The Small-Subunit Ribosomal RNA Gene Sequences from the Hypotrichous Ciliates Oxytriclha IWWZ and Stylonychia pustulata’ Hille J. Elwood,* Gary J. 01sen,tT2 and Mitchell L. Sogin* *Department o f Molecular and Cellular Biology, National Jewish Hospital and Research Center; TDepartment of Biochemistry, Biophysics and Genetics, University of Colorado School of Medicine; and $Department of Molecular and Cellular Biology, National Jewish Hospital and Research Center We have determined the complete nucleotide sequence of the small-subunit ribo- somal RNA genes for the ciliate protozoans Stylonychia pustulata and Oxytricha nova. The sequences are homologous and sufficiently similar that these organisms must be closely related. In a phylogeny inferred from comparisons of several eu- karyotic small-subunit ribosomal RNAs, the divergence of the ciliates from the eukaryotic line of descent is seen to coincide with the radiation of the plants, the animals, and the fungi. This radiation is preceded by the divergence of the slime mold, Dictyostelium discoideum. Introduction Eukaryotic microorganisms are classified in either the protoctista or the fungi. In contrast to the fungi, which are a relatively cohesive phylogenetic grouping, the protoctists are a heterogeneous collection of organisms that display enormous physi- ological, cytological, and biochemical diversity (Margulis and Schwartz 1982). Con- sequently, it has been very difficult to infer consistent phylogenies for these simple eukaryotes using classical taxonomic approaches, i.e., comparative studies of pheno- types. As an alternative, similarities between ribosomal RNA sequences can be used to define quantitative phylogenetic relationships for these organisms. We have pre- viously reported that comparisons of small-subunit ribosomal RNA sequences indicate that the protoctist, Dictyostelium discoideum, represents the deepest divergence in the eukaryotic line of descent yet characterized by molecular phylogeny (McCarroll et al. 1983). In this paper we have expanded our analysis to include the two closely related protoctists, Oxytricha nova and Stylonychia pustulata. In a phylogeny inferred from the small-subunit ribosomal RNA similarities, these ciliates are seen to diverge from the eukaryotic line of descent significantly later than the branching of D. discoideum. Material and Methods Reagents Restriction enzymes, bacterial alkaline phosphatase, DNA polymerase/IUenow fragment, and the DNA synthesis kit were purchased from New England Biolabs. [u- 35S]dATP was purchased from New England Nuclear. DNA ligase was prepared using 1. Key words: small-subunit ribosomal RNA genes, hypotrichous ciliates, evolutionary relationships. Address for correspondence and reprints: Dr. M. L. Sogin, Department of Molecular and Cellular Biology, National Jewish Hospital, 3800 East Colfax, Denver, Colorado 80206. 2. Present address: Biology Department, Indiana University, Bloomington, Indiana 47405. Mol. Biol. Evol. 2(5):399410. 1985. 0 1985 by The University of Chicago. All rights reserved. 07374038/85/0205-0655$02.00 399
Transcript
The Small-Subunit Ribosomal RNA Gene Sequences from the Hypotrichous Ciliates Oxytriclha IWWZ and Stylonychia pustulata’
Hille J. Elwood,* Gary J. 01sen,tT2 and Mitchell L. Sogin* *Department o f Molecular and Cellular Biology, National Jewish Hospital and Research Center; TDepartment of Biochemistry, Biophysics and Genetics, University of Colorado School of Medicine; and $Department of Molecular and Cellular Biology, National Jewish Hospital and Research Center
We have determined the complete nucleotide sequence of the small-subunit ribo- somal RNA genes for the ciliate protozoans Stylonychia pustulata and Oxytricha nova. The sequences are homologous and sufficiently similar that these organisms must be closely related. In a phylogeny inferred from comparisons of several eu- karyotic small-subunit ribosomal RNAs, the divergence of the ciliates from the eukaryotic line of descent is seen to coincide with the radiation of the plants, the animals, and the fungi. This radiation is preceded by the divergence of the slime mold, Dictyostelium discoideum.
