Plant Cell Advance Publication. Published on April 3, 2017, doi:10.1105/tpc.17.00020
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
RESEARCH ARTICLE
38
INTRODUCTION
39
The transcription start site (TSS) of genes transcribed by RNA polymerase II in
40
eukaryotes is thought to be primarily determined by the assembly of the pre-initiation
41
complex on recognizable sequences in the promoter (Danino et al. 2015; Thomas and
42
Chiang 2006; Vernimmen and Bickmore 2015). Because the term promoter in its
43
broadest sense includes all sequences that regulate expression, we will use the term
44
proximal promoter to refer to the sequences surrounding and immediately upstream of
45
the TSS. This process starts when the TATA-binding protein subunit of the general
Intron DNA Sequences Can Be More Important Than the Proximal
Promoter in Determining the Site of Transcript Initiation
Jenna E Gallegos1 & Alan B Rose1*
1
Department of Molecular and Cellular Biology, University of California, Davis, 1 Shields
Avenue, Davis, CA, 95616
*Corresponding Author: [email protected]
Short title: Introns affect transcript initiation
One-sentence summary: Introns boost mRNA accumulation without a proximal promoter and
influence the transcription start site, indicating that some introns play a previously unrecognized
role in controlling initiation.
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.plantcell.org) is: Alan B. Rose ([email protected]).
ABSTRACT
To more precisely define the positions from which certain intronic regulatory sequences
increase mRNA accumulation, the effect of a UBIQUITIN intron on gene expression was tested
from 6 different positions surrounding the transcription start site (TSS) of a reporter gene fusion
in Arabidopsis thaliana. The intron increased expression from all transcribed positions, but had
no effect when upstream of the 5'-most TSS. While this implies that the intron must be
transcribed to increase expression, the TSS changed when the intron was located in the 5'
UTR, suggesting that the intron affects transcription initiation. Remarkably, deleting 303 nt of the
promoter including all known TSSs and all but 18 nt of the 5’-UTR had virtually no effect on the
level of gene expression as long as an intron containing stimulatory sequences was included.
Instead, transcription was initiated in normally untranscribed sequences the same distance
upstream of the intron as when the promoter was intact. These results suggest that certain
intronic DNA sequences play unexpectedly large roles in directing transcription initiation and
constitute a previously unrecognized type of downstream regulatory element for genes
transcribed by RNA polymerase II.
©2017 American Society of Plant Biologists. All Rights Reserved
1
46
transcription factor TFIID binds TATA box sequences 30-40 nt upstream of the TSS.
47
Proximal promoters often contain additional recognizable sequences including the
48
initiator, which encompasses the TSS, the TFIIB recognition element, and the
49
downstream promoter element, which, along with the initiator, binds to TFIID (Butler and
50
Kadonaga 2002; Thomas and Chiang 2006). However, there are no universally
51
conserved promoter elements, and even the TATA box is found in a minority of genes in
52
plants (Morton et al. 2014), yeast (Basehoar et al. 2004), and humans (Yang et al.
53
2007). The common heterogeneity in start sites within a single gene further suggests
54
that there is considerable flexibility in the sequences that can support transcription
55
initiation (Morton et al. 2014). Many genes rely on additional promoter elements
56
(usually less than 1 Kb from the TSS) and distal enhancer elements to regulate gene
57
expression (Vernimmen and Bickmore 2015).
58
59
There are many examples in which a fully intact promoter with all its transcription factor
60
binding sites is inactive unless one or more of the gene’s endogenous introns is present
61
(Callis et al. 1987; Emami et al. 2013; Palmiter et al. 1991). In other cases, the
62
expression of genes that are weakly active without an intron can be increased tenfold or
63
more by the addition of an intron (Callis et al. 1987; Clancy et al. 1994; Kempe et al.
64
2013; Morello et al. 2011; Rose 2002). Furthermore, introns can change the spatial
65
expression patterns of tissue-specific genes (Emami et al. 2013; Giani et al. 2009;
66
Jeong et al. 2006). These examples demonstrate that some introns significantly boost
67
expression, even in the absence of prior promoter activity.
68
69
In some instances, the mechanism through which an intron increases gene expression
70
is well understood. For example, introns can contain enhancers or alternative TSSs
71
(Deyholos and Sieburth 2000; Kim et al. 2006; Morello et al. 2002). In addition,
72
interactions between the splicing machinery and other factors involved in mRNA
73
synthesis and maturation, as well as the exon-junction complex proteins deposited on
74
the mRNA during splicing, assist in mRNA production, stability, export, and translation
75
(Dahan et al. 2011; Le Hir et al. 2001; Maniatis and Reed 2002; Wiegand et al. 2003).
76
2
77
Certain introns that stimulate gene expression exhibit properties that suggest they must
78
operate by a different and poorly understood mechanism. First, the varying abilities of
79
efficiently spliced introns to increase mRNA accumulation indicate that only certain
80
introns boost mRNA levels beyond the general effects caused by the splicing machinery
81
or exon junction complexes (Rose 2002). These introns are unlike enhancers because
82
they have no effect when inserted into the 3’ UTR (Callis et al. 1987; Jeong et al. 2006;
83
Rose 2002; Snowden et al. 1996) or upstream of the promoter (Callis et al. 1987; Jeon
84
et al. 2000). The increase in mRNA accumulation caused by specific introns only when
85
located in transcribed sequences near the TSS will be referred to here as intron-
86
mediated enhancement (IME) (Gallegos and Rose 2015).
87
88
The question of whether splicing is necessary for IME is difficult to address because
89
unspliced intron sequences that remain in the mRNA have numerous secondary effects,
90
mostly due to stop codons and short open reading frames that interfere with translation
91
and can cause mRNA decay. Tests of introns engineered to maintain the reading frame
92
when retained in the mRNA came to conflicting conclusions about the need for splicing
93
in IME (Akua et al. 2010; Clancy and Hannah 2002; Rose and Beliakoff 2000).
