Proc. Nat. Acad. Sci. USA Vol. 72, No. 7, pp. 2743-2747, July 1975 Genetics
Cluster of genes in Escherichia coli for ribosomal proteins, ribosomal RNA, and RNA polymerase subunits* (transducing phage X/in vitro protein synthesis/protein identification/DNA.RNA hybridization)
LASSE LINDAHL, S. RICHARD JASKUNAS, PATRICK P. DENNIS, AND MASAYASU NOMURA Institute for Enzyme Research and Departments of Biochemistry and Genetics, University of Wisconsin, Madison, Wisc. 53706
Communicated by James F. Crow, April 21,1975
The transducing phage Xri$18 isolated by ABSTRACT Kirschbaum and Konrad [(1973) J. Bacteriol. 116, 517-526] was found to carry structural genes for several 50S ribosomal proteins and 16S and 23S rRNA. It has previously been demonstrated [Kirschbaum & Scaife (1974) Mol. Gen. Genet. 132, 193-201] that this phage carries genes for the DNA-dependent RNA polymerase (nucleosidetriphosphate:RNA nucleotidyltransferase; EC 2.7.7.6) subunits # and '. Thus, the region of the E. coli chromosome carried by XriP18 contains a cluster of genes essential for transcription and translation.
Knowledge of the genetic organization of the genes for ribosomal components is important for understanding the mechanism of regulation of ribosome biosynthesis. It is known that many ribosomal protein (r-protein) genes in Escherichia coli are clustered in the str-spc region of the chromosome at 64 min (for a review, see ref. 1). Many r-protein genes in this region have been identified by isolating several transducing phages that carry these genes (ref. 2, and our unpublished experiments). For example, one such phage, Xfus2, carries approximately 30 r-protein genes and the gene for an elongation factor EF-G (cited in ref. 2). However, not all the rprotein genes map at the str-spc region. We have previously shown that at least one r-protein gene, the structural gene for S18, maps outside this region (3, 4). Thus, we have initiated a search for additional structural genes for r-proteins outside the str-spc region. One suspected region was near the rif gene at 79 min, since experiments by Friesen and his coworkers indicated that relC, which maps near rif, could be a structural gene for a 50S r-protein (5). As will be described in this paper, we have in fact found that the transducing phage Xrifdl8 isolated by Kirschbaum and Konrad (6) carries the structural genes for r-proteins L1, L7, L12, and probably also the genes for L8, L10, and Lii, in addition to the gene rif. This finding further prompted us to examine the possibility that some other essential genes related to transcription or translation are also located in this same region. We found that genes for ribosomal RNA (rRNA) are also carried by the same transducing phage. It has previously been shown that this phage carries genes for the DNA-dependent RNA polymerase (nucleosidetriphosphate:RNA nucleotidyltransferase, EC 2.7.7.6) subunits , and jY' (7). Thus, genes for several r-proteins, rRNA, and RNA polymerase subunits #3 and ,B', all are clustered in the small chromosomal region that is carried by Xrifdl8. MATERIALS AND METHODS Transducing phage XCI857S7rifdi8 (called Xrifdl8 in this paper) was prepared from a strain H105 (XCI857S7, Abbreviations: r-protein, ribosomal protein; UV, ultraviolet. * This is paper no. 1846 from the Laboratory of Genetics. 2743
Xrifdl8) This strain (6) was obtained from Dr. J. B. Kirschbaum. Another phage XCI857St68 h8OglyT su+36 (called Ah80glyT) was prepared from a strain JC100 that carried this phage (8) and a helper phage XCI857h80St68. The strain was obtained from Dr. J. Carbon. Transducing phage Xfus2 was isolated in this laboratory (cited in ref. 2). XCI857S7 (called X) was used as a control phage. These phages were prepared by heat induction and purified by CsCl equilibrium centrifugation (2, 9). Preparation of antisera against individual 30S proteins was described previously (10). Antiserum against pure Li was prepared in the same way. Antiserum against L12 was a gift from Dr. H. Weissbach. This antiserum reacts with both L7 and L12, but no other r-proteins. L7 is an acetylated form of L12 (11). When we do not distinguish the two proteins, we use L7/Li2 to indicate these proteins. The nomenclature of r-proteins is according to ref. 11. Other methods are described in the table and figure legends. RESULTS
Stimulation of synthesis of ribosomal proteins by XriP1I8 We measured the stimulation of r-protein synthesis in E. coli cells after infection by the transducing phage XrifdI8. The synthesis of r-proteins by the host r-protein genes was reduced by prior UV-irradiation of the host cells. For comparison, we also infected the irradiated cells with X, which does not carry any r-protein genes, and Xfus2, which carries approximately 30 r-protein genes. The experimental procedure was similar to that used previously (2) and is described in the legend to Table 1. The data on the proteins whose synthesis was apparently stimulated by Xrifdl8 relative to cells infected with the control phage A are shown in Table 1. Among these proteins, the apparent stimulation of S5, S6, and L9 is probably due to contamination of these protein samples with Lii, L12, and L8, respectively (see legend to Table 1). In this connection, it should be noted that the synthesis of Lii was apparently stimulated by Xfus2, but stimulation of the synthesis of S5 by this phage was greater. The presence of the gene for S5 in Xfus2 as well as related transducing phages, Xspcl and Xspc2, has been proven previously (ref. 2, and our unpublished experiments). Incomplete separation of LII and S5 in the two-dimensional electrophoresis could explain the apparent stimulation of Lii by Xfus2. Thus, our conclusion is that Arifdl8 stimulates synthesis of LI, L7, L8, L10, Lii, and L12. These proteins are underlined in Table 1. As discussed in the previous paper (2), we cannot eliminate the possibility that stimulation of the synthesis of r-pro-
2744
Genetics: Lindahl et al.
Proc. Nat. Acad. Sci. USA 72 (1975)
Table 1. Stimulation of the synthesis of r-proteins by Xrifd18 3H/'4C ratio Proteins
S5 S6 Li L7 L8 L9 L10 L11 L12 Others
X
0.04 0.19 0.02 0.03 0.10 0.10 0.04 0.06 0.04 0.01-2.5
Xfus2 5.5 1.6 0.09 0.10 0.15 0.13 0.07 2.3 0.14 0.07-13.4
Xrifd 18 2.6 3.0 9.5 34.6 6.4 1.1 22.4 8.7 10.8
0.09-3.0
E. coli cells pre-labeled with ['4C]leucine were irradiated with UV light and infected with phages as previously described (2) except that the UV-sensitive bacterial strain was a Xpapa lysogen of strain 159 (26). The [3H]leucine was added 30 min after phage infeetion and the incorporation was stopped after 10 min. The cells were lysed and proteins were extracted and separated by 2-dimensional gel electrophoresis (27). The 3H/14C ratio of each r-protein spot, except S1 and L31, was determined. Details of the procedures were described previously (2, 28). The data are shown only for the proteins whose synthesis was apparently stimulated by Xrif18 as judged by 3H/14C ratio relative to the control 3H/14C ratio obtained after A infection. As described in the text, proteins S5 and L11, L8 and L9, and S6 and L12 overlap partially in the 2-dimensional gel electrophoresis, and this probably accounts for apparent stimulation of S5, S6, and L9 by Xrifl8.
