International Journal of Systematic and Evolutionary Microbiology (2014), 64, 1340–1350

DOI 10.1099/ijs.0.058909-0

Reclassification of rhizosphere bacteria including strains causing corky root of lettuce and proposal of Rhizorhapis suberifaciens gen. nov., comb. nov., Sphingobium mellinum sp. nov., Sphingobium xanthum sp. nov. and Rhizorhabdus argentea gen. nov., sp. nov. Isolde M. Francis,1 Kenneth N. Jochimsen,23 Paul De Vos3 and Ariena H. C. van Bruggen24 Correspondence

1

Ariena H.C. van Bruggen

2

[email protected]

Department of Plant Pathology, University of Florida, Gainesville, FL 32611, USA Department of Plant Pathology, University of California, Davis, CA 95616, USA

3

Laboratory of Microbiology, Ghent University, 9000 Ghent, Belgium

The genus Rhizorhapis gen. nov. (to replace the illegitimate genus name Rhizomonas) is proposed for strains of Gram-negative bacteria causing corky root of lettuce, a widespread and important lettuce disease worldwide. Only one species of the genus Rhizomonas was described, Rhizomonas suberifaciens, which was subsequently reclassified as Sphingomonas suberifaciens based on 16S rRNA gene sequences and the presence of sphingoglycolipid in the cell envelope. However, the genus Sphingomonas is so diverse that further reclassification was deemed necessary. Twenty new Rhizorhapis gen. nov.- and Sphingomonas-like isolates were obtained from lettuce or sow thistle roots, or from soil using lettuce seedlings as bait. These and previously reported isolates were characterized in a polyphasic study including 16S rRNA gene sequencing, DNA–DNA hybridization, DNA G+C content, whole-cell fatty acid composition, morphology, substrate oxidation, temperature and pH sensitivity, and pathogenicity to lettuce. The isolates causing lettuce corky root belonged to the genera Rhizorhapis gen. nov., Sphingobium, Sphingopyxis and Rhizorhabdus gen. nov. More specifically, we propose to reclassify Rhizomonas suberifaciens as Rhizorhapis suberifaciens gen. nov., comb. nov. (type strain, CA1T5LMG 17323T5ATCC 49355T), and also propose the novel species Sphingobium xanthum sp. nov., Sphingobium mellinum sp. nov. and Rhizorhabdus argentea gen. nov., sp. nov. with the type strains NL9T (5LMG 12560T5ATCC 51296T), WI4T (5LMG 11032T5ATCC 51292T) and SP1T (5LMG 12581T5ATCC 51289T), respectively. Several strains isolated from lettuce roots belonged to the genus Sphingomonas, but none of them were pathogenic.

3Present address: Zuckermann Farms, Stockton, California 95201, USA. 4Present address: Department of Plant Pathology and the Emerging Pathogens Institute, University of Florida, Gainesville, FL 32611, USA. Abbreviations: DDH, DNA–DNA hybridization; FAME, fatty acid methyl ester; PHB, polyhydroxybutyrate. The GenBank/EMBL/DDBJ accession numbers for the 16S rRNA gene sequences of Sphingobium mellinum WI4T, Rhizorhapis suberifaciens CA1T, Rhizorhabdus argentea SP1T and Sphingobium xanthum NL9T are KF437546, KF437561, KF437572 and KF437579, respectively, and those of other strains determined in this study are listed in Table S1. Four supplementary figures and three supplementary tables are available with the online version of this paper.

1340

The genus Rhizomonas was created for strains of Gramnegative bacteria causing corky root of lettuce (van Bruggen et al., 1990), a widespread and important lettuce disease in various parts of the USA, Europe and Australia (Datnoff & Nagata, 1990; van Bruggen & Jochimsen, 1992, 1993; van Bruggen et al., 1988, 1989). Only one species of the genus Rhizomonas was described, Rhizomonas suberifaciens, containing most pathogenic strains isolated in the USA, except for strain WI4T which was presumed to belong to a separate species of the genus Rhizomonas (van Bruggen et al., 1990). A non-pathogenic strain isolated from lettuce roots in California, strain CA16, was shown to be related to strain WI4T and was presumed to belong to the same unnamed species (van Bruggen et al., 1993).

Downloaded from www.sgmjournals.org by 058909 G 2014 IUMS IP: 93.91.26.109 On: Sat, 25 Jul 2015 11:47:58

Printed in Great Britain

Reclassification of lettuce corky root bacteria

Yabuuchi et al. (1999) expanded the genus Sphingomonas by reclassifying Rhizomonas suberifaciens along with Blastomonas natatoria (Sly & Cahill, 1997) and Erythromonas ursincola (Yurkov et al., 1997) based on the results of a phylogenetic study of 16S rRNA gene sequences and the presence of sphingoglycolipid in their cell envelopes. They included phytopathogenic as well as photosynthetic organisms into a genus originally defined as a collection of Gram-negative, chemoheterotrophic, strictly aerobic, nonmotile or motile rods that produce yellow-pigmented colonies which are characterized by the presence of 2hydroxymyristic acid (C14 : 0 2-OH), the absence of 3hydroxy fatty acids, and ubiquinone 10 (Q10) as their major respiratory quinone (Balkwill et al., 2006; Yabuuchi et al., 1990). The expanded genus Sphingomonas also included the former Pseudomonas paucimobilis, the former Flavobacterium capsulatum, and three novel species, Sphingomonas parapaucimobilis, Sphingomonas adhaesiva and Sphingomonas yanoikuyae. Sphingomonas paucimobilis, the type species of the genus Sphingomonas (Holmes et al., 1977; Yabuuchi et al., 1990), was originally isolated as an opportunistic pathogen from various parts of the human body (Holmes et al., 1977), but also rhizosphere organisms have been described as belonging to this species (Lambert et al. 1990). After the expansion, the genus Sphingomonas became very diverse. Hiraishi et al. (2000) returned Sphingomonas natatoria to the genus Blastomonas after the identification of photosynthetic genes in Sphingomonas natatoria. Based on phenotypic, chemotaxonomic and phylogenetic analyses of Blastomonas natatoria, Sly & Cahill (1997) emended the description of this species as an aerobic photosynthetic bacterium. Hiraishi et al. (2000) also reclassified Erythromonas ursincola (Yurkov et al., 1997) as Blastomonas ursincola and concluded that both species of the genus Blastomonas were clearly distinct from species of the genus Sphingomonas. Despite the removal of Sphingomonas natatoria and Sphingomonas ursincola from the genus Sphingomonas, this genus remained genetically diverse. Therefore, a proposal was submitted by Takeuchi et al. (2001), based on data collected from phylogenetic, polyamine, fatty acid, and other biochemical analyses, to divide the genus Sphingomonas into at least four distinct genera within the family Sphingomonadaceae, including Sphingomonas and three new genera: Sphingobium, Novosphingobium and Sphingopyxis. This proposal was not accepted by Yabuuchi et al. (2002) who considered these new genera as junior objective synonyms of Sphingomonas. Still, the genera Erythrobacter, Porphyrobacter and Blastomonas, as well as the phytopathogenic genus Rhizomonas did not fit into any of these Sphingomonas subgroups but clustered as separate lineages (Takeuchi et al., 2001). This conclusion supported our previously published data (van Bruggen et al., 1990). Although Rhizomonas suberifaciens and Sphingomonas paucimobilis have many characteristics in common (Holmes et al., 1977; van Bruggen et al., 1990), the most http://ijs.sgmjournals.org

