A Novel Plant in Vitro Assay System for Pre-mRNA
Cleavage during 3#-End Formation1[OA]
Hongwei Zhao2,3, Jun Zheng2, and Qingshun Quinn Li*
Department of Botany, Miami University, Oxford, Ohio 45056
Messenger RNA (mRNA) maturation in eukaryotic cells requires the formation of the 3# end, which includes two tightly
coupled steps: the committing cleavage reaction that requires both correct cis-element signals and cleavage complex formation,
and the polyadenylation step that adds a polyadenosine [poly(A)] tract to the newly generated 3# end. An in vitro biochemical
assay plays a critical role in studying this process. The lack of such an assay system in plants hampered the study of plant
mRNA 3#-end formation for the last two decades. To address this, we have now established and characterized a plant in vitro
cleavage assay system, in which nuclear protein extracts from Arabidopsis (Arabidopsis thaliana) suspension cell cultures can
accurately cleave different pre-mRNAs at expected in vivo authenticated poly(A) sites. The specific activity is dependent on
appropriate cis-elements on the substrate RNA. When complemented by yeast (Saccharomyces cerevisiae) poly(A) polymerase,
about 150-nucleotide poly(A) tracts were added specifically to the newly cleaved 3# ends in a cooperative manner. The
reconstituted polyadenylation reaction is indicative that authentic cleavage products were generated. Our results not only
provide a novel plant pre-mRNA cleavage assay system, but also suggest a cross-kingdom functional complementation of
yeast poly(A) polymerase in a plant system.
Gene expression in eukaryotes requires the transcription of DNA into mRNA in the nucleus. The
newly transcribed pre-mRNAs undergo extensive processing, such as 5#-end capping and removal of introns
and 3#-end polyadenylation before they are ready
to be transported to the cytoplasm for translation.
Among pre-mRNA processing events, 3#-end formation and polyadenylation are known to regulate transcription termination, affect intron splicing, promote
mRNA transportation and translation initiation, and
protect mature mRNAs from unregulated degradation
(Buratowski, 2005; Moore and Proudfoot, 2009). In
addition, studies on alternative polyadenylation show
that more than half of plant and mammal genes can be
alternatively processed at different locations, resulting
in distinct mature mRNAs from the same pre-mRNA
transcripts (Tian et al., 2005; Xing and Li, 2010; Wu
et al., 2011). More recently, the choice of alternative
polyadenylation sites at 3#-untranslated region (UTR)
of many genes has been linked to gene expression level
1
This work was supported by the U.S. National Institutes of
Health (grant no. GM07719201 to Q.Q.L.), the U.S. National Science
Foundation (grant no. IOS–0817829 to Q.Q.L.), and Miami University Botany Academic Challenge Grants (to J.Z. and H.Z.).
2
These authors contributed equally to the article.
3
Present address: Department of Plant Pathology and Microbiology, University of California, Riverside, CA 92521.
* Corresponding author; e-mail [email protected].
The author responsible for distribution of materials integral to the
findings presented in this article in accordance with the policy
described in the Instructions for Authors (www.plantphysiol.org) is:
Qingshun Quinn Li ([email protected]).
[OA]
Open Access articles can be viewed online without a subscription.
www.plantphysiol.org/cgi/doi/10.1104/pp.111.179465
control and cancer development (Mayr and Bartel,
2009). Thus, a theme of gene expression regulation
through pre-mRNA polyadenylation is emerging. Due
to its tight connections to transcription, translation,
and RNA decay, mRNA 3#-end polyadenylation appears to act as a hub for fine tuning gene expression
and forming a regulation network (Danckwardt et al.,
2008).
The 3#-end processing of mRNA includes two coupled steps: cleavage at a specific site at the 3#-UTR and
an addition of a polyadenosine [poly(A)] tract to the
newly formed 3# end. The biochemical process of
polyadenylation is relatively well studied in mammals
and yeast (Saccharomyces cerevisiae). The 3#-end cleavage and polyadenylation reaction is guided by sequence elements (cis-elements) in pre-mRNA, which
are recognized by a protein complex that carries out
enzymatic cleavage and polyadenylation reaction at a
specific site. This apparently simple process directly
involves several cis-elements, as well as more than 14
protein factors in mammals (Zhao et al., 1999; Mandel
et al., 2008). These 14 protein factors can be divided
into several subcomplexes, such as cleavage and polyadenylation specificity factor (CPSF), cleavage stimulation factor, cleavage factor I, and cleavage factor II.
Meanwhile, poly(A) polymerase (PAP), poly(A) binding protein-II, symplekin, and C-terminal domain of
the RNA polymerase II have also been found to be
necessary in this 3#-end processing complex (Mandel
et al., 2008). In yeast, a similar set of proteins, albeit
named differently, were also identified to be important
for 3#-end processing (Mandel et al., 2008). A recent
report revealed the involvement of over 80 proteins
that form a complex with functional poly(A) signals
(Shi et al., 2009). The functionality of this larger col-
1546 Plant PhysiologyÒ, November 2011, Vol. 157, pp. 1546–1554, www.plantphysiol.org Ó 2011 American Society of Plant Biologists. All Rights Reserved.
Downloaded from on June 18, 2017 - Published by www.plantphysiol.org
Copyright © 2011 American Society of Plant Biologists. All rights reserved.
Pre-mRNA Cleavage during 3#-End Formation
lection of associated proteins in polyadenylation, however, remains to be confirmed.
When investigating the mechanisms of mRNA 3#end formation in mammal and yeast, in vitro assay
systems were found to be critical in identifying and
characterizing components of the cleavage and polyadenylation complex (Zhao et al., 1999). These in vitro
assay systems simplified the isolation and identification of new protein factors involved in this process,
making it much easier to perform subsequent functional studies and test the functions of cis-elements.
