Infection, Genetics and Evolution xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Short communication
Population structure of Legionella spp. from environmental samples in Gabon, 2013 Jonas Ehrhardt a,b, Abraham S. Alabi a,b, Thorsten Kuczius c, Francis Foguim Tsombeng b, Karsten Becker d, Peter G. Kremsner a,b, Frieder Schaumburg b,d,⇑,1, Meral Esen a,b,1 a
Institut für Tropenmedizin, Eberhard Karls Universität Tübingen, Deutsches Zentrum für Infektionsforschung, Tübingen, Germany Centre de Recherches Médicales de Lambaréné (CERMEL), Lambaréné, Gabon Institute of Hygiene, University Hospital Münster, Münster, Germany d Institute of Medical Microbiology, University Hospital Münster, Münster, Germany b c
a r t i c l e
i n f o
Article history: Received 31 March 2015 Received in revised form 13 May 2015 Accepted 18 May 2015 Available online xxxx Keywords: Legionella Africa Gabon Population structure Hospital
a b s t r a c t Aquatic environments are the most important source for Legionella spp. infections such as Legionnaires’ disease and Pontiac fever. The reservoirs of Legionella spp. are mostly unclear in sub-Saharan Africa. The aim of this study, conducted in 2013, was to identify geographical areas of an increased risk for exposure to Legionella spp., and to describe the population structure of Legionella spp. from different water sources in a cross-sectional study in Gabon. Fresh water samples (n = 200) were cultured on Legionella selective agar; species were confirmed by MALDI-TOF, a Legionella pneumophila specific real-time PCR and 16S RNA gene sequencing. Serogroups were identified by agglutination test. The population structure was assessed by multilocus sequence typing (MLST). Legionella spp. isolates (n = 29) were frequently found in the hospital setting particularly in hot water systems. Open water bodies (i.e. rivers, lakes) were not contaminated with Legionella spp. Isolated L. pneumophila mainly belonged to serogroups 2–14 (n = 19) and MLST sequence type ST1, ST75 (and related STs) and ST1911. In conclusion, hospitalized patients might have an increased risk to become infected with Legionella spp. in the studied areas in Gabon, particularly if they have risk factors such as comorbidities. Both broadly extended (ST1, ST75) and local lineages (ST1911) were present in our setting. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Natural reservoirs of Legionella spp. are aquatic environments. Humans usually get infected by the inhalation of contaminated aerosols and might develop the severe pneumonic Legionnaires’ disease or the self-limited Pontiac fever (Whiley et al., 2014). Risk factors are comorbidities, immunosuppression and age (Whiley et al., 2014). Legionella pneumophila is the most common species causing legionellosis with a predominance of serogroup 1 which accounts for almost 80% of isolates from Legionnaires’ disease (Helbig et al., 2002). Other Legionella spp. (i.e. Legionella dumoffii, Legionella anisa) can also cause pneumonic infections (Muder, 2000).
⇑ Corresponding author at: Institute of Medical Microbiology, University Hospital Münster, Münster, Germany. E-mail address: [email protected] (F. Schaumburg). 1 Shared senior authorship.
Epidemiological data on Legionella spp. from environmental and clinical samples are limited and partly contradicting in sub-Saharan Africa. Few studies point towards a high prevalence in surface water in South Africa (73–75%) and Nigeria (67.3%) (Alli et al., 2011; Bartie et al., 2003; Dobrowsky et al., 2014). In contrast, legionellosis is infrequently reported. No case of Legionella infection was detected in patients with community acquired pneumonia in Gabon and only one case was detected in patients with influenza-like illness or severe acute respiratory infection in Kenya respectively (Kim et al., 2012; Lassmann et al., 2008). This is in contrast to South Africa where Legionella spp. was the fifth most frequent cause of pneumonia in patients that required hospital admission (Potgieter and Hammond, 1992). Knowledge about reservoirs of Legionella spp. is important to assess the potential risk of infection particularly in older people with comorbidities and to develop strategies for infection control and prevention. As this information is missing for Central Africa, we performed a cross sectional study on the burden of Legionella spp. in water bodies in Gabon. The aim was to identify
http://dx.doi.org/10.1016/j.meegid.2015.05.019 1567-1348/Ó 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: Ehrhardt, J., et al. Population structure of Legionella spp. from environmental samples in Gabon, 2013. Infect. Genet. Evol. (2015), http://dx.doi.org/10.1016/j.meegid.2015.05.019
2
J. Ehrhardt et al. / Infection, Genetics and Evolution xxx (2015) xxx–xxx
geographical areas of increased risk for exposure and to describe the population structure of Legionella spp. in the aquatic environment in Gabon. 2. Materials and methods 2.1. Water samples Water samples were drawn from randomly selected water sources at seven sites in four provinces in Gabon (Estuaire, Moyen Ogooué, Ogooué-Ivindo, Ngounié) from April to October 2013 (Table 1). Samples were taken from households or institutions after receiving permission from the person in charge. From each sampling site, 500 ml of water were collected in sterile, screw capped glass bottles containing a final concentration of 0.16 mM sodium thiosulphate. Samples from water taps and standpipes were aseptically collected after water was allowed to run for at least 1 min. In case of hot water systems, two samples were drawn (first cold then hot water). Samples were stored in a cool box and analyzed within 4 h after sampling. For each sample, the type of water source (e.g. tap water, shower, standpipe, borehole, open water body) and temperature of the water sample were recorded. The study protocol was approved by our local scientific review committee (SRC 2013.01). 2.2. Sample processing and culture Trained laboratory technicians performed the microbiological analysis of samples. The laboratory successfully participates in regular external quality assessments (Alabi et al., 2013). Water samples were analyzed using a modified protocol of the International Organization for Standardization (ISO 11731 and ISO 11731-2) (International Organization for Standardization (ISO), 1998). Briefly, 500 ml of each water sample was filtered using a 0.45 lm membrane filter (Millipore Merck, Billerica, Massachusetts, USA). To suppress the growth of non-Legionella spp. organisms, the filter was treated with 20 ml of 0.2 M hydrochloride/sodium chloride buffer (pH 2.2) for 5 min and was rinsed with sterile water. The membrane was shaken vigorously in 3 ml of sterile water to wash the organisms from the membrane. Of this, a volume of 0.75 ml was cultured on GVPC agar (Oxoid, Wesel, Germany) at 37 °C for up to 10 days under microaerophilic conditions. Fluorescence of colonies was examined under UV-light (254 nm/366 nm) after 3, 6, and 10 days. Contamination with Legionella spp. was reported as colony forming units (CFU) per 100 ml of unfiltered water. As concentrations of non-Legionella spp. and suspended matters were high in samples from open water bodies (i.e. borehole, river, lake), we filtered 100 ml instead of 500 ml of these samples.
