Biosci. Biotechnol. Biochem., 75 (3), 419–426, 2011

Neokomagataea gen. nov., with Descriptions of Neokomagataea thailandica sp. nov. and Neokomagataea tanensis sp. nov., Osmotolerant Acetic Acid of the -

Pattaraporn YUKPHAN,1 Taweesak MALIMAS,1 Yuki MURAMATSU,2 Wanchern POTACHAROEN,1 y Somboon TANASUPAWAT,3 Yasuyoshi NAKAGAWA,2 Morakot TANTICHAROEN,1 and Yuzo YAMADA1;

1BIOTEC Culture Collection (BCC), National Center for Genetic Engineering and Biotechnology (BIOTEC), Pathumthani 12120, Thailand 2Biological Resource Center (NBRC), Department of Biotechnology, National Institute of Technology and Evaluation (NITE), Kisarazu 292-0818, Japan 3Department of Microbiology, Faculty of Pharmaceutical Sciences, Chulalongkorn University, Bangkok 10330, Thailand

Received May 7, 2010; Accepted November 29, 2010; Online Publication, March 7, 2011 [doi:10.1271/bbb.100344]

Isolates AH11T and AH13T were isolated from Of the 12 genera of acetic acid bacteria described to flowers of lantana and candle bush respectively collected date,1–5) the genus Saccharibacter Jojima et al. 20046) in Thailand. In phylogenetic trees based on 16S rRNA is unique from the osmophilic point of view and does gene sequences, the two isolates formed an independent not show any growth on 1.0% D-glucose w/v, differing cluster, which was then connected to the type strain of from other acetic acid bacteria. In addition, the genus Saccharibacter floricola. The calculated pair-wise 16S Saccharibacter is characterized by no oxidation of rRNA gene sequence similarities of isolate AH11T were acetate and weak oxidation of lactate, and no growth is 95.7–92.3% to the type strains of the type of the shown in the presence of 0.35% acetic acid w/v.6) 12 genera of acetic acid bacteria. The DNA base During the course of studies on microbial diversity in composition was from 51.2 to 56.8 mol % G+C, with a the natural environment of Thailand, two isolates were range of 5.6 mol %. When isolate AH11T was labeled, obtained, related phylogenetically to strains of Saccha- DNA-DNA similarities were 100, 12, 4, 5, and 4% ribacter floricola Jojima et al. 2004,6) but the two respectively to isolates AH11T and AH13T and the isolates were not osmophilic but osmotolerant, being type strains of Saccharibacter floricola, Gluconobacter different in this respect from strains of Saccharibacter oxydans, and Acetobacter aceti. The two isolates were floricola. non-motile and did not oxidize either acetate or lactate. This paper proposes Neokomagataea gen. nov., along No growth was found in the presence of 0.35% acetic with descriptions of Neokomagataea thailandica sp. acid w/v. The two isolates were not osmophilic but nov. and Neokomagataea tanensis sp. nov., for the two osmotolerant, produced 2,5-diketo-D-gluconate from isolates, which were isolated at Tan Island, Hat D-glucose, and did not oxidize lactate, thus differing Khanom-Mu Ko Thale Tai National Park, Nakhon-Si- from strains of Saccharibacter floricola, which showed Thammarat, Thailand on March 13, 2007. weak lactate oxidation. The two isolates contained unsaturated C18:1!7c fatty acid as the major fatty acid, Materials and Methods and were unique in the presence of a considerable amount of straight-chain C18:12OH fatty acid. Q-10 Isolation of acetic acid bacteria and reference strains. Two isolates, was present as the major isoprenoid quinone. Neo- designated AH11T and AH13T, were isolated by an enrichment culture T komagataea gen. nov. was proposed with the two species, approach: isolate AH11 was isolated from a flower of lantana Neokomagataea thailandica sp. nov. for isolate AH11T (Lantana camera; hedge flower; pha-ka-krong in Thai) by the use of a T T glucose/ethanol medium containing 2.0% D-glucose w/v, 0.5% (¼ BCC 25710 ¼ NBRC 106555 ), which has 56.8 ethanol w/v, 0.5% peptone w/v, and 0.3% yeast extract w/v, and mol % G+C, and Neokomagataea tanensis sp. nov. was adjusted to pH 3.5, and isolate AH13T was isolated from a flower T T for isolate AH13T (¼ BCC 25711 ¼ NBRC 106556 ), of candle bush (Senna alata; ringworm bush; chum-het-thet in Thai) by which has 51.2 mol % G+C. the use of a glucose medium containing 10.0% D-glucose w/v, 0.5% peptone w/v, and 0.3% yeast extract w/v, and was adjusted to 7,8) Key words: Neokomagataea gen. nov.; Neokomagataea pH 3.5. Neither of the enrichment culture media contained any acetic acid, and they differed from that of Yamada et al.9) Acetobacter thailandica sp. nov.; Neokomagataea aceti NBRC 14818T (NBRC, NITE Biological Resource Center, tanensis sp. nov.; acetic acid bacteria; Department of Biotechnology, National Institute of Technology and Evaluation, Kisarazu, Japan), Gluconobacter oxydans NBRC 14819T, Gluconobacter cerinus NBRC 3267T, Gluconobactr frateurii NBRC

