Biochimie (1991) 73, 845-851 © Soci6t6 franqaise de biochimie et biologie mol6culaire l Elsevier, Pads

845

The nuclear:organelle distribution of chloroplast ribosomal proteins genes. Features of a cDNA clone encoding the cytoplasmic precursor of LII PM Smooker*, J Schmidt, AR Subramanian** Max-Planck-lnstitut fiir Molekuliire Genetik, Abteilung Wittmann, Ihnestrasse 73, Berlin-Dahlem, Germany

(Received 13 November 1990; accepted 13 March 1991)

Summary m The majority of chloroplast ribosomal proteins are encoded in the nuclear genome. In order to characterize these proteins through their mRNA, we have previously constructed a spinach cDNA expression library and raised antisera to several spinach chloroplast ribosomal proteins. Here we describe the immune isolation of cDNA clones encoding protein L 11 and its chloroplast-targeting presequence. The cytoplasmic precursor form of LI 1 is 224 amino acid residues long (Mr 23 662); the mature LI 1 and the transit sequence are predicted to be of = 159 and = 65 residues, respectively. The predicted chloroplast L 11 is significantly longer than the E coli L11, but similar (in size) to archaebacterial and yeast cytoplasmic LI 1. In sequence it is closer to E coli LI 1 (54% identity) than to the archaebacterial (32%) or yeast (23%) proteins. These results and the conservation of the contexts of the 3 methyl modified residues found in E coli L11 are discussed in the light of the endosymbiont theory and nuclear relocation of the rplKAJL gene cluster. Spinacia oleracea (spinach) / rplll gene / transit sequence / endosymbionttheory

Introduction Genes encoding the protein and RNA components of chloroplast ribosomes are located in 2 cell compartments [ 1]. Coding sequences for the rRNA and = 22 ribosomal proteins (r-proteins) have been identified in the completely sequenced chloroplast genomes of 3 land plants [2]. The number of chloroplast r-proteins is estimated to be > 60, ie, somewhat greater than the number on the bacterial ribosome [3]. Therefore, _> 40 r-protein genes are located in another genome compartment in the cell, generally assumed to be in the nucleus. Chloroplast ribosomes are similar to eubacterial ribosomes in their motifs of overall structure and function [4]. According to the endosymbiont theory [5, 6], the organelles originated, mono- or polyphyletically, from photosynthetic bacteria over 1 x 109 *Present address: The Murdoch Institute, Royal Childrens Hospital, Parkville, Victoria 3052, Australia **Correspondence and reprints Abbreviations: r-protein, ribosomal protein; (prefixes L and S refer to the large and the small subunit, respectively); pfu: plaque forming unit. The nucieotide sequence data i~ this paper have been submitted to the EMBL/GenBank Data Libraries under the accession number X56615.

years ago. Chloroplasts display this bacterial origin in their ribosomes, which are markedly different in many details from the cytoplasmic 80S or mitochondrial 75S ribosomes found in the same plant cells. The percent of identity in primary/secondary structure of the rRNAs [7] and of > 30 of the chloroplast r-proteins identified so far [1] to eubacterial counterparts is strikingly high. Some recent experiments are, however, beginning to change this picture. For example, 3 r-proteins have been identified in chloroplast ribosomes which have no homologues in Escherichia coli [8-10]. Also several chloroplast r-proteins have N- and C-terminal extensions not found in eubacterial r-proteins [1]. These results have suggested significant divergence in the evolution of chloroplast ribosomes since the original endosymbiotic event. The assembly of chloroplast ribosomes is expected to be as precisely ordered as the assembly of E coli ribosomes [ 11 ]. However, only = 22 r-proteins will be synthesized in situ inside the chloroplast. Genes for these proteins are organized into clusters resembling the operons in E coli [1]; they may therefore possess regulatory controls as in E coli [12]. The remainder of the r-proteins have to be co-ordinately synthesized on the cytoplasmic 80S ribosomes and transported into the organelle. Thus an original set of eubacterial type genes in this system has now to be transcribed by 2