teins in these experiments was a secondary effect of some phage-coded proteins on residual r-protein synthesis in irradiated cells. Also, identification of the stimulated protein was done only by co-electrophoresis with mature proteins, and more rigorous experiments are required for definitive identification. However, it is notable that there were no proteins whose synthesis was stimulated by both Xrifdl8 and Xfus2 except for S5 and Lii, which have been discussed above. It is highly likely that the observed stimulation of L1, L7, L8, L10, Lii, and L12 is due to the presence of the genes for these proteins on the Xrifdl8 phage genome. The rif gene is located at 79 min on the E. coli genetic map (12), and is very closely linked to genes supM and glyT that code for a tyrosine and a glycine tRNA, respectively. A transducing phage Xh80glyT isolated by Carbon and his coworker carries the supM and glyT genes (8). However, purified Xh80glyT failed to show stimulation of the synthesis of any r-proteins in UV-irradiated cells (data not shown). In vitro synthesis of 50S ribosomal proteins with XriP1i8 DNA. Synthesis of several 30S r-proteins in vitro was previously described using either E. coli DNA (13) or Xspci DNA (2) as template. We used the same DNA-dependent protein synthesizing system to see whether DNA extracted from Xrifdl8 can in fact code for the 50S r-proteins that were stimulated in the in vivo experiments described above. Proteins were synthesized in the presence of [a5S]methionine and DNA from Xrifdl8 as a template. For comparison, DNAs from X, Xfus2 and Xh80glyT were also used as templates. We used simultaneous reconstitution of 30S and "50S" ribosomal subunits to recover radioactive r-proteins synthesized in vitro (refs. 2, 13, and see the legend to Fig. 1).
Xh8Ogl~yT
Cc)
Cd)
30S4
X
Xfus 2
O50$
800n 30S"50SO
QD
60 40-
0.1
20
5
10
15 5 FRACTION NUMBER
10
15
FIG. 1. Ability of various DNAs to stimulate synthesis of rproteins. [35S]Methionine-labeled protein was synthesized in a DNA-dependent cell-free system, using DNA from X, Xrifdl8, Xh80glyT, and Xfus2 as template. To recover [35S]r-proteins synthesized, we simultaneously reconstituted 30S and "50S" particles (see text). This was performed as described previously (2, 13) with the following modification: the dialysis against TMAI-urea [10 mM Tris-HCl (pH 7.4), 10 mM MgCl2, 30 mM NH4Cl, 6 mM 2mercaptoethanol, 8 M urea] and the DEAE-cellulose steps were omitted. Nonradioactive protein and RNA prepared by the ureaLiCl method (29) from 70S ribosomes were added to give a final concentration of 0.1-0.5 mg/ml of rRNA and a ratio of 1.5 equivalents of protein per equivalent of RNA. Ionic conditions were adjusted to 30 mM Tris-HCl (pH 7.4), 20 mM MgCl2, and 0.33 M KCl (reconstitution buffer), and the mixture was incubated for 1 hr at 400. After chilling to 0° the mixture was centrifuged through a 5-20% sucrose gradient in reconstitution buffer at 24,000 rpm in a Spinco SW27 rotor for 11 hr. The gradient was then pumped through a flow photometer monitoring A2m6. and fractionated. An aliquot of each fraction was analyzed for radioactivity. A260 nm
(-) radioactivity (- 0 -).
It should be noted that "50S" particles reconstituted under the conditions described are not identical to the standard 50S subunits, but they contain most of 50S r-proteins and little or no SOS r-protein (our unpublished experiments). The technique is therefore specific in detecting radioactive 50S r-proteins synthesized in vitro. As shown in Fig. 1, some of the proteins synthesized in the presence of Xrifdl8 DNA were found in the reconstituted "50S" particles, but the amount of [s5S]proteins in the reconstituted 30S particles was insignificant. As expected (ref. 2, and our unpublished experiments), Xfus2 DNA stimulated the synthesis of both 30S and 50S r-proteins. In contrast, no significant amount of [a5S]protein was found in the reconstituted particles when DNA from X or Ah80glyT was used as a template. Radioactive proteins synthesized in the presence of Xrifdl8 or Xfus2 DNA were then extracted from the reconstituted particles (combined 30S and "50S" particles) and analyzed further. The radioimmunodiffusion method (2, 13) was employed, using antisera against each of the individual 30S r-proteins (except S1 and S17) and against Li and L12.
Genetics: Lindahl et al.
Proc. Nat. Acad. Sci. USA 72 (1975)
2745
b
a
L12! 4
L8,L1O,L11 I,
FIG. 2. Identification by radioimmunodiffusion of r-proteins synthesized in vitro. Proteins were extracted from the reconstituted particles (Fig. 1) and tested by radioimmunodiffusion (2, 13). Each set of pictures shows a gel (left) and an autoradiogram of the gel (right). The center wells received antiserum against (a) L12 or (b) Li. The top and bottom wells received protein extracted from the reconstituted ribosomes; top, )rijfd8 DNA as template, and bottom Xfus2 DNA as template. The upper right well of (b) received purified L1. Anti-L1 antiserum showed two precipitin bands with a 70S r-protein mixture. We believe that one of the precipitin bands is probably due to a reaction with an aggregate containing Li.
By this method
found that protein synthesized with Xrifdl8 DNA as template contained Li and L7/L12 or proteins immunologically related to them (Fig. 2), but none of the other proteins tested for. By contrast, neither Li nor L7/Li2 was detected among proteins synthesized with Xfus2 DNA as template (Fig. 2). Instead, a number of 30S proteins were detected in this case (data not shown). Radioactive proteins recovered from the reconstituted particles were also analyzed by one-dimensional polyacrylamide gel electrophoresis at pH 4.5 (14). An autoradiogram of the sliced gels is shown in Fig. 3. Five radioactive protein bands were clearly visible in the sample with Xrifdl8 DNA as a template. The upper two bands appeared at the same positions as the reference radioactive L7 and L12 bands. Apparently, both acetylated (L7) and nonacetylated (Li2) forms of the protein were synthesized in the in vitro system. However, rigorous proof of the presence of the acetyl group must await further experiments. Identity of the remaining three radioactive protein bands has not been established. Quite a few 50S proteins migrate in this area. However, it is noted that one of the three bands corresponded to the position of Li, and the remaining two bands were close to the positions where proteins L8, L10, and Lii migrate. Thus it seems likely that one or more of the three other proteins stimulated in the in vvo experiments (L8, LIO, and Lii) were also synthesized in the in vitro system. Many radioactive protein bands were seen in the sample with Xfus2 DNA as a template, but none of them corresponded to the position of L7, L12, or LI (Fig. Sc). To give further proof for the in vitro synthesis of L7/L12 directed by Xrifdl8 DNA, we digested the in vitro synthesized L7/Li2 with trypsin and a-chymotrypsin and analyzed the resultant [5sS]methionine-containing peptides. As shown in Fig. 4, the in vitro synthesized L7/Li2 gave a peptide map very similar to that of the in vivo synthesized L7/ L12. From all these experimental results, we conclude that Xrifdl8 DNA directs synthesis of proteins in vitro that are very similar or identical to protein L7/Li2 and LI. Hence, the structural genes for L7/L12 as well as LI are present on Xrifdl8 DNA. we
Presence of rRNA genes in XriP1i8. We have found that the Xrifdl8 chromosome contains bacterial DNA that is homologous to stable RNA in E. coli. In DNA-RNA hybridization experiments using [3H]uracil-la-
"w.. ,4
I.