important one being a relatively large proportion of C14 : 0 2-OH fatty acid in the whole-cell fatty acid profile (Dees et al., 1979; Kawahara et al., 1991; van Bruggen et al., 1990; Yabuuchi et al., 1990), there are several important differences distinguishing these bacteria. These include differences in G+C content of chromosomal DNA (6 %), several fatty acids, colony colour (cream for Rhizomonas suberifaciens and yellow for Sphingomonas paucimobilis), sugar utilization pattern (wider for Sphingomonas paucimobilis), and maximal growth temperature (40 uC for Sphingomonas paucimobilis and 36 uC for Rhizomonas suberifaciens) (van Bruggen et al., 1990). Furthermore, DNA–rRNA hybridizations showed that Rhizomonas constitutes a separate rRNA branch within rRNA superfamily IV, equidistantly related to three other separate branches formed by Sphingomonas paucimobilis, Sphingomonas capsulata (currently known as Novosphingobium capsulatum) and Zymomonas, respectively (van Bruggen et al., 1993). The emendation of the genus Sphingomonas by Yabuuchi et al. (1999) to encompass Rhizomonas suberifaciens, Blastomonas natatoria and the former Erythrobacter ursincola was based on a comparison of these strains with only few Sphingomonas strains, namely Sphingomonas paucimobilis ATCC 29837T and Sphingomonas yanoikuyae LMG 11252T (currently known as Sphingobium yanoikuyae), and 10 other more distantly related bacterial species, mostly belonging to the order Rhizobiales. A more recent study by Chen et al. (2013) reclassified Sphingomonas suberifaciens as Sphingobium suberifaciens based on the comparative characterization with their newly reported species of the genus Sphingobium, Sphingobium limneticum and Sphingobium boeckii. However, the level of DNA–DNA hybridization between Sphingomonas suberifaciens and other Sphingobium strains was low, and no other rhizosphere strains closely related to Sphingomonas suberifaciens were included in the study. Additional Rhizomonas-like strains, isolated from US, European and Australian soils, have been reported (van Bruggen & Jochimsen, 1992, 1993; van Bruggen et al., 1990). Although several of these strains are pathogenic to lettuce, many are non-pathogenic. In addition, several of these strains form yellow colonies and/or a brown diffusible pigment unlike the previously described strains of Rhizomonas suberifaciens. Moreover, some of these strains have a low or moderate level of DNA–DNA relatedness with the type strain of Rhizomonas suberifaciens and tested negative with monoclonal antibody mAb-Rs1 which is specific for Rhizomonas suberifaciens (van Bruggen & Jochimsen, 1992, 1993; van Bruggen et al., 1990, 1992). Fatty acid analyses indicated, however, that these strains do belong to the genus that was named Rhizomonas or to closely related genera (van Bruggen & Jochimsen, 1992). This variation in characteristics of previously described rhizosphere strains (Figs S1 and S2, available in the online Supplementary Material) taken together with a collection of newly isolated Rhizomonas- and Sphingomonas-like

Downloaded from www.sgmjournals.org by IP: 93.91.26.109 On: Sat, 25 Jul 2015 11:47:58

1341

I. M. Francis and others

100 100 99 0.02

1342

Agromonas oligotrophica LMG 10732T (JQ619230) Agrobacterium tumefaciens ATCC 33970 (NR074266) Erythrobacter longus DSM 6997T (NR041889) Porphyrobacter tepidarius DSM 10594T (D84429) Zymomonas mobilis subsp. mobilis ATCC 10988T Sphingopyxis terrae LMG 17326T (NR029123) Sphingopyxis macrogoltabida LMG 17324T (NR043392) Sphingopyxis taejonensis DSM 15583T (NR024999) 97 Sphingopyxis alaskensis LMG 18877T (NR074280) 97 Sphingopyxis chilensis LMG 20986T (NR024631) Sphingopyxis sp. CA28 (KF437565) 100 Sphingopyxis sp. CA31 (KF437566) Sphingopyxis sp. CA32 (KF437567) 96 Sphingopyxis sp. CA30 (KF437574) Novosphingobium resinovorum LMG 8367T (EF029110) 88 Novosphingobium stygium LMG 18305T (NR040826) 75 Novosphingobium subterraneum LMG 18306T (NR040827) 78 Novosphingobium capsulatum LMG 2830T (NR025838) Novosphingobium rosa LMG 17328T (D13945) Blastomonas natatoria DSM 3183T (NR040824) Blastomonas ursincola DSM 9006T (Y10677) 100 Sphingobium boeckii LMG 26901T (JN591315) 100 Rhizorhapis sp. NL8 (KF437548) Rhizorhapis sp. GB2 (KF437557) Rhizorhapis suberifaciens WI2 (KF437549) Rhizorhapis suberifaciens WI3 (KF437550) Rhizorhapis suberifaciens NY11 (KF437554) 99 Rhizorhapis suberifaciens CA1T (KF437561) Rhizorhapis suberifaciens FL2 (KF437551) 100 Rhizorhapis suberifaciens FL11 (KF437568) Rhizorhapis suberifaciens FL14 (KF437569) Rhizorhapis suberifaciens FL20 (KF437577) Sphingomonas trueperi LMG 2142T (NR026338) 89 Sphingomonas roseiflava DSM 15593T (NR043426) Sphingomonas parapaucimobolis LMG 10923T (GU272291) 100 Sphingomonas sanguinis LMG 10925 (AB680528) Sphingomonas paucimobilis LMG 1227T (AM237364) 98 Sphingomonas sp. LMG 10924 (AB193907) 99 Sphingomonas aquatilis DSM 15581T (NR024997) Sphingomonas asaccharolytica DSM 10564T (NR029327) Sphingomonas pruni ATCC 51838T (NR026373) 83 98 Sphingomonas mali DSM 10565T (NR026374) Sphingomonas koreensis DSM 15582T (NR024998) 77 Sphingomonas sp. GB1 (KF437559) Sphingomonas sp. CA23 (KF437563) Sphingomonas sp. GR2 (KF437558) 96 Sphingomonas sp. CA25 (KF437564) 89 Sphingomonas sp. AU71 (KF437570) Sphingomonas starnbergensis DSM 25077T (JN591314) Sphingomonas wittichii DSM 6014T (NR074268) 100 Rhizorhabdus argentea SP1T (KF437572) Rhizorhabdus sp. NL2 (KF437556) Rhizorhabdus sp. SP3 (KF437560) 100 96 Rhizorhabdus sp. CA15 (KF437573) Rhizorhabdus sp. NY4 (KF437576) Sphingobium scionense DSM 19371T (EU009209) 94 Sphingobium yanoikuyae LMG 11252T (Y72725) Sphingobium sp. NL4 (KF437575) Sphingobium olei DSM 18999T (AM489507) 99 Sphingobium abikoense DSM 23268T (AB021416) Sphingobium rhizovicinum DSM 19845T (EF465534) Sphingobium xenophagum DSM 6383T (NR026304) Sphingobium sp. FL24 (KF437553) 100 78 Sphingobium sp. GR1 (KF437557) 96 Sphingobium japonicum DSM 16413T (AF039168) 100 Sphingobium francense MTCC 6363T (AY519130) 83 Sphingobium chinhatense MTCC 8598T (EF190507) 71 Sphingobium indicum DSM 16412T (AY519129) 90 Sphingobium chungbukense DSM 16638T (EU679660) Sphingobium chlorophenolicum ATCC 33790T (NR026249) Sphingobium quisquiliarum MTCC 9472T (EU781657) Sphingobium fuliginis DSM 18781T (DQ092757) Sphingobium herbicidovorans DSM 11019T (NR040807) Sphingobium sp. NL1 (KF437555) Sphingobium mellinum NL7 (KF437547) 75 Sphingobium mellinum WI4T (KF437546) 100 Sphingobium mellinum CA16 (KF437562) Sphingobium qiguonii DSM 21541T (EU095328) Sphingopyxis vermicomposti DSM 21299T (AM998824) Sphingobium sp. FL18 (KF437552) Sphingobium vulgare LMG 24321T (FJ177535) 98 (KF437578) 97 Sphingobium xanthum FL21 100 Sphingobium xanthum NL9T (KF437579) Sphingobium amiense DSM 16289T (AB047364) Sphingobium limneticum LMG 26659T (JN591313) 78

Downloaded from www.sgmjournals.org by International Journal of Systematic and Evolutionary Microbiology 64 IP: 93.91.26.109 On: Sat, 25 Jul 2015 11:47:58

Reclassification of lettuce corky root bacteria

Fig. 1. Phylogenetic tree based on the neighbour-joining algorithm BIONJ for 16S rRNA gene sequences. Bootstrap values (calculated with 1000 resamplings; Felsenstein, 1985) ¢70 % are shown at the branch points. Agrobacterium tumefaciens strain ATCC 33970T, Agromonas oligotrophica ATCC 43045T, and three other members of the order Sphingomonadales, Erythrobacter longus DSM 6997T, Porphyrobacter tepidarius DSM 10594T and Zymomonas mobilis subsp. mobilis ATCC 10988T were used as outgroup organisms. Bar, 0.02 substitutions per nucleotide position.