846

PM Smooker et al

different poiymerases (an eubacterial chloroplast polymerase, and pol II in the nucleus), processed differently (eg poly(A) addition) and translated on 2 kinds of ribosomes; yet the synthesis must be so regulated as to yield the same protein stoichiometry as in E coli. Even the 4 copies of the L12 protein found in E coli are maintained in the chloroplast ribosome [13,14]. We have previously worked in this system first to characterize many of the organelle encoded r-protein genes in maize (eg [15, 16]; reviewed in [1]) and later to characterize nuclear-coded r-proteins and their cDNAs that also encode transit peptides and terminal overhangs. This work has identified LI2, L I 3 , L21, L35 [ 17-20] and a protein of the 30S subunit (named PSrp-l) having no E coli homologue [8]. In the present report we identify a cDNA clone specifying the precursor of chloroplast L11 protein. In E coli L 11 is encoded in the rplKAR, cluster and is the most heavily modified of the r-proteins [21]. Its presence controls thiostreptone sensitivity [22]. These aspects and the question of nuclear:organelle gene allocation are discussed. M a t e r i a l s a n d methods

Isolation of ribosomal proteins and preparation of antisera Spinach chloroplast ribosomes were isolated and extracted, and the proteins size-fractionated and pooled as previously described [23]. The pool E from this experiment was subdivided by chromatography on carboxymethyl Sepharose CL6B [19]. Fractions 64-71 contained mainly 2 proteins, of M, 27 and 22 kDa. They were pooled (pool 44) and used to raise antisera in rabbits [19].

Spinach cDNA expression library The construction of the library used for these experiments has been previously described [17]. It was prepared from the total poly(A+) RNA of spinach (Spinacia oleracea, cv Matador) using the expression cloning vector Xgtl 1.

lmmunoscreening procedure A standard procedure was used as described before [ 19]. Antiserum was diluted 1:1000 in PBS (phosphate-buffered saline, containing 100 mM NaH2PO4, pH 7.2, 150 mM NaCI) and reacted at room temperature for 3.5 h with expressed proteins from plaques lifted and immobilized on nitrocellulose filters. After incubation with secondary antiserum (peroxidase-conjugated goat antirabbit IgG), bound antibodies were revealed by the developer 0.5 mg/ml 4-chloro-l-naphthol/0.03% H202 in PBS.

DNA manipulations and sequencing Phage DNA was isolated [24], and the cDNA inserts excised by restriction enzyme digestion and agarose gel electrophoresis and subcloned into the plasmid vector pT7/T3-19U (Pharmacia) by standard techniques [25]. To obtain clones with longer inserts, the library was rescreened with the isolated cDNA,

I

t

,t

0

,~

I

I 4(I)

200

.i

I

i

Mature LI I protein (-159 codlms)

Transit peptide

(-65 codons)

j

I gill

fail

A(n) CLI I-i

II

~p.

,, f"'1 "

"~r

', .....

'][q*J~I~[TA']C¢'~ ' ' ~['I~[T.J~CII~?./Ul/~-I * T i ' l ~ , - I ~ , i i . - i ~ . ~ ' . . ~ , : i : i ~ , ~ , $ C T g ~ l ~ ~ T G S L S S N P R N S L Q F L N B K L F L S P P S T S H R R L S

~

~

~

~

,

~

~

~

A w ~ ~ ~ r e A ~ c c ~ e r A e a ~ v ~ . A G ~ K D P ~ M I C K ~

G~VAI T ~

GK

1

T

1

D Q

L

R G

....~~TTCTCCA,,[CAIGTAG,[CAATIff*ATAT~AT,~A,

. . . . . 'IGTATAA~ACA~TATATT~TITIGT~*~AAAK4A

814

A I

r A T

~ ~

E K L P

190 I

~ erte~vm:A 670

D L N C

ITIGT,~41.IGA~AAGTII~ 834

Fig 1. Schematic diagram of 2 LI 1 cDNA clones (cL11-1 and cL 11-2), sequencing strategy, nucleotide sequence, and derived amino acid sequence. Some restriction sites, the transit peptide/mature protein coding regions, and poly(A) tail are shown. The context of the putative initiating ATG codon is underlined. Arrows with rectangles indicate sequencing with synthetic oligonucleotides designed with data from previous rounds of sequencing. random-primer labeled with [ct-32p]dATP. Standard procedures were used for plaque hybridization, washing and autoradiography [25]. The nucleotide sequence was determined by the dideoxy procedure, modified for use with double-stranded DNA templates [26].

Computer analyses A VAX8600/VMS computer and UWGCG programs were used for data analysis. The NBRF, Swiss Protein and RIBO (this institute) data banks were searched for identification of homologous sequence.

