- Email: [email protected]
Acetobacter intermedius, sp. nov.
System. Appl. Microbial. 21, 220-229 (1998) _©_G_us_ta_v_Fi_sc_he_r¥_e_rl_ag_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
SYSTEI'VL4TIC AND APPLIED MICROBIOLOGY
Acetobacter intermedius, sp. nov. CORNELIA BOESCH!, JANJA TRCEK 2 , MARTIN SIEVERS 3 and MICHAEL TEUBER 1 lInstitut fur Lebensmittelwissenschaft, Labor fur Lebensmittelmikrobiologie, ETH-Zurich, Zuerich, Switzerland 2Biotechnical Faculty, Department of Food Science and Technology, University of Ljubljana, Ljubljana, Slovenia 3Ingenieurschule Wadenswil, Wadenswil, Switzerland Received August 26, 1997
Summary Strains of a new species in the genus Acetobacter, for which we propose the name A. intermedius sp. nov., were isolated and characterized in pure culture from different sources (Kombucha beverage, cider vinegar, spirit vinegar) and different countries (Switzerland, Slovenia). The isolated strains grow in media with 3% acetic acid and 3% ethanol as does A. europaeus, do, however, not require acetic acid for growth. These characteristics phenotypically position A. intermedius between A. europaeus and A. xylinus. DNA-DNA hybridizations of A. intermedius-DNA with DNA of the type strains of Acetobacter europaeus, A. xylinus, A. aceti, A. hansenii, A. liquefaciens, A. methanolicus, A. pasteurianus, A. diazotrophicus, Gluconobacter oxydans and Escherichia coli HB101 indicated less than 60% DNA similarity. The important features of the new species are described. Acetobacter intermedius strain TF2 (DSMl1804) isolated from the liquid phase of a tea fungus beverage (Kombucha) is the type strain. Key words: A. intermedius sp. nov.
Introduction Differentiation of Acetobacter-strains from industrial vinegar fermentations is difficult due to their unwillingness to grow well on semisolid media. Data of DNA similarities (30 to 100%) among industrial strains revealed their division into more than one group suggesting the occurrence of several new species. Based on these data, Acetobacter europaeus was described as a main species of industrial vinegar fermenters in central Europe (SIEVERS et aI., 1992). The main biochemical differentiation characteristics of the isolates from high percentage vinegar productions against the traditional species are both a strong tolerance to acetic acid, 4 to 8 % in AE-agar and 10-14 % in industrial fermentations, and requirement of acetic acid for growth. Due to the lack of suitable differentiating physiological markers, total cellular proteins of 44 Acetobacter and Gluconobacter strains were characterized by SDS-PAGE in our laboratory (KERSTERS, 1985) to provide a reproducible phenotypic property (KNEUBOHLER, 1994). Computerized numerical analysis of the protein profiles showed that the 25 different strains from industrial vinegar fermentations clustered in 4 phenons separated from all other investigated strains with a close linkage of A. xylinus (EUZEBY, 1997) to the A. europaeus group. The protein patterns correlated with the results of DNA-
DNA hybridizations with the A. europaeus type strain (SIEVERS and TEUBER, 1995). In the present reports the low (30-<70%) DNA-DNA hybridization similarity group (SIEVERS et aI., 1992) and the 4 phenons described by protein profiling (SIEVERS and TEUBER, 1995) are analyzed on a phenotypic and genotypic level and compared to an isolate (TF2) from a tea fungus beverage (SIEVERS et aI., 1995). Based on 16S rDNA sequence alignments, missing hybridization signals with an oligonucleotide probe based on the 23S rRNA of A. europaeus, DNA-DNA hybridization data and phenotypic properties, strain TF2 is described as the new species Acetobacter intermedius in the genus Acetobacter.
Materials and Methods Bacterial strains and cultivation: Bacterial strains used in this study are shown in Table 1. Escherichia coli strain HB101 was cultivated on Luria-Bertani medium. Acetobacter europaeus and Acetobacter intermedius strains were cultivated on a modified AE-medium (ENTANI et aI., 1985): 0.5% glucose, 0.3% yeast extract, 0.4% peptone, 0.9% agarose, 3%(v/v) ethanol and 3%(v/v) acetic acid. AEG20% is AE-medium with 20% glucose. Incubation was at 30°C and 92-96% relative air humidity. Ace-
Acetobacter intermedius sp. nov.
221
Table 1. Bacterial strains. Species
Strain designation
source
original substrate/ fermenter type)
Acetobacter aceti Acetobacter pasteurianus Acetobacter liquefaciens Acetobacter hansenii Acetobacter xylinus Acetobacter xylinus Acetobacter xylinus Acetobacter xylinus Acetobacter xylinus Acetobacter methanolicus Acetobacter diazotrophicus Gluconobacter oxydans Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter sp. Acetobacter sp. Acetobacter sp. Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter sp.
NCIB 8621 T LMD22.F IFO 12388 T NCIB 8746 T NCIB 11664T NCIB613 De Ley 25 ATCC 23768 ATCC 14851 MB58 T ATCC 49037T ATCC 19357T DES lIT DOSll TSA4 WW Sl TSA9 KO SW3 SSPR1 JK2 VI TSN5 TSN1 TSA8 TF2T JK3 JKD TSA10 0SSPR TSN3 TSA3 TSA7 KRI TSN4 E1 E2 0SW HE
DSM 3508 LMG 1262 LMG 1328 LMG 1527 LMG 1515 LMG 1510 LMG25 DSM2004 DSM46604 LMG 1668 LMG 7603 DSM 3503 DSM 6160 DSM 6161 LMETH LMETH LMETH LMETH LMETH LMETH LMETH ZIM B021 ZIMB026 LMETH LMETH LMETH DSM 11804 ZIM B022 ZIM B025 LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH
submersed submersed generator submersed submersed generator submersed submersed submersed cider vinegar/submersed wine vinegar/submersed generator generator generator Kombucha cider vinegar/submersed cider vinegar/submersed generator submersed generator generator generator submersed generator submersed submersed submersed submersed
T type strains ZIM LM ETH DSMZ LMG ')submersed generator
-
Zbirka Industrijskih Mikroorganizmov, Ljubljana, Slovenia Labor fur Lebensmittelmikrobiologie, ETH Zurich, Switzerland Deutsche Sammlung von Mikroorganismen und Zellkulturen Laboratorium voor Microbiologie, Rijksuniversiteit, Gent submersed fermentation trickle fermenter
tobacter and Gluconobacter type strains were cultivated in MRS medium or in GY medium containing 5% glucose and 1 % yeast extract; a concentrated glucose-solution was autoclaved separately to avoid Maillard reactions. Incubation was at 30°C. GYG20% is GY-medium containing 20% glucose. To analyze growth behaviour with different sugars, alcohols and acids, YPmedium containing 0.3% yeast extract and 0.4% peptone was supplemented with 20% of the according carbon source. Formation of 2- and 5-ketogluconic acid: The formation of 2- and 5-ketogluconic acid was assayed with strain TF2 cultivated in shaken culture in 2% glucose, 0.5% yeast extract, 0.3% peptone and 2.0% acetic acid for 14 to 21 days. 2- and
5-ketogluconic acid were determined by thin layer chromatography of growth medium (GOESSELE et al., 1980). In vitro amplification of the 16S rRNA gene: A loopful of cells from agar were suspended in a 10 mM Tris, 10 mM EDTA-buffer (pH 8.0). Cell-lysis was induced by addition of 20 pg Proteinase K and incubation at 55°C for 30 minutes. Nucleic acids were purified by extraction with 100 pi phenol (saturated with 3% NaCl) and subsequently by extraction with 100 pi methylene chloride/isoamyl alcohol (24:1 [vol.!vol.]). 1-3 pi of the aqueous phase were directly used for PCR. 16S rDNA was amplified using oligonucleotide primers complementary to highly conserved regions of bacterial rRNA genes
222
C. BOESCH et al.
(AMMAN et ai, 1990). The 5'- and 3'-terminal primers were GAGTTTGAT(Cff)(C/A)TGGCTCA (positions 9 to 26, according to the Escherichia coli numbering system (BROSIUS et. al., 1981)) and CA(GIT)AAAGGAGGTGATCC (positions 1545 to 1529), respectively. PCR was performed in a total volume of 100 JlI containing 1 U DynaZyme™ DNA polymerase F-501L (FinnZymes), 10 JlI lOx DynaZyme™-reaction buffer, 1 JlM of each primer and 300 JlM of each deoxynucleoside triphosphate. The mixture was amplified in a thermal cycler (Perkin Elmer Cetus 480) as follows: an initial denaturation at 94°C for 5 min. followed by: 92 °C, 1 min.; 47°C, 2 min.; 72 °C, 2 min. and 35 cycles. In vitro amplification of the 23S rRNA gene: Preparation for PCR was as described above for 16S rRNA. The 5'- and 3'-terminal primers were TGCGGCTGGATCACCTCC (positions 3039 to 3056 (BROSIUS et al., 1981)) and CCCGCTTAGATGCTTTCAGC (positions 6262 to 6243 (BROSIUS et al., 1981)), respectively. PCR was performed in a total volume of 100 JlI containing 4 U DynaZyme™ EXT DNA polymerase F-505S (FinnZymes), 10 JlI lOx DynaZyme™ EXT DNA-reaction buffer, 2.0 mM MgCI 2, 1 ]lM of each primer and 360 JlM of each deoxynucleoside triphosphate. The mixture was subjected to 30 cycles of amplification in a thermal cycler (Biometra Personal CycierTM). An initial denaturation at 93°C for 2 min. was followed by: 93 °C, 20 sec.; 60°C, 30 sec.; 68 °C, 3 min. Purification of PCR-product: PCR-products were directly purified with the NucleoSpin Extract kit (Macherey-Nagel, Oensingen, Switzerland). Elution from the DNA binding column was with bidistilled water. Direct Sequencing of PCR product: Automated cycle sequencing of 300 ng purified PCR products (16S rDNA and 23S rDNA) was performed using the Thermo Sequenase fluorescent
labeled primer cycle sequencing kit with 7-deaza-dGTP according to the manufacturer's instructions (Amersham International pic, Buckinghamshire, England). Synthetic oligonucleotide primers (Microsynth, Balgach, Switzerland) were Cy5™-Iabelled; the primers and their positions according to the Escherichia coli numbering system (BROSIUS et al., 1981) are listed in Table 2 and Table 3. Unless otherwise indicated sequences of primers were selected from published data. Cycle sequencing was performed on a Biometra Personal CycierTM (Biometra, Gottingen, Germany) with the following parameters: 95°C, 30 sec.; 50 °C, 30 sec.; 72 °C, 60 sec. and 25 cycles (DASEN et al., 1998). Detection of the denatured (85°C, 3 min.) cycle-sequenced samples was performed on an ALFexpress™ DNA sequencer (Pharmacia, Uppsala, Sweden) using a hydrolink long ranger gel (Long RangerTM, FMC BioProducts, Rockland, USA). Data Analysis: Computer analyses of the DNA sequences were performed using the GCG software package of the University of Wisconsin Genetics Computer Group (Madison, Wis., USA). The 16S rDNA and 23S rDNA nucleotide sequences of A. intermedius have been deposited in the EMBL Nucleotide Sequence Library (Cambridge, UK) under the accession numbers Y14694 and Y14680, respectively. DNA-DNA hybridization with chromosomal DNA: Total DNA of Acetobacter and Gluconobacter species was prepared according to a modified protocol of the method by Marmur (1961). DNA probes (0.1-0.2 Jlg) for hybridizations were internally labeled with 20 JlCi [a- 32 P]dATP (3000 Ci/mmol) using Klenow Polymerase and the hexanucleotide mixture Nr. 1277081 (Boehringer, Mannheim, Germany) according to the instructions of the manufacturer. Pharmacia NAP-10 columns (Pharmacia, Uppsala, Sweden) were used to remove unincorporated 32P-Iabeled nucleotides.