Columbia blood agar were considered cysteine non-dependent and reported as non-Legionella spp. (Centers For Disease Control and Prevention, 2005). Serogrouping was performed on pure cultures using a latex agglutination test which differentiates between serogroup 1, serogroup 2–14 and seven non-L. pneumophila species (Legionella longbeachae, Legionella bozemanii, Legionella dumoffii, Legionella gormanii, Legionella jordanis, Legionella micdadei and Legionella anisa, Oxoid, Wesel, Germany). Species identification was done in Germany using MALDI-TOF mass spectrometry (Biotyper, Bruker, Bremen, Germany) and the MALDI Biotyper library (Version 3.3.2.0). Species of L. pneumophila was confirmed by a L. pneumophila specific real-time PCR (GeneProof, medac Diagnostika, Wedel, Germany). All non-L. pneumophila species were confirmed with 16S RNA gene sequencing (Becker et al., 2004). 2.4. Genotyping From all confirmed Legionella spp. isolates, DNA was extracted with QIAmp DNA Mini Kit (Qiagen, Herne, Germany). Multilocus sequence typing (MLST) was performed for all L. pneumophila isolates using the protocol developed by the European Working Group for Legionella Infections (EWGLI) published by the ESCMID study group for Legionella infections (Gaia et al., 2005; Mentasti et al., 2014). MLST sequence types were deduced using the sequence based typing (SBT) Database of L. pneumophila (www.hpa-bioinformatics.org.uk/legionella/legionella_sbt/php/sbt_homepage.php) hosted at the Health Protection Agency, England. The SBT scheme has a good correlation with genetic lineages based on whole genome sequencing (Underwood et al., 2013). To assess the phylogenetic relation of our L. pneumophila isolates with African isolates published in the Legionella SBT Database (assessed 9 January 2015, one isolate per ST per country), we constructed a neighbor joining tree using the concatenated sequences of the seven MLST genes of the Legionella typing scheme and MEGA5 (www.megasoftware.net). 2.5. Statistics Comparisons between groups were performed by means of the Chi2-test and the Fisher exact test for binominal variables (i.e. contamination rates, hot/cold-water systems) where appropriate. Odds ratios were reported as measures of association. Student’s t-test and Mann–Whitney-U-test were calculated to detected differences between numerical variables according to the underlying distribution. The significance level was set at 0.05. Statistical analysis was performed with PASW Statistics Version 18 (SPSS Inc., Chicago, Illinois, USA).
2.3. Identification of Legionella spp.
2.6. Ethics statement
Presumptive Legionella spp. colonies were sub-cultured on GVPC agar and Columbia blood agar. Colonies that grew on
Permits are not required to collect water samples from public water sources in Gabon.
Table 1 Samples screened for contamination with Legionella spp. in Gabon in 2013. Place
Province
Number of samples (improved/unimproved)
Mean temperature of water (°C) ± SD
Detection of Legionella sp. (n)
Bengue Fougamou Lambaréné Libreville Lopé Makouké Sindara
Moyen-Ogooué Ngounié Moyen-Ogooué Estuaire Ogooué-Ivindo Moyen-Ogooué Ngounié
4 (2/2) 10 (7/3) 175 (169/6) 2 (1/1) 2 (1/1) 4 (3/1) 3 (1/2)
28.5 ± 1.2 26.7 ± 1.5 32.8 ± 11.8 27.8 ± 0.4 28.3 ± 0 28.4 ± 1.4 27.9 ± 1.3
1 0 22 0 0 0 0
Please cite this article in press as: Ehrhardt, J., et al. Population structure of Legionella spp. from environmental samples in Gabon, 2013. Infect. Genet. Evol. (2015), http://dx.doi.org/10.1016/j.meegid.2015.05.019
J. Ehrhardt et al. / Infection, Genetics and Evolution xxx (2015) xxx–xxx
3
Fig. 1. Phylogenetic relation of Legionella pneumophila from Gabon. A neighbor joining tree was constructed using the concatenated MLST sequences of the Gabonese L. pneumophila isolates (A). As a reference, the concatenated MLST sequences from isolates originating from other African countries were extracted from the Legionella SBT database (d). Bootstrap scores of P75% are shown. The sites where the major phylogenetic groups were found are indicated in the map (B).