Materials Restriction enzymes, DNA ligase and other enzymes and reaction kits (random-primer labeling, DNA sequencing) were purchased from Boehringer, GIBCO-BRL, New England Biolabs or Pharmacia; radioactive isotopes, nitrocellulose membrane and conjugated antibodies were purchased from Amersham, Schleicher and Sch(ill and Dianova GmbH, respectively.

Results Isolation of truncated LI 1 clones Previous screening results have shown that our spinach c D N A expression library in ~ g t l l contains

847

L 11 of chloroplast ribosome is nucleus-encoded

double digestion with either of these enzymes and EcoRI, the clones were inferred to be identical. The nucleotide sequence of one of the inserts was obtained and is given in figure 1 (eL11-2). "Ilae 836 bp insert is identical to the earlier 404 bp in ~ e 3' part but contains an additional 432 bp sequence at the 5' part. The complete sequence is composed of a 675 bp open reading frame encoding a polypeptide of 224 residues. The reading frame begins at a Met codon (position 3-5) and terminates at the ochre codon (position 675-677) identified earlier. The sequence thus includes 22 nucleotides of a poly(A +) tail, 147 bp of a 3' non-coding region, and (only) 2 bp of a 5' noncoding region.

clones for chloroplast r-proteins at the frequency of one in 7500 plaques [17, 27]. A total o f - 1.5 x 105 pfu were screened in the present experiment. The positive plaques were purified through several further rounds of screening, and phage DNA was prepared. Digestion with EcoRl and gel electrophoresis revealed (the cDNAs were attached to the vector through EcoRl linkers) that the size of the largest insert obtained was only = 400 bp. This is smaller than the expected size for a full-size L11 precursor eDNA. The EcoRI fragment was cloned into the plasmid vector pT7/T3-19U and the nucleotide sequence determined. The insert was found to be exactly 404 bp long with the last 22 nucleotides comprising a poly(A) tail. It contained a reading frame beginning from the 5' end nucleotide and terminating at an ochre codon TGA at position 241-243, encoding a 80-residue long polypeptide (eLl 1-1 in fig 1). Since this reading frame (Pro-ProAla .....Phe) did rot begin with a Met residue, it was obviously a partial eDNA clone.

Identification of Ll l homology Computer search with the protein sequences in the NBRF/Swiss Protein/RIPe data bases revealed that a 140-residues long region in the predicted proteinsequence (from Ala-73 to Pro-212) has considerable identity to the reported L11 sequences from eubacteria, archaebacteria and yeast. E coli L11, the only prokaryotic L l l characterized at the protein level [28], is 141 residues long. The percent of identity with the LI 1 sequence of E coli is 54%; and the identical residues are distributed throughout the sequence (fig 2). Several amino acid replacements are of the conservative type (eg Ser -~ Thr, Asp -~ Gin etc). This result supports the view that the isolated clone encodes the

Isolation of clones with a complete reading ])'ame The cDNA library was again screened, utilizing the radiolabeled 404 bp insert as the probe. Positive plaques were identified, purified, and inserts characterized by restriction enzyme digestions. Two clones were identified with an insert size of = 850 bp. Each of them contained single internal BamHI and HindlIl sites and, on the basis of the fragments released by

Tramit l~ptide 1 i

ChlcmoLU

oooo

oo

++

oo

o

oo

o

+

oo

o

o+

o

ooo

6'?

MAQPLVAAPSSSS I TSPI PRKLCSSLLTPSSLSLSSNPRNSLQFLI4SKLFLSPPSTS ,- l t i t u r e p r o t e i n

+++



E.oon

m

-~~' ~~"

'""'



,.,,

Il

I''"

*1cl ^

Pl , 70 amino acids) at the N-terminus of the spinach protein, with a striking amino acid composition: great abundance of serine/threonine and proline and absence of tyrosine/tryptophan and acidic amino acids (fig 2).