Table 2. Primers for 16S rDNA sequencing. Primer
Sequence
Target site 1
Specificity
References
Eubac338 MS350 uni515 uni775 uni1088 uni1392 uni1392r
5'ACTCCTACGGGAGGCAGO' 5'CTGCTGCCTCCCGTN' 5'ACCGCGGCTGCTGGCAC3' 5'GGMTTAGATACCCTGGTAGTCO' 5'GGTTAAGTCCCGCAACGAGC3' 5'GTACACACCGCCCGTCN' 5TGACGGGCGGTGTGTAC3'
338-355 350-336 515-498 775-796 1088-1107 1392-1408 1392-1376
Bacteria Bacteria Bacteria Bacteria Bacteria Bacteria Bacteria
AMANN et al., 1990 AMANN et al., 1990 this study this study AMANN et al., 1995 this study this study
1
positions according to the Escherichia coli numbering system (BROSIUS et al., 1981)
Table 3. Primers for 23S rDNA sequencing. Primer
Sequence
Target site 1
Specificity
References
b-1 bO blrev b1 b2rev b2 b3 b4 b5 b6 b7
5'ATACGGGGCTATCACCCG3' 5'CCTTTCCCTCACGGTACF 5TTTCGGGGAGAACCAGCTAF s'ATAGCTGGTTCTCCCCGAAA3' 5'GATGGCTGCTTCTAAGCCAAO' 5'GTTGGCTTAGAAGCAGCCATO' 5'CCCCTAAGGCGAGGCCGAAAG3' 5'AGAGAATACCAAGGCGCTTGl' 5'AAGTTCCGACCTGCACGAAT3' 5'GTAACGGAGGCGCGCGAF 5'AGAACGTCGTGAGACAGTTC3'
3841-3824 3974-3957 4321-4302 4321-4302 4579-4559 4559-4579 4848-4868 5130-5194 5452-5471 5757-5774 6027-6080
a-Proteobacteria Proteobacteria Bacteria Bacteria Bacteria Bacteria Bacteria Proteobacteria Bacteria a-Proteo bacteria a-Proteobacteria
this study this study this study this study this study this study this study this study this study this study this study
1
positions according to the Escherichia coli numbering system (BROSIUS et al., 1981)
Acetobacter intermedius sp. nov.
100-500 ng of Acetobacter intermedius DNA were restricted with EcoRV prior to labeling. Equal amounts (1-2 rg) of EcoRV-restricted DNA from various Acetobacter and Gluconobacter strains were sepatated by gelelectrophoresis and vacuum-blotted (Vacu-Blot, Biometra, Switzerland) to a ZetaProbe® GT blotting membrane (Biorad, Switzerland) according to the method of SOUTHERN (1975). The membrane was dried at 80°C and 30 min. to fix the DNA to the membrane. Hybridization with labeled A. intermedius probe was performed at 65°C (SIEVERS et aI., 1992) in 0.25 M Na 2 HP0 4 (pH 7.2) and 7% SDS according to the standard protocol for Zeta-Probe® GT blotting membranes. Washing of the membrane was twice 15 min. in 20 mM Na 2HP0 4 (pH 7.2) and 5% SDS and then twice in 20mM Na 2 HP0 4 (pH 7.2) and 1 % SDS, all washing steps were performed at 65°C. The wet membranes were sealed into plastic bags and the labeled bands were visualized on X-ray films (Fuji RX) with intensifying screens for 1 day to a week. Analysis of the X-ray films was performed with the program WIN CAM, Cybertech, Berlin. The hybridization signal obtained by restricted and labeled DNA-probe with the identical DNA on the membrane was defined as the 100% intensity signal, the intensities of the signals obtained with EcoRV restricted DNA of other Acetobacter and Gluconobacter strains were calculated by the program WIN CAM, calibration of the values was through linear regression. Oligonucleotide labeling: 10pmoll rl of the respective synthetic oligonucleotide primer (Microsynth, Balgach, Switzerland) was end-labeled with 25 rCi [y-3 2P]dATP (6000 Ci/mmol) using T4 Polynucleotide Kinase according to the manufacturer's instructions (Biolabs, Beverly, MA, USA). Pharmacia NICK columns (Pharmacia, Uppsala, Sweden) were used to remove unincorporated 12P-labeled nucleotides. The probes used in colony hybridization experiments are listed in Table 4 .. Preparation of dot blot filters: A loopful of freshly grown cells on solid media was suspended in 1ml of a 0.45% NaCland 0.5% Lysozyme-solution and incubated for 15 min. at 37°C. After centrifugation 0.9ml of the supernatant was removed and the cells resuspended in the remaining 0.1 ml. 10 rl of the suspension were pipetted on to a Zeta-Probe® GT Blotting Membrane (Biorad). Denaturation of the DNA was performed on 1M NaCl saturated 3MM paper (Whatman) for 4 min. Neutralization and washing steps were performed in the same manner with 1 M Tris pH 8 and 2xSSC (lxSSC is 0.15 M NaCI plus 0.015 M sodium citrate), respectively. Hybridization conditions: Hybridizations were carried out in a hybridization oven (Hybaid). Hybridization conditions were the relevant temperature for the respective oligonucleotide probe (Table 4) and 0.25 M Na2HP04 (pH 7.2) and 7% SDS according to the standard protocol for Zeta-Probe® GT blotting membranes. Washing of the membrane was twice 15 min. in 20 mM Na 2HP0 4 (pH 7.2) and 5% SDS and then twice in 20 mM Na2HP04 (pH 7.2) and 1 % SDS, the washing steps were performed at the same temperature as the hybridization. The wet membranes were sealed into plastic bags and the la-
beled bands were visualized on X-ray films (Fuji RX) with intensifying screens for 1 day to a week. Probe stripping and rehybridization: Reprobing was desired to confirm the presence of nucleic acids on the membranes. The Zeta-Probe® GT blotting membrane was prevented from drying between hybridizations. Stripping of the membrane was at 95°C for 1.5 h in O.lx SSC and 0.5% SDS. The presence of nucleic acids on the membranes was confirmed by hybridization with [a- 32 P]dATP-labeled 23S rDNA of A. europaeus at 65°C.
Results Phylogenetic positioning of Acetobacter intermedius The 165 rRNA sequence of A. intermedius (EMBL accession number is Y14694), was determined by direct seGoxydans
Gfmteurii G cerilUlS G asaii
A aceti A pasteurianus A methanolicus
A diozntrophicus A liquefaciens
R globifonnis
A hansenii
Aintennedius AxylilUlS Aell1VfXI£US
Fig. 1. Phylogenetic tree of all Acetobacter and Gluconobacter type strains. The tree was constructed with the GCG software package using the programs pileup, distances (Jin-Nei Gamma) and growtree (UPGMA). The bar indicates 1 % estimated sequence difference. Based on information from SIEVERS and TEUBER, 1995.
Table 4. Sequence of probes used in colony hybridization experiments. Probe
Sequence
Target organism
Target site I
Hybridization Temperature
Source or reference
oligotf2 23seu oligoxyl
5TCAGAAACAACCACTGGC3' 5'AATGCGCCAAAAGCCGGAT3' S'GACTGTTCCTGATATTACGC1'
A. intermedius type strain A. europaeus strains A. xylinus type strain
2272-2249 3455-3437 2348-2324
42°C 45°C 54°C
this study this study this study
1
223
positions according to the Escherichia coli numbering system (BROSIUS et aI., 1981)
C. BOESCH et al.
224
quencing of the PCR-amplified rRNA gene products. The derived 16S rDNA primary structure was compared with homologous sequences taken from published databases including the sequences of all Gluconobacter and Acetobacter species previously described (SIEVERS et al., 1995). A phylogenetic tree reflecting the close relationships of the type strains is shown in Figure 1. The tree is based on the information contained in the 16S rRNA primary structures of the Acetobacter and Gluconobacter type strains and was constructed using the
1
2
4
6
7
8
programs "pileup", "distances" and "growtree" from the GCG software package. The very close clustering of A. europaeus, A. intermedius and A. xylinus only represents their present evolutionary status, their respective positioning in their branch of the phylogenetic tree is no proof for their species status (LUDWIG W., 1998, personal communication). As shown in Table 5, the overall 16S rRNA nucleotide sequence similarities between A. europaeus, A. xylinus and A. intermedius are higher than 97% which made DNA pairing studies necessary as an
9 10 11 12 1
14 1
16
17
Fig. 2. DNA-DNA hybridization of EcoRV digested DNA with [a- 32P]dATP-labeled A. intermediusT-DNA as a probe after gel electrophoresis: lane 1, Escherichia coli HB101; lane 2, Gluconobacter oxydansT; lane 3, A. acett'T; lane 4, A. pasteurianusT; lane 5, A. methanolicusT; lane 6, A. liquefaciens T; lane ?, A. diazotrophicus\ lane 8, A. hansenit'T; lane 9, A. xylinusT; lane 10, A. europaeusT; lane 11, A. intermediusT; lane 12, A. europaeus (JK2); lane 13, A. europaeus (VI); lane 14, A. europaeus (KO); lane 15, A. intermedius (TSA10); lane 16, A. europaeus (TSA8); lanel?, A. intermedius (0SSPR). Top: Ethidium bromide stained chromosomal DNA pattern Bottom: Autoradiograph of chromosomal DNA pattern after hybridization with labeled A. inter-
mediusT-DNA
Acetobacter intermedius sp. nov.
