INTERNATIONALJOURNAL OF SYSTEMATICBACTERIOLOGY, July 1977, p. 222-240 Vol. 27, No. 3 Copyright 0 1977 International Association of Microbiological Societies Printed in U.S.A. Intra- and Intergeneric Similarities of Agrobacterium Ribosomal Ribonucleic Acid Cistrons J. DE SMEDT AND J. DE LEY Laboratory of Microbiology and Microbial Genetics, Faculty of Sciences, State University, B-9000 Ghent, Belgium We prepared hybrids between 14C-labeledribosomal ribonucleic acid (rRNA) from either Agrobacterium tumefaciens ICPB TTlll or A. rhizogenes ICPB TR7, and deoxyribonucleic acid (DNA) from a great variety of reference gram- negative and gram-positive bacteria. Each hybrid was described by (i) its TMe,, the temperature at which 50% of the hybrid was denatured, and (ii) percentage of rRNA binding, i.e., micrograms of 14C-labeledrRNA duplexed per 100 pg of filter-fixed DNA. Each taxon occupied a definite area on the rRNA similarity map. The size and shape of this area depended on the phenotypic and genetic heterogeneity of the taxon. There appeared to be a correlation between TMe,of the heterologous hybrids and the overall phenotypic similarities of the orga- nisms and taxa involved. TMe,values above 65°C were taxonomically most meaningful. DNA:rRNA hybridizations condensed all strains from a genus in one narrow cluster; the method had little resolution to distinguish species within a genus, but it seemed to be a very useful approach to detect remote relation- ships at the inter- and suprageneric level, for taxonomic and identification purposes. The hybrid parameters of Azotomonas fluorescens, Mycoplana bul- lata, Mycoplana dimorpha, Phyllobacterium, two misnamed “Chromobacter- ium liuidum” strains from leaf-nodulating plants, two misnamed agrobacteria from the Baltic Sea, and a few misnamed “Achromobacter” strains were all in the vicinity of Agrobacterium and Rhizobium. We suggest that all of these organisms are remote relatives and belong in the family of the Rhizobiaceae. Azotomonas insolita NCIB 9749 is misnamed; it is an Agrobacterium. Several organisms which had been misnamed Agro bacterium formed DNA:rRNA hy- brids with properties outside the Agrobacterium area. The nature and degree of the phylogenetic sequencing of bacterial cytochromes and other relationships between most bacterial genera small proteins might reveal the phylogenetic are not known, except for a limited number of relationships between bacterial genera (12). cases, where such relationships can be deduced This area is in full expansion (3). However, it indirectly from comparative biochemistry (131, may take a very long time and much effort to numerical analysis of phenotypic data (651, and solve this single aspect of bacterial relation- deoxyribonucleic acid (DNA):DNA hybridiza- ships. tions (14, 21). Most bacterial genera, however, It seemed to us that a faster way than deter- are evolutionarily too far removed from each mining amino acid sequences in homologous other to form stable DNA:DNA hybrids; the gene products consisted in comparing base se- last technique is in general useful either within quence similarities between homologous genes. a genus such as Agrobacterium (16, 171, or The advantage of the latter approach was that between genera which did not diverge too the actual sequences need not to be known. The much, such as in the Enterobacteriaceae (7). obvious choice was to compare ribosomal ribo- The detection of more remote relationships nucleic acid (rRNA) cistrons from several gen- between bacterial taxa will require other tech- era by DNArRNA hybridizations. rRNA cis- niques. One of them is the comparison of amino trons are conserved: their base sequences have acid sequences of small proteins. It is well es- changed less than those of the bulk of the tablished that the amino acid sequences of mi- genes constituting the bacterial genome (26, tochrondrial c cytochromes from animals corre- 27, 53, 71). Various degrees of rRNA sequence late well with the overall phylogenetic relation- similarities have been detected between pheno- ships of the organisms, clarified by other meth- typically quite different bacteria (39, 53, 54, ods (12, 49). The other way around, amino acid 56, 57), but extensive taxonomic improvements 222 VOL. 27, 1977 AGROBACTERIUM rRNA CISTRONS 223 have not been achieved. rRNA homology, scribed by De Ley et al. (19). Several gram-positive determined by preventing homologous DNA: and coryneform organisms lysed readily in the sol- rRNA duplexing in the presence of excess vent described by Crombach (11): 0.033 M tris- heterologous competing rRNA, is useful to sub- (hydroxymethy1)aminomethane-0.001 M ethylene- divide large and heterogeneous genera, such diaminetetraacetic acid (pH 8), containing 5 mg of lysozyme per g of wet cells. The molecular weight of Pseudomonas (58) and Clostridium (40). as the DNA fragments was 5 x lo6 to 10 x lo6. We shall examine whether