ON THE NATURE OF NUCLEOLAR RNA* BY MARGARET I. H. CHIPCHASE AND MAX L. BIRNSTIELt - PNAS
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
ON THE NATURE OF NUCLEOLAR RNA*
BY MARGARET I. H. CHIPCHASE AND MAX L. BIRNSTIELt
DIVISION OF BIOLOGY, CALIFORNIA INSTITUTE OF TECHNOLOGY
Communicated by James Bonner, October 10, 1963
It is now generally held that the structural RNA of cytoplasmic ribosomes is of
nuclear origin. Several kinds of studies, including autoradiography, have implied
that the nucleolus plays a role in ribosomal RNA economy. 1-4 Biochemical investi-
gations have shown that an RNA of base composition similar to that of cytoplasmic
ribosomal RNA5 and of identical molecular sizes, namely, 28 and 18 S, occurs in the
nucleolus.6 In addition, ribosome-like particles have been observed in the nucleolus
by electron microscopy7-1' and have been shown to have sedimentation constants
similar to those of cytoplasmic ribosomes and their subunits.6
The hybridization of RNA with denatured DNA provides a tool for the identifi-
cation of DNA complementary in base sequence to any given RNA.12-16 Applica-
tion of this technique has enabled us to discover that nucleolar RNA is identical to
cytoplasmic ribosomal RNA in base sequence and may therefore be the nuclear
precursor of cytoplasmic RNA. We will also show that the DNA regions, or
stretches complementary to cytoplasmic ribosomal RNA, are not confined to the
nucleolus but are distributed throughout the chromatin of the nucleus.
Materials and Methods.-P32 (as orthophosphate) was obtained from the Oak Ridge National
Laboratories, Oak Ridge, Tenn. RNase was purchased from Sigma Chemical Co., St. Louis, Mo.
T4 DNA was a gift from Dr. Roger Weil of the California Institute of Technology, Pasadena,
Calif. Membrane filters type A coarse, Lot no. 2427, were obtained from Schleicher and Schuell,
Keene, N. H.
Preparation of labeled RNA: Fifty gm of peas were sterilized with detergent, germinated for 12
hr, rinsed thoroughly with distilled water containing 1 mg penicillin-G per ml, and incubated in
a shallow dish 24-36 hr in the presence of 3-5 Cm 95% radiophosphorus (as sodium phosphate)
in 10 ml water containing 50 ,ug penicillin per ml. After incubation, the seedlings were washed, and
the P32 was chased for 12 hr with 0.001 M sodium phosphate, pH 7.0 (10,000-fold excess of p31).
The seedlings were then harvested and ribosomes prepared according to the protocol of Ts'o et al."7
The repeated recycling of the ribosomes selects 80 S particles and provides preparations virtually
free of sRNA (as determined by sucrose gradient centrifugation). The ribosomal RNA was de-
ploteinized twice according to the procedure of Wallace et al.,18 and once with an equal volume
of 88% phenol according to Kirby. The RNA preparations were then fractionated by the sucrose
gradient centrifugation described earlier.-9 The 28 and 18 S RNA components (peak fractions)
were collected, and each was further purified by a second sucrose gradient centrifugation. The
labeled ribosomal RNA subunits were then centrifuged through 3 ml of a CsCl solution, pH 7.0
(1.73 gem-3) in a Spinco SW 39 at 35,000 rpm for 12 hr in the cold20 in order to free the RNA of any
contaminating DNA. The specific activities of the two RNA components were identical in each
preparation and ranged from 5 X 106 to 30 X 106 cpm/mg in different experiments.
Preparation of DNA: Pea nuclei were prepared21 and purified6 as described earlier. Nucleoli
were obtained by grinding nuclei in sucrose media containing citrate22 and purified by sedimenta-
tion through 2.2 M sucrose at 20,000 rpm for 20 min in the SW 25 Spinco head.