Until now, however, an in vitro assay system that can
accurately cleave and polyadenylate pre-mRNA in
plants has never been established, to our knowledge,
and the lack of such a system has affected the study of
3#-polyadenylation machinery in plants (Rothnie,
1996; Li and Hunt, 1997). To address this, numerous
efforts were made in establishing a plant cleavage and
polyadenylation in vitro assay system by using the
nuclear extracts of different plants, such as peas (Pisum
sativum), cauliflower (Brassica oleracea), and Arabidopsis (Arabidopsis thaliana), but without success. Over the
past few years, through bioinformatics, phylogenetics,
protein interactions, and proteomic analysis, we and
others have identified the conserved polyadenylation
factors (Simpson et al., 2003; Herr et al., 2006; Hunt
et al., 2008; Zhao et al., 2009). However, the biochemical efficacy of these proteins remains largely elusive.
Here, we report a plant in vitro mRNA 3#-end assay
system that cleaves pre-mRNA at specific sites as
found in vivo. Nuclear protein extracts from Arabidopsis suspension cells have been shown to carry out
the committing cleavage step. When complemented
with a yeast PAP (yPAP), a full cleavage and polyadenylation activity was reconstituted. This assay system
will be useful for studying plant mRNA 3#-end formation mechanisms as well as other related RNA
processing events.
RESULTS
Establishment of a Plant in Vitro Pre-mRNA Cleavage
System Using Soluble Nuclear Proteins from
Arabidopsis Suspension Cells
While studying the proteomics of plant polyadenylation factors, we overexpressed Arabidopsis CPSF73-I
(for CPSF 73-kD subunit-I; At1g61010) using a tandem
affinity purification tagging approach and isolated the
protein complexes formed with it (Zhao et al., 2009).
Arabidopsis CPSF73-I is a homolog of mammalian
CPSF73 that was shown to be the endonuclease that
plays a direct role in the cleavage step of pre-mRNA
3#-end formation (Dominski et al., 2005; Mandel et al.,
2006). Thus, it was reasonable to examine whether
AtCPSF73-I plays the same role of cleaving premRNAs in plants. To accomplish this, the soluble
nuclear proteins from Arabidopsis suspension cells
overexpressing AtCPSF73-I were prepared from nu-
clei. This extract was free of organelle contamination
such as chloroplast (Fig. 1A), and was used for the
initial establishment of the assay conditions.
A 3#-UTR from the pre-mRNA of cauliflower mosaic
virus (CaMV) transcript clone named STS was used as
the substrate for this in vitro cleavage assay system.
The cleavage and polyadenylation profile of STS has
been extensively studied (Mogen et al., 1990; Rothnie
et al., 1994). Based on conventional genetic studies, it
has been demonstrated that STS is 200 nucleotides
(nts) in length and contains three well-defined ciselements: the far-upstream element (FUE), the nearupstream element (NUE), and a cleavage element (CE)
around the cleavage/polyadenylation site (Fig. 1B).
Accordingly, the nuclear extract was tested for its
cleavage activity using uniformly [g-32P] ATP-labeled
STS in assay conditions as described in the “Materials
and Methods” section. To our surprise, the nuclear
proteins could cleave STS into RNA about 182 nt in
length (Fig. 1C, lane 3), the expected size range as
revealed by in vivo assays (Rothnie et al., 1994). This
cleavage only occurred on the sense transcripts of STS
since the reverse (or antisense) transcript of it could
not be cleaved (compare lanes 1 and 3 in Fig. 1C),
indicating the observed cleavage is sequence specific.
To confirm that the observed cleavage is at the 3# end,
where the in vivo cleavage occurs, a 5#-end 32P-labeled
STS was generated and subjected to the same assay
conditions. Again, the cleavage reaction left a product
that was about 182 nts in length (Fig. 1D, the middle
lane). STS was also 3#-end labeled by [5#-32P] pCp,
using T4 RNA ligase, and then subjected to the same
assay. As shown in Figure 1E (lane 1), a 3#-end cleavage product of about 20 nts was detected. To further
confirm the cleavage site at the sequence level, the 182nt band was cloned and sequenced. In agreement with
the in vivo experiments, the results showed that the
cleavage occurred at the same two sites observed by
Rothnie et al. (1994) producing two different lengths,
180 and 182 nt (not seen as two bands on the gel
possibly due to resolution; Rothnie et al., 1994). These
results indicate that the cleavage was carried out by an
endonuclease generating both the 5# and 3# cleavage
products. This excludes the possibility that such products were generated by an exonuclease activity. Thus,
an in vitro assay system for pre-mRNA cleavage was
established by using the nuclear proteins from cells
overexpressing AtCPSF73-I.
Since we found cleavage activity based on nuclear
extracts overexpressing AtCPSF73-I in the wild-type
cell background, we further asked if the overexpression of AtCPSF73-I was required for this activity. To
address this question, we generated nuclear extracts
from nontransformed, wild-type culture cells, and
tested for their cleavage activity. Surprisingly, the
wild-type nuclear protein was also able to correctly
cleave the pre-mRNA (Fig. 1F). Therefore, our results
indicated that specific pre-mRNA cleavage activity
can be found in both the AtCPSF73-I overexpressing
and the wild-type nuclear protein extracts. Although
Plant Physiol. Vol. 157, 2011
1547
Downloaded from on June 18, 2017 - Published by www.plantphysiol.org
Copyright © 2011 American Society of Plant Biologists. All rights reserved.
Zhao et al.