species isolated from lettuce roots, sow thistle roots, or soil using lettuce seedlings as bait (Table S1) encouraged us to re-examine the phylogenetic position of the debated genus Rhizomonas. For this purpose we carried out a polyphasic study including 16S rRNA gene sequencing, DNA–DNA hybridization (DDH), DNA G+C content analysis, wholecell fatty acid composition, phenotypic characterization, substrate oxidation profiles, temperature and pH sensitivity, and pathogenicity to lettuce. The name Rhizomonas was rejected by the Judicial Opinion 14 (De Vos & Tru¨per, 2000) because it was found to be a later homonym of a name of a taxon of protozoa (Rhizomonas W. S. Kent, 1880). Based on the Greek word for root, rhiza, and the Greek word for rod, rhapis, we propose to assign strains identified as belonging to the illegitimate genus Rhizomonas to a more appropriately named new genus, Rhizorhapis gen. nov. The isolation sources and locations of the strains used in this study are listed in Table S1. Several reference strains of the genus Sphingomonas, including Sphingomonas parapaucimobilis LMG 10923T and Sphingomonas paucimobilis LMG 1227T, as well as Sphingobium yanoikuyae LMG 11252T were obtained from the BCCM/LMG bacteria collection (Laboratory of Microbiology, Ugent, Belgium). Agromonas oligotrophica ATCC 43045T, obtained from the American Type Culture Collection (ATCC), served as a positive control for the oligotrophy test, and Escherichia coli DH1 (obtained from C. I. Kado) was used as a positive control for anaerobiosis. All cultures were grown on solid S medium (van Bruggen et al., 1988), the only medium on which Rhizomonas suberifaciens grows satisfactorily, for 2–4 days at 30 uC unless stated otherwise. Chromosomal DNA of selected strains was isolated using the MasterPure Gram Positive DNA Purification kit (Epicentre Biotechnologies) according to the manufacturer’s instructions. Although the strains studied are Gramnegative, it is more difficult to isolate DNA with Gramnegative purification kits. The universal primers 27f (59AGAGTTTGATCCTGGCTCAG-39) and 1492r (59-TACGGTTACCTTGTTACGACTT-39) were used to amplify nearly full-length 16S rRNA gene sequences. Amplified DNA was purified from agarose gel (E.Z.N.A. Gel Extraction kit, Omega Bio-Tek), cloned into the pGEM-T Easy Vector (Promega) and sequenced by the Sanger method with commercial primers hybridizing to the T7 and SP6 promoter. Phylogenetic and molecular evolutionary analyses were conducted using SeaView version 4 (Gouy et al., 2010). Via this interface the obtained 16S rRNA gene sequences were aligned with sequences available from the GenBank http://ijs.sgmjournals.org

database, and phylogenetic trees were reconstructed computed by distance using the neighbour-joining algorithm BIONJ (Gascuel, 1997) (Fig. 1) and by the maximumlikelihood method using PhyML 3.0 (Guindon et al., 2010) (data not shown). All phylogenetic trees were calculated with 1000 resamplings (bootstrap analysis; Felsenstein, 1985). Agrobacterium tumefaciens ATCC 33970T, Agromonas oligotrophica ATCC 43045T, and three other members of the order Sphingomonadales, Erythrobacter longus LMG 6997T, Porphyrobacter tepidarius DSM 10594T and Zymomonas mobilis subsp. mobilis ATCC 10988T were used as outgroup organisms (Fig. 1). DDH of selected strains was performed as described previously by van Bruggen et al. (1993). The extent of hybridization was determined under moderately stringent (50 uC) conditions and by measuring counts min21 on the blots using the Ambis Radioanalytic Imaging System (Ambis Systems) (Table S2). The results of genetic analyses are shown in Fig. 1 and Table S2. A separate lineage of strains was formed clustering together with Rhizorhapis suberifaciens gen. nov., comb. nov. CA1T. The recommended cut-off values for species discrimination are 98.7 % for 16S rRNA gene sequence similarity and 70 % for DDH (Stackebrandt & Ebers, 2006). Even though several of the strains had high 16S rRNA gene sequence similarity values (100–99 %) to Rhizorhapis suberifaciens gen. nov., comb. nov. CA1T, they showed DDH values below 70 % (Table S2). This could indicate that they belong to different species within the genus (Gevers et al., 2005; Stackebrandt & Ebers, 2006) or point to differences in gene content reflecting strainspecific genes within a bacterial species. Variation in gene content is greatly enhanced by lateral gene transfer, a prevalent phenomenon among members of the phylum Proteobacteria (Kloesges et al., 2011). In particular strains NL6, NL8 and GB2 showed an exceptionally low DDH value (,21 %) with Rhizorhapis suberifaciens gen. nov., comb. nov. CA1T (Table S2). This distinction from Rhizorhapis suberifaciens gen. nov., comb. nov. is also reflected in the phylogenetic tree where they are positioned on a separate branch within the Rhizorhapis gen. nov. cluster (Fig. 1). We therefore propose to classify them for now as Rhizorhapis gen. nov. spp. NL6, NL8 and GB2. Nevertheless, strains NL6 and NL8 reacted positively with monoclonal antibody mAb-Rs1 (van Bruggen & Jochimsen, 1992), which is specific for strains of Rhizomonas suberifaciens (van Bruggen et al., 1992). As seen in the phylogenetic tree, strains SP1T, SP3, NL2, NY4 and CA15 clustered together with Sphingomonas wittichii and Sphingomonas starnbergensis and form a branch that is closely related but clearly distinct from the

Downloaded from www.sgmjournals.org by IP: 93.91.26.109 On: Sat, 25 Jul 2015 11:47:58

1343

I. M. Francis and others

main Sphingomonas cluster. We therefore propose a new genus, named Rhizorhabdus gen. nov., for our isolated strains, and a re-evaluation of the other strains that seem to be closely related to our Rhizorhabdus gen. nov. strains. The separation of the Rhizorhabdus gen. nov. strains from the Sphingomonas cluster is strengthened by the low DDH values between these strains and representative strains of other closely related genera, with maxima of only 11.2 % and 10.7 % to Sphingomonas paucimobilis LMG 1227T and Sphingobium xanthum sp. nov. NL9T, respectively (Table S2). Considering the low DDH values between strain SP1T and the rest of the Rhizorhabdus gen. nov. strains (Table S2), it is likely that the latter ones belong to a different species within this genus. With the addition of the strains reported in this study, the recently identified Sphingobium boeckii 469T (Chen et al., 2013) positioned itself more closely to Rhizorhapis gen. nov. and was clearly distinct from the rest of the species of the genus Sphingobium (Fig. 1). The new information provided here might shed more light on the position of these previously identified species within the family Sphingomonadaceae. Several other strains recently isolated from lettuce roots clustered together with species belonging to the genera Sphingopyxis, Sphingobium and Sphingomonas, which clearly formed distinct bacterial lineages (Fig. 1, Table S2), supporting the subdivision of the genus Sphingomonas (Takeuchi et al., 2001). Strains CA28 to CA32 showed a 99 % 16S rRNA gene sequence similarity with Sphingopyxis chilensis S37T, while strains GR1, GR3 and FL24 had 98–99 % 16S rRNA gene sequence similarity to Sphingobium xenophagum BN6T (data not shown). Further DDH studies will have to confirm the position of these strains within the respective species mentioned. In addition, several of the newly isolated strains clustered together forming two separate novel species within the genus Sphingobium (Fig. 1, Table S2). We propose the novel species Sphingobium xanthum sp. nov., with bright yellow colonies, formed by strains NL9T, CA20, FL21 and FL22, and Sphingobium mellinum sp. nov., with honey-coloured colonies, containing strains NL1, NL5 and NL7 together with WI4T and CA16, previously reported as very similar but clearly distinct from Rhizomonas suberifaciens (van Bruggen et al., 1990). Strains FL18 and NL1 seemed to be closely related to strains of Sphingobium xanthum sp. nov. and Sphingobium mellinum sp. nov., respectively (Fig. 1), but their low DDH values indicate that they probably belong to a novel yet unidentified species (Table S2). The same is true for strain NL4 which seems to be most similar to strains of Sphingobium yanoikuyae but still distinct enough to be classified as a representative of a separate species. Strains AU71, GB1, CA23, CA25 and GR2 grouped in the genus Sphingomonas, possibly forming one novel species (Fig. 1), which will have to be confirmed by DDH. Their 16S rRNA gene sequences showed high similarity to each other (98–99 %) while these values dropped to 97 % and lower when compared to other species of the genus Sphingomonas (Table S2). The DNA G+C content of strains representing newly characterized species or strains clustering together to form 1344