3. Results In total, 200 water samples were taken from two hospitals including their dental units (n = 49 samples), three health care centers including one HIV outpatient clinic (n = 12), five hotels (n = 35), private apartments (n = 54), public water access points (n = 40) and open water bodies (n = 10) in four provinces in Gabon (Table 1). In total, 29 Legionella spp. were found in 23 samples (11.6%), the median concentration was 11 CFU/100 ml. Legionella spp. was mainly found in hot water systems (OR = 21.9, 95% CI: 7.7–62.1, p < 0.001). Similarly, the median CFU count was significantly higher in samples from hot water systems compared to cold water systems (32 CFU/100 ml vs. 2.2 CFU/100 ml, p = 0.02).
All species identified with MALDI-TOF were either confirmed by the L. pneumophila specific real-time PCR or 16S RNA gene sequencing (Supplementary material, Table S1). From six samples, two phenotypically different isolates were detected which were either different species (L. pneumophila and L. dumoffii, n = 5) or the same species (L. pneumophila) but different STs (ST1911 and ST1912, n = 1). Positive samples were collected from showers (n = 11), water taps (n = 10), one borehole and the oral irrigator of one dental chair. Most of these samples were collected from the hospital setting (n = 11, 47.8%). The predominant species was L. pneumophila (n = 22), followed by L. dumoffii (n = 6) and L. anisa (n = 1). The L. dumoffii isolates were found in one hospital (n = 5) and in one borehole (n = 1). The L. anisa isolate was found in the hospital setting. The predominant serogroup of L.
Please cite this article in press as: Ehrhardt, J., et al. Population structure of Legionella spp. from environmental samples in Gabon, 2013. Infect. Genet. Evol. (2015), http://dx.doi.org/10.1016/j.meegid.2015.05.019
4
J. Ehrhardt et al. / Infection, Genetics and Evolution xxx (2015) xxx–xxx
pneumophila was 2–14 (n = 19), while others were also found (serogroup 1, n = 2 and serogroup other than 1–14, n = 1). We detected three new STs (ST1911, ST1912, ST1919) applying the SBT scheme which were unrelated to known STs from Africa (Fig. 1). In Lambaréné, the capital of the province ‘‘Moyen-Ogoo ué’’, we found distinct geographical clusters of related STs. Isolates of group 1 (ST1) and group 2 (ST75, ST1712) were solely detected in one hospital (Fig. 1). In contrast, isolates forming group 3 (ST1911, ST1919) were found only in samples from the stream island ‘‘Atongowanga’’ and from areas north of the river Ogooué. Isolates belonging to ST1 (group 1) cluster with other isolates from Africa (e.g. ST1735, ST152, ST7, ST1020). On the contrary, isolates belonging to group 2 and 3 form separate clades which are unrelated to known African L. pneumophila lineages.
4. Discussion To the best of our knowledge, this is the first study on Legionella spp. from environmental samples in Central Africa. We found a high prevalence of Legionella spp. in water samples from hospitals. These isolates displayed geographical clusters. Outbreaks are usually associated with aerosols created at open water sources (Whiley et al., 2014). Therefore, the frequent detection of Legionella spp. in water samples from hospital settings including a dental clinic could pose a risk for legionellosis. There is some evidence that dental instrumentation can be the source of Legionella sp. infections (Ricci et al., 2012). This is particularly true considering the median concentration of Legionella spp. isolates in our samples (11 CFU/100 ml) since the infectious dose is between 2.3 and 47 CFU (Whiley et al., 2014). The frequent colonization of hot water systems with Legionella spp. might point towards a marginal effect of hot water on prevention in our setting. On the other hand, the water temperature might not have been maintained above 55 °C in the whole hot water system which is mandatory for the control of colonization with Legionella spp. (Roig et al., 2003). In contrast to the hospital setting, domestic water sources do not seem to be a significant reservoir of Legionella spp. in our setting. Isolates belonging to ST1 are found around the globe and can be considered to be pandemic. Also in our dataset, ST1 was predominant (Fig. 1). More than half of the ST1 isolates published in the Legionella database are from clinical samples and it was suggested that isolates belonging to ST1 might have a higher pathogenic potential (Kozak-Muiznieks et al., 2014). Reports from other countries suggest that L. pneumophila belonging to ST75 and ST1712 (group 2) have also an increased ability to cause disease as 44% of all isolates belonging to these STs are from infection (www.hpa-bioinformatics.org.uk/legionella/legionella_sbt/php/sbt_homepage.php). In contrast, the pathogenic potential of isolates belonging to group 3 is unclear as only one single locus variant (ST1317) of ST1911 and ST1919 has been reported from environmental samples from South Africa (www.hpa-bioinformatics.org.uk/legionella/legionella_sbt/php/sbt_homepage.php). The frequent detection of STs which are usually associated with infection (i.e. ST1, ST75) suggests that our Legionella spp. isolates from environmental samples could be of clinical relevance. Future studies should therefore address if Legionella isolates from aquatic environments in Gabon have an impact on the morbidity and mortality in the local population. Although our study provides important data about Legionella spp. in Gabon, some limitations need to be addressed. First, our strain collection is not representative for the whole country as only a limited number of sites were sampled. Second, the distribution of
