THE JOURNAL OF BIOLOGICAL CHEMISTRY
© 2002 by The American Society for Biochemistry and Molecular Biology, Inc.
Vol. 277, No. 21, Issue of May 24, pp. 18665–18669, 2002
Printed in U.S.A.
The 3ⴕ-Untranslated Region of Chloroplast psbA mRNA Stabilizes
Binding of Regulatory Proteins to the Leader of the Message*
Received for publication, January 31, 2002
Published, JBC Papers in Press, March 19, 2002, DOI 10.1074/jbc.M201033200
Yael S. Katz‡ and Avihai Danon§
From the Department of Plant Sciences, Weizmann Institute of Science, Rehovot 76100, Israel
The 5ⴕ-leader and 3ⴕ-tail of chloroplast mRNAs have
been suggested to play a role in posttranscriptional regulation of expression of the message. The regulation is
thought to be mediated, at least in part, by regulatory
proteins that are encoded by the nuclear genome and
targeted to the chloroplast where they interact with
chloroplast mRNAs. Previous studies identified high affinity binding of the 5ⴕ-untranslated region (UTR) of the
chloroplast psbA mRNA by Chlamydomonas reinhardtii
proteins. Here we tested whether the 3ⴕ-UTR of psbA
mRNA alone or linked in cis with the 5ⴕ-UTR of the
mRNA affects the high affinity binding of the message in
vitro. We did not detect high affinity binding that is
unique to the 3ⴕ-UTR. However, we show that the cislinked 3ⴕ-UTR increases the stability of the 5ⴕ-UTR binding complex. This effect could provide a means for translational discrimination against mRNAs that are
incorrectly processed.
In initiation of translation, ribosomes bind to the 5⬘-untranslated region (UTR)1 of the mRNA and are consequently directed to the initiator codon of the open reading frame. The
5⬘-UTR of several chloroplast mRNAs in higher plants and in
the unicellular green alga Chlamydomonas reinhardtii contain
regions of inverted repeats and binding sites for nuclear encoded proteins. These interactions have been shown to be involved not only in regulation of translation but also in processing and stability of the message (1–10). It has been suggested
that ribosome association at the 5⬘-UTR may couple translation
with mRNA stability (11, 12). Most chloroplast mRNAs have an
AU-rich 3⬘-UTR with a terminal inverted repeat. This 3⬘-UTR
inverted repeat has been shown to play a role in the processing
and stabilization of the mRNA (13–17). Identification of interactions between the two termini of cytoplasmic transcripts (18,
19) and examples of 3⬘-UTR modulation of translation initiation in both eukaryotes (18, 20) and prokaryotes (21–23) raise
the possibility that interactions of the 5⬘- and 3⬘-UTR of chloroplast mRNAs may influence their expression. In support of
this notion, recent results have shown that correct processing
of the 3⬘-UTR can promote translation initiation and polysomal
association (24).
Expression of the D1 protein, encoded by the chloroplast
* This work was supported by Grant 651/00 from the Israel Science
Foundation and a grant from the Minerva Foundation. The costs of
publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
‡ Recipient of a Feinberg Graduate School Fellowship.
§ Holder of the Judith and Martin Freedman Career Developmental
Chair. To whom correspondence should be addressed. Tel.: 972-8-9342382; Fax: 972-8-934-4181; E-mail: [email protected].
1
The abbreviations used are: UTR, untranslated region; GMS, gel
mobility shift; UVCL, UV-cross-linking; r, recombinant.
This paper is available on line at http://www.jbc.org
psbA mRNA, is a prime example of the regulation of translation
involving the interaction of nuclear encoded factors and regions
of inverted repeats in the 5⬘-UTR of the transcript (3, 6, 25–27).