a new genus was determined. Thermal denaturation was performed as described by De Ley & Van Muylem (1963), and the mol% G+C content of chromosomal DNA was calculated using the equation of Marmur & Doty (1962) as modified by De Ley (1970). A mean of 64.0±0.2 mol% was measured for strains CA15 and NL2 belonging to the new genus Rhizorhabdus gen. nov. The DNA G+C contents of strains NL9T and FL22 representing the novel species Sphingobium xanthum sp. nov. had a mean of 64.7 mol%, and the mean value of Sphingobium mellinum sp. nov. strains NL7 and WI4T was 63.1±0.2 mol%. In this study several new strains belonging to the genus Sphingomonas were isolated. The DNA G+C contents of two of those strains, namely CA22 and GB1, were 65.8 and 66.9 mol%, respectively. These values together with the published values for strains of Sphingomonas paucimobilis and Sphingomonas parapaucimobilis, averaging 65.73±0.07 mol% and 64.5±0.5 mol%, respectively (Yabuuchi et al., 1990), and those of the newly reported Sphingobium limneticum (63.4 mol%) and Sphingobium boeckii (64.6 mol%) (Chen et al., 2013) show a clear discrepancy with the significantly lower mean value of 58.9±0.3 mol% previously reported for strains of Rhizomonas suberifaciens (van Bruggen et al., 1990), justifying the removal of Rhizorhapis gen. nov. from the Sphingomonas group. Fatty acids were extracted from strains in exponential phase, grown on solid S medium with 5 g l21 casein enzymic hydrolysate (N-Z-Amine AS, Sigma-Aldrich), for 3 days at 28±1 uC. The extracted fatty acids were methylated using the procedure of Lambert et al. (1983). Fatty acid methyl ester (FAME) fractions were separated on a 5 % methyl phenyl silicone capillary column of a gas chromatograph (Hewlett Packard) equipped with a flameionization detector. Standards containing known fatty acids were used for calibration and the peaks were identified with the Sherlock Microbial Identification System (MIS) software package (MIDI). The fatty acid profiles were subjected to cluster analysis using the same software. Fatty acid profiles of most strains consisted mainly of unsaturated and saturated straight-chain fatty acids with even numbers of carbon atoms (C18 : 1, C16 : 1 and C16 : 0), 2-hydroxy fatty acids (C14 : 0 2-OH, C15 : 0 2-OH and C16 : 0 2-OH), and one methylated fatty acid (C18 : 1 10CH3) (Table 1). The fatty acid profiles were separated into 11 clusters, which coincided quite well with the clusters based on 16S rRNA gene sequences. Yet, there were some notable discrepancies, in particular for strain FL18 which ended up in a separate FAME cluster, while it belongs to Sphingobium sp. according to the 16S rRNA gene sequence analysis (FAME cluster 11). Other strains could potentially be identified based on characteristic FAME compositions. Members of the genus Sphingopyxis (FAME cluster 1) are clearly distinct from the rest because of a lower amount of C18 : 1 fatty acids grouped in summed feature 7 (SUM7) (11.7±1.2 %) and higher amounts of C17 : 1v6c and C17 : 1v8c (23.8±1.9 % and 5.1±0.5 %, respectively). Sphingopyxis strains also contained C15 : 0 2-OH as the

Downloaded from www.sgmjournals.org by International Journal of Systematic and Evolutionary Microbiology 64 IP: 93.91.26.109 On: Sat, 25 Jul 2015 11:47:58

Reclassification of lettuce corky root bacteria

Table 1. Cellular fatty acid composition of the studied strains compared to related species Strains forming the FAME clusters: 1, Sphingopyxis sp. CA28, CA29, CA30, CA31 and CA32; 2, Sphingobium sp. FL24, GR1 and GR3; 3, Sphingobium sp. NL4 and Rhizorhabdus gen. nov. sp. SP1T, SP3, NL2, CA15 and NY4; 4, Sphingomonas paucimobilis LMG 1227T, Sphingomonas parapaucimobilis LMG 10923T, Sphingomonas sp. LMG 10924 and Sphingomonas sp. LMG 10925; 5, Sphingobium sp. FL18; 6, Sphingobium sp. NL1, Sphingobium mellinum sp. nov. WI4T, NL5, NL7, CA16 and CA33; 7, strains of Rhizorhapis suberifaciens gen. nov. comb. nov. and Rhizorhapis gen. nov. sp. listed in Table S1; 8, Sphingobium yanoikuyae LMG 11252T; 9, Sphingomonas sp. CA23; 10, Sphingomonas sp. AU71, GR2, GB1, CA22 and CA25; 11, Sphingobium xanthum sp. nov. NL9T, CA20, FL21 and FL22. Values shown are percentages of the total fatty acids. TR, Trace (,1.0 %); 2, not detected; SUM7, summed feature 7 representing fatty acids C18 : 1v7c, C18 : 1v9c and C18 : 1v12t that cannot be separated by the Microbial Identification System. Fatty acid C14 : 0 C14 : 0 2-OH C15 : 0 C15 : 0 2-OH C16 : 0 C16 : 0 2-OH iso-C16 : 0 3-OH C16 : 1v7c C16 : 1v5c C17 : 0 C17 : 1v8c C17 : 1v6c C18 : 1 2-OH C18 : 1v5c SUM7 C18 : 1 10-CH3 C19 : 0 cyclo

1 2 3.9±0.5 3.3±0.2 19.2±0.7 2.8±0.1 1.2±0.2 2 11.2±0.8 TR

1.0±0.1 5.1±0.5 23.8±1.9 2 2 11.7±1.2

2 TR

6.7±0.6 TR TR

6.7±1.0 2.0±0.5 1.2±0.2 28.7±1.3 TR

2 2 1.4±1.3 2 1.0±0.9 41.1±0.8

3

4

5

6

7

2.1±1.2 14.6±9.4

TR

9.3±3.0

2 9.7

1.0±0.7 7.7±1.6

1.1±0.4 9.3±0.9

TR

2 14.8±4.1 TR

2 14.3±4.1 2.2±0.9 TR TR TR TR TR

43.9±2.7

TR

TR

TR

TR

2 8.0±2.6 2 2 5.0±2.2

TR

TR

TR

TR TR

2 1.4±1.2 2 2.6±1.1 70.2±2.5

7.9 2 2 7.9 1.3 2

17.6±1.6

TR

2 1.4±0.4 2 2.5±0.4 53.4±2.9

3.3 2 2 66.0

TR

2 6.9±1.2 TR

TR

TR

TR

TR

TR

2

TR

2.3±1.3

TR

2

1.4±0.5

http://ijs.sgmjournals.org

TR TR

11.6±1.8 1.3±0.4

TR

TR

main hydroxylated fatty acid (19.2±0.7 %) in contrast to the rest which are characterized predominantly by C14 : 0 2OH. Whole-cell fatty acid profiles for FAME clusters 2 and 3, harbouring respectively Sphingobium strains FL24, GR1 and GR2, and Sphingobium sp. NL4 and the Rhizorhabdus gen. nov. strains, showed slightly lower amounts of SUM7 C18 : 1 fatty acids (41.1±0.8 % and 43.9±2.7 %, respectively), but can be distinguished by the higher amount of C16 : 1v7c fatty acids for FAME cluster 2 (28.7±1.3 %) than cluster 3 (14.3±4.1 %). FAME cluster 4 containing strains of Sphingomonas paucimobilis and Sphingomonas parapaucimobilis is characterized by a high amount of SUM7 C18 : 1 fatty acids (70.0±2.5 %), a trait that is shared by Sphingobium sp. FL18, the only member of cluster 5 (66 %). Strains in the Sphingobium mellinum sp. nov. cluster (FAME cluster 6), the Rhizorhapis suberifaciens gen. nov., comb. nov. cluster (FAME cluster 7), and Sphingobium yanoikuyae LMG 11252T (FAME cluster 8) seemed to be closely related according to their fatty acid profile. This is in agreement with the proposed classification of Chen et al. (2013) that strains representing the genus Rhizorhapis gen. nov. would belong to the genus Sphingobium based on their fatty acid analysis. However, their 16S rRNA gene sequence analysis results in a separate lineage for Rhizorhapis suberifaciens gen. nov., comb. nov. and Sphingobium boeckii, distinct from the other Sphingobium strains. Moreover, the FAME profiles of Rhizorhapis gen. nov. strains differed from

