Reprinted from the ProcEepines or THE NATIONAL ACADEMY OF SCIENCES Vol. 49, No. 4, pp. 588-544. April, 1963. DISTINCT CISTRONS FOR THE TWO RIBOSOMAL RNA COM PONENTS* By 8. A. YANKoFskyf AND S. SPIEGELMAN DEPARTMENT OF MICROBIOLOGY, UNIVERSITY OF ILLINOIS, URBANA Communicated by T. M. Sonneborn, February 25, 1968 Previous investigations!: ? have established that HF’. coli DNA contains sequences complementary to homologous ribosomal RNA. The proof depended on demon- strating the formation of specific RNAase resistant complexes between labeled ribo- somal RNA and heat-denatured DNA. It was further shown, by the use of two identifying isotopie labels, that nonribosomal RNA from the same organism does not compete for the DNA sites complementary to ribosomal RNA. All of these experiments were carried out with the 238 ribosomal RNA and they left unanswered the relation of these findings to the 165 RNA component. The similarity in base composition®: ¢ and the fact that the molecular weights? of the 235 and 168 are almost in the relation of 2:1 suggest the possibility of a common ori- gin, the 238 being perhaps a dimer of the 165 RNA. Definitive evidence on whether they do, in fact, derive from the same sequence can be readily obtained with the hy- bridizing technique of Hall and Spiegelman® as modified in the ribosomal RNA in- vestigations! ? cited. The following sorts of information are pertinent to a resolution. (a) Saturation plateaus: Tf the 168 and 238 are derived from the same sequence, the RNA/DNA ratio found in the hybrid at saturation should be the same for each RNA. (b) Additivity: At the saturation RNA/DNA ratio of either, the addition of the other should lead to no further complex formation if they are derived from the same se- quences. If the sequences of origin are different, additional hybrids should be ob- served. (c) Competitive interaction: By the use of two identifying isotopic labels the presence or absence of competition during hybridization can be established. Absence of competitive interaction would indicate distinct sequences and its exist- ence would argue for identity. The present paper describes experiments which provide the data necessary for a decision. To alleviate somewhat the current monotony of molecular biology and to extend our understanding of these matters beyond F. colz, the experiments to be described were performed with B. megatertwm. The results indicate that the sequences of the 238 and 168 RNA components are dissimilar. They must there- fore possess different genetic origins. VoL. 49, 1963 BIOCHEMISTRY: YANKOFSKY AND SPIEGELMAN 539 Materials and Methods.{a) Bacterial strain: Strain 219, a pyrimidine-requiring derivative of KM isolated by the technique of Mangalo and Wachsman’) § was kindly provided by Dr. J. T. Wachsman. (b) Media: A basal medium’ supplemented with 10 to 30 g/ml uridine was generally used. For P® incorporation experiments, the phosphate concentration was reduced from 0.024 M to 0.0012 M and 0.05 M Tris (pH 7.3) added for buffering. (c) Preparation of cells: Cells suspended in basal medium supplemented with 20-30 ug/ml in uridine were shaken overnight at 37°C, harvested, and resuspended at an O.D.¢6 of 0.200 in fresh medium containing 14 g/ml uridine. When they attained an O.D.¢0 of 0.400, the cultures were harvested, washed, and resuspended in basal medium to an O.D.s6) of about 1.000 for use as inocula in incorporation experiments. (d) Steady state isotope incorporation: (1) H*-uridine: Log phase cells, prepared as described, were suspended in basal medium to an O.D.¢69 of 0.085 and shaken at 37°C for 15 min. Then 10.3 ug/ml of H*-uridine (New England Nuclear Corp., 3.0 mc/mM) was added and the culture shaken at 37°C until growth stopped at an O.D.¢0 of 0.240. The culture was harvested, washed, and re- suspended in twice the original volume of basal medium containing 90 ug/ml of unlabeled uridine. The cells were incubated with aeration at 37°C for 0.8 generations to eliminate H* counts from the unstable RNA fraction, then harvested. (2) P*?