225
Table 5. Overall 16S rRNA sequence similarity values for Acetobacter species, Cluconobacter oxydans and Rhodopila globiformis. Abbreviations: Ain, Acetobacter intermedius; Aeu, Acetobacter europaeus; Aac, Acetobacter aceti; Axy, Acetobacter xylinus; Aha, Acetobacter hansenii; Adi, Acetobacter diazotrophicus; Ali, Acetobacter liquefaciens; Apa, Acetobacter pasteurianus; Cox, Cluconobacter oxydans; RgI, Rhodopila globiformis. %similarity Organism A. intermedius A. europaeus A. xylinus A. hansenii A. liquefaciens A. diazotrophicus A. pasteurianus A. aceti C.oxydans R. globiformis
Ain
Aeu 99.5
Axy 99.2 99.5
Aha 98.2 98.4 98.3
additional genotypic parameter to underline the species status of A. intermedius (STACKEBRANDT and GOEBEL, 1994). DNA-DNA hybridization studies Figure 2 shows the DNA-DNA hybridization of EcoRV-restricted and [a- 32 PldATP-Iabeled A. intermediusT-DNA as a probe with the type strains of A. europaeus, A. intermedius, A. xylinus, A. aceti, A. hansenii, A. liquefaciens, A. methanolicus, A. pasteurianus, Gluconobacter oxydans and Escherichia coli HBI0l and Acetobacter strains from our laboratory. The highest homology of the A. intermedius-type strain with all the type strains of Acetobacer and Gluconobacter was 60% with the type strain of A. europaeus. DNA-DNA hybridization of labeled A. intermediusT-DNA as a probe with other strains yielded the following results: 65% with strain ]K2, 40% with strain VI, 65% with strain KO, 65% with TSAI0, 99% with TSA8 and 90% with 0SSPR. Intense hybridization bands appear in the following strains in Figure 2: lane 3 (G. oxydansT), lane 9 (A. xylinusT), lane 11 (A. intermediusT), lane 12 (strain ]K2) and lane 16 (strain TSA8). Such strong signals could be attributed to the occurrence of similar IS-elements in these strains. The intensities of the radioactive signals on the developed film were determined by quantitative densitometry. Table 6. Overall 23S rRNA sequence similarity values for Acetobacter intermedius, Acetobacter europaeus and Acetobacter xylinus. Abbreviations: Ain, Acetobacter intermedius; Aeu, Acetobacter europaeus; Axy, Acetobacter xylinus. %similarity Organism A. intermedius A. europaeus A. xylinus
Ain
Aeu
Axy
99.2
99 98.9
Ali 97.0 97.4 97.2 97.1
Adi 96.2 97.0 96.7 97.0 98.4
Apa 95.1 95.1 94.8 94.5 95.7 95.3
Aac 95.6 95.4 95.2 94.2 95.8 95.5 96.8
Cox 95.0 94.8 94.6 94.4 94.2 94.2 95.0 95.6
Rgi 92.4 93.9 93.7 93.7 94.0 94.0 92.7 92.8 92.6
The DNA-DNA hybridization values in Table 9 for the strains other than the A. europaeus, A. intermedius and the A. xylinus type strains were obtained by repeated hybridization experiments, the average of the obtained intensities of the hybridization signals was chosen as the result. The strains ]K2 and VI isolated from slovene cider- and wine-vinegar are currently under investigation in a separate thesis (TRCEK, in preparation). Therefore, they are not further discussed in the present work. Generation of 23S rDNA oligonucleotide probes The alignment of the 23S rDNA sequences of the type strains of A. europaeus, A. xylinus and A. intermedius (EMBL accession numbers are X8977I, X898I2 and
Table 7. Acetobacter species that give a positive signal in hybridization experiments with the oligonucleotide probe 23seu derived from the A. europaeus type strain. Empty fields in the column titled "phenon" indicate that these strains were not included in the protein profile experiments performed by Kneubiihler (1994). strain
phenon
identity
source
A. europaeusT A. europaeus S3 TSA4 WW Sl TSA9 KO SW3 TSN5 TSN1 TSA8 JK2 V1 De Ley 25 ATCC 23768 ATCC 10245
1 1 1 1 1 2 2 2 2 4 4 4
A. europaeus A. europaeus A. europaeus A. europaeus A. europaeus A. europaeus A. europaeus A. europaeus A. europaeus Acetobacter sp. Acetobacter sp. Acetobacter sp. A. europaeus A. europaeus "A. xylinus" "A. xylinus" "A. xylinus"
DSM 6061 DSM 6161 LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH ZIM B021 ZIM B026 LMC25 DSM46604 DSM2004
226
C. BOESCH et al.
Y14680, respectively) revealed useful target sites for oligonucleotide probes (Table 4). An oligonucleotide probe (23seu) developed for A. europaeus strains was tested with Acetobacter strains from our culture collection isolated from generators and aceta tors in Switzerland, Germany, Slovenia and Spain. Acetobacter strains that gave a positive signal in hybridization experiments with the oligonucleotide probe 23seu derived from the A. europaeus type strain are listed in Table 7. The probe gave no hybridization signal with the A. xylinus type strain NCIB 11664 and the A. xylinus strain NCIB 613, however, hybridized with the following strains which are
classified as Acetobacter xylinus: De Ley 25, ATCC 23768 and ATCC 10245. No hybridization signals were obtained with the type strains of A. intermedius, A. xyli-
nus, A. aceti, A. hansen ii, A. liquefaciens, A. methanolicus, A. pasteurianus, Gluconobacter oxydans and Escherichia coli HB101 and also no hybridization signals with strains JK3 and JKD from cider vinegar, strain TF2 (A. intermedius, sp. nov) from a tea fungus beverage and all Acetobacter strains in protein profile phenon 3 (SIEVERS and TEUBER, 1995) (Table 9). The probe based on the 23S rRNA of the A. xylinus typestrain gave no signal with the A. europaeus- and
Table 8. Characteristics differentiating the type strains of A. intermedius, A. xylinus and A. europaeus. Growth on AEG20%
Growth on GYG20% A. xylinusT A. europaeusT A. intermedius T TF2
Growth onAE
+
Growth onGY
Growth on MRS
+ + +
+
+
+
Cellulose formation
GC content of DNA (mol%)
+
55-63 56-58 61.55
+
Table 9. Protein profile homology of the 4 phenons of strains isolated from generators and aceta tors in Europe and their DNADNA homology with A. intermedius and A. europaeus (SIEVERS and TEUBER, 1995). The given values (in %) indicate the values for all the strains in the respective phenon.
Phenon 1
Phenon 2
Phenon 3
Phenon 4
Strains
Protein profile homology
approximate DNA-DNA homology to A. europaeus
approximate DNA-DNA homology to A. intermedius
Hybridization with Probe for A. europaeus
DSM 6161 S3 DSM6160 TSA4 WW S1 TSA9 KO SW3 SSPR1 TSN3 0SSPR TSA7 TSA3 TSA6 KRI TSA10 TSN4 0SW HE' E1 E2 TSN5
77.1
90-100%
<60%
yes
87.7 ± 4.6%
ca. 75%
60%
yes
70
<70%
70-100%
no
<30%
ca. 80%
yes, except A. xylinus NCIB
TSN1 TSA8 NCIB 11664
±
64.5
±
4.4%
5.9%
±
1.6%
11664
1 strain HE has identical16S rDNA sequence as the A. europaeus typestrain, DNA association value to the A. intermedius typestrain is <30%.
Acetobacter intermedius sp. nov.
A. intermedius-type strains, no further testing was performed with other A. xylinus strains. The A. intermedius probe is specific only for the type strain - the difference in the 16S rDNA sequence, however, should allow for the construction of an A. intermedius-specific probe.
165 rONA sequences of strains in phenon 3 The 16S rDNA of the Acetobacter strains from protein profile phenon 3 (Table 9) and strains JK3 and JKD were sequenced. Except for strain HE (identical sequence to A. europaeus) the 16S rDNA sequences of these strains were identical to that of A. intermedius. Phenotypic characteristics Phenotypic characteristics differentiating A. intermedius- from A. europaeus- and A. xylinus-type strains are shown in Table 8. The A. intermedius type strain grows with or without acetic acid, survives glucose concentrations of 20% both on media with and without 3% acetic acid and 3 % ethanol, shows excellent growth on MRS medium and forms a cellulose pellicle on solid media and loose cellulose threads in liquid media which are sometimes suppressed. The A. europaeus type strain does not grow on GYG20%, on AEG20% or on MRS medium, does not form cellulose/acetan either in liquid or on solid media and requires acetic acid for growth. The A. xylinus type strain only grows on GYG20% and GY, cellulose formation is always present on solid medium and sometimes suppressed in liquid medium. Growth of A. intermedius and A. europaeus strains on AE, GY, AEG20%, GYG20%, selective MRS-medium for lactobacilli and on YP-media containing different carbon sources was tested. The growth of the cells on YP-media supplemented with different carbon sources gave no clear picture. Meaningful results on the use of carbon sources by the investigated strains would only be obtained in growth experiments on minimal media. As A. europaeus-strains generally need both acetic acid and an additional carbon source in their media, correct interpretation of results thus obtained would be tricky. The A. intermedius strains all grow on AE-medium, 85% of the investigated A. intermedius strains grow on GY-medium and all the A. intermedius strains tested on MRS-agar showed excellent growth on this medium. These results clearly position A. intermedius as a new species with phenotypic properties of both A. europaeus and A. xylinus: A. xylinus strains do not tolerate concentrations of 3 % acetic acid in the medium and the typical A. europaeus strain requires acetic acid for growth. Classification of the strains grouped according to protein profile homology and DNA reassociation values in Table 9 as belonging either to A. intermedius or A. europaeus according to phenotypic properties is not always clear as some strains like strain SW3 (clearly established as A. europaeus by its DNA-homology value and its rrna sequence) do exist that grow well on media without acetic acid. The A. intermedius type strain does not oxidize glucose to 5-ketogluconate or 2-ketogluconate. The
227
production of either 5-ketogluconate or 2-ketogluconate or both is a variable trait (KNEUBOHLER, 1994) and can therefore not be used for differentiation.
Discussion Isolation and cultivation of Acetobacter strains from generators and aceta tors is a difficult and slow procedure. A possibility for rapid identification of such strains would be by hybridization with a species-specific oligonucleotide probe. Development of an oligonucleotide probe for A. europaeus-strains on the basis of 16S rDNA sequences was not possible due to the high similarity with A. xylinus (99.5%, see Table 5) with only 7 varying nucleotides distributed over the length of the gene. Compared to the 16S rRNA primary structure, the 23S rRNA molecule exhibits an increased degree of sequence variability (STACKEBRANDT, 1988) and has served as an appropriate target for probes to differentiate between genetically similar organisms (RONNER and STACKEBRANDT, 1994). The alignment of the 23S rDNA sequences of the type strains of A. europaeus, A. xylinus and A. intermedius revealed equally high similarities (Table 6), however, with useful target sites for oligonucleotide probes. The oligonucleotide probe 23seu hybridized with all Acetobacter strains from the protein profile phenons 1, 2 and 4. It gave no hybridization signal with A. intermedius-strains, the A. xylinus type strain NCIB 11664 and the A. xylinus strain NCIB 613, however, hybridized with three strains which are classified as Acetobacter xylinus: De Ley 25, ATCC 23768 and ATCC 10245. These strains have previously been studied for their DNA-DNA homologies with A. europaeus and A. xylinus (BOESCH, 1994), all of them have DNA similiarities below 30% with both the A. europaeus and the A. xylinus type strain. These results plus the results obtained in the hybridization studies with the oligonucleotide probe 23seu show that A. europaeus, A. intermedius and A. xylinus belong to a branch of rapidly evolving Acetobacter strains. The oligonucleotide probe 23seu is not A. europaeus-specific but useful as a tool for the differentiation of A. intermedius from A. europaeus and A. xylinus. The overall 16S rRNA gene similarities between A. europaeus, A. xylinus and A. intermedius of more than 97% made DNA pairing studies necessary as an additional genotypic parameter to underline the species status of A. intermedius. According to WAYNE et al. (1987) a species would include strains with approximately 70% or greater DNA-DNA relatedness and which possess a common phenotypic characteristic. Previous DNA-DNA reassociation experiments with the A. europaeus type strain and all Acetobacter strains from the 4 phenons (SIEVERS and TEUBER, 1995, Kneubiihler, 1994 and unpublished results) had already resulted in the shifting of the strains of phenon 3 to subspecies level and grouping of the strains of phenon 4 in a possible new species. DNA-DNA homology below the 70% threshold value for species delineation as well as three phenotypic differ-