bacterial rRNA Fixation of single - stranded high - molecular- similarities are valuable parameters to reveal weight DNA on membrane filters. We followed the taxonomic relatedness at the generic and su- fixation procedure as described by De Ley and Tyt- prageneric levels. In the present paper we shall gat (24). We used Sartorius SM 11309 membrane compare the rRNA cistrons of Agrobacterium filters. The charged filters were preserved at 4°C in within the genus and with those of a number of vacuo (20). other bacterial genera. Several conclusions Saturation hybridization between 14C-labeled rRNA and filter-fixed DNA: thermal stability of the were confirmed by DNA:DNA hybridizations DNA:rRNA hybrids. The basic aspects of the hybrid- and by protein electropherograms. ization conditions, the nature of ribonuclease and its effect on hybridization, the effect of hybridi- MATERIALS AND METHODS zation temperature on DNA leaching, the condi- Bacterial strains and growth media. The strains tions of saturation hybridization, etc., have been used are listed in Table 1 (see also Table 6). The reported in previous papers from this department bacteriological purity was checked by plating, and (20, 24). For DNA:rRNA hybridization and hybrid by examination of living and gram-stained cells. For stability we followed the methods of De Ley and De mass cultures, cells were grown in Roux flasks on Smedt (20). A 10-pg amount of 14C-labeled rRNA media solidified with 2.5% agar for 2 to 3 days at was incubated in 1 ml of 2x SSC in 20% formam- 28°C. A few slow-growing strains were grown for at ide, with a membrane filter carrying about 50 pg of most 4 to 6 days. The compositions of all growth DNA, for 16 h at the stringent optimal hybridization media used are summarized in Table 2. After growth temperature of 50°C. After washing, the filters were the cells were harvested, washed, and lyophilized. treated with RNase. The thermal stability of the We received three Phyllobacterium strains from hybrid was determined in &degree steps from 50 to D. Knosel, P. rubiacearum strain LMGl and P. myr- 90°C in 1.5~SSC in 20% formamide. The amount of sinacearum LMG2 and LMG3. Upon plating we de- labeled material released at each step was counted tected two colony types in each strain, labeled t, and in a Tri-Carb 33 10 liquid scintillation spectrometer t2. Strain LMGl t2 (sequence number 11 [see Table (Packard Instrument Co.) at 2°C for 50 min. TnLce,is 11; henceforward, these numbers will be indicated the temperature at which 50% of the hybrid was as, for example, “number 11” following the strain, eluted. The total amount of 14C-labeled rRNA was identified in this laboratory as a contaminating bound, after RNase treatment, expressed in micro- Agrobacterium. The five remaining strains (num- grams of rRNA per 100 pg of DNA retained on the bers 50 to 54) were included as Phyllobacterium. filter, was called the “percentage of rRNA binding.” Preparation of I4C-labeled rRNA. The organisms Chemical determination of DNA on the filter. In were inoculated in 100 ml of liquid medium A (Table each series of hybridizations, a vial was included 2), in a broad-bottomed Erlenmeyer flask with a with buffer and a DNA filter, but without 14C-la- Klett tube as side arm. Shaking provided good aera- beled rRNA. After the simulated hybridization step, tion. Growth was followed turbidimetrically in a the remaining fixed DNA was released from the Klett colorimeter at 660 nm. A 100-pCi amount of [2- filter by the method of Meys and Schilperoort (51) *4C]uracil (The Radiochemical Centre, Amersham, and determined by the method of Burton (10). Buckinghamshire, England) in 5 ml of 0.01 M phos- Preparation of 14C-labeled DNA. Labeled DNA phate buffer (pH 7.0) was sterilized separately was prepared as described before (35). The specific through a membrane filter and added at the start of activities of 14C-labeled DNA from A. tumefaciens the log phase. The cells were harvested near the end ICPB TTlll and B6 were 4,670 and 3,965 cpm/wg of of the exponential phase and washed. 14C-labeled DNA, respectively. Both were sheared in a French rRNA was prepared and purified as described by pressure cell to fragments of about 4 x lo6 molecular De Ley and De Smedt (20). The 23s (see below) and weight and were heat-denatured before use. 16s fractions were collected separately and dialyzed DNA:DNA hybridizations: thermal stability of at 4°C three times against 2 liters of 2x SSC buffer. the DNA:DNA duplexes. We followed the method of The fractions were preserved at -12°C. The specific De Ley and Tytgat (24), slightly modified, with 10 activities of Agrobacterium tumefaciens TTlll I4C- pg of single-stranded, sheared, 14C-labeledDNA in labeled rRNA and of Agrobacterium rhizogenes solution, and a membrane filter with about 22 pg of TR7 14C-labeledrRNA were 3,000 cpm/pg and 2,400 single-stranded, high-molecular-weight, unlabeled cpm/pg, respectively.