Eur J Clin Microbiol Infect Dis (2016) 35:235–244 DOI 10.1007/s10096-015-2536-1
ORIGINAL ARTICLE
Detection of signature volatiles for cariogenic microorganisms M. Hertel 1 & R. Preissner 2 & B. Gillissen 3,4 & A. M. Schmidt-Westhausen 1 & S. Paris 5 & S. Preissner 5
Received: 29 October 2015 / Accepted: 18 November 2015 / Published online: 26 November 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract The development of a breath test by the identification of volatile organic compounds (VOCs) emitted by cariogenic bacteria is a promising approach for caries risk assessment and early caries detection. The aim of the present study was to investigate the volatile profiles of three major cariogenic bacteria and to assess whether the obtained signatures were species-specific. Therefore, the headspaces above cultures of Streptococcus mutans, Lactobacillus salivariusand Propionibacterium acidifaciens were analysed after 24 and 48 h of cultivation using gas chromatography and mass spectrometry. A volatile database was queried for the obtained VOC profiles. Sixty-four compounds were detected within the analysed culture headspaces and were absent (36) or at least only present in minor amounts (28) in the control headspace. For S. mutans 18, for L. salivarius three and for P. acidifaciens five compounds were found to be unique signature VOCs. Database matching revealed that the identified signatures of all bacteria were unique. Furthermore, 13 of the 64 detected substances have not been previously reported to be emitted by bacteria or fungi. Specific VOC signatures were
* S. Preissner [email protected] 1
Department of Oral Medicine, Dental Radiology and Oral Surgery, Charité - Universitätsmedizin Berlin, Berlin, Germany
2
Structural Bioinformatics Group, Institute for Physiology, Charité Universitätsmedizin Berlin, Berlin, Germany
3
German Cancer Consortium (DKTK), Heidelberg, Germany
4
Hematology, Oncology and Tumor Immunology, Charité Universitätsmedizin Berlin, Berlin, Germany
5
Department of Operative and Preventive Dentistry, Charité Universitätsmedizin Berlin, Aßmannshauser Str. 4-6, 14197 Berlin, Germany
found in all the investigated bacteria cultures. The obtained results encourage further research to investigate the transferability to in vivo conditions towards the development of a breath test.
Introduction Caries is one of the most frequent diseases worldwide. During the last decade, there has been a paradigm shift in the therapy of caries. For over 100 years, cavities were treated following the ‘extension for prevention’ concept of Black, involving removing sound tooth structure [1]. Minimal-invasive concepts were developed, based on on-going research and the improvement of restorative materials [2]. The idea of preserving tooth structure developed further and the focus is now on preventive dentistry and oral health maintenance. A new system to measure initial caries lesions called the International Caries Detection and Assessment System (ICDAS) was introduced in 2007 [3]. Furthermore, various non-invasive and micro-invasive therapy concepts have been introduced, such as nutrition management, fluoride application, sealing or infiltration [4]. An outstanding problem is caries risk assessment, as the only accurate predictor for future caries that is currently available is past caries experience [5]. The early detection of highrisk patients would facilitate individual prevention. In addition, costs might be saved extensively, because, currently, prevention is applied to collectives or groups, whereas only a minority of patients requires individual prophylaxis. A simple and reliable caries risk test is currently not available. Certain oral bacteria, especially Streptococcus mutans and Lactobacillus salivarius, have been reported to be the major causes of caries [6]. A recent in vivo study revealed that Propionibacterium acidifaciens is significantly more
236
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
Fig.
1 Signature compound assessment. VOC = volatile organic compound; GC/MS = gas chromatography/mass spectrometry. *= compound not evaluable due to superimposing GC/MS peaks
abundant in caries samples compared to other microorganisms [7]. Overall, analyses of the commensal microbiome revealed that it was indicative of a variety of pathological conditions [8], but a comprehensive sequencing analysis is elaborate and expensive. The detection of volatile organic compounds (VOCs) emitted by pathogenic microbiota has been performed [9]. Although gas chromatography and mass spectrometry (GC/ MS), which provide high reliability and accuracy, have been used to profile the VOCs emitted by a wide range of bacteria, inter alia Cyanobacteriae [10] and Rhizobacteriae [11]. Nevertheless, the VOCs emitted by cariogenic bacteria are unknown. Breath analyses are rapid and affordable, and are, therefore, already used in various disciplines; for instance, the identification of Helicobacter pylori via VOC detection is routinely carried out in gastroenterology [12]. Inflammation in the airways can be monitored by nitric oxide testing [13] and a breath test for Mycobacterium tuberculosis has been developed [14]. For caries screening and risk assessment, a breath test based on the detection of volatiles released by cariogenic bacteria would be promising. Additionally, breast tests are simple to perform, painless, non-invasive and repeatable, even in young children. Exposure to radiation might also be reduced, as screening for approximal caries lesions is predominantly performed using radiographic methods. The aim of the present study was to identify the VOCs released by the three major cariogenic bacteria, specifically S. mutans, L. salivarius and P. acidifaciens, using GC/MS. Furthermore, the study sought to investigate whether the obtained profiles were unique, by searching the mVOC database [15]. It was hypothesised that species-specific volatile signatures can be detected in the culture headspaces of the investigated bacteria.
Materials and methods Bacteria cultivation Three reference microorganisms, namely, S. mutans, L. salivarius and P. acidifaciens (DSM 20523, 20555 and 21887, DSMZ, Braunschweig, Germany), were cultivated for 24 h at 30 °C on Brain Heart Infusion-agar (BHI) (AppliChem, Darmstadt, Germany) plates under microaerophilic conditions. Subsequently, two colony-forming units (CFUs) per strain were transferred into 3.5 mL liquid BHI (VWR International, Darmstadt, Germany) and Erlenmeyer flasks containing 50 mL solid BHI were inoculated.