In C. reinhardtii, the 5⬘-UTR of psbA mRNA is the target for
the light-regulated binding of a complex of proteins of 38, 47,
55, and 60 kDa (3). These proteins interact with an inverted
repeat that is located upstream from a potential Shine-Dalgarno-like site in the 5⬘-UTR (3, 7) and possibly with a second
site after psbA mRNA undergoes 5⬘-processing (28). Assays of
D1 synthesis in C. reinhardtii mutants, in which the inverted
repeat has been partially deleted or mutated, indicate the
importance of the stem-loop fold for the light-responsive increase of D1 synthesis (7).
The central RNA-binding protein in the psbA 5⬘-binding complex is the 47-kDa protein (RB47). RB47 is a nuclear encoded
protein that shows high homology to poly(A)-binding proteins
(29). Mutants in which RB47 expression is low or lacking are
defective in D1 translation, implicating RB47 as a messagespecific translational factor (29 –31). The 60-kDa protein
(RB60) is a protein-disulfide isomerase-like protein also encoded by the nuclear genome (32). RB60 has been identified as
the regulatory redox-active protein of the psbA 5⬘-binding complex and is the likely candidate for perceiving the light-signal
and modulating the binding of the complex (33).
In addressing the potential modes of regulation of psbA expression, we examined whether the 3⬘-UTR affects protein
binding to psbA mRNA. We report here a comparison of binding
of C. reinhardtii proteins to RNA probes of 5⬘-UTR, 3⬘-UTR,
and cis-linked 5⬘-UTR-3⬘-UTR. Our analysis indicated that the
high affinity binding to psbA mRNA is primarily via its 5⬘-UTR,
whereas the presence of 3⬘-UTR in cis increases the affinity of
binding of the 5⬘-UTR-binding protein complex.
EXPERIMENTAL PROCEDURES
psbA RNA Constructs—To assess the influence of psbA 3⬘-UTR on
binding of the psbA 5⬘-binding complex, five types of RNA probes
containing the psbA UTRs were constructed as shown in Fig. 1. The 5⬘
transcript (5⬘-T) included the 5⬘-UTR and 36 bp downstream of the
initiator codon, and the 3⬘ transcript (3⬘-T) contained 30 bp preceding
the stop codon and the 3⬘-UTR. The fused 5⬘-UTR and 3⬘-UTR transcript (5⬘3⬘-T) was composed of the combined 5⬘-T and 3⬘-T, in “sense”
orientation. The 5⬘3⬘-NC-T transcript lacked the 30-bp coding region
proximal to the 3⬘-UTR and the 5⬘3⬘-LC-T transcript included a longer
segment of the 3⬘-coding region (Fig. 1). The psbA UTR constructs were
verified by sequence analysis.
Algae Growth Conditions and Purification of RNA-binding Proteins—C. reinhardtii 2137a cells were grown in TAP medium (34),
under continuous light at 25 °C, to a density of ⬃1 ⫻ 107 cells/ml. The
C. reinhardtii proteins were enriched for RNA-binding proteins by
liquid chromatography using a heparin-agarose column (3). Bound proteins were eluted from the column with a potassium acetate gradient of
0.1–1.6 M (3). The recombinant RB47 protein (rRB47) containing a His6
tag at its amino terminus was produced in Escherichia coli and purified
on a Ni-His trap column.
In Vitro Transcription, Gel Mobility Shift, and UV-cross-linking Assays—Each of the plasmids containing the psbA UTRs was digested
18665
18666
Regulatory Interactions of 5⬘-UTR and 3⬘-UTR
FIG. 1. Schematic representation of the three expression constructs used for in vitro transcription. The 92 bp of the psbA
5⬘-UTR (light gray box) and 36 bp from the coding region (white box)
were subcloned under the T7 promoter to form the 5⬘ transcript (5⬘-T).
The 52 bp of 3⬘-UTR (dark gray box) and 30 bp of upstream coding
region (hatch box) were subcloned to form the 3⬘ transcript (3⬘-T).
Fusion in sense orientation of the 5⬘-T and 3⬘-T created the 5⬘3⬘-T
transcript (5⬘3⬘-T). The 5⬘3⬘-NC-T transcript (5⬘3⬘ NC-T) lacked the
30-bp coding region proximal to the 3⬘-UTR, and the 5⬘3⬘-LC-T transcript (5⬘3⬘ LC-T) included a longer segment of the 3⬘-coding region.