abi a b c d e FIG. 3. Polyacrylamide gel electrophoresis of [35S]proteins synthesized in vitro. [35S]Proteins synthesized in vitro in the presence of DNA from Xrifdl8 (b) and Xfus2 (c) were extracted from reconstituted ribosomes and subjected to gel electrophoresis on 10% polyacrylamide gel in the presence of 8 M urea at pH 4.5 (14). For comparison, radioactive 50S proteins prepared from cells grown in the presence of [35S]methionine (d) and a mixture of pure L7 and L12 purified from the above 50S [ssS]proteins (e) were also analyzed in the same way. After the run, gels were processed as described previously (30, 31). The autoradiograms are shown in the figure (b-e). A stained gel of total 70S proteins is-also shown (a). The arrows indicate the positions of proteins L7, L12, L1, L8, L10, and L11. The L8, L10, and L11 bands are not resolved from other proteins on the reference stained gel.
beled stable RNA and an excess of Xrifdl8 DNA, as much as 70-80% of the input RNA can hybridize to Xrifdl8 DNA, but very little or none to Xfus2 DNA. Since 16S and 23S rRNA account for about 85% of the total stable RNA in the cell (15, 16) the results indicate that both 16S and 23S rRNA genes are carried on Xrifdl8 chromosome. Furthermore, as can be seen in Fig. 5, purified nonradioactive 16S and 23S RNA effectively competed with 90% or more of the [3H]uracil-labeled stable RNA that hybridized to Xrifdl8 DNA. The residual 10% radioactive RNA not affected by competition may represent 5S RNA and possibly tRNA. Hybridization of purified radioactive 16S RNA and 23S RNA to Xrifdi8 DNA was also demonstrated. These results clearly demonstrate that Xrifdi8 carries DNA homologous to 16S and 23S rRNA. In addition, hybridization experiments using a constant amount of Xrifdl8 DNA and excess amounts of RNA suggest that the number of copies of these genes on the phage is probably one (see the legend to Fig. 5). In preliminary experiments, we have also observed that Xrifdl8 stimulates the synthesis of 23S RNA and 16S RNA (or their "precursor" RNAs) in UV-irradiated E. coli cells. It therefore appears that a complete set of rRNA genes is carried by Xrifdl8, and that these genes map near the rif locus on the E. coli chromosome. DISCUSSION Results presented in this paper prove that the structural genes for 50S proteins L7/L12 and Li and for 16S and 23S rRNA are carried by Xrifdl8. The data also indicate that in all likelihood the structural genes for L8, L10, and Lii are also carried by Xrifdl8.
2746
Genetics: Lindahl et al. Rf 0.60.50.4
Proc. Nat. Acad. Sci. USA 72 (1975) 1004
a Rf
b
0.3F
t 0.8 30.6
0.2 F 0.1 F
J0.4
0
0
1 8
?
0
60_ _
0.2
- + I
4 3
I %_ I,
1 ° cm cm
I
I
L
X
k
a,
5 4 3 2 10
FIG. 4. Peptide maps of in vitro and in vivo synthesized L7/ L12. (a) In vitro L7/L12: [s5Sjmethionine labeled r-protein synthesized from Xrifdl8 was isolated and fractionated by gel electrophoresis (compare Fig. 3b). The L7 and L12 bands were cut out and minced, and the pH was adjusted to 8 with 1 M NH4HCO3. Trypsin was added to a final concentration of 0.1 mg/ml, and the mixture was incubated overnight at 37°. The liquid was removed, and the gel was suspended in 0.1 M NH4HCO3 containing trypsin (0.1 mg/ml) and again incubated overnight at 37°. The liquid was again collected and combined with the first extract. Then a-chymotrypsin was added at a ratio of protein to chymotrypsin 1:100. The mixture was incubated for 4 hr at 37° and lyophilized. (b) In vivo L7/L12: 505 subunits were isolated from E. coli PR13 grown in the presence of [ssS]methionine. Proteins L7 and L12 were extracted from the 50S by the salt-ethanol method (32) and subjected to electrophoresis as in Fig. 3. The L7 and L12 proteins were extracted and digested with trypsin and chymotrypsin as above. The digests were fractionated in two dimensions on silica gel thin-layer plates by electrophoresis at pH 4.7 followed by chromatography using a butanol-water-acetic acid solvent as previously described (4) except that the electrophoresis was for 2 hr at 750 V, and chromatography was for 14 hr (a) or 6 hr (b). Electrophoresis: horizontal; chromatography: vertical.
Protein L7/L12 is important in the ribosome-associated GTPase reactions that are dependent on various protein factors such as elongation factor G or elongation factor T (for a review, see ref. 17). Using partially protein-deficient 50S "core" particles, it has been recently demonstrated that L7/L12, L1O, and Lii are interrelated in the reconstitution of 50S subunits (ref. 18, see also ref. 17). It is interesting to note that all of the genes for these proteins are clustered near rif as indicated in the present results. Xrifdl8 carries the genes for fB and iB' subunits of RNA polymerase (7). We have confirmed the above conclusion by showing stimulation of the synthesis of f# and fl' subunits in UV-irradiated cells infected with Xrifdl8 (Jaskunas, Nomura, and Burgess, unpublished experiments). Both the r-protein genes and the rRNA genes we have found on this phage are probably located very close to the genes for /3 and ,B'. Heteroduplex analysis on Xrifdl8 DNA shows that about 52% of A genome is deleted and substituted by bacterial DNA that is about 59% of a A unit in length (our unpublished experiments). The sum of the molecular weights of the (3 and (3' subunits of RNA polymerase is about 320,000 (19) and the sum of the molecular weights of 50S proteins Li, L7/L12, L8, L10, and LII is about 100,000 (11). Therefore, the genes for these proteins occupy a minimum of about.11,700 base pairs of DNA. Combined with an additional 5000 base pairs for the 16S and 23S rRNA genes (2022), this represents a minimum of 16,700 base pairs or 36% of a A unit of DNA; this represents greater than 60% of the total bacterial DNA (59% of a A unit) in Xrifdl8. Such a calculation indicates that the r-protein genes as well as rRNA genes studied in this work are physically very close to the genes for the RNA polymerase subunits. Since neither the rprotein genes nor rRNA genes were used to select Xrifdl8,
20 20
0
25 5.0 7.5 COMPETING rRNA (C,/ASSAY)
1.0
FIG. 5. Hybridization of [3H]uracil-labeled stable RNA to Xrifdl8 DNA in the presence of nonradioactive rRNA as competitor. The [3H]uracil-labeled stable RNA was prepared as described previously (Dennis and Nomura, manuscript submitted for publication). Its specific activity was 9.5 X 104 cpm/hg. In the hybridization assays 0.5 gg of 3H-labeled stable RNA was hybridized to 20 sg of DNA prepared from Xrifdl8 phage. Various amounts of nonradioactive 16S and 23S rRNA were added as indicated. Hybridization was performed according to a method (Dennis and Nomura, manuscript submitted) similar to that of Gillespie and Spiegelman (33). Relative amounts of [3H]RNA hybridized are shown in the figure. Radioactivity in hybrid at 100% of maximum was 2.48 X 104 cpm. Competitor RNA was obtained from E. coli AS19 after 30 min of rifampicin treatment to eliminate unstable RNA. Stable RNA was prepared by a phenol method and separation of 16S and 23S rRNA from 4S and 5S RNA was accomplished by passage through Sephadex G-200. At saturation, approximately 0.35 zg of stable RNA complexed to 20 ,ug of Xrifdl8 DNA and saturation requires an input of more than 0.5 gg of RNA. The Xrifdl8 phage preparation was about 50% contaminated with helper phage. If the transducing phage carries a complete copy of 16S and 23S rRNA genes, it would be expected that about 5% of the pure Xrifdl8 DNA or about 2.5% (0.5 gg) of the DNA used for hybridization would be homologous to rRNA. The observed value (0.35 Mg) was close to the expected value, supporting the presence of one set of 16S and 23S rRNA genes on the phage.