-orthophosphate: Pyrophosphate-free, neu- tralized P**-orthophosphate was added (630 uc/ml) to log phase cells at an O.D.65 of 0.07 in basal medium, containing 0.0012 M phosphate, and incorporation continued until an O.D.s% of 0.400 was reached. The culture was then “chased” for one generation in a medium adjusted to 0.024 M in nonradioactive phosphate. (e) Conversion to spheroplasts: Log-phase cells were suspended to an O.D.g9 of about 1.2 in a medium consisting of 0.04 M Tris, pH 7.3 ~ 0.002 M MgSO, — 0.3 M sucrose, and equilibrated to 37°C. Armour’s lysozyme (200 »g/ml) was added and conversion to spheroplasts followed with a phase-contrast microscope. Conversion was virtually complete within 15 min. The spheroplasts were harvested, then washed once in the above medium. (f) Lysis and bulk RNA extraction: Washed spheroplast pellets were lysed by resuspension in 0.01 M Tris, pH 7.38 ~— 0.005 M MgCl (TM) buffer containing lysozyme (200 ug/ml) and 25 ug/ml of DNAase (Worthington Biochemical). The lysate was then subjected to three freeze- thaw cycles and total cellular RNA was isolated and purified, all as detailed by Hayashi and Spiegelman. (g) Purification of ribosomal RNA subclasses: The two ribosomal RNA components were separated from each other by repeated chromatography on methylated-albumin-kieselguhr (MAK) columns prepared according to Mandell and Hershey. All buffers used during chromatog- raphy contained 0.025 M NaH,PO, — 0.025 M Na,sHPO, — pH 6.9. RNA preparations were loaded at 50 ug/ml or less, and elution accomplished with linear NaCl gradients ranging from 0.6 M to 1.25 M NaCl. The total eluting volume was from 320 to 380 ml, and 5 to 7 ml fractions were collected. The resulting purified RNA fractions were pooled and concentrated to about 50 ug/ml as follows: the ionic strength of the solvent was first changed to 0.01 M Tris, pH 7.3 — 0,002 M MgCl. — 0.02 M NaCl by dialyzing against at least 100 volumes of this buffer for about 15 hr with two buffer changes. The preparations were next reduced to the appropriate volume in a flash-evaporator at reduced pressure. The sample flask was held at 28°C and the collecting flask at 0°C. Concentrated RNA preparations were finally dialyzed against TMS buffer (0.01 M Tris, pH 7.3 — 0.001 M MgCl, — 0.3 M NaCl). (h) Sucrose gradient analysis: The size distribution of RNA preparations was routinely deter- mined by centrifugation through linear sucrose density gradients.% 11 (i) DNA isolation: DNA was extracted and purified from spheroplasts and heat-denatured as previously described for E. coli.1_ Heat-denatured preparations will be designated by 1XDNA. (j) DNA sedimentation velocity analysis: Sedimentation coefficients were determined in the Spinco E analytical ultracentrifuge using UV optics. Runs were performed at 25 ug/ml accord- ing to the procedure of Marmur.!? Observed sedimentation coefficients were corrected to 20°C. No other corrections were applied. Molecular weights were estimated from the measured Sx using the empirical relationship of Doty, McGill, and Rice.14 (k) DNA-RNA hybridization: All experiments described were performed with a heat-de- natured DNA derived from a native preparation that had an Se of 21.8 and an estimated molec- 540 BIOCHEMISTRY: YANKOFSKY AND SPIEGELMAN Proc. N. A. 8. ular weight of 7.3 * 106% DNA from B. megateritum undergoes renaturation rather readily; hence, the slow cool from higher temperatures employed in the earlier studies’ ® was avoided. Hybridizations were always performed by incubation at 41°-43°C. Mixtures of IXDNA at 50 ug/ml and labeled RNA at various concentrations in 0.7 ml of TMS buffer were incubated at 41°-43°C for 12 to 16 hr. Saturated CsCl was added to a final volume of about 3 ml] and a density of 1.72. Centrifugation was carried out for 70 hr at 33,000 rpm in an SW 39 rotor of the Model L Spinco ultracentrifuge at a rotor temperature of 25°C. Fractions were collected from the bottom of the tube. Procedures for examining the DNA density region for RNAase resistant radioactivity on millipore membranes in a liquid scintillation spectrometer have been detailed by Yankofsky and Spiegelman. Results.