94
However, the ability of some modified introns to increase mRNA accumulation in the
95
absence of splicing suggests that the mechanism of IME is at least partially splicing
96
independent (reviewed in Gallegos and Rose 2015).
97
98
The properties of IME have been most thoroughly examined for the first intron of the
99
Arabidopsis (Arabidopsis thaliana) UBIQUITIN 10 (UBQ10) gene. The UBQ10 intron
100
increases mRNA accumulation over 12 fold, while an efficiently spliced intron from the
101
COLD-REGULATED 15a (COR15a) gene, which is nearly identical in length and GC
102
content, has a less than twofold effect on expression (Rose 2002; Rose et al. 2008).
103
Replacing segments of the COR15a intron with like-sized segments of the UBQ10
104
intron revealed that the sequences responsible for increasing mRNA accumulation are
105
redundant and dispersed throughout the UBQ10 intron (Rose et al. 2008), unlike the
106
discrete sequences to which transcription factors bind in promoters and enhancers.
107
Hybrid introns containing COR15a intron splice sites and an internal portion of the
3
108
UBQ10 intron in either orientation stimulated expression equally well (Rose et al. 2011),
109
suggesting that the effect on gene expression is caused by intron sequences in the
110
DNA, rather than the strand-specific RNA. Experiments varying the location of the
111
UBQ10 intron and the endogenous TRYPTOPHAN BIOSYNTHESIS 1 (TRP1) first
112
intron in TRP1:GUS coding sequences revealed that their ability to stimulate mRNA
113
accumulation declines when moved from roughly 250 nt to 550 nt downstream of the
114
major TSS, and is completely gone when 1100 nt or more from the major TSS (Rose
115
2004).
116
117
The observation that these introns must be near the TSS to exhibit IME inspired the
118
development of an algorithm known as the IMEter that assigns a score based on the
119
oligomer composition of a given intron compared to promoter-proximal introns genome-
120
wide (Rose et al. 2008). The utility of the IMEter in predicting the stimulating ability of
121
introns has been confirmed in Arabidopsis (Rose et al. 2008), soybeans (Zhang et al.
122
2015), and other angiosperms (Aguilar-Hernandez and Guzman 2013).
123
124
The intron sequences responsible for IME have proven difficult to identify because their
125
redundant and dispersed nature renders deletion analysis ineffective (Clancy and
126
Hannah 2002; Donath et al. 1995; Rose 2002; Rose et al. 2008). However, the strong
127
correlation between intron IMEter scores and their ability to increase expression allowed
128
a large number of potentially stimulating introns to be computationally searched for
129
over-represented sequences (Rose et al. 2008). A candidate motif, TTNGATYTG, was
130
shown to be sufficient for IME because it can convert the previously non-stimulating
131
COR15a intron into one that strongly boosts mRNA accumulation. Introns containing
132
either 6 or 11 copies of this motif, named COR15a6L and COR15a11L, increase mRNA
133
accumulation 15-fold and 24-fold respectively (Rose et al. 2016). This cannot be the
134
only sequence that mediates IME because an exact match to this motif occurs only
135
once in the UBQ10 intron, while stimulating sequences are distributed throughout this
136
intron (Rose et al. 2008).
137
4
138
While the ability of an intron to stimulate expression is known to vary with its location
139
within a gene, the exact positional requirements for IME have not been fully determined.
140
Furthermore, it is unknown if the stimulating ability of an intron continues to increase as
141
it gets closer to the TSS, or if introns have their maximum effect approximately 200 nt
142
downstream of the major TSS where genome-wide average intron IMEter scores peak
143
(Parra et al. 2011). Pinpointing the ideal location for a stimulating intron will help to
144
maximize gene expression in industrial and research applications. Additionally, the
145
differing effects of a stimulating intron at various locations could yield insight into the
146
mechanism of IME.
147
148
In this study, we completed the first gene-scale mapping of the effect of intron location
149
on IME by comparing the expression of TRP1:GUS fusions containing the UBQ10 intron
150
at different locations around the TSS. In the process, we discovered an unexpected role
151
for introns in determining the site of transcription initiation.
152
153
5
154
RESULTS
155
The UBQ10 intron has no effect on expression from upstream of the 5’-most TSS
156
To determine the 5’-limit from which an intron can stimulate expression by IME, the first
157
intron of UBQ10 was tested at six locations upstream of the normal second exon of a
158
TRP1:GUS reporter gene (designated positions 1-6 in Figure 1). Previously generated
159
transgenic TRP1:GUS plants with the first intron of UBQ10 at three additional locations
160
(Rose 2004), designated herein as positions 0, -1, and -2, were included in Figure 1 to
161
show the 3’ limits of intron position for IME. Positions 0, -1, and -2 were previously
162
designated “259”, “551”, and “1136” respectively according to their distance from the
163
most frequently used TSS. A different numbering system was used here because a
164
description of intron location relative to the TSS is complicated by the presence of
165
multiple TSSs in this gene. The main TSS is at -41 relative to the ATG, as determined
166
by primer extension and RNAse protection (Rose et al. 1992), RNAseq, and the 5’ ends
167
of ESTs and cDNAs in the TAIR database (https://www.arabidopsis.org/). There are
168
also a few minor TSSs, of which the one at -117 is the furthest from the start codon and
169
is the annotated TSS for this gene.
