RESEARCH PAPER |
Structural relationships between diverse cis-acting elements are critical for the functional properties of a rbcS minimal light regulatory unit
1Departamento de Ingeniería Genética de Plantas, Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional – Unidad Irapuato, Apartado Postal 629, Irapuato, Guanajuato 36500, Mexico
2Laboratorio Nacional de Genómica para la Biodiversidad, Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional – Unidad Irapuato, Apartado Postal 629, Irapuato, Guanajuato 36500, Mexico
To whom correspondence should be addressed: E-mail: lherrera{at}ira.cinvestav.mx
Received 26 September 2007; Accepted 5 November 2007
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Light-regulation of photosynthesis-associated nuclear genes is mediated by multipartite cis-regulatory units located in their promoter regions. The combination, spacing, and relative orientation of transcription factor binding sites in these units influences the assembly of specific multiprotein complexes that control gene expression. In this work, the functional architecture of the conserved modular array 5 (CMA5), a previously characterized minimal light-regulatory unit of rbcS gene promoters, has been analysed. With the aim of defining the sequences of CMA5 that, besides the I- and G-box elements, are essential for CMA5 responsiveness to light and chloroplast-derived signals, a series of mutations affecting the sequences flanking these elements was performed. It was found that an I-box associated module, named IbAM5, is essential for CMA5 functionality and is able to bind nuclear proteins in vitro. The spacing requirements of the I- and G-box elements in achieving adequate combinatorial interaction were also studied as well as the effect of interchanging the relative position of these elements in the native rbcS promoter arrangement. The results show that helical phasing and distance between the I- and G-box motifs are critical to determine the functionality and transcriptional strength of CMA5. Furthermore, it is shown that the relative position of the IbAM5/I-box composite element and the G-box element is not critical for the enhancer activity of CMA5, as long as the proper distance between them is maintained. Taken together, these results suggest that the light-responsive, plastid-dependent activity of CMA5 requires the synergistic interaction of several DNA-binding transcription factors assembled in a higher-order nucleoprotein complex.
Key words: Enhancer element, gene regulation, light regulation, nuclear–plastid signalling, photosynthesis genes, promoter architecture
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Plant responses to light are complex and affect multiple developmental processes, including seed germination, seedling photomorphogenesis, phototropism, gravitropism, chloroplast development, circadian rhythms, and flower induction, among others. To achieve these multiple responses, plants use at least three distinct sets of photoreceptors to monitor the environment: red and far-red light-absorbing phytochromes, the blue-light receptors cryptochromes and phototropins, and photoreceptors for ultraviolet B wavelengths (Lorrain et al., 2006; Jiao et al., 2007). The activation of photoreceptors, mainly phytochromes and cryptochromes, can significantly affect gene expression through downstream signalling networks and, in a few cases, by direct effects on transcription factors (Casal and Yanovsky, 2005; Jiao et al., 2007). In particular, light induces the expression of many nuclear-encoded photosynthetic genes, such as RBCS (encoding the small subunit of ribulose-1,5-bisphosphate carboxylase) and LHCB (formerly CAB, encoding light-harvesting chlorophyll a/b binding proteins) both belonging to the superfamily of photosynthesis-associated nuclear genes (PhANGs), whose expression is co-ordinated with chloroplast development (Gray et al., 2003). Besides the profound effects of light on gene transcription, several studies have shown that light-controlled gene expression is regulated at different levels, such as chromatin re-modelling and post-transcriptional and post-translational regulation, although these mechanisms remain relatively unexplored (Casal and Yanovsky, 2005). Expression-profiling experiments have demonstrated that light induces dramatic changes in the transcriptome of Arabidopsis, and early light-responsive genes include a large proportion of transcription factors (Casal and Yanovsky, 2005; Jiao et al., 2007).
Several transcription factors involved in light regulation of gene expression have been identified by their ability to bind to their cognate DNA sequences present in many plant promoters. These regulatory elements are known as light-responsive cis-elements (LREs), which are apparently essential for light-controlled transcriptional activity. Although several LREs and their binding proteins have been identified, no single element has been found in all of the photoregulated promoters, or demonstrated to confer light responsiveness to minimal heterologous promoters in gain-of-function experiments. This suggests that combinations of different cis-acting sequences composing light-responsive units (LRUs), rather than individual elements, are required to confer proper photoresponsiveness to a light-insensitive basal promoter (Terzaghi and Cashmore, 1995; Puente et al., 1996; Chattopadhyay et al., 1998). The composite structure of the LRUs, may explain differences in their responses to light in terms of intensity and spectral quality and/or differences in their developmental and organ-specific expression patterns.
