Proc. Nat. Acad. Sci. USAVol. 72, No. 6, pp. 2418-2422, June 1975

Phylogenetic Origin of the Chloroplast and Prokaryotic Nature ofIts Ribosomal RNA

(Euglena/16S rRNA/evolution)

L. B. ZABLEN*T, M. S. KISSILt, C. R. WOESEtT*, AND D. E. BUETOWtDepartments of t Physiology and Biophysics, $ Genetics and Development, and * Microbiology, University of Illinois, Urbana, Ill. 61801

Communicated by C. Ladd Prosser, March 28, 1976

ABSTRACT The 16S ribosomal RNA of the Euglenagracilis chloroplast has been characterized in terms of itstwo-dimensional electrophoretic "fingerprint" (TI ribo-nuclease). Results show it to be a typically prokaryotic16S rRNA. By the present criterion, different chloroplastsare shown to be related to one another and at least dis-tantly to blue-green algae and perhaps to Bacillaceae.These results argue in favor of an endosymbiont origin ofthe chloroplast.

Evidence suggests that various intracellular eukaryoticorganelles evolved from endosymbiotic prokaryotes (1). Forchloroplasts, similarities with prokaryotes are found forphotosynthetic membranes (2, 3), ribosomes (4), and path-ways of CO2 fixation and photoelectron flow (5, 6). Also,symbiosis between blue-green algae and a variety of eukaryoticcells occurs frequently (7). Thus, the most likely endo-symbionts postulated to give rise to the chloroplast are theCyanophytes (1, 7). Although there is much circumstantialevidence for both the general and the specific conjectures con-cerning endosymbiosis, a definitive molecular support for anyendosymbiotic origin of eukaryotic organelles has yet to befurnished.

Phylogenetic relationships among prokaryotes can beestablished by means of primary structural characterizationsof the larger ribosomal RNAs (8, 9). Thus, if the chloroplastarose from an ancestral prokaryote, the chloroplast ribosomeshould carry this organelle's pedigree. A primary structuralcharacterization of the chloroplast 16S rRNA should thentest the validity of the prokaryote-endosymbiont hypothesis.To date, more than 30 prokaryotic 16S rRNAs and eukaryotic18S rRNAs have been characterized in terms of their (T1ribonuclease) oligomer "fingerprints" (refs. 8-11; Woese et al.,unpublished). These rRNAs exhibit primary structural fea-tures common to all and/or common to various subgroups,but show little, if any, resemblance to the eukaryotic 18SrRNAs. Against this molecular background, then, it is possibleto make reasonably definitive conjectures about the originof the chloroplast.We present here the results of a "fingerprint" analysis of the

chloroplast 16S rRNA of the unicellular flagellate, Euglenagracilis. This organism was chosen since chloroplasts andchloroplast ribosomes can be isolated from it free of significantcontamination by ribosomes from any other cellular compart-ment (12, 13). Thus, any confusion with the mitochondrial16S rRNA, for example, is eliminated.

MATERIALS AND METHODS

Organism and Growth Conditions. E. gracilis, strain Z, wasgrown in 100-125 ml of defined medium with ethanol ascarbon source (12) in the dark at 27° with limiting phosphateconcentrations (15). At densities of 1 to 1.5 X 106 cells per

ml, cultures were supplemented with 10-25 mCi of carrier-free 32Pi and incubated in the light for 87-90 hr at 7.2 Jm-2 sec-1.

Isolation of Chloroplasts. Chloroplasts were isolated asdescribed (13). All buffers were supplemented with heparin,rat liver RNase-inhibitor, and 2-mercaptoethanol to inhibitRNase activity (12). Chloroplasts were lysed with 0.5%sodium deoxycholate, 0.5% sodium dodecyl sulfate, and 1%triisopropyl-naphthalene sulfonate.

Isolation of 16S Chloroplast rRNA. Total chloroplast RNAwas obtained by two extractions of lysed chloroplasts withwater-saturated redistilled phenol (16), layered on poly-acrylamide gels, and electrophoresed for 5-6 hr at 10 mA pertube (9, 17, 18). RNA was extracted from gels by phenol (19)and purified by chromatography on a (Whatman) CF11cellulose column (Zablen, Ph.D. dissertation, University ofIllinois, 1975).