different samples is not well balanced and might be subject to a bias towards improved water sources (e.g. piped water systems). Finally, we were not able to address the pathogenic potential of our isolates in more depth as clinical isolates for comparison are not yet available in our setting. In conclusion, Legionella spp. was frequently cultured from hot water pipes in the investigated hospitals. This could pose a risk of legionellosis for aged and immunosuppressed hospitalized patients. Genotyping revealed geographical clusters. Acknowledgment This study was supported by Grant EI 247/8-1 from the Deutsche Forschungsgemeinschaft. We thank Damayanti Kaiser and Melanie Bach for indispensable technical assistance. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.meegid.2015.05. 019. References Alabi, A., Frielinghaus, L., Kaba, H., Kosters, K., Huson, M.A., Kahl, B., Peters, G., Grobusch, M., Issifou, S., Kremsner, P., Schaumburg, F., 2013. Retrospective analysis of antimicrobial resistance and bacterial spectrum of infection in Gabon, Central Africa. BMC Infect. Dis. 13, 455. Alli, A.O., Olusoga, O.D., Adedokum, S.A., Ogundare, O.E., 2011. Isolation of Legionella pneumophila from surface and ground waters in Osogbo. Niger. Afr. J. Microbiol. Res. 5, 2779–2785. Bartie, C., Venter, S.N., Nel, L.H., 2003. Identification methods for Legionella from environmental samples. Water Res. 37, 1362–1370. Becker, K., Harmsen, D., Mellmann, A., Meier, C., Schumann, P., Peters, G., von Eiff, C., 2004. Development and evaluation of a quality-controlled ribosomal sequence database for 16S ribosomal DNA-based identification of Staphylococcus species. J. Clin. Microbiol. 42, 4988–4995. Centers for Disease Control and Prevention, 2005. Procedures for the Recovery of Legionella From the Environment. U.S. Department of Health and Human Services, Atlanta, Georgia, USA. Dobrowsky, P.H., De Kwaadsteniet, M., Cloete, T.E., Khan, W., 2014. Distribution of indigenous bacterial pathogens and potential pathogens associated with roofharvested rainwater. Appl. Environ. Microbiol. 80, 2307–2316. Gaia, V., Fry, N.K., Afshar, B., Lück, P.C., Meugnier, H., Etienne, J., Peduzzi, R., Harrison, T.G., 2005. Consensus sequence-based scheme for epidemiological typing of clinical and environmental isolates of Legionella pneumophila. J. Clin. Microbiol. 43, 2047–2052. Helbig, J., Bernander, S., Castellani Pastoris, M., Etienne, J., Gaia, V., Lauwers, S., Lindsay, D., Lück, P., Marques, T., Mentula, S., Peeters, M., Pelaz, C., Struelens, M., Uldum, S., Wewalka, G., Harrison, T., 2002. Pan-european study on cultureproven Legionnaires’ disease: distribution of Legionella pneumophila serogroups and monoclonal subgroups. Eur. J. Clin. Microbiol. Infect. Dis. 21, 710–716. International Organization for Standardization (ISO), 1998. Water Quality – Detection and Enumeration of Legionella ISO 11731. International Organization for Standardization, Geneva, Switzerland. Kim, C., Nyoka, R., Ahmed, J., Winchell, J., Mitchell, S., Kariuki Njenga, M., Auko, E., Burton, W., Breiman, R., Eidex, R., 2012. Epidemiology of respiratory infections caused by atypical bacteria in two Kenyan refugee camps. J. Immigr. Minor. Health 14, 140–145. Kozak-Muiznieks, N.A., Lucas, C.E., Brown, E., Pondo, T., Taylor, T.H., Frace, M., Miskowski, D., Winchell, J.M., 2014. Prevalence of sequence types among clinical and environmental isolates of Legionella pneumophila serogroup 1 in the United States from 1982 to 2012. J. Clin. Microbiol. 52, 201–211. Lassmann, B., Poetschke, M., Ninteretse, B., Issifou, S., Winkler, S., Kremsner, P.G., Graninger, W., Apfalter, P., 2008. Community-acquired pneumonia in children in Lambarene, Gabon. Am. J. Trop. Med. Hyg. 79, 109–114. Mentasti, M., Underwood, A., Lück, C., Kozak-Muiznieks, N.A., Harrison, T.G., Fry, N.K., 2014. Extension of the Legionella pneumophila sequence-based typing scheme to include strains carrying a variant of the N-acylneuraminate cytidylyltransferase gene. Clin. Microbiol. Infect. 20, O435–O441. Muder, R.R., 2000. Other Legionella species. In: Mandell, G.L., Bennett, J.E., Dolin, R. (Eds.), Principles and Practice of Infectious Diseases. Churchill Livingstone, Philadelphia, Pennsylvania, USA, pp. 2435–2441. Potgieter, P.D., Hammond, J.M., 1992. Etiology and diagnosis of pneumonia requiring ICU admission. Chest 101, 199–203.
Please cite this article in press as: Ehrhardt, J., et al. Population structure of Legionella spp. from environmental samples in Gabon, 2013. Infect. Genet. Evol. (2015), http://dx.doi.org/10.1016/j.meegid.2015.05.019
J. Ehrhardt et al. / Infection, Genetics and Evolution xxx (2015) xxx–xxx Ricci, M.L., Fontana, S., Pinci, F., Fiumana, E., Pedna, M.F., Farolfi, P., Sabattini, M.A.B., Scaturro, M., 2012. Pneumonia associated with a dental unit waterline. Lancet 379, 684. Roig, J., Sabria, M., Pedro-Botet, M.L., 2003. Legionella spp.: community acquired and nosocomial infections. Curr. Opin. Infect. Dis. 16, 145–151.