237
Incubation at 30 °C for 24 h was performed in microaerophilic conditions after sealing both open ends of the flasks using polytetrafluoroethylene (PTFE) plugs. Before headspace sampling, a standardised biochemical identification system (API 20 Strep, bioMérieux, Nürtingen, Germany) was used to confirm the genus of the cultured bacteria. Headspace sampling Following incubation, volatile compound sampling was carried out. To assess potential time-dependent differences in the VOC spectra, sampling was performed after 24 and 48 h. The headspaces above the bacteria cultures and from a blank BHI, which served as a control, were transferred to adsorbent (polyphenylene oxide containing) tubes (Tenax, Gerstel, Mühlheim, Germany). Therefore, a flexible PTFE hose was used to connect the culture flask and the Tenax tube. A second hose was installed to connect the Tenax tube to a suction pump (SG 350, GSA Messgerätebau, Neuss, Germany). Additionally, an activated carbon filter (Niosh Typ B, Dräger, Lübeck, Germany) was connected to the open end of the culture flask to avoid negative pressure. Subsequently, a total volume of 1 L was sucked through the Tenax tube in 10 min, applying a constant flow rate of 100 mL/min. Headspace analysis Headspace analysis was performed using a capillary gas chromatograph to which a mass spectrometer was coupled (6890N Network Gas Chromatograph System, Agilent Technologies, Santa Clara, CA, USA), with sample injection by means of thermal desorption. Closed-loop stripping analysis (CLSA) was utilized, using a 5 % phenyl methylpolysiloxane capillary column (HP-5ms, Agilent Technologies, Santa Clara, CA, USA). The applied acquisition mode conditions and parameters were as follows: Oven: initial temperature: 30 °C; initial time: 1.0 min; maximum temp: 350 °C; equilibration time: 0.5 min; post temperature: 30 °C; post time: 1.0 min; run time: 39.33 min; Inlet: mode: solvent vent; initial temperature: 250 °C; vent time: 0.1 min; vent flow: 50.0 mL/min; vent pressure: 41.0 psi; purge flow: 10.0 mL/min; purge time: 0.02 min; total flow: 14.5 mL/min; saver flow: 30.0 mL/min (gas type: helium); saver time: 1.00 min; Column: mode: constant flow; nominal length: 50 m; nominal inside diameter: 0.2 mm; nominal film thickness: 0.33 μm; maximum temperature: 350 °C; initial flow: 1.7 mL/min; initial pressure: 40.1 psi; average velocity: 36 cm/s; outlet pressure: vacuum; Temperature program: initial temperature: 30 °C; delay time: 1.1 min; rate: 60.0 °C/min; end temperature:
238 Table 1 Volatiles detected in Streptococcus mutans, Lactobacillus salivariusand Propionibacterium acidifaciens culture headspaces
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
VOC
CAS-Nr.
S.mutans
n-Hexane
110-54-3
n-Heptane
142-82-5
#
n-Octane
111-65-9
#
n-Nonane n-Pentadecane
111-82-2 629-62-9
# #
n-Heptadecane n-Octadecane
629-78-7 593-45-3
#
3-Methyl pentane
96-14-0
Toluene Ethyl benzene
108-88-3 100-41-4
* *
m-/p-Xylene
108-38-3
*
o-Xylene
106-42-3 95-47-7
*
1,3,5-Trimethyl benzene 1,2,4-Trimethyl benzene
108-67-8 95-63-6
# #
3-/4-Ethyl toluene
620-14-4
#
1,2,3,4-Tetrahydro naphthalene Naphthalene 2-Methyl naphthalene 1-Methyl naphthalene β-Pinene
622-96-8 119-64-2 91-20-3 91-57-6 92-12-0 19902-08-0
# * # # #
Delta-3-carene 2-Methyl-1-propanol 1-Butanol
18172-67-3 498-15-7 78-83-1 71-36-3
# * *
2-Ethyl-1-hexanol Benzyl alcohol Methyl acetate Ethyl acetate n-Propyl acetate
104-76-7 100-51-6 79-20-9 141-78-6 109-60-4
Isobutyl acetate n-Butyl acetate
110-19-0 123-86-4
Ethyl-2-methyl butyrate Methyl ethyl ketone Methyl isobutyl ketone Cyclohexanone Acetophenone 2-Methyl furane Ethylene glycol monobutyl ether Dipropylene glycol monomethyl ether Diethylene glycol monoethyl ether Diethylene glycol monobutyl ether 3-Methyl-1-butanal n-Hexanal n-Heptanal n-Octanal n-Nonanal n-Decanal n-Undecanal
7452-79-1 78-93-3 108-10-1 108-94-1 98-86-2 534-22-5 111-76-2 34590-94-8 111-90-0 112-34-5 590-86-3 66-25-1 111-71-7 124-13-0 124-19-6 112-31-2 112-44-7
L.salivarius
P.acidifaciens
* #
# #
* #
* # * *
* * * # *
*
#
*
# * *
* * #
* # * *
# *
*
# * *
* *
*
*
# # # # * * # * * # #
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
239
Table 1 (continued) VOC
CAS-Nr.
S.mutans
L.salivarius
P.acidifaciens
Benzaldehyde
100-52-7
*
Dimethyl sulfide
75-18-3
*
*
*
3-Methyl furane 3-Methyl-2-butanol
930-27-8 598-75-4
#
*
* #
2-Pentanone 3-Methyl-1-butanol
107-87-9 123-51-3
* #
* #
*
2-Methyl-1-butanol
137-32-6
#
#
Dimethyl disulfide 1-Hexanol
624-92-0 111-27-3
* #
2-Heptanone 1-Heptanol
110-43-0 110-70-6
* #
1-Octene-3-ol
3391-86-4
#
3-Octanone 3-Octanol
106-68-3 589-98-0
# #
#
1-Octanol
111-87-5
*
*
1-Nonanol 1-Decanol
143-08-8 112-30-1
# #
* *
Data retrieved from S. mutans and L. salivarius after 24 and 48 h were pooled #= volatile not found in the BHI control *= amounts detected in culture that exceeded those in the BHI control
260 °C; hold time: 8.00 min; transfer temperature: 300 °C; standby temperature: 50 °C. Selected ion monitoring (SIM)/scan mode was applied. A recorded, full-scan-mode chromatogram was used for qualitative analysis via library-assisted matching, and up to 150 peaks were automatically integrated within the chromatogram. Compounds were identified by the comparison of mass spectra to database spectra. Within our own database, all 227 tested substances were referenced as external standards based on multipoint calibration using at least six levels per substance. Based on algorithmic comparison (probability-based matching), the mass spectra of the obtained peaks were automatically assigned and reviewed by hand by an experienced investigator (S.P.). Finally, the matching reference spectra contained in the searched spectral libraries were assigned. To assess if volatiles that were detectable within both the cultures and the control exceeded the blank BHI, quantitative analysis was performed according to DIN ISO 16000-6. Compound quantification was performed by applying the method of external standards on comparative blends. VOC profiling Only compounds that were detected in the culture headspaces but not in the BHI control were included in the volatile profiling. Accordingly, VOC profiling was exclusively based on the results obtained from qualitative analysis. For this
purpose, the mVOC database [15], including PubMed, was searched for the identified signatures, in order to assess whether the VOCs had previously been identified in microbial cultures and to investigate whether species specificity was obtained. The obtained profiles were compared with the profiles of all bacteria and fungi emitting at least one of the detected volatiles. The date of access was 1 October 2015.
