Changes in RNA in relation to growth of the fibroblast: II. The lifetime of mRNA,rRNA, and tRNA in resting and growing cells

The stabilities of the principal classes of RNA have been studied in resting and exponentially growing mouse fibroblast lines 3T6 and 3T3. Cytoplasmic mRNA, labeled with tritiated uridine and isolated by virtue of its poly A content, is equally stable in resting and growing cells, displaying a half-life of about 9 hr. We conclude that the accumulation of poly A(+) mRNA during transition from resting to growing state is due not to an increase in its stability, but to an increase in its rate of formation.The stability of cytoplasmic rRNA was measured after labeling with ³H-methyl-methionine. In agreement with the results of previous studies, we found that rRNA is stable in growing cells and unstable in resting cells. Quite unexpectedly, the 18S and 28S rRNA of resting cultures were found to differ appreciably in turnover rate. In both 3T6 and 3T3, the half-life of 28S RNA is about 50 hr, and that of 18S RNA about 72 hr. For this reason, though growing cells should synthesize the two ribosomal subunits in equal numbers, resting cells should synthesize more of the larger subunits than of the smaller. tRNA is unstable under all conditions. Its half-life is 36 hr in resting cells and about 60 hr in growing cells.     

Ribosomal RNA Turnover in Contact Inhibited Cells

CONTACT inhibition of animal cell growth is accompanied by a decreased rate of incorporation of nucleosides into RNA^1–3. Contact inhibited cells, however, transport exogenously-supplied nucleosides more slowly than do rapidly growing cells^4,5, suggesting that the rate of incorporation of isotopically labelled precursors into total cellular RNA may be a poor measure of the absolute rate of RNA synthesis by these cells. Recently, Emerson⁶ determined the actual rates of synthesis of ribosomal RNA (rRNA) and of the rapidly labelled heterogeneous species (HnRNA) by labelling with ³H-adenosine and measuring both the specific activity of the ATP pool and the rate of incorporation of isotope into the various RNA species. He concluded that contact inhibited cells synthesize ribosomal precursor RNA two to four times more slowly than do rapidly growing cells, but that there is little if any reduction in the instantaneous rate of synthesis of HnRNA by the non-growing cells. We have independently reached the same conclusion from simultaneous measurements on the specific radioactivity of the UTP pool and the rate of ³H-uridine incorporation into RNAs (unpublished work of Edlin and myself). However, although synthesis of the 45S precursor to ribosomal RNA is reduced two to four times in contact inhibited cells, the rate of cell multiplication and the rate of rRNA accumulation are reduced ten times. This suggests either “wastage”⁷ of newly synthesized 45S rRNA precursor, or turnover of ribosomes in contact inhibited cells Two lines of evidence suggest that “wastage” of 45S RNA does not play a significant role in this system. (1) The rate of synthesis of 45S RNA in both growing and contact inhibited cells agrees well with that expected from the observed rates of synthesis of 28S and 18S RNAs (unpublished work of Edlin and myself). Emerson has made similar calculations⁶. (2) 45S RNA labelled with a 20 min pulse of ³H-uridine is converted in the presence of actinomycin D to 28S and 18S RNAs with the same efficiency (approximately 50%) in both growing and contact inhibited cells. These results indicate that, in order to maintain a balanced complement of ribosomal RNAs, contact inhibited cells must turn over their ribosomes. We present evidence here that rRNA is stable in rapidly growing chick cells, but begins to turn over with a half-life of approximately 35–45 h as cells approach confluence and become contact inhibited.     

Alteration in the processing of ribosomal ribonucleic acid inSaccharomyces cerevisiae caused by ethidium bromide

Ethidium bromide in a concentration of 200 μg/ml causes a full inhibition of RNA synthesis in aSaccharomyces cerevisiae ρ° strain, while protein synthesis continues at a reduced rate. Under these conditions, processing of rRNA is slowed down and part of the 37S rRNA precursor molecules are cleaved to a 32S RNA fraction (molecular weight 2.15×10⁶). The 32S RNA accumulates in cells treated with ethidium bromide but cannot be processed to mature 25S and 18S rRNA and is degraded. The 32S RNA fraction also appears when processing of rRNA occurs in cells starved for required amino acids. The degradation of 37S precursor molecules through 32S RNA may be a regulatory mechanism of rRNA biosynthesis in yeast, which operates when excess rRNA must be wasted.     