5
Underwood, A., Jones, G., Mentasti, M., Fry, N., Harrison, T., 2013. Comparison of the Legionella pneumophila population structure as determined by sequence-based typing and whole genome sequencing. BMC Microbiol. 13, 302. Whiley, H., Keegan, A., Fallowfield, H., Ross, K., 2014. Uncertainties associated with assessing the public health risk from Legionella. Front. Microbiol., 5
Please cite this article in press as: Ehrhardt, J., et al. Population structure of Legionella spp. from environmental samples in Gabon, 2013. Infect. Genet. Evol. (2015), http://dx.doi.org/10.1016/j.meegid.2015.05.019
Contents lists available at ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Short communication
Population structure of Legionella spp. from environmental samples in Gabon, 2013 Jonas Ehrhardt a,b, Abraham S. Alabi a,b, Thorsten Kuczius c, Francis Foguim Tsombeng b, Karsten Becker d, Peter G. Kremsner a,b, Frieder Schaumburg b,d,⇑,1, Meral Esen a,b,1 a
Institut für Tropenmedizin, Eberhard Karls Universität Tübingen, Deutsches Zentrum für Infektionsforschung, Tübingen, Germany Centre de Recherches Médicales de Lambaréné (CERMEL), Lambaréné, Gabon Institute of Hygiene, University Hospital Münster, Münster, Germany d Institute of Medical Microbiology, University Hospital Münster, Münster, Germany b c
a r t i c l e
i n f o
Article history: Received 31 March 2015 Received in revised form 13 May 2015 Accepted 18 May 2015 Available online xxxx Keywords: Legionella Africa Gabon Population structure Hospital
a b s t r a c t Aquatic environments are the most important source for Legionella spp. infections such as Legionnaires’ disease and Pontiac fever. The reservoirs of Legionella spp. are mostly unclear in sub-Saharan Africa. The aim of this study, conducted in 2013, was to identify geographical areas of an increased risk for exposure to Legionella spp., and to describe the population structure of Legionella spp. from different water sources in a cross-sectional study in Gabon. Fresh water samples (n = 200) were cultured on Legionella selective agar; species were confirmed by MALDI-TOF, a Legionella pneumophila specific real-time PCR and 16S RNA gene sequencing. Serogroups were identified by agglutination test. The population structure was assessed by multilocus sequence typing (MLST). Legionella spp. isolates (n = 29) were frequently found in the hospital setting particularly in hot water systems. Open water bodies (i.e. rivers, lakes) were not contaminated with Legionella spp. Isolated L. pneumophila mainly belonged to serogroups 2–14 (n = 19) and MLST sequence type ST1, ST75 (and related STs) and ST1911. In conclusion, hospitalized patients might have an increased risk to become infected with Legionella spp. in the studied areas in Gabon, particularly if they have risk factors such as comorbidities. Both broadly extended (ST1, ST75) and local lineages (ST1911) were present in our setting. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Natural reservoirs of Legionella spp. are aquatic environments. Humans usually get infected by the inhalation of contaminated aerosols and might develop the severe pneumonic Legionnaires’ disease or the self-limited Pontiac fever (Whiley et al., 2014). Risk factors are comorbidities, immunosuppression and age (Whiley et al., 2014). Legionella pneumophila is the most common species causing legionellosis with a predominance of serogroup 1 which accounts for almost 80% of isolates from Legionnaires’ disease (Helbig et al., 2002). Other Legionella spp. (i.e. Legionella dumoffii, Legionella anisa) can also cause pneumonic infections (Muder, 2000).
⇑ Corresponding author at: Institute of Medical Microbiology, University Hospital Münster, Münster, Germany. E-mail address: [email protected] (F. Schaumburg). 1 Shared senior authorship.
Epidemiological data on Legionella spp. from environmental and clinical samples are limited and partly contradicting in sub-Saharan Africa. Few studies point towards a high prevalence in surface water in South Africa (73–75%) and Nigeria (67.3%) (Alli et al., 2011; Bartie et al., 2003; Dobrowsky et al., 2014). In contrast, legionellosis is infrequently reported. No case of Legionella infection was detected in patients with community acquired pneumonia in Gabon and only one case was detected in patients with influenza-like illness or severe acute respiratory infection in Kenya respectively (Kim et al., 2012; Lassmann et al., 2008). This is in contrast to South Africa where Legionella spp. was the fifth most frequent cause of pneumonia in patients that required hospital admission (Potgieter and Hammond, 1992). Knowledge about reservoirs of Legionella spp. is important to assess the potential risk of infection particularly in older people with comorbidities and to develop strategies for infection control and prevention. As this information is missing for Central Africa, we performed a cross sectional study on the burden of Legionella spp. in water bodies in Gabon. The aim was to identify
http://dx.doi.org/10.1016/j.meegid.2015.05.019 1567-1348/Ó 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: Ehrhardt, J., et al. Population structure of Legionella spp. from environmental samples in Gabon, 2013. Infect. Genet. Evol. (2015), http://dx.doi.org/10.1016/j.meegid.2015.05.019
2
J. Ehrhardt et al. / Infection, Genetics and Evolution xxx (2015) xxx–xxx
geographical areas of increased risk for exposure and to describe the population structure of Legionella spp. in the aquatic environment in Gabon. 2. Materials and methods 2.1. Water samples Water samples were drawn from randomly selected water sources at seven sites in four provinces in Gabon (Estuaire, Moyen Ogooué, Ogooué-Ivindo, Ngounié) from April to October 2013 (Table 1). Samples were taken from households or institutions after receiving permission from the person in charge. From each sampling site, 500 ml of water were collected in sterile, screw capped glass bottles containing a final concentration of 0.16 mM sodium thiosulphate. Samples from water taps and standpipes were aseptically collected after water was allowed to run for at least 1 min. In case of hot water systems, two samples were drawn (first cold then hot water). Samples were stored in a cool box and analyzed within 4 h after sampling. For each sample, the type of water source (e.g. tap water, shower, standpipe, borehole, open water body) and temperature of the water sample were recorded. The study protocol was approved by our local scientific review committee (SRC 2013.01). 2.2. Sample processing and culture Trained laboratory technicians performed the microbiological analysis of samples. The laboratory successfully participates in regular external quality assessments (Alabi et al., 2013). Water samples were analyzed using a modified protocol of the International Organization for Standardization (ISO 11731 and ISO 11731-2) (International Organization for Standardization (ISO), 1998). Briefly, 500 ml of each water sample was filtered using a 0.45 lm membrane filter (Millipore Merck, Billerica, Massachusetts, USA). To suppress the growth of non-Legionella spp. organisms, the filter was treated with 20 ml of 0.2 M hydrochloride/sodium chloride buffer (pH 2.2) for 5 min and was rinsed with sterile water. The membrane was shaken vigorously in 3 ml of sterile water to wash the organisms from the membrane. Of this, a volume of 0.75 ml was cultured on GVPC agar (Oxoid, Wesel, Germany) at 37 °C for up to 10 days under microaerophilic conditions. Fluorescence of colonies was examined under UV-light (254 nm/366 nm) after 3, 6, and 10 days. Contamination with Legionella spp. was reported as colony forming units (CFU) per 100 ml of unfiltered water. As concentrations of non-Legionella spp. and suspended matters were high in samples from open water bodies (i.e. borehole, river, lake), we filtered 100 ml instead of 500 ml of these samples.
