| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
(Received for publication, August 5, 1994; and in revised form, October 24, 1994) From the
Poly(A) polymerase synthesizes poly(A) tails rapidly and
processively only when the substrate RNA is bound simultaneously by two
stimulatory proteins, the cleavage and polyadenylation specificity
factor (CPSF) and poly(A)-binding protein II (PAB II). A burst of
synthesis terminates after the addition of about 250 nucleotides, a
length corresponding to that of newly synthesized poly(A) tails in
vivo. Further elongation is slow. Length control can be reproduced
with premade poly(A) tails of different lengths and is insensitive to
large changes in the elongation rate. Thus, the control mechanism truly
measures the length of the poly(A) tail. The stimulatory action of PAB
II is similar on long and short tails. Coating of poly(A) with one PAB
II molecule for approximately 30 nucleotides is required, such that the
number of PAB II molecules in the polyadenylation complex is a direct
measure of poly(A) tail length. CPSF also stimulates poly(A) polymerase
on long and short tails. Long tails differ from short ones only in that
they do not permit the simultaneous stimulation of poly(A) polymerase
by CPSF and PAB II. Consequently, elongation of long tails is
distributive. Thus, length control is brought about by an interruption
of the interactions responsible for rapid and processive elongation of
short tails. The 3`-end of the poly(A) tail is not sequestered in the
protein-RNA complex when the correct length has been reached. Neither
ATP hydrolysis nor turnover of the polymerized AMP is involved in
length control. The poly(A) tails found at the 3`-ends of nearly all eukaryotic
mRNAs are added during RNA processing in the cell nucleus and are later
shortened after the RNA has been transported into the cytoplasm. The
poly(A) tail has at least two different functions: the initiation of
translation and the regulation of mRNA breakdown (reviewed by Sachs
(1990), Jackson and Standart(1990), Bachvarova(1992), Wickens(1992),
Sachs(1993), and Sachs and Wahle(1993)). A role in the export of mRNA
from the cell nucleus is also possible (Wickens and Stephenson, 1984;
Eckner et al., 1991). Evidence for a role in translation
initiation has been provided by genetic experiments in yeast (Sachs and
Davis, 1989, 1990), studies on RNA introduced by electroporation into a
number of different cell types (Gallie, 1991), in vitro translation experiments (Munroe and Jacobson, 1990), and
microinjection of RNAs into oocytes (Galili et al., 1988;
Vassalli et al., 1989; Paris and Richter, 1990; Sheets et
al., 1994). The oocyte studies suggest that not only the presence
or absence of a poly(A) tail is important but that its length can
determine translational efficiency. Although this does not prove that
tail length is also important for translation in other cells, regulated
changes in poly(A) tail length have been described that are correlated
with changes in translation in somatic cells (Robinson et al.,
1988). Evidence for a role of the poly(A) tail in the regulation of
mRNA degradation comes from the observation that deadenylation is the
first and rate-limiting step in the breakdown of at least some unstable
mRNAs both in mammals and in yeast (Wilson and Treisman, 1988; Shyu et al., 1991; Muhlrad and Parker, 1992; Decker and Parker,
1993; Muhlrad et al., 1994). Sequences in the 3`-untranslated
region and the coding sequence can control the half-life of the mRNA by
influencing the rate of deadenylation (Shyu et al., 1991;
Muhlrad and Parker, 1992; Lowell et al., 1992). A control
of the rate of deadenylation is useful only if the length of poly(A)
initially added in the cell nucleus is also controlled. In mammalian
cells, newly synthesized bulk poly(A) is approximately 200-250
nucleotides long (Brawerman, 1981). Individual mRNAs that have been
examined, c-fos and several c-fos- Nuclear polyadenylation is preceded by an endonucleolytic
cleavage of the primary transcript at the site of poly(A) addition. In vitro, specific polyadenylation of a
``precleaved'' RNA already ending at the poly(A) site has
been reconstituted from three proteins purified from mammalian tissue
(reviewed by Wahle and Keller (1992) and Sachs and Wahle(1993)).
Poly(A) polymerase, the enzyme that catalyzes polyadenylation, is
devoid of any pronounced specificity with regard to the RNA substrate
and is also almost inactive on its own (Wahle, 1991a). Two other
proteins activate the enzyme. One of these is the cleavage and
polyadenylation specificity factor (CPSF), ( The
polyadenylation apparatus reconstituted as described above also
reproduces the length control observed either in vivo or in
extracts. Once a poly(A) tail length of close to 250 nucleotides has
been reached, further elongation becomes very slow (Wahle, 1991b;
Bienroth et al., 1993). This paper describes the reproduction
of true length control in vitro and presents evidence that it
is due to a switch from processive to distributive synthesis.
Figure 1:
Poly(A) tail length control with
premade tails. A 25
Figure 2:
Length control requires CPSF and PAB II. A, a 33
Standard reaction conditions included 80 fmol of RNA, 100
fmol of CPSF, 200 fmol of PAB II, and 9 fmol of poly(A) polymerase.
When the concentration of CPSF in the reaction was increased, length
control was unchanged. Likewise, a 2-fold or 10-fold increase in PAB II
concentration did not relieve length control (data not shown, but see Fig. 4). Measurements of the initial rate of poly(A) chain
growth showed that the standard amount of PAB II was optimal with 9 or
180 fmol of poly(A) polymerase. Whereas at the higher concentration of
poly(A) polymerase a significantly larger fraction of precursor RNA was
elongated immediately after the start of the reaction, the initial rate
of chain growth was not affected; it was 20-25 nt/s at either
concentration of poly(A) polymerase ( Fig. 3and data not shown).