Figure 1. RNA substrate and endonuclease cleavage activity of nuclear extracts from Arabidopsis
culture cells. A, Examination of nuclear extracts for chloroplast contamination. Left section,
Coomassie-Blue-stained SDS-polyacrylamide
gel loaded with equal amounts of protein from
whole-cell extracts (WCE-CI), and nuclear extracts (Nu-CI) from culture cells overexpressing
AtCPSF73-I (or CI). Right section, Proteins from a
replicate gel were transferred to a membrane and
probed for the chloroplast stromal Rubisco large
subunit (RbcL, approximately 45 kD) using a
specific antibody (Agrisera). Protein size markers
(kD) are indicated on the right. B, A schematic
representation of the CaMV STS pre-mRNA substrate used in this study. The 3#-UTR region
includes the FUE, the NUE, and the CE (including
cleavage sites [CS], indicated by arrows), with the
size of cleavage products indicated (not drawn to
scale). The sizes of nucleotide deletions on the
FUE and NUE for mutational studies are also
indicated under these elements. CDS, Coding
sequence. C, Detection of cleavage activity with
uniformly labeled STS and reverse STS, with or
without Nu-CI. D, Cleavage of 5#-end-labeled
STS using Nu-CI. The arrow indicates the expected 182-nt cleavage product. E, Cleavage of
3#-end-labeled STS by Nu-CI. The arrow indicates
the expected 20-nt 3#-end product. F, Cleavage of
uniformly labeled STS pre-mRNA by nuclear
protein extracts from wild-type cells (Nu-WT).
The arrow indicates the expected 182-nt 5#-end
product. In D to F, the control in each section
refers to reactions incubated with buffer in place
of the nuclear extract. In C to F, substrates and
cleavage sites (marked by black triangles), and the
way the RNA substrates were labeled (either
uniform or end labeled; marked by stars) are
indicated at the left-hand side; RNA size standards (nt) are indicated on the right-hand side.
the overexpressed materials helped us in the establishment of the assay system, overexpression of the
plant polyadenylation factors was not necessary for
this activity. So some of the following experiments
were done using nuclear protein extracts from
AtCPSF73-I overexpressing cells (called Nu-CI). All
important key experiments were repeated using wildtype nuclear extracts (called Nu-WT), some of which
are shown as indicated.
The in Vitro Cleavage System Is cis-Element Dependent
To demonstrate that the accurate cleavage can also
happen to other transcripts than STS, we generated a
few transcripts with downstream extensions on the 3#UTR of the native CaMV STS, but without changing
the poly(A) signals. The results showed that the cleavage sites were maintained, despite that the pre-RNA
molecules extended to different lengths at their 3#
ends (Fig. 2A). This result indicates that the observed
cleavage depends on cis-elements embedded in the
sequence of the substrate upstream of the cleavage
site. The accuracy of cleavage is not affected by sequence downstream of the cleavage site. These results
agree with Rothnie and Mogen’s in vivo observation,
which showed both FUE and NUE but not other
downstream elements contribute to pre-mRNA cleavage site selection (Mogen et al., 1990; Rothnie et al.,
1994). To dissect the potential cis-elements [poly(A)
signals] located in the upstream of the pre-mRNA,
different mutations of the STS pre-mRNA were employed. These mutants were designed based on previous studies on published in vivo studies where a
disruption at either the FUE or the NUE region caused
utilization of cryptic polyadenylation sites (Mogen
et al., 1990; Rothnie et al., 1994). Similar observations
were made in our in vitro experiments where a deletion of FUE (22 nts) resulted in cleavage at an alterna-
1548
Plant Physiol. Vol. 157, 2011
Downloaded from on June 18, 2017 - Published by www.plantphysiol.org
Copyright © 2011 American Society of Plant Biologists. All rights reserved.
Pre-mRNA Cleavage during 3#-End Formation
viral one. A representative 3#-UTR of a Rubisco small
subunit gene (At5g38420) was chosen due to its high
expression level and abundance of available ESTs.
When this pre-mRNA was processed in vitro using our
cleavage assay prepared from Nu-CI, two bands (Fig.
2D) representing two groups of poly(A) sites were
detected. Sequencing results showed these two bands
corresponded to two authentic poly(A) sites supported
by ESTs that have poly(A) tails attached to them.
These EST data are from the 8K poly(A) site dataset as
described previously by Loke and coworkers (2005),
the National Center for Biotechnology Information
Arabidopsis EST collection (http://www.ncbi.nlm.
nih.gov), and the Arabidopsis Information Resources
databases (www.arabidopsis.org). Specifically, the two
poly(A) site groups were found at the 3#-UTR of
At5g38420 with TAIR9 genome coordinates 15381033
to 15381036 (for band 171–174 in Fig. 2D), and
15380980 to 15380988 (for the band marked 219–227
in Fig. 2D). Similar results were also obtained with
nuclear extract prepared from wild-type Arabidopsis
cell cultures (data not shown). Taken together, the
cleavage activity we found is an authentic mRNA 3#end processing component functioning on both a viral
RNA in a poly(A) signal-dependent manner and native plant pre-mRNA transcripts.
Figure 2. Properties of the cleavage activity. RNA substrates and
cleavage sites are indicated at the left-hand side of gels. A, Cleavage
of STS and its 3#-end extension variants. The total lengths are the STS
plus the indicated numbers of nts on the top of the lanes. Note that all
the STSs were cleaved at the same site, leaving the same 182-nt 5#-end
products as indicated by arrowheads. Reactions were incubated with
nuclear extracts (Nu-CI; +), and without nuclear extracts added (2). B
and C, Cleavage reactions require both FUE and NUE cis-elements.
Internal deletions of 22 nts of the FUE element from STS (B), or 6 nts of
the NUE element from STS58 (C; 58-nt 3#-end extension of STS) abolish
formation of site-specific cleavage products. The locations of expected
cleavage products are marked by arrows. Control reactions were
incubations without nuclear extract addition. D, Specific cleavage
products of the native Arabidopsis Rubisco small subunit (rbcS) transcript (At5g38420) 3#-UTR by Nu-CI. Cleavage products are indicated
by white triangles (sizes labeled by asterisk [*]). All substrate RNAs
were uniformly labeled. RNA size markers are in nts.
tive site upstream of the original site (Fig. 2B). Deletion
of NUE (6 nts) disrupted the cleavage at the correct
sites, but resulted in the use of nonspecific sites downstream of the original cleavage sites (Fig. 2C). Such
results can be typically found in genetic studies where
the utilities of alternative sites were reflected by read
through the length of detection probes or cryptic sites
(Mogen et al., 1990; Rothnie et al., 1994). These results
indicate that accurate cleavage requires correct poly
(A) signals, as it was revealed by previous studies.