Results Headspace analysis Of the 227 tested substances, 120 were found to be below the detection limit and a further 27 compounds were detected within the culture headspaces, but their amounts did not exceed that of the control. Sixteen volatiles were not evaluable due to the superimposition of GC/MS peaks and were, therefore, excluded from further analysis. Thus, 64 substances (alkanes, aromatics, terpenes, monovalent alcohols, esters of monovalent alcohols, ketones, cyclic ethers, glycols, glycol esters, glycol ethers and aldehydes) were detected in the headspaces of the investigated cultures (Fig. 1 and Table 1). Ethanol was ubiquitously detected, but could not be quantified due to excessive concentration. Alkenes, chloronaphthalenes, chlorinated hydrocarbons, siloxanes, esters of fatty acids, esters of dicarboxylic acids, heterocyclic compounds, phenols and acrylates were not found, at least not in amounts that
240
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
Fig. 2
Chromatograms of Streptococcus mutans (a), Lactobacillus salivarius (b) and Propionibacterium acidifaciens headspaces (c) after 48 h of cultivation
exceeded those in the BHI control. Chromatograms of S. mutans, L. salivarius and P. acidifaciens are shown in the Fig. 2a–c.
241
investigated bacteria differed from the 226 profiles and, therefore, were unique. Hence, species specificity was obtained in all cases. The signature patterns of S. mutans consisted of 18 compounds, whereas that of L. salivarius and P. acidifaciens was characterised by the presence of three and five volatiles, respectively (Fig. 3 and Table 2). Peak assignment within the mass spectra revealed matching scores of 100 % in all 36 compounds exclusively detected in the bacteria cultures.
VOC profiling Fifty-one out of the 64 detected compounds were found by searching the mVOC database using their PubChem IDs for unequivocal identification. Hence, these substances have been reported to be emitted by different bacteria and/or fungi. Thirteen molecules were not previously reported to be released by microbial cultures (Fig. 1). Volatile profile determination was performed, excluding 28 out of the 64 substances, as minor amounts were also found in the control headspace. Thus, 36 VOCs that were exclusively found in the headspaces of S. mutans, L. salivarius and P. acidifaciens were compiled and the mVOC database was again searched for the obtained volatile profiles (Fig. 1). Overall, 226 microorganisms were found that emit at least one of the signature VOCs, and 39 of these bacteria and fungi potentially colonise or infect humans. Database matching revealed that the profiles found for all the
Discussion The aim of the present study was to identify the VOC profiles of S. mutans, L. salivarius and P. acidifaciens. The emission signatures obtained from solid cultures were specific for all three bacteria and allowed their differentiation. Nevertheless, certain compounds (phenyl ethanol, 3-methyl-1-butanol, 2methyl-1-butanol and 1-decanol) were found to be emitted by more than five microbiota in addition to cariogenic bacteria. Hence, excerpts of the retrieved volatile signatures might be sufficient for their identification. Before excerpting the obtained VOC profiles, the extent of transferability of the retrieved results to in vivo conditions should form the focus of subsequent research. Growth conditions, including oxygen
Fig. 3 Venn diagram of signature VOCs of S. mutans, L. salivarius and P. acidifaciens
242
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
Table 2 The volatile organic compound (VOC) signatures of S. mutans, L. salivarius and P. acidifaciens (displayed in blue) compared with bacteria and fungi that infect humans, retrieved from the mVOC database
concentration, nutrient supply, temperature and pH, have been reported to influence bacterial metabolism and, therefore, VOC emission [16]. Accordingly, volatile release was found to be time-dependent in all the investigated cultures. In the present study, one culture per species was investigated as reproducible and standardized growth conditions were applied. Thus, no variability in metabolism and, hence, VOC emission was expected. However, subsequent research might apply varying culture conditions
to assess their influence on the release of bacterial volatiles. Further studies might, therefore, consider performing VOC analysis by applying conditions that mimic the environment of oral cavities. Additionally, artificial caries lesions appear to be suitable for further preclinical research. Clinical approaches that perform breath analyses in healthy and diseased subjects might also be undertaken. In addition to transferability to in vivo conditions, the specificity, sensitivity and the positive and
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
the negative predictive values for the detection of early caries remain unknown. Even though VOC quantification is feasible using GC/MS, it is of inferior relevance compared to qualitative analysis, at least within the intent of the present study. Quantification was only used to allow comparison between amounts detected in the cultures and blank BHI. Furthermore, the VOCs quantity does not necessarily correlate with the cell number [17, 18]. Accordingly, neither the CFU count nor the optical density measurement were performed. Volatile compounds released by bacteria act as quorum-sensing molecules, regulating inter alia metabolism and growth. Furthermore, they mediate interactions between different microbiota. As an example, different sulphur compounds and long-chain ketones inhibit the growth of fungi [19]. Dimethyl disulfide, an active compound involved in the mentioned interactions, was found in our study within the headspaces of S. mutans and P. acidifaciens. The origin and the related metabolic pathways of the 13 compounds that were not found by searching the mVOC database remain uncertain, with the exception of n-octadecane, which has been described to be released by Microcystis aeruginosa [20]. This cyanobacterium is not found in humans. The remaining compounds, n-heptane, 1,3,5-trimethyl benzene, 3ethyl toluene, 1,2,3,4-tetrahydro naphthalene, 2-methyl naphthalene, 1-methyl naphthalene, ethylene glycol monobutyl ether, diethylene glycol monoethyl ether and diethylene glycol monobutyl ether have been described to be sources of carbon and energy for certain bacterial species [21–26]. As a wide range of glycol ethers can be metabolized by bacteria [21], it can be assumed that this includes dipropylene glycol monoethyl ether, even if this has not been reported. The compounds 3-methyl pentane and β-pinene could not be related to bacterial metabolism; nevertheless, even when these 13 substances were excluded from the obtained profiles due to their unknown origin, species specificity was maintained in all the investigated species. As a consequence, the identified VOCs could be seen as biomarkers for cariogenic bacteria, potentially allowing caries risk assessment or screening for initial caries lesions based on their detection using a breath test.