170
171
Introns inserted at position 3 should be in the 5’-UTR of all TRP1:GUS transcripts, while
172
position 4 is upstream of the major TSS but downstream of the 5’-most TSS. The intron
173
at position 5 is upstream of all known TSSs, and roughly midway between
174
the TRP1 start codon and the stop codon of the upstream gene. This gene (At5g17980)
175
is a 3.15 Kb intronless gene of which only 1.8 Kb from the 3’ end is present in
176
the TRP1 promoter fragment in the TRP1:GUS fusions. The intron at position 6 is in
177
coding sequences of this gene 198 nt from the stop codon.
178
179
To insert the intron, a PstI site was added to the 5’ end of the intron, and the last six
180
nucleotides of the intron were converted into a PstI site. The intron was then cloned as
181
a PstI fragment into a PstI site created by site-directed mutagenesis at the desired
182
location (Figure 1D) as described (Rose and Beliakoff 2000). Introns inserted as PstI
183
sites leave no extraneous nucleotides in the mRNA (Rose 2004), and are efficiently
184
spliced, presumably because the 5’ splice site is unchanged, and the 3’ PstI site
6
185
(CTGCAG) conforms to the major 3’ splice site consensus (NNNYAG) in dicots.
186
7
187
For intron positions in coding sequences (positions 1 and 2), there were no places
188
where PstI sites could be made without changing an amino acid, so sites were chosen
189
such that changes were unlikely to disrupt protein function. The PstI sites introduced at
190
positions 1 and 2 convert an isoleucine and a glycine respectively into alanines, all of
191
which are in the family of nonpolar amino acids. Further, the first two exons of the TRP1
192
gene encode the chloroplast transit peptide (Rose et al. 1992). Transit peptides are not
193
stringent in their sequence requirements and can usually tolerate moderate changes
194
without losing function (Inoue and Glaser 2014). Because transit peptides are cleaved
195
off as the protein is imported into the chloroplast, conservative changes here are
196
unlikely to affect the enzymatic activity of the mature GUS reporter.
197
198
To qualitatively assess the positional effect of the UBQ10 intron on expression of the
199
TRP1:GUS reporter, pooled T2 transgenic seedlings of unknown transgene copy
200
number were histochemically stained for GUS activity (Figure 1B). The UBQ10 intron
201
clearly stimulated expression from all four transcribed locations but not from either
202
position upstream of the TSS, where it may reduce expression.
203
204
To quantitatively compare expression of reporter genes containing the intron at all
205
transcribed positions, GUS enzyme activity and mRNA accumulation were measured in
206
single-copy transgenic Arabidopsis lines (Figure 1C and Supplemental Tables 1 and 2).
207
Expression levels between independent single-copy lines of each construct were
208
similar, indicating that, for this transgene, the site of integration had little effect. This lack
209
of importance of the site of integration has also been demonstrated for other TRP1:GUS
210
fusions (Rose 2004; Rose and Beliakoff 2000), and likely stems from the location of the
211
TRP1 promoter in the middle of the T-DNA, where it is isolated from flanking plant
212
sequences by at least 2 Kb on each side.
213
214
GUS enzyme assays and RNA gel blots both indicated that TRP1:GUS mRNA levels
215
were similar when the UBQ10 intron was located at position 0, 1, 2, 3, or 4 (Figure 1
216
and Supplemental Tables 1 and 2). The activity of the intron at position 4 was surprising
217
because this location is 43 nt upstream of the major TSS, meaning that most of the
8
218
transcripts were not predicted to contain the intron. Inserting 304 nt of sequence into the
219
proximal promoter was expected to reduce expression.
220
221
Intron placement alters TSS usage
222
To determine if any transcripts initiated within the intron, cDNA from pooled seedlings
223
was amplified by 5’-RACE, cloned, and sequenced. None of the sequenced products
224
contained any intron sequences, but they did reveal an effect on the site of
225
initiation. When the intron was located at positions 1 or 2 (within coding sequences of
226
the first exon), the 5’ ends of most 5’-RACE products, and presumably the TSS,
227
mapped to within 10 nt of the major TSS (Figure 2). However, when the intron was
228
located at position 3 (within the 5’-UTR), transcription began almost exclusively within 5
229
nt of the furthest upstream TSS. When the intron was located at position 4 (upstream of
230
the major TSS), transcription began at the 5’-most TSS or at locations further upstream
231
where initiation does not normally occur, but not at any site downstream of the intron
232
including the major TSS (Figure 2A). These results suggest that introns may affect the
233
site of transcription initiation and explain how the gene could still be highly expressed
234
when the intron was located upstream of the major TSS.
235
236
A substantial promoter deletion does not diminish IME
237
The observation that inserting the intron into the 5’-UTR changed the location at which
238
transcription predominantly began suggested that transcription does not always start a
239
fixed distance downstream of conserved sequences in the proximal promoter. To
240
determine the extent of the flexibility in sequences that can support initiation, 303 nt of
241
the TRP1 promoter were deleted from the TRP1:GUS fusion (from position 3 to 5 in
242
Figure 1). The deletion encompassed all previously recognized TSSs and all but 18 nt
243
of the 5’-UTR (Figure 1D). The deletion was created in reporter gene fusions containing
244
no intron, the non-stimulating COR15a intron, or one of three stimulating
245
introns: UBQ10, COR15a6L, or COR15a11L (derivatives of the COR15a intron
246
containing 6 or 11 copies of the TTNGATYTG motif). Remarkably, deleting the proximal
247
promoter did not obviously reduce gene expression, as determined by histochemical
248
staining for GUS activity in seedlings of unknown transgene copy number (Figure 3). To
9
249
quantify potential subtle differences in expression, mRNA levels were compared
250
in single-copy transgenic Arabidopsis lines. Deleting all known TSSs did not diminish
251
mRNA levels from TRP1:GUS fusions containing any of the three stimulating introns
10
252
(Figure 3B). However, deleting these promoter sequences reduced but did not eliminate
253
expression of the constructs containing the non-stimulating COR15a intron or no intron
11
254
(Figure 3B). The size of the mature mRNA was not significantly changed by the
255
promoter deletion regardless of which intron was present.