DNA was prepared from nuclei and nucleoli by a modification of Marmur's method.22 After
5-10 successive deproteinizations, aliquots of the DNA were dissolved in 1 ml CsCl (pH 7.5, 1.80
gCm-3) and overlayered with 1 ml each of a CsCl solution at 1.7 and 1.6 gem. -3 After centrifuga-
tion at 36,000 rpm in the Spinco SW 39 for 36 hr to separate DNA from any lingering RNA con-
tamination, the tube was pierced, and the highly viscous DNA fractions were collected. The
pooled DNA fractions were then dialyzed for 8 hr against three changes of cold dilute saline citrate
buffer. The melting profile of the DNA showed a Tm of 70-71°C (in dilute saline citrate buffer)
and a hyperchromicity of 37%.
Downloaded by guest on October 22, 2021
11011102 BIOCHEMISTRY: CHIPCHASE AND BIRNSTIEL PROC. N. A. S.
Hybrid formation and determination: The annealing conditions used were those described by
Scherrer et al.16 Routinely, 50 ug DNA and 2-4 jg P32-labeled RNA were used. For the forma-
tion of complex, incubations of 2-3 hr at 630C and of 24 hr at 420C were equally effective, and
saturating values were reached in both cases. The RNA-DNA mixture was brought to a density
of 1.81 gcm-3 by the addition of CsCl, transferred to a Spinco SW 39 centrifuge tube, and over-
layered with more CsCl solution of the same density to a volume of 1.5 ml. The solution was
then overlayered with 1.5 ml of CsCl solution (1.63 gcm-3) and 1.5 ml of paraffin oil, and the
samples were centrifuged (in the cold) for 72 hr at 33,000 rpm or, alternatively, for 50 hr at 35,000
rpm. The centrifuge tubes were then pierced and five-drop fractions collected. O.D. at 260 mu
and p32 radioactivity were determined for each fraction. The fractions containing the hybrid peak
were collected, dialyzed against dilute saline citrate buffer, and RNased with 5 jug/ml DNase-free
RNase for 1-2 hr at room temperature. The RNase-resistant material was precipitated with
5% TCA in the presence of carrier RNA and its radioactivity determined with a Nuclear-Chicago
D-181 micromil window gas flow counting system. The contribution of the RNase-resistant core
of the nonhybridized P32-RNA was found to be 4%, and corresponding adjustments in the final
values were made.
For preliminary experiments, the filter disk method of Nygaard and Hall24 was employed.
After completion of the annealing process, the RNA-DNA mixture was incubated with RNase 5
,ug/ml for 30 min at 30'C. The mixture was then filtered through the membrane filter and washed
twice with 5 ml of 0.5 NaCl + 0.01 M tris pH 7.4 at 450C. The results obtained by this method
are in general agreement with those obtained by CsCl centrifugation.
Results.-The hybrid between ribosomal RNA and DNA is characterized by its
extreme stability toward RNase and its buoyant density which is intermediate be-
tween those of DNA and RNA. These two features may be easily demonstrated
with hybrids formed between P32-labeled ribosomal RNA from pea cytoplasm and
DNA from pea nuclei. Thus, the hybrid is resistant to ribonuclease treatment
(Fig. 1). In addition, the hybrid'3 possesses a buoyant density in CsCl significantly
higher than that of DNA (Fig. 2). DNA becomes saturated-that is, combines
with all of the ribosomal RNA with which it is capable of combining-at a level of
0.17 ,4g RNA/100 jig DNA in the case of the 28 S RNA subunit and at a level of
100 10,000 2.0
4
Z 0 -RNos*
A 4-RNosej COUNTS
0
u0E7,500
=' V
0
2.05 75 10 l 12 14 6 20 2
50 ->5,000 .0~
~i 2,500 / 0.5
0
2.5 5 7.5 10 I10 12 14 16 20 ?22
INCUBATION TIME FRACTION
FIG. 1.-RNase resistance of the FIG. 2.-CsCl density gradient anal-
hybrid between cytoplasmic ribosomal ysis of the DNA-RNA hybrid. De-
RNA and denatured pea DNA. DNA natured pea DNA was annealed with
was annealed with P32-labeled cyto- P32-labeled cytoplasmic ribosomal RNA
plasmic ribosomal RNA and the hybrid and the product centrifuged in a CsCJ
prepared as described in Materials and density gradient until equilibrium had
Methods. The hybrid was incubated been attained. Centrifugation and
with 5 jg/ml of DNase-free RNase at determination of radioactivity and ab-
30'C and aliquots were withdrawn and sorbancy were carried out as described
precipitated for counting with 5% in Materials and Methods. The peak
TCA in the presence of carrier RNA. of labeled RNA is shifted to a density
Incubation time in hours. higher than that of the DNA.