243 Acknowledgements The authors would like to thank Mss. Anja and Antje Richter for their most valuable assistance in the laboratory and Janette Nickel for her help with the mVOC database. Compliance with ethical standards Conflict of interest The authors state that they have no affiliations with or involvement in any organisation or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this article.
References 1. 2.
3.
4.
5.
6. 7.
8.
9.
10.
11.
12.
Conclusions 13.
As hypothesised, volatile profiles that were unique for Streptococcus mutans, Lactobacillus salivarius and Propionibacterium acidifaciens were obtained in the present study. Thus, developing a breath test based on volatile organic compounds (VOCs) identification remains a promising approach for caries risk assessment and early caries detection. These results encourage further studies to investigate the transferability of these findings to in vivo conditions.
14.
15.
16.
Osborne JW, Summitt JB (1998) Extension for prevention: is it relevant today? Am J Dent 11(4):189–196 Peters MC, McLean ME (2001) Minimally invasive operative care. I. Minimal intervention and concepts for minimally invasive cavity preparations. J Adhes Dent 3(1):7–16 Ismail AI, Sohn W, Tellez M, Amaya A, Sen A, Hasson H, Pitts NB (2007) The International Caries Detection and Assessment System (ICDAS): an integrated system for measuring dental caries. Community Dent Oral Epidemiol 35(3):170–178 Meyer-Lueckel H, Bitter K, Paris S (2012) Randomized controlled clinical trial on proximal caries infiltration: three-year follow-up. Caries Res 46(6):544–548 Mejàre I, Axelsson S, Dahlén G, Espelid I, Norlund A, Tranæus S, Twetman S (2014) Caries risk assessment. A systematic review. Acta Odontol Scand 72(2):81–91 Poole DF, Newman HN (1971) Dental plaque and oral health. Nature 234(5328):329–331 Wolff D, Frese C, Maier-Kraus T, Krueger T, Wolff B (2013) Bacterial biofilm composition in caries and caries-free subjects. Caries Res 47(1):69–77 Blaser M, Bork P, Fraser C, Knight R, Wang J (2013) The microbiome explored: recent insights and future challenges. Nat Rev Microbiol 11(3):213–217 Tait E, Perry JD, Stanforth SP, Dean JR (2014) Identification of volatile organic compounds produced by bacteria using HSSPME-GC-MS. J Chromatogr Sci 52(4):363–373 Höckelmann C, Jüttner F (2004) Volatile organic compound (VOC) analysis and sources of limonene, cyclohexanone and straight chain aldehydes in axenic cultures of Calothrix and Plectonema. Water Sci Technol 49(9):47–54 Farag MA, Ryu CM, Sumner LW, Paré PW (2006) GC-MS SPME profiling of rhizobacterial volatiles reveals prospective inducers of growth promotion and induced systemic resistance in plants. Phytochemistry 67(20):2262–2268 Peeters M (1998) Urea breath test: a diagnostic tool in the management of Helicobacter pylori-related gastrointestinal diseases. Acta Gastroenterol Belg 61(3):332–335 Smith AD, Cowan JO, Brassett KP, Herbison GP, Taylor DR (2005) Use of exhaled nitric oxide measurements to guide treatment in chronic asthma. N Engl J Med 352(21):2163–2173 Scott-Thomas A, Epton M, Chambers S (2013) Validating a breath collection and analysis system for the new tuberculosis breath test. J Breath Res 7(3):037108 Lemfack MC, Nickel J, Dunkel M, Preissner R, Piechulla B (2014) mVOC: a database of microbial volatiles. Nucleic Acids Res 42(Database issue):D744–D748 Blom D, Fabbri C, Connor EC, Schiestl FP, Klauser DR, Boller T, Eberl L, Weisskopf L (2011) Production of plant growth
244 modulating volatiles is widespread among rhizosphere bacteria and strongly depends on culture conditions. Environ Microbiol 13(11): 3047–3058 17. Zehm S, Schweinitz S, Würzner R, Colvin HP, Rieder J (2012) Detection of Candida albicans by mass spectrometric fingerprinting. Curr Microbiol 64(3):271–275 18. Bunge M, Araghipour N, Mikoviny T, Dunkl J, Schnitzhofer R, Hansel A, Schinner F, Wisthaler A, Margesin R, Märk TD (2008) On-line monitoring of microbial volatile metabolites by proton transfer reaction-mass spectrometry. Appl Environ Microbiol 74(7):2179–2186 19. Hunziker L, Bönisch D, Groenhagen U, Bailly A, Schulz S, Weisskopf L (2015) Pseudomonas strains naturally associated with potato plants produce volatiles with high potential for inhibition of Phytophthora infestans. Appl Environ Microbiol 81(3):821–830 20. Walsh K, Jones GJ, Dunstan RH (1998) Effect of high irradiance and iron on volatile odour compounds in the cyanobacterium Microcystis aeruginosa. Phytochemistry 49(5):1227–1239 21. Kawai F (1995) Bacterial degradation of glycol ethers. Appl Microbiol Biotechnol 44(3-4):532–538
Eur J Clin Microbiol Infect Dis (2016) 35:235–244 22.