Introduction
Eukaryotic microorganisms are classified in either the protoctista or the fungi. In contrast to the fungi, which are a relatively cohesive phylogenetic grouping, the protoctists are a heterogeneous collection of organisms that display enormous physi- ological, cytological, and biochemical diversity (Margulis and Schwartz 1982). Con- sequently, it has been very difficult to infer consistent phylogenies for these simple eukaryotes using classical taxonomic approaches, i.e., comparative studies of pheno- types. As an alternative, similarities between ribosomal RNA sequences can be used to define quantitative phylogenetic relationships for these organisms. We have pre- viously reported that comparisons of small-subunit ribosomal RNA sequences indicate that the protoctist, Dictyostelium discoideum, represents the deepest divergence in the eukaryotic line of descent yet characterized by molecular phylogeny (McCarroll et al. 1983). In this paper we have expanded our analysis to include the two closely related protoctists, Oxytricha nova and Stylonychia pustulata. In a phylogeny inferred from the small-subunit ribosomal RNA similarities, these ciliates are seen to diverge from the eukaryotic line of descent significantly later than the branching of D. discoideum.
Material and Methods Reagents
Restriction enzymes, bacterial alkaline phosphatase, DNA polymerase/IUenow fragment, and the DNA synthesis kit were purchased from New England Biolabs. [u- 35S]dATP was purchased from New England Nuclear. DNA ligase was prepared using
Address for correspondence and reprints: Dr. M. L. Sogin, Department of Molecular and Cellular Biology, National Jewish Hospital, 3800 East Colfax, Denver, Colorado 80206.
2. Present address: Biology Department, Indiana University, Bloomington, Indiana 47405. Mol. Biol. Evol. 2(5):399410. 1985. 0 1985 by The University of Chicago. All rights reserved. 07374038/85/0205-0655$02.00
399
400 Elwood, Olsen, and Sogin
the methods of Panet et al. (1973). The dideoxynucleotides were purchased from Pharmacia P-L Biochemicals.
Preparation of Plasmids Containing Ribosomal RNA Genes
The complete macronuclear ribosomal RNA transcription units (inserted into the PstI site of pBR322) from 0. nova and S. pustulata were provided by M. Swanton (Swanton et al. 1982). The recombinant plasmids were grown in Escherichia coli strain HB 10 1 and amplified in the presence of chloramphenicol. Plasmid DNA was isolated using the SDS-alkali lysis procedures described by Maniatis et al. (1982).
Subcloning of Small-Subunit rRNA Genes
Restriction fragments containing the small-subunit ribosomal RNA genes were isolated from the recombinant plasmids described above. The 3.27~Kb HindIII/HindIII restriction fragment from 0. nova and the 3.12-Kb HindIII/HindIII fragment from S. pustulata were electrophoretically fractionated in 0.75% agarose gels built in E buffer (40 mM Tris-acetate, pH 8.2, 20 mM sodium acetate, and 2 mM EDTA). The regions of the gel containing the DNA fragments, as defined by ethidium bromide staining, were excised and placed in vials containing 5 M NaI. The gels were dissolved by heating at 48 C, and the DNA was absorbed to glass beads as described by Vogelstein and Gillespie (1979). After elution from the beads, the DNA was concentrated by ethanol precipitation and suspended in LT buffer ( 10 mM Tris-HCl, pH 7.5, 10 mM NaCI, and 0.5 mM EDTA). These fragments were cloned into the multiple cloning site of the M 13/mp9 single-stranded phage (Messing 1983).
The phage-cloning vector was prepared by digesting 20 pg of the M 13/mp9 rep- licative form with HindIII. The linearized vector was treated with bacterial alkaline phosphatase and electrophoretically purified on agarose gels. Subsequent to extraction from the agarose gels, 20 ng of the vector plus 60 ng of the gel-purified DNA fragments containing the small-subunit rRNA genes was incubated for 18 h at 10 C with 10 units of DNA ligase in 10 pl of ligation buffer (50 mM Tris-HCl, pH 7.4, 10 mM MgC12, 10 mM dithiothreitol, 1 mM spermidine, 1 mM ATP, and 100 pg/ml bovine serum albumin). The recombinant Ml 3 vectors were used to transform Escherichia coli, strain JM103 (Messing 1983). Noncolored M 13 plaques were selected, and C- tests, as described by Messing (1983), were used to determine the size and orientation of the presumptive rDNA inserts.