The in Vitro Cleavage System Functions on
Native Transcripts
Furthermore, we tested if the cleavage activity also
works on endogenous Arabidopsis genes, besides the
Optimization of the Cleavage Reaction and
Cleavage Kinetics
To optimize the conditions for the cleavage assay,
different temperatures, pH gradients, and reaction
time were tested. The highest cleavage efficiency was
observed at 30°C, where more than 95% of pre-mRNA
was cleaved (Fig. 3A). The most efficient cleavage was
found at pH 8.5 (Fig. 3B), but it remained highly active
from pH 8.0 to pH 10.0. When a time course of the
cleavage reaction was carried out, we found that a
significant proportion of cleavage products was detected within 90 min and continued to accumulate
until 180 min (Fig. 3C). Thus, the overall reaction
condition can be set at 30°C, pH 8.0, for 120 min.
It has been shown that divalent metal ions, such as
Mg2+ and Mn2+, may affect cleavage and polyadenylation in both mammalian and yeast systems (Moore
and Sharp, 1985; Wahle, 1991; Ryan et al., 2004). To test
the efficacy of these ions in our cleavage system,
uniformly labeled STS was incubated with nuclear
extracts from AtCPSF73-I overexpressing cells under
the optimal reaction conditions described above, except with varying concentrations of Mg2+or Mn2+ as
indicated. The cleavage efficiencies were then compared. As shown in Figure 3D, cleavage efficiency
increased with increasing Mg2+ concentration, until
reaching the highest level at 6 mM. The cleavage efficiency remained high through the range of 3.5 to
6 mM Mg2+ (Fig. 3D). Thus, 3.5 mM was used as
the standard condition. In contrast to Mg2+, replacing
Mg2+ with Mn2+ (ranging from 0 to 20 mM) abolished
cleavage activity (Fig. 3E), although it slightly promoted
Plant Physiol. Vol. 157, 2011
1549
Downloaded from on June 18, 2017 - Published by www.plantphysiol.org
Copyright © 2011 American Society of Plant Biologists. All rights reserved.
Zhao et al.
Figure 3. Optimization of the cleavage reaction and cleavage kinetics. The reaction conditions of temperature (A), pH (B), time
(C), and ion concentrations (D and E) were tested as indicated. Nuclear extract prepared from wild-type Arabidopsis cell cultures
were used for the temperature, time course, and Mg2+ optimization, while the remaining determinations were performed with
extracts prepared from cultures overexpressing the AtCPSF73-I gene. F, The requirement of ATP for the cleavage reaction. The
bands corresponding to cleaved products and uncleaved substrates were measured by ImageQuant (GE Healthcare Inc.).
Cleavage ratio (CR) is calculated as the ratio between cleaved and uncleaved STS and is presented beneath the gels. G, The
cleavage kinetics were calculated from a Lineweaver-Burk plot of the cleavage velocity determined at various template
(substrate) concentrations. All the RNA substrates used in these assays were uniformly labeled.
general RNA degradation from 0.32 to 2.5 mM. These
results largely agree with those found in the in vitro
assays of mammalian and yeast systems (Manley,
1983; Moore and Sharp, 1985; Butler et al., 1990),
where Mg2+ is important for in vitro assays. ATP has
an inhibitory effect on the cleavage activity starting
from 1 mM, and total inhibition was reached at 2 mM,
as shown in Figure 3F.
After optimizing the cleavage assay conditions, the
reaction kinetics was studied. A series dilution of STS
substrate was assayed using an excessive amount of
the wild-type nuclear proteins under the standard
cleavage assay conditions. The reaction was stopped
after 30 min, the products were then resolved by a
sequencing gel, and the ratio of cleaved products to
uncleaved substrates were measured. Based on the
amounts of RNA added, the concentrations of RNA
substrate as well as the cleavage velocity were calculated. The Lineweaver-Burk plot shown in Figure 3G
was drawn using the reverse of substrate concentration (1/[S]) versus the reverse of respective cleavage
velocity (1/V). This exercise generated a Vmax of 8.4
nmol/min and Km of about 33.6 nM for STS pre-mRNA
(202 nt).
Reconstitution of Cleavage Reaction Coupled with
Polyadenylation by Supplementation with yPAP
Having successfully reconstituted cleavage activity,
we extended our search for polyadenylation activity.
However, after many attempts, no detectable poly(A)
tail was found coupled with the cleavage activity. As
in mammal and yeast mRNA 3#-end formation, the
cleavage step is fully committed and irreversible;
polyadenylation, on the other hand, requires additional protein factors, such as PAP and the poly(A)
binding protein-II (Zhao et al., 1999). Since our extracts
were made from the nucleus where polyadenylation
occurs, it is likely that these proteins were there, but
their activities were masked or inhibited. To circumvent this and reconstitute a fully functional 3#-end
processing assay, we tested if yPAP could supplement
the cleavage assay (Nu-CI) by adding poly(A) tails. To
our surprise, the addition of yPAP extended the substrates for about 150-nts long in the presence of ATP
(Fig. 4A, lane 2). When cordycepin (3#-dATP), an RNA
chain elongation inhibitor, was added to the system at
the same time yPAP was added, the extension of the
newly cleaved product was stopped (lane 5, Fig. 4A),
indicating that the observed extension is a product of
1550
Plant Physiol. Vol. 157, 2011
Downloaded from on June 18, 2017 - Published by www.plantphysiol.org
Copyright © 2011 American Society of Plant Biologists. All rights reserved.