Downloaded by guest on October 22, 2021VOL. 50, 1963 BIOCHEMISTRY: CHIPCHASE AND BIRNSTIEL 1103
0.30 jhg/100 ,g DNA for the combined 28 and }
18 S RNA fractions (Fig. 3). The amount of 8
ribosornal RNA complexed per unit DNA is J 0.3 Z
thus similar to that found in bacterial sys- 0 a
Dz
tems. 12-14 o O02
0.
< 4
/28S
alone
The following measures were taken to ensure M
no appreciable interference by informational .1/ no
RNA. In all experiments the time of incuba-
tion of tissue in P32 was sufficiently lengthy
(24-36 hr) to permit uniform labeling of all 2 3 4 5
RNA.25 At 50teLg Lg p32 INPUT PER DNA
~
cellular
2cellular the end of the incubation
incubation, RNA
FIG. 3.-Saturation curves for hybrid
the bulk of the radioactivity therefore resides formation between denatured pea
in the ribosomal RNA fraction which accounts DNA and ribosomal RNA subunits.
for approximately 80 per cent of the total annealedIn each case, 50 pg of DNA were
with varying amounts of P"2-
RNA. In addition, after the incubation with labeled RNA and the hybrids assayed
p32 a large excess (10,000-fold) of unlabeled as described in Materials and Methods.
orthophosphate was added, and the seedlings were incubated for an additional
12 hr. This provides time for the turnover of unstable RNA. Selection
against messenger RNA also occurs during preparation of the 80 S ribosomes and
the subsequent repetitive sucrose gradient centrifugations of the deproteinized RNA
in which only the 28 and 18 S peak fractions were collected. We find, too, that the
hybrid between P32-labeled ribosomal RNA and DNA has a buoyant density higher
than that of DNA. This is in agreement with the findings of Yanofsky and Spiegel-
man'3 who have shown that the hybrid between E. coli ribosomal RNA and E.
coli DNA is denser than is DNA. This is not true of hybrids between informational
RNA and DNA whose density coincides with that of DNA,'3 due perhaps to com-
plexing of only a very small portion of any given DNA molecule with its comple-
mentary RNA.
A fourfold excess of E. coli ribosomal RNA does not interfere with the annealing
of P32-labeled pea ribosomal RNA to pea DNA. This confirms previous findings
that heterologous RNA's do not compete for the same DNA since hybrid formation
is base sequence specific. There is, as we would expect, competition between
P32-labeled and unlabeled pea cytoplasmic ribosomal RNA for pea DNA. The
amount of P32-labeled RNA complexed is inversely proportional to the input of
unlabeled RNA (Table 1). Nucleolar (ribosomal) RNA prepared from the whole
TABLE 1
SATURATION OF DENATURED NUCLEAR AND NUCLEOLAR PEA DNA BY RNA OF RIBOSOMAL
SUBUNITS
p&g P32RNA/
DNA source p32 Pea RNA p31 RNA source 100 pg DNA
Pea nucleus 28 + 18 S ... 0.29
T4 28 + 18 S ... 0.02
Pea nucleus 28 + 18 S E. coli ribosomal RNA (4X) o.31
Pea nucleus 28 + 18 S Pea cytoplasmic ribosomal RNA 0.16
(1X)
Pea nucleus 28 + 18 S Pea cytoplasmic ribosomal RNA 0.08
(4X)
Pea nucleus 28 + 18 S Pea nucleolar 2M NaCi precipitable 0.10
RNA (4X)
Pea nucleus 28 + 18 S Pea nucleolar total RNA (4X) 0.13
Values were determined from saturation curves as in Fig. 2.