228
C. BOESCH et a!.
ences (growth with or without acetic acid, growth on media containing 20% glucose) clearly support the species status of A. intermedius. Further DNA-DNA hybridization experiments with A. intermedius and Acetobacter strains from vinegar fermenters with the A. intermedius type strain as a probe were performed more than once and produced the DNA-DNA homology values listed in Table 9. Outliers from the values are the strain TSA8 from phenon 4 which shows less than 30% DNA homology with A. europaeus (SIEVERS and TEUBER, 1995), hybridizes with the oligonucleotide probe 23seu and shows growth only on AE medium, yet has a DNAreassociation value of 99% with the A. intermedius type strain; this strain is therefore listed as Acetobacter sp. in Table 1 and Table 7. Further, the strain HE from ph en on 3 does not hybridize with the oligonucleotide probe 23seu, but its 16S rDNA sequence is identical to that of A. europaeus and its DNA reassociates at below 30% with the A. intermedius type strain and below 60% with the A. europaeus typestrain; this strain is listed as Acetobacter sp. in Table 1. Despite missing phenotypic features for some strains in the A. intermedius group and the mentioned exceptions, the 16S rDNA sequences and DNA-DNA reassociation experiments allow the division of all Acetobacter strains from our culture collection into either A. europaeus or A. intermedius, the exceptions being strain HE from phenon 3 and strains TSA8, TSN1 and TSN5 from phenon 4 (all Acetobacter sp.). Bacterial isolates derived from the tea fungus on YPM (mannitol, peptone, yeast extract)- and AE-medium were previously classified to belong to the genus Acetobacter (SIEVERS et aI., 1995). Based on the biosynthesis of cellulose/acetan, the results of the SDS-PAGE analysis and the growth on media without acetic acid, these isolates were tentatively identified as belonging to the species A. xylinus. The low clustering value of the A. xylinus type strain in the protein profile of 58% was attributed to the fact that this type strain is genotypically not thought to be representative for the species A. xylinus (unpublished results from our laboratory). Though these strains have been lost due to cultivation failure, the results based on the characterization of the newly isolated strain TF2 from the same Kombucha beverage indicate that the published strains (SIEVERS et aI., 1995) would belong to the newly described species A. intermedius. YAMADA et aI., 1997, have described the phylogenetic relationships of the acetic acid bacteria based on partial 16S rRNA gene sequences and propose the elevation of a subgenus Gluconoacetobacter to the generic level. It is generally accepted that only complete sequences allow reliable phylogenetic comparison with the available database of complete or almost complete sequences (STACKEBRANDT and GOEBEL, 1994). The construction of a phylogenetic tree based on partial base sequences must therefore be regarded as not valid. Also, one of the mentioned characteristics differentiating the 4 proposed genera, growth on mannitol agar only for the genera Gluconobacter and Gluconoacetobacter, is definitely wrong. In lists of bacterial cultures (Laboratorium Microbiologie Rijksuniversiteit Gent, BCCM Catalogue, 1989) YPM
agar is suggested as medium for Acetobacter and Gluconobacter strains. The main purpose of industrial strains is to produce high concentrations of acetic acid from ethanol. Industrial submerced fermenters are started by inoculation with 'seed vinegar' (microbiologically undefined remains from previous fermentations). Studies on plasmid profiles (SIEVERS and TEUBER, 1995) clearly show that the transfer of the vinegar microflora from one fermenter to another is possible which suggests that industrially used strains have over the last two hundred years and more been adapted to an extreme niche and thus subjected to high evolutionary pressure. The close phenotypic and genotypic relationship of the three species A. europaeus, A. intermedius and A. xylinus would indicate that they belong to a rapidly evolving branch of the acetic acid bacteria. Description of the new species Acetobacter intermedius (sp. nov.), L.adj.m. intermedius, in the middle between, to characterize the phenotype of this strain that has phenotypic properties of both Acetobacter europaeus and Acetobacter xylinus: growth in high concentrations of acetic acid and growth without acetic acid. Cells are Gram-negative, rod shaped, straight, 0.7 )lm by 2 )lm, occurring in pairs. Mobility and flagella not observed. Endospores are not formed. Obligately aerobic. Pale colonies, no pigments produced. Catalase positive. Oxidase negative. Oxidizes ethanol to acetic acid. Acetate is oxidized to CO 2 but only below an acetate concentration of 6%. No oxidation of glucose to 5-ketogluconate or 2-ketogluconate. DNA base composition is 61.55%. Acetobacter intermedius occurs in tea fungus beverages, spirit vinegar and cider vinegar. No significant DNA-DNA hybridization with the type strains of Acetobacter aceti, A. pasteurianus, A. liquefaciens, A. hansenii, A. xylinus, A. methanolicus, and Gluconobacter oxydans. DNA-DNA hybridization with the Acetobacter europaeus type strain of <60%. Type strain: Acetobacter intermedius TF2 (DSMl1804). Deposited at the Deutsche Sammlung von Mikroorganismen und Zellkulturen, Braunschweig. Isolated from a commercially available tea fungus beverage (Kombucha) in Switzerland.
Literature AMANN, R.L., BINDER, B.]., OLSON, R.]., CHISHOLM, S.W., DEVEREUX, R., STAHL, D.A.: Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. App!. Environ. Microbio!. 56, 1919-1925 (1990). AMANN, R.L., LUDWIG, W., SCHLEIFER, K.H.: Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59, 143-169, (1995). BOESCH, c.: Vergleich von spezifischen DNA-Sequenzen in der Taxonomie von Essigsaurebakterien. Diploma work, Institute of Food Science and Technology, ETH Zurich, Switzerland (1994).
Acetobacter intermedius sp. nov.
BROSIUS, J., DULL, T.]., SLEETER D.D., NOLLER, H.E: Gene organization and primary structure of a ribosomal RNA operon from Escherichia coli. J. Mol. Biol. 148, 107-127 (1981). DASEN, G., SMUTNY, J., TEUBER, M., MElLE, L.: Classification and identification of propionibacteria based on ribosomal RNA genes and PCR system. System. Appl. Microbiol. (in press). ENTANI, E., OHMORI, S., MASAI, H., SUZUKI, K.-I.: Acetobacter polyoxogenes sp. nov., a new species of an acetic acid bacterium useful for production of vinegar with high acidity. J. Gen. Appl. Microbiol. 31,475-490 (1985). EUZEBY, J.P.K: Revised nomenclature of specific or subspecific epithets that do not agree in gender with generic names that end in -bacter. Int. J. Syst. Bacteriol. 47, 585 (1997). GOESSELE, J., SWINGS, J., DE LEY, ].: A rapid, simple and simultaneous detection of 2-keto-, 5-keto- and 2,5-diketogluconic acid by thin layer chromatography in culture media of acetic acid bacteria. Zbl. Bakt. Hyg., I. Abt. Orig. Cl, 178-181 (1980). KERSTERS, K.: Numerical Methods in the classification of bacteria by protein electrophoresis. In: GOODFELLOW, M., JONES, D., PRIEST, EG. (Eds.), Computer-Assisted Bacterial Systematics (SCM Special Publication No. 15), Academic Press, London, 337-368 (1985). KNEUBDHLER, B.: Numerische Analyse von Proteinprofilen und DNA-DNA-Hybridisierung zur Charakterisierung und Differenzierung der Acetobacter-Flora industrieller Essigfermenter der Schweiz. Thesis no. 10674. ETH-Zurich, Switzerland (1994). RbNNER, S. and STACKEBRANDT, E.: Development of 23S rDNAoligonucleotide Probes for the Identification of Salmonella species. System. Appl. Microbiol. 17,257-264 (1994). SIEVERS, M., TEUBER, M.: The microbiology and taxonomy of Acetobacter europaeus in commercial vinegar production. ]. Appl. Bacteriol. Symposium Supplement 79, 84-95 (1995). SIEVERS, M., GABERTHDEL, c., BOESCH, c., LUDWIG, W., TEUBER, M.: Phylogenetic position of Gluconobacter species as a coherent cluster separated from all Acetobacter species on the basis of 16S ribosomal RNA sequences. FEMS Microbiol. Lett. 126,123-126(1995).
229
SIEVERS, M., LANINI, c., WEBER, A., SCHULER-SCHMID,U., TEUBER, M.: Microbiology and Fermentation Balance in a Kombucha Beverage Obtained from a Tea Fungus Fermentation. System. Appl. Microbiol. 18,590-594 (1995). SIEVERS, M., LUDWIG, W., and TEUBER, M.: Phylogenetic positioning of Acetobacter, Gluconobacter, Rhodopila and Acidiphilium species as a branch of acidophilic bacteria in the a-subclass of Proteobacteria based on 165 ribosomal DNA sequences. System. Appl. Microbiol. 17, 189-196 (1994). SIEVERS, M., SELLMER, S., TEUBER, M.: Acetobacter europaeus sp. nov., a main component of industrial vinegar fermenters in Central Europe. System. Appl. Microbiol. 15, 386-392 (1992). SOUTHERN, E.M.: Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. BioI. 98, 503-511 (1975). STACKEBRANDT, E., GOEBEL, B.M.: Taxonomic Note: A place for DNA-DNA reassociation and 16S rRNA sequence analysis in the present species definition in bacteriology. Int. J. Syst. Bacteriol. 44, 846-849 (1994). WAYNE. L.G., BRENNER, D.]., COLWELL, R.R., GRIMONT, P.A.D., KANDLER, 0., KRICHEVSKY, M.I., MOORE, L.H., MOORE, W.E.C., MURRAY, R.G.E., STACKEBRANDT, E., STARR, M.P. and TRDPER, H.G.: Report of the Ad Hoc Committee on Reconciliation of Approaches to Bacterial Systematics. Int. ]. Syst. Bacteriol. 37,463-464 (1987). YAMADA, Y., HOSHINO, K. and ISHIKAWA, T.: The Phylogeny of Acetic Acid Bacteria Based on the Partial Sequences of 16S Ribosomal RNA: The Elevation of the Subgenus Gluconoacetobacter to the Generic Level. Biosci. Biotech. Biochem. 61 (8), 1244-1251 (1997).