Kleemann R, Meckenstock RU (2011) Anaerobic naphthalene degradation by Gram-positive, iron-reducing bacteria. FEMS Microbiol Ecol 78(3):488–496 23. Kunz DA, Chapman PJ (1981) Catabolism of pseudocumene and 3ethyltoluene by Pseudomonas putida (arvilla) mt-2: evidence for new functions of the TOL (pWWO) plasmid. J Bacteriol 146(1): 179–191 24. Leahy JG, Tracy KD, Eley MH (2003) Degradation of mixtures of aromatic and chloroaliphatic hydrocarbons by aromatic hydrocarbon-degrading bacteria. FEMS Microbiol Ecol 43(2): 271–276 25. Musat F, Galushko A, Jacob J, Widdel F, Kube M, Reinhardt R, Wilkes H, Schink B, Rabus R (2009) Anaerobic degradation of naphthalene and 2-methylnaphthalene by strains of marine sulfate-reducing bacteria. Environ Microbiol 11(1):209–219 26. Sikkema J, de Bont JA (1993) Metabolism of tetralin (1,2,3,4tetrahydronaphthalene) in Corynebacterium sp. strain C125. Appl Environ Microbiol 59(2):567–572
ORIGINAL ARTICLE
Detection of signature volatiles for cariogenic microorganisms M. Hertel 1 & R. Preissner 2 & B. Gillissen 3,4 & A. M. Schmidt-Westhausen 1 & S. Paris 5 & S. Preissner 5
Received: 29 October 2015 / Accepted: 18 November 2015 / Published online: 26 November 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract The development of a breath test by the identification of volatile organic compounds (VOCs) emitted by cariogenic bacteria is a promising approach for caries risk assessment and early caries detection. The aim of the present study was to investigate the volatile profiles of three major cariogenic bacteria and to assess whether the obtained signatures were species-specific. Therefore, the headspaces above cultures of Streptococcus mutans, Lactobacillus salivariusand Propionibacterium acidifaciens were analysed after 24 and 48 h of cultivation using gas chromatography and mass spectrometry. A volatile database was queried for the obtained VOC profiles. Sixty-four compounds were detected within the analysed culture headspaces and were absent (36) or at least only present in minor amounts (28) in the control headspace. For S. mutans 18, for L. salivarius three and for P. acidifaciens five compounds were found to be unique signature VOCs. Database matching revealed that the identified signatures of all bacteria were unique. Furthermore, 13 of the 64 detected substances have not been previously reported to be emitted by bacteria or fungi. Specific VOC signatures were
* S. Preissner [email protected] 1
Department of Oral Medicine, Dental Radiology and Oral Surgery, Charité - Universitätsmedizin Berlin, Berlin, Germany
2
Structural Bioinformatics Group, Institute for Physiology, Charité Universitätsmedizin Berlin, Berlin, Germany
3
German Cancer Consortium (DKTK), Heidelberg, Germany
4
Hematology, Oncology and Tumor Immunology, Charité Universitätsmedizin Berlin, Berlin, Germany
5
Department of Operative and Preventive Dentistry, Charité Universitätsmedizin Berlin, Aßmannshauser Str. 4-6, 14197 Berlin, Germany
found in all the investigated bacteria cultures. The obtained results encourage further research to investigate the transferability to in vivo conditions towards the development of a breath test.
Introduction Caries is one of the most frequent diseases worldwide. During the last decade, there has been a paradigm shift in the therapy of caries. For over 100 years, cavities were treated following the ‘extension for prevention’ concept of Black, involving removing sound tooth structure [1]. Minimal-invasive concepts were developed, based on on-going research and the improvement of restorative materials [2]. The idea of preserving tooth structure developed further and the focus is now on preventive dentistry and oral health maintenance. A new system to measure initial caries lesions called the International Caries Detection and Assessment System (ICDAS) was introduced in 2007 [3]. Furthermore, various non-invasive and micro-invasive therapy concepts have been introduced, such as nutrition management, fluoride application, sealing or infiltration [4]. An outstanding problem is caries risk assessment, as the only accurate predictor for future caries that is currently available is past caries experience [5]. The early detection of highrisk patients would facilitate individual prevention. In addition, costs might be saved extensively, because, currently, prevention is applied to collectives or groups, whereas only a minority of patients requires individual prophylaxis. A simple and reliable caries risk test is currently not available. Certain oral bacteria, especially Streptococcus mutans and Lactobacillus salivarius, have been reported to be the major causes of caries [6]. A recent in vivo study revealed that Propionibacterium acidifaciens is significantly more
236
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
Fig.
1 Signature compound assessment. VOC = volatile organic compound; GC/MS = gas chromatography/mass spectrometry. *= compound not evaluable due to superimposing GC/MS peaks
abundant in caries samples compared to other microorganisms [7]. Overall, analyses of the commensal microbiome revealed that it was indicative of a variety of pathological conditions [8], but a comprehensive sequencing analysis is elaborate and expensive. The detection of volatile organic compounds (VOCs) emitted by pathogenic microbiota has been performed [9]. Although gas chromatography and mass spectrometry (GC/ MS), which provide high reliability and accuracy, have been used to profile the VOCs emitted by a wide range of bacteria, inter alia Cyanobacteriae [10] and Rhizobacteriae [11]. Nevertheless, the VOCs emitted by cariogenic bacteria are unknown. Breath analyses are rapid and affordable, and are, therefore, already used in various disciplines; for instance, the identification of Helicobacter pylori via VOC detection is routinely carried out in gastroenterology [12]. Inflammation in the airways can be monitored by nitric oxide testing [13] and a breath test for Mycobacterium tuberculosis has been developed [14]. For caries screening and risk assessment, a breath test based on the detection of volatiles released by cariogenic bacteria would be promising. Additionally, breast tests are simple to perform, painless, non-invasive and repeatable, even in young children. Exposure to radiation might also be reduced, as screening for approximal caries lesions is predominantly performed using radiographic methods. The aim of the present study was to identify the VOCs released by the three major cariogenic bacteria, specifically S. mutans, L. salivarius and P. acidifaciens, using GC/MS. Furthermore, the study sought to investigate whether the obtained profiles were unique, by searching the mVOC database [15]. It was hypothesised that species-specific volatile signatures can be detected in the culture headspaces of the investigated bacteria.
Materials and methods Bacteria cultivation Three reference microorganisms, namely, S. mutans, L. salivarius and P. acidifaciens (DSM 20523, 20555 and 21887, DSMZ, Braunschweig, Germany), were cultivated for 24 h at 30 °C on Brain Heart Infusion-agar (BHI) (AppliChem, Darmstadt, Germany) plates under microaerophilic conditions. Subsequently, two colony-forming units (CFUs) per strain were transferred into 3.5 mL liquid BHI (VWR International, Darmstadt, Germany) and Erlenmeyer flasks containing 50 mL solid BHI were inoculated.