Sequence Characterization of 16S rRNA. T1 RNase digests of16S rRNA were "fingerprinted" by the method of Sanger andcoworkers (20); modified by Uchida et al. (21) and Sogin andWoese (unpublished results). Sequence of oligomers in thevarious spots was determined by the modified method ofUchida et al. (21). Oligomers were quantitated by excisingand counting each spot on the primary fingerprint (21).

RESULTS AND DISCUSSION

Phylogenetic considerations

The Euglena chloroplast 16S ribosomal RNA yields theelectrophoretic fingerprint shown in Fig. 1. A comparison ofthese oligonucleotide sequences to those from typical pro-karyotic 16S rRNAs (Table 1) permits several conclusions:

(i) Chloroplast rRNA, and so by inference the chloroplastitself, is of prokaryotic origin. The large number of coincidentsequences between the chloroplast catalog and those of typicalprokaryotes can be given no other reasonable interpretation.Similar characterizations of the cytoplasmic 18S rRNAs from

yeast (Sogin, Sogin, Zablen, and Woese, unpublished), mouse

L-cells (Pace, Sogin, Bonen, and Woese, unpublished), and

Porphyridium (14) show negligible resemblance to prokaryotic16S rRNAs.

(ii) The chloroplast 16S rRNA from Euglena and the 16SRNA from Porphyridium show a higher level of oligomer co-

incidence than any other pairing of organisms in Table 1.

Three-fourths (23 of 31) of the pentamers found in the Euglenachloroplast occur in Porphyridium-four of these beingfound in no other organisms in the Table. No other pairwisecomparison of organisms in Table 1 yields more than 17

pentamer coincidences, and 10 of these are found in every

organism. Furthermore, 11 oligomers (hexamers and larger)

2418

Dow

nloa

ded

by g

uest

on

Aug

ust 3

0, 2

020

Phylogenetic Origin of Chloroplast 2419

are found only in the pair Euglena and Porphyridium or onlyin these two and in Anacystis. This is comparable to what isfound in the genus Bacillus, for example. In other words, in atypical pairwise comparison of Bacilli, approximately 70% ofthe pentamers are common. Also, among the larger oligomers,about 15 coincidences are found whose sequences are uniqueto the Bacilli or unique to the Bacilli and the Clostridia(Woese et al., unpublished). By contrast, no other pair oforganisms in Table 1 shows statistically significant levels ofthese sorts of (hexamer and larger) coincidences, exceptpossibly Rhodopseudomonas and the Vibrio-enteric group(four such), previously suggested to be distantly related forthis reason (9). It is reasonable to conclude, therefore, thatthe two examples of chloroplasts so far characterized do con-stitute the equivalent of a prokaryotic Genus, or perhaps, aFamily.

(iii) The postulated relationship between the chloroplastand the blue-green algae (represented here by Anacystisnidulans) can also be seen. As Table 1 shows, four oligomers(hexamer or larger) are confined solely to these rRNAs.Another two (two) are found only in Anacystis and the chloro-plast from Euglena (Porphyridium). Our results, together withthose of Bonen and Doolittle (14), indicate that the postulatedcommon ancestor of (extant) blue-green algae and chloroplastswould have predated the common ancestor of the chloro-plasts themselves. A larger sampling, however, of algae andchloroplasts is necessary to make this statement definitively,particularly in view of the results of Piggott and Carr (22).The present techniques of sequence comparison have been

used previously to suggest that blue-green algae and theBacillaceae are more closely related to one another than eithergroup is to many of the other bacteria (9). This conclusionwas based not only upon oligomer sequence commonality,but also upon a particular pattern of base modification foundso far only in Bacillaceae and the blue-green algae (9). § TheEuglena chloroplast, at least, shares this characteristic modi-

*

fication pattern; it contains the sequence U-C-A-C-A-C--U-

C-A G, found previously only in Bacillaceae and Anacystis-C-

(refs. 9 and 11; Woese et al., unpublished), and it lacks the* *

modified sequence A-A-C-C-U-G characteristic of the Vibrio-enteric group, Rhodopseudomonas, and many other gram-negative organisms (refs. 8, 9, 10, and 21; Zablen, Ph.D.dissertation, University of Illinois, 1975; Woese et al., unpub-lished results).

