Origin and evolution of organisms as deduced from 5S ribosomal RNA sequences.

Origin and Evolution of Organisms as Deduced from 5s Ribosomal RNA Sequences’ Hiroshi Hori* and Syozo Osawat *Department of Molecular Hiroshima University; and TLaboratory o f Genetics, GEN-KEN, Genetics,, Department of Biology, Nagoya University Introduction A phylogenetic tree of most of the major groups of organisms has been constructed from the 352 5s ribosomal RNA sequences now available. The tree suggests that there are several major groups of eubacteria that diverged during the early stages of their evolution. Metabacteria (=archaebacteria) and eukaryotes separated after the emergence of eubacteria. Among eukaryotes, red algae emerged first; and, later, thraustochytrids (a Proctista group), ascomycetes (yeast), green plants (green algae and land plants), “yellow algae” (brown algae, diatoms, and chrysophyte algae), basidiomycetes (mushrooms and rusts), slime- and water molds, various protozoans, and animals emerged, approximately in that order. Three major types of photored algae (=Chlorophyll a group), green plants (Chl. synthetic eukaryotes-i.e., a+b group) and yellow algae (Chl. a+c)-are remotely related to one another. Other photosynthetic unicellular protozoans-such as Cyanophora (Chl. a), Euglenophyta (Chl. a+b), Cryptophyta (Chl. a+c), and Dinophyta (Chl. a+c)-seem to have separated shortly after the emergence of the yellow algae. At present, the evolutionary relationships of the major groups of organisms are quite obscure, and the present systems of classification are mainly based on physiological and morphological characters. Since the evolutionary changes of such characters are very complicated and the rate of change is variable in different groups of organisms or in different evolutionary periods, not much confidence can be given to the systems. A more useful approach to this problem is to use DNA or RNA sequences, because the evolutionary change of these molecules is roughly proportional to evolutionary time. The 5s ribosomal RNA (5s rRNA) sequence is particularly useful for establishing the phylogenetic relationship of distantly related organisms (Kimura and Ohta 1973; Hori 1975) because of its low substitution rate (mean rfr SE 0.18 + 0.05 substitution/ nucleotide site/lo’ years; Hori et al. [ 19771) and because of its basic similarity of structure among all organisms, which makes it possible to align the sequences for the construction of a comprehensive phylogenetic tree. The 5s rRNA phylogenetic trees for many groups of organisms or organelles have been reported, e.g., for eubacteria (Dekio et al. 1984; Vandenberghe et al. 1985), “the purple eubacterial group” (Lane et al. 1985), the eubacterial family Vibrionaceae (MacDonell and Colwell 1985; MacDonell et al. 1986) Mycoplasmas (Rogers et al. 1985), metabacteria (Fox et al. 1982; Hori et al. 1982), green plants (Hori et al. 1985a), Ascomycota (Chen et al. 1984), Basidiomycota (Walker and Doolittle 1982; Huysmans et al. 1983; Gottschalk and Blanz 1984; Walker 1984), protozoans (Kumazaki et al. 1. Key words: 5s rRNA, simplified UPGMA trees, molecular phylogeny. Address for correspondence and reprints: Dr. Hiroshi Hori, Department of Genetics, GEN-KEN, Hiroshima University, Kasumi, Hiroshima, 734, Japan. Mol. Biol. Evol. 4(5):445-472. 1987. 0 1987 by The University of Chicago. All rights reserved. 0737-4038/87/0405-0001$02.00 445 446 Hori and Osawa 1983a), Meso- and Metazoa (Ohama et al. 1984), and organelles (Hori et al. 1982; Wolters and Erdmann 1984). However, a 5s rRNA tree for all groups of organisms has not been constructed. In the present paper, we have employed the 352 sequences of 5s rRNAs now available to construct a phylogenetic tree of a wide spectrum of extant organisms, including organelles, by means of a simplified unweighted-pairgroup (UPG) method. Material and Methods Sequence Alignment of 5s rRNA Construction of Phylogenetic Trees The 352 5s rRNA sequences from various organisms available as of January 1986 have been used in the present study. Representative organisms examined herein are taxonomically summarized in table 1. The alignment of these sequences was obtained mainly by juxtaposing the 5s rRNA secondary structures as described elsewhere (Hori et al. 1985b). The evolutionary distance, Knuc, between two sequences was calculated by means of the equation described by Kimura (1980). Knuc estimates the number of base substitutions per nucleotide site that have occurred since the separation of the two sequences. Knuc = -( 1/2)log,[( 1 - 2P- Q)( 1 - 2Q)‘12], (1) where P and Q are the fractions of nucleotide sites between two sequences showing transition- and transversion-type differences, respectively. The SE of the Knuc, SEK, was calculated by using Kimura’s (1980) equation. When a gap of length one was paired with one nucleotide, it was counted as equal to one transversion-type substitution. Large deletions in 5s rRNA sequence -e.g., those found in the sequences of Mycoplasma species -are likely to be due to single rare events rather than to the compound effect of several separate events. Therefore, a gap of two or more nucleotides was counted as two differences in determining Q. The G+C content of genomic DNA in eubacteria is diversified to a considerable extent, ranging from 25% to 75%. Since the G+C content of 5s rRNA more or less reflects the genomic G+C content in eubacteria, we introduced a parameter to cancel such an effect that might influence the rate of nucleotide substitution in 5s rRNA molecules. (In eukaryotes and metabacteria, the genomic G+C content does not correlate significantly with the G+C content of 5s rRNA.) To estimate the evolutionary distance between sequences i and j, the following equation was adopted from Hori and Osawa (1986). Dnuc = (cJcj)Knuc, (2) where Knuc is the value from equation (1) and ci and Cj(ci ;5 Cj)are the G+C contents of sequences i and j, respectively. With use of the Knuc or Dnuc values, a phylogenetic tree was constructed by means of a “simplified” method of the UPG method by using arithmetic averages (Sneath and Sokal 1973). For the estimation of the SE of each branching point in the tree, the variance of each branching point was calculated by means of the equation described by Nei et al. (1985). This is given by Evolution of Organisms from 5s rRNA Sequences 447 where dkl is the inter-cluster distance between the kth species in cluster A and the Zth species in cluster B and Yand s are the numbers of species in clusters A and B, respectively; V and Cov are the variance and covariance, respectively. In the actual computation, however, to avoid excessive computational time owing to the large number of 5s rRNA sequences (352 in this case), (KS)~was conventionally kept < 16 by using representative sequences in each cluster and was used for tree construction by means of the UPG method (=“simplified” UPG method). Results and Discussion Validity of Phylogenetic Trees Deduced from 5s rRNA Sequences As mentio (...truncated)