Mechanism and regulation of bacterial ribosomal RNA processing.

ANNUAL REVIEWS

Annu . Rev. Microbial. 1990. 44:/05-29 Copyright © 1990 by Annual Reviews Inc.

Further

Quick links to online content All rights reserved

MECHANISM AND REGULATION OF Annu. Rev. Microbiol. 1990.44:105-129. Downloaded from www.annualreviews.org by Duke University on 08/06/12. For personal use only.

BACTERIAL RIBOSOMAL RNA PROCESSING Anand

K.

Srivastava and David Schlessinger

Department of Molecular Microbiology. Washington University School of Medicine,

St. Louis, Missouri KEY WORDS:

63110

ribosome formation, tRNA in ribosome biosynthesis, RNases, precursor RNA, 16S rRNA, 2 3 S rRNA

CONTENTS INTRODUCTION . .... . . . . . . . .... . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . ... . . . . . . .

105

RIBOSOMAL RNA OPERONS AND THEIR TRANSCRIPTS. . . . .. . . . ... . . . . . . . . . . . . . . . . .

106 106 110

rRNA Genes in Prokaryotes . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . Transcripts and Processed Transcripts . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . PROCESSING PATHWAYS .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . .. RNaseIll Cleavage of Primary Transcripts . . .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . Maturation of 235 rRNA . . . . . . . . ... . . . . . . . . . . . .. .. . . . . . . . . . . . . . . . . . . . .. .. . . . . . . . . . . . . . . . .. . . . . . . Maturation of 165 rRNA . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .. Maturation of 5S rRNA . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .

1 10 1 12 1 13 1 14 1 16

FEATURES OF PRECURSORS REQUIRED FOR RECOGNITION DURING PROCESSING . . . . . . . . . . . .. . . . . . . . . . . . . . . . . .. . . . .. . . . . . . . . . . . . . .. . . . . . . .. . . . . . . . . . . . . . . . . . . . . . ..

1 17

MATURATION IN POLYSOMES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. ..

How Functional Are Pre-rRNAs? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . .. The Polysomal Su bstrate . .. . . . .......... . . . . . . . ... . . . . . . .. ... . . .. . . . . .... . . . .. . . . . .. . . . . . . . . . . . . Coregulation of rRNA Processing and Protein Synthesis . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . Possible Link of Processing to Autoregulation of Ribosome Content . . . . . . . . . . . . . . . . . . PROSPECTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . ..

120 120 120 121 122 124

INTRODUCTION

During the assembly of functional ribosomes, rRNA undergoes a variety of metabolic steps included under the general heading of "processing." Studies 0066-4227/90/1001-0105$02. 00

105

Annu. Rev. Microbiol. 1990.44:105-129. Downloaded from www.annualreviews.org by Duke University on 08/06/12. For personal use only.

106

SRIVASTAVA & SCHLESSINGER

of processing are now venerable parts of molecular biology, but the signifi cance of rRNA processing-and therefore the significance of studies of processing-has remained questionable. By mid 1971, both indirect measure ments of the transcription time of rRNA and direct observation of species of rRNA found in growing and chloramphenicol-treated cells suggested that rRNA is transcribed in large precursor molecules (19, 70). As in other RNA molecules, whether stable structural entities or unstable mRNAs, extra se quences must be removed to produce mature rRNA. In the intervening years, many research lines have suggested that precursor sequences and processing are both important in various ways, but the evi dence remains fragmentary. Emerging evidence presents the especially in teresting possibility that processing steps in bacteria not only form mature RNA, but may also regulate ribosome biosynthesis. In growing cells the amount of ribosomes and thus rRNA synthesis is regulated. Several factors may affect the rate of formation of complete active ribosomes. Among the most attractive suggestions are: (a) inhibition of RNA synthesis by inhibition of RNA polymerase transcription in presence of guanosine tetraphosphate (ppGpp) (47, 48); (b) regulation by the rRNA transcription product through a negative feedback loop (42); and (c) a possible limitation of the rate of protein synthesis by the rate of processing (85), which could in tum regulate the production of ribosomes. RIBOSOMAL RNA OPERONS AND THEIR TRANSCRIPTS

Because the level of their expression is universally high and their function is indispensable, rRNA genes constitute a reliable model system for the study of organization and expression of organelle genomes throughout phylogeny. Among the range of eubacteria, archaebacteria, and eukaryotes, rRNA from the eubacterium Escherichia coli has been studied in greater detail. Although significant progress has been made in the study of rRNA from other organ isms, that information is still much less incomplete. Here we treat the rRNA operons rrn in E. coli in more detail and compare some relevant data for rrn operons from other species and groups.

rRNA Genes in Prokaryotes The E. coli genome contains seven rRNA transcriptional units (rrnA, B, C, D, E, G, and If), each at a different chromosomal location (for details, see 46, 72, and their references). DNA for all the operons have been isolated and characterized, and most have been sequenced. Each has two promoters, P I and P2, about 120 bp apart (7, 15, 107). The 16S rRNA gene is located about 190 nucleotides downstream of promoter P2, is followed by a spacer region

rRNA PROCESSING Escherichia

�

G) �}D;

��1 =8

__

®

i

or

@ -,�

�23� 8

______

107

____

\

L.o--!J-l

@)

De Ala

/

or

\

� ! 58 Asp Trp :

®� ! ! @ Lo-o---J Asp 5SThr5S


58

� sUbtllls

rIRNAs,

238 ��16S =-�H�--�� --�� 5S

(!)� lie Ala

� thermophllu8

58 168 23S ��--'=--�n-/r----yr--�----'="------f-O-' 58 Gly 238 ...yr----=.c---f-G{)-

Halobacterlum cutlrubrum

Ala ,168 �L�_-_��=�KHL_

2_3 _ 8

__

-. �1�l\S ��

__

________

Desulfurococcus mobilis

Figure 1

Schematic of some eubacterial and archaebacterial rRNA operon structures. At the 5'

promoter sitt:s, veritcal lines indicate the number and relative position of multiple promoters. For

E. coli, operons BCEG have variant I for spacer tRNA; operons ADH have variant 2 . At the 3' end, operons A, B, E, and G have variant 3; operons C, D, and H have variants 4, 5, and 6, respectively. For B. subtilis, variant 7 (2 spacer tRNAs) is only found in two operon s .

containing genes for tRNA, then by the 23S rRNA, and then by another spacer region with a 5S rRNA sequence. In rrnB, for example, the two promoters are 180 nuc1eotides upstream of the 5' end of mature 16S rRNA (7). One spacer tRNAGIU2 gene resides between 16S and 23S rRNA gene, 171 nuc1eotides from the 3' end of 16S rRNA and 193 nuc1eotides upstream of the 5' end of the 23S rRNA gene. The gene for 5S rRNA is 92 bp downstream of the 23S rRNA gene, and two terminator signals follow, one adjacent to its 3' terminus, the other approximately 175 nucleotides downstream. In every E. coli rrn operon, 16S and 238 rRNA are each present in one copy, but the numbers of tRNA and 58 RNA genes vary (Figure 1). For example, in contrast to the single copy of 58 rRNA at the end of rrnB (7), two

108


5S rRNA copies are present in rrnD, with a tRNA gene between them (18). S everal different tRNAs (spacer tKNAs) are found between 16S and 23S rRNA, and still others can reside distal to the 5S rRNA gene (61, 62, 66, 67). In the internal spacer, four operons (rrnB C, E, and G) contain tRNAG1u2; the other three, (rrnA, D, andH) contain copies of both tRNAllel and tRNAAla lB. Distal to the 5S RNA sequence, rmC has both tRNAAsp I and tRNAT'1'; rrnD and rrnH have, respectively, tRNAThe and tRNA Asp I gene copies. The possibility that pre-rRNAs fall into discrete functional classes that correspond to the structural variants is intriguing, but no plausible notion for such a function has been made. Any such function must occur at the level of the pre-rRNA, since the final rRNA species are presumed identical. In fact, the final steps in the processing of the tRNAs in the spacer region, for example, are similar to those for other cellular tRNAs, using RNases, P, D, etc. (1, 46), and are not be discussed further here. In a typical ribosomal RNA operon of Bacillus subtilis (Figure 1), the lack of terminator hairpin structures proximal to the tRNA genes indicates that one polycistronic unit can run through any of the distal tRNA regions ( 100). E. coli, however, differs greatly. For example, although the genome size in E. coli and B. subtilis is approximately the same, B. subtilis contains ten rrn operons (4, 55), and only two of these contain tRNAs in the spacer between 16S and 23S rRNA (60). The tRNA genes also reside as clusters in the spacer between tandem ribosomal gene sets (4, 41, 55, 99), which is an even more striking difference from E. coli, where the rrn operons are dispersed (72). One of the consequences of the clustering is a marked genetic instability of the numbers of rRNA operons in B. subtilis resulting from deletion of closely spaced operons (102). Looping out and homologous recombination most likely cause the deletion. The distribution of clustered vs dispersed rrn operons among prokaryotes remains to be determined, but B. subtilis resem bles the eukaryotic paradigm. There, tandem repeats of a single unit are seen in essentially all cases (59), and the number is quite plastic, increasing or decreasing in Drosophila, for example, by several mechanisms (75). The variations in growth rate of E. coli and B. subtilis in nature are probably comparable. In any case, if each copy of the operon indeed contains its own promoter, clustering in one organism compared with another seems to have no intrinsic advantage. Without data, speculation has an open field. For example, the clusters may be significant in the sporulation of the Bacillus; or alternatively, interactions of distant rrn operons may occur during growth that are not tolerated in linear chromosomes like those found in Bacilli and eukaryotes. Operon organization is similar in many other eubacteria. Mycoplasma, which is the smallest genome among self-replicating prokaryotes, contains only one or two copies of the rRNA operon compared to the many copies in