The functional characterization of the Conserved Modular Array 5 (CMA5), the shortest native light-responsive unit of a photosynthetic gene promoter characterized experimentally to date (Martínez-Hernández et al., 2002), constitutes a clear example of this complexity. CMA5 corresponds to a 52 bp fragment of the Nicotiana plumbaginifolia rbcS 8B promoter, which has been shown to act as a functional light-responsive enhancer in Nicotiana tabacum and Arabidopsis thaliana transgenic plants. CMA5 contains an I-box and a G-box element, both of which are essential for the activation of a minimal heterologous promoter in a phytochrome-, cryptochrome-, and plastid signal-dependent manner (Martínez-Hernández et al., 2002). Furthermore, it has recently been shown that CMA5 is able to respond not only to light and chloroplast signals, but also to sugar signals in a pathway involving ABA (Acevedo-Hernández et al., 2005). The latter response is mediated by direct binding of the ABI4 transcription factor to the S-box, a conserved element found in close association with the G-box motif of CMA5 (Martínez-Hernández et al., 2002; Acevedo-Hernández et al., 2005).
Although the light-dependent activity of CMA5 has been shown to rely on the presence of both I- and G-box elements (Martínez-Hernández et al., 2002), it is also possible that sequences flanking these elements are important for activity of this LRU. One of those flanking sequences is an element adjacent to the I-box core motif found in the rbcS promoters from several dicotyledons, formerly designated GA-motif or ARGATGA-motif, and named CMA5 I-box Associated Module in this work (IbAM5 in Fig. 1).
|
To gain further insights into the complexity and architecture of light-responsive enhancer elements, a detailed dissection of the functional architecture of CMA5 is presented here. The results show that IbAM5 is an essential element for CMA5 activity in addition to the I- and G-box motifs, that the spacing between these individual cis-acting elements is important for achieving effective combinatorial interaction, and that the relative position of the I- and G-boxes in the regulatory unit does not seem to be critical, as long as optimal spacing is maintained. In addition, evidence is provided suggesting that all the cis-acting elements required for the light-responsive activity of CMA5 are bound by a DNA-binding multiprotein complex rather than by individual transcription factors.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Plant material, reporter constructs, plant transformation, and growth conditions
The Arabidopsis transgenic lines used in this study were generated by Agrobacterium-mediated transformation of ecotype RLD, except for Line 1 of CMA5-GUS, included for comparison and described elsewhere (Martínez-Hernández et al., 2002). Oligonucleotides corresponding to distinct versions of CMA5 were synthesized and fused in one copy to the –46/+8 CaMV 35S promoter upstream of the coding sequence of the uidA (GUS) reporter gene in pBI146S (Martínez-Hernández et al., 2002). Arabidopsis plants were transformed by Agrobacterium tumefaciens strains containing the constructs through the floral dip method reported by Clough and Bent (1998). Ten independent transgenic lines homozygous for each construct, including for pBI146S, were analysed. Three representative lines were selected for subsequent experiments, unless otherwise stated.
Arabidopsis seeds were surface-sterilized and then plated on media containing 1x Murashige and Skoog (MS) basal salt mixture (Gibco-BRL, Grand Island, NY, USA) supplemented with 1% (w/v) sucrose and 6.5 g l–1 phytagar (Gibco-BRL). Seeds underwent a 16 h treatment of continuous white light (WL) in controlled growth chambers (MOD AR-32L; Percival Scientific Inc., Boone, IA, USA) at 24 °C, with a light fluence of 100 µmol m–2 s–1 to induce germination. Subsequently, plates were covered with aluminium foil, stored for 3 d at 4 °C, and transferred to the appropriate experimental light conditions.