256
257
Stimulating introns activate alternative TSSs upstream of the promoter deletion
258
To determine where transcription was initiating in the deletion-containing constructs,
259
and to verify that transcription of the TRP1:GUS fusion was similar to that of the
260
endogenous TRP1 gene, cDNA from pooled seedlings was amplified by 5’-RACE,
261
cloned, and sequenced. For TRP1:GUS constructs containing the COR15a11L intron in
262
which the promoter was intact, the apparent TSSs mapped within the range of -114 to
263
+1 relative to the ATG (Figure 2B), consistent with the start sites of the endogenous
264
TRP1 gene.
265
266
When the promoter sequences were deleted, most of the apparent start sites from
267
constructs containing either the COR15a11L or the UBQ10 intron mapped upstream of
268
the deletion in normally untranscribed sequences (Figure 2B). Even though these
269
initiation sites were far upstream of the normal TSS, transcription was initiating a similar
270
distance upstream of the intron (223-311 nt) as it did when the promoter was intact. This
271
can be seen in the average length of the 5’-UTRs in the sequenced 5’-RACE clones,
272
which did not significantly differ between the promoter-deletion constructs containing
273
either the COR15a11L or UBQ10 intron, the intact promoter with the COR15a11L
274
intron, or the previously sequenced cDNAs and ESTs from the endogenous TRP1 gene
275
(Figure 2C).
276
277
To verify that deleting all known TSSs caused transcription to start in a new location,
278
RNA gel blots were probed with a 709 nt fragment spanning the PstI sites at positions 4
279
and 6 (Figure 3C). The probe hybridized much more strongly with mRNA from lines in
280
which the proximal promoter was deleted and a stimulating intron was present than with
281
mRNA from lines in which the promoter was intact regardless of whether or not an
282
intron was present. This finding confirms that transcripts derived from promoter deletion
283
constructs with stimulating introns contain upstream sequences that are not transcribed
284
when the proximal promoter is intact. The smaller band present in all lanes is of
12
285
unknown origin.
286
287
Expression requires at least some sequences from the intergenic region
288
To further test the limits of sequences that can support transcript initiation, a larger
289
deletion (spanning from position 3 to position 6 in Figure 1) was created in TRP1:GUS
290
fusions containing no intron or the UBQ10 intron. This deletion removed the entire
291
intergenic region from 18 nt upstream of the TRP1 ATG into coding sequences near the
292
3’-end of the upstream gene. Transgenic plants containing TRP1:GUS constructs with
293
this complete promoter deletion had undetectably low levels of GUS activity even when
294
a stimulating intron was included (Figure 3D). This result suggests that even though the
295
promoter sequences that can support initiation in response to a stimulating intron are
296
surprisingly flexible, there are some features of the TRP1 promoter, either sequences or
297
chromatin structure, that are absolutely required for expression.
298
299
13
300
DISCUSSION
301
The effect of intron position on gene expression
302
Including the results presented here, the effect of the UBQ10 intron on expression has
303
been measured from a total of 14 locations within the TRP1:GUS reporter gene. The
304
intron increased mRNA accumulation from only six positions near the start of the gene.
305
Of the six active locations, the effect of the intron on mRNA accumulation was greatest
306
at -18 from the start codon and least from +510. In practical applications where introns
307
are used to maximize gene expression, for both efficacy and ease it might be best to
308
insert the intron into the 5’-UTR near the start codon.
309
310
It was previously unclear whether introns could affect gene expression if they were
311
upstream of but near the TSS. In earlier studies in which an intron was placed upstream
312
of various promoters (Callis et al. 1987; Jeon et al. 2000), the distance between the
313
intron and the TSS was usually more than 550 nt, so the introns may have been too far
314
away to affect expression. While these experiments clearly demonstrated that introns
315
are unlike enhancer elements whose influence extends several kilobases, they did not
316
establish if introns must be transcribed to have an effect. Here we showed that the
317
UBQ10 intron clearly did not stimulate expression when located 204 nt upstream of the
318
5’-most TSS. This finding is consistent with the idea that introns must be transcribed to
319
affect expression, although this conclusion is complicated by the possibility that
320
stimulating introns might cause initiation upstream of themselves, as discussed below.
321
The inability of the intron to affect expression from untranscribed sequences does not
322
necessarily demonstrate a role for splicing, because the mechanism of IME could
323
involve direct interactions between the transcription machinery and the intronic DNA.
324
325
The effect of intron position on the location of the TSS
326
The finding that moving a stimulating intron into the 5’-UTR, or deleting all known TSSs,
327
causes transcription to begin in sequences that do not normally support initiation
328
suggests that start sites are not determined by proximal promoter sequences alone.
329
Further, there are no obviously conserved sequences between the various TSS’s
330
observed in this study (Figure 4). There must be additional conditions that can be
14
331
influenced by intron sequences to allow initiation at competent sites that are normally
332
inactive.
333
15
334
The transcription stimulated by UBQ10 intron sequences did not initiate a fixed distance
335
upstream of the intron, as relocating the intron in constructs with intact promoters did
336
not appreciably change the TSS unless the intron was inserted into the 5’-UTR (Figure
337
2). In yeast, splicing occurs during transcription very soon after the nascent RNA
338
emerges from RNA polymerase II (Carrillo Oesterreich et al. 2016); moving the intron
339
into the 5’-UTR might push the predominant site of initiation further upstream if there is
340
steric interference between the bulky spliceosome and the transcription machinery as it
341
assembles on the promoter for the next round of transcription.