Preparation of Primers
Primers for the dideoxynucleotide chain termination sequencing protocols that are complementary to evolutionarily conserved coding and noncoding strands in the small-subunit ribosomal RNA genes were prepared using the phosphotriester protocols (Matteucci and Caruthers 198 1). The deoxyoligonucleotides were purified on 40 X 20 cm X 0.8 mm thick 20% polyacrylamide gels that had been prepared in 8 M urea and 1 X NNB solution (134 mM Tris base, 45 mM boric acid, and 2.5 mM EDTA). After electrophoresis at 40 W and room temperature, the primers were located by UV shad- owing (Hassur and Whitlock 1974) and eluted from the gels with TE buffer ( 10 mM Tris-HCl, pH 7.5, and 0.5 mM EDTA). The eluted primers were bound to a C8 Bond Elut column (Analytichem International) in TE buffer plus 50 mM NH40Ac. After elution with 50% acetonitrile, the primers were lyophilized and suspended in 10 mM Tris-HCl, pH 7.2.
18 S rRNA !3equences From 0. nova and S. pustulata 40 1
Dideoxynucleotide Sequencing
Template DNA was prepared from the recombinant M 13 clones as described by Messing ( 1983). Six nanograms of M 13 primer (New England Biolabs) or the synthetic primers complementary to evolutionarily conserved regions of the ribosomal RNA genes were annealed to 6 pg of template DNA in annealing solution ( 10 mM Tris- HCl, pH 7.2, 10 mM MgC12, 1 mM dithiothreitol) by heating to 65 C for 5 min and slow cooling to room temperature over 30 min. Klenow fragment of DNA polymerase and 30 l&i [a-35S]dATP were added. This mix was distributed to each of five tubes containing the deoxynucleotide triphosphates (dNTPs) plus one dideoxynucleotide triphosphate (ddNTP). The dNTPs were present at a concentration of 0.4 mM, and the ddNTPs as follows: ddA, 0.176 mM; ddG, 0.68 mM; ddT, 1 .O mM; ddC, 1 .O mM; and ddG/dI, 0.35 mM/0.4 mM. (Band-compression artifacts on polyacrylamide se- quencing gels occur with a frequency of l-2 errors/ 100 residues. These artifacts occur primarily in dideoxyguanosine chain termination reactions and often result in erro- neous sequence interpretations. Band compressions are caused by strong secondary- structure interactions that distort gel sieving patterns or effect premature chain ter- mination in the dideoxynucleotide sequencing reactions. The error rate can be reduced to -0.5% by substituting deoxyinosine for deoxyguanosine in an additional dideox- yguanosine chain termination sequencing reaction. Because the stacking interactions of deoxyinosine are weaker than those of deoxyguanosine, the secondary-structure stabilities are altered. The reduced secondary-structure stabilities minimize the band compressions that can be detected by comparing dideoxyguanosine-terminated re- actions containing deoxyinosine with similar reactions containing deoxyguanosine.) After incubation for 20 min at 37 C, a nonradioactive chase mix (1 mM in all dNTPs) plus additional Klenow enzyme was added and incubation was continued for 15 min. The reactions were halted by addition of EDTA to a concentration of 10 mM. The samples were dried under vacuum and then resuspended in 10 ~1 of gel-loading buffer (0.1% xylene cyanol/O. 1% bromphenol blue in formamide). Two microliters of each sample were loaded onto 6% or 8% polyacrylamide sequencing gels (Sanger and Coulson 1975) that had been prepared in 8 M urea with a salt gradient from 2.5 X NNB (bottom) to 0.5 X NNB (top). After electrophoresis at 40 W and room temperature, the gels were soaked for 30 min in 10% methanol/ 10% acetic acid/ 1% glycerol and then vacuum dried onto a sheet of 3-mm paper. The radioactive bands were located by autoradiography using Kodak XL1 film.
Results
Eukaryotic, small-subunit ribosomal RNAs encoded by the nucleus vary in length from 1,77 1 nucleotides in Stylonychia pustulata (present paper) to more than 2,450 nucleotides in Trypanosoma brucei (Hasan et al. 1982; M.L.S. and H.J.E., unpublished data). Comparisons of five eukaryotic and 20 prokaryotic small-subunit ribosomal RNA sequences as well as Ti oligonucleotide catalogues representing more than 200 prokaryotic organisms (Fox et al. 1980) reveal that universal or eukaryote-specific sequences (regions that are conserved among all organisms or among all eukaryotes, respectively) are interspersed among semiconserved sequences (regions of intermediate conservation) and nonconserved sequences (regions that display very high rates of genetic drift). The semiconserved sequences are useful for the construction of quan- titative molecular phylogenies involving distantly related organisms, whereas the non-
402 Elwood, Olsen, and Sogin
conserved regions are valuable for resolving close phylogenetic relationships. The highly conserved regions, because of a lack of sequence variation, do not contribute infor- mation about sequence divergence; however, they are potentially useful for rapidly sequencing small-subunit ribosomal RNA genes.