This confirms the processive nature of the reaction. Other experiments
showed that, in contrast, the elongation of long tails was proportional
to the concentration of poly(A) polymerase, as expected for a
distributive reaction (see below).
Figure 4:
PAB II dependence of burst elongation with
different initial poly(A) tail lengths. A 26
Figure 3:
The
initial rate of burst elongation. An 8
The requirement for only a
2-3-fold molar excess of PAB II over RNA was surprising. An
estimate of a packing density of 1 PAB II molecule/23 adenylate
residues (Wahle et al., 1993) suggests that the amount of PAB
II used in these experiments is sufficient to coat the short oligo(A)
tails present at the beginning of the reaction but not the long tails
generated during elongation. Therefore, the PAB II requirement was
tested during burst elongation of different initial poly(A) tail
lengths (Fig. 4). As mentioned above, burst elongation at
optimal PAB II concentrations always generated the same final poly(A)
tail length, irrespective of the initial tail length. Higher
concentrations of PAB II did not overcome the length limitation; in
fact, they were inhibitory with respect both to the tail length
generated and the number of molecules elongated. Suboptimal
concentrations of PAB II also led to shorter average poly(A) tails.
With 80 fmol of an A
Figure 5:
PAB II dependence of poly(A) elongation in
the absence of CPSF. A 34
Figure 6:
Competition of polyadenylated and
oligoadenylated RNA. A 25
Figure 7:
Stability of poly(A) tails. A 10
Figure 8:
Competition between short and long tails
in the presence of ATP analogs. An 18
The cell controls the length of poly(A) tails in three ways:
during synthesis in the nucleus, during shortening in the cytoplasm,
and, at least in special cases, during readenylation in the cytoplasm.
The first type of control, limitation of poly(A) tail synthesis to a
length of approximately 250 nucleotides, can be reconstituted from
three purified proteins, poly(A) polymerase, CPSF, and PAB II. The two
stimulatory factors, which are also responsible for the primer
specificity of the polyadenylation reaction, act by forming a complex
that includes poly(A) polymerase and the substrate RNA. This complex is
able to synthesize a full-length poly(A) tail processively without
disintegration (Bienroth et al., 1993). The length of poly(A)
tails generated during this processive burst of synthesis was higher in
this study than in previous experiments (Wahle, 1991b). The reason
seems to be that length control depends on certain properties of
poly(A) polymerase. Once a full-length tail has been made, polyadenylation
essentially terminates by a transition from processive to distributive
elongation. Direct evidence for a distributive elongation of long
poly(A) tails is a synchronous elongation of all RNA molecules under
conditions of limiting poly(A) polymerase and a dependence of the
elongation rate on the concentration of poly(A) polymerase. By both
criteria, elongation of short tails is processive. The processivity of
the latter reaction has also been demonstrated by a primer challenge
experiment (Bienroth et al., 1993). Also, under conditions
that only allow distributive elongation of short tails (presence of
only one stimulatory factor), there is no significant difference
between the elongation rates of short and long tails. Thus, a switch
from processive to distributive elongation as the mechanism for length
control also explains the protein requirements of the reaction; only
those conditions that are appropriate for processive elongation of
short tails show length control. This includes the presence of both
CPSF and PAB II and the use of a homologous poly(A) polymerase. Under all other conditions, long and short tails behave quite
similarly. Also, under conditions appropriate for length control, long
tails still grow at a rate far exceeding the unstimulated rate observed
with poly(A) polymerase alone. Thus, there is no lack of stimulation by
individual factors and no inhibition of the elongation of long tails.
In particular, the two RNA binding factors do not sequester the 3`-end
of the RNA in a nonaccessible complex once the correct length has been
reached. There is only a specific inability of long tails to mediate
the interaction of all three proteins that is responsible for
processive elongation. The length control mechanism involves a true
measurement of poly(A) tail length, as evidenced by the function of
length control with premade poly(A) tails of different lengths, the
lack of sensitivity to large changes in the rate of elongation, and the
ability of the polyadenylation machinery to discriminate between short
and long tails when challenged with a mixture of the two. What is
the yard stick that the polyadenylation complex uses to measure poly(A)
tail length? The most obvious possibility is that the complex senses
the number of PAB II molecules it contains. PAB II can coat high
molecular weight poly(A) at a stoichiometry of nearly 25 nt/protein
(Wahle et al., 1993), and there is no reason why it should not
do so during poly(A) tail elongation. In the absence of CPSF,
elongation of a poly(A) tail clearly requires its binding of multiple
molecules of PAB II. The optimal amount of PAB II tends to be slightly
less than expected from measurements of binding stoichiometry; thus
elongation may be most efficient when tails are not completely coated.
The requirement for a similar amount of PAB II in the presence of CPSF
is more apparent when the poly(A) tails present at the start of the
reaction are longer (Fig. 4). This reveals that the experiment
is probably unable to measure the true PAB II optimum. The higher
concentrations that are optimal for the long tails generated during the
reaction are inhibitory for the short ones present initially ( Fig. 4and Fig. 5). If the measurement of poly(A) tail
length involves coating the tail with PAB II, one might predict that,
at the low PAB II concentrations sufficient to induce processive
elongation of short tails, length control should not be functional.