10.3±1.6

TR TR

1.3±1.7 TR

1.6±0.4 57.9±2.3 3.5±2.9 1.3±0.7

8 TR

9.5±1.1 2 TR

9.9±1.2 1.0±0.2 TR

12.4±2.7 1.4±0.2 2 2 1.9±0.4 2 2.6±0.3 53.7±2.7

9 2 31.3 2 2 6.9 2 2 2.6 2 2 2 2 2 2 54.5

10

11

TR 2 14.9±3.3 19.0±1.0 TR 2 1.3±0.8 3.6±1.9 11.1±2.3 8.9±2.6 1.1±0.3 2 TR 2 1.5±0.2 8.5±2.5 TR

TR

TR

2

TR

TR

2.9±0.8

7.2±2.0 2 1.4±1.0 2 59.6±1.9 52.4±2.5 TR

TR

TR

TR

TR

1.3±0.3

2.0

1.3±0.6

2

those of the Sphingobium mellinum sp. nov. and Sphingobium xanthum sp. nov. clusters (FAME clusters 6 and 8) by having a lower amount of C18 : 1v5c (1.6±0.4 % compared to 2.5±0.4 % and 2.6±0.3 %, respectively). Strains of the Sphingobium mellinum sp. nov. cluster differed from those of the Rhizorhapis gen. nov. cluster by having a lower percentage of C16 : 1v7c (6.9±1.2 % compared to 11.6± 1.8 %) and C14 : 0 2-OH (7.7±1.6 % compared to 9.3± 0.9 %), and a C16 : 1v5c fatty acid content below the detection limit. Sphingomonas sp. CA23 (FAME cluster 9) distinguishes itself from other Sphingomonas strains with 31.3 % of C14 : 0 2-OH fatty acid, which is double the amount measured for its phylogenetically related strains AU71, GR2, GB1, CA22 and CA25 (14.9±3.3 %) (FAME cluster 10) presumably forming a novel species of the genus Sphingomonas. General shape and cell size were determined for a representative subset of the strains studied by means of a phase-contrast microscope and an ocular micrometer. Cells were rod-shaped often occurring in strings or clumps of several cells (Fig. S3) (except for Sphingomonas strains, Sphingobium yanoikuyae LMG 11252T and Sphingobium sp. NL4 which were not clumped). Cells of Sphingobium sp. NL4 and Sphingobium yanoikuyae LMG 11252T were relatively large (2.060.9 mm), while those of Sphingobium xanthum sp. nov. were slightly smaller (1.560.8 mm) and cells of Sphingobium mellinum sp. nov. were very small (0.960.5 mm). Cells of Rhizorhabdus gen. nov. were similar

Downloaded from www.sgmjournals.org by IP: 93.91.26.109 On: Sat, 25 Jul 2015 11:47:58

1345

I. M. Francis and others

to cells of Sphingobium xanthum sp. nov., measuring approximately 1.560.8 mm. Cells of strains belonging to the Rhizorhapis gen. nov. cluster were slightly smaller than cells of Rhizorhabdus gen. nov., and were all similar in size to the type strain Rhizorhapis suberifaciens gen. nov., comb. nov. CA1T (1.460.4 mm). None of the strains, except for Escherichia coli DH1, grew on S medium in anaerobiosis jars (as described previously, van Bruggen et al., 1990). Nevertheless, most strains survived as they started to grow when relieved from the anaerobic conditions. To test for oligotrophy, colonies of a representative subset of the studied strains were scraped from S medium plates and diluted in sterile distilled water to a density of approximately 108 c.f.u. ml21. Fifty microlitres suspension was added to 4 ml trypticase soy broth (TSB; BBL Microbiology Systems), 10- and 100-fold diluted TSB, and S medium broth. After 2 weeks of incubation at 30 uC the optical density was measured at 650 nm and converted to c.f.u. ml21 using standard curves for three classes of bacterial sizes. Agromonas oligotrophica ATCC 43045T was included as a positive control. All strains tested grew to a concentration of 2.36109–2.861010 c.f.u. ml21 in S medium broth. Strains belonging to the studied species of the genus Sphingobium and Sphingobium yanoikuyae LMG 11252T were non-oligotrophic and grew to a concentration of 6.96108–1010 c.f.u. ml21 on TSB, 6.96108– 3.56109 c.f.u. ml21 on 10-fold diluted TSB, and attained a density of 2.86108–6.96108 c.f.u. ml21 on 100-fold diluted TSB. In contrast, strains representing Rhizorhapis gen. nov., Rhizorhabdus gen. nov., as well as Sphingobium sp. NL4 resembled Agromonas oligotrophica ATCC 43045T as they hardly grew on TSB (0.0–26107 c.f.u. ml21), but reached 46107–1.46109 c.f.u. ml21 and 86107–2.46108 c.f.u. ml21 on 10- and 100-fold diluted TSB, respectively. Carbon oxidation patterns for the strains listed in Table S1 were determined by means of the Biolog system using the Biolog GN MicroPlates. Absorbance values were subjected to cluster analyses using the Statistical Analysis System (SAS Institute), which resulted in 12 distinct clusters which closely resembled the clusters formed by 16S rRNA gene sequence analysis (Fig. 1, Table S2). Typical oxidation reactions for the main clusters of the 16S rRNA gene sequence analysis are given in Table S3. Strains of Rhizorhabdus gen. nov., Rhizorhapis gen. nov., Sphingobium mellinum sp. nov. and Sphingobium sp. NL1 (Biolog clusters 1–5) oxidized relatively few substrates, while Sphingobium sp. FL18, Sphingobium sp. NL4, Sphingobium xanthum sp. nov. and Sphingobium yanoikuyae LMG 11252T (Biolog clusters 6–9) oxidized about the same number of substrates as strains of the Sphingomonas clusters 10–12. Nevertheless, several substrates could be used to differentiate the Sphingobium strains of clusters 8 and 9 from the Sphingomonas 16S rRNA gene cluster (Biolog clusters 10–12). Similarly, strains of clusters 6 and 7 (various species of the genus Sphingobium) could be easily distinguished from the 1346

Sphingomonas strains and the rest of the clusters by differential substrate oxidation. Strains of Rhizorhabdus gen. nov. oxidized only 10 substrates on Biolog plates, which enabled the distinction from Rhizorhapis gen. nov. strains (Table S3). In general, Rhizorhapis gen. nov. strains can be distinguished from the rest of the strains particularly because of poor oxidation of maltose, DL-lactic acid, Laspartic acid, L-glutamic acid and glycyl L-glutamic acid. Growth temperature limits were assessed visually after one and two weeks of growth on solid S medium at 30, 34, 37 and 40 uC (Table S3). After two weeks all strains had grown at 30 and 34 uC, except for Sphingobium sp. NL4 which did not grow at 34 uC. At 37 uC most strains grew well, except for members of the genera Rhizorhapis gen. nov. and Rhizorhabdus gen. nov. while only Sphingobium yanoikuyae LMG 11252T grew at 40 uC. Growth was also estimated by determining the OD650 of liquid S medium cultures with the pH adjusted to 4, 5, 6, 7, 8 or 9 with HCl or KOH before autoclaving (Table S3). None of the strains grew at pH 4, while pH 6–8 supported growth for all of the strains. Members of the genus Sphingomonas, Sphingobium yanoikuyae LMG 11252T as well as Sphingobium mellinum sp. nov. CA16 and WI4T were able to tolerate a pH as low as 5. These strains together with Sphingopyxis sp. CA29 and CA32, Sphingobium xanthum sp. nov. FL22, and Sphingobium sp. AU71, CA22 and GR3 even grew when the pH was raised to 9. All strains used in this study (except the reference strains Escherichia coli DH1 and Agromonas oligotrophica ATCC 43045T) were tested for pathogenicity to lettuce (Lactuca sativa L. ‘Salinas’). Lettuce seedlings were raised in vermiculite in 4 cm-wide pots placed inside insect-proof cages (one cage for each isolate or control) in a greenhouse with supplementary lighting by 400 W multivapour lamps for 14 h per day. One-week-old seedlings were inoculated at the stem base with 2 ml bacterial suspension (107– 108 c.f.u. ml21) scraped from S agar into sterile distilled water. Non-inoculated control plants received sterile distilled water. The plants received water, 0.56 Hoagland solution, or a solution of Ca(NO3)2+KNO3 (each at 5 mM) on alternate days (van Bruggen et al., 1989). Mean maximum and minimum temperatures as measured with a min–max thermometer were 31 and 19 uC, respectively. After 3 weeks, plants were uprooted and rated for corky root severity on a scale from 0–9 (Brown & Michelmore, 1988). The results for this assay are listed in Table S1. Most of the strains clustering with Rhizorhapis suberifaciens gen. nov., comb. nov. CA1T on the basis of 16S rRNA gene sequence and fatty acid analysis induced typical symptoms of corky root on lettuce cultivar Salinas. Notable exceptions were strains GB2, NL6 and NL8, which most likely represent a novel species of the genus Rhizorhapis gen. nov. Of the other genera tested, Rhizorhabdus gen. nov., Sphingopyxis and Sphingobium had few pathogenic strains, whereas no pathogenic strains were observed among species of the genus Sphingomonas. This is the first report of corky root-inducing