SacI, XbaI, AccI, and BamHI denote restriction sites used for building
the constructs.
with either XbaI or BamHI restriction enzyme and used as a template
in run-off transcription reactions containing [32P]UTP (3). Purified radiolabeled RNA probes were used in gel mobility shift (GMS) assays to
compare the binding of the psbA 5⬘, 3⬘- or 5⬘-3⬘-UTR RNA to proteins
extracted from C. reinhardtii.
In GMS experiments, the proteins were first incubated, at room
temperature, with 15 units of Prime-RNase inhibitor in 10 ␮l of buffer
containing 20 mM Tris (pH 7.5), 100 mM potassium acetate, 0.2 mM
EDTA, 5 mM ␤-mercaptoethanol, 4.5 mM MgCl2, 1.5 mM dithiothreitol,
and 20% glycerol. Next, the proteins were incubated with the 32Plabeled RNA transcripts and, where specified, with nonlabeled RNA
competitors. Nonspecific RNA competitors that included 20 ␮g of tRNA
(Sigma) and 2 ␮g of total RNA isolated from a mutant cell line of C.
reinhardtii that lacks the psbA gene (FuD7) were included in both GMS
and UV-cross-linking assays. The products of GMS assays were separated on 5% native polyacrylamide gels (3), and formation of RNAprotein complexes was visualized by autoradiography or by
phosphorimaging.
In the UV-cross-linking assays, 32P-labeled RNA-protein complexes
were placed, on ice, 5 cm from the lamps of a Stratalinker device and
irradiated for 30 min. The RNA transcripts were digested with 50 ␮g of
RNase A (Sigma) for 30 min in the presence of 3 M urea, 3 mM EDTA at
55 °C. After adding SDS and ␤-mercaptoethanol to 2.5% final concentrations, proteins were fractionated by 12% SDS-PAGE (35), and radiolabeled proteins were detected by autoradiography.
RESULTS
C. reinhardtii Proteins Exhibit High Affinity Binding to cislinked UTRs—C. reinhardtii proteins partially purified by heparin affinity chromatography were incubated with 32P-labeled
5⬘-T transcript and subjected to GMS assay. Total RNA, extracted from the C. reinhardtii mutant cell line (FuD7) lacking
the psbA gene, was included in the binding reactions to compete out binding that was not psbA mRNA-specific. Protein
fractions, previously shown to contain the 47- and 60-kDa
proteins (3), demonstrated prominent binding activity to the
5⬘-UTR of psbA mRNA as expected (Fig. 2A). When the same
protein fractions were incubated with an equimolar concentration of 32P-labeled 3⬘-T transcript, only faint binding was observed (Fig. 2B). This suggests that the protein(s) binding to
the 3⬘-UTR are either much less abundant or have significantly
lower affinity than those that bind the 5⬘-UTR.
Interestingly, a notably high binding signal was detected
when concentration of an equimolar 32P-labeled 5⬘3⬘-T transcript was included in the GMS assay (Fig. 2C). This result
implies that the presence in cis of the 3⬘ terminus of the
message resulted in formation of a higher affinity protein-RNA
complex. To establish whether this phenomenon is caused by
FIG. 2. Gel mobility assays showing binding of C. reinhardtii
proteins to the three psbA transcripts. Protein fractions from heparin affinity chromatography were incubated with radiolabeled 5⬘-T (A),
3⬘-T (B), 5⬘3⬘-T (C), 5⬘3⬘-NC-T (D), or 5⬘3⬘-LC-T (E) RNAs, in the
presence of non-labeled tRNA and FuD7 RNA, and subjected to GMS
assay. Fractions 14 and 15 contain high affinity binding activity to 5⬘-T
and 5⬘3⬘-T. Fractions 10 –12 have RNase activity that causes degradation of the free RNA probe. Longer incubation periods of radiolabeled
RNA and Fractions 14 and 15 prior to GMS assays verified lack of
RNase activity in these fractions (data not shown). RNA denotes sample
containing the labeled RNA transcript without protein. Column fractions represent samples containing C. reinhardtii proteins of the indicated fractions of heparin column chromatography. Complex and Free
mark the mobility of the RNA-protein complex and the free RNA,
respectively.