Evidently, this is not the case. This might mean that the termination
of processive elongation depends on poly(A) tail length in a manner not
involving the number of PAB II molecules. However, detailed kinetics
always showed a lack of synchrony during burst elongation (Fig. 3). Thus, the number of PAB II molecules bound to an
individual poly(A) tail may differ from what is calculated from the
molar ratio in the entire reaction, because the population of RNA
molecules is probably heterogeneous. Some molecules may recruit a
sufficient amount of PAB II to terminate processive synthesis, whereas
others may be devoid of PAB II and thus also fail to be elongated
efficiently. Regardless of the exact mechanism, a disruption of the
polyadenylation complex must take place once 250 As have been
polymerized. The nature of this event remains to be determined. From
careful in vivo pulse-labeling studies of HeLa cells, Sawicki et al.(1977) concluded that there were two distinct modes of
adenosine incorporation into nuclear poly(A) tails. Synthesis of a
full-length poly(A) tail of about 230 nucleotides was found to take
less than a minute. In addition, poly(A) tails that were already
full-length at the time of pulse labeling were subject to a slower end
addition with 5-10 nucleotides added in a 2-min pulse. It is
reassuring that the kinetics described here for the purified in
vitro system can fully account for the in vivo incorporation kinetics.
Volume 270,
Number 6,
Issue of February 10, 1995 pp. 2800-2808
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
-globin
chimeras, have initial poly(A) tails of 150-300 nucleotides
(Schiavi et al., 1994). A limitation of poly(A) tail synthesis
to approximately 200 nucleotides has been observed during in vitro polyadenylation in crude nuclear extracts (Sheets and Wickens,
1989).
)which binds the
polyadenylation signal AAUAAA that is present in almost every mRNA
precursor (Bienroth et al., 1991; Keller et al.,
1991; Murthy and Manley, 1992). The second stimulatory factor is a
poly(A)-binding protein (PAB II), which binds the growing poly(A) tail
(Wahle, 1991b; Wahle et al., 1993). Each of these two factors
directs poly(A) polymerase specifically to RNAs carrying the respective
protein binding site, either AAUAAA or an oligo(A) tail of at least 10
nucleotides. However, poly(A) polymerase works most efficiently when it
is stimulated by both factors simultaneously (Wahle, 1991b). The enzyme
on its own is entirely distributive. When held on the RNA by the
combined efforts of CPSF and PAB II, the enzyme attains sufficient
processivity to synthesize a complete poly(A) tail in one single
binding event. Either stimulatory factor alone leads only to a very
small increase in processivity (Bienroth et al., 1993).
Polyadenylation Proteins
PAB II was taken from
the peak fraction of the preparation described earlier (Wahle et
al., 1993). Protein concentration was 0.47 mg/ml as determined by
a Bradford assay (Bradford, 1976) and 0.54 mg/ml as determined by
quantitative amino acid analysis of a dialyzed aliquot (Wahle et
al., 1993). Calculations were based on a concentration of 0.5
mg/ml. CPSF was purified by a modification of the procedure of Bienroth et al.(1991). Spermine-agarose chromatography was left out.
The pool of the poly(A)-Sepharose column was concentrated by
centrifugation through Centricon-30 and further purified by gel
filtration on a Superose 6 prep grade fast protein liquid
chromatography column. Active fractions were pooled and concentrated on
Centricon-30. Purity and activity were comparable with those of
published preparations (Bienroth et al., 1991). Total protein
concentration as determined by a Lowry assay (Scopes, 1987) was 0.52
mg/ml. Concentration measurements by densitometric scanning of the
73-kDa subunit in a SDS-polyacrylamide gel stained with Coomassie
Brilliant Blue and comparison to a BSA standard gave 0.13 mg/ml,
assuming equimolar amounts of all subunits. Active CPSF as determined
in a gel retardation assay (Bienroth et al., 1991) with excess
L3pre RNA was at least 200 pmol/ml, corresponding to 0.07 mg/ml. The
latter concentration was the basis of calculations. Poly(A) polymerase
was a gift from Georges Martin. The protein was expressed in Escherichia coli from the bovine cDNA (Wahle et al.,
1991; Raabe et al., 1991) under the control of a T7 promoter
with 6 N-terminal histidine residues. It was purified on a
Ni-nitrilotriacetic acid column (Hochuli et
al., 1987) followed by chromatography on a MonoQ-fast protein
liquid chromatography column. (
)The protein was nearly
homogeneous on a SDS-polyacrylamide gel. The specific activity in the
presence of Mn and saturating amounts of oligo(A)
primer was 19
10
units/mg, very similar to the
activity of the 60-kDa proteolytic fragment purified from calf thymus
(Wahle, 1991a). (
)The concentration of poly(A) polymerase
(0.15 mg/ml) was based on a comparison with a BSA standard on a
SDS-polyacrylamide gel stained with Coomassie Brilliant Blue. Yeast
poly(A) polymerase purified to homogeneity from recombinant E. coli (Lingner et al., 1991b) was a gift from Pascal Preker.