Pre-mRNA Cleavage during 3#-End Formation
Figure 4. Reconstitution of cleavage reaction coupled with polyadenylation by supplementation with yPAP. A, Polyadenylation
of cleaved STS by yPAP. Lane 1: Control reaction where buffer in place of Nu-CI or yPAP was added to template pre-mRNA; lane
2: incubation with both Nu-CI and yPAP for 2 h; lane 3: 2-h incubation with Nu-CI only; lane 4: 1-h incubation with yPAP after
2-h cleavage reaction by Nu-CI; lane 5: 1-h incubation with cordycepin (or 3#-dATP) and yPAP after 2-h cleavage reaction by
Nu-CI; lane 6: 1-h incubation with yPAP without Nu-CI. B, A time course of cleavage and polyadenylation product formation
when both Nu-CI and yPAP were added simultaneously to initiate the reaction. Pre-mRNA (uncleaved STS) and cleavage
products are indicated by arrowheads. Polyadenylated RNAs are indicated by bracket. Incubation time (min) is noted on the top
of each lane, and RNA size markers in nts are labeled on the right-hand side of each section.
yPAP [poly(A) tract]. To investigate whether the poly
(A) addition was on the cleaved products, or the intact
pre-mRNA, the polyadenylation products were cloned
and sequenced after oligo(dT)-mediated reverse transcription-PCR. The results showed that, indeed, most
of the poly(A) tails were added to the newly cleaved
ends. Out of the 20 colonies sequenced, only four were
from polyadenylation product of not cleaved substrate, while the remaining 16 were all from correctly
cleaved substrates. These data indicated that 80% of
the polyadenylated RNA were from substrates that
were correctly cleaved.
PAP alone, however, could not add poly(A) to the
pre-mRNA without the help of nuclear proteins (Fig.
4A, lane 6) under our assay conditions. This suggests
that the specific addition of poly(A) is a result of the
cooperation between the plant cleavage apparatus and
yeast polyadenylation factors. It should be noted that
similar results were also obtained by cleavage system
using nuclear extract from wild-type cell culture (data
not shown). The complementation further supports
the idea that the cleavage activity described here is an
authentic pre-mRNA cleavage since another polyadenylation factor, yPAP, recognizes and cooperates with
it. In yeast, only the correct protein-protein interaction
can lead to such functional cooperation (Meinke et al.,
2008). Our study may indicate that a similar proteinprotein cooperation is conserved in plant and yeast
polyadenylation machinery. Such a cross-kingdom
cooperative reconstitution of polyadenylation factors
between yeast and plants opens up the possibility of
complementation of other poly(A) factors, some of
which were found to be impossible between mammal
and yeast (Jenny et al., 1994, 1996).
The cooperative working model is further supported by the observation that the interaction of plant
cleavage machinery with yPAP is required for efficient
poly(A) tail addition. When yPAP was added after the
cleavage step had already completed, poly(A) tails
were shorter with uneven sizes, and many cleavage
products remained unadenylated (compare Fig. 4A
lanes 2 and 4). Such a progressive reaction by yPAP
was also seen by a time-course study, where Nu-CI
and yPAP were added at the same time and showed a
uniform addition of poly(A) tails (Fig. 4B). It was also
found that the reaction was progressive at 5 to 10 nt/10
min, but faster at the beginning at a rate of 60 nt/10
min (Fig. 4B).
DISCUSSION
We report here the successful establishment of an in
vitro pre-mRNA cleavage assay system in Arabidopsis, in which yPAP can be add to the cleavage assay to
achieve specific poly(A) tail addition. This assay system will not only promote the biochemical study of
plant polyadenylation, but also shed light on the
establishment of other mRNA processing in vitro
reactions, e.g. splicing and transcription termination.
Although genetic studies can explain how genes function in many cases, biochemical investigation certainly
will lead to important functional information at the
molecular level. With an increasing number of cases
documenting gene expression regulation through alternative processing of pre-mRNA, such as alternative
splicing and alternative polyadenylation, in vitro stud-
Plant Physiol. Vol. 157, 2011
1551
Downloaded from on June 18, 2017 - Published by www.plantphysiol.org
Copyright © 2011 American Society of Plant Biologists. All rights reserved.
Zhao et al.
ies will be indispensable in understanding mechanisms at the biochemical level.
Compared to previous attempts to set up such an in
vitro assay system, our success may be attributed to
some distinct features such as the protein source, the
method of preparing nuclear extracts, reaction conditions, and choice of pre-mRNA substrates. First, the
nuclear proteins used in our study came from Arabidopsis suspension cells, which are actively dividing
and growing cells, in contrast to materials used previously, such as young seedlings or cauliflower heads,
both of which have only a small percentage of actively
dividing/growing cells. The gene expression machinery in actively growing and dividing cells is continually active, and their nuclear proteins may therefore be
more enriched with polyadenylation factors. Second,
the method we used in our nuclear protein extraction
may result in more abundant polyadenylation factors.
Because the extraction began with a considerable
amount of material (about 50-g cells), we can obtain
about 0.5 to 1 g/L nuclear proteins, which is a relatively high concentration for in vitro assays, even after
two centrifugations through Percoll gradients (to ensure purity). Meanwhile, we minimized the volume of
nuclear extracts by directly adding ammonium sulfate
to a final concentration of 0.5 M to break the nuclei.
In the process, we also eliminated the need of using
high-concentration ammonium sulfate to precipitate
nuclear proteins, thus possibly help to preserve functional protein complexes that are required for the in
vitro activity. Third, the assay conditions we used were
carefully optimized to achieve maximum activity. In
particular, the use of crowding agents such as polyvinyl alcohol and glycerol in the cleavage buffer may
further increase the effective reaction concentration.
Finally, the pre-mRNA substrate we used, CaMV STS
pre-mRNA, is well characterized (Mogen et al., 1990;
Rothnie et al., 1994). This 200-nt substrate contains
both strong AAUAAA as a NUE, as well as the typical
FUE, with an extension of a newly discovered CE
(Loke et al., 2005) covering both sides of the cleavage
site. These cis-elements may help cleavage site recognition and processing. Taken together, incremental
modifications of all of these factors may contribute to
a successful reconstitution of the cleavage assay.