Corresponding author: M. TEUBER, Institut fur Lebensmittelwissenschaft, Labor fur Lebensmittelmikrobiologie, ETH-Zurich, Schmelbergstrasse 9, 8092 Zurich, Switzerland e-mail: [email protected]
SYSTEI'VL4TIC AND APPLIED MICROBIOLOGY
Acetobacter intermedius, sp. nov. CORNELIA BOESCH!, JANJA TRCEK 2 , MARTIN SIEVERS 3 and MICHAEL TEUBER 1 lInstitut fur Lebensmittelwissenschaft, Labor fur Lebensmittelmikrobiologie, ETH-Zurich, Zuerich, Switzerland 2Biotechnical Faculty, Department of Food Science and Technology, University of Ljubljana, Ljubljana, Slovenia 3Ingenieurschule Wadenswil, Wadenswil, Switzerland Received August 26, 1997
Summary Strains of a new species in the genus Acetobacter, for which we propose the name A. intermedius sp. nov., were isolated and characterized in pure culture from different sources (Kombucha beverage, cider vinegar, spirit vinegar) and different countries (Switzerland, Slovenia). The isolated strains grow in media with 3% acetic acid and 3% ethanol as does A. europaeus, do, however, not require acetic acid for growth. These characteristics phenotypically position A. intermedius between A. europaeus and A. xylinus. DNA-DNA hybridizations of A. intermedius-DNA with DNA of the type strains of Acetobacter europaeus, A. xylinus, A. aceti, A. hansenii, A. liquefaciens, A. methanolicus, A. pasteurianus, A. diazotrophicus, Gluconobacter oxydans and Escherichia coli HB101 indicated less than 60% DNA similarity. The important features of the new species are described. Acetobacter intermedius strain TF2 (DSMl1804) isolated from the liquid phase of a tea fungus beverage (Kombucha) is the type strain. Key words: A. intermedius sp. nov.
Introduction Differentiation of Acetobacter-strains from industrial vinegar fermentations is difficult due to their unwillingness to grow well on semisolid media. Data of DNA similarities (30 to 100%) among industrial strains revealed their division into more than one group suggesting the occurrence of several new species. Based on these data, Acetobacter europaeus was described as a main species of industrial vinegar fermenters in central Europe (SIEVERS et aI., 1992). The main biochemical differentiation characteristics of the isolates from high percentage vinegar productions against the traditional species are both a strong tolerance to acetic acid, 4 to 8 % in AE-agar and 10-14 % in industrial fermentations, and requirement of acetic acid for growth. Due to the lack of suitable differentiating physiological markers, total cellular proteins of 44 Acetobacter and Gluconobacter strains were characterized by SDS-PAGE in our laboratory (KERSTERS, 1985) to provide a reproducible phenotypic property (KNEUBOHLER, 1994). Computerized numerical analysis of the protein profiles showed that the 25 different strains from industrial vinegar fermentations clustered in 4 phenons separated from all other investigated strains with a close linkage of A. xylinus (EUZEBY, 1997) to the A. europaeus group. The protein patterns correlated with the results of DNA-
DNA hybridizations with the A. europaeus type strain (SIEVERS and TEUBER, 1995). In the present reports the low (30-<70%) DNA-DNA hybridization similarity group (SIEVERS et aI., 1992) and the 4 phenons described by protein profiling (SIEVERS and TEUBER, 1995) are analyzed on a phenotypic and genotypic level and compared to an isolate (TF2) from a tea fungus beverage (SIEVERS et aI., 1995). Based on 16S rDNA sequence alignments, missing hybridization signals with an oligonucleotide probe based on the 23S rRNA of A. europaeus, DNA-DNA hybridization data and phenotypic properties, strain TF2 is described as the new species Acetobacter intermedius in the genus Acetobacter.
Materials and Methods Bacterial strains and cultivation: Bacterial strains used in this study are shown in Table 1. Escherichia coli strain HB101 was cultivated on Luria-Bertani medium. Acetobacter europaeus and Acetobacter intermedius strains were cultivated on a modified AE-medium (ENTANI et aI., 1985): 0.5% glucose, 0.3% yeast extract, 0.4% peptone, 0.9% agarose, 3%(v/v) ethanol and 3%(v/v) acetic acid. AEG20% is AE-medium with 20% glucose. Incubation was at 30°C and 92-96% relative air humidity. Ace-
Acetobacter intermedius sp. nov.
221
Table 1. Bacterial strains. Species
Strain designation
source
original substrate/ fermenter type)
Acetobacter aceti Acetobacter pasteurianus Acetobacter liquefaciens Acetobacter hansenii Acetobacter xylinus Acetobacter xylinus Acetobacter xylinus Acetobacter xylinus Acetobacter xylinus Acetobacter methanolicus Acetobacter diazotrophicus Gluconobacter oxydans Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter europaeus Acetobacter sp. Acetobacter sp. Acetobacter sp. Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter intermedius Acetobacter sp.
NCIB 8621 T LMD22.F IFO 12388 T NCIB 8746 T NCIB 11664T NCIB613 De Ley 25 ATCC 23768 ATCC 14851 MB58 T ATCC 49037T ATCC 19357T DES lIT DOSll TSA4 WW Sl TSA9 KO SW3 SSPR1 JK2 VI TSN5 TSN1 TSA8 TF2T JK3 JKD TSA10 0SSPR TSN3 TSA3 TSA7 KRI TSN4 E1 E2 0SW HE
DSM 3508 LMG 1262 LMG 1328 LMG 1527 LMG 1515 LMG 1510 LMG25 DSM2004 DSM46604 LMG 1668 LMG 7603 DSM 3503 DSM 6160 DSM 6161 LMETH LMETH LMETH LMETH LMETH LMETH LMETH ZIM B021 ZIMB026 LMETH LMETH LMETH DSM 11804 ZIM B022 ZIM B025 LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH
submersed submersed generator submersed submersed generator submersed submersed submersed cider vinegar/submersed wine vinegar/submersed generator generator generator Kombucha cider vinegar/submersed cider vinegar/submersed generator submersed generator generator generator submersed generator submersed submersed submersed submersed
T type strains ZIM LM ETH DSMZ LMG ')submersed generator
-
Zbirka Industrijskih Mikroorganizmov, Ljubljana, Slovenia Labor fur Lebensmittelmikrobiologie, ETH Zurich, Switzerland Deutsche Sammlung von Mikroorganismen und Zellkulturen Laboratorium voor Microbiologie, Rijksuniversiteit, Gent submersed fermentation trickle fermenter
tobacter and Gluconobacter type strains were cultivated in MRS medium or in GY medium containing 5% glucose and 1 % yeast extract; a concentrated glucose-solution was autoclaved separately to avoid Maillard reactions. Incubation was at 30°C. GYG20% is GY-medium containing 20% glucose. To analyze growth behaviour with different sugars, alcohols and acids, YPmedium containing 0.3% yeast extract and 0.4% peptone was supplemented with 20% of the according carbon source. Formation of 2- and 5-ketogluconic acid: The formation of 2- and 5-ketogluconic acid was assayed with strain TF2 cultivated in shaken culture in 2% glucose, 0.5% yeast extract, 0.3% peptone and 2.0% acetic acid for 14 to 21 days. 2- and
5-ketogluconic acid were determined by thin layer chromatography of growth medium (GOESSELE et al., 1980). In vitro amplification of the 16S rRNA gene: A loopful of cells from agar were suspended in a 10 mM Tris, 10 mM EDTA-buffer (pH 8.0). Cell-lysis was induced by addition of 20 pg Proteinase K and incubation at 55°C for 30 minutes. Nucleic acids were purified by extraction with 100 pi phenol (saturated with 3% NaCl) and subsequently by extraction with 100 pi methylene chloride/isoamyl alcohol (24:1 [vol.!vol.]). 1-3 pi of the aqueous phase were directly used for PCR. 16S rDNA was amplified using oligonucleotide primers complementary to highly conserved regions of bacterial rRNA genes
222
C. BOESCH et al.
(AMMAN et ai, 1990). The 5'- and 3'-terminal primers were GAGTTTGAT(Cff)(C/A)TGGCTCA (positions 9 to 26, according to the Escherichia coli numbering system (BROSIUS et. al., 1981)) and CA(GIT)AAAGGAGGTGATCC (positions 1545 to 1529), respectively. PCR was performed in a total volume of 100 JlI containing 1 U DynaZyme™ DNA polymerase F-501L (FinnZymes), 10 JlI lOx DynaZyme™-reaction buffer, 1 JlM of each primer and 300 JlM of each deoxynucleoside triphosphate. The mixture was amplified in a thermal cycler (Perkin Elmer Cetus 480) as follows: an initial denaturation at 94°C for 5 min. followed by: 92 °C, 1 min.; 47°C, 2 min.; 72 °C, 2 min. and 35 cycles. In vitro amplification of the 23S rRNA gene: Preparation for PCR was as described above for 16S rRNA. The 5'- and 3'-terminal primers were TGCGGCTGGATCACCTCC (positions 3039 to 3056 (BROSIUS et al., 1981)) and CCCGCTTAGATGCTTTCAGC (positions 6262 to 6243 (BROSIUS et al., 1981)), respectively. PCR was performed in a total volume of 100 JlI containing 4 U DynaZyme™ EXT DNA polymerase F-505S (FinnZymes), 10 JlI lOx DynaZyme™ EXT DNA-reaction buffer, 2.0 mM MgCI 2, 1 ]lM of each primer and 360 JlM of each deoxynucleoside triphosphate. The mixture was subjected to 30 cycles of amplification in a thermal cycler (Biometra Personal CycierTM). An initial denaturation at 93°C for 2 min. was followed by: 93 °C, 20 sec.; 60°C, 30 sec.; 68 °C, 3 min. Purification of PCR-product: PCR-products were directly purified with the NucleoSpin Extract kit (Macherey-Nagel, Oensingen, Switzerland). Elution from the DNA binding column was with bidistilled water. Direct Sequencing of PCR product: Automated cycle sequencing of 300 ng purified PCR products (16S rDNA and 23S rDNA) was performed using the Thermo Sequenase fluorescent
labeled primer cycle sequencing kit with 7-deaza-dGTP according to the manufacturer's instructions (Amersham International pic, Buckinghamshire, England). Synthetic oligonucleotide primers (Microsynth, Balgach, Switzerland) were Cy5™-Iabelled; the primers and their positions according to the Escherichia coli numbering system (BROSIUS et al., 1981) are listed in Table 2 and Table 3. Unless otherwise indicated sequences of primers were selected from published data. Cycle sequencing was performed on a Biometra Personal CycierTM (Biometra, Gottingen, Germany) with the following parameters: 95°C, 30 sec.; 50 °C, 30 sec.; 72 °C, 60 sec. and 25 cycles (DASEN et al., 1998). Detection of the denatured (85°C, 3 min.) cycle-sequenced samples was performed on an ALFexpress™ DNA sequencer (Pharmacia, Uppsala, Sweden) using a hydrolink long ranger gel (Long RangerTM, FMC BioProducts, Rockland, USA). Data Analysis: Computer analyses of the DNA sequences were performed using the GCG software package of the University of Wisconsin Genetics Computer Group (Madison, Wis., USA). The 16S rDNA and 23S rDNA nucleotide sequences of A. intermedius have been deposited in the EMBL Nucleotide Sequence Library (Cambridge, UK) under the accession numbers Y14694 and Y14680, respectively. DNA-DNA hybridization with chromosomal DNA: Total DNA of Acetobacter and Gluconobacter species was prepared according to a modified protocol of the method by Marmur (1961). DNA probes (0.1-0.2 Jlg) for hybridizations were internally labeled with 20 JlCi [a- 32 P]dATP (3000 Ci/mmol) using Klenow Polymerase and the hexanucleotide mixture Nr. 1277081 (Boehringer, Mannheim, Germany) according to the instructions of the manufacturer. Pharmacia NAP-10 columns (Pharmacia, Uppsala, Sweden) were used to remove unincorporated 32P-Iabeled nucleotides.