Sequence Analysis of the Small-Subunit Ribosomal RNA Genes
The dideoxynucleotide chain termination protocols were used to sequence por- tions of the coding and noncoding strands of the S. pustulata and 0. nova small- subunit rDNA genes cloned into the single-stranded phage M 13/mp9. We synthesized 13 oligonucleotides (15-17-mers) that are complementary to coding and noncoding strands of universal and eukaryote-specific regions. These regions, strategically located in all eukaryotic small-subunit ribosomal RNA genes, were used to initiate synthesis in the dideoxynucleotide chain termination sequencing protocols. The eukaryote-spe- cific and universal oligonucleotide primer sequences as well as their locations in eu- karyotic and prokaryotic small-subunit rRNAs (as represented by D. discoideum and Escherichia coli, respectively) are listed in table 1. From a given primer site it was generally possible to determine the sequence of 300-500 nucleotides. The sequencing strategies for the two ciliate small-subunit ribosomal RNA genes presented in this paper are shown in figure 1. Figure 2 displays the small-subunit ribosomal RNA gene sequences from 0. nova and 5’. pustulata aligned with the previously reported small- subunit ribosomal RNA genes from D. discoideum (McCarroll et al. 1983; Ozaki et
Table 1 Synthetic DNA Oligonucleotides Complementary to Conserved Regions in Eukaryotic Small-Subunit Ribosomal RNA Gene Sequences
Eukaryotic Location a Prokaryotic Location b Sequence
NOTE-The locations of phylogenetically conserved sequences in eukaryotic small-subunit ribosomal RNAs were identified in comparisons of five eukaryotic and 20 prokaryotic small-subunit ribosomal RNA sequences as well as from T, oligonucleotide catalogues representing more than 200 prokaryotic organisms.
a Nucleotide positions of the synthetic DNA oligomers in eukaryotic small-subunit ribosomal RNAs as rep resented by Dictyostelium discoideum.
b Analogous nucleotide positions of the synthetic DNA oligomers in prokaryotic small-subunit ribosomal RNAs as represented by E. cob.
’ Synthetic DNA oligonucleotides complementary to evolutionarily conserved regions of the coding strand of eukaryotic small-subunit ribosomal RNA genes.
d Synthetic DNA oligonucleotides complementary to evolutionarily conserved regions of the noncoding strand of eukaryotic small-subunit ribosomal RNA genes.
18 S rRNA Sequences From 0. nova and S. pustdata 403
Base Pairs FIG. I.-Restriction map and strategies used to determine the DNA sequences of the Stylonychia
pustulata and Oxytricha nova small-subunit ribosomal RNA genes. The small-subunit ribosomal RNA coding region for S. pus&data resides within a 3.12-Kb Hind111 restriction fragment (top panel) and that for 0. nova within a 3.27-Kb Hind111 restriction fragment (bottom panel). Synthetic DNA oligomers that are complementary to evolutionarily conserved regions in eukaryotic small-subunit RNAs were used to prime the dideoxynucleotide chain termination sequencing protocols. The arrows indicate the extent of sequence data read from a particular primer site. The location of the primers in eukaryotic small-subunit rRNAs, as represented by Dictyostelium discoideum, are indicated on the arrows, and the analogous positions in the prokaryotic small-subunit rRNAs, as represented by E. coli, are included in parentheses.
al. 1984), Saccharomyces cerevisiae (Rubtsov et al. 1980; Mankin et al. 198 1) and E. coli (Brosius et al. 1978).