237
Incubation at 30 °C for 24 h was performed in microaerophilic conditions after sealing both open ends of the flasks using polytetrafluoroethylene (PTFE) plugs. Before headspace sampling, a standardised biochemical identification system (API 20 Strep, bioMérieux, Nürtingen, Germany) was used to confirm the genus of the cultured bacteria. Headspace sampling Following incubation, volatile compound sampling was carried out. To assess potential time-dependent differences in the VOC spectra, sampling was performed after 24 and 48 h. The headspaces above the bacteria cultures and from a blank BHI, which served as a control, were transferred to adsorbent (polyphenylene oxide containing) tubes (Tenax, Gerstel, Mühlheim, Germany). Therefore, a flexible PTFE hose was used to connect the culture flask and the Tenax tube. A second hose was installed to connect the Tenax tube to a suction pump (SG 350, GSA Messgerätebau, Neuss, Germany). Additionally, an activated carbon filter (Niosh Typ B, Dräger, Lübeck, Germany) was connected to the open end of the culture flask to avoid negative pressure. Subsequently, a total volume of 1 L was sucked through the Tenax tube in 10 min, applying a constant flow rate of 100 mL/min. Headspace analysis Headspace analysis was performed using a capillary gas chromatograph to which a mass spectrometer was coupled (6890N Network Gas Chromatograph System, Agilent Technologies, Santa Clara, CA, USA), with sample injection by means of thermal desorption. Closed-loop stripping analysis (CLSA) was utilized, using a 5 % phenyl methylpolysiloxane capillary column (HP-5ms, Agilent Technologies, Santa Clara, CA, USA). The applied acquisition mode conditions and parameters were as follows: Oven: initial temperature: 30 °C; initial time: 1.0 min; maximum temp: 350 °C; equilibration time: 0.5 min; post temperature: 30 °C; post time: 1.0 min; run time: 39.33 min; Inlet: mode: solvent vent; initial temperature: 250 °C; vent time: 0.1 min; vent flow: 50.0 mL/min; vent pressure: 41.0 psi; purge flow: 10.0 mL/min; purge time: 0.02 min; total flow: 14.5 mL/min; saver flow: 30.0 mL/min (gas type: helium); saver time: 1.00 min; Column: mode: constant flow; nominal length: 50 m; nominal inside diameter: 0.2 mm; nominal film thickness: 0.33 μm; maximum temperature: 350 °C; initial flow: 1.7 mL/min; initial pressure: 40.1 psi; average velocity: 36 cm/s; outlet pressure: vacuum; Temperature program: initial temperature: 30 °C; delay time: 1.1 min; rate: 60.0 °C/min; end temperature:
238 Table 1 Volatiles detected in Streptococcus mutans, Lactobacillus salivariusand Propionibacterium acidifaciens culture headspaces
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
VOC
CAS-Nr.
S.mutans
n-Hexane
110-54-3
n-Heptane
142-82-5
#
n-Octane
111-65-9
#
n-Nonane n-Pentadecane
111-82-2 629-62-9
# #
n-Heptadecane n-Octadecane
629-78-7 593-45-3
#
3-Methyl pentane
96-14-0
Toluene Ethyl benzene
108-88-3 100-41-4
* *
m-/p-Xylene
108-38-3
*
o-Xylene
106-42-3 95-47-7
*
1,3,5-Trimethyl benzene 1,2,4-Trimethyl benzene
108-67-8 95-63-6
# #
3-/4-Ethyl toluene
620-14-4
#
1,2,3,4-Tetrahydro naphthalene Naphthalene 2-Methyl naphthalene 1-Methyl naphthalene β-Pinene
622-96-8 119-64-2 91-20-3 91-57-6 92-12-0 19902-08-0
# * # # #
Delta-3-carene 2-Methyl-1-propanol 1-Butanol
18172-67-3 498-15-7 78-83-1 71-36-3
# * *
2-Ethyl-1-hexanol Benzyl alcohol Methyl acetate Ethyl acetate n-Propyl acetate
104-76-7 100-51-6 79-20-9 141-78-6 109-60-4
Isobutyl acetate n-Butyl acetate
110-19-0 123-86-4
Ethyl-2-methyl butyrate Methyl ethyl ketone Methyl isobutyl ketone Cyclohexanone Acetophenone 2-Methyl furane Ethylene glycol monobutyl ether Dipropylene glycol monomethyl ether Diethylene glycol monoethyl ether Diethylene glycol monobutyl ether 3-Methyl-1-butanal n-Hexanal n-Heptanal n-Octanal n-Nonanal n-Decanal n-Undecanal
7452-79-1 78-93-3 108-10-1 108-94-1 98-86-2 534-22-5 111-76-2 34590-94-8 111-90-0 112-34-5 590-86-3 66-25-1 111-71-7 124-13-0 124-19-6 112-31-2 112-44-7
L.salivarius
P.acidifaciens
* #
# #
* #
* # * *
* * * # *
*
#
*
# * *
* * #
* # * *
# *
*
# * *
* *
*
*
# # # # * * # * * # #
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
239
Table 1 (continued) VOC
CAS-Nr.
S.mutans
L.salivarius
P.acidifaciens
Benzaldehyde
100-52-7
*
Dimethyl sulfide
75-18-3
*
*
*
3-Methyl furane 3-Methyl-2-butanol
930-27-8 598-75-4
#
*
* #
2-Pentanone 3-Methyl-1-butanol
107-87-9 123-51-3
* #
* #
*
2-Methyl-1-butanol
137-32-6
#
#
Dimethyl disulfide 1-Hexanol
624-92-0 111-27-3
* #
2-Heptanone 1-Heptanol
110-43-0 110-70-6
* #
1-Octene-3-ol
3391-86-4
#
3-Octanone 3-Octanol
106-68-3 589-98-0
# #
#
1-Octanol
111-87-5
*
*
1-Nonanol 1-Decanol
143-08-8 112-30-1
# #
* *
Data retrieved from S. mutans and L. salivarius after 24 and 48 h were pooled #= volatile not found in the BHI control *= amounts detected in culture that exceeded those in the BHI control
260 °C; hold time: 8.00 min; transfer temperature: 300 °C; standby temperature: 50 °C. Selected ion monitoring (SIM)/scan mode was applied. A recorded, full-scan-mode chromatogram was used for qualitative analysis via library-assisted matching, and up to 150 peaks were automatically integrated within the chromatogram. Compounds were identified by the comparison of mass spectra to database spectra. Within our own database, all 227 tested substances were referenced as external standards based on multipoint calibration using at least six levels per substance. Based on algorithmic comparison (probability-based matching), the mass spectra of the obtained peaks were automatically assigned and reviewed by hand by an experienced investigator (S.P.). Finally, the matching reference spectra contained in the searched spectral libraries were assigned. To assess if volatiles that were detectable within both the cultures and the control exceeded the blank BHI, quantitative analysis was performed according to DIN ISO 16000-6. Compound quantification was performed by applying the method of external standards on comparative blends. VOC profiling Only compounds that were detected in the culture headspaces but not in the BHI control were included in the volatile profiling. Accordingly, VOC profiling was exclusively based on the results obtained from qualitative analysis. For this
purpose, the mVOC database [15], including PubMed, was searched for the identified signatures, in order to assess whether the VOCs had previously been identified in microbial cultures and to investigate whether species specificity was obtained. The obtained profiles were compared with the profiles of all bacteria and fungi emitting at least one of the detected volatiles. The date of access was 1 October 2015.