the 3⬘-UTR or is unique to the 5⬘3⬘-T, two additional 5⬘3⬘
transcripts, 5⬘3⬘-NC-T (with no 3⬘-proximal coding region) and
5⬘3⬘-LC-T (with long 3⬘-proximal coding region), were tested
(Fig. 2, D and E). These two 5⬘3⬘ transcripts also caused stabilization of the complex in comparison with the 5⬘-T, assigning
a primary role to the 3⬘-UTR in this phenomenon. These results
suggested three main alternatives: 1) the presence in cis of
3⬘-UTR in the 5⬘3⬘-T transcript increases the affinity of the
5⬘-UTR binding activity; 2) the presence in cis of 5⬘-UTR increases the affinity of a novel binding to the 3⬘-UTR; 3) a novel
binding activity with high specificity to the 5⬘3⬘-T was detected
in the GMS assay. To differentiate between these alternatives
the following experiments were performed.
Binding of cis-linked UTRs Is via RB47—In a UV-crosslinking (UVCL) assay, a radioactive label is transferred from
the RNA to the protein(s) that comes into close contact with the
RNA molecule. The labeled protein is then identified by autoradiography of SDS-PAGE-fractionated proteins. We reasoned
that if the high binding activity observed with the 5⬘3⬘-T transcript is due to a novel binding protein, specific either to the
3⬘-UTR or to the cis-linked 5⬘3⬘-T transcript, then the UVCL
assay might detect it. UVCL assay with 32P-labeled 5⬘-T transcript, in the presence of tRNA and FuD7 total RNA as competitors, typically labeled the RB47 protein (Fig. 3A).
In agreement with the results of the GMS analysis, only a
faint signal corresponding to the 47-kDa protein was detected
with 32P-labeled 3⬘-T transcript in the UVCL assay (Fig. 3B).
The pattern of proteins labeled by cross-linking to the 5⬘- or
5⬘-3⬘-RNA was similar (Fig. 3, A and C). There was no notable
difference of protein bands to account for the influence of the
Regulatory Interactions of 5⬘-UTR and 3⬘-UTR
18667
FIG. 3. UVCL assay showing binding of C. reinhardtii proteins
to the three transcripts. Protein fractions from heparin affinity chromatography (column fractions) were incubated with radiolabeled 5⬘-T
(A), 3⬘-T (B), or 5⬘3⬘-T (C), in the presence of non-labeled tRNA and
FuD7 RNA and subjected to UVCL assay. The pattern of radiolabeled
proteins is similar for the 5⬘-T and 5⬘3⬘-T, with a primary label transfer
to a 47-kDa protein (RB47) from fractions 14 and 15.
3⬘-UTR on the 5⬘3⬘-T binding in GMS assays. However, this
does not exclude binding that may occur in an RNA region that
does not have enough uridine bases to cause strong labeling of
the protein(s) or protein-protein interactions that may affect
the RNA binding. These results suggest that the binding to the
5⬘3⬘-T is also by RB47 and its associated proteins and mainly
via the 5⬘-UTR. To confirm this, the specificity of the binding to
the transcripts was addressed further using affinity and competition assays.
Specificity and Affinity of cis-linked UTR Binding Activity—In competition assays, the specificity of RNA binding is
evaluated by including increasing amounts of non-labeled competitor RNAs in the binding reactions. A higher capacity of a
non-labeled RNA to compete with the radioactive RNA for the
binding activity ranks it higher in specificity. If the binding to
the transcript containing the cis-linked 5⬘- and 3⬘-UTRs is
principally via the 5⬘-UTR, then this RNA should compete
better than the 3⬘-UTR. To assess the specificity of binding to
the 5⬘3⬘-T RNA, GMS assays containing either the 32P-labeled
5⬘-T transcript or the 5⬘3⬘-T transcript were performed in the
presence of 5-, 25-, 50-, 100-, 200-, and 400-fold excess of nonlabeled 5⬘-T or 3⬘-T RNAs (Fig. 4A). The non-labeled 5⬘-UTR
transcript competed with the radiolabeled 5⬘-T or 5⬘3⬘-T RNAs.