Our in vitro cleavage assay system is quite similar to
assays used for mammals and yeast, including assay
conditions, a small volume reaction of about 10 mL,
reaction over a period of 2 h at 30°C, and similar
concentrations of Mg2+, K+, and Na+ ions. However,
we did not include any EDTA in the cleavage buffer
(not more than 0.10 mM carried over from the nuclear
suspension buffer), and used much less radiolabeled
RNA (only about 2,000 cpm or 2 pmol/reaction).
Capping of the substrate RNA is not required for the
cleavage activity (data not shown). While the poly(A)
tail addition can be achieved by adding yPAP, the
endogenous polyadenylation activity was not found in
our assay. Possible reasons for this are that Arabidopsis PAP was not active or was inhibited by some
factors, such as putative polyadenylation factor-B,
which has been shown to inhibit nonspecific PAP
activity (Forbes, 2004). It may be also due to the lack of
correct interactions between PAP and other different
polyadenylation factors, or the phosphorylation state
of PAP (Bond et al., 2000). Meanwhile, other not-yetidentified PAP inhibitors may exist to prevent in vivo
polyadenylation from happening. It is interesting to
note that, initially, PAP activity in yeast in vitro polyadenylation assay system was very low when setting
up; however, researchers ultimately determined how
to achieve higher PAP activity (Butler and Platt, 1988;
Butler et al., 1990).
With the supplement of yPAP, we managed to
complement the polyadenylation activity in our assay
system and provide further validation of our cleavage
reaction at the same time. Some early work with
polyadenylation factors indicated that such complementation with CPSF 73 and 100 and their counterparts in yeast (Jenny et al., 1994, 1996) was not
successful. However, functional complementation of
plant protein with yeast protein, or vice versa, is not
uncommon (e.g. among exosome proteins [Chekanova
et al., 2000]). It seems that whether a yeast polyadenylation factor can complement its plant counterpart is
determined by how much these factors have remained
structurally conserved during evolution. In the case of
PAP, yeast and plants are relatively conserved with
40% identity in the first 400 amino acid sequences
(Hunt et al., 2000). Our results provided direct evidence that at least some yeast polyadenylation-related
proteins can work cooperatively in the plant system. It
would be interesting to see if plant polyadenylation
factors can complement the functions of other yeast
counterparts.
Clearly, the establishment of such an in vitro cleavage assay system opens up a new avenue to study
plant polyadenylation. Future directions should
include the identification of the plant cleavage/
polyadenylation factors that are responsible for the
activities, and the roles of those potential factors revealed through bioinformatic and genetic studies, as
confirmed by protein-protein interactions (Xu et al.,
2006; Hunt et al., 2008; Xing et al., 2008a, 2008b; Zhao
et al., 2009). Several approaches can be employed to
identify the plant polyadenylation factors through this
type of biochemical assay. Moreover, with the broad
spectrum of tools we have generated, such as genetic
mutants, gene expression, and protein interaction profiles, the assay system will undoubtedly promote the
synergistic study of plant mRNA alternative polyadenylation, and posttranscriptional gene expression
regulation in general.
MATERIALS AND METHODS
Arabidopsis Suspension Cell Cultures
Arabidopsis (Arabidopsis thaliana) suspension cell culture (MM1 from
Landsberg erecta) was used in this research. Culture conditions, construction
1552
Plant Physiol. Vol. 157, 2011
Downloaded from on June 18, 2017 - Published by www.plantphysiol.org
Copyright © 2011 American Society of Plant Biologists. All rights reserved.
Pre-mRNA Cleavage during 3#-End Formation
of AtCPSF73-I-TAP fusion proteins, and cultured-cell transformation were
described elsewhere (Zhao et al., 2009).
Nuclear Protein Extract Preparation
Nucleus isolation from Arabidopsis suspension cultures was modified
from Escobar et al. (2001) and Folta and Kaufman (2006). About 50-g
suspension cells were ground in liquid nitrogen with acid-washed sand
(Sigma-Aldrich Inc.) and then resuspended in the extraction buffer, and
filtered through two layers of Miracloth twice. The filtrate was centrifuged at
25,000g for 5 min. The pellet was collected, resuspended in 30% Percoll (GE
Healthcare), then overlaid on the top of 30% and 80% Percoll double layers,
and centrifuged again at 2,000g for 30 min using a swing rotor. The middle
layer between the 30% and 80% Percoll layers was collected, washed twice
with buffer B (25 mM Tris-HCl pH 8.0; 10 mM MgCl2; 0.46 M Suc; 0.5 mM
phenylmethylsulfonyl fluoride (PMSF); 6 mM b-mercaptal ethanol; 0.5% Triton
X-100), and the nuclei were resuspended in 5 mL buffer C (25 mM Tris-HCl pH
8.0; 10 mM MgCl2; 0.46 M Suc; 0.5 mM PMSF; 6 mM b-mercaptal ethanol; 75%
Percoll). The nuclei were then collected by centrifuge at 5,000g for 30 min. The
concentrated nuclei were resuspended in buffer D (20 mM HEPES pH 8.0, 25%
glycerol, 0.4 mM EDTA, 0.5 mM PMSF, 1 mM dithiothreitol, and 100 mM NaCl)
and lysed by slowly adding a 2.0-M ammonium sulfate solution to a final
concentration of 0.5 M. Lysed nuclei were centrifuged at 13,000 rpm for 30 min,
and the supernatant (soluble nuclear protein extracts) was recovered and
stored at 280°C freezer for future use.
Labeling of RNA Substrates for in Vitro Assays
The 3#-UTR of CaMV 35S RNA STS clone (Mogen et al., 1990) was a gift
from Dr. Arthur Hunt (University of Kentucky). The target region (Fig. 1B)
was amplified with a primer fused with a T7 promoter sequence at the 5# end.