,


rRNA PROCESSING

109

other prokaryotes (74), but most of the rRNA genes studied thus far in Mycoplasma have a topology similar to other prokaryotes (e.g. an order of 5' 16S-23S and 5S). In one minor variation, Mycoplasma hyponeumoniae, the 16S and 23S rRNA genes are again very close, but the 5S rRNA gene is displaced about 4 kb (96). In Chlamydia, where two rRNA operons have been found, each encodes at least one 16S and one 23S rRNA, and both operons are transcribed from tandem upstream promoters and contain no intervening sequences (21). The number of rRNA operons varies from one to four in different species of archaebacteria (17). The organization of single rrn genes in Halobacterium cutirubum is typical (Figure 1). The rrn contain a spacer tRNAAla and a distal tRNAcys gene (39). In most archaebacteria, in contrast to the two tandem promoters of E. coli. expression is driven by multiple tandemly arranged promoters (16, 64), which presumably operate with different efficiencies. A different structural organization has recently been reported for Mycoplas ma gallisepticum. in which one locus contains 168. 238, and 58 rRNAs; a second contains 23S and presumably 5S rRNA; and a third appears to have only a 16 8 rRNA gene (10). The significance of this arrangement is not known. Another very different organization is seen in sulphur-dependent thermo acidophils (like Desu/furococcus mobilis), in which an operon has an intron within the 23S rRNA gene and has no spacer tRNA (Figure 1; 49). In this case also, the 5S rRNA gene is unlinked to the 16S and 23S rRNA genes. The 5S rRNA is then separately transcribed (50), a situation somewhat similar to the archaebacteria Thermoproteus tenax (51) and typical eukaryotes. A unique feature is seen in the thermophile Thermus thermophilus (Figure 1). It has two copies of the rRNA gene, but the 168 rRNA gene is separated by at least 7. 8 and 6. 4 kb from the 5' end of the 238 rRNA-58 rRNA sequences (36), and is transcribed as an isolated transcription unit (35). (The other transcription unit contains 238 rRNA, a 5S rRNA sequence, and a tRNAG1y gene. ) No direct evidence relates the number of rrn operon copies in an organism to growth-rate regulation, but slowly growing organisms have been shown to have fewer copies. The internal structure of individual operons tends to vary according Ito a common theme. Features like the juxtaposition of 168 and 23S rRNAs may have arisen evolutionarily-perhaps from a single entity. This operon then served as the raw material for variations in structure and for the superposition of additional advantages based on the juxtaposition-for ex ample, more efficient ribosome formation or coregulation of 30S and 50S ribosome formation. The formation of eukaryotic rRNA, where transcription by RN A polymerase I is the rule may parallel this development, but species specific factors make rRNA transcription quite idiosyncratic (84).

11 0

SRIVASTAVA

&

SCHLESSINGER


Transcripts and Processed Transcripts In all the bacteria studied to date, regardless of the details of organization, the complex rrn region is transcribed into long precursor molecules that include all of the spacer elements. In E. coli, the 30S pre-rRNA contains 22% more extra sequences than 16S and 23S rRNA (29, 68, 70). The 146 nucleotides immediately preceding the mature 16S rRNA are essentially identical in the five operons studied in detail (7, 15, 106). The spacer regions flanking the mature RNA sequences are also highly conserved (104). Precursor sequences at the 5' end and 3' end of 16S and 23S rRNA contain complementary sequence tracts that form strong base-paired stems enclosing the sequence of the mature species (5, 106). The 146 nucleotides upstream of 16S rRNA include 131 involved in stem formation. Those and the 43 nucleotides immediately following mature 16S rRNA are identical in the four operons studied (7, 67, 107). The stem bracketing 23S rRNA involves 1 14 nucleotides on the 5' side and 71 nucleotides 3' to the 23S rRNA, and actually includes eight base pairs involving the 5' and 3' terminal nucleotides of mature 23S rRNA ( 14). Once again, the sequences are conserved in the various operons.

Because the operons are scattered in the genome, there is probably not a rectification mechanism based on unequal crossing-over in tandem repeats that homogenizes all the copies (as would be conceivable, for example, in the replication of B. subtilis DNA; see above). Instead, a selection process may produce an optimized sequence that is then spread through the genome by rare but selected recombination events. The conservation of features of the operon and its transcripts thus extends to most though not all eubacteria studied thus far: in particular, cotranscrip tion of 16S and 23S rRNA is very frequent, and in evolutionarily distant organisms, the long transcripts contain spacer tRNAs and sequences that enclose the mature 16S and 23S rRNA species in double-stranded stems. Not surprisingly, these features tend to be involved in the regulation of transcrip tion and processing, and thus of ribosome formation. The analysis is limited to only a few cases but some features are detailed here. The integration of processing in cell metabolism beginning to emerge will be intriguingly varied in the bacteria whose rrn structure differs.

PROCESSING PATHWAYS

In E. coli cells, processing is rapid and most rRNA is mature. Only 1-2% of rRNA is present as long precursors (44). The processing is multistep: (a) primary processing leads to formation of intermediate precursors of rRNA from the long transcripts; and (b) maturation steps produce mature termini of


rRNA PROCESSING

58 RNA

5

Figure 2

111

Distal IRNA

12fl12-� ��3'

S,chematic of the structure of an rrn operon and major processing steps for 16S and

23S RNA in relation to spacer tRNA. The drawing is not to scale: numbers indicate the position of primary processing cleavages by RNase III (3, 4, 7, 8 and 9) and secondary processing that produces the mature termini of 16S [1 (5' end), 2 (3' end)]. 23S rRNA [10 (5' end), 11 (3' end)]. and 5S rRNA at 12. The RNase P cleavage site (5) is shown at the 5' end of the tRNAs; additional points of nuclease cleavage (6) at the 3' ends are also indicated. Mature 16S and 23S rRNA sequences are indicated by a solid line; other precursor sequences by a hatched line. Two precursor seq[uences are required for 16S rRNA formation, (single and double asterisk) and one is required for 23S rRNA formation (triple asterisk), as indicated by a thicker filled line superim posed on the: hatched one. The secondary structures are detailed further in Ref. 21 and 41.

rRNA from precursors, either by direct action of single enzymes or with the preliminary action of several additional enzymes (see Figure 2). S oluble factors and more or less complex reaction conditions are required for final maturation steps. Preribosome formation is required for maturation cleavages to occur. Ribosomal proteins and other components are thus in ferred to promote a conformation that exposes or strains bonds at the mature terminus (see below). Some form of autocatalysis may thereby be enhanced, as in the case of Tetrahymena (9), but no strong evidence as yet supports any such spontaneous formation of a mature terminus of rRNA. In the discussion below, we assume that the steps involved are all catalyzed by enzymes. Initial situdies of processing used sucrose gradient and gel electrophoretic fractionation of processing products. Newer technology now allows the study of processing steps at the nucleotide level, using Northern hybridization, nuclease protection assays, primer extension analysis, fingerprinting, etc. Two approaches are especially useful: (a) detection of processing in termediates in steady-state cellular RNA (i.e. following maturation reactions that had occurred in vivo), and (b) use of isolated precursor particles to define actual minimum requirements for maturation in vitro, which is assessed by studying the release of fragments of pre-rRNA. The studies are facilitated using plasmids containing cloned rRNA operons. Systems have been de-


112


veloped that permit selective and conditional expression of the plasmid-borne rRNA under the control of E. coli, lambda, or T7 promoters (54, 93). Such plasmids can be re-introduced into the cell in intact or mutated form, and the expression of the rRNA can be studied without interference with or effect on endogenous rRNA metabolism. The isolation of an E. coli mutant (AB 301/105) defective in RNase ill (43) allowed identification of the endoribonuclease RNase ill as an early partici pant in in vivo processing of rRNA and permitted the detection of the full 30S pre-rRNA (19, 70) that reaches appreciable steady-state levels in the absence of RNase III. In addition, 30S pre-rRNA was cleaved with purified RNase III in vitro to produce species slightly larger than mature 16S and 23S rRNA (19, 69). These later proved to be the same as major processing intermediates found in wild-type cells. Major features of precursor RNA structure can be seen directly in partially denatured RNA molecules. In particular, the double-stranded stems at the base of 16S and 23S rRNA loops show up easily in the electron microscope (52, 53, 82). Furthermore, when the unprocessed 23S RNA from 50S ribo somes of the RNase III-deficient strain is examined, the stem can still be clearly seen in conditions where all other prominent features of secondary structure are denatured (82).