In contrast, the 3⬘-UTR did not effectively compete with binding of either the 5⬘-UTR or 5⬘3⬘-T transcript. Similar results
were obtained when 5⬘-T or 3⬘-T competitors were included in
UVCL assays (Fig. 4B). These results show that specificity of
RB47 protein and its associated proteins is primarily to the
5⬘-UTR, rather than the 3⬘-UTR, and strongly argue for its
binding the 5⬘3⬘-T transcript mainly via the segment containing the 5⬘-UTR.
If RB47 and its associated proteins bind the psbA 5⬘-UTR
segment of 5⬘-T and 5⬘3⬘-T RNAs then what is the mechanism
for enhanced binding of 5⬘3⬘-T? One possibility is that the
cis-linked 3⬘-UTR increases the affinity of the native protein
complex to the RNA. To compare the affinity of 5⬘-T and 5⬘3⬘-T
binding, decreasing amounts of C. reinhardtii proteins were
mixed with equimolar concentrations of each type of RNA and
subjected to GMS assay. The results in Fig. 5 show that an
⬃3-fold lower concentration of proteins, containing an estimated concentration of 0.07 ␮M RB47, was required for 50%
FIG. 4. Specificity of binding of 5ⴕ-UTR and 5ⴕ3ⴕ-T. Comparison
of the competition by non-labeled 5⬘- or 3⬘-RNA transcripts was analyzed by GMS assay performed with radiolabeled 5⬘-T or 5⬘3⬘-T (A).
Minimal protein concentrations required for complex formation with
each of the 5⬘-T and 5⬘3⬘-T transcripts were used as determined in Fig.
5 (a 3-fold higher protein concentration in 5⬘-T assays than in 5⬘3⬘-T
assays). Non-labeled 5⬘-T competed effectively for the binding of C.
reinhardtii proteins, whereas 3⬘-T did not. Native complex and Free
mark the mobility of the native RNA-protein complex and the free RNA,
respectively. Competition assays performed in a UV-cross-linking assay
(B) also demonstrated that the 5⬘-T, but not the 3⬘-T, effectively competed for RB47 labeling.
native complex formation with the radiolabeled 5⬘3⬘-T RNA
than with the 5⬘-UTR alone. This shows that the RB47 and its
associated proteins bind the 5⬘3⬘-T with higher affinity than
the 5⬘-T alone. Therefore, although the 3⬘-UTR itself exhibits
only minor protein binding in GMS or UVCL assays, its fusion
to the 5⬘-UTR seems to stimulate the binding of the native
protein complex.