342
343
The effect of proximal promoter deletions on IME
344
The sequences between positions 5 and 6 are capable of supporting initiation, while
345
those upstream of position 6 are not. The region between positions 5 and 6 may have
346
distinguishing features beyond specific sequence composition, such as a propensity to
347
form open chromatin. The DNase-sensitive region that includes the TRP1 promoter
348
does not extend as far upstream as position 6 (Figure 5). Further, the sequences
349
between positions 3 and 6 are noticeably more AT-rich than sequences upstream of
350
position 6 (Figure 5B), and histone associations are favored by GC-richness (Segal and
351
Widom 2009).
352
353
These results may help to explain why some promoters are completely dependent on
354
introns for expression. For some genes, the entire promoter may include regulatory
355
sequences in the first intron. A subset of these intron-dependent genes may even lack
356
proximal promoters in the traditional sense but rather rely on a cryptic downstream
357
promoter element within the intron that, unlike previously identified regulatory elements
358
for RNA polymerase II, specifically activates transcription a few hundred base pairs
359
upstream of itself. The ability of stimulating introns to override the tissue specificity of
360
promoters further suggests that the mechanism of IME does not depend on prior
361
promoter activity, and inherently leads to constitutive expression throughout the plant.
362
Intron-caused ubiquitous expression that is independent of normal promoter activity is
363
more consistent with a DNA-based than an RNA-based mechanism of IME.
364
16
365
If certain introns can drive expression in the absence of the normal TSSs, it is puzzling
366
that proximal promoter sequences have not been found more generally dispensable in
367
the large number of publications reporting promoter deletion analyses. One possible
17
368
explanation is that introns are rarely included in the genes used to assess promoter
369
activity. One exception is a study of the unc-54 gene of Caenorhabditis elegans in which
370
the expression of different versions of the gene was measured by their ability to rescue
371
an unc-54 null mutation (Okkema et al. 1993). While unc-54 genomic DNA rescued
372
efficiently, unc-54 cDNA under the control of the unc-54 promoter did not,
373
demonstrating a role for introns in controlling the expression of this gene. An intron-
374
containing but promoterless gene was able to rescue, and transcripts derived from this
375
construct initiated in the plasmid sequences newly fused upstream of the start codon.
376
This finding suggests that the first intron of unc-54 causes initiation upstream of itself,
377
although the possibility cannot be ruled out that some bacterial sequences can
378
fortuitously act as a promoter in C. elegans.
379
380
There is suggestive evidence from the ENCODE project that some introns might cause
381
initiation upstream of themselves in humans as well. Using unique EST ends as
382
genome-wide indicators of TSSs, initiation was observed not only at the annotated TSS
383
but also 250-300 nt upstream of the first intron in genes with long first exons
384
(Bieberstein et al. 2012). Additionally, genes with a naturally occurring intron very near
385
the TSS tend to be more actively transcribed, as demonstrated by association of these
386
genes with RNA Polymerase II and TFIID (Bieberstein et al. 2012). Thus, the apparent
387
ability of intron sequences to stimulate transcript initiation upstream of themselves and
388
for promoter proximal introns to increase gene expression may be conserved across
389
kingdoms.
390
391
Possible mechanisms of IME
392
The following speculation incorporates previous data from plants and other eukaryotes,
393
focusing largely on the UBQ10 intron. The observations that some introns increase
394
mRNA accumulation in the absence of splicing (Rose and Beliakoff 2000),that UBQ10
395
intron sequences increase expression to an equal degree in either orientation (Rose et
396
al. 2011), and that prior promoter activity is unnecessary for the UBQ10 intron to
397
activate expression (Emami et al. 2013) all point to a DNA-based mechanism of IME for
398
this intron. Genome-wide IMEter scores suggest that the expression of as many as 15%
18
399
of genes in Arabidopsis and other plants may be influenced by a similar mechanism
400
(Gallegos and Rose 2015). However, there are other known ways in which different
401
introns increase expression (Akua and Shaul 2013; Bianchi et al. 2013), and a single
402
intron could have multiple effects.
403
404
The main requirement for sequences to act as the promoter of intron-regulated genes
405
such as TRP1 might be a region of open chromatin that allows access of the
406
transcription machinery to the DNA. Within that region of open chromatin there may be
407
DNA sequences that favor initiation, but others can readily substitute when the preferred
408
sites are deleted. Intron sequences may also help to create the local chromatin state
409
necessary for initiation through the interaction between DNA sequences within the
410
intron and histone-modifying or nucleosome-positioning factors. Transcription from
411
many or most promoters may be inherently bidirectional (Bagchi and Iyer 2016), and
412
introns have been shown to decrease antisense transcription in yeast (Agarwal and
413
Ansari 2016). Introns also favor re-initiation by facilitating DNA looping that brings the 3’
414
end of the gene, and thus the transcription machinery, in proximity to the promoter
415
(Moabbi et al. 2012). In short, stimulating introns may boost mRNA production by
416
creating local chromatin conditions that favor initiation, decreasing the proportion of
417
RNA polymerase II molecules that transcribe away from the gene, and positive
418
feedback that amplifies the accumulation of mRNA through re-initiation.
419
420
The limitations to the positions from which the UBQ10 intron is capable of stimulating
421
mRNA accumulation could be caused by a combination of the relatively short distances
422
(1 Kb or less) over which the postulated intron-driven changes in chromatin structure
423
extend, and the location of potential translational start codons (Gallegos and Rose
424
2015). The intron at all locations might increase transcription initiation upstream of itself,
425
but transcripts that initiate outside of the 220 nt window in which the TRP1 start codon is
426
the first ATG are likely to be unstable (Figure 5C). This might also explain why there
427
was a slight decrease in expression when the intron was located at position 5 or 6 if
428
transcription was initiating preferentially upstream of the intron. The apparent lack of
429
change in TSS when the UBQ10 intron is moved through TRP1:GUS coding sequences
19
430
is consistent with the idea that only transcripts that start within the 220 nt window
431
upstream of the ATG accumulate.