the proximity relation of these genes to the RNA polymerase genes found in the transducing phage probably applies to their "natural" locations on the E. coli chromosome. Deonier and his coworkers previously suggested that a set of rRNA genes (rrnB) may be located between argC and glyT, which is close to rif (20). It remains to be seen whether the gene set found on Xrifdl8 is identical to rrnB studied by these workers. The rate of synthesis of both rRNA and r-proteins is regulated and is related to growth rate in various growth media (for a review, see ref. 15). Synthesis of RNA polymerase ,B and fl' subunits is also coordinately regulated and appears to vary depending on growth media (23-25). It remains to be seen whether the clustering of r-protein, rRNA, and RNA polymerase genes has any functional significance in the regulation of the expression of these important genes. The observed clustering might also be related to a topographical feature of the E. colh chromosome; the genes that are transcriptionally active, such as rRNA genes, r-protein genes, and RNA polymerase genes, may be located at a place where continuous transcription by RNA polymerases could be efficiently maintained.
Note Added in Proof. In collaboration with R. Burgess, we have also discovered that Xrifdl8 carries a gene for elongation factor Tu.
Genetics: Lindahl et al. We thank M. Stroud, L. Sadowski, K. Ryan, G. D. Strycharz, W. Taylor, Dr. J. Ingversen, and Dr. F. Harada for their assistance; Drs. J. B. Kirschbaum and J. Carbon for providing us with bacterial strains and transducing phages; and Dr. H. Weissbach for providing us with antiserum against L12. This work was supported in part by the College of Agriculture and Life Sciences, University of Wisconsin, and by grants from the National Science Foundation (GB31086) and the National Institute of General Medical Sciences (GM-20427). L.L. is a recipient of a Postdoctoral fellowship from the European Molecular Biology Organization.
1. Jaskunas, S. R., Nomura, M. & Davies, J. (1974) in Ribosomes, eds. Nomura, M., Tissieres, A. & Lengyel, P. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 333-368. 2. Jaskunas, S. R., Lindahl, L. & Nomura, M. (1975) Proc. Nat. Acad. Sci. USA 72,6-10. 3. Bollen, A., Faelen, M., Lecocq, J. P., Herzog, A., Zengel, J., Kahan, L. 4 Nomura, M. (1973) J. Mol. Biol. 76,463-472. 4. Kahan, L., Zengel, J., Nomura, M., Bollen, A. & Herzog, A. (1973) J. Mol. Biol. 76, 473-483. 5. Friesen, J. D., Fiil, N. P., Parker, J. M. & Haseltine, W. A. (1974) Proc. Nat. Acad. Sci. USA 71, 3465-3469. 6. Kirschbaum, J. B. & Konrad, E. B. (1973) J. Bacteriol. 116, 517-526. 7. Kirschbaum, J. B. & Scaife, J. (1974) Mol. Cen. Cenet. 132, 193-201. 8. Squires, C., Konrad, B., Kirschbaum, J. & Carbon, J. (1973) Proc. Nat. Acad. Sdi. USA 70,438-441. 9. Miller, J. H. (1972) Experiments in Molecular Genetics (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.). 10. Kahan, L., Held, W. A. & Nomura, M. (1974) J. Mol. Biol. 88, 797-808. 11. Wittmann, H. G. (1974) in Ribosomes, eds. Nomura, M., Tiss- ieres, A. & Lengyel, P. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 93-114. 12. Taylor, A. L. & Trotter, C. D. (1972) Bacteriol. Rev. 36, 504524.
Proc. Nat. Acad. Sci. USA 72 (1975)
2747
13. Kaltschmidt, E., Kahan, L. & Nomura, M. (1974) Proc. Nat. Aced. Sci. USA 71, 446-450. 14. Leboy, P. S., Cox, E. C. & Flaks, J. G. (1964) Proc. Nat. Acad. Sci. USA 52, 1367. 15. Kjeldgaard, N. 0. & Gausing, K. (1974) in Ribosomes, eds. Nomura, M., Tissieres, A. & Lengyel, P. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 369-392. 16. Dennis, P. P. & Bremer, H. (1974) J. Bacteriol. 119, 270-281. 17. Moller, W. (1974) in Ribsomes, eds. Nomura, M., Tissieres, A. & Lengyel, P. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 711-731. 18. Highland, J. H. & Howard, G. A. (1975) J. Biol. Chem. 250, 831-834. 19. Burgess, R. R. (1971) Annu. Rev. Biochem. 40,711-740. 20. Deonier, R. C., Ohtsubo, E., Lee, H. J. & Davidson, N. (1974) J. Mol. Biol. 89,619-629. 21. Fellner, P. (1974) in Ribosomes, eds. Nomura, M., Tissieres, A. & Lengyel, P. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 169-191. 22. Dunn, J. J. & Studier, F. W. (1973) Proc. Nat. Acad. Scd. USA
70,32964300. 23. Matzura, H., Hansen, B. S. & Zeuthen, J. (1973) J. Mol. Biol.
74,9-20. 24. Dalbow, D. G. (1973) J. Mol. Biol. 75, 181-184. 25. Iwakura, Y., Ito, K. & Ishihama, A. (1974) Mol. Gen. Cenet. 133, 1-23. 26. Murialdo, H. & Siminovitch, L. (1971) in The Bacteriophage Lambda, ed. Hershey, A. D. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 711-723. 27. Kaltschmidt, E. & Wittman, H. G. (1970) Anal. Biochem. 36, 401-412. 28. Dennis, P. (1974) J. Mol. Blot. 88,25-41. 29. Nomura, M. & Erdmann, V. A. (1970) Nature 228,744-748. 30. Sidikaro, J. & Nomura, M. (1975) J. Biol. Chem. 250, 11231131. 31. Fairbanks, C., Jr., Levinthal, C. & Reeder, R. H. (1965) Biochem. Biophys. Res. Commun. 20,393-399. 32. Hammel, E., Koka, M. & Nakamoto, T. (1972) J. Biol. Chem.