Dilution buffer for all proteins was 5 mM Tris-HCl, pH 8.0,
10% (v/v) glycerol, 50 mM NaCl, 0.5 mM EDTA, 0.2
mg/ml methylated BSA, 0.01% (v/v) Nonidet P-40, 1 mM dithiothreitol.RNA
The standard polyadenylation substrate L3pre,
derived from the L3 polyadenylation site of adenovirus-2, and
conditions for its synthesis by SP6 RNA polymerase have been described
(Christofori and Keller, 1988; Bienroth et al., 1993). RNA was
made without a cap. Products were run on 6% polyacrylamide gels in 1
TBE, 8.3 M urea (Sambrook et al., 1989), and
the RNA was eluted overnight at 37 °C in 750 mM ammonium
acetate, 10 mM magnesium acetate, 0.1 mM EDTA, 0.5%
SDS, mixed with 0.1 volume of buffer-saturated phenol. The RNA was
extracted once with phenol/chloroform and precipitated with ethanol. No
carrier tRNA was added. The amount of RNA obtained was calculated from
the specific activity of [
-P]UTP and the
number of U residues in the RNA. Transcription reactions also produced
a small amount of a slightly shorter transcript. Because it was
polyadenylated with the same efficiency as the correct RNA, no attempt
was made to separate the two. Polyadenylated RNAs were made from
gel-purified L3pre by the nonspecific Mn
-dependent
polyadenylation reaction essentially as described (Wahle, 1991b;
Bienroth et al., 1993). RNAs with poly(A) tails of the desired
lengths were gel-purified as above. Poly(A) tail lengths were
determined on analytical polyacrylamide gels with end-labeled DNA
fragments as markers. L3pre lacking poly(A) (expected size, 65
nucleotides) had an apparent length of approximately 70 nucleotides.
tRNA (Boehringer Mannheim) was dissolved, extracted with
phenol/chloroform, precipitated with ethanol, washed with 70% ethanol,
and dissolved in water. Concentration measurements assumed that A
= 1 corresponded to 40 µg/ml.
Poly(A) (Boehringer Mannheim) was dissolved in water, extracted with
phenol/chloroform, and dialyzed exhaustively against water. The
concentration was determined spectrophotometrically with
=9800. Measurements after complete alkaline hydrolysis
with
= 15,400 (3`-AMP, pH 13) were in
excellent agreement (less than 4% deviation). As analyzed by 5`-end
labeling with polynucleotide kinase and gel electrophoresis under
denaturing conditions, the poly(A) had a very heterogeneous size
distribution. Most chains were longer than 200 nucleotides with a large
fraction longer than the longest marker (622 nucleotides).
Other Materials
SP6 RNA polymerase, yeast
hexokinase, proteinase K, ATPS, and AMPPNP were from Boehringer
Mannheim. RNAguard, Superose 6 prep grade, and other fast protein
liquid chromatography columns and equipment were from Pharmacia Biotech
Inc., radiolabeled nucleotides were from Amersham, Centricon-30
filtration devices were from Amicon, Inc., and polyethyleneimine plates
for thin-layer chromatography were from Merck. Polyvinyl alcohol (cold
water soluble) was from Sigma, and
Ni
-nitrilotriacetic acid resin was from Diagen
(Hilden, FRG). Methylated BSA was made as described (Wahle, 1991b).
Polyadenylation Reactions
Reactions contained, in
25 µl, 25 mM Tris-HCl, pH 7.9 (measured at 1 M and room temperature), 10% (v/v) glycerol, 2.6% (w/v) polyvinyl
alcohol, 50 mM KCl, 2 mM MgCl, 0.05
mM EDTA, 0.01% (v/v) Nonidet P-40, 0.4 mg/ml methylated BSA,
0.5 mM ATP, 1 mM dithiothreitol, 4 units of RNAguard,
80 fmol of radiolabeled substrate RNA, 9 fmol of poly(A) polymerase,
100 fmol of CPSF, and 200 fmol of PAB II. For some experiments, tRNA
was added at 2.5 µg/reaction, and 18 fmol of poly(A) polymerase was
used. For longer poly(A) tails, the amount of PAB II was increased as
indicated for individual experiments. Incubations were at 37 °C for
the time indicated. For kinetic experiments, larger reaction mixtures (e.g. 10 = 250 µl) were assembled on ice in
the absence of ATP and then prewarmed at 37 °C. Polyadenylation was
started by the addition of ATP. Reactions were terminated by the
addition of 2-fold concentrated proteinase K digestion buffer (Wahle,
1991a) and water to a final volume of 100 µl. Digestion was carried
out with 10 µg of proteinase K and 2.5 µg of tRNA for 30 min at
37 °C. RNA was precipitated with 2.5 volumes of ethanol, washed
with 70% ethanol, dried, dissolved in formamide loading buffer, and
analyzed on 6% polyacrylamide gels (40 cm long) in 1 TBE, 8.3 M urea (Sambrook et al., 1989). Rates of
polyadenylation were calculated from estimates of average poly(A) tail
length after different times of reaction. Incorporation of
[P]AMP into unlabeled RNA was measured by
adsorption to DE81 paper (Stayton and Kornberg, 1983). In experiments
in which the production of ADP and AMP in polyadenylation reactions was
measured by polyethyleneimine thin-layer chromatography, incorporated
nucleotides were taken to be those remaining at the origin.
Other Procedures
SDS-polyacrylamide gels were run
according to Laemmli(1970). Nucleotides were analyzed by thin-layer
chromatography on polyethyleneimine plates with 1 M formic
acid, 0.5 M LiCl as a solvent.