LI 1 of chloroplast ribosome is nucleus-encoded

The presequences of all chloroplast r-proteins we identified lack Tyr/Trp, have high content of Ser/Thr/ Pro, and few Asp/Glu residues [8, 17-20]. We therefore propose this presequence to be the transit peptide, responsible for routing the L11 protein into the chloroplast. It would be removed, prior to assembly of L11 into the ribosome, by a common stromal processing peptidase (SPP) which has been recently described [35]. The point at which the L11 presequence would be cleaved is at present unknown. However, as the match with E coli LI 1 begins at Ala-73, the cleavage is unlikely to be after this residue. Recently Gavel and yon Heijne [36] have proposed a 'loosely conserved' motif, V/I-X-A/C,I,A, with a cluster of Arg residues 6-10 residues upstream, as the consensus cleavage site for transit peptides. Two such sites occur in the LI 1 precursor at positions Ala-65 and Met-66 (fig 2). Since the characteristic composition pattern of the transit peptide seems to end at Ser-62, the cleavage is very likely to be in this region. Its precise position will be known when the L I 1 protein is purified and sequenced at the N-terminus. Terminal overhangs The mature forms of nuclear-coded chloroplast rproteins generally contain peptide extensions (overhangs) at the ends of their E coli homologous region, the largest so far being a 66-residue overhang with an unusual composition in L21 [19]. From its predicted sequence, L I 1 contains a very basic 12-residue overhang at the C-terminus (fig 2), and possibly a shorter one at the N-terminus. Whether such overhangs have a specific chloroplast-related function is an open question at the moment. Work at the protein level is, however, necessary to confirm the proposed overhangs. The initiating Met codon in LI 1 As the proposed reading frame in eL11-2 begins at nucleotide number 3 (fig 1), it is unlikely to be a full length clone. The 5' noncoding regions in the previously characterized eDNA clones are much longer, eg 78 nucleotides in the LI2 clone [17]. The following observations, however, indicate the first Met codon in the reading frame to be an initiating Met codon: i), it has the context CAATGGCA, identical to the initiating context in the L12 and LI3 cDNAs [17, 18]. The methionine is followed by alanine as in all our eDNA clones ensuring a G at +4, a requirement for efficient initiation [37]. ii), The context is part of the most favored initiation context, AACAATGGC, for plant mRNAs [37]. This point will, however, be conclusively answered only by isolating a much longer eDNA clone, or clones of the L11 gene from a genomic library.

849

Chloroplast L11 protein In E coil L11 is the most heavily modified r-protein, containing a total of 9 methyl groups in 3 trimethylated residues, ie (t~Me3)Ala-1, (eMea)Lys-3 and (eMe3)Lys-39 [28]. The 3 residues (shown underlined in fig 2) as well as their immediate sequence contexts are conserved in chloroplast L l 1. It would therefore be of interest to analyze for posttranslational modification in the chloroplast L l l protein. We have previously reported the first exan-tple of a methyl modification in the chloroplast system in protein L2 [23]. Since most, if not all, r-protein modifications in E coli are apparently non-essential for ribosome function [38], it would be significant if they are conserved in the chloroplasts. The L 11 protein is localized near L 10 and L6 at the base of the L I2 stalk in the 50S ribosomal subunit [39], and functionally involved in the factor-dependent GTP hydrolysis catalyzed by this domain. Mutants lacking L11 protein have, however, been isolated from several eubacterial species: in Bacillus through thiostreptone resistance and in E coli as reverrants from erythromycin dependence (eg [22]). Thus L11 is probably involved only in enhancing or stabilizing this GTP domain. The presence of a rather conserved L l l pretein in chloroplasts would argue that such a function is important in itself and not dispensable. The known archaebactefial L I l ' s and the yeast cytoplasmic L11 (table I) are of the same chain length (= 165 residues) while the eubacterial L l l ' s are significantly shorter (141 residues). The precise length of the mature chloroplast LI 1 is presently unknown, but is likely to be = 160 residues (see earlier discussion). The latter is thus in the same size category as the archaebacterial LI l's. This finding may support the view that the chloroplast LI 1, like its archaebacterial counterpart, has retained the presumably long original size whereas the E coli L11 has been shortened under evolutionary pressure. The relocation of the L I 1 gene into the large plant nuclear genome would arguably reduce the evolutionary pressure for size reduction. The Ll l gene In E coli the L11 gene (rplK) is part of a cluster [12, 40] that includes 7 genes of the transcription translation apparatus (tufB-rplKAJL-rpoBC). In higher plants the rpoBC genes of this cluster are located in the organelle DNA [2], while the r-protein genes ([17]; and this paper) are relocated in the nucleus. As we have pointed out recently [1], this is an unusual case: more commonly the prokaryotic r-protein gene clusters are divided and distributed between the chloroplast and the nucleus. With the availability of

850

PM Smooker et al

probes for both LI1 (this paper) and L12 [17] it should be possible to investigate the chromosomal locations, sequence contexts and other features o f the r p l K A J L cluster in its nuclear situation.