The gel-purified PCR product was used as a template for in vitro transcription
using the AmpliScrib T7 high yield transcription kit (Epicentre, Inc.), according to the manufacturer’s instructions. The same kit was used for the
transcriptions of cold and [a-32P] ATP-labeled (using one-tenth of cold ATP)
RNA. The cold STS RNA was 5#-end labeled by [g-32P] ATP using RNA kinase
(Epicentre, Inc.), and 3#-end labeled by [5#-32P] pCp using T4 RNA ligase
(New England Biolabs, Inc.). All other templates (including At5g38420 and
STS variants) were amplified by PCR, except the deletion mutants of FUE and
NUE, which were produced by overlap extension PCR (Warrens et al., 1997).
The labeled RNAs were purified using 7 M urea 6% polyacrylamide gel. The
corresponding gel bands were cut out, eluted, precipitated, and stored for
further use.
In Vitro Cleavage and Polyadenylation Assay
For an in vitro assay, in a 0.6-mL Eppendorf tube, 4.5 mL of cleavage buffer
(5 mM MgCl2; 41.67 mM phosphocreatine disodium; 1.67 mM ATP; 3.3%
glycerol; 0.8% polyvinyl alcohol; 3.3 mM HEPES pH 8.0), 0.5-mL RNaseOut (an
RNase inhibitor; Invitrogen Inc.), and 0.5-mL nuclear protein extracts (about
0.5 mg/mL), diluted pre-mRNA substrate (1,000–2,000 cpm/reaction; typically
about 2 pmol of RNA), and water were added to a total final volume of 7.5 mL.
The reaction was incubated at 30°C for 2 h. For polyadenylation assays, 1 mL
diluted (0.013, about 6 units) yPAP (US Biochemical Inc.) was added. When
the reaction was done, 7.5 mL 23 RNA gel-loading buffer was added, then all
15 mL was loaded to a 6% sequencing gel and run at 1,000 V for 1.5 to 2 h. The
gel was transferred to Whatman filter paper, dried, autoradiographed, and
scanned by a PhosphorImager scanner (Molecular Dynamics Inc.).
Sequencing of the Cleavage and
Polyadenylation Products
Cleavage products of cold RNA were gel purified, and a 3#-end RNA linker
(miRCat33; Integrated DNA Technology, Inc.) with a preactivated 5# end and a
blocked 3# end was ligated to the 3# end of the purified cleavage product, as
described by the manufacturer. The ligation product was then reverse transcribed using a primer against the 3#-end linker. The resulting products were
PCR amplified by primers; one matched the 3#-end linker, and the other
matched the 5# end of RNA. The amplified fragments were cloned into pTopo
TA cloning vector (Invitrogen) and sequenced to reveal the cleavage sites. The
polyadenylation product was reverse transcribed with oligo(dT)18VN primer
with adaptor sequence, PCR amplified using an STS-specific primer and a
primer annealed to the adaptor sequence, and then cloned using pTopo TA
vector and sequenced.
Sequence data from this article can be found in the GenBank/EMBL data
libraries under accession number AY140900.
ACKNOWLEDGMENTS
We thank Arthur Hunt for the CaMV STS construct, other Li lab members
for suggestions, David Martin and Tommy Li for language editorial assistance, and Miami University Instrumentation Laboratory and the Center for
Bioinformatics and Functional Genomics for technical support.
Received May 3, 2011; accepted September 8, 2011; published September 9,
2011.
LITERATURE CITED
Bond GL, Prives C, Manley JL (2000) Poly(A) polymerase phosphorylation is dependent on novel interactions with cyclins. Mol Cell Biol 20:
5310–5320
Buratowski S (2005) Connections between mRNA 3# end processing and
transcription termination. Curr Opin Cell Biol 17: 257–261
Butler JS, Platt T (1988) RNA processing generates the mature 3# end of
yeast CYC1 messenger RNA in vitro. Science 242: 1270–1274
Butler JS, Sadhale PP, Platt T (1990) RNA processing in vitro produces
mature 3# ends of a variety of Saccharomyces cerevisiae mRNAs. Mol
Cell Biol 10: 2599–2605
Chekanova JA, Shaw RJ, Wills MA, Belostotsky DA (2000) Poly(A) taildependent exonuclease AtRrp41p from Arabidopsis thaliana rescues 5.8
S rRNA processing and mRNA decay defects of the yeast ski6 mutant
and is found in an exosome-sized complex in plant and yeast cells. J Biol
Chem 275: 33158–33166
Danckwardt S, Hentze MW, Kulozik AE (2008) 3# end mRNA processing:
molecular mechanisms and implications for health and disease. EMBO J
27: 482–498
Dominski Z, Yang XC, Marzluff WF (2005) The polyadenylation factor
CPSF-73 is involved in histone-pre-mRNA processing. Cell 123: 37–48
Escobar C, Aristizéabal F, Navas A, Del Campo F, Fenoll C (2001) Isolation
of active DNA-binding nuclear proteins from tomato galls induced by
root-knot nematodes. Plant Mol Biol Rep 19: 375–376
Folta KM, Kaufman LS (2006) Isolation of Arabidopsis nuclei and measurement of gene transcription rates using nuclear run-on assays. Nat
Protoc 1: 3094–3100
Forbes KP (2004) Characterization of Plant Polyadenylation Transacting
factors—Factors That Modify Poly(A) Polymerase Activity. University
of Kentucky, Lexington, KY, p 147
Herr AJ, Molnàr A, Jones A, Baulcombe DC (2006) Defective RNA
processing enhances RNA silencing and influences flowering of Arabidopsis. Proc Natl Acad Sci USA 103: 14994–15001
Hunt AG, Meeks LR, Forbes KP, Das Gupta J, Mogen BD (2000) Nuclear
and chloroplast poly(A) polymerases from plants share a novel biochemical property. Biochem Biophys Res Commun 272: 174–181
Hunt AG, Xu R, Addepalli B, Rao S, Forbes KP, Meeks LR, Xing D, Mo M,
Zhao H, Bandyopadhyay A, et al (2008) Arabidopsis mRNA polyadenylation machinery: comprehensive analysis of protein-protein interactions and gene expression profiling. BMC Genomics 9: 220
Jenny A, Hauri H-P, Keller W (1994) Characterization of cleavage and
polyadenylation specificity factor and cloning of its 100-kilodalton
subunit. Mol Cell Biol 14: 8183–8190
Jenny A, Minvielle-Sebastia L, Preker PJ, Keller W (1996) Sequence
similarity between the 73-kilodalton protein of mammalian CPSF and a
subunit of yeast polyadenylation factor I. Science 274: 1514–1517
Li QQ, Hunt AG (1997) The polyadenylation of RNA in plants. Plant
Physiol 115: 321–325
Loke JC, Stahlberg EA, Strenski DG, Haas BJ, Wood PC, Li QQ (2005)
Compilation of mRNA polyadenylation signals in Arabidopsis revealed
a new signal element and potential secondary structures. Plant Physiol
138: 1457–1468
Plant Physiol. Vol. 157, 2011
1553
Downloaded from on June 18, 2017 - Published by www.plantphysiol.org
Copyright © 2011 American Society of Plant Biologists. All rights reserved.