The GenBank/EMBL/DDBJ accession numbers for the 16S rRNA gene sequences are AB513364 for Neokomagataea thailandica isolate AH11T and AB513363 for Neokomagataea tanensis isolate AH13T. y To whom correspondence should be addressed. JICA Senior Overseas Volunteer, Japan International Cooperation Agency (JICA), Shibuya-ku, Tokyo 151-8558, Japan; Professor Emeritus, Shizuoka University, Suruga-ku, Shizuoka 422-8529, Japan; Tel/Fax: +81-54-635-2316; E-mail: [email protected] 420 P. YUKPHAN et al. 3264T, Gluconobacter thailandicus BCC 14116T (BCC, BIOTEC bootstrap value of 66% (Fig. 1). Between the two Culture Collection, National Center for Genetic Engineering and isolates, the calculated bootstrap value was 100%. Biotechnology, National Science and Technology Development In a phylogenetic tree based on 16S rRNA gene Agency, Pathumthani, Thailand), Gluconobacter japonicus NBRC 3271T, Gluconobacter wancherniae BCC 15775T, Acidomonas meth- sequences constructed by the maximum parsimony 16) anolica NRIC 0498T (NRIC, NODAI Research Institute Center, Tokyo method, almost the same clustering was found in the University of Agriculture, Tokyo, Japan), Gluconacetobacter liquefa- isolates and the genera Saccharibacter and Glucono- ciens NBRC 12388T, Asaia bogorensis NBRC 16594T, bacter (Fig. 2). The calculated bootstrap values were 50, baliensis NBRC 16664T, Swaminathania salitolerans LMG 21291T 66, and 99% respectively. In a phylogenetic tree based (LMG, Laboratorium voor Microbiologie, Universiteit Gent, Ghent, T on 16S-rRNA gene sequences constructed by the Belgium), Saccharibacter floricola BCC 16445 , Neoasaia chiang- maximum likelihood method,17) similar small clusters maiensis BCC 15763T, Granulibacter bethesdensis ATCC BAA 1260T (ATCC, American Type Culture Collection, Rockville, MD, USA), and a similar large cluster were found, with bootstrap Tanticharoenia sakaeratensis BCC 15772T, and Ameyamea chiag- values respectively of 74, 80, and 100% (Fig. 3). maiensis BCC 15744T were used as reference strains. The calculated pair-wise sequence similarities of isolate AH11T were 94.2, 95.7, 95.3, 95.0, 95.0, 94.5, Sequencing of 16S rRNA genes and phylogenetic analyses. PCR 94.5, 94.1, 94.1, 93.9, 93.9, and 92.3% respectively to the amplification of 16S rRNA genes was performed, and amplified type strains of the type species of the respective genera, 16S rRNA genes were sequenced and analyzed as described previ- Saccharibacter floricola, Gluconobacter oxydans, Asaia ously.7,8,10–12) Multiple sequence alignment was performed with the program CLUSTAL X (version 1.8) (Thompson et al.13)). Alignment bogorensis, Tanticharoenia sakaeratensis, Ameyamaea gaps and unidentified bases were eliminated. Distance matrices for the chiangmaiensis, Acetobacter aceti, Swaminathania sal- aligned sequences were calculated by Kimura’s two-parameter itolerans, Neoasaia chiangmaiensis, Kozakia baliensis, method.14) Phylogenetic trees for 1,356 bases of 16S rRNA genes Acidomonas methanolica, Gluconacetobacter liquefa- were constructed by the neighbor-joining method,15) the maximum ciens, and Granulibacter bethesdensis. The sequence parsimony method,16) and the maximum likelihood method.17) Robust- 18) similarity was 99.3% between the two isolates. Since the ness for individual branches was estimated by bootstrapping with two isolates were closely related to Gluconobacter 1,000 replications by means of the program MEGA (version 4.0).19) Bootstrap values below 50% were not shown. To construct the species phylogenetically (Fig. 1), the pair-wise sequence phylogenetic tree by the maximum likelihood method,17) the program similarities of the isolates were additionally calculated PHYLIP (version 3.6; J. Felsenstein, University of Washington) was to representative strains of the genus Gluconobacter. used. Pair-wise sequence similarities were calculated in 16S rRNA Isolates AH11T and AH13T had 96.0–97.1% similarity gene sequences of 1,382 bases. A computerized 16S rRNA gene to the type strains of Gluconobacter albidus, Glucono- restriction analysis was made using NEBcutter (version 2.0; New bacter cerinus, Gluconobacter frateurii, Gluconobacter England BioLabs, Ipswich, MA, USA) for isolates AH11T and AH13T. The 16S rRNA gene PCR products of the isolates were prepared and thailandicus, Gluconobacter japonicus, and Glucono- digested with restriction endonucleases TaqI and BccI.5,8,10,11,20) bacter wancherniae. 16S rRNA gene restriction analysis using NEBcutter DNA base composition determination and DNA-DNA hybridization. (version 2.0) theoretically discriminated the two isolates Chromosomal DNA was prepared as described previously.4,5,7,8) The from the type strains of the type species of the 12 DNA base composition was determined by the method of Tamaoka and 21) genera of acetic acid bacteria using TaqI, and the two Komagata. DNA-DNA hybridization was carried out by the photo- isolates were distinguished from each other using BccI. biotin-labeling method using microplate wells, as described by Ezaki et al.22) Percent similarities in DNA-DNA hybridization were In TaqI digestion, the two isolates had the same determined colorimetrically.23) The color intensity was measured at restriction pattern, designated the Nkth type, comprising A450 on a model VersaMax microplate reader (Molecular Devices, 357, 350, 203, 112, 105, 104, 86, 83, and 43-bp Sunnyvale, CA, USA). Isolated, single-stranded, and labeled DNAs restriction fragments (data not shown). In BccI diges- were hybridized with DNAs from test strains in 2SSC containing tion, the two isolates were discriminated from each 50% formamide at 49.0 C for 3 h. The highest and lowest values other by different restriction patterns, designated the obtained in each sample were excluded, and the mean of the remaining T three values was taken as the similarity value. Nkth type in isolate AH11 , comprising 1,071 and 372- bp fragments, and designated the Nkta type in isolate Phenotypic and chemotaxonomic characterizations. Phenotypic AH13T, comprising 947, 372, and 124-bp fragments characteristics were determined by the methods of Asai et al.,24) (data not shown). Yamada et al.,9,25,26) Katsura et al.,27) Lisdiyanti et al.,28) Yukphan et al.,4,5,7,8) and Kersters et al.2) Cellular fatty acid composition was DNA base composition and DNA-DNA similarity of determined in cells grown on NBRC 804 agar at 30 C for 24 h. Methyl the two isolates esters of cellular fatty acids were prepared and identified following the instructions for the Microbial Identification System (MIDI, Hewlett The DNA base compositions were 56.8 mol % G+C T Packard, Palo Alto, CA, USA). A major isoprenoid quinone was for isolate AH11 and 51.2 mol % G+C for isolate estimated by the method of Yamada et al.29) AH13T, with a range of 5.6 mol %. The calculated DNA G+C contents of the two isolates were included in a Results group of lower G+C content acetic acid bacteria (Table 2). Phylogenetic characteristics of the two isolates A single-stranded and labeled DNA from isolate In a phylogenetic tree based on 16S rRNA gene AH11T represented 100, 12, 4, 5, and 4% DNA-DNA sequences constructed by the neighbor-joining method,15) similarity respectively to isolates AH11T and AH13T isolates AH11T and AH13T formed an independent and the type strains of Saccharibacter floricola, Gluco- cluster, which was then connected to the type strain of nobacter oxydans, and Acetobacter aceti, which was Saccharibacter floricola with a bootstrap value of 55%, used as a reference strain. When isolate AH13T and the and the resulting cluster produced a large cluster along type strain of Saccharibacter floricola were labeled, the with the cluster of the genus Gluconobacter with a calculated DNA-DNA similarities were 13, 100, 6, 6, Neokomagataea gen. nov. 421