—Purification of ribosomal RNA subclasses: Separation of labeled 168 and 238 RNA components was achieved by repeated chromatography on MAK columns. The purification was monitored by centrifugation in sucrose linear density gradients with unlabeled bulk RNA of £. colt added as size markers. The degree of cross contamination is readily determined by comparison of the radio- activity and O.D.»6¢ profiles. An example of bulk B. megaleritum RNA separation on a MAK column is shown in Figure 1. The profile is similar to those obtained in this laboratory with F. colz RNA preparations” except that the 165 region appears to be partially resolved into two components. 0600) 100 Fig. 1.—Chromatographic separation of bulk RNA. B. megaterium RNA 0400 1 Nact was uniformly labeled with H#-uridine asin Methods. The column was equil- 080 ibrated at 0.66 M NaCl; the RNA 0? 260 loaded at 50 xg/ml in 0.66 M NaCl and eluted with a 360 ml linear gradient run- ning from 0.66 M to 1.25 M NaCl. 5 ml fractions collected. 0.200 10.60 TUBE NUMBER The 235 RNA region, indicated by the arrows in Figure 1, was chromatographed repeatedly, and the profile on the fourth column is shown in Figure 2A. Here, the arrows denote the region pooled and concentrated for experimental use, and Figure 2B shows its size distribution in a sucrose gradient. As can be seen, the purified labeled component is virtually confined to the 238 region of the carrier bulk RNA added. The 16S component of Figure 1 was similarly treated, and Figure 34 shows a representative profileon MAK. Again, the arrows indicate the region pooled, con- centrated, and analyzed for size. Figure 3B shows the size distribution of this re- gion compared to that of #. coli marker RNA. Although of interest, the abnor- mality seen in the 168 profile both in Figure 1 and Figure 34 is not directly pertinent to the present investigation and its discussion will be deferred for a subsequent pub- lication. Comparison of 3A and 3B indicates that the asymmetry observed is not due to significant contamination with 238 RNA. All preparations employed in the VoL. 49, 1963 BIOCHEMISTRY: YANKOFSKY AND SPIEGELMAN 5AL 0300 120 Fig. 2A.—MAK column: Chromatographic Nacl profile after the fourth chromatography of the 23S 100 region shown by the arrows in Figure 1. The column was equilibrated at 0.68 1 NaCl; 600 ug RNA loaded at 50 ug/ml in 0.68 MW NaCl and eluted with a 320 ml linear gradient from 0.72 to 1.22 M NaCl. Fie. 28.—Sucrose density gradient centrif- 0.200 0.100;-- 0.80 2260 ugation. An aliquot (0.5 ug RNA, 60,000 cpm) of the pooled tubes indicated by the 0.600 arrows in (A) was used. 0.6 mg E. coli bulk RNA added as O.D. marker. 1.2 ml fractions 0400 collected and 0.3 ml samples from each tube : plated for radioactive counts. The O.D. pro- file identified the known components in the 0.200 added carrier material. The first major peak on the left is the 238, the second the 168, and ° the last corresponds to the 48 component. lo 20 TUBE NUMBER present study were examined before use in sucrose gradients for cross contamination or evidence of breakdown. Samples showing evidence of either were discarded. We now consider the details of the three types of experiments which can illumi- nate the origins of the 16S and 238 RNA components. (1) Saturation plateaus: The proportion of RNAase resistant hybrid formed by incubating a fixed amount of 1X DNA with increasing amounts of each ribosomal RNA component are shown in Figure 4. The 238 RNA reaches a plateau when ap- proximately 0.18% of the DNA is oceupied, while about 0.14% of the DNA is capable of complexing with 165 RNA. In six repetitions, mean values of 0.179 + 0.0072 and 0.136 + 0.014 were obtained for the respective saturation values of 238 and 16S RNA. The fact that the saturation plateaus for the two are different supports the conclusion that the two types of RNA have different origins. (2) Additivity: It will be noted from Figure 4 that for 50 wg DNA, saturation for the 235 RNA is achieved at 3 ug/ml, and 2 ug/ml saturates for the 168 component. We now inquire whether the addition of both at saturating levels to the same reaction mixture increases the amount of complex observed, and, if so, to what extent. The results of such an experiment are presented in Table 1. Addi- tion of the values obtained when saturating amounts of each RNA subclass is com- plexed alone (mixture 1 + mixture 2) indicates that 0.303 per cent of the DNA would be hybridized. The amount of complex formed when both are incubated together (mixture 3) is within 4 per cent of this value. These results are difficult to reconcile Fie. 34.