Table 2. Primers for 16S rDNA sequencing. Primer
Sequence
Target site 1
Specificity
References
Eubac338 MS350 uni515 uni775 uni1088 uni1392 uni1392r
5'ACTCCTACGGGAGGCAGO' 5'CTGCTGCCTCCCGTN' 5'ACCGCGGCTGCTGGCAC3' 5'GGMTTAGATACCCTGGTAGTCO' 5'GGTTAAGTCCCGCAACGAGC3' 5'GTACACACCGCCCGTCN' 5TGACGGGCGGTGTGTAC3'
338-355 350-336 515-498 775-796 1088-1107 1392-1408 1392-1376
Bacteria Bacteria Bacteria Bacteria Bacteria Bacteria Bacteria
AMANN et al., 1990 AMANN et al., 1990 this study this study AMANN et al., 1995 this study this study
1
positions according to the Escherichia coli numbering system (BROSIUS et al., 1981)
Table 3. Primers for 23S rDNA sequencing. Primer
Sequence
Target site 1
Specificity
References
b-1 bO blrev b1 b2rev b2 b3 b4 b5 b6 b7
5'ATACGGGGCTATCACCCG3' 5'CCTTTCCCTCACGGTACF 5TTTCGGGGAGAACCAGCTAF s'ATAGCTGGTTCTCCCCGAAA3' 5'GATGGCTGCTTCTAAGCCAAO' 5'GTTGGCTTAGAAGCAGCCATO' 5'CCCCTAAGGCGAGGCCGAAAG3' 5'AGAGAATACCAAGGCGCTTGl' 5'AAGTTCCGACCTGCACGAAT3' 5'GTAACGGAGGCGCGCGAF 5'AGAACGTCGTGAGACAGTTC3'
3841-3824 3974-3957 4321-4302 4321-4302 4579-4559 4559-4579 4848-4868 5130-5194 5452-5471 5757-5774 6027-6080
a-Proteobacteria Proteobacteria Bacteria Bacteria Bacteria Bacteria Bacteria Proteobacteria Bacteria a-Proteo bacteria a-Proteobacteria
this study this study this study this study this study this study this study this study this study this study this study
1
positions according to the Escherichia coli numbering system (BROSIUS et al., 1981)
Acetobacter intermedius sp. nov.
100-500 ng of Acetobacter intermedius DNA were restricted with EcoRV prior to labeling. Equal amounts (1-2 rg) of EcoRV-restricted DNA from various Acetobacter and Gluconobacter strains were sepatated by gelelectrophoresis and vacuum-blotted (Vacu-Blot, Biometra, Switzerland) to a ZetaProbe® GT blotting membrane (Biorad, Switzerland) according to the method of SOUTHERN (1975). The membrane was dried at 80°C and 30 min. to fix the DNA to the membrane. Hybridization with labeled A. intermedius probe was performed at 65°C (SIEVERS et aI., 1992) in 0.25 M Na 2 HP0 4 (pH 7.2) and 7% SDS according to the standard protocol for Zeta-Probe® GT blotting membranes. Washing of the membrane was twice 15 min. in 20 mM Na 2HP0 4 (pH 7.2) and 5% SDS and then twice in 20mM Na 2 HP0 4 (pH 7.2) and 1 % SDS, all washing steps were performed at 65°C. The wet membranes were sealed into plastic bags and the labeled bands were visualized on X-ray films (Fuji RX) with intensifying screens for 1 day to a week. Analysis of the X-ray films was performed with the program WIN CAM, Cybertech, Berlin. The hybridization signal obtained by restricted and labeled DNA-probe with the identical DNA on the membrane was defined as the 100% intensity signal, the intensities of the signals obtained with EcoRV restricted DNA of other Acetobacter and Gluconobacter strains were calculated by the program WIN CAM, calibration of the values was through linear regression. Oligonucleotide labeling: 10pmoll rl of the respective synthetic oligonucleotide primer (Microsynth, Balgach, Switzerland) was end-labeled with 25 rCi [y-3 2P]dATP (6000 Ci/mmol) using T4 Polynucleotide Kinase according to the manufacturer's instructions (Biolabs, Beverly, MA, USA). Pharmacia NICK columns (Pharmacia, Uppsala, Sweden) were used to remove unincorporated 12P-labeled nucleotides. The probes used in colony hybridization experiments are listed in Table 4 .. Preparation of dot blot filters: A loopful of freshly grown cells on solid media was suspended in 1ml of a 0.45% NaCland 0.5% Lysozyme-solution and incubated for 15 min. at 37°C. After centrifugation 0.9ml of the supernatant was removed and the cells resuspended in the remaining 0.1 ml. 10 rl of the suspension were pipetted on to a Zeta-Probe® GT Blotting Membrane (Biorad). Denaturation of the DNA was performed on 1M NaCl saturated 3MM paper (Whatman) for 4 min. Neutralization and washing steps were performed in the same manner with 1 M Tris pH 8 and 2xSSC (lxSSC is 0.15 M NaCI plus 0.015 M sodium citrate), respectively. Hybridization conditions: Hybridizations were carried out in a hybridization oven (Hybaid). Hybridization conditions were the relevant temperature for the respective oligonucleotide probe (Table 4) and 0.25 M Na2HP04 (pH 7.2) and 7% SDS according to the standard protocol for Zeta-Probe® GT blotting membranes. Washing of the membrane was twice 15 min. in 20 mM Na 2HP0 4 (pH 7.2) and 5% SDS and then twice in 20 mM Na2HP04 (pH 7.2) and 1 % SDS, the washing steps were performed at the same temperature as the hybridization. The wet membranes were sealed into plastic bags and the la-
beled bands were visualized on X-ray films (Fuji RX) with intensifying screens for 1 day to a week. Probe stripping and rehybridization: Reprobing was desired to confirm the presence of nucleic acids on the membranes. The Zeta-Probe® GT blotting membrane was prevented from drying between hybridizations. Stripping of the membrane was at 95°C for 1.5 h in O.lx SSC and 0.5% SDS. The presence of nucleic acids on the membranes was confirmed by hybridization with [a- 32 P]dATP-labeled 23S rDNA of A. europaeus at 65°C.
Results Phylogenetic positioning of Acetobacter intermedius The 165 rRNA sequence of A. intermedius (EMBL accession number is Y14694), was determined by direct seGoxydans
Gfmteurii G cerilUlS G asaii
A aceti A pasteurianus A methanolicus
A diozntrophicus A liquefaciens
R globifonnis
A hansenii
Aintennedius AxylilUlS Aell1VfXI£US
Fig. 1. Phylogenetic tree of all Acetobacter and Gluconobacter type strains. The tree was constructed with the GCG software package using the programs pileup, distances (Jin-Nei Gamma) and growtree (UPGMA). The bar indicates 1 % estimated sequence difference. Based on information from SIEVERS and TEUBER, 1995.
Table 4. Sequence of probes used in colony hybridization experiments. Probe
Sequence
Target organism
Target site I
Hybridization Temperature
Source or reference
oligotf2 23seu oligoxyl
5TCAGAAACAACCACTGGC3' 5'AATGCGCCAAAAGCCGGAT3' S'GACTGTTCCTGATATTACGC1'
A. intermedius type strain A. europaeus strains A. xylinus type strain
2272-2249 3455-3437 2348-2324
42°C 45°C 54°C
this study this study this study
1
223
positions according to the Escherichia coli numbering system (BROSIUS et aI., 1981)
C. BOESCH et al.
224
quencing of the PCR-amplified rRNA gene products. The derived 16S rDNA primary structure was compared with homologous sequences taken from published databases including the sequences of all Gluconobacter and Acetobacter species previously described (SIEVERS et al., 1995). A phylogenetic tree reflecting the close relationships of the type strains is shown in Figure 1. The tree is based on the information contained in the 16S rRNA primary structures of the Acetobacter and Gluconobacter type strains and was constructed using the
1
2
4
6
7
8
programs "pileup", "distances" and "growtree" from the GCG software package. The very close clustering of A. europaeus, A. intermedius and A. xylinus only represents their present evolutionary status, their respective positioning in their branch of the phylogenetic tree is no proof for their species status (LUDWIG W., 1998, personal communication). As shown in Table 5, the overall 16S rRNA nucleotide sequence similarities between A. europaeus, A. xylinus and A. intermedius are higher than 97% which made DNA pairing studies necessary as an
9 10 11 12 1
14 1
16
17
Fig. 2. DNA-DNA hybridization of EcoRV digested DNA with [a- 32P]dATP-labeled A. intermediusT-DNA as a probe after gel electrophoresis: lane 1, Escherichia coli HB101; lane 2, Gluconobacter oxydansT; lane 3, A. acett'T; lane 4, A. pasteurianusT; lane 5, A. methanolicusT; lane 6, A. liquefaciens T; lane ?, A. diazotrophicus\ lane 8, A. hansenit'T; lane 9, A. xylinusT; lane 10, A. europaeusT; lane 11, A. intermediusT; lane 12, A. europaeus (JK2); lane 13, A. europaeus (VI); lane 14, A. europaeus (KO); lane 15, A. intermedius (TSA10); lane 16, A. europaeus (TSA8); lanel?, A. intermedius (0SSPR). Top: Ethidium bromide stained chromosomal DNA pattern Bottom: Autoradiograph of chromosomal DNA pattern after hybridization with labeled A. inter-
mediusT-DNA
Acetobacter intermedius sp. nov.