rRB47 Does Not Bind the cis-linked UTRs with Preferential
Affinity—RB47 was the primary binding protein identified in
assays containing 5⬘-T and 5⬘3⬘-T RNAs. To determine whether
binding of RB47 on its own is sufficient for the preferential
affinity to the cis-linked transcript, we assayed whether purified recombinant RB47 (rRB47) can mimic the binding of the
native complex containing RB47 in GMS assays. Binding of
rRB47 to 5⬘-T or 5⬘3⬘-T formed complexes that migrated faster
than the corresponding native complex (Fig. 6A), suggesting
that the complex formed by the intrinsic binding contained
additional proteins. Further, in contrast to the preferential
affinity of the native complex to the 5⬘3⬘-T, the binding of
rRB47 to both 5⬘- and 5⬘3⬘-RNA is of similar affinity (Fig. 6, B
and C). In fact, on its own rRB47 displayed a slight preference
to the 5⬘-T (Fig. 6C), further supporting the different nature of
binding of rRB47 and the native complex. Together, these
results suggest that an additional protein(s) associated with
RB47 is necessary for the preferential binding to the cis-linked
5⬘-3⬘-UTRs as compared with the 5⬘-UTR and that the presence
in cis of the 3⬘-UTR promotes the binding of the native complex
comprised of RB47 and its cognate proteins. We tested whether
the inclusion of recombinant RB60 (rRB60) with rRB47 mimics
the binding of the native complex. However, rRB60 and rRB47
18668
Regulatory Interactions of 5⬘-UTR and 3⬘-UTR
FIG. 5. Affinity of 5ⴕ-UTR and cis-linked UTR binding activities. The protein concentration that resulted in 50% binding of either
the 5⬘-UTR (A) or the cis-linked 5⬘-3⬘-UTRs (B) was determined by
incubating each radiolabeled RNA with decreasing amounts of C. reinhardtii proteins and analyzing the formation of protein-RNA complexes
by GMS assay. The combined results of three independent experiments
are presented in a graph (C). Approximately 3-fold higher concentrations of proteins were required for 50% binding of the native complex to
5⬘-T than to 5⬘3⬘-T. RNA denotes sample containing the labeled RNA
transcript without protein. ␮g protein represents samples containing C.
reinhardtii proteins of Fraction 14 of heparin column chromatography.
Native complex and Free mark the mobility of the native RNA-protein
complex and the free RNA, respectively.
together also did not result in preferential binding of the cislinked 5⬘3⬘-T RNA (data not shown) suggesting that the addition of RB60 was either not required or not sufficient.
DISCUSSION
Here we show that the presence in cis of the 3⬘-UTR of psbA
mRNA is required for high affinity binding of the native complex, containing the RB47 protein, to the 5⬘-UTR. Several
modes of action could explain the effect of the 3⬘-UTR on the
binding of the proteins to the 5⬘-UTR of psbA mRNA. A change
in the affinity of an RNA-protein complex may result from a
conformational shift in the RNA and/or from an altered binding
of the protein. Our results, so far, do not exclude either possibility. The 3⬘-UTR of psbA mRNA may interact directly with
the 5⬘-UTR of the message, thereby inducing it to form a
structural motif with high affinity to the native complex containing RB47. Such long range RNA-RNA interactions have
been found to affect translational efficiency in prokaryotes (22).
It should be noted that the MFOLD computer program for RNA
structure analysis (36) did not predict any such 5⬘-UTR-3⬘-UTR
interactions (data not shown). Alternatively, because low affinity binding of RB47 to the 3⬘-UTR was also found (Fig. 3), it is
possible that RB47 may bind both UTRs of psbA mRNA when
they are present in cis. This may not be entirely surprising as
the 3⬘-UTR is AU-rich, and RB47 shares high homology with
poly(A)-binding proteins (31). In this scheme, the simultaneous
binding of RB47 and its associated proteins to the 5⬘- and
3⬘-UTRs forms a high affinity complex (Fig. 2).
Previous characterization of the 5⬘-UTR binding activity by
GMS, UVCL, and psbA RNA-affinity chromatography assays
FIG. 6. rRB47 binding to 5ⴕ-T and 5ⴕ3ⴕ-T transcripts. Comparison
of the binding of recombinant RB47 (rRB47) to the 5⬘-T and 5⬘3⬘-T
RNAs with binding of C. reinhardtii heparin column chromatography
fraction 14 (Hep) by GMS assay (A). RNA denotes sample containing the
labeled RNA transcript without protein. The affinity of rRB47 binding
to the 5⬘ of 5⬘-3⬘-RNA was examined by determining the concentration
of rRB47 that results in 50% binding of either the 5⬘-T and 5⬘3⬘-T RNAs
(B). The combined results of three independent experiments are presented in a graph (C). Binding of rRB47 to 5⬘-T and 5⬘3⬘-T transcripts
displayed similar affinities. Native complex, rRB47 complex, and Free
mark the mobility of the native RNA-protein complex, the RNA-rRB47
complex, and the free RNA, respectively.
suggested that it is composed of four proteins (3). The UVCL
assay, which labels proteins that are in direct contact with the
RNA, showed that RB47 is the primary RNA-binding protein of
the complex (Fig. 3 and Ref. 3), and cloning of RB47 showed
that it contains RNA-binding domains (31). Extraction of the
proteins that were complexed with the 5⬘-UTR of psbA mRNA
in the GMS assay revealed that RB60 is also associated with
the RNA, potentially via protein-protein interactions with
RB47 (3). Based on these results, it was suggested that the
complex contains four proteins of which only binding of RB47
and RB60 was resistant to the conditions of the GMS (3).