Similarity Calculations and Tree Construction
The ribosomal RNA sequences were aligned using a nonrigorous procedure that considers the phylogenetic conservation of both primary- and secondary-structural features (McCarroll et al. 1983). Initially, short subregions of identical or similar pri- mary structure in approximately homologous positions were aligned for the sequences shown in figure 2 and for incomplete or unpublished sequences from Euplotes aedi- culatus, and Paramecium tetraureka (M.L.S., J. Gunderson, and H.J.E., unpublished data). Alignment gaps were placed by eye to juxtapose regions of high similarity in the various sequences. The procedure was repeated in order to detect regions of weaker similarity. The alignments in regions of length variation were further refined by lining up those secondary structures that appear to be evolutionarily conserved in all taxa
FIG. 2.-Sequence of the Stylonychia pustulata and Oxytricha nova small-subunit ribosomal RNA coding regions aligned with other small-subunit rRNAs. The sequences of the S. pustulata and the 0. nova small-subunit rRNAs are shown aligned with those from Dictyostelium discoideum (McCarroll et al. 1983), Saccharomyces cervisiae (Rubtsov et al. 1980; Mankin et al. 198 I), and E. coli (Brosius et al. 1978). The S. pustulata and 0. nova sequences were determined as described under Experimental Procedures using the sequencing strategies depicted in fig. 1. Initially the small-subunit sequences shown as well as those from Xenopus laevis (Salim and Maden 1981), Zea mays (Messing et al. 1984), rat (Chan et al. 1984) rice
whose sequences have been examined. In any alignment procedure, the merits of improved sequence similarity versus the introduction of alignment gaps must be weighed. Our method of alignment is not rigorously defined, but we believe that the positions aligned by this procedure have a higher probability of being in homologous alignment than those aligned by maximizing the number of matching nucleotides. We will employ the word “structural” to distinguish similarity defined using our type
18 S rRNA Sequences From 0. nova and S. pustulata 405
(Takaiwa et al. 1984), Tetrahymena thermophila (Spangler and Blackburn 1985), Paramecium tetraurelia, and Euplotes aediculatus (M.L.S., J. Gunderson, and H.J.E., unpublished data) were aligned according to primary structure. The locations of evolutionarily conserved structures were then used to refine the alignment where length variation occurred. The differences in sequence length were compensated by introducing ap propriate gaps (-) in the sequences. A number system for the aligned sequences, as well as number systems for each sequence, is provided.
of alignment from those using maximal matches. We define structural similarity, s, as
s = m/(m + u + g/2),
where m is the number of sequence positions with matching nucleotides in the two sequences, u is the number of sequence positions with nonmatching nucleotides, and
406 Elwood, Olsen, and Sogin
g is the number of sequence positions that have a gap in one sequence opposite a nucleotide in the other sequence. A special case is the occurrence of large insertions and deletions. These events are likely to be the result of single rare events rather than of the compounding of large numbers of single nucleotide events. Therefore, only the first five sequence gaps in a string of gaps were counted in determining g.
Pairwise comparisons of all homologous nucleotide positions for the small-subunit ribosomal RNA sequences shown in figure 2 as well as those of Zea mays (Messing et al. 1984), rice (Takaiwa et al. 1984), Xenopus Zaevis (Salim and Maden 198 l), Halobacterium volcanii (Gupta et al. 1983), and rat (Chan et al. 1984) were used to compute similarity values. If one treats the structural similarities as representing the fraction of sites that are identical, then h = 1 - s is the fraction of sites that are different and may be used in the formula of Jukes and Cantor (1969) to get a corresponding structural distance expressed in terms of nucleotide substitutions/site. Both the sim- ilarity values and the computed structural distances are presented in table 2.
The structural distances were converted to phylogenetic trees by a variation of the method of Fitch and Margoliash (1967). The evaluation of alternative phylogenetic trees was based on the agreement of the structural distance data separating pairs of organisms and the sum of the tree segment lengths joining the organisms in the tree. The difference between the sum of the tree segment lengths and the structural distance for each pair of organisms was squared. This error was divided by the variance of the structural distance estimate (Olsen 1983).
The variance o2 is defined as
(3= w-0-v n[ 1 - (4/3)h12 ’
where o* is the variance, s is the structural similarity, and n = m + u + g/2 (Kimura and Ohta 1972; Hori and Osawa 1977). The summation of weighted errors for all pairs of organisms is defined as the tree error. For a given tree topology, the tree segment lengths that minimize the tree error were determined. If a topology yielded a negative length segment (negative length segments are mathematical artifacts that have no evolutionary meaning), the tree error was penalized by a factor of 3 for every negative length segment in the tree.