—MAK column: Chroma- tographic profile of B. megaterium 168 RNA (steady state P*? label) after third chromatography. Column equilibrated at 0.6 M NaCl; 760 we RNA loaded at 50 ug/ml in 0.6 M NaCl and eluted with a 340 ml linear gradient from 0.6 M to 1.2 M NaCl. Fig. 3B.—Sucrose density gradient cen- trifugation: Analysis of an aliquot (0.5 yg of RNA, 80,000 cpm) from the pooled tubes shown under the arrows in (A). 0.6 mg E. < colt bulk RNA added as O.D. marker. 20 All other details as in Figure 2B. 0.400) 0.200) 02260 1.000 0.600} 0.200) 10 TUBE NUMBER 0 10 20 30 40 g/ml INPUT Fie. 4.—Saturation pla- teaus: Saturation curves of B. megatertum 168 and 238 RNA hybridized with 50 yg/ml B. megatertumheat-denatured DNA. Each point represents RN Aase resistant counts found in the DNA region after CsCl equilibrium density gradient centrifugation. Annealing and analytical procedures as de- scribed under Methods. 542 BIOCHEMISTRY: YANKOFSKY AND SPIEGELMAN Proc. N. A. S. 3 5 | TOTAL HYBRID (P7245) Z 0280;~ 0 2 g — a = ” mn a ~~ / z g o240y y a rs / Fs < 0.2001¢ 3 Se ° = Z 0./60,- 235 HYBRID (H3) a 0 40 20 30 }g/ml INPUT P32-16S-RNA Fia. 5.— Competitive interaction: Tests for competitive interactions between mixtures of 23S-H®-RNA and 168-P?-RNA from B. mega- tertum complexed with homologous heat-denatured DNA. All mix- tures contain 50 ug/ml DNA, 3 ug/ml purified 238-H?RNA (see Fig. 2) and the indicated concen- trations of purified 16S-P3-RNA (see Fig. 3). Annealing and ana- lytical procedures as described un- der Methods. with a common origin and are clearly consistent with the existence of distinct com- plementary regions. We come now to the final available experimental test. (3) Competitive interaction: Increasing amounts of P?*labeled 165 RNA were incubated in three tubes, each containing 50 ug of LXDNA and H*labeled 238 RNA at its saturation level (3.0 ug). Because of the identifying isotopic labels, it was possible to determine independently the per cent hybrid formed by each of the two ribosomal RNA size classes in the three mixtures. From the data shown in Figure 5 it is evident that the addition of the P#-16S RNA results in no significant displacement of H?-235 RNA. Furthermore, as more 165 RNA is added, the total hybrid approaches a level of saturation near that expected for the sum of the two subclasses incubated alone. There is no evidence of competition between the two ribosomal RNA subclasses for common DNA sites. Discussion.—In previous studies! ? we have shown that. specific complexes are formed between homologous ribosomal RNA and DNA in bacterial species having intermediate (52%) and high (64%) contents of guanosine-cytosine (GC) in their DNA. The present study establishes a similar sequence complementarity in B. megaterium which has a low (38%) GC content. The fact that ribosomal RNA TABLE 1 Trsr ror ADDITIVITY DURING HyBRIDIZATION aT SATURATION LEVELS or 168 AND 238 RNA RN Aase resistant hybrid Mixture Contents ug RNA fixed/100 pe DNA 1 50 ug/ml 1XDNA + 3.06 ug/ml 238 RNA 0.186 2 50 wg/ml IXDNA + 2.21 pg/ml 168 RNA 0.117 Sum = 0.303 3 50 ug/ml IXDNA + 3.06 ug/ml 288 RNA + 2.21 pg/ml 168 RNA 0.291 The addition mixture (3) contained the same DNA and RNA preparations as the control mixtures (1 and 2). All three mixtures were annealed under identical conditions, centrifuged together, and the raw hybrids tested for RNAase resistance with the same enzyme preparation. Details of annealing and analytical procedures are given in Methods. Vou. 49, 1963 BIOCHEMISTRY: YANKOFSKY AND SPIEGELMAN 543 complementarity to DNA obtains in organisms of diverse DNA composition, lends credence to its generality. These findings raise rather forcibly an interesting problem. The base composi- tion of ribosomal RNA shows virtually no correlation with that of homologous DNA... 16 Evidently the ribosomal RNA cistrons have been kept within narrow limits while the rest of the genome has undergone the widest variation in base composition permissible within a triplet coding mechanism. The specification of the selective mechanism which can produce this remarkable outcome poses an interesting problem for experimental resolution. Sequence complementarity has previously been shown for the unstable messen- ger RNA,® transfer RNA,” 38 and 288 RNA.'