Ribonucleic Acid Synthesis in Cells Infected with Herpes Simplex Virus: I. Patterns of Ribonucleic Acid Synthesis in Productively Infected Cells

HEp-2 cells were pulse-labeled at different times after infection with herpes simplex virus, and nuclear ribonucleic acid (RNA) and cytoplasmic RNA were examined. The data showed the following: (i) Analysis by acrylamide gel electrophoresis of cytoplasmic RNA of cells infected at high multiplicities [80 to 200 plaque-forming units (PFU)/cell] revealed that ribosomal RNA (rRNA) synthesis falls to less than 10% of control (uninfected cell) values by 5 hr after infection. The synthesis of 4S RNA also declined but not as rapidly, and at its lowest level it was still 20% of control values. At lower multiplicities (20 PFU), the rate of inhibition was slower than at high multiplicities. However, at all multiplicities the rates of inhibition of 18S and 28S rRNA remained identical and higher than that of 4S RNA. (ii) Analysis of nuclear RNA of cells infected at high multiplicities by sucrose density gradient centrifugation showed that the synthesis and methylation of 45S rRNA precursor continued at a reduced but significant rate (ca. 30% of control values) at times after infection when no radioactive uridine was incorporated or could be chased into 28S and 18S rRNA. This indicates that the inhibition of rRNA synthesis after herpesvirus infection is a result of two processes: a decrease in the rate of synthesis of 45S RNA and a decrease in the rate of processing of that 45S RNA that is synthesized. (iii) Hybridization of nuclear and cytoplasmic RNA of infected cells with herpesvirus DNA revealed that a significant proportion of the total viral RNA in the nucleus has a sedimentation coefficient of 50S or greater. The sedimentation coefficient of virus-specific RNA associated with cytoplasmic polyribosomes is smaller with a maximum at 16S to 20S, but there is some rapidly sedimenting RNA (> 28S) here too. (iv) Finally, there was leakage of low-molecular weight (4S) RNA from infected cells, the leakage being approximately three-fold that of uninfected cells by approximately 5 hr after infection.     

Kern-Plasma-Transport neusynthetisierter rRNS in Zellen einer Suspensionskultur von Petroselinum sativum

Summary A rapidly labelled rRNA precursor can be detected in callus cells of Petroselinum sativum grown on a liquid synthetic medium. Its molecular weight has been calculated to be 2.3×10⁶. This value agrees with that of the rRNA precursor from other plant material. In order to follow the synthesis and processing of rRNA in time and to correlate single steps in this process with cell organelles it was necessary to obtain pure fractions of nuclei and ribosomes. The isolation method for nuclei is given in detail. The nucleic acids are separated on polyacrylamide gels of low acrylamide concentration. Pulse-chase experiments show that the rRNA precursor is split into two fragments within the nucleus: an 18S and a 25S component. The 18S RNA leaves the nucleus rapidly. It is already found quantitatively in the ribosomal fraction after 30–60 min chase. At that time the 25S RNA is still within the nucleus; it appears much later in the ribosomes. Since the increase in ribosomal label occurs simultaneously with the decrease in nuclear label, it is concluded that there is no degradation of 18S RNA within the nucleus. Apparently there are two distinct transport mechanisms with different kinetics for the two RNA components.     

Use of a DNA probe for mapping by electron microscopy the ribosomal sequences in ribosomal RNA precursors from duck cells

This report describes the use of purified ribosomal DNA to map by electron microscopy the relative positions of the 18 S and 28 S RNA regions within the duck rRNA precursor and their relationship to the non-conserved portions of the precursor molecule. By repeated fractionation of the total DNA, based on the relative reassociation rates of the DNA sequences with different degrees of repetition, a fraction of the rapidly renaturing DNA was obtained which comprised only 6% of the total DNA, but contained 71% of the rRNA cistrons. Further purification of the rDNA was achieved by saturation hybridization with rRNA and separation of the rRNA-rDNA hybrids by banding in CsCl. In this manner, an rDNA-rRNA fraction was obtained which had a buoyant density of 1.805 g/cm³, an RNA to DNA ratio of 1.01, and a base composition for the RNA present in the hybrid identical to that of an equimolar mixture of 18 S and 28 S rRNA. The final yield of rDNA isolated by this procedure is 32%. When the purified rDNA was annealed with a mixture of 18 S and 28 S rRNA and the hybrids spread for electron microscopy, they appeared as two distinct populations with a number-average length of 0.62 ± 0.13 μm and 1.37 ± 0.18 μm, respectively. Likewise, hybrids between the rRNA precursor, isolated from duck embryo fibroblasts, and the rDNA appeared as structures containing two duplex regions of lengths 0.60 ± 0.11 μm and 1.38 ± 0.15 μm, separated from each other by a single-stranded region appearing as a large bush: this represents a portion of the precursor molecule not conserved during processing of the parent molecule. From these observations a model of the structure of the duck rRNA precursor is proposed.     