RNase

III

Cleavage of Primary Transcripts

In wild-type cells, processing begins even before transcription of the rrn operon is complete (3, 44). RNase ill separates p16S, p23S, ptRNA, and p5S species. Secondary processing events then produce the mature rRNAs. A critical difference between the initial RNase III action and final maturation is that accurate RNase ill cleavage occurs in vitro with naked pre-rRNA or with preribosomes as substrate, but final maturation depends on the preformation of the complex of rRNA and r-proteins (77). Thus, while all steps of rRNA processing occur at the level of ribonucleoprotein particles, r-proteins are only required for the final steps. RNase III cuts in the stems that bracket the mature 16S and 23S rRNA sequences (5, 106). Comparison of the RNase III processing sites in different operons shows close sequence homology; sequence variation between differ ent operons tends to be at positions thought to be unpaired; but much of the unpaired sequences outside of the stems also show strong conservation (see above). The specific requirements for RNase III cleavage are not yet clear. RNase III cleavage also occurs in bacteriophage T7 at specific sites in mRNA (19), which, like all known RNase III recognition sites, show secondary structural features without exhibiting any common primary sequence information. The enzyme has been shown to make cleavages at staggered positions in double-


rRNA PROCESSING

113

stranded stems, probably involving a separate cut in each strand (5, 106). In one case, a change in structure outside of the double-stranded region limits RNase III action to one cut instead of two (94; see below), and in our studies with rRNA, deletions in one strand prevent any RNase III cut in the other strand (87). Most likely, the specifity is primarily detennined by the double-stranded stem itself, which is consistent with results for 23S rRNA processing, where even a small deletion in the stem eliminates RNase III cleavage (91). Never theless, the sites characterized to date show considerable variation in secon dary structure. In phage mRNA, a single cleavage site can consist of four to five mismatched nucleotides (20); RNase III makes a single cut in that zone. Such data have led to the suggestion that the tertiary structure of the cleavage site may be the critical factor in RNase III recognition. But, of course, the secondary structure largely delimits possible tertiary structure. p16S RNA has much longer extra sequences than p23S at both its 5' and 3' termini. After RNase III cleavage, 115 nucleotides remain at the 5' end and 33 at the 3 I end, with a surviving base-paired stem of 26 nucleotides. In contrast, p23S rRNA is left with only 3 or 7 nucleotides at its 5' end and 7 to 9 at its 3' end (see Figure 2; 81). This species retains a base-paired stem of only 17 nucleotides, including 8 bp formed by complementary terminal nucleotides of mature sequence. The difference in the amounts of precursor sequence in the two RNAs may be related to the locations of their termini in ribosomal particles-the 5' and 3' ends of mature 16S rRNA are well separated (6), whereas the tennini of 23S rRNA are in close proximity (82). E. coli cells lacking RNase III do not excise the nonnal RNA precursors from nascent transcripts as do wild-type cells. These cells nonetheless form p16S and p23S RNAs somewhat larger than their wild-type counterparts (26). These species contain additional sequences at both termini and extend beyond the normal sites of RNase III cleavage. The species are produced from larger precursors in the absence of RNase III by nucleases acting either specifically, at normal processing sites bordering tRNA and 5S RNA sequences, or nonspecifically, at single-stranded regions. A large lIlumber of enzymes other than RNase III have also been implicated in RNA processing. Only a few of these are well characterized, however. They include RNase P, RNase D, RNase E, and RNase M5 (46). Several mutants defective in one or more activities are now available, and are useful in elucidating the processing pathways of specific rRNAs.

Maturation of 23S rRNA Initial cleavage by RNase III is indispensible for the maturation of 23S rRNA; maturation fails completely in the absence of the enzyme (45). [Fortunately for the cell and the investigator, the unprocessed 23S RNA in RNase III-


114


deficient cells is functional enough to maintain strain viability (see below). ] Studies of 23S rRNA processing have exploited the RNase III-deficient strain for both in vivo and in vitro experi ments. This strain's heterogenous population of 23S precursors have from 20 to more than 97 additional nucleotides at their 5' ends (45) in a range of RNA species produced in vivo by cleavage in RNA segments. In addition, a species 4 nucleotides shorter at the 5' terminus than mature 23S rRNA has also been found in both wild-type and RNase III deficient strain (80). Any significance of this species is not clear, but it might be functionally active, as evidenced by its presence in the polysomal fraction in cells. Thus it may have some role in cell me tabolism. Mature 5' termini are not formed in vitro when isolated pre-23S rRNA, 50S, or 70S ribosomes are treated with purified RNase III or a salt-extracted preparation of ribosomal proteins from wild-type cells (81). Such incubations produce only the species characteristic of RNase III action, three or seven nucleotides longer than the mature 5' end. Instead, mature 5' termini form when incubations are carried out in protein synthetic conditions�r more simply, with polysomes as a substrate (81, 85). In the latter case, the reaction proceeds in ordinary buffered salts, without the need for protein synthetic conditions (85). These findings imply that: 1. A soluble enzyme or factor seems to be required to convert the RNase III-cleaved product to mature form. Most likely the reaction is endonucleolytic, since no intermediate species are detected. And 2. ribosomes in polysomes adopt a conformation that facilitates the maturation cleavage, and since free ribosomes are not detectably matured, ribosomes apparently lose the necessary conformation when they are free in the cytoplasm (see below). At the 3' end of p23S rRNA, exonucleolytic action produces the mature terminus after RNase III has cleaved the stem (81). Intermediates are observed lacking successive nucleotides up to the stem remaining in mature 23S rRNA. The observation of the same intermediate species at low levels in RNA from wild-type cells supports the inferred mechanism. The exonucleo lytic action occurs in buffered salt solutions, but seems to be more efficient in protein synthetic conditions (81), perhaps coordinated with maturation of the 5' terminus.

Maturation of 16S rRNA Unlike 23S rRNA, 16S rRNA matures without ordered processing reactions. At f irst, this type of maturation seemed improbable because cleavage of nascent transcripts by RNase III in wild-type cells is certainly much faster than the formation of mature termini (44). The faster kinetics, however, did not indicate an obligate route of processing, which became clear when the pathway was detailed in the RNase III-deficient strain. In the absence of


rRNA PROCESSING

115

RNase III, no cleavages were seen in precursor-specific sequences in the vicinity of cleavage sites, but 16S molecules formed with normal mature termini at the same rate as in the wild-type strain (44, 86). These results implied that enzymes involved in final maturation steps act independently of RNase III. The mechanism of 16S rRNA processing is not known in detail, but intact precursor fragments that extend from the mature 5' and 3' termini to se quences far beyond the RNase III cleavage sites are released during process ing reactions in vitro (86). Identical fragments are found in the RNase III-deficient strain in vivo. Thus, mature termini are apparently formed by single direct endonucleolytic cleavage at both 5' and 3' ends with the concomitant release of precursor fragments, and without formation of any substantially longer intermediate species. These results extend the findings that mature 16S rRNA forms at the same rate whether or not prior RNase III cleavage has occurred. Maturation enzymes can seemingly function in either case. Interestingly enough, precursor sequences were anticipated to degrade as soon as they were cleaved off pre-rRNA, but such fragments have been found to remain associated with ribosomes for some time after cleavage (86, 87). Once released, the fragments are very unstable and are rapidly metabolized by both exo- and endonucleases. Intact fragments have been detected easily, however, as products from transcripts of rrnB operons modified in various ways (87). The enzymes involved in the formation of 5' and 3' termini of 16S rRNA have been partially isolated and characterized. At the 5' end of 16S rRNA, an endonuclease (RNase M16) has been studied ( 1 4). It seems to be deficient in a particular mutant that accumulates a "16.3S" p16S rRNA precursor with 66 extra 5' nucleotides. The enzyme preparation does not cleave purified 16.3S rRNA or longer precursors, but cleaves the precursor in 30S or 70S ribo somes. This action demonstrates that maturation reactions depend on ribo some assembly. The 16.3S rRNA has a mature 3' end ( 1 4), suggesting that separate enzymes form the two termini of 16S rRNA. An enzymatic activity that forms 3' mature termini in vitro has been partially purified from crude extracts (37). In vitro, the 3' end of 16S rRNA, like that of 23S rRNA, matures efficiently in protein synthetic conditions (37). Correct in vitro maturation of the 5' end of rRNA aliso occurs efficiently in protein synthetic conditions. The signifi cance of this effect on processing is discussed below. The maturation of 1 6S rRNA does not show any order of reaction as 23S rRNA does (see above), but in wild-type cells rapid cleavage by RNase III in precursor specific sequences (double-stranded stem) precedes the subsequent formation of the direct mature ends, without involving any intermediate


116

SRIVASTAVA

& SCHLESSINGER

species. Fragments detected in cells in the steady-state by nuclease protection assays also show that precursor species produced by cleavage downstream of the 5' RNase III site and upstream of the 3' RNase III site are present at a very low level in wild-type cells and are much more prominent in an RNase III-deficient strain (86). Possibly, this results from the action of enzymes that have lower affinity or are present at a lower level than RNAse III, and are usually preempted by RNase III action in wild-type strains. As 30S ribosomes containing p l 6S rRNA are not biologically active, maturation is essential for function in protein synthesis (but see section on maturation in polysomes).