§ Anacystis and the Bacilli share five large (unmodified) oli-gomers not found in Rhodopseudomonas or the Vibrio-entericgroup-A-A-U-U-A-U-U-G, A-A-A-C-U-C-A-A-A-G, A-A-C-C-U-U-A-C-C-A-G, A-C-A-A-A-C-C-G, and A-U-C-C-U-G (ref-erences 9 and 14; Woese et al.,. unpublished). The first three ofthese are found in the Euglena chloroplast, while the last fourare found in Porphyridium 16S RNA. Porphyridium addition-ally contains two sequences, C-A-U-U-A-G and U-A-A-C-A-C-G,found only in (all) Bacilli (ref. 14; Woese et al., unpublished).To recapitulate the above, for sequences hexamers and larger,there are seven found only in the two chloroplasts, four foundonly in both chloroplasts and Anacystis (eight occur only in one

of the chloroplasts and Aniacystis), and two are found only in theBacilli and all three members of the Anacystis-chloroplastgroup (seven additional oligomers occur only in Bacilli and one

or two members of that group).

_ ,_ _pH 3.50 * 43,W

eS vv %, 6 &

S.so a 40 3'

S J6 0

9I6 i%1b*

*5

~~ a

16S rRNACHLOROPLAST 4Eug/ena a

0

FIG. 1. Two-dimensional electrophoretic fingerprint of aTi ribonuclease digest of chloroplast 16S ribosomal RNA. Post-transcriptionally modified oligomers are indicated by arrows,as are the 3' and 5' termini.

The chloroplast ribosome

Because of their obligate intracellular status, organelles arenot self-replicating entities in the strict sense of the term.It is open to question, therefore, whether, both in kind andnumber, the constraints operative on organelle ribosomes areequivalent to those operating on the ribosomes of free-livingprokaryotes. Functional chimeras can be formed betweenchloroplast and Escherichia coli 30S and 50S ribosomal sub-units, indicating similarity between the two (23).Here we can add two more points of similarity: (i) The

chloroplast has retained the normal prokaryotic capacity tomodify, post-transcriptionally, particular sequences. Althoughno specific roles have been assigned to these highly conservedsequences, it appears that they function in general in subtleways in translation, a subtlety the chloroplast now can besaid to share with its free-living counterparts (24, 25).

(ii) Prokaryotic 16S rRNA appears to contain nine regionsof high sequence conservation, six rather regularly spaced inthe 3' half and three in the 5' half of the molecule (26).Fig. 2 shows the "sequence conservation map" for E. coli16S rRNA, upon which are placed those chloroplast oligomerscommon to the E. coli rRNA catalog. The majority of thesecommon oligomers coincide with regions of high sequenceconservation. Also, all such regions are well represented in theEuglena chloroplast 16S rRNA with the exception of regionno. 6, which shows only one of its three characteristic oli-gomers. In the Porphyridium 16S RNA region, no. 6 is not

Proc. Nat. Acad. Sci. USA 72 (1975)

Dow

nloa

ded

by g

uest

on

Aug

ust 3

0, 2

020

2420 Biochemistry: Zablen et al.

TABLE 1. Catalog of oligonucleotides produced by Ti ribonuclease digestion of Eugiena chioroplast 16S ribosomal RNA

-D

A B C D E F G Comments

neoZ

0~~~~~

A BOCDE F G Comments

Modified oligomers

GCCG*

OCCCG*

CAACG* *

AACCUG*

UAACAAG*

UCACACCAUG

*

UCACACCACG*

UCACACCACG

PentamersCAACGAACAGAAACGAAAAGUCCCGCCUAGCUCAGCAUCGCACUGUCAAGCAAUGAUCAGAAUCGUAAAGAAAUGCCUUGUACUGUUAAGACUUGAUUAG

AAUUGUCUUGAUUUG

1 1 1

1 1 1

0 1/2 1

0 0 0

1 1 1

1 ? 1

0

(a

1

1

1

1

1

0

0 0 0 + 0

0 0 O 0 1

I 1 1 +1 1 0 01 1 1 01 1 0 01 1 1 +1 0 0 01 1 1 +1 1 0 01 0 0 02 lt 1 01 1 1 p1 1 1 +1 1 1 +2 2 1-2 +3 3 1 +1 1 1 +1 1 0 0