Results Headspace analysis Of the 227 tested substances, 120 were found to be below the detection limit and a further 27 compounds were detected within the culture headspaces, but their amounts did not exceed that of the control. Sixteen volatiles were not evaluable due to the superimposition of GC/MS peaks and were, therefore, excluded from further analysis. Thus, 64 substances (alkanes, aromatics, terpenes, monovalent alcohols, esters of monovalent alcohols, ketones, cyclic ethers, glycols, glycol esters, glycol ethers and aldehydes) were detected in the headspaces of the investigated cultures (Fig. 1 and Table 1). Ethanol was ubiquitously detected, but could not be quantified due to excessive concentration. Alkenes, chloronaphthalenes, chlorinated hydrocarbons, siloxanes, esters of fatty acids, esters of dicarboxylic acids, heterocyclic compounds, phenols and acrylates were not found, at least not in amounts that
240
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
Fig. 2
Chromatograms of Streptococcus mutans (a), Lactobacillus salivarius (b) and Propionibacterium acidifaciens headspaces (c) after 48 h of cultivation
exceeded those in the BHI control. Chromatograms of S. mutans, L. salivarius and P. acidifaciens are shown in the Fig. 2a–c.
241
investigated bacteria differed from the 226 profiles and, therefore, were unique. Hence, species specificity was obtained in all cases. The signature patterns of S. mutans consisted of 18 compounds, whereas that of L. salivarius and P. acidifaciens was characterised by the presence of three and five volatiles, respectively (Fig. 3 and Table 2). Peak assignment within the mass spectra revealed matching scores of 100 % in all 36 compounds exclusively detected in the bacteria cultures.
VOC profiling Fifty-one out of the 64 detected compounds were found by searching the mVOC database using their PubChem IDs for unequivocal identification. Hence, these substances have been reported to be emitted by different bacteria and/or fungi. Thirteen molecules were not previously reported to be released by microbial cultures (Fig. 1). Volatile profile determination was performed, excluding 28 out of the 64 substances, as minor amounts were also found in the control headspace. Thus, 36 VOCs that were exclusively found in the headspaces of S. mutans, L. salivarius and P. acidifaciens were compiled and the mVOC database was again searched for the obtained volatile profiles (Fig. 1). Overall, 226 microorganisms were found that emit at least one of the signature VOCs, and 39 of these bacteria and fungi potentially colonise or infect humans. Database matching revealed that the profiles found for all the
Discussion The aim of the present study was to identify the VOC profiles of S. mutans, L. salivarius and P. acidifaciens. The emission signatures obtained from solid cultures were specific for all three bacteria and allowed their differentiation. Nevertheless, certain compounds (phenyl ethanol, 3-methyl-1-butanol, 2methyl-1-butanol and 1-decanol) were found to be emitted by more than five microbiota in addition to cariogenic bacteria. Hence, excerpts of the retrieved volatile signatures might be sufficient for their identification. Before excerpting the obtained VOC profiles, the extent of transferability of the retrieved results to in vivo conditions should form the focus of subsequent research. Growth conditions, including oxygen
Fig. 3 Venn diagram of signature VOCs of S. mutans, L. salivarius and P. acidifaciens
242
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
Table 2 The volatile organic compound (VOC) signatures of S. mutans, L. salivarius and P. acidifaciens (displayed in blue) compared with bacteria and fungi that infect humans, retrieved from the mVOC database
concentration, nutrient supply, temperature and pH, have been reported to influence bacterial metabolism and, therefore, VOC emission [16]. Accordingly, volatile release was found to be time-dependent in all the investigated cultures. In the present study, one culture per species was investigated as reproducible and standardized growth conditions were applied. Thus, no variability in metabolism and, hence, VOC emission was expected. However, subsequent research might apply varying culture conditions
to assess their influence on the release of bacterial volatiles. Further studies might, therefore, consider performing VOC analysis by applying conditions that mimic the environment of oral cavities. Additionally, artificial caries lesions appear to be suitable for further preclinical research. Clinical approaches that perform breath analyses in healthy and diseased subjects might also be undertaken. In addition to transferability to in vivo conditions, the specificity, sensitivity and the positive and
Eur J Clin Microbiol Infect Dis (2016) 35:235–244
the negative predictive values for the detection of early caries remain unknown. Even though VOC quantification is feasible using GC/MS, it is of inferior relevance compared to qualitative analysis, at least within the intent of the present study. Quantification was only used to allow comparison between amounts detected in the cultures and blank BHI. Furthermore, the VOCs quantity does not necessarily correlate with the cell number [17, 18]. Accordingly, neither the CFU count nor the optical density measurement were performed. Volatile compounds released by bacteria act as quorum-sensing molecules, regulating inter alia metabolism and growth. Furthermore, they mediate interactions between different microbiota. As an example, different sulphur compounds and long-chain ketones inhibit the growth of fungi [19]. Dimethyl disulfide, an active compound involved in the mentioned interactions, was found in our study within the headspaces of S. mutans and P. acidifaciens. The origin and the related metabolic pathways of the 13 compounds that were not found by searching the mVOC database remain uncertain, with the exception of n-octadecane, which has been described to be released by Microcystis aeruginosa [20]. This cyanobacterium is not found in humans. The remaining compounds, n-heptane, 1,3,5-trimethyl benzene, 3ethyl toluene, 1,2,3,4-tetrahydro naphthalene, 2-methyl naphthalene, 1-methyl naphthalene, ethylene glycol monobutyl ether, diethylene glycol monoethyl ether and diethylene glycol monobutyl ether have been described to be sources of carbon and energy for certain bacterial species [21–26]. As a wide range of glycol ethers can be metabolized by bacteria [21], it can be assumed that this includes dipropylene glycol monoethyl ether, even if this has not been reported. The compounds 3-methyl pentane and β-pinene could not be related to bacterial metabolism; nevertheless, even when these 13 substances were excluded from the obtained profiles due to their unknown origin, species specificity was maintained in all the investigated species. As a consequence, the identified VOCs could be seen as biomarkers for cariogenic bacteria, potentially allowing caries risk assessment or screening for initial caries lesions based on their detection using a breath test.