Light and norflurazon (NF) treatments
White light (WL) treatments were at a fluence of 100 µmol m–2 s–1 with a 16 h photoperiod for 2 weeks, in a growth chamber as described above. Dark-grown seedlings (Dark) were in the same conditions, except that the plates were covered with aluminium foil and kept in black plastic boxes. These seedlings were harvested using green safe-light conditions at 0.01 µmol m–2 s–1. For de-etiolation experiments (24 h-WL), 3-d-old dark-grown seedlings were exposed to a 24 h treatment of continuous white light at 100 µmol m–2 s–1. During this process cotyledons opened and turned green. For daily light-pulses (WLP), 3-d-old etiolated seedlings were treated with five pulses, one per day, of 5 min of white light at 600 µmol m–2 s–1. Seedlings were harvested 24 h after the onset of the last irradiation. For NF treatments, seeds were germinated and grown for 2 weeks in white light at 100 µmol m–2 s–1 in MS medium supplemented with 1 µM NF, which was kindly provided by the Sandoz Chemical Company (Des Plaines, IL).
GUS fluorometric analyses
Protein content was quantified using the Bradford reagent (Bio-Rad Laboratories, Hercules, CA, USA). For each assay, 5 µg of total protein were used, and GUS activity performed according to Gallagher (1992), using a TKO 100 fluorometer (Hoefer Scientific Instruments, San Francisco, CA, USA). GUS activity is reported as picomoles of 4-methylumbelliferone (MU) per milligram of protein per minute.
Gel-shift assays
Nuclear protein extracts from 2-week-old light-grown Arabidopsis plants were prepared as described by Giuliano et al. (1988), and used for gel-shift assays as reported by Borello et al. (1993), except that samples were run on a 7% acrylamide gel, 0.25x TBE buffer was used for electrophoresis, and competitors were added in a 50x or 100x molar excess. Complementary oligonucleotides corresponding to each sequence element were synthesized, annealed into double-stranded DNA, and phosphorylated with 32P using T4-polynucleotide kinase (Invitrogen, Carlsbad, CA, USA). For the competition assays, non-labelled oligonucleotides were used. For the non-specific competition assays, the polylinker region of plasmid pBluescript was obtained by digestion with restriction endonucleases HindIII–BamHI and purified. Images of gel-shift assays were processed using Adobe photoshop 7 and contrasted with Gene-Spring software.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Phylogenetically conserved elements other than the I- and G-boxes are present in CMA5
A previous informatics analysis of light-responsive promoters from monocots, dicots, and gymnosperms revealed that a tripartite modular array of I-G-I boxes is present in rbcS promoters of the three main lineages of vascular plants (Argüello-Astorga and Herrera-Estrella, 1996, 1998; Fig. 1A). Divergence in the structural organization of this ancestral I-G-I array is evident by the finding of several Conserved Modular Arrays (CMAs). This diversification has apparently taken place by the evolutionary integration of new specific DNA motifs closely associated to the central regulatory sequences and by the secondary loss of some of the DNA motifs present in this original CMA (Fig. 1A).
The rbcS CMA5, an I-G modular arrangement shown to function independently in gene regulation as a light-responsive enhancer (Martínez-Hernández et al., 2002), has been maintained as a single arrangement in the Solanaceae family (Fig. 1B). The incorporation of new specific modules into this I-G array can be illustrated in the tomato rbcS multigene family. The rbcS1, rbcS2, and rbcS3A promoters contain a combination of I- and G-box elements in identical spatial arrangement. The intermediate sequence between the I- and G- boxes differs among the three promoters, suggesting that these sequences are not functionally relevant; however, these spacer sequences are conserved in orthologous rbcS genes from different plant species in the Solanaceae family (Fig. 1A, B), where family member-specific modules Sp1, Sp2, and Sp3 correspond to the rbcS1, rbcS2, and rbcS3A tomato promoters, respectively. Interestingly, the spacer sequence of the rbcS3A promoter was shown to be the target of a negative fruit-specific factor, associated with the reduced expression of the corresponding gene in young tomato fruits (Meier et al., 1995). Further analysis of several CMA5-containing promoter sequences revealed that incorporation of new elements within CMA5 has also taken place in the form of an I-box associated module with the consensus TRGGATGA (herewith named IbAM5; Fig. 1C; Argüello-Astorga and Herrera-Estrella, 1996, 1998). Data derived from these analyses strongly suggest that other conserved elements present in CMA5, in addition to the I-, G-, and S-boxes, might be cis-acting elements involved in modulating the transcriptional activity of this complex light-responsive unit.