Downloaded by guest on October 22, 20211104 BIOCHEMISTRY: CHIPCHASE AND BIRNSTIEL PROC. N. A. S.
TABLE 2
HYBRID FORMATION BETWEEN DENATURED PEA DNA AND HETEROLOGOUS AND HOMOLOGOUS
RNA's
DNA source p32 RNA pg p32 RNA/100 pg DNA
Nucleus 28 + 18 S 0.30
Nucleus 28 S alone 0.17
18 S (by difference) 0.13
Nucleoli 28 + 18 S 0.32
Nucleoli 28 S alone 0.19
Annealing procedure and hybrid assay as described in Materials and Methods.
nucleolar RNA by 2 M NaCl precipitation27-29 also competes effectively with p32_
labeled ribosomal RNA in complex formation. Whole nucleolar RNA is some-
what less effective in this function since about half of it is low-molecular-weight
RNA (Fig. 4) identified as transfer RNA26 by its capacity to engage in aminoacyl
RNA formation. That the RNA extractable from the nucleolus is not derived
from contamination by cytoplasmic ribosomes has been previously shown.6
DNA obtained from nucleus-free nucleolar prep-
arations also complexes with cytoplasmic ribosomal
RNA and reaches saturation at levels only slightly
higher than those characteristic of the whole nuclear
o DNA (Table 2). The DNA stretches complementary
¶/1 \^to t
\VOL. 50, 1963 BIOCHEMISTRY: CHIPCHASE AND BIRNSTIEL 1105
plexing with ribosomal RNA, 0.3 per cent, indicates that in the nucleus, as in
bacteria, there is a multiplicity of DNA stretches complementary to ribosomal RNA.
Since some 10-15 per cent of the nuclear DNA is intimately associated with the
nucleolus,6, 22, 30-33 exclusive intranucleolar localization of the DNA complementary
to ribosomal RNA would reveal itself by causing nucleolar DNA to become satu-
rated with ribosomal RNA at a level 7-10 times higher than that characteristic of
total genomal DNA. Our results show, however, that the nuclear and nucleolar
DNA become saturated with ribosomal RNA at approximately the same' level.
It is clear therefore that while a portion of the cistrons complementary to ribosomal
RNA is found in the nucleolus, the great majority of cistrons are not, but are
to be found elsewhere in the genome. Base complementarity of DNA to cytoplas-
mic ribosomal RNA is therefore a characteristic not of the nucleolus, but of the
chromatin which may be calculated to contain 90 per cent of all the cistrons com-
plementary to ribosomal RNA and possibly more, since the 10 per cent of the
nuclear DNA present in the nucleolar preparations are known to include some
random chromatin contamination.34 The possibility of an exchange and therefore
randomization in situ between nucleolar and extranucleolar DNA may be dis-
missed as shown by the experiments of Harries.31
Our data suggest that ribosomal RNA is synthesized principally by the extra-
nucleolar chromatin and is only subsequently transferred to the nucleolus. Such
transfer of newly synthesized RNA from chromatin to the nucleolus has been re-
ported by Rho and Bonner2 for pea embryo tissue.