Determining the tree geometry and branch lengths that best fit the structural- similarity data is an optimization problem of considerable magnitude (there are 2 X 1 O6 possible unrooted trees for 10 organisms). Because it is not practical to test all possible trees, we have used an algorithm in which the effects of a given set of rearrangements on a given phylogenetic tree are tested, and then the best of all tested alternatives (i.e., the most improved tree) is maintained and is used as the starting point for another round of optimization. Two simple classes of tree rearrangements are tested by the optimization algorithm (Olsen 1983). Both regard the current tree as sets of subtrees connected by segments. A subtree can range from a single sequence to N - 3 sequences, where N is the number of organisms represented in the tree. A subtree can be moved to a new location by removing its nearest node from the tree and inserting this node into an alternative tree segment. In the first class of rearrange- ments, the effect of moving each possible subtree (one at a time) to every alternative location in the tree is systematically tested. The second class of rearrangements tested involves interchanging the locations of a pair of subtrees. The effect of all possible
Table 2 Structural Similarity and Distance between Small-Subunit Ribosomal RNA Gene Sequences
Z. mays . . . . . . S. cerevisiae . S. pustulata . . . 0. nova . . D. discoideum . . . H. volcanii . . E. coli . . . . . .
0.044
0.240
0.247
0.265
0.316
0.316
0.425
0.715
0.845
0.958
0.236
0.240
0.263
0.313
0.312
0.398
0.706
0.843
0.794
0.797
0.023
0.207
0.225
0.224
0.350
0.687
0.845
0.790
0.795
0.977
0.206
0.22 1
0.223
0.350
0.682
0.840
0.777
0.778
0.819
0.820
0.209
0.208
0.337
0.684
0.820
0.742
0.744
0.805
0.809
0.817
0.015
0.328
0.666
0.780
0.742
0.745
0.806
0.807
0.818
0.985
0.324
0.670
0.774
0.676
0.69 1
0.720
0.720
0.729
0.734
0.737
0.696
0.815
0.539
0.542
0.550
0.552
0.55 1
0.559
0.557
0.547
0.56 1
0.493
0.494
0.493
0.495
0.501
0.515
0.517
0.503
0.605
NOTE.-The upper-right half of the table gives s values for all pairs of aligned small-subunit rRNA sequences. If 1 - s is considered to be the fraction of sites that are identical, the formula of Jukes and Cantor ( 1969) can be used to compute the structural distances (average number of base changes per sequence position), which are shown in the lower-left half of the table.
’ Sequence data from rat (Chan et al. 1984), X. fuevis (Salim and Maden 198 l), rice (Takaiwa et al. 1984), 2. muys (Messing et al. 1984), S. cerevisiue (Rubtsov et al. 1980), D. discodieum (McCarroll et al. 1983), H. volcunii (Gupta et al. 1983), and E. coli (Brosius et al. 1978).
408 Elwood, Olsen, and Sogin
pairwise interchanges of subtrees (one pair at a time) is tested. The rearrangement that leads to the most improved tree is used as the starting point for a new round of optimization. For 10 organisms there are on the order of 250 independent trees per round of optimization. The number of rounds of optimization required for a convergent solution depends on the topology of the initial tree. Computer programs for calculating similarities and structural distances and for implementing our phylogenetic-tree eval- uation and optimization algorithm have been written in FORTRAN for execution on the Digital VAX 1 l/750. The program is useful for trees with less than 30 organisms. It appears to have successfully found the optimum of all sequence sets tested (as evaluated by the “optimal” tree being independent of the initial tree).
The computer-assisted optimization algorithm described above was used to infer the phylogenetic tree shown in figure 3 from the similarity and structural distance data presented in table 2. This optimized tree (tree error = 0.30) is consistent with S. pustulata and 0. nova being closely related but places the divergence of D. discoideum from the eukaryotic line of descent prior to the branching of these ciliates. The tree with the second-lowest error (tree error = 0.32) was constructed by interchanging the branching order of S. cervisiae and the ciliates. We interpret the slight difference in tree errors and the small negative segment (-0.01) between the cilate and the fungal branchings in the second-best tree to mean that the branching order for these two groups is not statistically significant.