Reconstitution of Length Control with Premade Poly(A)
Tails
Derivatives of the precleaved polyadenylation substrate
RNA L3pre carrying poly(A) tails of different lengths were prepared
(see ``Experimental Procedures'') and used in polyadenylation
assays with purified poly(A) polymerase, CPSF, and PAB II. These
reactions with premade poly(A) tails faithfully reproduced the length
control seen during de novo poly(A) synthesis. An L3pre
derivative carrying 30 adenylate residues at its 3`-end was elongated
to an average tail length of approximately 250 nucleotides within 1 min
or less. Further elongation was slow. This corresponds to the kinetics
described previously (Wahle, 1991b; Bienroth et al., 1993). A
derivative of the same RNA already carrying 250 adenylate residues did
not undergo the initial burst of rapid elongation; it was only
elongated in the slow mode (Fig. 1). RNAs with intermediate tail
lengths were elongated rapidly to a tail length of 250 nucleotides
before the elongation rate decreased (see below). Thus, the length
control mechanism measures the length of the poly(A) tail present,
rather than the time that poly(A) polymerase has spent elongating the
substrate RNA. Further evidence for this conclusion will be presented
below. From the same experiment one may also conclude that, whereas the
burst of elongation of short tails is processive, the slow elongation
of long tails is distributive. In the reaction, substrate RNA was
present in excess over poly(A) polymerase. Elongation of a small
fraction of the short tails to a length of 250 nucleotides with no
extension of the majority of the substrate at early times is indicative
of the processive nature of the reaction as described before (Bienroth et al., 1993). Under the same conditions, the entire
population of RNA molecules carrying long tails was elongated in a
synchronous manner, demonstrating a distributive behavior of the
polymerase. Some shortening of the substrate, evident in Fig. 1,
was due to an apparently nonspecific 3`-exonuclease activity that
contaminates CPSF preparations. Exonucleolytic degradation of the
labeled substrate RNA could be suppressed by the addition of tRNA (see
below). However, for the sake of simplicity, this was not done in most
experiments.
reaction mixture lacking RNA and ATP was set
up on ice and divided into three equal portions. Two of these were
mixed with the substrate RNAs indicated at the bottom, and a
25 µl aliquot was withdrawn from each (0 min time point). The rest
of each mixture was prewarmed for 3 min at 37 °C. Reactions were
started by the addition of ATP, and additional aliquots were taken at
various times after ATP addition as indicated. The size of DNA markers
(in nucleotides) is indicated on the left.
Protein Requirements
Length control was only
observed in reactions containing poly(A) polymerase with both
stimulatory factors, CPSF and PAB II. When the reaction was carried out
with either factor alone, elongation was slow even though a 10-fold
higher concentration of poly(A) polymerase was used compared with the
reaction containing both stimulatory factors (Fig. 2). The low
rate of polyadenylation is due to the distributive nature of the
reaction under these conditions (Bienroth et al., 1993). Most
importantly, when poly(A) tail length was plotted versus time (Fig. 2B), no sudden drop in chain growth rate was
observed at any particular tail length. A progressive decrease in the
chain growth rate in the CPSF-dependent reaction is probably caused by
the increasing distance between the CPSF binding site and the 3`-end to
be elongated. The decrease in rate in the PAB II-dependent reaction is
likely to be due to the decreasing ratio of PAB II to poly(A) (see
below).
standard reaction mixture containing
L3pre-A was set up in the absence of all polyadenylation
factors and ATP and divided into four equal portions. Polyadenylation
factors were added to three of these as indicated. The two tubes
containing either only PAB II or only CPSF received a 10-fold higher
amount of poly(A) polymerase (90 fmol/25 µl). A 25-µl aliquot
was withdrawn from each tube (0 min time point), and the rest was
prewarmed for 3 min at 37 °C. The reactions were started by the
addition of ATP, and additional aliquots were taken as indicated. The
size of DNA markers (in nucleotides) is indicated on the left. B, poly(A) tail lengths were estimated from A and
plotted against reaction time. Open circles, reactions
containing CPSF and PAB II; filled circles, reactions
containing CPSF; squares, reactions containing PAB
II.
reaction mixture
was assembled in the absence of RNA, ATP, and PAB II. Poly(A)
polymerase was used at 90 fmol/25 µl. The mixture was split into
three equal portions, which were mixed with RNAs as indicated at the bottom. 25 µl aliquots were then dispensed, and PAB II was
added as indicated. One aliquot from each mixture was mixed with
proteinase K digestion buffer without prior incubation to show the
unreacted substrate (first lane in each set of eight). The
others were prewarmed to 37 °C for 2 min. Reactions were started by
the addition of ATP and stopped after 30 s by the addition of
proteinase K digestion buffer. The size of DNA markers (in nucleotides)
is indicated on the left.
reaction mixture was
assembled with L3pre-A in the absence of ATP. Compared
with the standard conditions, a 20-fold higher concentration of poly(A)
polymerase was used (180 fmol/25 µl). A 25-µl aliquot was
withdrawn (0 min time point). The rest of the mixture was prewarmed for
3 min at 37 °C and then kept at this temperature. 25-µl
aliquots were drawn up into a pipette tip, expelled into prewarmed
tubes containing ATP, drawn up into the same pipette tip, and
transferred to proteinase K digestion buffer. The size of DNA markers
(in nucleotides) is indicated on the left.
tail, 200 fmol of PAB II were optimal
(8 adenylates/PAB II based on initial tail length), with an A
tail, 300 fmol of PAB II were optimal (20 adenylates/PAB II), and
with an A
tail, 400 fmol of PAB II were optimal (30
adenylates/PAB II). Although it is apparent that optimal elongation
requires more than a single molecule of PAB II for each poly(A) tail,
the effect of different PAB II concentrations was not very strong, and
the interpretation of PAB II titrations in this type of reaction is not
straightforward (see ``Discussion''). Thus, the PAB II
requirement was also tested in the elongation of poly(A) tails in the
absence of CPSF. This simpler reaction showed a much more pronounced
dependence on the exact PAB II concentration. The elongation of 80 fmol
of an RNA with an A
tail was most efficient with 1200
fmol of PAB II. This corresponds to a ratio of between 1 PAB II/23
adenylate residues, based on the initial tail length, and 1 PAB II/30
adenylate residues, based on the final tail length of 450 nucleotides.