The IbAM5 element is necessary for CMA5 functionality
It has previously been shown that mutation of either the I- or G-boxes abolished CMA5 activity, showing that both elements are necessary for the enhancer activity of this CMA (Martinez-Hernandez et al., 2002). However, since, in addition to the I- and G- boxes, other conserved sequences, such as IbAM5, are present in several CMA5-containing rbcS promoters, it was important to determine whether those additional motifs are also functional components of this LRE.
With this aim, an altered version of CMA5 containing the I- and G-boxes in the original orientation and spacing, but in which the rest of the sequence was replaced by a random choice of nucleotides, was synthesized (mSS; Fig. 2A). Moreover, oligonucleotides for two different versions of CMA5 where the spacer sequence was replaced by random sequences, and an additional version in which only the IbAM5 was mutated, were also generated (mSp1, mSp2, and mIbAM5, respectively; Fig. 2A). These synthetic DNAs were fused to the –46/+8 CaMV 35S minimal promoter upstream of the coding sequence of the GUS reporter gene and used to produce Arabidopsis transgenic lines, in order to determine the ability of these CMA5 mutants to activate light-regulated transcription, compared with that of the native CMA5 sequence contained in CMA5-GUS Arabidopsis transgenic plants (Martínez-Hernández et al., 2002).
|
Arabidopsis transgenic lines harbouring the constructs described above, were either grown under continuous darkness or germinated in darkness and subjected to a 24 h treatment of white light before their GUS activity was determined by fluorometric assays. As previously reported, GUS expression in CMA5-GUS lines is induced 3–5 times in seedlings subjected to a 24 h light treatment (Martínez-Hernández et al., 2002). In mSS-GUS lines in which the flanking sequences of the I- and G-boxes of CMA5 was changed by a random sequence, no GUS activity, above that present in lines containing the –46/+8 35S promoter alone (46S-GUS lines), could be detected in either light or dark conditions (Fig. 2B). A total of eight transgenic mSS-GUS lines were analysed in detail along different developmental stages, but significant reporter gene expression was not detected at any age (Fig. 2B; data not shown). These results suggest that sequences other than the I- and G- boxes are required for the CMA5 light-responsive enhancer function.
Substitution of the spacer sequence between the I- and G-boxes of CMA5 in mSp1-GUS and mSp2-GUS lines resulted in a 30–40% reduction in the level of GUS activity compared with that in the native CMA5-GUS lines in both light and dark conditions. However, these lines with altered sequences clearly displayed a 3–4.5-fold light-induced increase of GUS activity, which is very similar to the induction observed for CMA5-GUS (Fig. 2B). These results show that the spacer sequence between the I- and G-boxes of CMA5 could participate in modulating the strength of this enhancer, but not its light-regulated activity.
GUS activity in IbAM5-GUS lines, in which IbAM5 is missing, was similar to that present in control 46S-GUS lines and no effect due to the light treatment was observed (Fig. 2B). A total of 12 IbAM5-GUS lines were analysed and no enhancer activity was detected at any developmental stage (Fig. 2B; data not shown). Alterations of the 3' region of CMA5 were not included in this study since it was previously demonstrated that this region was not necessary for the light-responsiveness of this CMA (Acevedo-Hernández et al., 2005).
Taken together, these results show that the I- and G-boxes are necessary, but not sufficient in the context of the native CMA5, to constitute an active LRE able to enhance transcription above background levels of a minimal heterologous promoter and that IbAM5, but not the spacer sequence, is required for CMA5 light-regulated enhancer activity.
The spacer sequence is not involved in chloroplast-dependent CMA5 activity
CMA5-mediated expression has been shown to be dependent on chloroplast development (Martínez-Hernández et al., 2002). To determine whether alterations in the spacer sequence affected the CMA5 dependence on plastids, seedlings of the mSp1-GUS and mSp2-GUS transgenic lines were grown in media containing the herbicide norflurazon (NF), which inhibits carotenoid biosynthesis and arrests chloroplast development in light-grown plants (Oelmüller et al., 1986). As previously reported, the expression directed by CMA5 in NF treated seedlings was reduced to a level similar to that observed in seedlings grown in darkness (Fig. 2C). mSp1-GUS and mSp2-GUS transgenic lines had a similar response to that displayed by CMA5-GUS lines, showing a 3–5-fold reduction in GUS activity when treated with NF (Fig. 2C). These results demonstrate that the spacer sequence is not involved in the chloroplast-dependent activity of CMA5. Moreover, they strongly suggest that the regulatory properties of CMA5 displayed in response to light- and plastid-derived signals are mediated by the same cis-acting elements.