Maturation of 5S rRNA 5S rRNA matures through a multistep process. The product of RNase III cleavage in wild-type cells is a 9S precursor molecule (p5S) with 85 extra nucleotides at its 5' end and extra 3' nucleotides extending to the terminator site (76, 79a). In operons with distal tRNA at the end of transcripts 5S is cleaved from the tRNA moiety, probably by RNase P action at the 5' terminus of tRNA. The 9S precursor is not normally seen in wild-type cells but accumulates in a temperature-sensitive mutant of RNase E grown at nonper missive temperature (27). RNase E rapidly cleaves precursor 5S rRNA in wild-type cells to produce a species with 3 extra nucleotides adjacent to each end of the 5S RNA sequence (76). Additional species found in cells in which protein synthesis is inhibited have 1, 2, or 3 extra nucleotides, suggesting that final maturation of 5S RNA probably involves 5' and 3' exonucleases (22). Similar to 16S and 23S rRNA, p5S rRNA has been reported in polysomes in cells (22), where maturation most likely occurs. Although initial studies of 5S rRNA processing have been done using an RNase E-temperature-sensitive mutant, recently 9S precursor molecules of similar structure have been produced in vivo from a construct with a T7 promoter (11). Use of this extrachromosomal product has shown that RNase E seems to recognize a part ic ular RNA sequence rather than secondary or tertiary structures. Also, the substrate requirement of RNase E could be a helix terminating in a GC/UA dinucleotide pair followed by a single-stranded stretch containing the sequence AAU end. In the proposed model, cleavage at the 5' end is independent of cleavage at the 3' end (11). A entirely different situation in which processing of 5S rRNA is defined in greater biochemical detail is seen in B. subtilis. In this case, initial cleavage by RNase III is followed by endonucleolytic cuts by RNase M5 at both 5' and 3' mature termini. Here, the reaction does not depend on the assembly of a complete ribosome, but strongly depends on the presence of r-protein BL16, which binds to the pre-5S RNA (83, 89). ,

rRNA PROCESSING

117


FEATURES OF PRECURSORS REQUIRED FOR RECOGNITION DURING PROCESSING

Only a few of the enzymes implicated in RNA processing have been studied in detail, and those include none involved in the final maturation steps that produce 16S and 23S rRNA termini (see 46 for list of enzymes). Neverthe less, more and more information is available about possible structural cues for processing steps. One hint comes from universal structural features. These include the double-stranded stems that bracket the mature rRNAs and the tRNA se quences beyond the 3' end of the 16S rRNA. The conservation of these features during prokaryotic evolution suggests that they might be important for the formation or function of rRNA. No role for spacer tRNA has been suggested that might rationalize its presence in pre-rRNA, but the suggestion has been made (see 46) that precursor-specific stem sequences might aid in ribosome biosynthesis. 16S rRNA must be mature to function (73), and the activation (mergy for ribosome formation from mature rRNA is very high (97). Precursor sequences might help to promote a conformation required for assembly. The extension of comparative partial denaturation mapping of precursor and mature 16S rRNA in the electron microscope suggests another possibility. The loop pattern seen in mature 16S rRNA (52, 53) was totally compatible with that inferred from evolutionary comparisons and chemical substitution studies (71), but the pre-rRNA exhibited particular alternate conformational features (52). This pattern suggests that the 16S rRNA sequence can at least have a more plastic secondary structure when it is part of a precursor species, but it does not demonstrate any involvement of such plasticity in ribosome assembly. After assembly is complete and ribosomes are functional, extra sequences and any effect they have on conformation become dispensible, and they are discarded, which is consistent with the apparent promotion of processing when the substrate is in polysomes (see below). As discussed above, the precise requirements for RNase III cleavage are still not clear, but the specificity is primarily determined by the double stranded stf:m itself (87). However, the primacy of double-stranded structure does not preclude alteration in RNase III action when the conformation of the rest of the rRNA is altered. For example, cleavage by RNase III is restricted when the distal spacer tRNA in the transcript is deleted in whole or in part (94). Any requirement for the stem structure for rRNA maturation must therefore be more subtle because processing can proceed at both the 5' and 3' termini when the base-paired stem is totally absent, and 5' termini are formed independent of the sequence or processing events at the 3' end (54, 87).


118


E. Morgan (personal communication) has presented some provocative speculations about the juxtaposition, in both rDNA and the early leftward transcription unit of lambda, of BoxA sequences (23, 56), and sequences involved in double-stranded RNAse III sites. He proposes that at a boxA sequence, the RNA polymerase grabs hold of the distal nascent RNA se quence and keeps hold of the 5' portion until the 3' portion of the double stranded region has been synthesized. With no free ends available during the synthetic process, susceptibility of the transcript to RNases would be les sened, and the formation of the intramolecular stem at which RNase III acts would be facilitated by the continued proximity of the 5' stem-forming sequence when the polymerase reached the complementary 3' sequence tract. In this way, the RNA polymerase would help direct the formation of the secondary structure of pre-rRNA molecules, and thus would help stabilize nascent molecules until the more protective double-stranded stem were formed. This model is somewhat analogous to Horwitz et aI's (38) for boxA function in lambda phage development. The model is consistent with data in the lambda system and with Morgan's studies of effects of mutations in the boxA sequence on rDNA expression, and it is attractive. As Morgan (personal communication) points out, however, these effects on ribosome formation must be second-order because stable constructs can be formed and properly processed in the absence of the double-stranded stem (see above). In contrast to the apparent dispensibility of the double-stranded stems, the distal tRNA must be present-and at a minimum spacer distance-to permit formation of the 3' terminus of 16S pre-rRNA (Figure 2; 87). Whether an intact tRNA sequence is required for the processing event is not clear because not many changes in tRNA sequence have been assayed for their effects on function. In a first approach to test this possibility we modified the tRNA structure by inserting 4 extra nucleotides in its D-loop, but transcripts from this construct were processed as well as the parental rRNA (87). Proper maturation also requires part but not all of the 16S rRNA sequence itself. For example, deletion of the large stem structure close to the 3' end of mature 16S rRNA (at residues 1409-1490, where the 3' terminus is at nucleotide 1542) did not block the formation of the 3' terminus, but deletions and even single base changes at a number of sites in the 16S rRNA sequence led to a failure of processing (32, 90). The maturation of 16S pre-rRNA, which requires mature rRNA, spacer RNA, and a distal tRNA sequence at a proper distance (Figure 2), is thus very different from the model of 5S RNA of B. subtilis. The recognition signals for cleavage of that 5S pre-rRNA lie completely in the mature sequence, and precursor-specific segments adjacent to either terminus can be replaced by oligonucleotides without blocking processing (2, 65). To assess the possible role of various spacer sequences in rRNA formation, plasmid constructs with no precursor-specific sequence features were tested

rRNA PROCESSING pre-rRNA

+

119

r-proteins

/ pSOS"'"(p23S, pSS)


p30S(p16S, p4S')

Figure 3

POLYSOMES Mature ribosomes

Representation of the processing of a typical bacterial pre-rRNA in relation to

ribosome synthesis. The initial long transcript is cleaved as in Figure 2 to yield the pRNA species indicated. Primary processing can occur with cues for enzymatic cleavage in the pre-rRNA itself,

producing the p30S and p50S pre-ribosomes with the addition of ribosomal proteins. An asterisk indicates that the p4S spacer sequence may remain bound to the nascent p30S particle and

contribute to its maturation (see text). Secondary processing (maturation) occurs later, possibly in polysomes, with the speculative suggestion of an initiation complex that may contain both p30S and p50S pre-ribosomes (double asterisk; see text).

for their capacity to form mature rRNA. Since mutants that accumulate abnormal quantities of precursors may be inviable, a useful approach has been to create mutant rDNAs in vitro and introduce them into cells as ex trachromosomal elements whose transcription is conditional. When transcrip tion is put under the control of an inducible T7 promoter (87, 92, 93), the formation of T7 polymerase is itself induced by the addition of �-isopropyl thio-D-galactoside (IPTG), and the polymerase in tum transcribes the plas mid-borne rDNA. Thus, a series of deletions in the rDNA spacer can be carried on [he plasmid without damaging the cell, and any effect of a deletion on RNA maturation can be assessed following the induction of the T7 promoter function. Stem-forming sequences were dispensible for both 5' and 3' terminus formation, whereas an intact spacer tRNA positioned more than 24 nucleo tides downstream of the 16S RNA sequence was required for correct 3' -end maturation (87). These results sllggest that spacer tRNA at an appropriate location helps form a conformation obligate for pre-rRNA processing, per haps by binding to a nascent binding site in pre-ribosomes. Thus, spacer tRNAs may be an obligate participant in ribosome formation. In the absence of the stem structure, the tRNA must be more than 23 nucleotides from the 3' end of the 16S rRNA sequence, but 39 nucleotides is at least sufficient to support accurate processing. Presumably, the tRNA is folded into a standard clover-leaf configuration even in pre-rRNA, and the spacer distance permits the tRNA to interact properly with other parts of the pre-rRNA (Figure 3). Pre-tRNA could interact with or even help form the P site in a precursor to the 30S ribosome.