We have previously reported the existence in the nucleolus of a large pool of
protein which resembles ribosomal protein in amino acid composition.3' We have
also shown that this protein fraction becomes labeled when whole cells, nuclei,36
or isolated nucleoli3' are incubated with labeled amino acids. Movement of pro-
tein from nucleolus to extranucleolar chromatin is indicated by pulse-chase experi-
ments in vivo37 and in vitro.30 We conceive of the possibility that ribosomal pro-
tein of nucleolar origin complexes with newly synthesized ribosomal RNA on or
near the chromatin and that the resulting ribonucleoprotein is then transferred
to the nucleolar periphery. The considerable heterogeneity of the nucleolar ribo-
nucleoproteins, the large proportion of subunits present,6 and their inability to sum
port protein synthesis6' 36 38 all suggest that the nucleolar ribosomes are in fact un-
finished ribosomal precursors whose accumulation in the nucleolar region may be
for the final modification and completion required for production of functional
ribosomal units.
Summary.-It has been shown that the structural RNA of cytoplasmic ribosomes
of pea seedlings can be hybridized with denaturated pea-seedling DNA. Such hy-
bridization, at RNA saturation, involves 0.3 per cent of the total genomal DNA
indicating that a multiplicity of DNA stretches are involved in the production of
ribosomal RNA. The RNA of the nucleolus also hybridizes with the DNA stretches
which are complementary to ribosomal RNA. Nucleolar RNA would appear there-
fore to contain ribosomal RNA. Only a minority of ribosomal RNA cistrons occur
in the nucleolar DNA while the vast majority is to be found in the chromatin. It
is concluded that the bulk of the ribosomal RNA-if not all-is manufactured by
nonnucleolar regions of the chromatin and that the ribosomal RNA is then trans-
ferred into the nucleolus for final assembly into ribosomes.
Downloaded by guest on October 22, 20211106 BIOCHEMISTRY: CHIPCHASE AND BIRNST'IEL PROC. N. A. S.
*
Report of work supported by grants from ACS (IN-39C) and USPHS (GM-03977-10 and
AM-3102). We are grateful to Dr. Jerome Vinograd for his help in running the CsCl band cen-
trifugations and to Dr. Melvin Green for his advice concerning the disk method of hybrid deter-
mination. We would also like to thank Professor James Bonner for his support and help in pre-
paring the manuscript, and Dr. Gary Flamm for his advice and criticism.
t Present address: Institute of Animal Genetics, University of Edinburgh, Edinburgh, Scotland.
I Sirlin, J. L., in The Cell Nucleus, ed. J. S. Mitchell (London: Butterworths, 1960), p. 35.
2 Rho, J. H., and J. Bonner, these PROCEEDINGS, 47, 1611 (1961).
3Sirlin, J. L., Progr. Biophys. Biophys. Chem., 12, 25, 319 (1962).
4 Busch, H., P. Byvoet, and K. Smetana, Cancer Res., 23, 313 (1963).
'Edstr6m, J. E., W. Grapp, and N. Schor, J. Biophys. Biochem. Cytol., 11, 549 (1961).
6 Birnstiel, M. L., M. I. H. Chipchase, and B. B. Hyde, Biochim. et Biophys. Acta, in press.
7 Swift, H., in Structure and Function of Genetic Elements, Brookhaven Symposia in Biology,
No. 12 (1959), p. 134.
8 Porter, K. R., in Proceedings of the Fourth International Conference on Electron Microscopy
(Berlin: Springer Verlag, 1960), vol. 2., p. 189.
9 Marinozzi, V., in Proceedings of the Fifth International Congress on Electron Microscopy (New
York: Academic Press, 1962).
10Jacob, J., and J. L. Sirlin, J. Cell Biol., 17, 153 (1963).
11 Lafontaine, J. G. and L. A. Couinard, J. Cell Biol., 17, 167 (1963).
12 Yanofsky, S. A. and S. Spiegelman, these PROCEEDINGS, 48, 1069 (1962).
13Ibid., 1466.