T 65Gluconobacter albidus NBRC 3250 (AB178392) 96 Gluconobacter kondonii NBRC 3266T (AB178405) 84 Gluconobacter sphaericus NBRC 12467T (AB178431) 97 Gluconobacter kanchanaburiensis BCC 15889T (AB459530) Gluconobacter oxydans NBRC 14819T (X73820) 95 T 100 Gluconobacter roseus NBRC 3990 (AB178429) Gluconobacter cerinus NBRC 3267T (AB063286) Gluconobacter wancherniae isolate AC42T (AB511060) 82 66 Gluconobacter thailandicus F149-1T (AB128050) Gluconobacter frateurii NBRC 3264T (X82290) 94 T 85Gluconobacter japonicus NBRC 3271 (AB253435) 56 Saccharibacter floricola strain S-877T (AB110421) Neokomagataea tanensis isolate AH13T (AB513364) 55 T 100 Neokomagataea thailandica isolate AH11 (AB513363) Acetobacter aceti NBRC 14818T (X74066) T 93 Acetobacter pasteurianus LMD 22.1 (X71863) Tanticharoenia sakaeratensis strain AC37T (AB304087) T 91 Ameyamaea chiangmaiensis isolate AC04 (AB303366) T 93 Kozakia baliensis Yo-3 (AB056321) Neoasaia chiangmaiensis AC 28T (AB524503) T 85 Swaminathania salitolerans PA 51 (AF459454) Asaia lannensis AB92T (AB286050) 100 Asaia krungthepensis AA 08T (AB102953) 51 Asaia bogorensis strain 71T (AB025928) Asaia siamensis S60-1T (AB035416) T 97 Asaia spathodeae GB 23-2 (AB511277) T 68 Acidomonas methanolica LMG 1668 (X77468) Gluconacetobacter liquefaciens IFO 12388T (X75617) T 90 Gluconacetobacter xylinus NCIB 11664 (X75619) Granulibacter bethesdensis CGDNIH1T (AY788950) Acidocella facilis ATCC 35904T (D30774) Acidiphilium cryptum ATCC 33463T (D30773) 100 T 100 Acidiphilium multivorum strain AIU 301 (AB006711) Zymomonas mobilis ZM4 (NC_006526)

Knuc 0.02

Fig. 1. Phylogenetic Relationships of Neokomagataea thailandica Isolate AH11T and Neokomgataea tanensis Isolate AH13T. The phylogenetic tree based on 16S rRNA gene sequences was constructed by the neighbor-joining method. Zymomonas mobilis ZM4 was used as the outgroup. Numerals at nodes indicate bootstrap values (%) derived from 1,000 replications.

Gluconobacter albidus NBRC 3250T (AB178392) Gluconobacter kondonii NBRC 3266T (AB178405) 62 Gluconobacter sphaericus NBRC 12467T (AB178431) 91 Gluconobacter kanchanaburiensis BCC 15889T (AB459530) Gluconobacter oxydans NBRC 14819T (X73820) T 97 Gluconobacter roseus NBRC 3990 (AB178429) 72 Gluconobacter wancherniae isolate AC42T (AB511060) Gluconobacter cerinus NBRC 3267T (AB063286) Gluconobacter thailandicus F149-1T (AB128050) 66 Gluconobacter frateurii NBRC 3264T (X82290) Gluconobacter japonicus NBRC 3271T (AB253435) Saccharibacter floricola strain S-877T (AB110421) Neokomagataea tanensis isolate AH13T (AB513364) 50 T 99 Neokomagataea thailandica isolate AH11 (AB513363) Acetobacter aceti NBRC 14818T (X74066) T 78 Acetobacter pasteurianus LMD 22.1 (X71863) Tanticharoenia sakaeratensis strain AC37T (AB304087) T 73 Ameyamaea chiangmaiensis isolate AC 04 (AB303366) T 75 Kozakia baliensis Yo-3 (AB056321) Neoasaia chiangmaiensis AC 28T (AB524503) Swaminathania salitolerans PA 51T (AF459454) Asaia lannensis AB92T (AB286050) 90 Asaia bogorensis strain 71T (AB025928) Asaia krungthepensis AA 08T (AB102953) 53 Asaia siamensis S60-1T (AB035416) T 76 Asaia spathodeae GB 23-2 (AB511277) T 66 Acidomonas methanolica LMG 1668 (X77468) Gluconacetobacter liquefaciens IFO 12388T (X75617) Gluconacetobacter xylinus NCIB 11664T (X75619) Granulibacter bethesdensis CGDNIH1T (AY788950) Acidocella facilis ATCC 35904T (D30774) Acidiphilium cryptum ATCC 33463T (D30773) 88 T 100 Acidiphilium multivorum strain AIU 301 (AB006711) Zymomonas mobilis ZM4 (NC_006526) Nucleotide substitution

