin. Microbiol. Infect. Dis. 29:191–205 [DOI] [PubMed] [Google Scholar]
- 23. Stockman LJ, Wright CJ, Visvesvara GS, Fields BS, Beach MJ. 2011. Prevalence of Acanthamoeba spp. and other free-living amoebae in household water, Ohio, U.S.A.—1990–1992. Parasitol. Res. 108:621–627 [DOI] [PubMed] [Google Scholar]
- 24. Travis TC, et al. 2012. Survey of Legionella species found in Thai soil. Int. J. Microbiol. 2012:218791. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Velonakis EN, et al. 2010. First isolation of Legionella species, including L. pneumophila serogroup 1, in Greek potting soils: possible importance for public health. Clin. Microbiol. Infect. 16:763–766 [DOI] [PubMed] [Google Scholar]
- 26. Zhou H, et al. 2010. Optimization of pulsed-field gel electrophoresis for Legionella pneumophila subtyping. Appl. Environ. Microbiol. 76:1334–1340 [DOI] [PMC free article] [PubMed] [Google Scholar]