14 Ibid., 538.
15 Goodman, H. M., and A. Rich, these PROCEEDINGS, 48, 2101 (1962).
16Scherrer, K., H. Latham, and J. E. Darnell, these PROCEEDINGS, 49, 240 (1963).
17 Ts'o, P. 0. P., J. Bonner, and J. Vinograd, Biochim. et Biophys. Acta, 30, 570 (1958).
18 Wallace, J. M., R. F. Squires, and P. O. P. Ts'o, Biochim. et Biophys. Acta, 49, 130 (1961).
19 Chipchase, M. I. H., and M. L. Birnstiel, these PROCEEDINGS, 49, 692 (1963).
20 Vinograd, J., R. Brunner, R. Kent, and J. Weigle, these PROCEEDINGS, 49, 902 (1963).
21 Rho, J. H., and M. I. H. Chipchase, J. Cell Biol., 14, 183 (1962).
22 Birnstiel, M. L., J. H. Rho, and M. I. H. Chipchase, Biochim. et Biophys. Acta, 55, 734 (1962).
23 Marmur, J., J. Mol. Biol., 3, 208 (1961).
24 Nygaard, A. P., and B. D. Hall, Biophys. Biochem. Res. Commun., 12, 98 (1963).
25 Ts'o, P. 0. P., and C. S. Sato, Exptl. Cell Research, 17, 237 (1959).
26 Birnstiel, M. L., E. Fleissner, and E. Borek, Science, in press.
27Smith, K. C., Biochim. et Biophys. Acta, 40, 360 (1960).
8 Osawa, S., Biochim. et Biophys. Acta, 43, 110 (1960).
29 Rosenbaum, M., and R. A. Brown, Anal. Biochem., 2, 15 (1961).
30 Birnstiel, M. L., M. I. H. Chipchase, and J. Bonner, Biophys. Biochem. Res. Commun., 6,
161 (1961).
31 Birnstiel, M. L., and B. B. Hyde, J. Cell Biol., 18, 41 (1963).
32 Monty, K. J., M. Litt, E. R. M. Kay, and A. L. Dounce, J. Biophys. Biochem. Cytol., 2, 127
(1956).
33 Maggio, R., P. Siekevitz, and G. E. Palade, Proceedings of the 2nd Meeting, American Society
of Cell Biology (1962).
34 Our data do not show how many of the DNA cistrons complementary to ribosomal RNA are
active in RNA synthesis since we do not yet know whether all of the multiple DNA cistrons which
code for ribosomal RNA are alike in base sequence or whether they are different and produce a
diversity of ribosomal RNA's. If the latter were to be the case, then the hybridization technique
might ultimately yield an unequivocal answer concerning which of the DNA cistrons which code
for ribosomal RNA are active in the cell and where they are located in the nucleus. If, however,
all ribosomal RNA's are identical, synthesis at one locus (e.g., the nucleolus) will yield RNA
capable of annealing with DNA of cistrons other than those from which it was derived (e.g.,
chromatin).
There is in our data (see also ref. 13) an indication that the ribosomal RNA's are diverse. This
is the fact that the RNA input required for saturation of DNA is far in excess of the RNA ac-
Downloaded by guest on October 22, 2021VOL. 50, 1963 MATHEMATICS: I. FLEISCHER 1107
tually complexed. This could be due to low concentrations of particular ribosomal species in a
population of ribosomal RNA. Such a diversity of ribosomal RNA could arise from mutation and
would parallel the degeneracy of sRNA.
35 Harries, H., Biochem. J., 73, 362 (1959).
3 Birnstiel, M. L., and W. G. Flamm, in preparation.
3 De, D. N., The Nucleus, 4, 1 (1961).
38 Flamm, W. G., and M. L. Birnstiel, Biology and Chemistry of the Histones (San Francisco:
Holden-Day, 1963), in press.
A CHARACTERIZATION OF LEXICOGRAPHICALLY
ORDERED nYou can also read