Higher concentrations of PAB II were inhibitory. Additional titrations,
some of them with shorter poly(A) tails, gave comparable ratios (Table 1), their accuracy being limited by the number of
nucleotides added during the reaction. Saturation of the same molar
quantities of shorter tails at lower concentrations of PAB II is not
only further evidence that coating of the tail is required but also
demonstrates that the experiments measured the stoichiometry of the PAB
II requirement rather than the affinity of PAB II for poly(A). In the
elongation of a simple unlabeled poly(A) primer with radiolabeled ATP,
the amount of PAB II required for maximum incorporation was also
proportional to the amount of poly(A) present. Saturation was achieved
near 35 nucleotides/PAB II monomer (Fig. 5). Inhibition became
apparent with amounts of PAB II exceeding 1/30 adenylate residues. In
these experiments, AMP incorporation at the optimum levels of PAB II
was between 20 and 38% of the poly(A) added as a primer. The optimal
ratio of protein to poly(A) based on the initial poly(A) concentration
may thus be slightly overestimated.
reaction was assembled lacking CPSF,
PAB II, RNA, and ATP. Poly(A) polymerase was included at 360 fmol/25
µl. The mixture was split into three equal aliquots, which received
unlabeled high molecular weight poly(A) at 273, 137, and 55 pmol of
AMP/25 µl, respectively. 25-µl aliquots of each mixture were
mixed with PAB II as indicated. Reactions were started by the addition
of ATP and transfer to 37 °C. After 30 min, they were stopped by
application to DE81 paper. Background was determined by application of
ice-cold reaction mixtures lacking PAB II to DE81 paper immediately
after the addition of ATP. The value obtained (0.8 pmol of AMP) was
subtracted from the incorporation measured in all other reactions. Open circles, 273 pmol of poly(A)/25 µl; filled
circles, 137 pmol of poly(A)/25 µl; triangles, 55
pmol of poly(A)/25 µl.
Elongation of Long Poly(A) Tails
Although poly(A)
tails of more than 250 adenylate residues fail to undergo the
processive elongation typical of short tails, they are nevertheless
good primers for the PAB II-stimulated activity of poly(A) polymerase.
In the experiments described in Table 1, excessively long tails
were elongated by 100 or more nucleotides in the presence of the
optimal PAB II concentration, whereas PAB II-independent elongation was
barely detectable. In fact, more precise measurements showed that the
elongation rate of long tails was, if anything, slightly higher than
the elongation rate of short tails at their respective PAB II optima (Table 2). Similarly, poly(A) polymerase assisted by CPSF was
only 5-fold slower on long compared with short tails. This may reflect
the distance between the 3`-end being elongated and the CPSF binding
site. A serious deficiency of long tails was apparent only in the
presence of both PAB II and CPSF; long tails were extended 300-fold
more slowly than short tails. Whereas the simultaneous presence of both
factors increased the elongation rate for short tails by a factor of
300-1000 compared with either factor alone, the effect was only
2-5-fold for long tails (Table 2). When, in the presence of
both stimulatory factors, the concentration of poly(A) polymerase was
increased up to 10-fold, the elongation rate of long tails increased
proportionally (data not shown). This is further evidence for a
distributive mechanism as discussed above. The rate of extension of
long tails in the presence of both CPSF and PAB II was at least as high
as with either factor alone and about 50-fold higher than in the
absence of any stimulatory factor (near 0.1 nt/min under these
conditions). Thus, length control is not due to an inhibition of the
elongation of long poly(A) tails. Neither is it due to a lack of
interaction between poly(A) polymerase and CPSF or PAB II. Rather, it
is caused by the inability of long tails to mediate the simultaneous
effects of PAB II and CPSF on poly(A) polymerase that are the basis for
the burst of processive elongation of short tails.
Accessibility of 3`-Ends
A simple possibility for
length control would be the length-dependent formation of an
RNA-protein complex that prevents further elongation by sequestering
the 3`-end of the RNA. Although the lack of inhibition evident from Table 2does not suggest this model, it can be more rigorously
tested by the use of an enzyme that elongates the RNA nonspecifically.
One such enzyme is poly(A) polymerase of yeast. Because the yeast
enzyme does not respond to mammalian CPSF and PAB II (Lingner et
al., 1991a, 1991b) (additional data not shown), a 90-fold larger
amount of yeast poly(A) polymerase compared with the mammalian enzyme
was used for reasonable chain growth rates. No control of poly(A) tail
length was found. With an RNA carrying 30 A residues, the initial rate
of elongation was 14 nt/min until a total tail length of 70 As had been
reached. The elongation rate was 8 nt/min in the interval between 125
and 300 adenylates and 6 nt/min between 300 and 500 adenylates. In a
control reaction with a second aliquot of the same reaction mixture,
mammalian poly(A) polymerase showed the expected kinetics. When a
mixture of short and long premade tails was used, yeast poly(A)
polymerase in the presence of CPSF and PAB II elongated both substrates
at similar rates (25 nt/min for short tails and 10 nt/min for long
tails), whereas the mammalian enzyme first elongated the short tails to
nearly 250 adenylates before it touched the long tails (Fig. 6).