.? The present study demonstrates that it also holds for 165 RNA. Thus, the synthesis of all known cellular RNA com- ponents can be explained in terms of a DNA mediated reaction. The present study had as its primary purpose to provide evidence which could decide whether the 16S and 238 ribosomal components are derived from the same or different. complementary DNA sequences. The experiments reported indicate a difference in saturation plateaus, additivity of hybrid formation at saturation levels of each type, and absence of competitive interaction during hybrid formation. These findings are difficult to reconcile with a common sequence. They provide consistent evidence for distinct cistronic origins. The further analysis into the nature of ribosomal RNA will require an examina- tion for heterogeneity within each class. We have already pointed out!: ? that the level (0.2%) at which DNA is saturated by hybridizing with homologous 238 RNA would suggest that H. cold contains about 10 complementary cistrons for this com- ponent. The data presented here would suggest that 0.18% of B. megatertum DNA is complementary to its 235 ribosomal RNA and 0.14% to its 165 compo- nent. On the basis of the DNA content per ‘nuclear body’’!® one would estimate that the DNA contains approximately 35 stretches complementary to 238 RNA and 45 complementary to the 168 component. The significance of this apparent re- dundancy may be related to the rather large number of strands required for a full ribosomal complement which constitutes 85 per cent of the total cellular RNA. However, the existence of multiple copies in the genome provides a possibility for variation. It is of no little interest to determine whether this potentiality was exploited. It is evident that the use of column fractionation and competition ex- periments with identifying labels should provide data pertinent to this problem. Summary.—The experiments reported were designed to decide whether the 16S and 238 ribosomal RNA components are derived from the same or different com- plementary sequences in the DNA. Specific hybrid formation, coupled with iso- topic labeling was employed as the analytical device. The data establish that (a) the maximal amount of RNA which ean hybridize per unit of DNA is different for the two; (b) at saturation concentrations of each, the amount of hybrid formed is additive when 168 and 23S RNA are both present; (c) no evidence of competitive interaction between the two for the same sites can be detected. All these findings are difficult to reconcile with a common origin. We conclude that 165 and 238 ribosomal RNA are derived from DNA sequences unique to each. * This investigation was aided by grants-in-aid from the U.S. Public Health Service and the National Science Foundation. 544 BIOCHEMISTRY: YANKOFSKY AND SPIEGELMAN Proc.N.A.8, { Predoctoral fellow trainee in Molecular Genetics (USPH 2G-319). 1 Yankofsky, 8S. A., and 8. Spiegelman, these ProcrEpINGs, 48, 1069 (1962). 2 Yankofsky, 8. A., and S. Spiegelman, these ProceEpINGs, 48, 1466 (1962). 3 Spahr, P. F., and A. Tissiéres, J. Mol. Biol., 1, 237 (1959). 4 Yankofsky, 8. A., and 8. Spiegelman, unpublished observations (1962). 5 Kurland, C. G., J. Mol. Biol., 2, 83 (1960). 6 Hall, B. D., and 8. Spiegelman, these ProcrEepinas, 47, 137 (1961). 7 Mangalo, R., and J. T. Wachsman, J. Bacteriol., 83, 27 (1962). 8 Mangalo, R., and J. T. Wachsman, J Bacteriol., 83, 35 (1962). 9 Hayashi, M., and S. Spiegelman, these Procerpines, 47, 1564 (1961). 10 Mandell, J. D., and A. D. Hershey, Analyt. Biochem., 1, 66 (1960). 4 Britten, R. J., and R. B. Roberts, Science, 131, 32 (1960). 2 Marmur, J., J. Mol. Biol., 3, 208 (1961). 18 Doty, P., B. McGill, and 8S. A. Rice, these Procreprnes, 44, 432 (1958). “4 Kano-Sueoka, T., and 8. Spiegelman, these ProckEDINGs, 48, 1942 (1962). 4 Spiegelman, S., in Cellular Regulatory Mechanisms, Cold Spring Harbor Symposia on Quantita- tive Biology, 26, 75 (1961). 16 Woese, C. R., Nature, 189, 918 (1961). 7 Giacomoni, D., and 8. Spiegelman, Science, 138, 1328 (1962). 18 Goodman, H. M., and A. Rich, these ProcrEpines, 48, 2101 (1962). 18 Spiegelman, 8., A. I. Aronson, and P. C. Fitz-James, J. Bacteriol., 75, 102 (1958).