247,805-814. 33. Gillespie, D. & Spiegelman, S. (1965) J. Mol. Bwl. 12,829.
Cluster of genes in Escherichia coli for ribosomal proteins, ribosomal RNA, and RNA polymerase subunits* (transducing phage X/in vitro protein synthesis/protein identification/DNA.RNA hybridization)
LASSE LINDAHL, S. RICHARD JASKUNAS, PATRICK P. DENNIS, AND MASAYASU NOMURA Institute for Enzyme Research and Departments of Biochemistry and Genetics, University of Wisconsin, Madison, Wisc. 53706
Communicated by James F. Crow, April 21,1975
The transducing phage Xri$18 isolated by ABSTRACT Kirschbaum and Konrad [(1973) J. Bacteriol. 116, 517-526] was found to carry structural genes for several 50S ribosomal proteins and 16S and 23S rRNA. It has previously been demonstrated [Kirschbaum & Scaife (1974) Mol. Gen. Genet. 132, 193-201] that this phage carries genes for the DNA-dependent RNA polymerase (nucleosidetriphosphate:RNA nucleotidyltransferase; EC 2.7.7.6) subunits # and '. Thus, the region of the E. coli chromosome carried by XriP18 contains a cluster of genes essential for transcription and translation.
Knowledge of the genetic organization of the genes for ribosomal components is important for understanding the mechanism of regulation of ribosome biosynthesis. It is known that many ribosomal protein (r-protein) genes in Escherichia coli are clustered in the str-spc region of the chromosome at 64 min (for a review, see ref. 1). Many r-protein genes in this region have been identified by isolating several transducing phages that carry these genes (ref. 2, and our unpublished experiments). For example, one such phage, Xfus2, carries approximately 30 r-protein genes and the gene for an elongation factor EF-G (cited in ref. 2). However, not all the rprotein genes map at the str-spc region. We have previously shown that at least one r-protein gene, the structural gene for S18, maps outside this region (3, 4). Thus, we have initiated a search for additional structural genes for r-proteins outside the str-spc region. One suspected region was near the rif gene at 79 min, since experiments by Friesen and his coworkers indicated that relC, which maps near rif, could be a structural gene for a 50S r-protein (5). As will be described in this paper, we have in fact found that the transducing phage Xrifdl8 isolated by Kirschbaum and Konrad (6) carries the structural genes for r-proteins L1, L7, L12, and probably also the genes for L8, L10, and Lii, in addition to the gene rif. This finding further prompted us to examine the possibility that some other essential genes related to transcription or translation are also located in this same region. We found that genes for ribosomal RNA (rRNA) are also carried by the same transducing phage. It has previously been shown that this phage carries genes for the DNA-dependent RNA polymerase (nucleosidetriphosphate:RNA nucleotidyltransferase, EC 2.7.7.6) subunits , and jY' (7). Thus, genes for several r-proteins, rRNA, and RNA polymerase subunits #3 and ,B', all are clustered in the small chromosomal region that is carried by Xrifdl8. MATERIALS AND METHODS Transducing phage XCI857S7rifdi8 (called Xrifdl8 in this paper) was prepared from a strain H105 (XCI857S7, Abbreviations: r-protein, ribosomal protein; UV, ultraviolet. * This is paper no. 1846 from the Laboratory of Genetics. 2743
Xrifdl8) This strain (6) was obtained from Dr. J. B. Kirschbaum. Another phage XCI857St68 h8OglyT su+36 (called Ah80glyT) was prepared from a strain JC100 that carried this phage (8) and a helper phage XCI857h80St68. The strain was obtained from Dr. J. Carbon. Transducing phage Xfus2 was isolated in this laboratory (cited in ref. 2). XCI857S7 (called X) was used as a control phage. These phages were prepared by heat induction and purified by CsCl equilibrium centrifugation (2, 9). Preparation of antisera against individual 30S proteins was described previously (10). Antiserum against pure Li was prepared in the same way. Antiserum against L12 was a gift from Dr. H. Weissbach. This antiserum reacts with both L7 and L12, but no other r-proteins. L7 is an acetylated form of L12 (11). When we do not distinguish the two proteins, we use L7/Li2 to indicate these proteins. The nomenclature of r-proteins is according to ref. 11. Other methods are described in the table and figure legends. RESULTS
Stimulation of synthesis of ribosomal proteins by XriP1I8 We measured the stimulation of r-protein synthesis in E. coli cells after infection by the transducing phage XrifdI8. The synthesis of r-proteins by the host r-protein genes was reduced by prior UV-irradiation of the host cells. For comparison, we also infected the irradiated cells with X, which does not carry any r-protein genes, and Xfus2, which carries approximately 30 r-protein genes. The experimental procedure was similar to that used previously (2) and is described in the legend to Table 1. The data on the proteins whose synthesis was apparently stimulated by Xrifdl8 relative to cells infected with the control phage A are shown in Table 1. Among these proteins, the apparent stimulation of S5, S6, and L9 is probably due to contamination of these protein samples with Lii, L12, and L8, respectively (see legend to Table 1). In this connection, it should be noted that the synthesis of Lii was apparently stimulated by Xfus2, but stimulation of the synthesis of S5 by this phage was greater. The presence of the gene for S5 in Xfus2 as well as related transducing phages, Xspcl and Xspc2, has been proven previously (ref. 2, and our unpublished experiments). Incomplete separation of LII and S5 in the two-dimensional electrophoresis could explain the apparent stimulation of Lii by Xfus2. Thus, our conclusion is that Arifdl8 stimulates synthesis of LI, L7, L8, L10, Lii, and L12. These proteins are underlined in Table 1. As discussed in the previous paper (2), we cannot eliminate the possibility that stimulation of the synthesis of r-pro-
2744
Genetics: Lindahl et al.
Proc. Nat. Acad. Sci. USA 72 (1975)
Table 1. Stimulation of the synthesis of r-proteins by Xrifd18 3H/'4C ratio Proteins
S5 S6 Li L7 L8 L9 L10 L11 L12 Others
X
0.04 0.19 0.02 0.03 0.10 0.10 0.04 0.06 0.04 0.01-2.5
Xfus2 5.5 1.6 0.09 0.10 0.15 0.13 0.07 2.3 0.14 0.07-13.4
Xrifd 18 2.6 3.0 9.5 34.6 6.4 1.1 22.4 8.7 10.8
0.09-3.0
E. coli cells pre-labeled with ['4C]leucine were irradiated with UV light and infected with phages as previously described (2) except that the UV-sensitive bacterial strain was a Xpapa lysogen of strain 159 (26). The [3H]leucine was added 30 min after phage infeetion and the incorporation was stopped after 10 min. The cells were lysed and proteins were extracted and separated by 2-dimensional gel electrophoresis (27). The 3H/14C ratio of each r-protein spot, except S1 and L31, was determined. Details of the procedures were described previously (2, 28). The data are shown only for the proteins whose synthesis was apparently stimulated by Xrif18 as judged by 3H/14C ratio relative to the control 3H/14C ratio obtained after A infection. As described in the text, proteins S5 and L11, L8 and L9, and S6 and L12 overlap partially in the 2-dimensional gel electrophoresis, and this probably accounts for apparent stimulation of S5, S6, and L9 by Xrifl8.