In conclusion, these experiments suggest that there is no sequestration
of the 3`-end of the poly(A) tail to terminate elongation. When in an
experiment as shown in Fig. 6both yeast and mammalian poly(A)
polymerase were present simultaneously, no sequestration was observed
either. In more general terms, CPSF and PAB II are not sufficient for
length control. The mammalian poly(A) polymerase is needed, because, in
the presence of the two stimulatory factors, this enzyme but not the
yeast enzyme is capable of processive burst elongation.
reaction mixture containing 40 fmol/25
µl each of L3pre-A and L3pre-A
was
assembled in the absence of poly(A) polymerase and ATP. PAB II was used
at 1200 fmol/25 µl. A 4
aliquot was mixed with 36 fmol of
bovine poly(A) polymerase (PAP), and a 5 aliquot was
mixed with 3900 fmol of yeast poly(A) polymerase. 25-µl aliquots
were withdrawn from each tube (0 min time point). The rest was
prewarmed for 2 min at 37 °C before the reactions were started by
the addition of ATP. Additional aliquots were withdrawn as indicated.
The size of DNA markers (in nucleotides) is indicated on the left.
Stability of Poly(A) Tails
DNA polymerases as well
as RNA polymerases have exonucleolytic activities (Brutlag and
Kornberg, 1972; Kornberg and Baker, 1992; Surratt et al.,
1991; Kassavetis and Geiduschek, 1993), and the same has been claimed
for poly(A) polymerase (Abraham and Jacob, 1978; Tarui and Minamikawa,
1989). Therefore, the possibility was considered that exonucleolytic
shortening of the poly(A) tail might play a role in length control. For
these experiments, the exonuclease activity contaminating the CPSF
preparation (see above) had to be suppressed by tRNA as a nonspecific
competitor. This did not cause any change in length control. When under
these conditions poly(A) synthesis was abruptly stopped at a late stage
by ATP depletion with hexokinase and glucose, the poly(A) tails that
had been synthesized were completely stable, even though they exceeded
a length of 250 nucleotides (data not shown). When a mixture of short
and excessively long tails was exposed to the polyadenylation factors
in the absence of ATP, both were completely stable during extended
incubations. Upon addition of ATP, short tails were preferentially
extended in a processive manner (Fig. 7). Finally, when under
similar conditions [-P]ATP was used to
elongate unlabeled L3pre RNA, AMP production (as a result of
exonucleolytic excision of incorporated nucleotides) was at the limit
of detection, at most 5% of the net incorporation. Analysis of
additional aliquots of the same reaction by gel electrophoresis
demonstrated that length control was normal, and both the gel pattern
and calculations showed that incorporation was almost exclusively into
the polyadenylation substrate and not the tRNA used to suppress the
exonuclease (data not shown). In conclusion, exonuclease activity is
not involved in poly(A) tail length control. Tests under stringent
conditions have also not detected any poly(A)-degrading activity in
poly(A) polymerase itself.
reaction mixture containing 40 fmol/25 µl each of L3pre-A and L3pre-A
was assembled in the absence of poly(A)
polymerase and ATP. Reactions contained tRNA and PAB II at 2000 fmol/25
µl. A 25-µl aliquot was withdrawn (0 min time point). The rest
was mixed with poly(A) polymerase (18 fmol/25 µl) and transferred
to 37 °C. Aliquots were taken after 5, 10, and 20 min. After 21
min, ATP was added, and additional aliquots were withdrawn as
indicated. The size of DNA markers (in nucleotides) is indicated on the left.
Role of ATP
We have previously reported that
poly(A) polymerase can use ATP analogs with nonhydrolyzable
- bonds (Bienroth et al., 1993). Length
control is also functional with these analogs; in a kinetic experiment,
elongation with AMPPNP proceeded approximately 50-fold more slowly than
with ATP and slowed down at a tail length of about 200 nucleotides.
After 90 min, when elongation had become very slow, an aliquot of the
reaction mixture was added to fresh oligoadenylated precursor RNA. This
fresh substrate was still elongated, effectively competing out the
previously made tails (data not shown). Thus, in spite of the long
incubation time, inactivation of proteins was not the reason for the
decrease in the elongation rate. Experiments with ATP
S also showed
normal length control with an elongation rate intermediate between ATP
and AMPPNP (data not shown). In a competition between long and short
tails, the polyadenylation machinery almost exclusively selected the
short RNA in the presence of ATP, ATP
S, or AMPPNP (Fig. 8).
Thus, the hydrolysis of the
- bond of ATP is not
required for length control. (All experiments involving AMPPNP were
done in the presence of hexokinase and glucose to exclude possible ATP
contaminations. In control experiments with ATP, hexokinase completely
abolished polyadenylation within seconds.) From the same experiments,
one may also conclude that length control is essentially insensitive to
large changes in the rate of chain elongation. This confirms that, as
discussed above, the length control mechanism truly measures the length
of the poly(A) tail present.
reaction mixture was set
up with 40 fmol/µl each of L3pre-A and L3pre-A
in the absence of ATP. PAB II was used at 2000 fmol/25 µl.