20

Fig. 2. Phylogenetic Relationships of Neokomagataea thailandica Isolate AH11T and Neokomgataea tanensis Isolate AH13T. The phylogenetic tree based on 16S rRNA gene sequences was constructed by the maximum parsimony method. Zymomonas mobilis ZM4 was used as the outgroup. Numerals at nodes indicate bootstrap values (%) derived from 1,000 replications. 422 P. YUKPHAN et al.

Gluconobacter kondonii NBRC 3266T (AB178405) Gluconobacter sphaericus NBRC 12467T (AB178431) Gluconobacter albidus NBRC 3250T (AB178392) 63 80 Gluconobacter kanchanaburiensis BCC 15889T (AB459530) T 52 Gluconobacter oxydans NBRC 14819 (X73820) T 96 Gluconobacter roseus NBRC 3990 (AB178429) Gluconobacter wancherniae isolate AC42T (AB511060) T 72Gluconobacter japonicus NBRC 3271 (AB253435) Gluconobacter frateurii NBRC 3264T (X82290) Gluconobacter thailandicus F149-1T (AB128050) Gluconobacter cerinus NBRC 3267T (AB063286) 80 Saccharibacter floricola strain S-877T (AB110421) T 74 Neokomagataea tanensis isolate AH13 (AB513364) 100 Neokomagataea thailandica isolate AH11T (AB513363) Acidomonas methanolica LMG 1668T (X77468) T 83 Kozakia baliensis Yo-3 (AB056321) Neoasaia chiangmaiensis AC 28T (AB524503) Swaminathania salitolerans PA 51T (AF459454) 53 Asaia lannensis AB92T (AB286050) 83 Asaia krungthepensis AA 08T (AB102953) Asaia bogorensis strain 71T (AB025928) Asaia spathodeae GB 23-2T (AB511277) T 78 Asaia siamensis S60-1 (AB035416) Tanticharoenia sakaeratensis strain AC37T (AB304087) T 57 76 Ameyamaea chiangmaiensis isolate AC04 (AB303366) Acetobacter aceti NBRC 14818T (X74066) T 77 Acetobacter pasteurianus LMD 22.1 (X71863) Gluconacetobacter liquefaciens IFO 12388T (X75617) 61 Gluconacetobacter xylinus NCIB 11664T (X75619) Granulibacter bethesdensis CGDNIH1T (AY788950) Acidocella facilis ATCC 35904T (D30774) 75 T 100 Acidiphilium cryptum ATCC 33463 (D30773) Acidiphilium multivorum strain AIU 301T (AB006711) Zymomonas mobilis ZM4 (NC_006526)

Knuc 0.02

Fig. 3. Phylogenetic Relationships of Neokomagataea thailandica Isolate AH11T and Neokomgataea tanensis Isolate AH13T. The phylogenetic tree based on 16S rRNA gene sequences was constructed by the maximum likelihood method. Zymomonas mobilis ZM4 was used as the outgroup. Numerals at nodes indicate bootstrap values (%) derived from 1,000 replications. and 7% and 6, 6, 100, 2, and 6% respectively. Since Discussion isolate AH11T had a calculated pair-wise 16S rRNA gene sequence similarity of 97.1% to the type strain of Phylogenetically, the two isolates, AH11T and Gluconobacter japonicus, DNA-DNA hybridization was AH13T, had calculated pair-wise sequence similarities made with representative strains of the genus Glucono- of 94.2 and 95.7% respectively to the type strains of bacter. Isolate AH11T showed DNA-DNA similarities Saccharibacter floricola, which was phylogenetically of 100, 8, 7, 5, 6, 7, and 7% respectively to isolate related, and Gluconobacter oxydans, the type species AH11T and the type strains of Gluconobacter cerinus, of the genus Gluconobacter (Fig. 1). Additionally, they Gluconobacter frateurii, Gluconobacter thailandicus, showed 97.1 and 97.0% sequence similarity to the type Gluconobacter japonicus, Gluconobacter wancherniae, strain of Gluconobacter japonicus, one of the most and Acetobacter aceti. The genetic data obtained closely phylogenetically related acetic acid bacteria. In indicated that the two isolates are to be classified in each of the phylogenetic trees, however, the branching separate species and are to be distinguished at the point between the lineage of the two isolates and the generic level. lineage of the type strain of Saccharibacter floricola was closer than the branching point between the two isolates Cellular fatty acid compositions of the two isolates and the type strains of the species of the genus Isolates AH11T and AH13T contained 48.9 and 50.2% Gluconobacter, despite the higher sequence similarities unsaturated fatty acid of C18:1!7c as the major fatty to the species of the genus Gluconobacter. The acid, 15.6 and 13.9% straight-chain fatty acid of C16:0, phylogenetic data obtained suggested that the two and 14.4 and 12.7% straight chain 2-hydroxy fatty acid isolates are distinguished at the generic level. 5) of C18:12OH (Table 1). The considerable amounts of In a previous study, the type strain of Saccharibacter C18:12OH acid distinguished the two isolates chemo- floricola was located outside, when a phylogenetic trees taxonomically from the type strains of Gluconobacter based on 16S rRNA gene sequences was constructed oxydans, Gluconobacter japonicus, and Saccharibacter using the type strain of Stella humosa as the outgroup. floricola. The two isolates were additionally distin- In contrast, it was located inside in the present study guished from the type strain of Saccharibacter floricola (Fig. 1). To confirm the difference of the phylogenetic by the absence of C19:0cyclo!8c acid. positions, namely, either the outside or the inside, a phylogenetic tree was newly constructed by the neighbor- Phenotypic characteristics of the two isolates joining method15) using the type strain of Stella humosa The phenotypic characteristics determined for isolates as the outgroup. In the resulting phylogenetic tree, the AH11T and AH13T are described in the genus and the type strain of Saccharibacter floricola was located species descriptions. outside as reported previously,5) and the cluster of the Neokomagataea gen. nov. 423 Table 1. Cellular Fatty Acid Compositions of the Two Isolates