3-2 2 2 +1-2 0 0 02 2 2 +

10001011010222010102b

0

0 U, no. 75

0 U, no. 205

0 U, no. 159

no. 221

0 U, no. 217

0 a Present inClostridia

0 0

+ 0

+

0+0+000+

+++++++

++0++

1 11+ 2 +

1 0 0 + 0 p2 0 0 0 0 0

no. 207

b Includes*

AUUAG

HexamersCACAAGACACAG

UCCACGCCUAAGCAAUAGUCAAAGUAAACGAAUACGAAACUGAAUAAGAAAAUG

UUCCCGCAUCUGUAAUCGAUCCUGCUAAUGAUCUAGUUAAAGAAUCUG

CCUUUGUUCUCG

UAUUUG

HeptamersCAAACAGCAACUCGAAACUCGUAUCCCGCUAUACGUAAUACGUUAAAAGUUAUCCGUAUCUAGAUCCUUGUAAUAUGAAUUUCGAUCUAUGUUUAAUG

1 1 1 + 1 + 0 U, no. 1311 0 0 0 0 0 0

0 0 0 + 1 + 0 U, no. 1111 it O 0 0 0 01 0 0 0 0 0 01 0 1 0 0 0 00 1 1 + 1 + 0 U, no. 1131 0 1 + 1 + 0 U, no. 1970 0 1 + 1 + 1 U,no.151 1 1 0 0 0 11 0 0 0 0 0 0

0 1 1 + 1 + 0 U, no. 1991 0 0 0 0 0 01 1 1 + 1 + 0 U,no.1910 1 1 + 1 0 1 U

1 0 0 0 0 0 01 0 0 0 1 0 01 1 0 0 0 0 01 1 1 0 0 0 0

1 0 0 0 1 0 11 0 0 0 0 0 0

1 0 0 0 0 0 0

1 0 0 + 1 + O U0 0 1 + 1 + 0 U, no. 1851 1 0 0 0 0 01 1 0 0 0 0 0ito 0 0 0 0 01 1 1 + 1 + 0 U, no. 771 0 0 0 0 0 01 1 1 0 0 0 01 0 0 0 0 0 01 0 0 0 0 0 01 0 0 0 0 0 01 0 0 0 0 0 01 0 0 0 0 0 01 0 0 0 0 0 0

representedrepresented.

at all, and region no. 3 may also be sparsely

In two respects, however, chloroplast 16S rRNA may notbe typical of extant prokaryotes. All free-living prokaryotesexamined exhibit a high degree of sequence homology near

the 3' terminus of the 16S rRNA. Almost all have the se-

quence A-U-C-A-C-C-U... followed by two-four uridylateand several cytidylate residues before termination in the finalAOH (refs. 21 and 27-29; Woese et al., unpublished). It hasbeen suggested that this area of the molecule functions ininitiation and/or termination of mRNA translation (throughbase pairing with specific sequences in the latter) (30). TheEuglena chloroplast 16S rRNA terminates in the sequence

A-A-C-A-A-C-U-C-NoH, which is only vaguely reminiscent

of the customary prokaryotic termination sequence. If themRNA-recognition hypothesis is correct, it is possible thenthat the Euglena chloroplast ribosome is designed to recognizea different mRNA "index" and, so, a restricted class ofmRNAs.A surprising feature of the chloroplast rRNAs is that each

lacks a relatively large number of those oligomers (Ti ribo-nuclease) considered to be very highly conserved, or "uni-versal" (9, 26). Any given prokaryotic Family may lack a fewof these universal sequences, and if so, the missing sequencesare the same throughout that Family (Zablen, Ph.D. disserta-tion, University of Illinois, 1975; Woese et al., unpublished).The two chloroplasts each lack nine such oligomers. Yet onlyfour of these nine are missing in both chloroplasts. Also, the

Proc. Nat. Acad. Sci. USA 72 (1975)