The IbAM5 element binds nuclear protein factors in vitro
Previously, Borello et al. (1993) reported binding activities from tomato nuclear extracts that recognize the I-box and related elements localized in the rbcS3A promoter of tomato. To determine whether Arabidopsis nuclear protein extracts also contained similar binding activities recognizing the conserved elements found in CMA5, a sequence containing the I-box along with IbAM5 was used as a probe in gel-shift assays (composite I-box sequence or Im probe; Fig. 3A, B). A pattern of binding complexes was observed that was very similar to that previously reported by Borello et al. (1993), where two retardation bands are clearly distinguished. Similarly, the gel shift patterns obtained with Arabidopsis nuclear protein extracts were determined for a probe containing the G-box of CMA5 (G probe; Fig. 3A, B). Interestingly, the competition properties of these complexes were also similar to those reported for tomato extracts by Borello et al. (1993), since an excess of unlabelled G oligonucleotide is able to compete for the slow migrating complex obtained when the composite I-box element is used as a probe, but unlabelled Im does not compete for the complexes binding to the G probe (Fig. 3B).
|
To determine whether the IbAM5 alone is able to bind nuclear protein factors, gel shift assays were carried out using a probe containing this motif (AM probe, Fig. 3A). It was observed that this sequence is able to bind proteins present in the Arabidopsis nuclear extract, forming two complexes (Fig. 3C). The gel-shift pattern observed was very similar to that obtained using the Im probe or a probe lacking IbAM5 (I-box core sequence or Ic probe; Fig. 3A, C). The protein nature of these complexes was demonstrated since their binding activity was Proteinase K-sensitive and RNase-resistant (data not shown). These binding activities were sensitive to the protein dissociating agent sodium deoxycholate (Fig. 3C, lane D), suggesting that they might be multimeric factors in which protein–protein interaction is important for DNA-binding activity. Furthermore, an excess of unlabelled AM oligonucleotide is able to compete with the AM labelled probe, albeit not completely. Interestingly, the Im probe is also able to compete with the AM probe in the formation of the observed complexes, suggesting that in vitro, IbAM5 itself is able to bind the same complexes as the broader sequence containing the IbAM5/I-box. Altogether, these results demonstrate that IbAM5 is recognized by nuclear factors from Arabidopsis, probably of the same nature as those recognizing the I-box alone.
The enhancer activity of CMA5 is phase-sensitive and distance-sensitive
Although there is little sequence homology in the intermediate DNA sequence between the I- and G-boxes in the CMA5 of rbcS promoters, its length is well conserved in the Solanaceae family (13 bp), and the distance varies slightly in other plant species including Arabidopsis, in which CMA5 has been proven to be functional (Fig. 1C). In a composite element in which efficient assembly of a functional transcription complex depends on optimally spaced protein–protein interactions among the cognate DNA-binding factors and auxiliary proteins, distance constraints between the binding sites are expected.
To determine whether spacing restrictions between the I- and G-boxes influence CMA5 activity, several versions of CMA5 where the I-G spacer length is altered were constructed and experimentally analysed. Mutated versions of CMA5 in which 5, 10, 15, or 20 additional base pairs were inserted between the I- and G-boxes were generated (Fig. 4A). These altered versions of CMA5 were fused to the –46/+8 CaMV 35S minimal promoter upstream of the coding sequence of the GUS reporter gene and used to produce Arabidopsis transgenic lines. Analysis of transgenic lines containing these chimeric gene constructs showed that I-5-G lines displayed a 4-fold decrease in GUS activity when compared with that observed in CMA5 lines. Interestingly, GUS activity was partially restored in I-10-G lines, which, on average, had activities equivalent to 50% of that observed in CMA5-GUS lines (Fig. 4B). When the length of the spacer sequence was increased by 15 and 20 bp, no GUS activity was detected in the Arabidopsis transgenic lines carrying either of these constructs.
|
These results show that the distance between the I- and G-boxes present in CMA5 is an important architectural feature of this array and that these DNA motifs cannot be separated by more than 23–25 bases. Moreover, since 10 bp represent one turn of DNA-B helix, the results obtained for I-5-G and I-10-G lines also suggest that proper helical phasing between the I- and G-boxes is necessary for the assembly of a functional gene-transcription activating complex.