Zhao et al.
Mandel CR, Bai Y, Tong L (2008) Protein factors in pre-mRNA 3#-end
processing. Cell Mol Life Sci 65: 1099–1122
Mandel CR, Kaneko S, Zhang H, Gebauer D, Vethantham V, Manley JL,
Tong L (2006) Polyadenylation factor CPSF-73 is the pre-mRNA 3#-endprocessing endonuclease. Nature 444: 953–956
Manley JL (1983) Accurate and specific polyadenylation of mRNA precursors in a soluble whole-cell lysate. Cell 33: 595–605
Mayr C, Bartel DP (2009) Widespread shortening of 3’UTRs by alternative
cleavage and polyadenylation activates oncogenes in cancer cells. Cell
138: 673–684
Meinke G, Ezeokonkwo C, Balbo P, Stafford W, Moore C, Bohm A (2008)
Structure of yeast poly(A) polymerase in complex with a peptide from
Fip1, an intrinsically disordered protein. Biochemistry 47: 6859–6869
Mogen BD, MacDonald MH, Graybosch R, Hunt AG (1990) Upstream
sequences other than AAUAAA are required for efficient messenger
RNA 3#-end formation in plants. Plant Cell 2: 1261–1272
Moore CL, Sharp PA (1985) Accurate cleavage and polyadenylation of
exogenous RNA substrate. Cell 41: 845–855
Moore MJ, Proudfoot NJ (2009) Pre-mRNA processing reaches back to
transcription and ahead to translation. Cell 136: 688–700
Rothnie HM (1996) Plant mRNA 3#-end formation. Plant Mol Biol 32: 43–61
Rothnie HM, Reid J, Hohn T (1994) The contribution of AAUAAA and the
upstream element UUUGUA to the efficiency of mRNA 3#-end formation in plants. EMBO J 13: 2200–2210
Ryan K, Calvo O, Manley JL (2004) Evidence that polyadenylation factor
CPSF-73 is the mRNA 3# processing endonuclease. RNA 10: 565–573
Shi Y, Di Giammartino DC, Taylor D, Sarkeshik A, Rice WJ, Yates JR III,
Frank J, Manley JL (2009) Molecular architecture of the human premRNA 3# processing complex. Mol Cell 33: 365–376
Simpson GG, Dijkwel PP, Quesada V, Henderson I, Dean C (2003) FY is
an RNA 3# end-processing factor that interacts with FCA to control the
Arabidopsis floral transition. Cell 113: 777–787
Tian B, Hu J, Zhang H, Lutz CS (2005) A large-scale analysis of mRNA
polyadenylation of human and mouse genes. Nucleic Acids Res 33:
201–212
Wahle E (1991) Purification and characterization of a mammalian polyadenylate polymerase involved in the 3# end processing of messenger
RNA precursors. J Biol Chem 266: 3131–3139
Warrens AN, Jones MD, Lechler RI (1997) Splicing by overlap extension by
PCR using asymmetric amplification: an improved technique for the
generation of hybrid proteins of immunological interest. Gene 186:
29–35
Wu X, Liu M, Downie B, Liang C, Ji G, Li QQ, Hunt AG (2011) Genomewide landscape of polyadenylation in Arabidopsis provides evidence
for extensive alternative polyadenylation. Proc Natl Acad Sci USA 108:
12533–12538
Xing D, Li QQ (2010) Alternative polyadenylation and gene expression
regulation in plants. Wiley Interdisciplinary Reviews: RNA 2: 445–458
Xing D, Zhao H, Li QQ (2008a) Arabidopsis CLP1-SIMILAR PROTEIN3, an
ortholog of human polyadenylation factor CLP1, functions in gametophyte, embryo, and postembryonic development. Plant Physiol 148:
2059–2069
Xing D, Zhao H, Xu R, Li QQ (2008b) Arabidopsis PCFS4, a homologue of
yeast polyadenylation factor Pcf11p, regulates FCA alternative processing and promotes flowering time. Plant J 54: 899–910
Xu R, Zhao H, Dinkins RD, Cheng X, Carberry G, Li QQ (2006) The 73 kD
subunit of the cleavage and polyadenylation specificity factor (CPSF)
complex affects reproductive development in Arabidopsis. Plant Mol
Biol 61: 799–815
Zhao H, Xing D, Li QQ (2009) Unique features of plant cleavage and
polyadenylation specificity factor revealed by proteomic studies. Plant
Physiol 151: 1546–1556
Zhao J, Hyman L, Moore C (1999) Formation of mRNA 3# ends in
eukaryotes: mechanism, regulation, and interrelationships with other
steps in mRNA synthesis. Microbiol Mol Biol Rev 63: 405–445
1554
Plant Physiol. Vol. 157, 2011
Downloaded from on June 18, 2017 - Published by www.plantphysiol.org
Copyright © 2011 American Society of Plant Biologists. All rights reserved.