120


The evolutionarily conserved double-stranded stem might also participate in this process. Although it is dispensible for 16S rRNA processing, the stem might increase the efficiency. For example, the stem might have evolved to angle the nascent tRNA in pre-rRNA toward its site of interaction with the ribosome. According to this speculative model , tRNA tends to remain attached to pre-16S rRNA through an intact intervening spacer until cleavage occurs at the 3' end of 16S rRNA (or at RNase III or RNase P cleavage sites). Thus the processing steps provide a route for the mutation of ribosomes and a check point to insure that only properly assembled ribosomes are added to the cellular pool. MATURATION IN POLYSOMES

How Functional Are Pre-rRNAs? Because unmatured 23S rRNA molecules function in 50S ribosomes of the RNase III-deficient mutant, 23S precursor RNA functions in vivo (45). In contrast, ribosomes containing pre-16S rRNA are not biologically active: processing to mature 16S rRNA molecules is obligate for competence in protein synthesis (73, 103). This inactivity may be because the 16S rRNA and 30S ribosome are more intimately involved in the initiation phase of protein synthesis (13a, 67a), as they have demanding conformational requirements for the Shine-Delgamo interaction, attachment of initiation factors, etc. It may be relevant that the 10 nucleotides of precursor sequence immediately preceding the 5 mature terminus are base-paired with neighboring mature 16S rRNA sequence, which could help ensure nonfunctionality until process ing is complete. Though they are probably not as efficient as mature ribosomes, ribosomes with p23S rRNA function well enough for cells to survive. Therefore that p23S rRNAs are found in poly somes extracted from cells is not surprising (82). It is quite surprising, however, that the inactive p16S rRNA is also found in polysomes (63). These results question the extent to which pre cursors of structural rRNAs can function in cells before processing is com pleted. I

The Polysomal Substrate The relation of the protein synthesis to maturation has been somewhat clar ified by the findings that mature termini of both 16S and 23S rRNA can be formed from preribosomes incubated with enzyme preparations in vitro, but maturation of both termini of 16S rRNA and the 3' end of 23S rRNA is more efficient under conditions of in vitro protein synthesis (37, 81). The depen-


rRNA PROCESSING

121

dence is total for the 5' end of 23S rRNA: it matures only when pre-50S ribosomes are added to 30S ribosomes and other components required for protein synthesis (81). The relation between maturation events and protein synthesis could hold if one of the components of protein synthetic mixture, like GTP, were required for maturation. Alternatively, preribosomes might be found in polysomes because they can participate in some partial reaction of protein synthesis, and thereby achieve a RNA conformation required for maturation reaction. The latter alternative is supported by the finding that polysomes rather than free ribosomes are the preferred substrate for matura tion (85). With polysomes instead of 50S particles as substrate, maturation can be carried out in simple buffers in the absence of protein synthesis conditions (85). This implies that the final maturation steps occur only after pre-ribosomes somehow join in polysomes (see Figure 3). The fonnation of ribosomes is thus directly linked to their incorporation into the protein synthetic machinery. The in vitro requirement of protein synthesis for maturation with free ribosomes presumably serves to form the substrate polysome. The exact nature of the the substrate for maturation is still unclear. Howev er, 30S pre-ribosomes, which cannot translate mRNA, possibly can form 70S initiation complexes and mature in that fonn. The complex normally fonns by the success.ive accretion of mRNA, initiation factors, initiator tRNA, and 50S ribosomes combined with 30S ribosomes; an analogous complex containing pre-ribosomes may be an adequate substrate for the processing reaction. Direct evidence for such a complex is lacking, but if it exists, the continued movement of active ribosomes to free up initiation sites on mRNA would be obligate to pennit the continued binding and rapid maturation of pre ribosomes.

Coregulation of rRNA Processing and Protein Synthesis Cleavage and trimming of mature rRNA's large precursor naturally precede function of the rRNA in ribosomes, and the production and activity of ribosomes are probably distinct and noninteractive processes. However, re cent findings in E. coli show a connection between the synthesis of proteins and maturation of rRNA, suggesting that processing and function of ribo somes must be reciprocally dependent (85). One might expect that extra sequences and their removal from the business parts of RNA would be a kind of second-order refinement during evolution. For example, mature rRNA would of course be active, and precursor se quences might secondarily increase the efficiency of ribosome assembly, facilitate the targetting of nascent ribosomes to a site in the cell, etc. The fonnation of 23S rRNA shows some such characteristics: the pre-rRNA is already acltive in ribosomes. The processing pathway then is a kind of


122


evolutionary refinement; it is helpful but not obligatory for cell survival, and there is no need for alternative pathways. Once processing becomes indispensible, however, as it is for 16S rRNA formation, its relation to cellular metabolism becomes more complex. Apparently, once this commitment was made, continued processing of 16S rRNA became a given on an evolutionarily scale. Other processes, like the rate or extent of translation, could then be linked to processing in a regulatory loop. It may be relevant that pre-rRNA accumulates rapidly when protein synthe sis is blocked by antibiotics like chloramophenicol and processing of rRNA chains stops short (13, 78). At first, rRNA maturation was believed to stop because r-proteins were not being made-and thus the ribonucleoprotein particles that are the substrates for final maturation steps could not be formed. But the inhibition of rRNA maturation when protein synthesis is blocked has unexpected features. First, the inhibition of maturation is essentially im mediate in vivo, even though cells contain appreciable pools of already assembled pre-ribosomes (up to 10% of the total ribosome population) and free pools of individual ribosomal proteins (28). Second, as discussed above, protein synthetic conditions are required in vitro for formation of mature termini.

Possible Link of Processing to Autoregulation of Ribosome Content All prokaryotic 16S, 23S, and 5S rRNA mature after ribosomes are formed (8, 22, 63, 85). Even in eukaryotic yeast cells, in which ribosome assembly occurs in nucleoli, the final maturation of 18S rRNA has been reported to occur in the cytoplasm (98). In all of these cases, preribosomes may join to rnRNA before maturation is complete. This implies that the rate of protein synthesis and rRNA processing are interrelated or mutually controlled. In this way, bacteria might regulate the quantity and rate of production of ribosomes at the level of maturation. The rate of protein synthesis may directly limit the rate of processing, and immature ribosomes, in tum, may even limit the movement of mature component ribosomes on mRNA. Such a link between processing and ribosome function in the regulation of cell physiology is significant only if polysomes containing pre-rRNA translate less efficiently than matured polysomes. Thus far the evidence is indirect but suggestive; the RNase III-deficient strain, which contains polysomes with only pre-23S rRNA, grows more slowly and shows defects in translation of f3-galactosidase and other mRNAs (30, 79, 95). [The results thus far are inconclusive, however. Defects are corrected when RNase III is restored to the strains, but the effects can be indirect rather than based on altered polysome function in the mutant.]


rRNA PROCESSING

123

In exponentially growing E. coli, the number of ribosomes per amount of cellular protein increases with growth rate (see 58, 72). At any time, the number of active ribosomes in the cells is proportional to the rate of synthesis of total rRNA. All the rRNA and r-proteins (except L7/L12, which occurs in 4 copies) are present in one copy per ribosome (34). Since only limited quanti ties of free r-proteins and rRNA exists in growing cells (24, 25, 57), the rate of synthesis is roughly proportional to their incorporation into ribosomes, and the rate of net synthesis is just sufficient to meet the needs for new ribosomes. The functional activity of ribosomes seems somehow connected with their own production. Several factors have been suggested to affect the rate of formation of complete active ribosomes. Nomura et al (72) have deciphered the mechanism of coregulation of rRNA and r-proteins. The regulation of rRNA in tum has recently been discussed in terms of possible inhibition of RNA transcription by the action of guanosine tetraphosphate (ppGpp) on RNA polymerase (47, 48); or more generally, by an autoregulatory process through a negative feedback loop responding to the level of functional ribosomes (31, 33). In their general formulation, Nomura and coworkers proposed a feedback control of rRNA synthesis based on the cellular pool of free, nontranslating ribosomes. According to this proposal, an equilibrium balances free, nontranslating ribosomes and translating ribosomes, and environmental fac tors shift the equilibrium towards the buildup of the free fraction resulting in the repression of rRNA and tRNA synthesis. At a rate consistent with the feedback model, rRNA synthesis is gene-dosage independent when multiple copies of rDNA are introduced into cells on plasmids (42). The molecular mechanism involved in the feedback regulation remains in doubt, but recent work has focussed on translational capacity rather than the number of ribosomes as the important variable (12, 105). For example, ribosomes unable to participate in initiation should be unable to effect feed back. In support of this notion, (a) ribosomes with a mutational alteration in the anti-Shine-Dalgamo region at the 3' end of 16S rRNA are unable to participate in feedback regulation in vivo ( 105), and (b) lowering the cellular concentration of protein synthesis initiation factor 2 (IF2) to a level sufficient to support growth at only about one third of the normal growth rate results in an increase in rRNA content and an accumulation of large numbers of nontranslating ribosomes ( 12). In addition, induction of IF2 synthesis in IF2-deficient cells causes a repression of rRNA synthesis as ribosomes be come more active. Probably all the components involved in translation initia tion, and not just excess free ribosomes and IF2, are needed for feedback regulation. We have considered that the rate of ribosome formation and protein synthe sis are coregulated at the step where preribosomes join the translational

124


machinery to become mature (88). Lowered levels of IF2 or altered Shine Dalgamo sequences might interfere with the formation of initiation complexes on preribosomes. This would affect the rate of maturation, and thus the function of ribosomes, and hence rRNA synthesis by a feedback mechanism. No details of such a process are known, but the notion is an intriguing pointer.