Downloaded from www.sgmjournals.org by International Journal of Systematic and Evolutionary Microbiology 64 IP: 93.91.26.109 On: Sat, 25 Jul 2015 11:47:58

Reclassification of lettuce corky root bacteria

strains that belong to genera outside of Rhizorhapis gen. nov. and research is ongoing to examine if pathogenicity genes are located on a plasmid. Our data support the conclusion drawn by Takeuchi et al. (2001) that species should not be assigned to the genus Sphingomonas solely on the basis of the presence of glycosphingolipids in their cell envelopes. We therefore propose a division of the genus Sphingomonas into Sphingomonas sensu stricto plus five additional genera, Sphingopyxis, Novosphingobium, Rhizorhapis gen. nov., Rhizorhabdus gen. nov. and Sphingobium. Thus, we suggest with Takeuchi et al. (2001) retaining Rhizorhapis gen. nov. as a separate genus, and therefore undoing the reclassification of Rhizomonas suberifaciens as part of the genus Sphingomonas (Yabuuchi et al., 1999) or Sphingobium (Chen et al., 2013). The current study demonstrated again that taxonomic studies with a limited number of strains leads to undesirable clumping of taxa. A summary of the main characteristics for differentiation among the genera and species that are described below, is given in Tables 2 and 3.

Cells are Gram-negative, straight or slightly curved rods that are motile by one lateral, subpolar or polar flagellum (Fig. S4), or are non-motile (van Bruggen et al., 1990). Resting stages are unknown. Cell division takes place by binary fission. Cells accumulate polyhydroxybutyrate (PHB) granules and are arginine dihydrolase-negative. Colonies are non-fluorescent, white or cream–white, and are smooth or wrinkled. Cells are obligate aerobes and have an oxidative metabolism. The optimum growth temperature is 28–32 uC; the maximum growth temperature is 37 uC. Ethanol is not converted to acetic acid. Oxidase and catalase are produced. Denitrification to N2 gas does not occur. Ubiquinone Q10 is present. Whole-cell fatty acids consist mainly of even-numbered unsaturated (C18 : 1 and C16 : 1) and saturated (C16 : 0) straight-chain fatty acids, as well as C14:O 2-OH. The C18:l fatty acids represent at least 50 % of the total fatty acids. The G+C content of the DNA ranges from 58 to 60 mol%. The type species is Rhizorhapis suberifaciens formerly known as Rhizomonas suberifaciens (van Bruggen et al., 1990).

Description of Rhizorhapis gen. nov.

Description of Rhizorhapis suberifaciens comb. nov.

Rhizorhapis (Rhi.zo.rha9pis. Gr. n. rhiza root; Gr. fem. n. rhapis rod; N.L. fem. n. Rhizorhapis a rod associated with roots).

Rhizorhapis suberifaciens (su.be.ri.fa9ci.ens. L. gen. n. suberis of cork, corky; L. part. adj. faciens making, producing; N.L. part. adj. suberifaciens cork making).

Previous illegitimate name: Rhizomonas van Bruggen et al. 1990.

Basonym: Sphingobium suberifaciens (van Bruggen et al. 1990) Chen et al. 2013.

Table 2. Differential characteristics of Rhizorhapis gen. nov. and Rhizorhabdus gen. nov., and the closely related genera Sphingopyxis, Sphingobium and Sphingomonas, with the latter two genera represented by the type species Sphingobium yanoikuyae and Sphingomonas paucimobilis, respectively Genera: 1, Rhizorhapis gen. nov.; 2, Rhizorhabdus gen. nov.; 3, Sphingobium; 4, Sphingopyxis; 5, Sphingomonas. +, Positive; 2, negative; ND, no data; SUM7, summed feature 7 representing fatty acids C18 : 1v7c, C18 : 1v9c, and C18 : 1v12t that cannot be separated by the Microbial Identification System. Characteristic Cell size (mm) Major fatty acids

1

2

1.460.4 1.560.8 C16 : 0, C16 : 1v7c, C14 : 0 2-OH, C16 : 0, SUM7 C16 : 1v7c, SUM7 DNA G+C content (mol%) 58.9±0.3 64.0±0.2 Colony colour (Cream) white White–bright white Oligotrophic + + Growth at 37 uC 2 2 Plant pathogenic + +/2 Assimilation of: 2 2 N-acetyl-D-glucosamine D-Fructose 2 + Maltose 2 2 + 2 Methyl b-D-glucoside Sucrose 2 2 2,3-Butanediol + 2

3

4

5

2.060.9 C16 : 1v7c, SUM7

0.960.4* C15 : 0 2-OH, C16 : 1v7c, SUM7, C17 : 1v6c 64±1.0* Yellowish-white

1.560.6* C14 : 0 2-OH, SUM7

ND

ND

63.8±0.9 Cream–yellow 2 +/2 +/2 + + + + + 2

ND

+/2 2 ND

+ ND ND ND

64.9±0.9 Cream–yellow + 2 + + + + + 2

*Data from Takeuchi et al. (2001). http://ijs.sgmjournals.org

Downloaded from www.sgmjournals.org by IP: 93.91.26.109 On: Sat, 25 Jul 2015 11:47:58

1347

I. M. Francis and others

Table 3. Differential characteristics of strains WI4T and NL9T, the type strains of the two novel species of the genus Sphingobium, and Sphingobium yanoikuyae LMG 11252T, the type species of the genus Sphingobium Strains: 1, Sphingobium mellinum sp. nov. WI4T; 2, Sphingobium xanthum sp. nov. NL9T; 3, Sphingobium yanoikuyae LMG 11252T. +, Positive; 2, negative; SUM7, summed feature 7 representing fatty acids C18 : 1v7c, C18 : 1v9c, and C18 : 1v12t that cannot be separated by the Microbial Identification System. Characteristic Cell size (mm) Major fatty acids DNA G+C content (mol%) Colony colour Growth at 37 uC Plant pathogenic Assimilation of: N-Acetyl-D-glucosamine D-Fructose D-Galactose D-Mannose Sucrose a-Ketobutyric acid D-Alanine L-Asparagine L-Threonine 2,3-Butanediol

1

2

3

0.960.5 C16 : 0, SUM7 63.0 Dark yellow + +

1.560.8 C14 : 0 2-OH, SUM7 64.7 Bright yellow 2 2

2.060.9 C16 : 1v7c, SUM7 61.7* Creamy white + 2

2 2 2 2 2 + 2 + + +

2 2 + + 2 + + 2 2 +

+ + + + + 2 2 + 2 2

*Data from Yabuuchi et al. (1990).