Fatty acid 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Saturated fatty acid C14:0 0.8 0.6 0.4 2.1 2.0 1.3 0.6 4.9 1.7 1.0 1.3 0.7 0.4 1.2 C16:0 15.6 13.9 8.8 18.2 20.0 14.4 13.0 9.5 9.4 16.8 11.1 17.6 8.6 8.3 16.1 C17:0 0.4 0.8 0.3 1.1 C18:0 6.1 8.7 2.2 3.0 1.5 1.3 1.2 1.0 1.6 4.8 5.0 1.8 1.0 1.4 3.2 C19:0cyclo!8c 1.0 7.0 34.3 1.7 3.4 0.7 4.3 1.6 1.0 2.2 1.6 1.2 1.8 Hydroxy fatty acid C10:03OH 1.2 C14:02OH 0.4 5.5 9.8 4.6 4.6 5.5 3.6 2.1 2.2 1.5 4.2 C16:02OH 3.6 3.2 10.1 6.2 5.9 10.7 9.1 7.5 6.8 4.9 8.1 1.1 5.7 5.2 5.5 C16:03OH 2.9 2.4 2.6 2.6 2.1 3.1 3.1 3.0 2.3 1.6 2.5 2.1 1.9 2.1 1.7 C18:03OH 3.5 4.6 1.5 2.2 1.6 1.1 1.1 0.9 2.1 1.0 1.1 0.5 0.9 0.7 C18:12OH 14.4 12.7 1.4 0.7 0.4 2.7 0.7 1.9 1.2 10.2 4.8 8.3 Monounsaturated fatty acid C17:1!6c 0.6 0.6 1.5 C18:1!7c 48.9 50.2 69.6 57.1 30.7 53.2 52.6 66.0 61.1 58.7 64.1 57.4 68.0 65.8 61.8 Summed feature 2a 1.8 1.4 2.1 2.0 0.2 2.4 2.8 2.3 1.8 1.5 1.5 2.2 1.4 2.1 1.6

Abbreviations:1,Neokomagataea thailandica isolate AH11T;2,Neokomagataea tanensis isolate AH13T;3,Gluconobacter oxydans NBRC 14819T;4,Gluconobacter japonicus NBRC 3271T;5,Saccharibacter floricola BCC 16445T;6,Neoasaia chiangmaiensis BCC 15763T;7,Swaminathania salitolerans LMG 21291T;8,Asaia bogorensis NBRC 16594T;9,Gluconacetobacter liquefaciens NBRC 12388T; 10, Acetobacter aceti NBRC 14818T; 11, Tanticharoenia sakaeratensis BCC 15772T; 12, Granulibacter bethesdensis ATCC BAA 1260T; 13, Ameyamaea chiangmaiensis BCC 15744T; 14, Acidomonas methanolica NRIC 0498T; 15, Kozakia baliensis NBRC 16664T. a Comprised alde-C12:0 acid and/or an unknown acid. Unknown fatty acids below 0.5% were omitted. two isolates was located inside and was connected to the isolates were osmotolerant but not osmophilic and grew cluster of the type strains of Gluconobacter species with on 1.0% D-glucose w/v, but strains of Saccharibacter a bootstrap value of 45% (data not shown). No such floricola were osmophilic6) but not osmotolerent and did strange phenomena were seen, when the type strain of not grow on 1.0% D-glucose w/v (data not shown). Stella humosa was replaced by another strain3,6) or when Additionally, the former grew on glutamate agar con- a strain like the type strain of Acidiphilium cryptum was taining 0.5% or 1.0% L-glutamate w/v and did not used along with the type strain of Stella humosa as a oxidize lactate, but the latter did not grow on glutamate total of two outgroups.4) The cause of these problematic agar containing 1.0% L-glutamate w/v but grew on phenomena is obscure, but it is probable that an glutamate agar containing 7.0% L-glutamate w/v and unknown affinity exists, especially between the type had weak lactate oxidation.6) The former produced strains of Saccharibacter floricola and Stella humosa, 2,5-diketo-D-gluconate from D-glucose but the latter did among acetic acid bacteria. not (Table 2). The two isolates were especially unusual physiolog- In contrast to the genus Gluconobacter Asai 1935, the ically in showing no oxidation of acetate or lactate, two isolates did not grow in the presence of 0.35% acetic growth on 30% and 1.0% D-glucose w/v, no growth in acid w/v and did not produce dihydroxyacetone from the presence of 0.35% acetic acid w/v, and weak glycerol, but the strains assigned to the genus Glucono- production of acetic acid from ethanol (Table 2). bacter did. In acetic acid production from ethanol, weak Assimilation of ammoniac nitrogen on D-glucose, on production was found in the former, but a great deal of D-mannitol, and on ethanol was negative or very acetic acid was produced in the latter. As for acetate and weakly positive. 2-Keto-D-gluconate, 5-keto-D-gluco- lactate oxidation, no oxidation was found in the two nate, and 2,5-diketo-D-gluconate were produced from isolates or the strains of the genus Gluconobacter, but D-glucose, but no water-soluble brown pigment was the reactions were greatly different from each other.3,24) produced. The isolates had Q-10 as the major isopre- The former did not show any color change, but the latter noid quinone. showed a unique, clear, characteristic color change, Morphologically, isolates AH11T and AH13T were namely, from dark green to bright yellow.3,24) Especially, distinguished due to the absence of motility from the calculated DNA G+C contents were different from peritrichously and polarly flagellated, motile strains of each other. The two isolates showed 51.2 to 56.8 mol % the genera Acetobacter, Gluconobacter, Acidomonas, G+C, with a range of 5.6 mol %, but the type strains of Gluconacetobacter, Asaia,andSwaminathania (Table 2). the 11 Gluconobacter species showed 55.1 to 60.5 Physiologically, the absence of acetate and lactate mol % G+C, with a range of 5.4 mol %,30) and the range oxidation distinguished the two isolates from strains amounted to 9.2 mol % between isolate AH13T and the of Acetobacter, Acidomonas, Gluconacetobacter, Asaia, type strain of Gluconobacter oxydans, the type species Kozakia, Swaminathania, Saccharibacter, and Granuli- of the genus Gluconobacter (Table 2). bacter species. In addition, they were distinguished Chemotaxonomically, the two isolates were distin- from strains of the genera Gluconobacter and Neoasaia guished in cellular fatty acid composition from the type by the absence of growth in the presence of 0.35% strains of Saccharibacter floricola, Gluconobacter D-glucose w/v. oxydans, and Gluconobacter japonicus by the presence In comparison with the genus Saccharibacter Jojima of a considerable amount of a straight chain 2-hydroxy et al. 2004, which is phylogenetically related, the two fatty acid of C18:12OH. The former had 12.7–14.4% 424 P. YUKPHAN et al. Table 2. Characteristics Differentiating the Genus Neokomagataea of Acetic Acid Bacteria