teins in these experiments was a secondary effect of some phage-coded proteins on residual r-protein synthesis in irradiated cells. Also, identification of the stimulated protein was done only by co-electrophoresis with mature proteins, and more rigorous experiments are required for definitive identification. However, it is notable that there were no proteins whose synthesis was stimulated by both Xrifdl8 and Xfus2 except for S5 and Lii, which have been discussed above. It is highly likely that the observed stimulation of L1, L7, L8, L10, Lii, and L12 is due to the presence of the genes for these proteins on the Xrifdl8 phage genome. The rif gene is located at 79 min on the E. coli genetic map (12), and is very closely linked to genes supM and glyT that code for a tyrosine and a glycine tRNA, respectively. A transducing phage Xh80glyT isolated by Carbon and his coworker carries the supM and glyT genes (8). However, purified Xh80glyT failed to show stimulation of the synthesis of any r-proteins in UV-irradiated cells (data not shown). In vitro synthesis of 50S ribosomal proteins with XriP1i8 DNA. Synthesis of several 30S r-proteins in vitro was previously described using either E. coli DNA (13) or Xspci DNA (2) as template. We used the same DNA-dependent protein synthesizing system to see whether DNA extracted from Xrifdl8 can in fact code for the 50S r-proteins that were stimulated in the in vivo experiments described above. Proteins were synthesized in the presence of [a5S]methionine and DNA from Xrifdl8 as a template. For comparison, DNAs from X, Xfus2 and Xh80glyT were also used as templates. We used simultaneous reconstitution of 30S and "50S" ribosomal subunits to recover radioactive r-proteins synthesized in vitro (refs. 2, 13, and see the legend to Fig. 1).
Xh8Ogl~yT
Cc)
Cd)
30S4
X
Xfus 2
O50$
800n 30S"50SO
QD
60 40-
0.1
20
5
10
15 5 FRACTION NUMBER
10
15
FIG. 1. Ability of various DNAs to stimulate synthesis of rproteins. [35S]Methionine-labeled protein was synthesized in a DNA-dependent cell-free system, using DNA from X, Xrifdl8, Xh80glyT, and Xfus2 as template. To recover [35S]r-proteins synthesized, we simultaneously reconstituted 30S and "50S" particles (see text). This was performed as described previously (2, 13) with the following modification: the dialysis against TMAI-urea [10 mM Tris-HCl (pH 7.4), 10 mM MgCl2, 30 mM NH4Cl, 6 mM 2mercaptoethanol, 8 M urea] and the DEAE-cellulose steps were omitted. Nonradioactive protein and RNA prepared by the ureaLiCl method (29) from 70S ribosomes were added to give a final concentration of 0.1-0.5 mg/ml of rRNA and a ratio of 1.5 equivalents of protein per equivalent of RNA. Ionic conditions were adjusted to 30 mM Tris-HCl (pH 7.4), 20 mM MgCl2, and 0.33 M KCl (reconstitution buffer), and the mixture was incubated for 1 hr at 400. After chilling to 0° the mixture was centrifuged through a 5-20% sucrose gradient in reconstitution buffer at 24,000 rpm in a Spinco SW27 rotor for 11 hr. The gradient was then pumped through a flow photometer monitoring A2m6. and fractionated. An aliquot of each fraction was analyzed for radioactivity. A260 nm
(-) radioactivity (- 0 -).
It should be noted that "50S" particles reconstituted under the conditions described are not identical to the standard 50S subunits, but they contain most of 50S r-proteins and little or no SOS r-protein (our unpublished experiments). The technique is therefore specific in detecting radioactive 50S r-proteins synthesized in vitro. As shown in Fig. 1, some of the proteins synthesized in the presence of Xrifdl8 DNA were found in the reconstituted "50S" particles, but the amount of [s5S]proteins in the reconstituted 30S particles was insignificant. As expected (ref. 2, and our unpublished experiments), Xfus2 DNA stimulated the synthesis of both 30S and 50S r-proteins. In contrast, no significant amount of [a5S]protein was found in the reconstituted particles when DNA from X or Ah80glyT was used as a template. Radioactive proteins synthesized in the presence of Xrifdl8 or Xfus2 DNA were then extracted from the reconstituted particles (combined 30S and "50S" particles) and analyzed further. The radioimmunodiffusion method (2, 13) was employed, using antisera against each of the individual 30S r-proteins (except S1 and S17) and against Li and L12.
Genetics: Lindahl et al.
Proc. Nat. Acad. Sci. USA 72 (1975)
2745
b
a
L12! 4
L8,L1O,L11 I,
FIG. 2. Identification by radioimmunodiffusion of r-proteins synthesized in vitro. Proteins were extracted from the reconstituted particles (Fig. 1) and tested by radioimmunodiffusion (2, 13). Each set of pictures shows a gel (left) and an autoradiogram of the gel (right). The center wells received antiserum against (a) L12 or (b) Li. The top and bottom wells received protein extracted from the reconstituted ribosomes; top, )rijfd8 DNA as template, and bottom Xfus2 DNA as template. The upper right well of (b) received purified L1. Anti-L1 antiserum showed two precipitin bands with a 70S r-protein mixture. We believe that one of the precipitin bands is probably due to a reaction with an aggregate containing Li.
By this method
found that protein synthesized with Xrifdl8 DNA as template contained Li and L7/L12 or proteins immunologically related to them (Fig. 2), but none of the other proteins tested for. By contrast, neither Li nor L7/Li2 was detected among proteins synthesized with Xfus2 DNA as template (Fig. 2). Instead, a number of 30S proteins were detected in this case (data not shown). Radioactive proteins recovered from the reconstituted particles were also analyzed by one-dimensional polyacrylamide gel electrophoresis at pH 4.5 (14). An autoradiogram of the sliced gels is shown in Fig. 3. Five radioactive protein bands were clearly visible in the sample with Xrifdl8 DNA as a template. The upper two bands appeared at the same positions as the reference radioactive L7 and L12 bands. Apparently, both acetylated (L7) and nonacetylated (Li2) forms of the protein were synthesized in the in vitro system. However, rigorous proof of the presence of the acetyl group must await further experiments. Identity of the remaining three radioactive protein bands has not been established. Quite a few 50S proteins migrate in this area. However, it is noted that one of the three bands corresponded to the position of Li, and the remaining two bands were close to the positions where proteins L8, L10, and Lii migrate. Thus it seems likely that one or more of the three other proteins stimulated in the in vvo experiments (L8, LIO, and Lii) were also synthesized in the in vitro system. Many radioactive protein bands were seen in the sample with Xfus2 DNA as a template, but none of them corresponded to the position of L7, L12, or LI (Fig. Sc). To give further proof for the in vitro synthesis of L7/L12 directed by Xrifdl8 DNA, we digested the in vitro synthesized L7/Li2 with trypsin and a-chymotrypsin and analyzed the resultant [5sS]methionine-containing peptides. As shown in Fig. 4, the in vitro synthesized L7/Li2 gave a peptide map very similar to that of the in vivo synthesized L7/ L12. From all these experimental results, we conclude that Xrifdl8 DNA directs synthesis of proteins in vitro that are very similar or identical to protein L7/Li2 and LI. Hence, the structural genes for L7/L12 as well as LI are present on Xrifdl8 DNA. we
Presence of rRNA genes in XriP1i8. We have found that the Xrifdl8 chromosome contains bacterial DNA that is homologous to stable RNA in E. coli. In DNA-RNA hybridization experiments using [3H]uracil-la-
"w.. ,4
I.