Cells are Gram-negative, straight or slightly curved rods that are non-motile or motile with one lateral, subpolar or polar flagellum (Fig. S4) and measure 1.19±0.236 0.46±0.05 mm. Filaments (up to 12 mm long) may occur. Cell division takes place by binary fission. Growth is better on diluted than on full-strength common bacteriological media. Colonies on S medium are non-pigmented, nonfluorescent, initially circular, and pulvinate or umbonate with entire margins, but later irregular and wrinkled with undulate margins. All strains are obligate aerobes. Optimal growth occurs at 28–32 uC; strains do not grow at 37 uC. The oxidase reaction is positive, and the catalase reaction is weakly positive. Nitrate is reduced to nitrite and ammonia, but not to nitrogen gas. Nitrogenase is not produced. Ethanol is not converted to acetic acid. Cells accumulate PHB granules, and arginine dihydrolase is not produced. All strains produce acid from oxidative fermentation of glucose, maltose, salicin and mannose. Whole-cell fatty acid profiles consist mainly of C18 : 1, C16 : 1 and C16 : O fatty acids. The presence of a large amount of C14 : 0 2-OH fatty acid is characteristic; C18 : 1 10-CH3 and C19 : 0 cyclo fatty acids are usually found. The G+C content of the DNA ranges from 58.2 to 59.5 mol%. Most strains are pathogenic for several members of the family Asteraceae tribe Cichoriae and induce corkiness or swellings or both on the main roots. The type strain, CA1T (5LMG 17323T5ATCC 49355T), was isolated from lettuce roots in the Salinas Valley of California, USA. 1348

Description of Rhizorhabdus gen. nov. Rhizorhabdus (Rhi.zo.rhab9dus. Gr. n. rhiza root; Gr. fem. n. rhabdus rod; N.L. fem. n. Rhizorhabdus a rod associated with roots). Cells are Gram-negative, non-sporulating rods that are motile or non-motile. On S medium, colonies are smooth and convex or compact and rough. Colony colour is white or cream–white. Growth is obligate aerobic and oligotrophic. Few carbon sources are utilized. Maximum growth occurs at 34 uC. Major cellular fatty acids are C18 : 1, C16 : 1 and C16 : 0. The main hydroxylated fatty acid is C14 : 0 2-OH. C18 : 1 10-CH3 and C19 : 0 cyclo fatty acids occur in trace and moderate amounts, respectively. The G+C content of chromosomal DNA ranges from 63–65 mol%. Some strains are pathogenic to Lactuca spp. and other genera in the tribe Cichoriae of the Asteraceae, inducing yellow banded lesions and brown corky areas on main and lateral roots. The type species is Rhizorhabdus argentea. Description of Rhizorhabdus argentea sp. nov. Rhizorhabdus argentea (ar.gen9te.a. L. fem. adj. argentea silvery). Cells are Gram-negative, flagellated or non-motile rods (approx. 1.560.8 mm). Colonies on S medium are smooth and convex or compact and rough. Colonies are bright white. Growth is obligately aerobic and oligotrophic. Few substrates are oxidized, namely D-fructose, a-D-glucose,

Downloaded from www.sgmjournals.org by International Journal of Systematic and Evolutionary Microbiology 64 IP: 93.91.26.109 On: Sat, 25 Jul 2015 11:47:58

Reclassification of lettuce corky root bacteria

a-ketobutyric acid, alaninamide, L-alanine, L-alanyl glycine, L-glutamic

acid, glycyl L-glutamic acid, L-proline, L-serine and glycerol. Maximum growth occurs at 34 uC. Major cellular fatty acids are C18 : 1, C16 : 1 and C16 : 0. The main hydroxylated fatty acid is C14 : 0 2-OH. C18 : 1 10-CH3 and C19 : 0 cyclo fatty acids occur in trace and moderate amounts, respectively. The G+C content of chromosomal DNA ranges from 63.6 to 64.4 mol%. The type strain is SP1T (5LMG 12581T5ATCC 51289T), isolated from soil using lettuce seedlings as bait in Montes de Toledo, Spain. Description of Sphingobium mellinum sp. nov. Sphingobium mellinum (mel.li9num. L. neut. adj. mellinum honey-like). Cells are Gram-negative, flagellated rods (approx. 0.96 0.5 mm) that grow on solid S medium in compact but smooth, dark yellow colonies, sometimes producing a brown diffusible pigment. All strains are obligate aerobes and not oligotrophic. Relatively few carbon sources are utilized. The following substrates are oxidized by all strains: a-D-glucose, maltose, methyl b-D-glucoside, DL-lactic acid, L-aspartic acid, L-glutamic acid, L-threonine, glycyl Lglutamic acid and 2,3-butanediol. Nitrate is reduced to nitrite and ammonia, but not to nitrogen gas (van Bruggen et al., 1990). Growth occurs at 37 uC but not at 40 uC. Profuse growth occurs between pH 6 and pH 8, but slight growth occurs at pH 5 and pH 9. Whole-cell fatty acid profiles consist mainly of C18 : 1, C16 : 0 and C16 : 1. The main hydroxylated fatty acid is C14 : 0 2-OH. C18 : 1 10-CH3 and C19 : 0 cyclo are found in trace amounts. The G+C content of chromosomal DNA varies from 62.7 to 63.5 mol%. Some strains are pathogenic to Lactuca spp. and other genera in the tribe Cichoriae of the Asteraceae, inducing yellow-banded lesions and brown corky areas on main and lateral roots. The type strain, WI4T (5LMG 11032T5ATCC 51292T), was isolated from soil using lettuce seedlings as bait in Wisconsin, USA. Description of Sphingobium xanthum sp. nov. Sphingobium xanthum (xan9thum. Gr. adj. xanthos yellow). Cells are Gram-negative, flagellated rods (approx. 1.56 0.8 mm). Colonies on solid S medium are compact, smooth and bright yellow. All strains are obligate aerobes and not oligotrophic. The following substrates are oxidized by all strains: D-galactose, a-D-glucose, maltose, D-mannose, Lrhamnose, DL-lactic acid, quinic acid, L-alanyl glycine, Laspartic acid, glycyl L-aspartic acid, glycyl L-glutamic acid, L-leucine, L-proline and inosine. Some strains grow at 37 uC but none grow at 40 uC. Profuse growth occurs between pH 6 and pH 8, with slight growth by some strains at pH 9. Some strains are pathogenic to Lactuca spp. and other genera in the tribe Cichoriae of the http://ijs.sgmjournals.org

Asteraceae, inducing yellow-banded lesions on main and lateral roots. Whole-cell fatty acid profiles consist mainly of C18 : 1, C17 : 1, C16 : 0 and C16 : 1. The main hydroxylated fatty acid is C14 : 0 2-OH; C15 : 0 2-OH also occurs. C18 : 1 10-CH3 is only found in trace amounts in some strains. The type strain is strain NL9T (5LMG 12560T5ATCC 51296T), isolated from soil using lettuce seedlings as bait in Maasland, The Netherlands. The chromosomal DNA G+C content of the type strain is 64.7 mol%.

Acknowledgements We thank Joe Wakeman and Oscar de Vos for lyophilizing and further handling of the cultures. We also thank Arjen Speksnijder for attempting to sequence the DNA from the lyophilized cultures. We would like to thank Professor Bernard Schink for assistance with nomenclature. We are grateful to Paul Segers, Marc Vancanneyt and Monique Gillis of the Laboratory of Microbiology at Ghent University, Belgium, for their contributions to the early part of this research and an earlier version of this manuscript. This work was supported by research grants from the California Iceberg Lettuce Research Board, travel grants from NATO and NSF to A. H. C. v. B., and research and personnel grants from the National Fund for Medical Scientific Research, Belgium to P. D. V.

References Balkwill, D. L., Fredrickson, J. K. & Romine, M. F. (2006).

Sphingomonas and related genera. In The Prokaryotes: a Handbook on the Biology of Bacteria, vol. 7, pp. 605–629. Edited by M. Dworkin, S. Falkow, E. Rosenberg, K.H. Schleifer & E. Stackebrandt. New York: Springer. Brown, P. R. & Michelmore, R. W. (1988). The genetics of corky root resistance in lettuce. Phytopathology 78, 1145–1150. Chen, H., Jogler, M., Rohde, M., Klenk, H.-P., Busse, H.-J., Tindall, B. J., Spro¨er, C. & Overmann, J. (2013). Sphingobium limneticum sp.

nov. and Sphingobium boeckii sp. nov., two freshwater planktonic members of the family Sphingomonadaceae, and reclassification of Sphingomonas suberifaciens as Sphingobium suberifaciens comb. nov. Int J Syst Evol Microbiol 63, 735–743. Datnoff, L. E. & Nagata, R. T. (1990). Isolation of Rhizomonas

suberifaciens in Florida. Plant Dis 74, 394. De Ley, J. (1970). Reexamination of the association between melting

point, buoyant density, and chemical base composition of deoxyribonucleic acid. J Bacteriol 101, 738–754. De Ley, J. & Van Muylem, J. (1963). Some applications of

deoxyribonucleic acid base composition in bacterial taxonomy. Antonie van Leeuwenhoek 29, 344–358. De Vos, P. & Tru¨per, H. G. (2000). Judicial Commission of the

International Committee on Systematic Bacteriology IXth International (IUMS) Congress of Bacteriology and Applied Microbiology. Minutes of the meetings, 14, 15 and 18 August 1999, Sydney, Australia. Int J Syst Evol Microbiol 50, 2239–2244. Dees, S. B., Moss, C. W., Weaver, R. E. & Hollis, D. (1979). Cellular

fatty acid composition of Pseudomonas paucimobilis and groups IIk-2, Ve-1, and Ve-2. J Clin Microbiol 10, 206–209. Felsenstein, J. (1985). Confidence limits on phylogenies: an approach

using the bootstrap. Evolution 39, 783–791. BIONJ: an improved version of the NJ algorithm based on a simple model of sequence data. Mol Biol Evol 14, 685–695.