A comparison of mobility shift of 5⬘-T and 5⬘3⬘-T RNAs by
either rRB47 or by the partially purified C. reinhardtii proteins
showed that the native proteins form complexes with slower
mobility (Fig. 6), further indicating that the native complex
contains additional protein(s), such as RB60, RB38, or RB55.
The higher affinity of the binding of the native complex to the
cis-linked 5⬘- and 3⬘-UTRs than to the 5⬘-T transcript indicates
that the presence in cis of the 3⬘-UTR promotes the binding of
the native complex comprised of RB47 and its cognate proteins.
Interestingly, the recombinant RB47 binds the 5⬘-T and the
5⬘3⬘-T at similar high affinities (compare Figs. 5 and 6), suggesting that the lower affinity of the native complex in the
absence of the 3⬘-UTR is a result of action of the proteins
associated with RB47.
The activity of the proteins that bind the 5⬘-UTR psbA
mRNA correlates with the translation rate of the D1 protein in
Regulatory Interactions of 5⬘-UTR and 3⬘-UTR
C. reinhardtii cells (3). Therefore, the increased affinity of
5⬘-UTR psbA mRNA-binding proteins conferred by the 3⬘-UTR
may suggest that the 3⬘-UTR influences D1 synthesis by stimulating binding of the native complex, comprised of RB47 and
its cognate proteins, to the 5⬘-UTR. Correct processing of the
3⬘-UTR was suggested to be required for high levels of translation initiation and polysomal association in C. reinhardtii
cells (24). Recent results from tobacco transformants in which
the influence of the psbA UTRs on translation of a reporter
gene were studied indicated that including the psbA 3⬘-UTR
resulted in a 3– 4-fold enhancement of translation (25). In another study, deletion of the inverted repeat of the 3⬘-UTR of
petD mRNA led to a reduction in petD expression beyond that
expected by the decrease in mRNA accumulation alone, indicating that the 3⬘-UTR may also contribute to efficient translation (37). These findings suggest that the 3⬘-UTR of chloroplast messages is required for optimal translation, in addition
to its role in determination of mRNA stability. Collectively,
these results suggest that the impaired binding of the native
complex, comprised of RB47 and its associated proteins, to a
psbA mRNA lacking an intact 3⬘-UTR may discriminate it from
efficient translation.