abi a b c d e FIG. 3. Polyacrylamide gel electrophoresis of [35S]proteins synthesized in vitro. [35S]Proteins synthesized in vitro in the presence of DNA from Xrifdl8 (b) and Xfus2 (c) were extracted from reconstituted ribosomes and subjected to gel electrophoresis on 10% polyacrylamide gel in the presence of 8 M urea at pH 4.5 (14). For comparison, radioactive 50S proteins prepared from cells grown in the presence of [35S]methionine (d) and a mixture of pure L7 and L12 purified from the above 50S [ssS]proteins (e) were also analyzed in the same way. After the run, gels were processed as described previously (30, 31). The autoradiograms are shown in the figure (b-e). A stained gel of total 70S proteins is-also shown (a). The arrows indicate the positions of proteins L7, L12, L1, L8, L10, and L11. The L8, L10, and L11 bands are not resolved from other proteins on the reference stained gel.
beled stable RNA and an excess of Xrifdl8 DNA, as much as 70-80% of the input RNA can hybridize to Xrifdl8 DNA, but very little or none to Xfus2 DNA. Since 16S and 23S rRNA account for about 85% of the total stable RNA in the cell (15, 16) the results indicate that both 16S and 23S rRNA genes are carried on Xrifdl8 chromosome. Furthermore, as can be seen in Fig. 5, purified nonradioactive 16S and 23S RNA effectively competed with 90% or more of the [3H]uracil-labeled stable RNA that hybridized to Xrifdl8 DNA. The residual 10% radioactive RNA not affected by competition may represent 5S RNA and possibly tRNA. Hybridization of purified radioactive 16S RNA and 23S RNA to Xrifdi8 DNA was also demonstrated. These results clearly demonstrate that Xrifdi8 carries DNA homologous to 16S and 23S rRNA. In addition, hybridization experiments using a constant amount of Xrifdl8 DNA and excess amounts of RNA suggest that the number of copies of these genes on the phage is probably one (see the legend to Fig. 5). In preliminary experiments, we have also observed that Xrifdl8 stimulates the synthesis of 23S RNA and 16S RNA (or their "precursor" RNAs) in UV-irradiated E. coli cells. It therefore appears that a complete set of rRNA genes is carried by Xrifdl8, and that these genes map near the rif locus on the E. coli chromosome. DISCUSSION Results presented in this paper prove that the structural genes for 50S proteins L7/L12 and Li and for 16S and 23S rRNA are carried by Xrifdl8. The data also indicate that in all likelihood the structural genes for L8, L10, and Lii are also carried by Xrifdl8.
2746
Genetics: Lindahl et al. Rf 0.60.50.4
Proc. Nat. Acad. Sci. USA 72 (1975) 1004
a Rf
b
0.3F
t 0.8 30.6
0.2 F 0.1 F
J0.4
0
0
1 8
?
0
60_ _
0.2
- + I
4 3
I %_ I,
1 ° cm cm
I
I
L
X
k
a,
5 4 3 2 10
FIG. 4. Peptide maps of in vitro and in vivo synthesized L7/ L12. (a) In vitro L7/L12: [s5Sjmethionine labeled r-protein synthesized from Xrifdl8 was isolated and fractionated by gel electrophoresis (compare Fig. 3b). The L7 and L12 bands were cut out and minced, and the pH was adjusted to 8 with 1 M NH4HCO3. Trypsin was added to a final concentration of 0.1 mg/ml, and the mixture was incubated overnight at 37°. The liquid was removed, and the gel was suspended in 0.1 M NH4HCO3 containing trypsin (0.1 mg/ml) and again incubated overnight at 37°. The liquid was again collected and combined with the first extract. Then a-chymotrypsin was added at a ratio of protein to chymotrypsin 1:100. The mixture was incubated for 4 hr at 37° and lyophilized. (b) In vivo L7/L12: 505 subunits were isolated from E. coli PR13 grown in the presence of [ssS]methionine. Proteins L7 and L12 were extracted from the 50S by the salt-ethanol method (32) and subjected to electrophoresis as in Fig. 3. The L7 and L12 proteins were extracted and digested with trypsin and chymotrypsin as above. The digests were fractionated in two dimensions on silica gel thin-layer plates by electrophoresis at pH 4.7 followed by chromatography using a butanol-water-acetic acid solvent as previously described (4) except that the electrophoresis was for 2 hr at 750 V, and chromatography was for 14 hr (a) or 6 hr (b). Electrophoresis: horizontal; chromatography: vertical.
Protein L7/L12 is important in the ribosome-associated GTPase reactions that are dependent on various protein factors such as elongation factor G or elongation factor T (for a review, see ref. 17). Using partially protein-deficient 50S "core" particles, it has been recently demonstrated that L7/L12, L1O, and Lii are interrelated in the reconstitution of 50S subunits (ref. 18, see also ref. 17). It is interesting to note that all of the genes for these proteins are clustered near rif as indicated in the present results. Xrifdl8 carries the genes for fB and iB' subunits of RNA polymerase (7). We have confirmed the above conclusion by showing stimulation of the synthesis of f# and fl' subunits in UV-irradiated cells infected with Xrifdl8 (Jaskunas, Nomura, and Burgess, unpublished experiments). Both the r-protein genes and the rRNA genes we have found on this phage are probably located very close to the genes for /3 and ,B'. Heteroduplex analysis on Xrifdl8 DNA shows that about 52% of A genome is deleted and substituted by bacterial DNA that is about 59% of a A unit in length (our unpublished experiments). The sum of the molecular weights of the (3 and (3' subunits of RNA polymerase is about 320,000 (19) and the sum of the molecular weights of 50S proteins Li, L7/L12, L8, L10, and LII is about 100,000 (11). Therefore, the genes for these proteins occupy a minimum of about.11,700 base pairs of DNA. Combined with an additional 5000 base pairs for the 16S and 23S rRNA genes (2022), this represents a minimum of 16,700 base pairs or 36% of a A unit of DNA; this represents greater than 60% of the total bacterial DNA (59% of a A unit) in Xrifdl8. Such a calculation indicates that the r-protein genes as well as rRNA genes studied in this work are physically very close to the genes for the RNA polymerase subunits. Since neither the rprotein genes nor rRNA genes were used to select Xrifdl8,
20 20
0
25 5.0 7.5 COMPETING rRNA (C,/ASSAY)
1.0
FIG. 5. Hybridization of [3H]uracil-labeled stable RNA to Xrifdl8 DNA in the presence of nonradioactive rRNA as competitor. The [3H]uracil-labeled stable RNA was prepared as described previously (Dennis and Nomura, manuscript submitted for publication). Its specific activity was 9.5 X 104 cpm/hg. In the hybridization assays 0.5 gg of 3H-labeled stable RNA was hybridized to 20 sg of DNA prepared from Xrifdl8 phage. Various amounts of nonradioactive 16S and 23S rRNA were added as indicated. Hybridization was performed according to a method (Dennis and Nomura, manuscript submitted) similar to that of Gillespie and Spiegelman (33). Relative amounts of [3H]RNA hybridized are shown in the figure. Radioactivity in hybrid at 100% of maximum was 2.48 X 104 cpm. Competitor RNA was obtained from E. coli AS19 after 30 min of rifampicin treatment to eliminate unstable RNA. Stable RNA was prepared by a phenol method and separation of 16S and 23S rRNA from 4S and 5S RNA was accomplished by passage through Sephadex G-200. At saturation, approximately 0.35 zg of stable RNA complexed to 20 ,ug of Xrifdl8 DNA and saturation requires an input of more than 0.5 gg of RNA. The Xrifdl8 phage preparation was about 50% contaminated with helper phage. If the transducing phage carries a complete copy of 16S and 23S rRNA genes, it would be expected that about 5% of the pure Xrifdl8 DNA or about 2.5% (0.5 gg) of the DNA used for hybridization would be homologous to rRNA. The observed value (0.35 Mg) was close to the expected value, supporting the presence of one set of 16S and 23S rRNA genes on the phage.