Dow

nloa

ded

by g

uest

on

Aug

ust 3

0, 2

020

Phylogenetic Origin of Chloroplast 2421

TABLE 1. (continued)

: o

II Q ~~~00

A B C D E F G

to

31 OI I 00

v42 -OD m 1: - 400.AE

D~~~~-1 G:C

OctamersAACACCAGCCACACUGCAAUACCGAAUAACAGACAAUAAGCCUACAUGUUAACACGAUACCCUGCUACAAUGAAAUACUGAUACUAAGUACCUUCGUCAUCAUGUCAUUUAGAAUCUUUGUAAUUUAGAAUUAUUG

NonamersCUACACACGUACACACCGCCAAUACUGCUUACCAAGACUCCUACGCUAAUUCCGACUUCUACG

0 it 1 +1 it 1 +11 0 01 0 0 01 0 0 01 1 0 01 0 0 01 0 0 +1 1 0 +1 0 0 01 0 0 01 0 0 01 0 1 01 0 0 0ito o 01 0 0 01 0 1 1

11000001100000000

1? 1+ 111 1+ 110 0 0 010 0 0 001 1+ 110 1 0 010 0 0 0

0+00000

++000+0000

++00+00

00000

00000000000

UU, no. 43

U, no. 109U, no. 177

U, no. 175

U, no. 187

Nonamers (cont'd)AAUUUCCAGUAAACUAUGUUAUUACCGAAUUUUCCGUUUAAUUCG

Decamers and largerAACACCAAUGAAACUCAAAGAACCUUACCAGAAAUACCUAAGAAAUUACUAGCUUAACACAUGCAUUAAAAACUGCUUAACUUUGCACUUUUCUAGCCCCWUAU-AUCCUG

UUAUUAUCUUGUncharacterizedTermini

5' pUG3' AUCACCU-

[U2-4CXJAOHAACAACUCNoH

10 0 0 0 010 0 0 0 010 0 0 0 011 0 0 0 010 1+ 1 +

1 001 111 it 1

1 001 001 it 11 001 001 001 00

1

4

1

0

1

0 0 0

0 0

1 01 00 0

0 0

0 0

0 0

0 0

0 0

0 0

0

0

0

0

0

0

0

0

0

0

0

0

1 + 0 +0 00 0

Column A lists the sequences and approximate molar occurrence for all oligomers, pentamers and larger, present in Fig. 1. The se-

quences of four large oligonucleotides containing >15 nucleotides including > 5 uridylate residues are as yet undetermined and so notincluded. An asterisk over a base indicates post-transcriptional modification. Hyphens are omitted for brevity.Column B indicates whether a given oligomer is found in Porphyridium 16S RNA (14); column C, the same for Anacystis nidulans

(11); column E, the same for Rhodopseudomonas spheroides (9). Columns D and F indicate the presence of any given oligomer in theBacilli or the Vibro-enteric group, respectively (Woese et al., unpublished; Zablen, Ph.D. thesis Univ. of Illinois, 1975; refs. 8, 10, and 21).In these cases + (+ +) indicates that one (two or more) copies of a given oligomer is in 80-100% of the organisms in the group; "p"indicates its presence in 50-80% of the organisms; "0" indicates its presence in less than 50% of the organisms in the group. ColumnG indicates the presence (+) or absence (0) of an oligomer in the 18S rRNA of yeast (Sogin, Sogin, Zablen, & Woese, unpublished). The"Comments" column indicates by "U" that an oligomer is "universal" in prokaryotes (9, 26). Numbers indicate the position of an

oligomer of that sequence in the overall E. coli 16S rRNA sequence (26)-see Fig. 2, abscissa.Space requirements of this journal do not permit publication of the data from which these sequences were derived.t Sequence not completely certain, but probable.

Euglena chloroplast contains three of the four "universal"oligomers missing in Anacystis (11).

It is not possible to give a simple phylogenetic explanationof this phenomenon. It is our opinion that it may indicatethat some of the constraints operative on the free-living pro-

karyotic ribosome are not operative on its chloroplast counter-part, thus causing or permitting otherwise "universal"sequences to change during evolution.