Gascuel, O. (1997).

Downloaded from www.sgmjournals.org by IP: 93.91.26.109 On: Sat, 25 Jul 2015 11:47:58

1349

I. M. Francis and others

Gevers, D., Cohan, F. M., Lawrence, J. G., Spratt, B. G., Coenye, T., Feil, E. J., Stackebrandt, E., Van de Peer, Y., Vandamme, P. & other authors (2005). Opinion: Re-evaluating prokaryotic species. Nat Rev

chemotaxonomic analyses. Int J Syst Evol Microbiol 51, 1405– 1417.

Microbiol 3, 733–739.

Rhizomonas sp. causing corky root of lettuce in Europe. Neth J Plant Pathol 98, 45–56.

Gouy, M., Guindon, S. & Gascuel, O. (2010). SeaView version 4: a

multiplatform graphical user interface for sequence alignment and phylogenetic tree building. Mol Biol Evol 27, 221–224. Guindon, S., Dufayard, J. F., Lefort, V., Anisimova, M., Hordijk, W. & Gascuel, O. (2010). New algorithms and methods to estimate maximum-

van Bruggen, A. H. C. & Jochimsen, K. N. (1992). First report of

van Bruggen, A. H. C. & Jochimsen, K. N. (1993). First report of

Rhizomonas suberifaciens causing corky root of lettuce in Australia. Australas Plant Pathol 22, 14–19. van Bruggen, A. H. C., Grogan, R. G., Bogdanoff, C. P. & Waters, C. M. (1988). Corky root of lettuce in California caused by a Gram-

likelihood phylogenies: assessing the performance of PhyML 3.0. Syst Biol 59, 307–321.

negative bacterium. Phytopathology 78, 1139–1145.

Hiraishi, A., Kuraishi, H. & Kawahara, K. (2000). Emendation of

van Bruggen, A. H. C., Brown, P. R. & Jochimsen, K. N. (1989). Corky

the description of Blastomonas natatoria (Sly 1985) Sly and Cahill 1997 as an aerobic photosynthetic bacterium and reclassification of Erythromonas ursincola Yurkov et al. 1997 as Blastomonas ursincola comb. nov. Int J Syst Evol Microbiol 50, 1113–1118. Holmes, B., Owen, R. J., Evans, A., Malnick, H. & Willcox, W. R. (1977). Pseudomonas paucimobilis, a new species isolated from human

clinical specimens, the hospital environment, and other sources. Int J Syst Bacteriol 27, 133–146. Kawahara, K., Seydel, U., Matsuura, M., Danbara, H., Rietschel, E. T. & Za¨hringer, U. (1991). Chemical structure of glycosphingolipids

isolated from Sphingomonas paucimobilis. FEBS Lett 292, 107–110.

root of lettuce caused by strains of a Gram-negative bacterium from muck soils of Florida, New York, and Wisconsin. Appl Environ Microbiol 55, 2635–2640. van Bruggen, A. H. C., Jochimsen, K. N. & Brown, P. R. (1990).

Rhizomonas suberifaciens gen. nov., sp. nov., the causal agent of corky root of lettuce. Int J Syst Bacteriol 40, 175–188. van Bruggen, A. H. C., Jochimsen, K. N., Benedict, A. A. & Pollard, L. W. (1992). Monoclonal antibodies for the detection of Rhizomonas

suberifaciens, causal agent of corky root of lettuce, with enzyme immunoassays. Food Agric Immunol 4, 27–40.

Kent, W. S. (1880). A Manual of the Infusoria: Including a Description

van Bruggen, A. H. C., Jochimsen, K. N., Steinberger, E. M., Segers, P. & Gillis, M. (1993). Classification of Rhizomonas suberifaciens, an

of All Known Flagellate, Ciliate and Tentaculiferous Protozoa, British and Foreign, and an Account of the Organization and Affinities of the Sponges, vol. I. London: David Bogue.

unnamed Rhizomonas species, and Sphingomonas spp. in rRNA superfamily IV. Int J Syst Bacteriol 43, 1–7.

Kloesges, T., Popa, O., Martin, W. & Dagan, T. (2011). Networks of

gene sharing among 329 proteobacterial genomes reveal differences in lateral gene transfer frequency at different phylogenetic depths. Mol Biol Evol 28, 1057–1074. Lambert, M. A., Hickman-Brenner, F. W., Farmer, J. J., III & Moss, C. W. (1983). Differentiation of Vibrionaceae species by their cellular

fatty acid composition. Int J Syst Bacteriol 33, 777–792. Lambert, B., Meire, P., Joos, H., Lens, P. & Swings, J. (1990). Fast-

growing, aerobic, heterotrophic bacteria from the rhizosphere of young sugar beet plants. Appl Environ Microbiol 56, 3375–3381. Marmur, J. & Doty, P. (1962). Determination of the base composition

of deoxyribonucleic acid from its thermal denaturation temperature. J Mol Biol 5, 109–118. Sly, L. I. & Cahill, M. M. (1997). Transfer of Blastobacter natatorius (Sly 1985) to the genus Blastomonas gen. nov. as Blastomonas natatoria comb. nov. Int J Syst Bacteriol 47, 566–568. Stackebrandt, E. & Ebers, J. (2006). Taxonomic parameters revisited:

tarnished gold standards. Microbiol Today 33, 152–155. Takeuchi, M., Hamana, K. & Hiraishi, A. (2001). Proposal of the genus

Sphingomonas sensu strictu and three new genera, Sphingobium, Novosphingobium and Sphingopyxis, on the basis of phylogenetic and

1350

Yabuuchi, E., Yano, I., Oyaizu, H., Hashimoto, Y., Ezaki, T. & Yamamoto, H. (1990). Proposals of Sphingomonas paucimobilis gen.

nov. and comb. nov., Sphingomonas parapaucimobilis sp. nov., Sphingomonas yanoikuyae sp. nov., Sphingomonas adhaesiva sp. nov., Sphingomonas capsulata comb. nov., and two genospecies of the genus Sphingomonas. Microbiol Immunol 34, 99–119. Yabuuchi, E., Kosako, Y., Naka, T., Suzuki, S. & Yano, I. (1999).

Proposal of Sphingomonas suberifaciens (van Bruggen, Jochimsen and Brown 1990) comb. nov., Sphingomonas natatoria (Sly 1985) comb. nov., Sphingomonas ursincola (Yurkov et al. 1997) comb. nov., and emendation of the genus Sphingomonas. Microbiol Immunol 43, 339–349. Yabuuchi, E., Kosako, Y., Fujiwara, N., Naka, T., Matsunaga, I., Ogura, H. & Kobayashi, K. (2002). Emendation of the genus

Sphingomonas Yabuuchi et al. 1990 and junior objective synonymy of the species of three genera, Sphingobium, Novosphingobium and Sphingopyxis, in conjunction with Blastomonas ursincola. Int J Syst Evol Microbiol 52, 1485–1496. Yurkov, V., Stackebrandt, E., Buss, O., Vermeglio, A., Gorlenko, V. & Beatty, J. T. (1997). Reorganization of the genus Erythromicrobium:

description of ‘‘Erythromicrobium sibiricum’’ as Sandaracinobacter sibiricus gen. nov., sp. nov., and of ‘‘Erythromicrobium ursincola’’ as Erythromonas ursincola gen. nov., sp. nov. Int J Syst Bacteriol 47, 1172–1178.

Downloaded from www.sgmjournals.org by International Journal of Systematic and Evolutionary Microbiology 64 IP: 93.91.26.109 On: Sat, 25 Jul 2015 11:47:58