432
433
In conclusion, the ability of some introns to influence the location of the TSS and to
434
stimulate expression of a gene lacking its proximal promoter indicates that certain intron
435
sequences constitute a previously unrecognized type of downstream regulatory element
436
for genes transcribed by RNA polymerase II. These sequences can affect expression
437
from more than 500 nt downstream of the TSS, suggesting they are unlike the
438
downstream transcription factor binding sites found in genes transcribed by RNA
439
polymerase III. The inability of the same introns to stimulate expression from upstream
440
or more than 1100 nt downstream of the TSS also differentiates these sequence
441
elements from enhancers. Further characterization of the sequences that support
442
initiation in response to an intron, and of the role of chromatin structure, should lead to
443
better understanding of the mechanism through which these sequences increase gene
444
expression.
445
446
MATERIALS AND METHODS
447
Cloning of reporter gene fusions
448
The starting template for all constructs was an intronless TRP1:GUS fusion containing
449
2.4 Kb of sequence stretching from the 3’ end of the gene (At5g17980) upstream
450
of TRP1 through the first 8 amino acids of the 3rd exon of TRP1 fused to the E. coli
451
uidA (GUS) gene (Rose and Last 1997). PstI sites were introduced at 6 locations using
452
PCR mutagenesis, and confirmed by sequencing (Rose 2004). To delete promoter
453
sequences, a promoter fragment whose 3’ end was the PstI site at position 5 (in the
454
intergenic region) was ligated to an exon fragment whose 5’ end was the PstI site at
455
position 3 in the 5’-UTR, thereby deleting the sequences between positions 5 and 3.
456
The same procedure was used to generate the larger promoter deletion, except that the
457
distal fragment had the PstI site at position 6 (in the upstream gene). All fusions were
458
then cloned into the binary vector pEND4K, transformed into Agrobacterium
459
tumefaciens by electroporation, and then introduced into Arabidopsis thaliana ecotype
20
460
Columbia (Col) by floral dip as described (Rose and Beliakoff 2000).
461
462
GUS expression assays
463
For qualitative comparisons of expression (as in Figure 1B), T2 seedlings from multiple
464
lines for each construct were histochemically stained for GUS activity. The plate of
465
seedlings in buffer (10 mM EDTA, 100 mM NaPO4 pH 7.0, 0.1% Triton X-100)
466
containing 0.5 mg/mL 5-bromo-4-chloro-3-indolyl ß-D-glucuronic acid (Calbiochem, La
467
Jolla, CA, USA) was incubated at 37° for 30 minutes to 2 hours, the plants were rinsed
468
in water, and chlorophyll was removed with ethanol. Quantitative measurements of GUS
469
enzyme activity in leaf extracts were performed as described (Rose, 2004).
470
471
Identification of single-copy transgenic lines
472
Single-copy transgenic lines were identified for quantitative measurements
473
of GUS enzyme activity and mRNA levels. For each construct, 100 seeds from each of
474
18-72 T2 lines were screened for kanamycin resistance (Rose 2004). Genomic DNA
475
was extracted from lines that exhibited a 3:1 segregation ratio (kanamycin
476
resistant:sensitive) indicative of a single locus of transgene insertion. Gel blots of the
477
DNA digested with BamHI or PstI were probed with the GUS gene. All lines for which
478
both enzymes generated a single band were propagated to the T3 or T4 generation, and
479
homozygous plants were used for quantitative comparisons. Plants were grown in
480
Professional Growing Mix (Sun Gro Horticulture) at a density of 500 seeds per 150 cm2
481
pot in continuous light (~125 μmol/m2/s) under cool white fluorescent bulbs for 21 days
482
at 20°C. RNA was extracted from leaf tissue using the RNeasy Plant Mini kit (Qiagen)
483
following the manufacturer’s protocol. Expression levels were compared to the
484
analogous intronless control pAR281 (Rose and Beliakoff 2000). Other previously
485
generated constructs used in this study include TRP1:GUS fusions containing the
486
UBQ10, COR15a, COR15a6L, or COR51a11L intron at position 0 (Rose et al. 2008,
487
Rose et al. 2016), or the UBQ10 intron at position -1 or -2 (position 551 or 1136 in Rose
488
2004).
489
490
Mapping TSSs with 5’-RACE
21
491
The 5’ ends of TRP1:GUS mRNAs were mapped by 5’-RACE as described (Scotto-
492
Lavino et al. 2006). Briefly, RNA extracted from transgenic plants was reverse
493
transcribed (Reverse Transcription System; Promega, Madison, WI, USA) using
494
the GUS-specific primer OAR37 (5’-TAACGCGCTTTCCCACCAACG-3’). The resulting
495
cDNA was polyadenylated with terminal deoxynucleotidyltransferase and used as the
496
template for PCR with the primers OAR37, QT (5’-
497
CCAGTGAGCAGAGTGACGAGGACTCGAGCTCAAGCTTTTTTTTTTTTTTTTT-3’) and
498
QO (5’-CCAGTGAGCAGAGTGACG-3’). A dilution of the first round PCR product was
499
used as the template for a second round of PCR with the nested primers QI (5’-
500
GAGGACTCGAGCTCAAGC-3’) and OAR29 (5’-GGTTGGGGTTTCTACAGGACG-3’).
