Although the applicability of small subunit ribosomal RNA (16S rRNA) sequences for bacterial classification is now well accepted, the general use of these molecules has been hindered by the technical difficulty of obtaining their sequences. A protocol is described for rapidly generating large blocks of 16S rRNA sequence data without isolation of the 16S rRNA or cloning of its gene. The 16S rRNA in bulk cellular RNA preparations is selectively targeted for dideoxynucleotide-terminated sequencing by using reverse transcriptase and synthetic oligodeoxynucleotide primers complementary to universally conserved 16S rRNA sequences. Three particularly useful priming sites, which provide access to the three major 16S rRNA structural domains, routinely yield 800-1000 nucleotides of 16S rRNA sequence. The method is evaluated with respect to accuracy, sensitivity to modified nucleotides in the template RNA, and phylogenetic usefulness, by examination of several 16S rRNAs whose gene sequences are known. The relative simplicity of this approach should facilitate a rapid expansion of the 16S rRNA sequence collection available for phylogenetic analyses.

The Bacillus subtilis ribonuclease P consists of a protein and an RNA. At high ionic strength the reaction is protein-independent; the RNA alone is capable of cleaving precursor transfer RNA, but the turnover is slow. Kinetic analyses show that high salt concentrations facilitate substrate binding in the absence of the protein, probably by decreasing the repulsion between the polyanionic enzyme and substrate RNAs, and also slow product release and enzyme turnover. It is proposed that the ribonuclease P protein, which is small and basic, provides a local pool of counter-ions that facilitates substrate binding without interfering with rapid product release.

The 16S rRNAs from the bacterial endosymbionts of six marine invertebrates from diverse environments were isolated and partially sequenced. These symbionts included the trophosome symbiont of Riftia pachyptila, the gill symbionts of Calyptogena magnffica and Bathymodiolus thermophilus (from deep-sea hydrothermal vents), and the gill symbionts of Lucinoma annulata, Lucinoma aequizonata, and Codakia orbicularis (from relatively shallow coastal environments). Only one type of bacterial 16S rRNA was detected in each symbiosis. Using nucleotide sequence comparisons, we showed that each of the bacterial symbionts is distinct from the others and that all fall within a limited domain of the gamma subdivision of the purple bacteria (one of the major eubacterial divisions previously defined by 16S rRNA analysis [C. R. Woese, Microbiol. Rev. 51:221-271, 1987]). Two host specimens were analyzed in five of the symbioses; in each case, identical bacterial rRNA sequences were obtained from conspecific host specimens. These data indicate that the symbioses examined are species specific and that the symbiont species are unique to and invariant within their respective host species.The recent discovery of the unique fauna of the hydrothermal-vent communities has brought to light an unexpected mode of animal nutrition in which chemoautotrophic bacterial symbionts are maintained within specialized cells of the host animal (4,7,8). The bacterial symbionts obtain energy by oxidizing reduced sulfur compounds from the environment (6, 17, 23) and can fix inorganic carbon (8,21). A portion of this fixed carbon is utilized by the eucaryotic host (5,18,24). This type of symbiosis is now known to exist in members of a wide variety of marine invertebrate taxa encompassing at least four phyla. In addition to hydrothermal vents, a broad range of marine environments, from intertidal to deep sea, are now known to support symbioses of this kind (3,19).To date, the bacterial symbionts of all known sulfur-based invertebrate-bacterial endosymbioses have eluded cultivation. Thus, until recently there has been no satisfactory means of investigating many fundamental questions concerning the identity of these symbionts and the nature of the host-symbiont relationships. Among such questions are the following. Are the endosymbionts of an individual host organism monospecific or a mixed population? Does the composition of the endosymbiotic population vary among individuals of a given host species? Do the symbionts of different sulfur-based, animal-bacterial symbioses belong to a single, closely related phylogenetic group, or are they drawn from a variety of bacterial groups? How are these symbionts related to well-characterized, cultured bacterial species? Do the phylogenies of the bacterial symbionts parallel those of their respective hosts, or do their relationships reflect, for example, the geographic distribution or environment of the hosts? In this investigation we used newly developed methods of symbiont purification (D. L. California; (iii) Codakia...

Some 37 reverse transcriptase, partial 16S rRNA sequences from sulfur-and/or iron-oxidizing eubacteria, including sequences from species of the genera Thiobacilus, Thiothrix, Thiomicrospira, Acidophiium, "LeptospiriUum," Thiovulum, and Chlorobium, have been determined. In addition, 16S sequences from a number of unnamed sulfur-and/or iron-oxidizing bacteria from hydrothermal vent sites, from invertebratebacterial endosymbioses, and from various mineral recovery operations also have been determined. The majority of sequences place their bacterial donors in one or another of the subdivisions of the Proteobacteria.However, three unnamed facultatively thermophilic iron-oxidizing isolates, Alv, BC, and TH3, are affiliated with the gram-positive division. One H2S-oxidizer, from the genus Thiovulum, is affiliated with Campylobacter, WolineUla, and other genera in what appears to be a new subdivision of the Proteobacteria. Three "Leptospirillum"-helical vibrioid isolates, BU-1, LfLa, and Z-2, exhibit no clear phylum level affiliation at all, other than their strong relationship to each other. A picture is emerging of an evolutionary widespread capacity for sulfur and/or iron oxidation among the eubacteria.Sulfur-and iron-oxidizing bacteria have been subjects of some interest for more than 100 years (35). Interest in their evolution stems, at least in part, from a realization of the enormous influence of chemolithotrophic metabolisms in shaping our planet. The notion that such chemolithotrophic metabolisms might be primitive traits has since gone in and out of vogue a number of times (3,16,27,28). The existence of sulfur-metabolizing archaebacteria, evolutionarily so far removed from the more familiar thiobacilli, and more recent discoveries of (arguably) more primitive variations of sulfur oxidation-reduction metabolisms (15, 16) have added new dimensions to evolutionary thinking about lithotrophic metabolism. The phylogenetic diversity of the iron-and sulfurmetabolizing phenotypes clearly suggests that a closer molecular inspection of the genes encoding the enzymes of these pathways could provide some important insights into their evolution and the evolution of the organisms harboring them. A 16S rRNA phylogeny of the sulfur-and ironoxidizing bacteria should provide a useful framework for focusing such analyses.The 16S rRNA analyses presented here focus on the phylogenetic affiliations of eubacterial iron and sulfur oxidizers. Partial or complete 16S rRNA nucleotide sequences from 37 iron-and/or sulfur-oxidizing strains have been obtained for these analyses. Genera represented are Thiobacillus, Thiomicrospira, Thiothrix, Acidophilium, "Leptospirillum," Thiovulum, Chromatium, and Chlorobium. In addition, fully half of the sequences in the collection derive from unculturable or unclassified isolates. Thus, a fairly