Acknowledgments We thank P Padas for excellent technical assistance, K Giese for constructing the cDNA library, and K yon Knoblauch, S Bantz and KH Rak for assistance with protein purification/ antisera preparation.

References I

2 3 4

5 6 7

8

9 10

!1

12

Subramanian AR, Smooker PM, Giese K (1990) Chloroplast ribosomal proteins and their genes. In: The Ribosome: Structures, Function, and Evolution (Hill WE, Dahlberg A, Garrett RA, Moore PB, Schlessinger D, Warner JR, eds) Am Soc Microbioi Publ, Washington DC, 655-663 Sugiura M (1989) The chloroplast chromosomes in land plants. Annu Rev Cell Biol 5, 51-70 Wittmann HG (1982) Components of bacterial ribosomes. Annu Rev Bio,.'hem 51,155-183 Boynton JE, Gillham NW, Lambowitz AM (1980) Biogenesis of chloroplast and mitochondrial ribosomes. In: Ribosomes: Structure, Function and Genetics (Chambliss G, Craven GR, Davies J, Davis K, Kahan L, Nomura M, eds) University Park Press, Baltimore, MD, 903950 Bogorad L (1975) Evolution of organelles and eukaryotic genomes. Science 188, 891-898 Gray MW (1989) The evolutionary origins of organelles. Trends Genet 5,294-299 Brimacombe R, Greuer B, Mitchell P, Osswald M, RinkeAppel J, Schiller D, Stade K (1990) Three-dimensional structure and function of Escherichia coil 16S and 23S rRNA as studies by cross-linking techniques. In: The Ribosome: Structure, Function, and Evolution (Hill WE, Dahlberg A, Garrett RA, Moore PB, Schlessinger D, Warner JR, eds) Am Soc Microbiol Publ, Washington DC, 93-106 Johnson CH, Kruft V, Subramanian AR (1990) Identification of a plastid-specific ribosomal protein in the 30S subunit of chloroplast ribosomes and isolation of the cDNA clone encoding its cytoplasmic precursor. J Bioi Chem 265, 12790-12795 Gantt JS (1988) Nucleotide sequences of cDNAs encoding four complete nuclear-coded plastid ribosomal proteins. Curr Genet 14, 519-528 Zhou DX, Mache R (1989) Presence in the stroma of chloroplast of a large pool of a ribosomal protein not structurally related to any Escherichia coli ribosomal protein. Mol Gen Genet 219, 204-208; Erratum, ibid (1990) 223, 167 Herold M, Nierhaus KH (1987) Incorporation of six additional proteins to complete the assembly map of the 50S subunit from Escherichia coli ribosomes. J Biol Chem 262, 8826--8833 Nomura M, Gourse R, Baughman G (1984) Regulation of the synthesis of ribosomes and ribosomal components. Annu Rev Biochem 53, 75-117