PROSPECTS

Recent work emphasizes the power of the available technology, which seems adequate to resolve most or all remaining questions about the mechanism of processing. For example, with respect to the portions of transcripts required for processing, the extension of the use of deletion constructs should permit , one to find out: 1. whether intact spacer tRNA is required for 16S rRNA maturation (and if not, which portions are required); 2. whether the maturation process will accept any tRNA in the spacer region, or only the one in the corresponding operon, or any of the group present in the seven rrn operons; and 3 . whether spacer tRNA, or the 5S RNA, or the 3 ' distal tRNA have any corresponding role in the maturation of 23S rRNA. Similarly, the availability of substrates and rapid assays for maturation in vitro provides a direct route to the purification of the enzymes or other factors involved in the formation of mature 1 6S and 23S rRNA termini. This availability aids in the subsequent investigation of the interactions of the processes involved in gene expression. As the field begins to move toward the challenge presented by eukaryotic systems, others (40, 101) have reviewed the way in which compartmentaliza tion into nucleus, cytoplasm, and mitochondria can lead to independent evolution of the pathways that lead to ribosomes in eukaryotic cells . To try to extend the incomplete results with E. coli is premature, but both ordered and stochastic mechanisms will certainly be encountered; the machinery for pro cessing will certainly include ribonucleoprotein complexes (complex particles containing small nuclear RNAs, instead of RNase P or polysomal complexes); and, as a result, precursor-specific sequences will again be studied in the mechanism of processing and in the integration of ribosome formation into cell metabolism. ACKNOWLEDGMENTS

Our own work has been sustained by NSF grant PMS peM 8406949. Much of the discussion for E . coli is in a chapter in the forthcoming book edited by the w. E. Hill (88).

rRNA PROCESSING

1 25


Literature Cited I . Altman, S . , 1 98 1 . Transfer RNA pro cessing enzymes. Cell 23:3-4 2. Altman, S., Guerrier-Takada, C . , FrankfOJ1, H. M. , Robertson, H . D. 1 982. RNA processing nuclease. In Nuclease, ed. S. M. Linn, R. J. Roberts, pp. 243-74. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press 3. Apirion, D . , Gegenheimer, P. 1984. Molecular biology of RNA processing in prokaryotic cells. In Processing of RNA , ed. D. Apirion, pp. 36-62. Boca Raton , FL: CRe Press 4. Bott, K . , Stewart, G. C. , Anderson, A. G . 1 984. Genetic mapping of cloned ribosomal RNA genes. In Syntro Con ference on Genetics and Biotechnology of Bacilli, ed. J. A. Hoch, A. T. Gane san, pp. 1 9-34. NY: Academic 5. Bram, R. J . , Young, R. A . , Steitz, 1. A . 1 980. The Ribonuclease III site flanking 23S sequence in the 30S ribosomal pre cursor RNA of E. coli. Cell 19:393-401 6. Brimacombe, R. , Atmadja, J . , Stiege, W. , Schuler, D . 1988 . A detailed model of the three-dimensional structure of Es cherichia coli 16S ribosomal RNA in situ in the 30S subunit. J. Mol. Bioi.

199: 1 1 5·-36 7. Brosius, J . , Dull, T. J . , Sleeter, D . D . , Noller, H. F . 1 98 1 . Gene organization

8.

9.

10.

II.

12.

13.

and primary structure of a ribosomal RNA operon from Escherichia coli. J. Mol. BioI. 1 48 : 107-27 Ceccarelli, A . , Dotto, G. P . , AItruda, F. , Perlo, c . , Silengo, L. , et al. 1978. Immature 50S subunits in Escherichia coli polysomes. FEBS Lett. 93:348-50 Cech, T. R . , Bass, B. L. 1 986. Biolo gical catalysis by RNA. Annu. Rev. Biochem. 55:599-629 Chen, X . , Finch, L. R. 1989. Novel arrangement of rRNA genes in Myco plasma gallisepticum: Separation of the 16S gene of one set from the 23S and 5S genes. J. Bacteriol. 1 7 1 :2876--7 8 Christiansen, J . 1988. The 9S RNA pre cursor of Escherichia coli 5S RNA has three structural domains: implications for processing. Nucleic Acids Res. 1 6:7457·-76 Cole, J. R. , Olsson, C. L. , Hershey, J. W. B . , Grunberg-Manago , M . , No mura, M. 1987. Feedback regulation of rRNA synthesis in Escherichia coli: Requirement for initiation factor IF2. J. Mol. Bioi. 198:383-92 Cremer, K . , Silcngo, L . , Schlessinger, D. 1974. Polypeptide formation and polyribosomes in E. coli treated with chloramphenicol . J. Bacterial. 1 1 8 :58291

13a. Dahlberg, A. E. 1989. The functional role of ribosomal RNA in protein syn thesis. Cell 57:525-29 1 4 . Dahlberg, A. E. , Dahlberg, J. E. , Lund, E . , Tokimatsu, H . , Rabson, A. B . , et al. 1 978. Processing of the 5' end of Es cherichia coli 16S ribosomal RNA. Proc. Natl. Acad. Sci. USA 75:35983602 1 5 . de Boer, H. A . , Gilbert, S. F. , Nomura, M. 1979. DNA sequences of promoter

region of rRNA operon rrnE and rrnA in E. coli. Cell 1 7:201-9 16. Dennis, P. P. 1985. Multiple promoters for the transcription of the ribosomal RNA gene cluster in Halobacteriam cutirubrum. J. Mol. Bioi. 1 86:457-61 1 7 . Dennis, P. P. 1986. Molecular biology of archaebacteria. J. Bacteriol. 168: 47 1-78 1 8 . Duester, G. L . , Holmes, W. M 1980.

The distal end of the ribosomal RNA operon rrnD of Escherichia coli contains a tRNATh, gene, two 55 rRNA genes and a transcription terminator. Nucleic Acids Res. 8:3793-3807 1 9 . Dunn, J. J . , 5tudier, F. W. 1 973 . T7 early RNAs and Escherichia coli ribo somal RNAs are cut from large pre cursor RNAs in vitro by Ribonuclease III. Proc. Natl. Acad. Sci. USA

70:3296--3300 20. Dunn, J. J . , Studier, F. W. 1983 . Com

plete nucleotide sequence of bacterio phage T7 DNA and the location of T7 genetic elements. J. Mol. Bioi.

166:477-535 2 1 . Engel, J. N. , Ganem, D. 1 987. Chlamy

dial rRNA operons: Gene organization and identification of putative tandem promoters. J. Bacteriol. 169:5678-85 22. Feunteun, J . , Jordan, B. R. , Monier, R . 1972. Study o f maturation o f 5S pre cursors in Escherichia coli. J. Mol. Bioi. 70:465-74 23. Friedman, D. I . , Olson, E. R.

1983 .

Evidence that a nucleotide sequence, "boxA," is involved in the action of the nasA protein. Cell 34: 1 43-49 24. Gausing, K . 1 974. Ribosomal protein in E. coli: Rate of synthesis and pool size at different growth rates. Mol. Gen. Genet. 1 29:61-75 25. Gausing, K . 1977. Regulation of ribo some production in Escherichia coli: Synthesis and stability of ribosomal RNA and ribosomal protein messenger RNA at different growth rates. J. Mol. Bioi. 1 1 5:335-54 26. Gegenheimer, P . , Watson, N . , Apirion, D. 1 977. Multiple pathways for primary processing of ribosomal RNA in Es-

1 26

27.

28.


29.

30.

31.

32.

33.

34. 35.

SRIVASTAVA & SCHLESSINGER cherichia coli. J. Bioi. Chem. 252: 3064-73 Ghora, B. K . , Apirion, D. 1979. Identification of a novel RNA molecule in a new RNA processing mutant of Es cherichia coli which contains 5S rRNA sequence. J. Bioi. Chem. 254: 195 1-56 Gierer, L . , Gierer, A. 1 968. Synthesis of ribosomal proteins and formation of ribosomes in Escherichia coli. J. Mol. Bioi. 34:293-303 Ginsberg, D . , Steitz, J. A. 1975. The 30S ribosomal precursor RNA from Es cherichia coli. J. Bioi. Chern. 250: 5647-54 Gitelman, D. R . , Apirion, D. 1980. The synthesis of some proteins is affected in RNA processing mutant of Escherichia coli. Biochern. Biophys. Res. Cornrnun. 96: 1063-70 Gourse, R . L . , de Boer, H. A . , Nomura, M. 1986. DNA determinants of rRNA synthesis in E. coli: Growth rate de pendent regulation, feedback inhibition, upstream activation, antitermination . Cell 44: 1 97-205 Gourse, R. L . , Stark, M. J. R . , Dahl berg, A. E. 1982. Site-directed mutagenesis of ribosomal RNA: Con struction and characterization of deletion mutants. J. Mol. Bioi. 159:397-4 1 6 Gourse, R. L . , Takebe, Y . , Sharrock, R. A . , Nomura, M. 1985 . Feedback regulation of rRNA and tRNA synthesis and accumulation of free ribosomes after conditional expression of rRNA genes. Proc. Natl. Acad. Sci. USA 82: 1069-73 Hardy , S . J. S. 1975. The stoichiometry of the ribosomal proteins of Escherichia coli. Mol. Gen. Genet. 140:253-74 Hartmann, R . K . , Erdmann, V . A. 1 989. Therrnus therrnophilus 1 6S rRNA is transcribed from an isolated transcrip

36.

37.

38.