Columbia blood agar were considered cysteine non-dependent and reported as non-Legionella spp. (Centers For Disease Control and Prevention, 2005). Serogrouping was performed on pure cultures using a latex agglutination test which differentiates between serogroup 1, serogroup 2–14 and seven non-L. pneumophila species (Legionella longbeachae, Legionella bozemanii, Legionella dumoffii, Legionella gormanii, Legionella jordanis, Legionella micdadei and Legionella anisa, Oxoid, Wesel, Germany). Species identification was done in Germany using MALDI-TOF mass spectrometry (Biotyper, Bruker, Bremen, Germany) and the MALDI Biotyper library (Version 3.3.2.0). Species of L. pneumophila was confirmed by a L. pneumophila specific real-time PCR (GeneProof, medac Diagnostika, Wedel, Germany). All non-L. pneumophila species were confirmed with 16S RNA gene sequencing (Becker et al., 2004). 2.4. Genotyping From all confirmed Legionella spp. isolates, DNA was extracted with QIAmp DNA Mini Kit (Qiagen, Herne, Germany). Multilocus sequence typing (MLST) was performed for all L. pneumophila isolates using the protocol developed by the European Working Group for Legionella Infections (EWGLI) published by the ESCMID study group for Legionella infections (Gaia et al., 2005; Mentasti et al., 2014). MLST sequence types were deduced using the sequence based typing (SBT) Database of L. pneumophila (www.hpa-bioinformatics.org.uk/legionella/legionella_sbt/php/sbt_homepage.php) hosted at the Health Protection Agency, England. The SBT scheme has a good correlation with genetic lineages based on whole genome sequencing (Underwood et al., 2013). To assess the phylogenetic relation of our L. pneumophila isolates with African isolates published in the Legionella SBT Database (assessed 9 January 2015, one isolate per ST per country), we constructed a neighbor joining tree using the concatenated sequences of the seven MLST genes of the Legionella typing scheme and MEGA5 (www.megasoftware.net). 2.5. Statistics Comparisons between groups were performed by means of the Chi2-test and the Fisher exact test for binominal variables (i.e. contamination rates, hot/cold-water systems) where appropriate. Odds ratios were reported as measures of association. Student’s t-test and Mann–Whitney-U-test were calculated to detected differences between numerical variables according to the underlying distribution. The significance level was set at 0.05. Statistical analysis was performed with PASW Statistics Version 18 (SPSS Inc., Chicago, Illinois, USA).
2.3. Identification of Legionella spp.
2.6. Ethics statement
Presumptive Legionella spp. colonies were sub-cultured on GVPC agar and Columbia blood agar. Colonies that grew on
Permits are not required to collect water samples from public water sources in Gabon.
Table 1 Samples screened for contamination with Legionella spp. in Gabon in 2013. Place
Province
Number of samples (improved/unimproved)
Mean temperature of water (°C) ± SD
Detection of Legionella sp. (n)
Bengue Fougamou Lambaréné Libreville Lopé Makouké Sindara
Moyen-Ogooué Ngounié Moyen-Ogooué Estuaire Ogooué-Ivindo Moyen-Ogooué Ngounié
4 (2/2) 10 (7/3) 175 (169/6) 2 (1/1) 2 (1/1) 4 (3/1) 3 (1/2)
28.5 ± 1.2 26.7 ± 1.5 32.8 ± 11.8 27.8 ± 0.4 28.3 ± 0 28.4 ± 1.4 27.9 ± 1.3
1 0 22 0 0 0 0
Please cite this article in press as: Ehrhardt, J., et al. Population structure of Legionella spp. from environmental samples in Gabon, 2013. Infect. Genet. Evol. (2015), http://dx.doi.org/10.1016/j.meegid.2015.05.019
J. Ehrhardt et al. / Infection, Genetics and Evolution xxx (2015) xxx–xxx
3
Fig. 1. Phylogenetic relation of Legionella pneumophila from Gabon. A neighbor joining tree was constructed using the concatenated MLST sequences of the Gabonese L. pneumophila isolates (A). As a reference, the concatenated MLST sequences from isolates originating from other African countries were extracted from the Legionella SBT database (d). Bootstrap scores of P75% are shown. The sites where the major phylogenetic groups were found are indicated in the map (B).