501
The 2nd round PCR product was cleaned up with a PCR Purification Kit (Qiagen,
502
Hilden, Germany), digested with restriction enzymes XhoI and BamHI, and ligated into
503
the vector pBluescript KS+. The DNA from transformants was screened by digestion
504
with SacI. Clones with inserts long enough to contain the SacI site that begins 24 nt
505
downstream of the TRP1 start codon were sequenced.
506
507
Accession numbers
508
Sequence data from this article can be found in the Arabidopsis Genome Initiative or
509
EMBL/GenBank data libraries under the following accession numbers: COR15a
510
(At2g42540), TRP1 (At5g17990), and UBQ10 (At4g05320).
511
512
Supplemental Data
513
Supplemental Table 1. Detailed mRNA expression data.
514
Supplemental Table 2. Detailed enzyme expression data.
515
516
517
Acknowledgements
518
This research was supported by funding from the UC Davis PI Bridge Program. Jenna E
519
Gallegos was supported by an NSF Graduate Research Fellows Program Grant
520
1148897.
22
521
522
Author Contributions
523
524
J.E.G. and A.B.R. jointly conceived the project, performed the experiments, and wrote
the manuscript.
525
526
23
Parsed Citations
Agarwal N, Ansari A (2016) Enhancement of Transcription by a Splicing-Competent Intron Is Dependent on Promoter Directionality
PLoS Genet. 12:e1006047
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Aguilar-Hernandez V, Guzman P (2013) Spliceosomal introns in the 5' untranslated region of plant BTL RING-H2 ubiquitin ligases
are evolutionary conserved and required for gene expression BMC Plant Biol. 13:179
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Akua T, Berezin I, Shaul O (2010) The leader intron of AtMHX can elicit, in the absence of splicing, low-level intron-mediated
enhancement that depends on the internal intron sequence BMC Plant Biol. 10:93
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Akua T, Shaul O (2013) The Arabidopsis thaliana MHX gene includes an intronic element that boosts translation when localized in a
5' UTR intron J. Exp. Bot. 64:4255-4270
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Bagchi DN, Iyer VR (2016) The Determinants of Directionality in Transcriptional Initiation Trends Genet. 32:322-333
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Basehoar AD, Zanton SJ, Pugh BF (2004) Identification and distinct regulation of yeast TATA box-containing genes Cell 116:699-709
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Bianchi M, Crinelli R, Giacomini E, Carloni E, Radici L, Magnani M (2013) Yin Yang 1 intronic binding sequences and splicing elicit
intron-mediated enhancement of ubiquitin C gene expression PLoS One 8:e65932
Bieberstein NI, Carrillo Oesterreich F, Straube K, Neugebauer KM (2012) First exon length controls active chromatin signatures
and transcription Cell Rep. 2:62-68
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Butler JE, Kadonaga JT (2002) The RNA polymerase II core promoter: a key component in the regulation of gene expression
Genes Dev. 16:2583-2592
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Callis J, Fromm M, Walbot V (1987) Introns increase gene expression in cultured maize cells. Genes Dev. 1:1183-1200
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Carrillo Oesterreich F, Herzel L, Straube K, Hujer K, Howard J, Neugebauer KM (2016) Splicing of Nascent RNA Coincides with
Intron Exit from RNA Polymerase II Cell 165:372-381
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Clancy M, Hannah LC (2002) Splicing of the Maize Sh1 First Intron Is Essential for Enhancement of Gene Expression, and a T-Rich
Motif Increases Expression without Affecting Splicing Plant Physiol. 130:918-929.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Clancy M, Vasil V, Hannah LC, Vasil IK (1994) Maize Shrunken-1 intron and exon regions increase gene expression in maize
protoplasts Plant Sci. 98:151-161
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Dahan O, Gingold H, Pilpel Y (2011) Regulatory mechanisms and networks couple the different phases of gene expression Trends
Genet. 27:316-322
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Danino YM, Even D, Ideses D, Juven-Gershon T (2015) The core promoter: At the heart of gene expression Biochim. Biophys. Acta
1849:1116-1131
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Deyholos MK, Sieburth LE (2000) Separable whorl-specific expression and negative regulation by enhancer elements within the
AGAMOUS second intron Plant Cell 12:1799-1810
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Donath M, Mendel R, Cerff R, Martin W (1995) Intron-dependent transient expression of the maize GapA1 gene Plant Mol. Biol.
28:667-676
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Emami S, Arumainayagam D, Korf I, Rose AB (2013) The effects of a stimulating intron on the expression of heterologous genes in
Arabidopsis thaliana Plant Biotechnol. J. 11:555-563
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Gallegos JE, Rose AB (2015) The enduring mystery of intron-mediated enhancement Plant Sci. 237:8-15
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Giani S, Altana A, Campanoni P, Morello L, Breviario D (2009) In trangenic rice, a- and ß-tubulin regulatory sequences control GUS
amount and distribution through intron mediated enhancement and intron dependent spatial expression Transgenic Res. 18:151162
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Inoue K, Glaser E (2014) Processing and degradation of chloroplast extension peptides. In: Theg SM, Wollman F-A (eds) Plastid
Biology. Springer Science, New York, pp 305-323
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Jeon JS, Lee S, Jung KH, Jun SH, Kim C, An G (2000) Tissue-preferential expression of a rice alpha-tubulin gene, OsTubA1,
mediated by the first intron Plant Physiol. 123:1005-1014
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Jeong YM, Mun JH, Lee I, Woo JC, Hong CB, Kim SG (2006) Distinct roles of the first introns on the expression of Arabidopsis
profilin gene family members Plant Physiol. 140:196-209
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Kempe K, Rubtsova M, Riewe D, Gils M (2013) The production of male-sterile wheat plants through split barnase expression is
promoted by the insertion of introns and flexible peptide linkers Transgenic Res. 22:1089-1105
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Kim MJ, Kim H, Shin JS, Chung CH, Ohlrogge JB, Suh MC (2006) Seed-specific expression of sesame microsomal oleic acid
desaturase is controlled by combinatorial properties between negative cis-regulatory elements in the SeFAD2 promoter and
enhancers in the 5'-UTR intron Mol. Genet. Genomics 276:351-368
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Le Hir H, Gatfield D, Izaurralde E, Moore MJ (2001) The exon-exon junction complex provides a binding platform for factors
involved in mRNA export and nonsense-mediated mRNA decay EMBO J. 20:4987-4997.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Maniatis T, Reed R (2002) An extensive network of coupling among gene expression machines Nature 416:499-506.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Moabbi AM, Agarwal N, El Kaderi B, Ansari A (2012) Role for gene looping in intron-mediated enhancement of transcription Proc.