243 Acknowledgements The authors would like to thank Mss. Anja and Antje Richter for their most valuable assistance in the laboratory and Janette Nickel for her help with the mVOC database. Compliance with ethical standards Conflict of interest The authors state that they have no affiliations with or involvement in any organisation or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this article.
References 1. 2.
3.
4.
5.
6. 7.
8.
9.
10.
11.
12.
Conclusions 13.
As hypothesised, volatile profiles that were unique for Streptococcus mutans, Lactobacillus salivarius and Propionibacterium acidifaciens were obtained in the present study. Thus, developing a breath test based on volatile organic compounds (VOCs) identification remains a promising approach for caries risk assessment and early caries detection. These results encourage further studies to investigate the transferability of these findings to in vivo conditions.
14.
15.
16.
Osborne JW, Summitt JB (1998) Extension for prevention: is it relevant today? Am J Dent 11(4):189–196 Peters MC, McLean ME (2001) Minimally invasive operative care. I. Minimal intervention and concepts for minimally invasive cavity preparations. J Adhes Dent 3(1):7–16 Ismail AI, Sohn W, Tellez M, Amaya A, Sen A, Hasson H, Pitts NB (2007) The International Caries Detection and Assessment System (ICDAS): an integrated system for measuring dental caries. Community Dent Oral Epidemiol 35(3):170–178 Meyer-Lueckel H, Bitter K, Paris S (2012) Randomized controlled clinical trial on proximal caries infiltration: three-year follow-up. Caries Res 46(6):544–548 Mejàre I, Axelsson S, Dahlén G, Espelid I, Norlund A, Tranæus S, Twetman S (2014) Caries risk assessment. A systematic review. Acta Odontol Scand 72(2):81–91 Poole DF, Newman HN (1971) Dental plaque and oral health. Nature 234(5328):329–331 Wolff D, Frese C, Maier-Kraus T, Krueger T, Wolff B (2013) Bacterial biofilm composition in caries and caries-free subjects. Caries Res 47(1):69–77 Blaser M, Bork P, Fraser C, Knight R, Wang J (2013) The microbiome explored: recent insights and future challenges. Nat Rev Microbiol 11(3):213–217 Tait E, Perry JD, Stanforth SP, Dean JR (2014) Identification of volatile organic compounds produced by bacteria using HSSPME-GC-MS. J Chromatogr Sci 52(4):363–373 Höckelmann C, Jüttner F (2004) Volatile organic compound (VOC) analysis and sources of limonene, cyclohexanone and straight chain aldehydes in axenic cultures of Calothrix and Plectonema. Water Sci Technol 49(9):47–54 Farag MA, Ryu CM, Sumner LW, Paré PW (2006) GC-MS SPME profiling of rhizobacterial volatiles reveals prospective inducers of growth promotion and induced systemic resistance in plants. Phytochemistry 67(20):2262–2268 Peeters M (1998) Urea breath test: a diagnostic tool in the management of Helicobacter pylori-related gastrointestinal diseases. Acta Gastroenterol Belg 61(3):332–335 Smith AD, Cowan JO, Brassett KP, Herbison GP, Taylor DR (2005) Use of exhaled nitric oxide measurements to guide treatment in chronic asthma. N Engl J Med 352(21):2163–2173 Scott-Thomas A, Epton M, Chambers S (2013) Validating a breath collection and analysis system for the new tuberculosis breath test. J Breath Res 7(3):037108 Lemfack MC, Nickel J, Dunkel M, Preissner R, Piechulla B (2014) mVOC: a database of microbial volatiles. Nucleic Acids Res 42(Database issue):D744–D748 Blom D, Fabbri C, Connor EC, Schiestl FP, Klauser DR, Boller T, Eberl L, Weisskopf L (2011) Production of plant growth
244 modulating volatiles is widespread among rhizosphere bacteria and strongly depends on culture conditions. Environ Microbiol 13(11): 3047–3058 17. Zehm S, Schweinitz S, Würzner R, Colvin HP, Rieder J (2012) Detection of Candida albicans by mass spectrometric fingerprinting. Curr Microbiol 64(3):271–275 18. Bunge M, Araghipour N, Mikoviny T, Dunkl J, Schnitzhofer R, Hansel A, Schinner F, Wisthaler A, Margesin R, Märk TD (2008) On-line monitoring of microbial volatile metabolites by proton transfer reaction-mass spectrometry. Appl Environ Microbiol 74(7):2179–2186 19. Hunziker L, Bönisch D, Groenhagen U, Bailly A, Schulz S, Weisskopf L (2015) Pseudomonas strains naturally associated with potato plants produce volatiles with high potential for inhibition of Phytophthora infestans. Appl Environ Microbiol 81(3):821–830 20. Walsh K, Jones GJ, Dunstan RH (1998) Effect of high irradiance and iron on volatile odour compounds in the cyanobacterium Microcystis aeruginosa. Phytochemistry 49(5):1227–1239 21. Kawai F (1995) Bacterial degradation of glycol ethers. Appl Microbiol Biotechnol 44(3-4):532–538
Eur J Clin Microbiol Infect Dis (2016) 35:235–244 22.
Kleemann R, Meckenstock RU (2011) Anaerobic naphthalene degradation by Gram-positive, iron-reducing bacteria. FEMS Microbiol Ecol 78(3):488–496 23. Kunz DA, Chapman PJ (1981) Catabolism of pseudocumene and 3ethyltoluene by Pseudomonas putida (arvilla) mt-2: evidence for new functions of the TOL (pWWO) plasmid. J Bacteriol 146(1): 179–191 24. Leahy JG, Tracy KD, Eley MH (2003) Degradation of mixtures of aromatic and chloroaliphatic hydrocarbons by aromatic hydrocarbon-degrading bacteria. FEMS Microbiol Ecol 43(2): 271–276 25. Musat F, Galushko A, Jacob J, Widdel F, Kube M, Reinhardt R, Wilkes H, Schink B, Rabus R (2009) Anaerobic degradation of naphthalene and 2-methylnaphthalene by strains of marine sulfate-reducing bacteria. Environ Microbiol 11(1):209–219 26. Sikkema J, de Bont JA (1993) Metabolism of tetralin (1,2,3,4tetrahydronaphthalene) in Corynebacterium sp. strain C125. Appl Environ Microbiol 59(2):567–572