Neokomagataea Characteristic Acetobacter Gluconobacter Acidomonas Gluconacetobacter Asaia Kozakia Swaminathania Saccharibacter Neoasaia Granuibacter Tanticharoenia Ameyamaea

1 2 345678 9a 10b 11c 12d 13e 14f Flagellation n n perm polm polh perm pern pernnnnpol Oxidation of Acetate þþþww w w þ Lactate þþww w wþw Growth on: 30% D-Glucose (w/v) þþn þh þw þþþnd þ 1% D-Glucose (w/v) þ þ þþþþþþ nd nd nd nd þ Glutamate agar þ þ þþ þ l þþww Mannitol agar þþvw þ w þþþ þ þþw þþ Raffinose þw þwndndþ nd Growth in the presence of 0.35% acetic acid (w/v) þþþþþ þ þnd þþ a 1% KNO3 (w/v) þ þ nd nd Production of acetic acid w w þþþþþ þ w/þ vw þþ from ethanol Water-soluble brown þ þ nd þ pigment production Production of dihydroxy- þþþww þw þw acetone from glycerol Production of levan-like þ nd nd polysaccharide Assimilation of ammoniac nitrogen on D-Glucose vw þw þþ nd þvw D-Mannitol þw þþ nd wnd vw Ethanol vw w w nd nd vw Production of 2-Keto-D-gluconate þþþþnd þþþ nd þþnd þþ 5-Keto-D-gluconate þþþþnd þþþ nd þþnd þþ 2,5-Diketo-D-gluconate þþn nd þo nd k nd þ Acid production from D-Mannitol þw þ þw D-Sorbitol þþ(d) þþ(d) Dulcitol w þ(d) v w Glycerol þþþþþþ þ þw/þ w Raffinose þ þþ nd þnd w Ethanol w þþþþþ þ þþþþ Major isoprenoid quinone Q-10 Q-10 Q-9 Q-10 Q-10 Q-10 Q-10 Q-10 Q-10 Q-10 Q-10 Q-10f Q-10 Q-10 DNA G+C (mol %) 56.8 51.2 58.6g 60.6g 62h 64.5g 60.2i 57.2j 57.6–59.9 52.3 63.1 59.1 65.6 66.0

Abbrevations: pol, polar; per, peritrichous; n, none; þ, positive; , negative; w, weakly positive; vw, very weakly positive; d, delayed; v, variable; nd, not determined; 1, Neokomagataea thailandica isolate AH11T;2,Neokomagataea tanensis isolate AH13T;3,Acetobacter aceti NBRC 14818T;4,Gluconobacter oxydans NBRC 14819T;5,Acidomonas methanolica NRIC 0498T;6,Gluconacetobacter liquefaciens NBRC 12388T;7,Asaia bogorensis NBRC 16594T;8,Kozakia baliensis NBRC 16664T;9,Swaminathania salitolerans strain PA51T; 10, Saccharibacter floricola strain S-877T; 11, Neoasaia chiangmaiensis strain AC28T; 12, Granulibacter bethesdensis CGDNIH1T; 13, Tanticharoenia sakaeratensis strain AC37T; 14, Ameyamaea chiangmaiensis strain AC04T. Cited from aLoganathan and Nair,31) bJojima et al.,6) cYukphan et al.,8) dGreenberg et al.,32) eYukphan et al.,4) f Yukphan et al.,5) gYamada et al.,33) hYamashita et al.,34) iYamada et al.,26) jLisdiyanti et al.,28) and kthe present study. lAccording to Jojima et al.,6) growth occurred at 7% glutamate w/v but not at 1% glutamate w/v. mSome strains in the genus are non-motile. nSome strains in the genus are positive. oSome strains in the genus are negative.