General Conclusion. The present data permit then, thesegeneral conclusions: The chloroplast 16S rRNA is clearlyrelated structurally and, therefore, functionally, to pro-

karyotic 16S rRNA. Chloroplasts in a sense constitute a

Genus or Family within the prokaryotes, and are related to

the blue-green algae, perhaps also to the Bacillaceae. Thedata argue in favor of an endosymbiont origin for chloro-plasts. The questions of when, how, or how often an archetypechloroplast formed an endosymbiotic relationship with "eu-karyotic cytoplasm" remain unanswered. These questions can

be approached, however, given comprehensive phylogenies ofchloroplasts and blue-green algae on the one hand, and chloro-plast-bearing cytoplasms and related organisms on the other.

Note Added In Proof. Bonen and Doolittle (14) conclude thatthe two chloroplasts are not, as we claim, phylogenetically closerto one another than to all the other above organisms. Using themethod of analysis, one is almost tempted to agree, for the twochloroplasts have in common only a few more large oligomers than

U, no. 137

Proc. Nat. Acad. Sci. USA 72 (1975)

Dow

nloa

ded

by g

uest

on

Aug

ust 3

0, 2

020

2422 Biochemistry: Zablen et al.

r9-. F8- 7-, 6- 5 - - 323-

5' 3

281

26ia)(i 24,

22

01

hil!.ii

toil Al.II

1.

.I

I.1

h'Ii I

-H1i

rIi.1i ick

tULL i~87 93II10

Reiao,ve OlIgorrer Posit on, n 6 S rR.N D mory Structure

FIG. 2. Distribution of chloroplast oligomers on the E. coli 16S ribosomal RNA "sequence conservation map" (26). Distance on theabscissa is proportional to position in the E. coli 16S rRNA sequence as determined by Fellner and coworkers (26, 31). Each bar rep-resents a (T1 ribonuclease) oligomer found in E. coli 16S rRNA. Its position on the graph corresponds to that oligomer's position in theE. coli 16S rRNA sequence (26). Its width is proportional to the oligomer's size, and its height (the ordinate) is determined by the number,of different 16S rRNAs that have been shown to contain that oligomer (26). The maximum height, i.e., the number of 16SrRNAs characterized, is '27 units. Numbers on the abscissa do not signify distance, but identify various oligomers (26). Theprecise midpoint is located at oligomer no. 109. Black bars correspond to E. coli oligomers, hexamer and larger, and post-transcriptionally modified oligomers found in both Euglena chloroplast and Porphyridium 16S rRNAs; dotted bars represent thepentamers found in both Euglena and Porphyridium 16S rRNA. Bars diagonally hatched or cross-hatched correspond to E. coli oligomers,hexamer and larger, found in either Euglena chloroplast or Porphyridium 16S rRNA, respectively, but not both. The nine regions of highsequence conservation identified previously (26) are bracketed and numbered (above each).

either one of them (particularly Porphyridiurt) has in commonwith Anacystis or the Bacilli. However, one must distinguishbetween characteristics derived from the ancestor of an entiregroup and characteristics evolved during a particular line ofdescent. In the present instance, the widely distributed, "uni-versal," oligomers tend to be of the former type, and oligomersconfined to small groupings of organisms tend to be of the latter.Whereas absence of an otherwise universal oligomer in a par-ticular line of descent could have little phylogenetic significance,the presence of a significant number of large oligomers uniqueto a small group of organisms must have strong phylogeneticsignificance. Given that at least 11 (7) pentamers (hexamers) orlarger are unique to the two chloroplasts and neither chloroplastshares even half this number of (unique) oligomers when pairedwith any other of the above organisms, it is difficult not to con-clude that the two chloroplasts constitute a distinct line ofdescent-i.e., are, as we have stated, (relatively) closely relatedto one another. In any case, it must be recognized that thisproblem is separate from that of single compared to multipleorigin(s) of the chloroplast, and that sufficient data do not existat this time to decide the latter issue.

This study was supported by NSF Grants (GB 41928 to D.B.and GB 16298 to C.W.), NIH Grant (GM 19641 to D.B.),and NASA Grant (NS6-7044 to C.W.). M.S.K. is an NIH pre-doctoral trainee (GM 941). We thank L. Bonen and F. Doolittlefor exchanging unpublished information and C. Hershberger forhelpful discussion of labeling techniques.