Discussion
Ciliates are unicellular heterokaryotic organisms that have cilia at some point in their life cycle. Classification schemes for these organisms are generally based on char- acterization of ciliature and infraciliature. Corliss (1979) has proposed a taxonomy that includes 23 orders divided among three major classes. The hypotrichous ciliates Stylonychia pustulata and Oxytricha nova are members of the Polyhymenophora and are considered to be members of the suborder Sporadotrichina. As shown in our inferred phylogeny (fig. 3), the S. pustulata/O. nova ribosomal RNA distance is con-
-
FIG. tree was i Results. The evolutionary distance between nodes of the tree is given alongside them and is represented in the horizontal component of their separation.
the segment connecting
18 S rRNA Sequences From 0. nova and S, pus&data 409
&tent with this placement and is comparable to the divergence between rice and Zea mays.
As in the case of Dictyostelium discoideum, the ciliophora are taxonomically treated as members of the protoctista and are considered to be a very ancient phylum. Our phylogeny indicates that the emergence of the ciliophora was preceded by the rhizopodea as represented by D. discoideum, thus supporting the notion that the pro- toctista are not monophyletic. In fact, based on the structural distances of ribosomal RNA shown in table 2, the evolutionary distance between these two major protoctistan groups is comparable to the evolutionary distance between plants and D. discoideum or to that between animals and D. discoideum. We have previously argued that the large distance between the rRNA of D. discoideum and published sequences from other eukaryotes represents an early branching in the eukaryotic line of descent rather than an unusually high rate of genetic drift (fast evolutionary-clock speed) or convergent evolution in the rRNAs of other eukaryotes (see McCarroll et al. 1983).
Finally, the phylogeny shown in figure 3 suggests that the early branching of D. discoideum was followed by a radiative period that gave rise to the animals, the plants, the fungi, and the ciliates. It will be of interest to characterize other representatives of the protoctista to determine whether major phylogenetic groups radiated from the eukaryotic line of descent at a similar time.
Acknowledgements
This research was supported by grant GM32964 from the National Institutes of Health.
LITERATURE CITED
BROSIUS, J., M. L. PALMER, P. J. KENNEDY, and H. F. NOLLER sequence of a 16s ribosomal RNA gene from Escherichia coli 75:4801-4805.
197 8. Complete nucleotide Proc. Natl. Acad. Sci. USA
CHAN, Y., R. GUTELL, H. F. NOLLER, and I. G. WOOL. 1984. The nucleotide sequence of a rat 18s ribosomal ribonucleic acid gene and a proposal for the secondary structure of 18s ribosomal ribonucleic acid. J. Biol. Chem. 259:224-230.
CORLISS, J. 0. 1979. The ciliated protozoa: characterization, classification, and guide to literature, 2nd ed. Pergamon, New York. 455 pp.
FITCH, W. M., and E. MARGOLIASH. 1967. Construction of phylogenetic trees. Science 155: 279-284.
Fox, G. E., E. STACKEBRANDT, R. B. HESPELL, J. GIBSON, J. MANILOFF, T. A. DYER, R. S. WOLFE, W. E. BALCH, R. S. TANNER, L. J. MAGRUM, L. B. ZABLEN, R. BLAKEMORE, R. GUFTA, L. BONEN, B. J. LEWIS, D. A. STAHL, K. R. LEUHRSEN, K. N. CHEN, and C. R. WOESE. 1980. The phylogeny of prokaryotes. Science 209:457-463.
GUPTA, R., J. M. LANTER, and C. R. WOESE. 1983. Sequence of the 16s ribosomal RNA from Halobacterium volcanii, an archaebacterium. Science 221:656-659.
HASAN, G., J. J. TURNER, and J. S. CORDINGLEY. 1982. Ribosomal RNA genes of Trypanasoma brucei: cloning of a rRNA gene containing a mobile element. Nucleic Acids Res. 10:6747- 6761.
HASSUR, S. M., and H. W. WHITLOCK. 1974. UV shadowing-a new and convenient method for the location of ultraviolet-absorbing species in polyacrylamide gels. Anal. Biochem. 59: 162-164.
HORI, H., and S. OSAWA. 1979. Evolutionary change in 5S RNA secondary structure and a phylogenic tree of 54 5S RNA species. Proc. Natl. Acad. Sci. USA 76:381-385.
410 Elwood, Olsen, and Sogin
JUKES, T. H., and C. R. CANTOR. 1969. Evolution of protein molecules. Pp. 2 l-l 32 in H. N. MUNRO, ed. Mammalian protein metabolism. Academic Press, New York.