2-hydroxy acid, but the latter had only 0.4–1.4% thailandica sp. nov. for isolate AH11T and Neokoma- 2-hydroxy acid. gataea tanensis sp. nov. for isolate AH13T (Table 2). Isolates AH11T and AH13T, which are supposed to occupy an evolutionarily intermediate position between Description of Neokomagataea gen. nov. the genera Saccharibacter and Gluconobacter and to Neokomagataea (Ne.o.ko.ma.ga.ta’e.a. N. L. fem. n. link the two genera, are thus distinguished at the generic Neokomagataea after Dr. Kazuo Komagata, Professor level from the type strains of the type species of the 12 Emeritus, The University of Tokyo, Tokyo, Japan, who genera of acetic acid bacteria phylogenetically, genet- contributed to bacterial systematics and phylogeny, ically, morphologically, physiologically, and biochemi- especially of acetic acid bacteria). cally, and can be classified in a separate genus with two Gram-negative rods, non-motile, measuring 0.6–0:8 species. The name Neokomagataea gen. nov. is there- 1:0–1.6 mm. Colonies are creamy and smooth with entire fore proposed for the two isolates, with Neokomagataea margins, when grown on glucose/ethanol/peptone/yeast Neokomagataea gen. nov. 425 extract/calcium carbonate agar. Production of acetic acid dulcitol, meso-erythritol, glycerol, maltose, lactose, from ethanol is weakly positive. Does not oxidize acetate melibiose, or ethanol. 16S rRNA gene restriction analy- or lactate. Grows on glutamate agar or mannitol agar. sis shows the Nkth and Nkta patterns, when digested Growth on 30% D-glucose w/v is positive. Does not with TaqI and BccI respectively. The type strain is grow in the presence of 0.35% acetic acid w/v or in the isolate AH13T (¼ BCC 25711T ¼ NBRC 106556T). presence of 1.0% or 2.0% NaCl w/v, or 1.0% KCl w/v. The DNA base composition of the type strain is Does not grow on methanol. Ammoniac nitrogen is not 51.2 mol % G+C. The strain was isolated from a flower assimilated on D-glucose, D-mannitol, or ethanol as a of candle bush. source of carbon, but the assimilation is very weakly positive on D-glucose and ethanol in isolate AH11T. As described above, the DNA base composition of the Production of dihydroxyacetone from glycerol is neg- genus Neokomagataea was from 51.2 to 56.8 mol % ative. Produces 2-keto-D-gluconate, 5-keto-D-gluconate, G+C, with a range of 5.6 mol %, which corresponds to and 2,5-diketo-D-gluconate from D-glucose. No water- that of the genus Gluconobacter, whose range was soluble brown pigment is produced. Cellular fatty acids 5.4 mol % from 55.1 to 60.5 mol % G+C and of which 30) are composed of C18:1!7c, C16:0,C18:12OH, and C18:0 11 species have been described. This suggests that the acids, and so on. The major isoprenoid quinone is genus Neokomagataea is not small or narrow but large Q-10. The DNA base composition is 51.2–56.8 mol % or wide phylogenetically. Additional strains should be G+C, with a range of 5.6 mol %. The type species is isolated and many species should be reported in the Neokomagataea thailandica sp. nov. future.