13 Subramanian AR (1975) Copies of proteins L7 and LI2 and heterogeneity of the large subunit of Escherichia coli ribosome. J Mol Bio195, 1-8 14 Bartsch M, Kimura M, Subramanian, AR (1982) Purification primary structure, and homology relationships of a chloroplast ribosomal protein. Proc Nati Acad Sci USA 79, 6871-6875 15 Subramanian AR, Steinmetz A, Bogorad L (1983) Maize chloroplast DNA encodes a protein sequence homologous to the bacterial ribosome assembly protein $4. Nucleic Acids Res 15, 5277-5286 16 Markmann-Mulisch U, Subramanian AR (1987) Nucleotide sequence and linkage map position of the genes for ribosomal proteins L14 and $8 in the maize chloroplast genome. Eur J Biochem 170, 507-514 17 Giese K, Subramanian AR (1989) Chloroplast ribosomal protein L12 is encoded in the nucleus: construction and identification of its cDNA clones and nucleotide sequence including the transit peptide. Biochemistry 28, 3525-3529 18 Phua SH, Srinivasa BR, Subramanian AR (1989) Chloroplast ribosomal protein LI3 is encoded in the nucleus and is considerably larger than its bacterial homologue. J Biol Chem 264, 1968-1971 19 Smooker PM, Kruft V, Subramanian AR (1990) A ribosomal protein is encoded in the chloroplast DNA in a lower plant but in the nucleus in angiosperms. Isolation of the spinach L21 protein and cDNA clone with transit and an unusual repeat sequence. J Biol Chem 265, 1669916703 20 Smooker PM, Choli T, Subramanian AR (1990) Ribosomal protein L35: identification in spinach chloroplasts and isolation of a. cDNA clone encoding its cytoplasmic precursor. Biochemistry 29, 9733-9736 21 Wittmann-Liebold B (1986) Ribosomal proteins: their structure and evolution. In: Structure, Function and Genetics of Ribosomes (Hardesty B, Kramer G, eds) Springer-Veflag, Berlin, 326-361 22 StOffler (3, Cundliffe E, St6ffler-Meilicke M, Dabbs ER (1980) Mutants of Escherichia coil lacking ribosomal protein LI I. Biol Chem 255, 10517-10522 23 Kamp RM, Srinivasa BR, yon Knoblauch K, Subramanian AR (1987) Occurrence of a methylated protein in chloroplast ribosomes. Biochemistry 26, 5866-5870 24 Patterson TA, Dean M (1987) Preparation of high titer lambda phage lysates. Nucleic Acids Res 15, 6298 25 Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning. Cold Spring Harbor Laboratory Press, NY, 2nd edn 26 Chen EY, Seeburg PH (1985) Supercoil sequencing: a fast and simple method for sequencing plasmid DNA DNA 4, 165-170 27 Giese K (1990) PhD Dissertation, Free University of Berlin, Germany 28 Dognin M J, Wittmann-Liebold B (1980) Purification and primary structure determination of the N-terminal blocked protein L 11 from Escherichia coli ribosomes. Eur J Biochem 112, 131-151 29 Sor F, Nomura M (1987) Cloning and DNA sequence determination of the L I 1 ribosomal protein operon of Serratia marcescens and Proteus vulgaris: translational feedback regulation of the Escherichia coli L11 operon by heterologous LI 1 proteins. Mol Gen Genet 210, 52-59 30 Shimmin LC, Dennis PP (1989) Characterization of the LI 1, LI, LI0 and L12 equivalent ribosomal protein gene cluster of the halophilic archaebacterium Halobacterium cutirubrum. EMBO J 8, 1225-1235

L11 of chloroplast ribosome is nucleus-encoded 31

Amdt E, Weigel C (1990) Nucleotide sequence of the genes encoding the L l l , LI, L10 and LI2 equivalent ribosomal proteins from the archaebacterium Halobacterium marismortui. Nucleic Acids Res 18, 1285 32 Ramirez C, Shimmin LC, Newton CH, Matheson AT, Dennis PP (1989) Structure and evolution of the LI 1, L1, L10 and L12 equivalent ribosomal proteins in eubacteria, archaebacteria, and eucaryotes. Can J Microbiol 35, 234-244 33 PucciareUi MG, Remacha M, Vilella MD, Ballesta JPG (1990) The 26S rRNA binding ribosomal protein equivalent to bacterial protein L l l is encoded by unspliced duplicated genes in Saccharomyces cerevisiae. Nucleic Acids Res 18, 4409--4416 34 Dayhoff MO (1978) Atlas of Protein Sequence and Structure. Natl Biomed Res Found, Washington, DC, vol 5 35 Smeekens S, Weisbeek P, Robinson C (1990) Protein transport into and within chloroplasts. Trends Biochem Sci 15, 73-76

851

36 Gavel Y, yon Heijne G (1990) A conserved cleavage-site motif in chloroplast transit peptides. FEBS Lett 261, 455-458 37 Liitcke HA, Chow KC, Mickel FS, Moss KA, Kern HF, Scheele GA (1987) Selection of AUG initiation codons differs in plants and animals. EMBOJ 6, 4348 38 Isono K (1980) Genetics of ribosomal proteins and their modifying and processing enzymes in Escherichia coli. In: Ribosomes: Structures, Function and Genetics (Chambliss G, Craven GR, Davies J, Kahan L, Nomura M, eds) University Park Press, Baltimore, MD, USA, 641--669 39 Walleczek J, Schiller D, StOffler-Meilicke M, Brimacombe R, St6ffler G (1988) A model for the spatial arrangement of the proteins in the large subunit of the Escherichia coli ribosome. EMBO J 7, 3571-3576 40 Bachmann BJ (1990) Linkage map of Escherichia coli K- 12, edition 8. Microbiol Rev 54, 130-197