REFERENCES
1. Alexander, C., Faber, N., and Klaff, P. (1998) Nucleic Acids Res. 26, 2265–2272
2. Boudreau, E., Nickelsen, J., Lemaire, S. D., Ossenbuhl, F., and Rochaix, J. D.
(2000) EMBO J. 19, 3366 –3376
3. Danon, A., and Mayfield, S. P. (1991) EMBO J. 10, 3993– 4001
4. Drager, R. G., Higgs, D. C., Kindle, K. L., and Stern, D. B. (1999) Plant J. 19,
521–531
5. Fargo, D. C., Boynton, J. E., and Gillham, N. W. (2001) Plant Cell 13, 207–218
6. Hirose, T., and Sugiura, M. (1996) EMBO J. 15, 1687–1695
7. Mayfield, S. P., Cohen, A., Danon, A., and Yohn, C. B. (1994) J. Cell Biol. 127,
1537–1545
8. McCormac, D. J., Litz, H., Wang, J., Gollnick, P. D., and Berry, J. O. (2001)
J. Biol. Chem. 276, 3476 –3483
9. Ossenbuhl, F., and Nickelsen, J. (2000) Mol. Cell. Biol. 20, 8134 – 8142
10. Vaistij, F. E., Goldschmidt-Clermont, M., Wostrikoff, K., and Rochaix, J. D.
18669
(2000) Plant J. 21, 469 – 482
11. Bruick, R. K., and Mayfield, S. P. (1999) Nature 4, 190 –195
12. Mayfield, S. P., Yohn, C. B., Cohn, A., and Danon, A. (1995) Annu. Rev. Plant.
Physiol. Plant. Mol. Biol. 46, 147–166
13. Hayes, R., Kudla, J., Schuster, G., Gabay, L., Maliga, P., and Gruissem, W.
(1996) EMBO J. 15, 1132–1141
14. Lisitsky, I., Klaff, P., and Schuster, G. (1996) Proc. Natl. Acad. Sci. U. S. A. 93,
13398 –13403
15. Memon, A. R., Meng, B., and Mullet, J. E. (1996) Plant Mol. Biol. 30,
1195–1205
16. Monde, R. A., Schuster, G., and Stern, D. B. (2000) Biochimie (Paris) 82,
573–582
17. Schuster, G., and Gruissem, W. (1991) EMBO J. 10, 1493–14502
18. Gallie, D. R. (1998) Gene (Amst.) 216, 1–11
19. Mathews, M. B., Sonenberg, N., and Hershey, J. W. B. (1996) Translational
Control (Hershey, J. W. B., Mathews, M. B., and Sonenberg, N., eds) pp.
1–29, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
20. Gingras, A. C., Raught, B., and Sonenberg, N. (1999) Annu. Rev. Biochem. 68,
913–963
21. Franch, T., and Gerdes, K. (1996) Mol. Microbiol. 21, 1049 –1060
22. Lindahl, L., and Hinnebusch, A. (1992) Curr. Opin. Genet. Dev. 2, 720 –726
23. Voorma, H. O. (1996) in Translational Control (Hershey, J. W. B., Mathews, M.
B., and Sonenberg, N., eds) pp. 759 –777, Cold Spring Harbor Laboratory
Press, Cold Spring Harbor, NY
24. Rott, R., Levy, H., Drager, R. G., Stern, D. B., and Schuster, G. (1998) Mol.
Cell. Biol. 18, 4605– 4611
25. Eibl, C., Zou, Z., Beck, A., Kim, M., Mullet, J., and Koop, H.-U. (1999) Plant J.
19, 333–345
26. Girard-Bascou, J., Pierre, Y., and Drapier, D. (1992) Curr. Genet. 22, 47–52
27. Staub, J. M., and Maliga, P. (1994) Plant J. 6, 547–553
28. Bruick, R. K., and Mayfield, S. P. (1998) J. Cell Biol. 143, 1145–1153
29. Yohn, C. B., Cohen, A., Danon, A., and Mayfield, S. P. (1996) Mol. Cell. Biol.
16, 3560 –3566
30. Yohn, C. B., Cohen, A., Rosch, C., Kuchka, M. R., and Mayfield, S. P. (1998)
J. Cell Biol. 142, 435– 442
31. Yohn, C. B., Cohen, A., Danon, A., and Mayfield, S. P. (1998) Proc. Natl. Acad.
Sci. U. S. A. 95, 2238 –2243
32. Kim, J., and Mayfield, S. P. (1997) Science 278, 1954 –1957
33. Trebitsh, T., Levitan, A., Sofer, A., and Danon, A. (2000) Mol. Cell. Biol. 20,
1116 –1123
34. Gorman, D. S., and Levine, R. P. (1965) Proc. Natl. Acad. Sci. U. S. A. 54,
1665–1669
35. Laemmli, U. K. (1970) Nature 227, 680 – 685
36. Mathews, D. H., Sabina, J., Zuker, M., and Turner, D. H. (1999) J. Mol. Biol.
288, 911–940
37. Monde, R. A., Greene, J. C., and Stern, D. B. (2000) Plant Mol. Biol. 44,
529 –542