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

The processing of ribosomal RNA has been studied in a temperature sensitive mutant of the Syrian hamster cell line BHK 21. At 39 degrees C, these cells are unable to synthesize 28S RNA, and 60S ribosomal subunits, while 18S RNA, and 40S subunits are produced at both temperatures. At 39 degrees C the 45S RNA precursor is transcribed and processed as in wild type cells. The processing of the RNA precursors becomes defective after the cleavage of the 41S RNA, and the separation of the 18S and 28S RNAs sequences in two different RNA molecules. The 36S RNA precursor, which is always present in very small quantity in the nucleoli of wild type cells and of the mutant at 33 degrees C, is found in very large amounts in the mutant at 39 degrees C. The 36S RNA can be, however, slowly processed to 32S RNA. The 32S RNA cannot be processed at 39 degrees C, and it is degraded soon after its formation. Only a small proportion accumulates in the nucleoli. The 32S RNA synthesized at 39 degrees C cannot be processed to 28S RNA upon shift to the permissive temperature, even when the processing of the newly synthesized rRNA has returned to normal. The data suggest that the 36S and 32S RNAs are contained in aberrant ribonucleoprotein particles, leading to a defective processing of the particles as a whole.     

Wasting of 18 S ribosomal RNA by human myeloma cells cultured in adenosine.

When human myeloma cells are pulsed for one hour with 3H-uridine and chased for six hours in fresh medium containing unlabeled uridine, the processing of 45 S rRNA precursor into the stable 28 S and 18 S rRNA components can be followed. However, when the cells are chased in exogenous adenosine instead of uridine, the accumulation of 18 S rRNA is selectively inhibited. Cells pulsed with 3H-adenosine and chased in the absence of exogenous nucleosides exhibit normal rRNA precursor processing, while cells pulsed simultaneously with 3H-uridine and 3H-adenosine and chased with uridine and adenosine are deficient in labeled 18 S rRNA. Consequently, the inhibition of 18 S rRNA accumulation by adenosine is not an artifact of labeling nor is it relieved by an equal molar concentration of uridine. The wasting of 18 S rRNA in human myeloma cells is similar to that reported to occur in normal lymphocytes during the quiescent state.     

The relative amounts of the cytoplasmic RNA species in normal, transformed and senescent cultured cell lines.

We have examined the relative quantities of 18S and 28S rRNA, 4S RNA and poly (A) + mRNA in the following cultured cells: the mouse fibroblast lines 3T3 and 3T6 in the resting (contact inhibited) and growing (sparse) states, 3T3 clones transformed with SV40 (SV3T3) and with both SV40 and polyoma SV-Py 3T3), hamster lung fibrobalsts (v79), human cervical carcinoma cells (HeLa), and human diploid fibroblasts at early and late passage. The relative quantities of the RNA species were determined by labeling the cells to equilibrium with 32PO4 and measuring the amount of label in each RNA species. The ratio of mRNA to rRNA varied form 1.1% to 2.7% in the different cell lines, the more rapidly growing cell lines usually giving a higher ratio. In cells experiencing growth limitation either by contact inhibition or due to senescence, the ratio of mRNA to rRNA was about 30% lower than in the corresponding cells in the growing state. In most cell lines the ratio of 4S RNA to 18S rRNA was between 0.8 and 1.2, but in seescent fibroblasts, this ratio increased to greater than 1.7. Senescent fibroblasts also contained much more total RNA per unit of DNA than the same cells at early passage or than 3T6 or 3T3 cells.     