39.

tion unit. J. Bacteriol. 1 7 1 :2933-4 1 Hartmann, R. K . , Kroger, B . , Vogel, D. W . , Ulbrich, N . , Erdmann, V. A. 1 987. Characterization of cloned rDNA from Thermus thermophilus HB8 . Biochem. Int. 14:267-75 Hayes, F. , Vasseur, M. 1976. Process ing of the 17S Escherichia coli precursor RNA in the 27-S pre -ribosomal p arti cle . Eur. J. Biochern. 6 1 :433-42 Horwitz, R. J. , Li, J . , Greenblatt, J. 1 987. An elongation control particle containing the N gene transcriptional antitermination protein of bacteriophage lambda. Cell 5 1 :631-41 Hui, I. , Dennis, P. P. 1985. Characterization of the ribosomal RNA gene clusters in Halobacteriurn cutirub rum. 1. Bioi. Chem. 260:899-906

40. Jacobs , H. D. 1989. Do ribosomes regu late mitochondrial RNA synthesis? Bioassay 1 1 :27-34 4 1 . Jarvis, E. D . , Widom, R. L . , Lafauci, G . , Setoguchi, Y . , Richter, I . R . , Rud ner, R. 1988 . Chromosomal organiza tion of rRNA operon in Bacillus subtilis. Genetics 1 20:625-35 42. Jinks-Robertson, S . , Gourse, R . L . , Nomura, M . 1983 . Expression of rRNA and tRNA genes in E. coli: Evidence for feedback reg ulation by products of rRNA operons. Cell 33:865-76 43. Kindler, P . , KieI, T. U . , Hofschneider, P. H. 1973. Isolation and characteriza tion of a ribonuclease III deficient mutant of Escherichia coli. J. Bioi. Chern. 25 1 :53-69 44. King, T. C. , Schlessinger, D. 1 983. S I nuclease mapping analysis o f ribosomal RNA processing in wild-type and pro cessing deficient Escherichia coli. J. Bioi. Chern. 258 : 1 2034-42 45. King, T. c . , Sirdeshmukh, R . , Schles singer, D. 1984. RNase III cleavage is obligate for maturation but not for func tion of Escherichia coli pre-23S rRNA. Proc. Natl. Acad. Sci. USA 8 1 : 1 8588 4 6 . King, T. C., Sirdeshmukh , R., Schlessinger, D . 1986. Nucleolytic pro cessing of ribonucleic acids transcripts in prokaryotes. Microbial. Rev. 50:42851 47. Kingston, R . E . , Chamberlin, M . J. 1 98 1 . Pausing and attenuation of in vitro transcription in the rmB operon of E. coli. Cell 27:523-3 1 48. Kingston, R. E . , Nierman, W. c . , Chamberlin, M . J . 1 98 1 . A direct effect of guanosine tetraphosphate on pausing of Escherichia coli RNA polymerase during RNA chain e longation . 1. Biol. Chern. 256:2787-97 49. Kjems, J . , Garrett, R. A. 1985 . An in tron in the 23S ribosomal RNA gene of the archaebacterium Desulfurococcus mobilis. Nature 3 1 8:675-77 50. Kjems, J . , Garrett, R. A. 1 987. Novel expression of the ribosomal RNA genes in the extreme thermophile and archaebacterium Desulfurococcus mo bilis. EMBO J. 6:352 1-30 5 1 . Kjems, J . , Leffers, H . , Garrett , R. A . , Wich, G . , Leinfelder, W . , Bock, A . 1 987. Gene organization, transcription signals and processing of the single ribo somal RNA operon of the archaebacteri urn Thermoproteus tenax. Nucleic Acids Res. 15:4821-35 52. Klein, B. K . , Staden, A . , Schlessinger, D. 1 985. Alternative conformations in

rRNA PROCESSING


Escherichia coli 16S ribosomal RNA. Proc. Natl. Acad. Sci. USA 82:3539-42 5 3 . Klein, B . K . , Staden, A . , Schlessinger, D. 1985. Electron microscopy of secon dary structure in partially denatured pre cursor and mature Escherichia coli 1 6S and 23S rRNA. J. Bioi. Chem. 260: 8 1 14-20 54. Krych , M . , Sirdeshmukh, R . , Gourse , R . , Schlessinger, D. 1987. Processing of Escherichia coli 1 6S rRNA with bac teriopha.ge lambda leader sequences. J. Bacteriol. 169:5523-29 5 5 . La Fauc i, G . , Widom, R. L . , Eisner, R. L., Jarvis, E. D ., Rudner, R . 1986. Mapping of rRNA genes with integrable plasmid in Bacillus subtilis. J. Bacteri al. 1 65 : 204-1 4 5 6 . L i , S . C . , Squires, C . L . , Squires, C . 1984. Antitermination o f E. coli rRNA transcription is caused by a control re gion segment containing lambda nut-like sequences. Cell 38:85 1-60 57. Lindahl, L. 1975. Intermediates and time kinetics of the in vivo assembly of Escherichia coli ribosomes. J. Mol. Bioi. 92: 1 5-37 58. Lindahl, L . , Zengel , M. 1986. Ribo somal RNA gene in Escherichia coli. Annu. Rev. Genet. 20:297-326 59. Long, E. 0 . , David, I. B . 1 980. Repe ated genes in eukaryotes. Annu. Rev. Biochem. 49:727-64 60. Loughney, K . , Lund, E . , Dahlberg, J. E. 1982 . tRNA genes are found between the 16S and 23S rRNA genes in B. sub

tilis. Nucleic Acids Res. 1 0 : 1 607-24 6 1 . Lund, E . , Dahlberg, J . E. 1 977. Spacer transfer RNAs in ribosomal RNA tran scripts of E. coli: processing of 30S ribosomal RNA in vitro. Cell 1 1 :247-62 62. Lund, E . , Dahlberg , J. E . , Lindahl, L . , JaskunUlas, S . R . , Dennis, P. P . , Nomura, M. 1976. Transfer RNA genes between 16S and 23S rRNA genes in rRNA transcription units of E. coli. Cell 7 : 1 65-77 6 3. Mangiarotti , G . , Turco, E . , Ponzetto, A . , Altruda, F. 1974. Precursor 16S RNA in active 30S ribosomes. Nature 247 : 1 47-48 64. Mankin , A. S . , Kagramanova, V. K . 1 988. Complex promoter pattern o f the single ribosomal operon of an archaebacterium Halobacterium halo bium. Nucleic Acids Res. 1 6:4679-92 65. Meyhaek, B . , Pace, B . , Ullenbeck, O. C., Pac:e, N. R . 1978. Use of T4 RNA ligase to construct model substrates for a ribosomal RNA maturation endo nuclease. Proc. Natl. Acad. Sci. USA 75:3045-49

127

66. Morgan, E. A . , Ikemura, T . , Nomura, M . 1977. Identification of spacer tRNA gene in individual ribosomal RNA transcription unit of E. coli. Proc. Natl. Acad. Sci. USA 74:2710-14 67. Morgan, E. A . , Ikemura, T., Post, L. E., Nomura, M . 1980. tRNA genes in rRNA operons of Escherichia coli. In Transfer RNA : Biological Aspects, ed. P. R. Schimmel, D . Soil, J. N . Abelson, pp. 259-66. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press 67a. Nierhaus, K. H . , Brimacombe, R . , Nowotny, V . , Pon, C . L . , Rheinberger, H. J. , et al. 1987. New aspects of struc ture, assembly, evolution, and function of ribosomes. Cold Spring Harbor Symp. Quant. Bioi. 52:665-74 68. Nikolaev, N., Birenbaum, M., Schlessinger, D. 1 97 5. 30S pre ribosomal RNA of Escherichia coli: pri mary and secondary processing. Biochim. Biophys. Acta. 395:478-89 69. Nikolaev, N . , Schlessinger, D . , Wel lauer, P. K. 1974. 30S pre-ribosomal RNA of Escherichia coli and products of cleavage by RNase TIT: length and molecular weight. J. Mol. BioI. 86:74 1 47 70. Nikolaev, N . , Silengo, L . , Schlessinger, D. 1 973. Synthesis of large precursor to ribosomal RNA in a mutant of Es

cherichia coli. Proc. Natl. Acad. Sci. USA 70:336 1-65 7 1 . Noller, H. F. 1984. Structure of ribo somal RNA. Annu. Rev. Biochem. 53: 1 1 9-62

72. Nomura , M . , Gourse, R. , Baughman, G. 1 984. Regulation of the synthesis of ribosomes and ribosomal components.

Annu. Rev. Biochem. 53:75- 1 1 7 7 3 . Nomura, M . , Held, W. A. 1 974. Reconstitution of ribosomes: Studies of ribosomes structure , function and assembly . In Ribosomes. ed. M. Nomura, A . Tissieres, P. Lengyel, pp. 1 93-223. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press 74. Razin, S . 1 985. Molecular biology and genetics of mycoplasm. Microbiol. Rev. 49:419-55 75. Ritossa, F. 1976. The bobbed locus. In

The Genetics and Biology of Drosophi la. Vol. l a, ed. M. Ashbruner, E.

Novitsk i, p p . 801-946. N Y : Academic 76. Roy, M. K . , Singh, B . , Ray, B. K . , Apirion , D . 1983 . Maturation of 5S rRNA: Ribonuclease E cleavages and their dependence on precursor se quences. Eur. J. Biochem. 1 3 1 : 1 1 9-27 77. Schlessinger, D. 1 980. Processing of ribosomal RNA transcripts in bacteria.