KIMURA, M., and T. OHTA. 1973. On the stochastic model for estimation of mutational distances between homologous proteins. J. Mol. Evol. 2:87-90.
MANIATIS, T., E. F. FRITSCH, and J. SAMBROOK. 1982. Large-scale isolation of plasmid DNA. Pp. 86-94 in Molecular cloning: a laboratory manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
MANKIN, A. S., A. M. KOPYLOV, and A. A. BOGDANOV. 198 1. Modification of 18s rRNA in the 40s ribosomal subunit of yeast with dimethylsulfate. FEBS Lett. 134:11-14.
MARGULIS, L., and K. N. SCHWARTZ. 1982. Protoctista. P. 7 1 in P. BREWER, ed. Five kingdoms. W. H. Freeman, New York.
MATTEUCCI, M. D., and M. H. CARUTHERS. 198 1. Synthesis of deoxyoligonucleotides on a polymer support. J. Am. Chem. Sot. 103:3 185-3 19 1.
MCCARROLL, R., G. J. OLSEN, Y. D. STAHL, C. R. WOESE, and M. L. SOGIN. 1983. Nucleotide sequence of Dictyostelium discoideum small-subunit ribosomal ribonucleic acid inferred from the gene sequence: evolutionary implications. Biochemistry 22:5858-5868.
MESSING, J. 1983. New M 13 vectors for cloning. Pp. 20-78 in R. WV, L. GROSSMAN, and K. MOLDAVE, eds. Methods in enzymology: recombinant DNA. Part C. Academic Press, New York.
MESSING, J., J. CARLSON, G. HAGEN, I. RUBENSTEIN, and A. OLESON. 1984. Cloning and sequencing of the ribosomal RNA genes in maize: the 17s region. DNA 3:3 l-40.
OLSEN, G. J. 1983. Comparative analysis of nucleotide sequence data. Ph.D. thesis. University of Colorado, Boulder.
OZAKI, T., Y. KOSHIKAWA, Y. IIDA, and M. IWABUCHI. 1984. Sequence analysis of the transcribed and 5’ non-transcribed regions of the ribosomal RNA gene in Dictyostelium discoideum. Nucleic Acids Res. 12:4 17 l-4 184.
PANET, A., J. H. VAN DE SANDE, P. C. LOEWEN, H. G. KHORANA, A. J. RAAE, J. R. LILLEHAUG, and K. KLEPPE. 1973. Physical characterization and simultaneous purification of bacterio- phage T4 induced polynucleotide kinase, polynucleotide ligase, and deoxyribonucleic acid polymerase. Biochemistry 12:5045-5050.
RUBTSOV, P. M., M. M. MUSAKHANOV, V. M. ZAKHARYEV, A. S. KRAYEV, K. G. SKRYABIN, and A. A. BAYEV. 1980. The structure of the yeast ribosomal RNA genes. I. The complete nucleotide sequence of the 18s ribosomal RNA gene from Saccharomyces cerevisiae. Nucleic Acids Res. 8:5779-5794.
SALIM, M., and B. E. H. MADEN. 198 1. Nucleotide sequence of Xenopus laevis 18s ribosomal RNA inferred from gene sequence. Nature 291:205-208.
SANGER, F., and A. R. COULSON. 1975. A rapid method for determining sequences in DNA by primed synthesis with DNA polymerase. J. Mol. Biol. 94:441-448.
SPANGLER, E. A., and E. H. BLACKBURN. The nucleotide sequence of the 17s ribosomal RNA gene of Tetrahymena thermophila and the identification of point mutations resulting in resistance to the antibiotics paromomycin and hygromycin. J. Biol. Chem. 260:6334-6340.
SWANTON, M. T., R. M. MCCARROLL, and B. B. SPEAR. 1982. The organization of macronuclear rDNA molecules of four hypotrichous ciliated protozoans. Chromosoma 85: l-9.
TAKAIWA, F., K. 00~0, and M. SUGIURA. 1984. The complete nucleotide sequence of a rice 17s rRNA gene. Nucleic Acids Res. 12:5441-5448.
VOGELSTEIN, B., and D. GILLESPIE. 1979. Preparative and analytic purification of DNA from agarose. Proc. Natl. Acad. Sci. USA 76:6 15-6 19.
WALTER M. FITCH, reviewing editor
Received January 23, 1985; revision received May 14, 1985.