Three tubes each received a 5
aliquot of the mixture, one of
which also received hexokinase (1.25 units) and glucose (10
mM). 25-µl aliquots were withdrawn from each tube (0 min
time points), and the rest was prewarmed for 3 min at 37 °C.
Reactions were started by the addition of ATP, ATPS, or AMPPNP,
respectively. AMPPNP was added to the mixture containing hexokinase and
glucose. Additional aliquots were taken as indicated. The rest of the
original 18
mixture was used to demonstrate that hexokinase and
glucose prevented polyadenylation (not shown). The size of DNA markers
(in nucleotides) is indicated on the left.
Whereas previously the proteolytically
shortened poly(A) polymerase from calf thymus (Wahle, 1991a) was used,
the experiments reported here were carried out with a full-length
protein purified from recombinant E. coli. This observation
suggests that length control is not rigid and may be subject to
regulation.
),-imidodiphosphate; ATP
S, adenosine
5`-O-(3-thiotriphosphate); BSA, bovine serum albumin; PAB II,
poly(A) binding protein II; nt, nucleotide(s).
))
I thank Georges Martin and Pascal Preker for their
gifts of purified poly(A) polymerases, Walter Keller for his support,
and Witek Filipowicz, Andreas Jenny, Walter Keller, Lionel Minvielle,
and Helen Rothnie for reading the manuscript.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Fronz, S. Otto, K. Kolbel, U. Kuhn, H. Friedrich, A. Schierhorn, A. G. Beck-Sickinger, A. Ostareck-Lederer, and E. Wahle Promiscuous Modification of the Nuclear Poly(A)-binding Protein by Multiple Protein-arginine Methyltransferases Does Not Affect the Aggregation Behavior J. Biol. Chem., July 18, 2008; 283(29): 20408 - 20420. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Viphakone, F. Voisinet-Hakil, and L. Minvielle-Sebastia Molecular dissection of mRNA poly(A) tail length control in yeast Nucleic Acids Res., April 1, 2008; 36(7): 2418 - 2433. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Addepalli and A. G. Hunt A novel endonuclease activity associated with the Arabidopsis ortholog of the 30-kDa subunit of cleavage and polyadenylation specificity factor Nucleic Acids Res., July 26, 2007; 35(13): 4453 - 4463. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Perreault, C. Lemieux, and F. Bachand Regulation of the Nuclear Poly(A)-binding Protein by Arginine Methylation in Fission Yeast J. Biol. Chem., March 9, 2007; 282(10): 7552 - 7562. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. LIANG and C. S. LUTZ p54nrb is a component of the snRNP-free U1A (SF-A) complex that promotes pre-mRNA cleavage during polyadenylation. RNA, January 1, 2006; 12(1): 111 - 121. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Wang, D. D. Mosser, and J. Bag Induction of HSP70 expression and recruitment of HSC70 and HSP70 in the nucleus reduce aggregation of a polyalanine expansion mutant of PABPN1 in HeLa cells Hum. Mol. Genet., December 1, 2005; 14(23): 3673 - 3684. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Dheur, K. R. Nykamp, N. Viphakone, M. S. Swanson, and L. Minvielle-Sebastia Yeast mRNA Poly(A) Tail Length Control Can Be Reconstituted in Vitro in the Absence of Pab1p-dependent Poly(A) Nuclease Activity J. Biol. Chem., July 1, 2005; 280(26): 24532 - 24538. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Venkataraman, K. M. Brown, and G. M. Gilmartin Analysis of a noncanonical poly(A) site reveals a tripartite mechanism for vertebrate poly(A) site recognition Genes & Dev., June 1, 2005; 19(11): 1315 - 1327. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. TAVANEZ, P. CALADO, J. BRAGA, M. LAFARGA, and M. CARMO-FONSECA In vivo aggregation properties of the nuclear poly(A)-binding protein PABPN1 RNA, May 1, 2005; 11(5): 752 - 762. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Seli, M. D. Lalioti, S. M. Flaherty, D. Sakkas, N. Terzi, and J. A. Steitz An embryonic poly(A)-binding protein (ePAB) is expressed in mouse oocytes and early preimplantation embryos PNAS, January 11, 2005; 102(2): 367 - 372. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. S. Chen and T. L. Sheppard Chain Termination and Inhibition of Saccharomyces cerevisiae Poly(A) Polymerase by C-8-modified ATP Analogs J. Biol. Chem., September 24, 2004; 279(39): 40405 - 40411. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Calzado, R. Sancho, and E. Munoz Human Immunodeficiency Virus Type 1 Tat Increases the Expression of Cleavage and Polyadenylation Specificity Factor 73-Kilodalton Subunit Modulating Cellular and Viral Expression J. Virol., July 1, 2004; 78(13): 6846 - 6854. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Wallace, T. L. Denison, E. N. Attaya, and C. C. MacDonald Developmental Distribution of the Polyadenylation Protein CstF-64 and the Variant {tau}CstF-64 in Mouse and Rat Testis Biol Reprod, April 1, 2004; 70(4): 1080 - 1087. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Kuhn, A. Nemeth, S. Meyer, and E. Wahle The RNA Binding Domains of the Nuclear poly(A)-binding Protein J. Biol. Chem., May 2, 2003; 278(19): 16916 - 16925. [Abstract] [Full Text] [PDF]
|
|
|
|
</ |