1. Margulis, L. (1970) Origin of Eukaryotic Cells (Yale Univ.Press, New Haven), 349 p.

2. Bell, P. R. (1970) Symp. Soc. Exp. Biol. 24, 109-128.3. Arntzen, C. J. & Briantis, J. M. (1975) in Bioenergetics of

Photosynthesis, ed. Govindjee (Academic Press, New York),pp. 51-113.

4. Loening, U. E. & Ingle, J. (1967) Nature 215, 363-367.5. Taylor, D. L. (1970) Int. Rev. Cytol. 27, 29-64.6. Norris, L., Norris, R. E. & Calvin, M. (1955) J. Exp. Bot.

6,64-74.7. Fogg, G. E., Stewart, W. D. P., Fay, P. & Walsby, A. E.

(1973) in The Blue-Green Algae (Academic Press, NewYork), pp. 343-357.

8. Zablen, L., Bonen, L., Meyer, R. & Woese, C. (1975) J.Mol. Evol., 4, 347-358.

9. Zablen, L. & Woese, C. R. (1975) J. Mol. Evol., in press.10. Woese, C. R., Sogin, M. L. & Sutton, L. A. (1974) J. Mol.

Evol. 3, 293-299.11. I)oolittle, W. F., Woese, C. It., Sogin, M. L., Bonen, L. &

Stahl, D. (1975) J. Mol. Evol., 4, 307-315.12. Avadhani, N. G. & Buetow, 1). E. (1972) Biochem. J. 140,

73-78.13. Kissil, M. S. & Buetow, D. E. (1974) J. Cell Biol. 63, 169a.14. Bonen, L. & Doolittle, W. F. (1975) Proc. Nat. Acad. Sci..

USA 72, 2310-2314.15. Buetow, D. E. & Schuit, K. E. (1968) J. Protozool. 15,

770-773.16. Gierer, A. & Schramm, G. (1956) Nature 177, 702-703.17. Bishop, D. H. L., Claybrook, J. R. & Spiegelman, S. (1967)

J. Mol. Biol. 26, 373-387.18. Loening, U. E. (1967) Biochem. J. 102, 231-2.57.19. Cory, S., Adams, J. M., Spahr, P. F. & Rensing, U. (1972)

J. Mol. Biol. 63, 41-56.20. Sanger, F., Brownlee, G. G. & Barrell, B. G. (1965) J. Mol.

Biol. 13, 373-398.21. Uchida, T., Bonen, L., Schaup, H. W., Lewis, B. J., Zablen,

L. & Woese, C. (1974) J. Mol. Evol. 3, 63-77.22. Piggott, G. H. & Carr, N. G. (1972) Science 175, 1259-1261.23. Lee, S. G. & Evans, W. R. (1971) Science 173, 241-242.24. Helser, T. L., Davies, J. & Dahlberg, J. E. (1972) Nature

New Biol. 235, 6-9.25. Bjork, G. R. & Isaksson, L. A. (1970) J. Mol. Biol. 51,

83-100.26. Woese, C. R., Fox, G. E., Zablen, L., Uchida, T., Bonen, L.,

Pechman, K., Lewis, B. J. & Stahl, D. (1975) Nature 254,83-86.

27. Noller, H. F. & Herr, W. (1974) Mol. Biol. Rep. 1, 437-439.28. Santer, U. V. & Santer, M. (1972) FEBS Lett. 21, 311-314.29. Ehresmann, C., Steigler, T., Mackie, G., Zimmermann,

R. A., Ebel, J.-P. & Fellner, P. (1974) in Handbook ofNucleic Acid Sequences (Joynson-Bruvvers, Ltd., Oxford,England), p. 73.

30. Shine, J. & Dalgarno, L. (1974) Proc. Nat. Acad. Sci. USA71, 1342-1346.

31. Fellner, P., Ehresmann, C., Stiegler, P. & Ebel, J.-P.(1972) Nature New Biol. 239, 1-5.

Proc. Nat. Acad. Sci. USA 72 (1975)

Dow

nloa

ded

by g

uest

on

Aug

ust 3

0, 2

020