Isolates AH11T and AH13T were genetically sepa- Acknowledgments rated from each other. The former had 56.8 mol % G+C, and the latter had 51.2 mol % G+C. Phenotypically, This study was supported in part by the Biodiversity the former did not produce acid from melibiose and Research and Training Program, Bangkok, Thailand. raffinose and did not grow on raffinose, but the latter did. References Description of Neokomagataea thailandica sp. nov. 1) Skerman VBD, McGowan V, and Sneath PHA, Int. J. Syst. Neokomagataea thailandica (tha.i.lan’di.ca. N. L. Bacteriol., 30, 225–420 (1980). fem. adj. thailandica of Thailand, where the type strain 2) Kersters K, Lisdiyanti P, Komagata K, and Swings J, ‘‘The was isolated). Prokaryotes’’ 3rd edn. Vol. 5, eds. Dworkin M, Falcow S, Characteristics are the same as those described for the Rosenberg E, Schleifer K-H, and Stackebrandt E, Springer, New genus. Acid is produced from D-glucose, D-galactose York, pp. 163–200 (2006). Int. J. Food Microbiol. 125 (weakly positive), D-xylose, L-arabinose (weakly posi- 3) Yamada Y and Yukphan P, , , 15–24 (2008). tive), D-fructose, or sucrose, but not from D-mannose, D- 4) Yukphan P, Malimas T, Muramatsu Y, Takahashi M, Kaneyasu arabinose, L-rhamnose, L-sorbose, D-mannitol, D-sorbitol, M, Tanasupawat S, Nakagawa Y, Suzuki K, Potacharoen W, and dulcitol, meso-erythritol, glycerol, maltose, lactose, Yamada Y, Biosci. Biotechnol. Biochem., 72, 672–676 (2008). melibiose, raffinose, or ethanol. Grows on D-glucose, 5) Yukphan P, Mallimas T, Muramatsu Y, Takahashi M, Kaneyasu L-rhamnose (weakly positive), or sucrose, but not on M, Potacharoen W, Tanasupawat S, Nakagawa Y, Hamana K, D-mannose, D-galactose, D-xylose, D-arabinose, L- Tahara Y, Suzuki K, Tanticharoen M, and Yamada Y, Biosci. Biotechnol. Biochem., 73, 2156–2162 (2009). arabinose, D-fructose, L-sorbose, D-mannitol, D-sorbitol, dulcitol, meso-erythritol, glycerol, maltose, lactose, 6) Jojima Y, Mihara Y, Suzuki S, Yokozeki K, Yamanaka S, and Fudou R, Int. J. Syst. Evol. Microbiol., 54, 2263–2267 (2004). melibiose, raffinose, or ethanol. 16S rRNA gene restric- 7) Yukphan P, Potacharoen W, Tanasupawat S, Tanticharoen M, tion analysis shows the Nkth and Nkth patterns, when and Yamada Y, Int. J. Syst. Evol. Microbiol., 54, 313–316 digested with TaqI and BccI respectively. The type strain (2004). is isolate AH11T (¼ BCC 25710T ¼ NBRC 106555T). 8) Yukphan P, Malimas T, Potacharoen W, Tanasupawat S, The DNA base composition of the type strain is Tanticharoen M, and Yamada Y, J. Gen. Appl. Microbiol., 51, 56.8 mol % G+C. The strain was isolated from a flower 301–311 (2005). 9) Yamada Y, Okada Y, and Kondo K, J. Gen. Appl. Microbiol., of lantana. 22, 237–245 (1976). 10) Yukphan P, Malimas T, Takahashi M, Potacharoen W, Busabun Description of Neokomagataea tanensis sp. nov. T, Tanasupawat S, Nakagawa Y, Tanticharoen M, and Yamada Neokomagataea tanensis (ta.nen’sis. N. L. fem. adj. Y, J. Gen. Appl. Microbiol., 50, 189–195 (2004). tanensis of or pertaining to Tan Island, Nakhon- 11) Yukphan P, Potacharoen W, Nakagawa Y, Tanticharoen M, and si-thammarat, Thailand, where the type strain was Yamada Y, J. Gen. Appl. Microbiol., 50, 9–15 (2004). isolated). 12) Takahashi M, Yukphan P, Yamada Y, Suzuki K, Sakane T, and Nakagawa Y, J. Gen. Appl. Microbiol., 52, 187–193 (2006). Characteristics are the same as those described for the 13) Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, and genus. Acid is produced from D-glucose, D-mannose Higgins DG, Nucleic Acids Res., 25, 4876–4882 (1997). (weakly positive), D-galactose, D-xylose, L-arabinose, 14) Kimura M, J. Mol. Evol., 16, 111–120 (1980). D-fructose, meso-erythritol (weakly positive), melibiose, 15) Saitou N and Nei M, Mol. Biol. Evol., 4, 406–425 (1987). sucrose, raffinose, or ethanol (weakly positive), but not 16) Felsenstein J, Annu. Rev. Ecol. Syst., 14, 313–333 (1983). 17) Felsenstein J, J. Mol. Evol., 17, 368–376 (1981). from D-arabinose, L-rhamnose, L-sorbose, D-mannitol, 18) Felsenstein J, Evolution, 39, 783–791 (1985). D-sorbitol, dulcitol, glycerol, maltose, or lactose. Grows 19) Tamura K, Dudley J, Nei M, and Kumar S, Mol. Biol. Evol., 24, on D-glucose, L-rhamnose, sucrose, or raffinose, but 1596–1599 (2007). not on D-mannose, D-galactose, D-xylose, D-arabinose, 20) Trcˇek J and Teuber M, FEMS Microbiol. Lett., 208, 69–75 L-arabinose, D-fructose, L-sorbose, D-mannitol, D-sorbitol, (2002). 426 P. YUKPHAN et al. 21) Tamaoka J and Komagata K, FEMS Microbiol. Lett., 25, 125– 28) Lisdiyanti P, Kawasaki H, Widyastuti Y, Saono S, Seki T, 128 (1984). Yamada Y, Uchimura T, and Komagata K, Int. J. Syst. Evol. 22) Ezaki T, Hashimoto Y, and Yabuuchi E, Int. J. Syst. Bacteriol., Microbiol., 52, 813–818 (2002). 39, 224–229 (1989). 29) Yamada Y, Aida K, and Uemura T, J. Gen. Appl. Microbiol., 23) Verlander CP, ‘‘Nonisotopic DNA Probe Techniques,’’ 15, 186–196 (1969). ed. Kricka LJ, Academic Press, New York, pp. 185–201 30) Yukphan P, Mallimas T, Lundaa T, Muramatsu Y, Takahashi (1992). M, Kaneyasu M, Tanasupawat S, Nakagawa Y, Suzuki K, 24) Asai T, Iizuka H, and Komagata K, J. Gen. Appl. Microbiol., 10, Tanticharoen M, and Yamada Y, J. Gen. Appl. Microbiol., 56, 95–126 (1964). 67–73 (2010). 25) Yamada Y, Hosono R, Lisdiyanti P, Widyastuti Y, Saono S, 31) Loganathan P and Nair S, Int. J. Syst. Evol. Microbiol., 54, Uchimura T, and Komagata K, J. Gen. Appl. Microbiol., 45, 23– 1185–1190 (2004). 28 (1999). 32) Greenberg DE, Porcella SF, Stock F, Wong A, Conville PS, 26) Yamada Y, Katsura K, Kawasaki H, Widyastuti Y, Saono S, Murray PR, Holland SM, and Zelazny AM, Int. J. Syst. Evol. Seki T, Uchimura T, and Komagata K, Int. J. Syst. Evol. Microbiol., 56, 2609–2616 (2006). Microbiol., 50, 823–829 (2000). 33) Yamada Y, Ishikawa T, Yamashita M, Tahara Y, Yamasato K, 27) Katsura K, Kawasaki H, Potacharoen W, Saono S, Seki T, and Kaneko T, J. Gen. Appl. Microbiol., 27, 465–475 (1981). Yamada Y, Uchimura T, and Komagata K, Int. J. Syst. Evol. 34) Yamashita S, Uchimura T, and Komagata K, Int. J. Syst. Evol. Microbiol., 51, 559–563 (2001). Microbiol., 54, 865–870 (2004).