3. Results In total, 200 water samples were taken from two hospitals including their dental units (n = 49 samples), three health care centers including one HIV outpatient clinic (n = 12), five hotels (n = 35), private apartments (n = 54), public water access points (n = 40) and open water bodies (n = 10) in four provinces in Gabon (Table 1). In total, 29 Legionella spp. were found in 23 samples (11.6%), the median concentration was 11 CFU/100 ml. Legionella spp. was mainly found in hot water systems (OR = 21.9, 95% CI: 7.7–62.1, p < 0.001). Similarly, the median CFU count was significantly higher in samples from hot water systems compared to cold water systems (32 CFU/100 ml vs. 2.2 CFU/100 ml, p = 0.02).
All species identified with MALDI-TOF were either confirmed by the L. pneumophila specific real-time PCR or 16S RNA gene sequencing (Supplementary material, Table S1). From six samples, two phenotypically different isolates were detected which were either different species (L. pneumophila and L. dumoffii, n = 5) or the same species (L. pneumophila) but different STs (ST1911 and ST1912, n = 1). Positive samples were collected from showers (n = 11), water taps (n = 10), one borehole and the oral irrigator of one dental chair. Most of these samples were collected from the hospital setting (n = 11, 47.8%). The predominant species was L. pneumophila (n = 22), followed by L. dumoffii (n = 6) and L. anisa (n = 1). The L. dumoffii isolates were found in one hospital (n = 5) and in one borehole (n = 1). The L. anisa isolate was found in the hospital setting. The predominant serogroup of L.
Please cite this article in press as: Ehrhardt, J., et al. Population structure of Legionella spp. from environmental samples in Gabon, 2013. Infect. Genet. Evol. (2015), http://dx.doi.org/10.1016/j.meegid.2015.05.019
4
J. Ehrhardt et al. / Infection, Genetics and Evolution xxx (2015) xxx–xxx
pneumophila was 2–14 (n = 19), while others were also found (serogroup 1, n = 2 and serogroup other than 1–14, n = 1). We detected three new STs (ST1911, ST1912, ST1919) applying the SBT scheme which were unrelated to known STs from Africa (Fig. 1). In Lambaréné, the capital of the province ‘‘Moyen-Ogoo ué’’, we found distinct geographical clusters of related STs. Isolates of group 1 (ST1) and group 2 (ST75, ST1712) were solely detected in one hospital (Fig. 1). In contrast, isolates forming group 3 (ST1911, ST1919) were found only in samples from the stream island ‘‘Atongowanga’’ and from areas north of the river Ogooué. Isolates belonging to ST1 (group 1) cluster with other isolates from Africa (e.g. ST1735, ST152, ST7, ST1020). On the contrary, isolates belonging to group 2 and 3 form separate clades which are unrelated to known African L. pneumophila lineages.
4. Discussion To the best of our knowledge, this is the first study on Legionella spp. from environmental samples in Central Africa. We found a high prevalence of Legionella spp. in water samples from hospitals. These isolates displayed geographical clusters. Outbreaks are usually associated with aerosols created at open water sources (Whiley et al., 2014). Therefore, the frequent detection of Legionella spp. in water samples from hospital settings including a dental clinic could pose a risk for legionellosis. There is some evidence that dental instrumentation can be the source of Legionella sp. infections (Ricci et al., 2012). This is particularly true considering the median concentration of Legionella spp. isolates in our samples (11 CFU/100 ml) since the infectious dose is between 2.3 and 47 CFU (Whiley et al., 2014). The frequent colonization of hot water systems with Legionella spp. might point towards a marginal effect of hot water on prevention in our setting. On the other hand, the water temperature might not have been maintained above 55 °C in the whole hot water system which is mandatory for the control of colonization with Legionella spp. (Roig et al., 2003). In contrast to the hospital setting, domestic water sources do not seem to be a significant reservoir of Legionella spp. in our setting. Isolates belonging to ST1 are found around the globe and can be considered to be pandemic. Also in our dataset, ST1 was predominant (Fig. 1). More than half of the ST1 isolates published in the Legionella database are from clinical samples and it was suggested that isolates belonging to ST1 might have a higher pathogenic potential (Kozak-Muiznieks et al., 2014). Reports from other countries suggest that L. pneumophila belonging to ST75 and ST1712 (group 2) have also an increased ability to cause disease as 44% of all isolates belonging to these STs are from infection (www.hpa-bioinformatics.org.uk/legionella/legionella_sbt/php/sbt_homepage.php). In contrast, the pathogenic potential of isolates belonging to group 3 is unclear as only one single locus variant (ST1317) of ST1911 and ST1919 has been reported from environmental samples from South Africa (www.hpa-bioinformatics.org.uk/legionella/legionella_sbt/php/sbt_homepage.php). The frequent detection of STs which are usually associated with infection (i.e. ST1, ST75) suggests that our Legionella spp. isolates from environmental samples could be of clinical relevance. Future studies should therefore address if Legionella isolates from aquatic environments in Gabon have an impact on the morbidity and mortality in the local population. Although our study provides important data about Legionella spp. in Gabon, some limitations need to be addressed. First, our strain collection is not representative for the whole country as only a limited number of sites were sampled. Second, the distribution of