the proximity relation of these genes to the RNA polymerase genes found in the transducing phage probably applies to their "natural" locations on the E. coli chromosome. Deonier and his coworkers previously suggested that a set of rRNA genes (rrnB) may be located between argC and glyT, which is close to rif (20). It remains to be seen whether the gene set found on Xrifdl8 is identical to rrnB studied by these workers. The rate of synthesis of both rRNA and r-proteins is regulated and is related to growth rate in various growth media (for a review, see ref. 15). Synthesis of RNA polymerase ,B and fl' subunits is also coordinately regulated and appears to vary depending on growth media (23-25). It remains to be seen whether the clustering of r-protein, rRNA, and RNA polymerase genes has any functional significance in the regulation of the expression of these important genes. The observed clustering might also be related to a topographical feature of the E. colh chromosome; the genes that are transcriptionally active, such as rRNA genes, r-protein genes, and RNA polymerase genes, may be located at a place where continuous transcription by RNA polymerases could be efficiently maintained.
Note Added in Proof. In collaboration with R. Burgess, we have also discovered that Xrifdl8 carries a gene for elongation factor Tu.
Genetics: Lindahl et al. We thank M. Stroud, L. Sadowski, K. Ryan, G. D. Strycharz, W. Taylor, Dr. J. Ingversen, and Dr. F. Harada for their assistance; Drs. J. B. Kirschbaum and J. Carbon for providing us with bacterial strains and transducing phages; and Dr. H. Weissbach for providing us with antiserum against L12. This work was supported in part by the College of Agriculture and Life Sciences, University of Wisconsin, and by grants from the National Science Foundation (GB31086) and the National Institute of General Medical Sciences (GM-20427). L.L. is a recipient of a Postdoctoral fellowship from the European Molecular Biology Organization.
1. Jaskunas, S. R., Nomura, M. & Davies, J. (1974) in Ribosomes, eds. Nomura, M., Tissieres, A. & Lengyel, P. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 333-368. 2. Jaskunas, S. R., Lindahl, L. & Nomura, M. (1975) Proc. Nat. Acad. Sci. USA 72,6-10. 3. Bollen, A., Faelen, M., Lecocq, J. P., Herzog, A., Zengel, J., Kahan, L. 4 Nomura, M. (1973) J. Mol. Biol. 76,463-472. 4. Kahan, L., Zengel, J., Nomura, M., Bollen, A. & Herzog, A. (1973) J. Mol. Biol. 76, 473-483. 5. Friesen, J. D., Fiil, N. P., Parker, J. M. & Haseltine, W. A. (1974) Proc. Nat. Acad. Sci. USA 71, 3465-3469. 6. Kirschbaum, J. B. & Konrad, E. B. (1973) J. Bacteriol. 116, 517-526. 7. Kirschbaum, J. B. & Scaife, J. (1974) Mol. Cen. Cenet. 132, 193-201. 8. Squires, C., Konrad, B., Kirschbaum, J. & Carbon, J. (1973) Proc. Nat. Acad. Sdi. USA 70,438-441. 9. Miller, J. H. (1972) Experiments in Molecular Genetics (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.). 10. Kahan, L., Held, W. A. & Nomura, M. (1974) J. Mol. Biol. 88, 797-808. 11. Wittmann, H. G. (1974) in Ribosomes, eds. Nomura, M., Tiss- ieres, A. & Lengyel, P. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 93-114. 12. Taylor, A. L. & Trotter, C. D. (1972) Bacteriol. Rev. 36, 504524.
Proc. Nat. Acad. Sci. USA 72 (1975)
2747
13. Kaltschmidt, E., Kahan, L. & Nomura, M. (1974) Proc. Nat. Aced. Sci. USA 71, 446-450. 14. Leboy, P. S., Cox, E. C. & Flaks, J. G. (1964) Proc. Nat. Acad. Sci. USA 52, 1367. 15. Kjeldgaard, N. 0. & Gausing, K. (1974) in Ribosomes, eds. Nomura, M., Tissieres, A. & Lengyel, P. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 369-392. 16. Dennis, P. P. & Bremer, H. (1974) J. Bacteriol. 119, 270-281. 17. Moller, W. (1974) in Ribsomes, eds. Nomura, M., Tissieres, A. & Lengyel, P. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 711-731. 18. Highland, J. H. & Howard, G. A. (1975) J. Biol. Chem. 250, 831-834. 19. Burgess, R. R. (1971) Annu. Rev. Biochem. 40,711-740. 20. Deonier, R. C., Ohtsubo, E., Lee, H. J. & Davidson, N. (1974) J. Mol. Biol. 89,619-629. 21. Fellner, P. (1974) in Ribosomes, eds. Nomura, M., Tissieres, A. & Lengyel, P. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 169-191. 22. Dunn, J. J. & Studier, F. W. (1973) Proc. Nat. Acad. Scd. USA
70,32964300. 23. Matzura, H., Hansen, B. S. & Zeuthen, J. (1973) J. Mol. Biol.
74,9-20. 24. Dalbow, D. G. (1973) J. Mol. Biol. 75, 181-184. 25. Iwakura, Y., Ito, K. & Ishihama, A. (1974) Mol. Gen. Cenet. 133, 1-23. 26. Murialdo, H. & Siminovitch, L. (1971) in The Bacteriophage Lambda, ed. Hershey, A. D. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 711-723. 27. Kaltschmidt, E. & Wittman, H. G. (1970) Anal. Biochem. 36, 401-412. 28. Dennis, P. (1974) J. Mol. Blot. 88,25-41. 29. Nomura, M. & Erdmann, V. A. (1970) Nature 228,744-748. 30. Sidikaro, J. & Nomura, M. (1975) J. Biol. Chem. 250, 11231131. 31. Fairbanks, C., Jr., Levinthal, C. & Reeder, R. H. (1965) Biochem. Biophys. Res. Commun. 20,393-399. 32. Hammel, E., Koka, M. & Nakamoto, T. (1972) J. Biol. Chem.
247,805-814. 33. Gillespie, D. & Spiegelman, S. (1965) J. Mol. Bwl. 12,829.