128

SRIVASTAVA

& SCHLESSINGER

In Ribosomes, ed. G. Chambiliss, G. R. Craven, K. Davies, L. Khan, M. Nomura, pp. 767-80. Baltimore: Uni versity Park Press 78. Schlessinger, D . , Ono, M . , Nikolaev, N . , Silengo, L. 1974. Accumulation of 30S preribosomal ribonucleic acid in an Escherichia coli mutant treated with chlorarnphenical. Biochemistry 13: 4268-7 1 79. Silengo, L. , Nikolaev, N . , Schlessinger, D . , Imarnato, F. 1974. Stabilization of mRNA with polar effects in an Es cherichia coli mutant. Mol. Gen. Genet. 1 34:7- 1 9 79a. Sing, B . , Apirion, D . 1 982. Primary and secondary structure in a precursor of 5 S rRNA. Biochim. Biophys . Acta 698:252-59 80. Sirdeshmukh, R . , Krych, M . , Schlessin ger, D. 1 985. Escherichia coli 23S ribo somal RNA truncated at its 5 ' terminus. Nucleic Acids Res. 1 3 : 1 1 85-92 8 1 . Sirdeshmukh, R . , Schlessinger, D . 1 985. Ordered processing o f Es cherichia coli 23S rRNA in vitro. Nucle ic Acids Res. 1 3 :504 1-54 82. S irdeshmukh , R . , Schlessinger, D. 1 985. Why is processing of 23S ribo somal RNA in Escherichia coli not obli gate for its function? J. Mol. Bioi. 1 86:669-72 83. Sogin, M. L . , Pace, B . , Pace, N. R. 1 977. Partial purification and properties of a ribosomal RNA maturation en donuclease from Bacillus subtilis. J. Bioi. Chem. 252: 1 350--57 84. Sollner-Webb, B . , Tower, J. 1986. Transcription of cloned eukaryotic ribo somal RNA genes. Annu. Rev. Biochem . 55:801-30 85. Srivastava, A. K . , Schlessinger, D. 1988. Coregulation of processing and translation: Mature 5 ' termini of Es cherichia coli 23S rRNA form in poly somes. Proc. Nat!. Acad. Sci. USA 74:7144-48 86. Srivastava, A. K . , Schlessinger, D. 1 989. Processing pathways of Es cherichia coli pre- 1 6S rRNA. Nucleic Acids Res. 1 7 : 1 649-63 87. Srivastava, A. K . , Schlessinger, D. 1 989. Escherichia coli 16S rRNA 3 ' end formation requires a distal transfer RNA sequence at a proper distance. EMBO J. 8:3 1 59-66 88. Srivastava, A . , Schlessinger, D. 1 990. Ribosomal RNA processing in E. coli. In Structure Function and Evolution of Ribosomes. ed. W . Hill. Washington, DC: American Society for Microbiolo gy. In press

89. Stahl, D. A . , Pace, B . , Marsh, T . , Pace, N. R. 1984. The ribonucleoprotein sub strate for a ribosomal RNA processing nuclease. 1. Bioi. Chem. 259: 1 144853 90. Stark, M . J. R., Gourse, R . L., Dahl berg . A. E. 1982. Site-directed muta genesis of ribosomal rRNA: Analysis of ribosomal RNA deletion mutant using maxicells. J. Mol. Bioi. 159:4 1 7-39 9 1 . Stark, M. J. R . , Gourse , R. L . , Jemiolo, D. K . , Dahlberg, A. E. 1 98 5 . A muta tion in Escherichia coli ribosomal RNA operon that blocks the production of pre cursor 23S ribosomal RNA by RNase III in vivo and in vitro. 1. Mol. Bioi. 1 82:205-1 6 9 2 . Steen, R . , Dahlberg, A. E . , Lade, B . N . , Studier, F . W . , Dunn, J . 1986. n RNA polymerase directed expression of the Escherichia coli rmB operon. EMBO J. 5 : 1 099- 1 103 93 . Steen, R . , Jemiolo, D . K . , Skinner, R. H . , Dunn, J . J . , Dahlberg , A. E. 1986. Expression of plasmid-coded mutant ribosomal RNA in E. coli: Choice of plasmid vectors and gene expression systems. Prog . Nucleic Acid Res. Mol . Bioi. 33 : 1 - 1 8 94. Szymkowiak, ch. , Reynolds, R. L . , Chamberlin, M . J . , Wagner, R . 1988. The tRNAG1u2 gene in the rmB operon of E. coli is a pre-requisite for correct RNase III processing in vitro. Nucleic A cids Res. 16:7885-99 9 5 . Talkad , V . , Achord, D . , Kennell , D . 1 978. Altered mRNA metabolism i n Ribonuclease III-deficient strains o f Es cherichia coli. J. Bacterial. 1 35 : 52841 96. Taschke, C . , K linkert , M. Q . , Walters, J . , Herrmann, R. 1986. Organization of the ribosomal RNA genes in Mycoplas ma hyopneumoniae: The 5S rRNA gene is separated from the 16S and 23S rRNA genes. Mol . Gen. Genet. 205:428-33 97. Traub, P . , Nomura, M. 1969. Structure and function of E. coli ribosomes studies in vitro. J. Mol. Bioi. 40:391-4 1 3 9 8 . Udem, S . A . , Warner, J. R . 1 973. The cytoplasmic maturation of a ribosomal precursor ribonucleic acid in yeast. J. BioI. Chern. 248 : 1 4 1 2-16 99. VoId, B . S . 1985 . Structure and organ ization of genes for transfer ribonucleic acid in Bacillus subtilis. Microbial. Rev. 49:7 1-80 100. Void, B. S . , Green, C. J . , Narasimhan, N . , Strem, M . , Hansen, J. N . 1988. Transcriptional analysis of Bacillus sub litis rRNA-tRNA operons . J. Bioi. Chern. 263 : 1 4485-90


rRNA PROCESSING 1 0 1 . Warner, 1. R. 1 989. Synthesis of ribo somes in Saccharomyces cerevisiae. Microbial. Rev. 53:256-71 102. Widom, R . L. , Jarvis, E. D . , La Fauci , G . , Rudner, R. 1988. Instability of rRNA operons in Bacillus subtilis. 1. Bacterial. 1 70:605- 1 0 1 0 3 . Wireman, J . W . , Sypherd, P. S . 1974. In vitro assembly of 30S ribosomal parti cles from precursor 1 6S RNA of Es cherichia coli. Nature 247:552-54 104. Woese, C. R. 1987. Bacterial evolution. Microbial. Rev. 5 1 :221-71 105 . Yamagishi, M., de Boer, H. A . ,

1 29

Nomura, M. 1987. Feedback regulation of RNA synthesis: A mutational altera tion in the anti-Shine Dalgarno region of the 16S rRNA gene abolishes regulation. J. Mol. Biol. 198:547-50 106. Young, R. A . , Steitz, 1. A. 1978. Com plementary sequences 1 700 nucleotides apart from a ribonuclease III cleavage site in Escherichia coli ribosomal pre cursor RNA. Proc. Natl. Acad. Sci. USA 75:3593-97 107. Young, R. A . , Steitz, 1. A. 1979. Tan dem promoters direct E. coli ribosomal RNA synthesi s . Cell 1 7:225-34

5'-Terminal processing of ribosomal 28S RNA.

On the biosynthesis of bacterial ribosomal RNA.

Cisplatin Targeting of Bacterial Ribosomal RNA Hairpins.

Structural signatures of thermal adaptation of bacterial ribosomal RNA, transfer RNA, and messenger RNA.

An overview of pre-ribosomal RNA processing in eukaryotes.

Ultraviolet irradiation inhibits RNA decay and modifies ribosomal RNA processing in Trypanosoma brucei.

Structural basis for regulation of ribosomal RNA 2'-o-methylation.

Association between central auditory processing mechanism and cardiac autonomic regulation.

Conservation of ribosomal RNA during compensatory renal hypertrophy. A major mechanism in RNA accretion.

Ribosomal RNA and translation.

In vitro synthesis and processing of ribosomal RNA in the housefly ovary.

Trypanosoma brucei mitochondrial ribosomal RNA synthesis, processing and developmentally regulated expression.

Ribotoxin recognition of ribosomal RNA and a proposal for the mechanism of translocation.

Changes in ribosomal RNA processing paths in resting and phytohemagglutinin-stimulated guinea pig lymphocytes.

Alterations in the processing of rat-liver ribosomal RNA caused by cycloheximide inhibition of protein synthesis.

Processing of ribosomal RNA in a temperature sensitive mutant of BHK cells.

Processing of the 5' end of Escherichia coli 16S ribosomal RNA.

In vivo disruption of Xenopus U3 snRNA affects ribosomal RNA processing.

Regulation of transcription of genes of ribosomal rna during amphibian oogenesis. A biochemical and morphological study.

Analysis of two domains with novel RNA-processing activities throws light on the complex evolution of ribosomal RNA biogenesis.

Cluster of genes in Escherichia coli for ribosomal proteins, ribosomal RNA, and RNA polymerase subunits.

Ribosomal and non-ribosomal RNA synthesis in vitro.

Regulation of ribosomal RNA synthesis in T cells: requirement for GTP and Ebp1.

Molecular diagnosis of periprosthetic joint infection by quantitative RT-PCR of bacterial 16S ribosomal RNA.