different samples is not well balanced and might be subject to a bias towards improved water sources (e.g. piped water systems). Finally, we were not able to address the pathogenic potential of our isolates in more depth as clinical isolates for comparison are not yet available in our setting. In conclusion, Legionella spp. was frequently cultured from hot water pipes in the investigated hospitals. This could pose a risk of legionellosis for aged and immunosuppressed hospitalized patients. Genotyping revealed geographical clusters. Acknowledgment This study was supported by Grant EI 247/8-1 from the Deutsche Forschungsgemeinschaft. We thank Damayanti Kaiser and Melanie Bach for indispensable technical assistance. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.meegid.2015.05. 019. References Alabi, A., Frielinghaus, L., Kaba, H., Kosters, K., Huson, M.A., Kahl, B., Peters, G., Grobusch, M., Issifou, S., Kremsner, P., Schaumburg, F., 2013. Retrospective analysis of antimicrobial resistance and bacterial spectrum of infection in Gabon, Central Africa. BMC Infect. Dis. 13, 455. Alli, A.O., Olusoga, O.D., Adedokum, S.A., Ogundare, O.E., 2011. Isolation of Legionella pneumophila from surface and ground waters in Osogbo. Niger. Afr. J. Microbiol. Res. 5, 2779–2785. Bartie, C., Venter, S.N., Nel, L.H., 2003. Identification methods for Legionella from environmental samples. Water Res. 37, 1362–1370. Becker, K., Harmsen, D., Mellmann, A., Meier, C., Schumann, P., Peters, G., von Eiff, C., 2004. Development and evaluation of a quality-controlled ribosomal sequence database for 16S ribosomal DNA-based identification of Staphylococcus species. J. Clin. Microbiol. 42, 4988–4995. Centers for Disease Control and Prevention, 2005. Procedures for the Recovery of Legionella From the Environment. U.S. Department of Health and Human Services, Atlanta, Georgia, USA. Dobrowsky, P.H., De Kwaadsteniet, M., Cloete, T.E., Khan, W., 2014. Distribution of indigenous bacterial pathogens and potential pathogens associated with roofharvested rainwater. Appl. Environ. Microbiol. 80, 2307–2316. Gaia, V., Fry, N.K., Afshar, B., Lück, P.C., Meugnier, H., Etienne, J., Peduzzi, R., Harrison, T.G., 2005. Consensus sequence-based scheme for epidemiological typing of clinical and environmental isolates of Legionella pneumophila. J. Clin. Microbiol. 43, 2047–2052. Helbig, J., Bernander, S., Castellani Pastoris, M., Etienne, J., Gaia, V., Lauwers, S., Lindsay, D., Lück, P., Marques, T., Mentula, S., Peeters, M., Pelaz, C., Struelens, M., Uldum, S., Wewalka, G., Harrison, T., 2002. Pan-european study on cultureproven Legionnaires’ disease: distribution of Legionella pneumophila serogroups and monoclonal subgroups. Eur. J. Clin. Microbiol. Infect. Dis. 21, 710–716. International Organization for Standardization (ISO), 1998. Water Quality – Detection and Enumeration of Legionella ISO 11731. International Organization for Standardization, Geneva, Switzerland. Kim, C., Nyoka, R., Ahmed, J., Winchell, J., Mitchell, S., Kariuki Njenga, M., Auko, E., Burton, W., Breiman, R., Eidex, R., 2012. Epidemiology of respiratory infections caused by atypical bacteria in two Kenyan refugee camps. J. Immigr. Minor. Health 14, 140–145. Kozak-Muiznieks, N.A., Lucas, C.E., Brown, E., Pondo, T., Taylor, T.H., Frace, M., Miskowski, D., Winchell, J.M., 2014. Prevalence of sequence types among clinical and environmental isolates of Legionella pneumophila serogroup 1 in the United States from 1982 to 2012. J. Clin. Microbiol. 52, 201–211. Lassmann, B., Poetschke, M., Ninteretse, B., Issifou, S., Winkler, S., Kremsner, P.G., Graninger, W., Apfalter, P., 2008. Community-acquired pneumonia in children in Lambarene, Gabon. Am. J. Trop. Med. Hyg. 79, 109–114. Mentasti, M., Underwood, A., Lück, C., Kozak-Muiznieks, N.A., Harrison, T.G., Fry, N.K., 2014. Extension of the Legionella pneumophila sequence-based typing scheme to include strains carrying a variant of the N-acylneuraminate cytidylyltransferase gene. Clin. Microbiol. Infect. 20, O435–O441. Muder, R.R., 2000. Other Legionella species. In: Mandell, G.L., Bennett, J.E., Dolin, R. (Eds.), Principles and Practice of Infectious Diseases. Churchill Livingstone, Philadelphia, Pennsylvania, USA, pp. 2435–2441. Potgieter, P.D., Hammond, J.M., 1992. Etiology and diagnosis of pneumonia requiring ICU admission. Chest 101, 199–203.
Please cite this article in press as: Ehrhardt, J., et al. Population structure of Legionella spp. from environmental samples in Gabon, 2013. Infect. Genet. Evol. (2015), http://dx.doi.org/10.1016/j.meegid.2015.05.019
J. Ehrhardt et al. / Infection, Genetics and Evolution xxx (2015) xxx–xxx Ricci, M.L., Fontana, S., Pinci, F., Fiumana, E., Pedna, M.F., Farolfi, P., Sabattini, M.A.B., Scaturro, M., 2012. Pneumonia associated with a dental unit waterline. Lancet 379, 684. Roig, J., Sabria, M., Pedro-Botet, M.L., 2003. Legionella spp.: community acquired and nosocomial infections. Curr. Opin. Infect. Dis. 16, 145–151.
5
Underwood, A., Jones, G., Mentasti, M., Fry, N., Harrison, T., 2013. Comparison of the Legionella pneumophila population structure as determined by sequence-based typing and whole genome sequencing. BMC Microbiol. 13, 302. Whiley, H., Keegan, A., Fallowfield, H., Ross, K., 2014. Uncertainties associated with assessing the public health risk from Legionella. Front. Microbiol., 5
Please cite this article in press as: Ehrhardt, J., et al. Population structure of Legionella spp. from environmental samples in Gabon, 2013. Infect. Genet. Evol. (2015), http://dx.doi.org/10.1016/j.meegid.2015.05.019