Natl. Acad. Sci. USA 109:8505-8510
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Morello L, Bardini M, Sala F, Breviario D (2002) A long leader intron of the Ostub16 rice beta-tubulin gene is required for highlevel gene expression and can autonomously promote transcription both in vivo and in vitro Plant J. 29:33-44
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Morello L, Giani S, Troina F, Breviario D (2011) Testing the IMEter on rice introns and other aspects of intron-mediated
enhancement of gene expression J. Exp. Bot. 62:533-544
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Morton T et al. (2014) Paired-end analysis of transcription start sites in Arabidopsis reveals plant-specific promoter signatures
Plant Cell 26:2746-2760
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Okkema PG, Harrison SW, Plunger V, Aryana A, Fire A (1993) Sequence requirements for myosin gene expression and regulation in
Caenorhabditis elegans Genetics 135:385-404
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Palmiter RD, Sandgren EP, Avarbock MR, Allen DD, Brinster RL (1991) Heterologous introns can enhance expression of
transgenes in mice Proc. Natl. Acad. Sci. USA 88:478-482
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Parra G, Bradnam K, Rose AB, Korf I (2011) Comparative and functional analysis of intron-mediated enhancement signals reveals
conserved features among plants Nucleic Acids Res. 39:5328-5337
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Rose AB (2002) Requirements for intron-mediated enhancement of gene expression in Arabidopsis RNA 8:1444-1453
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Rose AB (2004) The effect of intron location on intron-mediated enhancement of gene expression in Arabidopsis Plant J. 40:744751
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Rose AB, Beliakoff JA (2000) Intron-mediated enhancement of gene expression independent of unique intron sequences and
splicing Plant Physiol. 122:535-542
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Rose AB, Carter A, Korf I, Kojima N (2016) Intron sequences that stimulate gene expression in Arabidopsis Plant Mol. Biol. 92:337346
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Rose AB, Casselman AL, Last RL (1992) A phosphoribosylanthranilate transferase gene is defective in blue fluorescent
Arabidopsis thaliana tryptophan mutants. Plant Physiol. 100:582-592
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Rose AB, Elfersi T, Parra G, Korf I (2008) Promoter-proximal introns in Arabidopsis thaliana are enriched in dispersed signals that
elevate gene expression Plant Cell 20:543-551
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Rose AB, Emami S, Bradnam K, Korf I (2011) Evidence for a DNA-based mechanism of intron-mediated enhancement Front. Plant
Sci. 2:98
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Rose AB, Last RL (1997) Introns act post-transcriptionally to increase expression of the Arabidopsis thaliana tryptophan pathway
gene PAT1 Plant J. 11:455-464
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Scotto-Lavino E, Du G, Frohman MA (2006) 5' end cDNA amplification using classic RACE Nat. Protoc. 1:2555-2562
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Segal E, Widom J (2009) What controls nucleosome positions? Trends Genet. 25:335-343
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Snowden KC, Buchholz WG, Hall TC (1996) Intron position affects expression from the tpi promoter in rice Plant Mol. Biol. 31:689692
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Thomas MC, Chiang CM (2006) The general transcription machinery and general cofactors Crit. Rev. Biochem. Mol. Biol. 41:105178
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Vernimmen D, Bickmore WA (2015) The Hierarchy of Transcriptional Activation: From Enhancer to Promoter Trends Genet. 31:696708
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Wiegand HL, Lu S, Cullen BR (2003) Exon junction complexes mediate the enhancing effect of splicing on mRNA expression Proc.
Natl. Acad. Sci. USA 100:11327-11332
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Yang C, Bolotin E, Jiang T, Sladek FM, Martinez E (2007) Prevalence of the initiator over the TATA box in human and yeast genes
and identification of DNA motifs enriched in human TATA-less core promoters Gene 389:52-65
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Zhang N, McHale LK, Finer JJ (2015) Isolation and characterization of "GmScream" promoters that regulate highly expressing
soybean (Glycine max Merr.) genes Plant Sci. 241:189-198
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Intron DNA Sequences Can Be More Important Than the Proximal Promoter in Determining the
Site of Transcript Initiation
Jenna E Gallegos and Alan B. Rose
Plant Cell; originally published online April 3, 2017;
DOI 10.1105/tpc.17.00020
This information is current as of June 17, 2017
Supplemental Data
/content/suppl/2017/04/05/tpc.17.00020.DC1.html
/content/suppl/2017/04/28/tpc.17.00020.DC2.html
Permissions
https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X
eTOCs
Sign up for eTOCs at:
http://www.plantcell.org/cgi/alerts/ctmain
CiteTrack Alerts
Sign up for CiteTrack Alerts at:
http://www.plantcell.org/cgi/alerts/ctmain
Subscription Information
Subscription Information for The Plant Cell and Plant Physiology is available at:
http://www.aspb.org/publications/subscriptions.cfm
© American Society of Plant Biologists
ADVANCING THE SCIENCE OF PLANT BIOLOGY