225
Table 5. Overall 16S rRNA sequence similarity values for Acetobacter species, Cluconobacter oxydans and Rhodopila globiformis. Abbreviations: Ain, Acetobacter intermedius; Aeu, Acetobacter europaeus; Aac, Acetobacter aceti; Axy, Acetobacter xylinus; Aha, Acetobacter hansenii; Adi, Acetobacter diazotrophicus; Ali, Acetobacter liquefaciens; Apa, Acetobacter pasteurianus; Cox, Cluconobacter oxydans; RgI, Rhodopila globiformis. %similarity Organism A. intermedius A. europaeus A. xylinus A. hansenii A. liquefaciens A. diazotrophicus A. pasteurianus A. aceti C.oxydans R. globiformis
Ain
Aeu 99.5
Axy 99.2 99.5
Aha 98.2 98.4 98.3
additional genotypic parameter to underline the species status of A. intermedius (STACKEBRANDT and GOEBEL, 1994). DNA-DNA hybridization studies Figure 2 shows the DNA-DNA hybridization of EcoRV-restricted and [a- 32 PldATP-Iabeled A. intermediusT-DNA as a probe with the type strains of A. europaeus, A. intermedius, A. xylinus, A. aceti, A. hansenii, A. liquefaciens, A. methanolicus, A. pasteurianus, Gluconobacter oxydans and Escherichia coli HBI0l and Acetobacter strains from our laboratory. The highest homology of the A. intermedius-type strain with all the type strains of Acetobacer and Gluconobacter was 60% with the type strain of A. europaeus. DNA-DNA hybridization of labeled A. intermediusT-DNA as a probe with other strains yielded the following results: 65% with strain ]K2, 40% with strain VI, 65% with strain KO, 65% with TSAI0, 99% with TSA8 and 90% with 0SSPR. Intense hybridization bands appear in the following strains in Figure 2: lane 3 (G. oxydansT), lane 9 (A. xylinusT), lane 11 (A. intermediusT), lane 12 (strain ]K2) and lane 16 (strain TSA8). Such strong signals could be attributed to the occurrence of similar IS-elements in these strains. The intensities of the radioactive signals on the developed film were determined by quantitative densitometry. Table 6. Overall 23S rRNA sequence similarity values for Acetobacter intermedius, Acetobacter europaeus and Acetobacter xylinus. Abbreviations: Ain, Acetobacter intermedius; Aeu, Acetobacter europaeus; Axy, Acetobacter xylinus. %similarity Organism A. intermedius A. europaeus A. xylinus
Ain
Aeu
Axy
99.2
99 98.9
Ali 97.0 97.4 97.2 97.1
Adi 96.2 97.0 96.7 97.0 98.4
Apa 95.1 95.1 94.8 94.5 95.7 95.3
Aac 95.6 95.4 95.2 94.2 95.8 95.5 96.8
Cox 95.0 94.8 94.6 94.4 94.2 94.2 95.0 95.6
Rgi 92.4 93.9 93.7 93.7 94.0 94.0 92.7 92.8 92.6
The DNA-DNA hybridization values in Table 9 for the strains other than the A. europaeus, A. intermedius and the A. xylinus type strains were obtained by repeated hybridization experiments, the average of the obtained intensities of the hybridization signals was chosen as the result. The strains ]K2 and VI isolated from slovene cider- and wine-vinegar are currently under investigation in a separate thesis (TRCEK, in preparation). Therefore, they are not further discussed in the present work. Generation of 23S rDNA oligonucleotide probes The alignment of the 23S rDNA sequences of the type strains of A. europaeus, A. xylinus and A. intermedius (EMBL accession numbers are X8977I, X898I2 and
Table 7. Acetobacter species that give a positive signal in hybridization experiments with the oligonucleotide probe 23seu derived from the A. europaeus type strain. Empty fields in the column titled "phenon" indicate that these strains were not included in the protein profile experiments performed by Kneubiihler (1994). strain
phenon
identity
source
A. europaeusT A. europaeus S3 TSA4 WW Sl TSA9 KO SW3 TSN5 TSN1 TSA8 JK2 V1 De Ley 25 ATCC 23768 ATCC 10245
1 1 1 1 1 2 2 2 2 4 4 4
A. europaeus A. europaeus A. europaeus A. europaeus A. europaeus A. europaeus A. europaeus A. europaeus A. europaeus Acetobacter sp. Acetobacter sp. Acetobacter sp. A. europaeus A. europaeus "A. xylinus" "A. xylinus" "A. xylinus"
DSM 6061 DSM 6161 LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH LMETH ZIM B021 ZIM B026 LMC25 DSM46604 DSM2004
226
C. BOESCH et al.
Y14680, respectively) revealed useful target sites for oligonucleotide probes (Table 4). An oligonucleotide probe (23seu) developed for A. europaeus strains was tested with Acetobacter strains from our culture collection isolated from generators and aceta tors in Switzerland, Germany, Slovenia and Spain. Acetobacter strains that gave a positive signal in hybridization experiments with the oligonucleotide probe 23seu derived from the A. europaeus type strain are listed in Table 7. The probe gave no hybridization signal with the A. xylinus type strain NCIB 11664 and the A. xylinus strain NCIB 613, however, hybridized with the following strains which are
classified as Acetobacter xylinus: De Ley 25, ATCC 23768 and ATCC 10245. No hybridization signals were obtained with the type strains of A. intermedius, A. xyli-
nus, A. aceti, A. hansen ii, A. liquefaciens, A. methanolicus, A. pasteurianus, Gluconobacter oxydans and Escherichia coli HB101 and also no hybridization signals with strains JK3 and JKD from cider vinegar, strain TF2 (A. intermedius, sp. nov) from a tea fungus beverage and all Acetobacter strains in protein profile phenon 3 (SIEVERS and TEUBER, 1995) (Table 9). The probe based on the 23S rRNA of the A. xylinus typestrain gave no signal with the A. europaeus- and
Table 8. Characteristics differentiating the type strains of A. intermedius, A. xylinus and A. europaeus. Growth on AEG20%
Growth on GYG20% A. xylinusT A. europaeusT A. intermedius T TF2
Growth onAE
+
Growth onGY
Growth on MRS
+ + +
+
+
+
Cellulose formation
GC content of DNA (mol%)
+
55-63 56-58 61.55
+
Table 9. Protein profile homology of the 4 phenons of strains isolated from generators and aceta tors in Europe and their DNADNA homology with A. intermedius and A. europaeus (SIEVERS and TEUBER, 1995). The given values (in %) indicate the values for all the strains in the respective phenon.
Phenon 1
Phenon 2
Phenon 3
Phenon 4
Strains
Protein profile homology
approximate DNA-DNA homology to A. europaeus
approximate DNA-DNA homology to A. intermedius
Hybridization with Probe for A. europaeus
DSM 6161 S3 DSM6160 TSA4 WW S1 TSA9 KO SW3 SSPR1 TSN3 0SSPR TSA7 TSA3 TSA6 KRI TSA10 TSN4 0SW HE' E1 E2 TSN5
77.1
90-100%
<60%
yes
87.7 ± 4.6%
ca. 75%
60%
yes
70
<70%
70-100%
no
<30%
ca. 80%
yes, except A. xylinus NCIB
TSN1 TSA8 NCIB 11664
±
64.5
±
4.4%
5.9%
±
1.6%
11664
1 strain HE has identical16S rDNA sequence as the A. europaeus typestrain, DNA association value to the A. intermedius typestrain is <30%.
Acetobacter intermedius sp. nov.
A. intermedius-type strains, no further testing was performed with other A. xylinus strains. The A. intermedius probe is specific only for the type strain - the difference in the 16S rDNA sequence, however, should allow for the construction of an A. intermedius-specific probe.
165 rONA sequences of strains in phenon 3 The 16S rDNA of the Acetobacter strains from protein profile phenon 3 (Table 9) and strains JK3 and JKD were sequenced. Except for strain HE (identical sequence to A. europaeus) the 16S rDNA sequences of these strains were identical to that of A. intermedius. Phenotypic characteristics Phenotypic characteristics differentiating A. intermedius- from A. europaeus- and A. xylinus-type strains are shown in Table 8. The A. intermedius type strain grows with or without acetic acid, survives glucose concentrations of 20% both on media with and without 3% acetic acid and 3 % ethanol, shows excellent growth on MRS medium and forms a cellulose pellicle on solid media and loose cellulose threads in liquid media which are sometimes suppressed. The A. europaeus type strain does not grow on GYG20%, on AEG20% or on MRS medium, does not form cellulose/acetan either in liquid or on solid media and requires acetic acid for growth. The A. xylinus type strain only grows on GYG20% and GY, cellulose formation is always present on solid medium and sometimes suppressed in liquid medium. Growth of A. intermedius and A. europaeus strains on AE, GY, AEG20%, GYG20%, selective MRS-medium for lactobacilli and on YP-media containing different carbon sources was tested. The growth of the cells on YP-media supplemented with different carbon sources gave no clear picture. Meaningful results on the use of carbon sources by the investigated strains would only be obtained in growth experiments on minimal media. As A. europaeus-strains generally need both acetic acid and an additional carbon source in their media, correct interpretation of results thus obtained would be tricky. The A. intermedius strains all grow on AE-medium, 85% of the investigated A. intermedius strains grow on GY-medium and all the A. intermedius strains tested on MRS-agar showed excellent growth on this medium. These results clearly position A. intermedius as a new species with phenotypic properties of both A. europaeus and A. xylinus: A. xylinus strains do not tolerate concentrations of 3 % acetic acid in the medium and the typical A. europaeus strain requires acetic acid for growth. Classification of the strains grouped according to protein profile homology and DNA reassociation values in Table 9 as belonging either to A. intermedius or A. europaeus according to phenotypic properties is not always clear as some strains like strain SW3 (clearly established as A. europaeus by its DNA-homology value and its rrna sequence) do exist that grow well on media without acetic acid. The A. intermedius type strain does not oxidize glucose to 5-ketogluconate or 2-ketogluconate. The
227
production of either 5-ketogluconate or 2-ketogluconate or both is a variable trait (KNEUBOHLER, 1994) and can therefore not be used for differentiation.
Discussion Isolation and cultivation of Acetobacter strains from generators and aceta tors is a difficult and slow procedure. A possibility for rapid identification of such strains would be by hybridization with a species-specific oligonucleotide probe. Development of an oligonucleotide probe for A. europaeus-strains on the basis of 16S rDNA sequences was not possible due to the high similarity with A. xylinus (99.5%, see Table 5) with only 7 varying nucleotides distributed over the length of the gene. Compared to the 16S rRNA primary structure, the 23S rRNA molecule exhibits an increased degree of sequence variability (STACKEBRANDT, 1988) and has served as an appropriate target for probes to differentiate between genetically similar organisms (RONNER and STACKEBRANDT, 1994). The alignment of the 23S rDNA sequences of the type strains of A. europaeus, A. xylinus and A. intermedius revealed equally high similarities (Table 6), however, with useful target sites for oligonucleotide probes. The oligonucleotide probe 23seu hybridized with all Acetobacter strains from the protein profile phenons 1, 2 and 4. It gave no hybridization signal with A. intermedius-strains, the A. xylinus type strain NCIB 11664 and the A. xylinus strain NCIB 613, however, hybridized with three strains which are classified as Acetobacter xylinus: De Ley 25, ATCC 23768 and ATCC 10245. These strains have previously been studied for their DNA-DNA homologies with A. europaeus and A. xylinus (BOESCH, 1994), all of them have DNA similiarities below 30% with both the A. europaeus and the A. xylinus type strain. These results plus the results obtained in the hybridization studies with the oligonucleotide probe 23seu show that A. europaeus, A. intermedius and A. xylinus belong to a branch of rapidly evolving Acetobacter strains. The oligonucleotide probe 23seu is not A. europaeus-specific but useful as a tool for the differentiation of A. intermedius from A. europaeus and A. xylinus. The overall 16S rRNA gene similarities between A. europaeus, A. xylinus and A. intermedius of more than 97% made DNA pairing studies necessary as an additional genotypic parameter to underline the species status of A. intermedius. According to WAYNE et al. (1987) a species would include strains with approximately 70% or greater DNA-DNA relatedness and which possess a common phenotypic characteristic. Previous DNA-DNA reassociation experiments with the A. europaeus type strain and all Acetobacter strains from the 4 phenons (SIEVERS and TEUBER, 1995, Kneubiihler, 1994 and unpublished results) had already resulted in the shifting of the strains of phenon 3 to subspecies level and grouping of the strains of phenon 4 in a possible new species. DNA-DNA homology below the 70% threshold value for species delineation as well as three phenotypic differ-