Synthesis and maturation of ribosomal ribonucleic acids in isolated HeLa cell nuclei. A tracer study on the topology of the 45S precursor of ribosomal ribonucleic acids

The synthesis and processing of RNA by isolated HeLa cell nuclei was studied at low ionic strength in the presence of alpha-amanitin. The RNA polymerase reaction, with endogenous template and enzyme, rapidly reaches a plateau dependent on the amount of nuclei. Evidence is presented that incorporation of [(3)H]UMP proceeds only in growing RNA chains, whereas initiation of new RNA chains is arrested. The product formed contains all the main components of the 45S pre-rRNA (precursor of rRNA) maturation pathway (45S, 32S and 20S pre-rRNA; 28S and 18S rRNA). Most of the labelled material is in the mature rRNA components and their immediate precursors, even at very short times of incubation (2min). Small, but definite, 5S and 4S RNA peaks are also observed. At shorter incubation times a substantial amount of [(3)H]UMP is incorporated into RNA molecules in the 24S and 10-16S zones. This RNA material is considered to represent the non-conserved segments of 45S pre-rRNA in the process of nucleolytic degradation. A model for the tracer study of the topology of 45S pre-rRNA, on arrest of rRNA initiation, is discussed. The experimental evidence obtained supports the following structure of 45S pre-rRNA: 5'-end-28S rRNA unit-18S rRNA unit-nonconserved segment-3'-end.     

Secondary structure maps of ribosomal RNA. II. Processing of mouse L-cell ribosomal RNA and variations in the processing pathway

Secondary structure mapping in the electron microscope was applied to ribosomal RNA and precusor ribosomal RNA molecules isolated from nucleoli and the cytoplasm of mouse L-cells. Highly reproducible loop patterns were observed in these molecules. The polarity of L-cell rRNA was determined by partial digestion with 3′-exonuclease. The 28 S region is located at the 5′-end of the 45 S rRNA precursor. Together with earlier experiments on labeling kinetics, these observations established a processing pathway for L-cell rRNA. The 45 S rRNA precursor is cleaved at the 3′-end of the 18 S RNA sequence to produce a 41 S molecule and a spacer-containing fragment (24 S RNA). The 41 S rRNA is cleaved forming mature 18 S rRNA and a 36 S molecule. The 36 S molecule is processed through a 32 S intermediate to the mature 28 S rRNA. This pathway is similar to that found in HeLa cells, except that in L-cells a 36 S molecule occurs in the major pathway and no 20 S precusor to 18 S RNA is found. The processing pathway and its intermediates in L-cells are analogous to those in Xenopus laevis, except for a considerable size difference in all rRNAs except 18 S rRNA.The arrangement of gene and transcribed spacer regions and of secondary structure loops, as well as the shape of the major loops were compared in L-cells, HeLa cell and Xenopus rRNA. The over-all arrangement of regions and loop patterns is very similar in the RNA from these three organisms. The shapes of loops in mature 28 S RNA are also highly conserved in evolution, but the shapes of loops in the transcribed spacer regions vary greatly. These observations suggest that the sequence complementarity that gives rise to this highly conserved secondary structure pattern may have some functional importance.     

Coordination of ribosomal RNA synthesis in vertebrate cells

Xenopus embryo cells and HeLa cells were investigated under various conditions to test for coordinate synthesis of high molecular weight (28S and 18S) and low molecular weight (5S) rRNA. Xenopus embryos initiate 28S and 18S rRNA synthesis at gastrulation (Brown and Littna, '64); we found that 5S rRNA synthesis is coordinately initiated with the 28S and 18S rRNAs at the same time in development. Dissociated Xenopus blastula cells were cultured in vitro for several hours to condition the medium; post-gastrula cells were then grown in the conditioned medium to test for the existence of an inhibitor of rRNA synthesis. No inhibitor was detected. Low doses of actinomycin D profoundly inhibit the synthesis of 28S and 18S rRNA in HeLa cells, while 5S rRNA synthesis is less affected by this treatment. Therefore, actinomycin D does not produce a coordinate inhibition of all rRNA species. Similar effects of the antibiotic were found in cultured amphibian cells. Synchronized HeLa cells reinitiating RNA synthesis following mitosis also respond to actinomycin D in a non-coordinate manner.