228
C. BOESCH et a!.
ences (growth with or without acetic acid, growth on media containing 20% glucose) clearly support the species status of A. intermedius. Further DNA-DNA hybridization experiments with A. intermedius and Acetobacter strains from vinegar fermenters with the A. intermedius type strain as a probe were performed more than once and produced the DNA-DNA homology values listed in Table 9. Outliers from the values are the strain TSA8 from phenon 4 which shows less than 30% DNA homology with A. europaeus (SIEVERS and TEUBER, 1995), hybridizes with the oligonucleotide probe 23seu and shows growth only on AE medium, yet has a DNAreassociation value of 99% with the A. intermedius type strain; this strain is therefore listed as Acetobacter sp. in Table 1 and Table 7. Further, the strain HE from ph en on 3 does not hybridize with the oligonucleotide probe 23seu, but its 16S rDNA sequence is identical to that of A. europaeus and its DNA reassociates at below 30% with the A. intermedius type strain and below 60% with the A. europaeus typestrain; this strain is listed as Acetobacter sp. in Table 1. Despite missing phenotypic features for some strains in the A. intermedius group and the mentioned exceptions, the 16S rDNA sequences and DNA-DNA reassociation experiments allow the division of all Acetobacter strains from our culture collection into either A. europaeus or A. intermedius, the exceptions being strain HE from phenon 3 and strains TSA8, TSN1 and TSN5 from phenon 4 (all Acetobacter sp.). Bacterial isolates derived from the tea fungus on YPM (mannitol, peptone, yeast extract)- and AE-medium were previously classified to belong to the genus Acetobacter (SIEVERS et aI., 1995). Based on the biosynthesis of cellulose/acetan, the results of the SDS-PAGE analysis and the growth on media without acetic acid, these isolates were tentatively identified as belonging to the species A. xylinus. The low clustering value of the A. xylinus type strain in the protein profile of 58% was attributed to the fact that this type strain is genotypically not thought to be representative for the species A. xylinus (unpublished results from our laboratory). Though these strains have been lost due to cultivation failure, the results based on the characterization of the newly isolated strain TF2 from the same Kombucha beverage indicate that the published strains (SIEVERS et aI., 1995) would belong to the newly described species A. intermedius. YAMADA et aI., 1997, have described the phylogenetic relationships of the acetic acid bacteria based on partial 16S rRNA gene sequences and propose the elevation of a subgenus Gluconoacetobacter to the generic level. It is generally accepted that only complete sequences allow reliable phylogenetic comparison with the available database of complete or almost complete sequences (STACKEBRANDT and GOEBEL, 1994). The construction of a phylogenetic tree based on partial base sequences must therefore be regarded as not valid. Also, one of the mentioned characteristics differentiating the 4 proposed genera, growth on mannitol agar only for the genera Gluconobacter and Gluconoacetobacter, is definitely wrong. In lists of bacterial cultures (Laboratorium Microbiologie Rijksuniversiteit Gent, BCCM Catalogue, 1989) YPM
agar is suggested as medium for Acetobacter and Gluconobacter strains. The main purpose of industrial strains is to produce high concentrations of acetic acid from ethanol. Industrial submerced fermenters are started by inoculation with 'seed vinegar' (microbiologically undefined remains from previous fermentations). Studies on plasmid profiles (SIEVERS and TEUBER, 1995) clearly show that the transfer of the vinegar microflora from one fermenter to another is possible which suggests that industrially used strains have over the last two hundred years and more been adapted to an extreme niche and thus subjected to high evolutionary pressure. The close phenotypic and genotypic relationship of the three species A. europaeus, A. intermedius and A. xylinus would indicate that they belong to a rapidly evolving branch of the acetic acid bacteria. Description of the new species Acetobacter intermedius (sp. nov.), L.adj.m. intermedius, in the middle between, to characterize the phenotype of this strain that has phenotypic properties of both Acetobacter europaeus and Acetobacter xylinus: growth in high concentrations of acetic acid and growth without acetic acid. Cells are Gram-negative, rod shaped, straight, 0.7 )lm by 2 )lm, occurring in pairs. Mobility and flagella not observed. Endospores are not formed. Obligately aerobic. Pale colonies, no pigments produced. Catalase positive. Oxidase negative. Oxidizes ethanol to acetic acid. Acetate is oxidized to CO 2 but only below an acetate concentration of 6%. No oxidation of glucose to 5-ketogluconate or 2-ketogluconate. DNA base composition is 61.55%. Acetobacter intermedius occurs in tea fungus beverages, spirit vinegar and cider vinegar. No significant DNA-DNA hybridization with the type strains of Acetobacter aceti, A. pasteurianus, A. liquefaciens, A. hansenii, A. xylinus, A. methanolicus, and Gluconobacter oxydans. DNA-DNA hybridization with the Acetobacter europaeus type strain of <60%. Type strain: Acetobacter intermedius TF2 (DSMl1804). Deposited at the Deutsche Sammlung von Mikroorganismen und Zellkulturen, Braunschweig. Isolated from a commercially available tea fungus beverage (Kombucha) in Switzerland.
Literature AMANN, R.L., BINDER, B.]., OLSON, R.]., CHISHOLM, S.W., DEVEREUX, R., STAHL, D.A.: Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. App!. Environ. Microbio!. 56, 1919-1925 (1990). AMANN, R.L., LUDWIG, W., SCHLEIFER, K.H.: Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59, 143-169, (1995). BOESCH, c.: Vergleich von spezifischen DNA-Sequenzen in der Taxonomie von Essigsaurebakterien. Diploma work, Institute of Food Science and Technology, ETH Zurich, Switzerland (1994).
Acetobacter intermedius sp. nov.
BROSIUS, J., DULL, T.]., SLEETER D.D., NOLLER, H.E: Gene organization and primary structure of a ribosomal RNA operon from Escherichia coli. J. Mol. Biol. 148, 107-127 (1981). DASEN, G., SMUTNY, J., TEUBER, M., MElLE, L.: Classification and identification of propionibacteria based on ribosomal RNA genes and PCR system. System. Appl. Microbiol. (in press). ENTANI, E., OHMORI, S., MASAI, H., SUZUKI, K.-I.: Acetobacter polyoxogenes sp. nov., a new species of an acetic acid bacterium useful for production of vinegar with high acidity. J. Gen. Appl. Microbiol. 31,475-490 (1985). EUZEBY, J.P.K: Revised nomenclature of specific or subspecific epithets that do not agree in gender with generic names that end in -bacter. Int. J. Syst. Bacteriol. 47, 585 (1997). GOESSELE, J., SWINGS, J., DE LEY, ].: A rapid, simple and simultaneous detection of 2-keto-, 5-keto- and 2,5-diketogluconic acid by thin layer chromatography in culture media of acetic acid bacteria. Zbl. Bakt. Hyg., I. Abt. Orig. Cl, 178-181 (1980). KERSTERS, K.: Numerical Methods in the classification of bacteria by protein electrophoresis. In: GOODFELLOW, M., JONES, D., PRIEST, EG. (Eds.), Computer-Assisted Bacterial Systematics (SCM Special Publication No. 15), Academic Press, London, 337-368 (1985). KNEUBDHLER, B.: Numerische Analyse von Proteinprofilen und DNA-DNA-Hybridisierung zur Charakterisierung und Differenzierung der Acetobacter-Flora industrieller Essigfermenter der Schweiz. Thesis no. 10674. ETH-Zurich, Switzerland (1994). RbNNER, S. and STACKEBRANDT, E.: Development of 23S rDNAoligonucleotide Probes for the Identification of Salmonella species. System. Appl. Microbiol. 17,257-264 (1994). SIEVERS, M., TEUBER, M.: The microbiology and taxonomy of Acetobacter europaeus in commercial vinegar production. ]. Appl. Bacteriol. Symposium Supplement 79, 84-95 (1995). SIEVERS, M., GABERTHDEL, c., BOESCH, c., LUDWIG, W., TEUBER, M.: Phylogenetic position of Gluconobacter species as a coherent cluster separated from all Acetobacter species on the basis of 16S ribosomal RNA sequences. FEMS Microbiol. Lett. 126,123-126(1995).
229
SIEVERS, M., LANINI, c., WEBER, A., SCHULER-SCHMID,U., TEUBER, M.: Microbiology and Fermentation Balance in a Kombucha Beverage Obtained from a Tea Fungus Fermentation. System. Appl. Microbiol. 18,590-594 (1995). SIEVERS, M., LUDWIG, W., and TEUBER, M.: Phylogenetic positioning of Acetobacter, Gluconobacter, Rhodopila and Acidiphilium species as a branch of acidophilic bacteria in the a-subclass of Proteobacteria based on 165 ribosomal DNA sequences. System. Appl. Microbiol. 17, 189-196 (1994). SIEVERS, M., SELLMER, S., TEUBER, M.: Acetobacter europaeus sp. nov., a main component of industrial vinegar fermenters in Central Europe. System. Appl. Microbiol. 15, 386-392 (1992). SOUTHERN, E.M.: Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. BioI. 98, 503-511 (1975). STACKEBRANDT, E., GOEBEL, B.M.: Taxonomic Note: A place for DNA-DNA reassociation and 16S rRNA sequence analysis in the present species definition in bacteriology. Int. J. Syst. Bacteriol. 44, 846-849 (1994). WAYNE. L.G., BRENNER, D.]., COLWELL, R.R., GRIMONT, P.A.D., KANDLER, 0., KRICHEVSKY, M.I., MOORE, L.H., MOORE, W.E.C., MURRAY, R.G.E., STACKEBRANDT, E., STARR, M.P. and TRDPER, H.G.: Report of the Ad Hoc Committee on Reconciliation of Approaches to Bacterial Systematics. Int. ]. Syst. Bacteriol. 37,463-464 (1987). YAMADA, Y., HOSHINO, K. and ISHIKAWA, T.: The Phylogeny of Acetic Acid Bacteria Based on the Partial Sequences of 16S Ribosomal RNA: The Elevation of the Subgenus Gluconoacetobacter to the Generic Level. Biosci. Biotech. Biochem. 61 (8), 1244-1251 (1997).
Corresponding author: M. TEUBER, Institut fur Lebensmittelwissenschaft, Labor fur Lebensmittelmikrobiologie, ETH-Zurich, Schmelbergstrasse 9, 8092 Zurich, Switzerland e-mail: [email protected]