To evaluate further whether the versions of CMA5 altered in I-G spacing by insertions of 5 bp and 10 bp were still able to respond to light treatments, white-light pulse experiments (WLP) were used to test the light-responsiveness of these CMA5 derivatives. It was observed that CMA5 directs a 3.5–4.5-fold increase in GUS activity with this treatment, compared with that observed in dark-grown seedlings (Fig. 4C). In the case of I-5-G and I-10-G lines, it was determined that both altered versions of CMA5 are still able to display a significant response to light pulses (Fig. 4C). Interestingly, these transgenic lines also retained chloroplast-dependent gene expression, since plants grown on NF-containing medium, exhibited a significant decrease in GUS activity when compared with untreated controls (Fig. 4D).
Taken together, these results indicate that, even though helical phasing and distance between the I- and G-boxes are important to determine the functionality and transcriptional strength of CMA5, these characteristics do not directly affect the light-responsiveness and dependence on plastid-derived signals of this composite cis-regulatory element.
The relative position of the I- and G-boxes is not critical for CMA5 activity
Several CMAs that presumably are evolutionary derivatives of the proposed I-G-I ancestral array present in rbcS promoters of vascular plants have been described (Argüello-Astorga and Herrera-Estrella, 1996, 1998; Fig. 1). Since some of these derivatives display a G-I arrangement instead of the I-G organization observed in CMA5, it was important to determine whether the relative position of the I- and G-boxes in CMA5 is a critical feature for determining the regulatory properties of this cis-acting complex unit. To assess this, oligonucleotides were designed in which the I-G structure of CMA5 was changed to a G-I arrangement. Two ‘shuffled’ versions of CMA5 were produced maintaining a spacer length similar to the original version, whether between the G-box and the IbAM5 or between the G-box and the I-box (CMA5_SHUF1 and CMA5_SHUF2, respectively; Fig. 5A). These oligonucleotides were fused to the –46/+8 CaMV 35S minimal promoter upstream of the coding sequence of the GUS reporter gene and used to produce Arabidopsis transgenic lines. It was found that CMA5_SHUF2 lines displayed a light-regulated expression pattern similar to that of CMA5, whereas CMA5_SHUF1 was unable to activate the basal transcription of the –46/+8 minimal promoter. These lines were also grown in NF-containing medium and it was observed that CMA5_SHUF2 displays a chloroplast-dependent response resembling that of CMA5 (Fig. 5C). Since CMA5_SHUF1 and CMA5_SHUF2 contain identical I- and G-boxes placed in the same order and only differ in the spacer length, this demonstrates that the relative position of these elements is not a determinant for the regulatory properties of CMA5, as long as the optimal distance between them is maintained.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The rbcS CMA5 minimal light-responsive unit has proved to be a useful tool for unravelling the complex mechanisms of transcriptional regulation of PhANGs as well as a paradigm for the study of combinatorial control of transcription in plants. The analysis of this compact yet complex enhancer has allowed us to evaluate in a native context the contribution of single cis-acting elements to the transcriptional activity of an artificial gene promoter in response to environmental and physiological signals such as light, sugar, ABA, and plastid-derived signals (Martínez-Hernández et al., 2002; Acevedo-Hernández et al., 2005). Earlier studies defined this LRU as a common target for different signals involved in light regulation and tissue specificity of PhANGs, being able to respond to both positive and negative regulatory mechanisms in a fashion resembling the functional properties of the corresponding complete rbcS promoter (Martínez-Hernández et al., 2002; Acevedo-Hernández et al., 2005). Furthermore, employing CMA5 the S-box was identified, a previously uncharacterized regulatory element mediating sugar responsiveness of rbcS promoters, which corresponds to a binding site for the ABI4 transcription factor. This finding allowed a model to be proposed for the ABI4-mediated repression of PhANGs in response to sugars by preventing binding of transcription factors to the G-box element (Acevedo-Hernández et al., 2005). Interestingly, the same mechanism of inhibition of G-box-mediated expression of photosynthetic genes by ABI4 was very recently postulated by Koussevitzky et al. (2007) to explain the transcriptional regulation of PhANGs mediated by plastid-to-nucleus retrograde signalling. The examples above demonstrate the importance of studying the architectural and functional properties of this apparently simple regulatory unit as an approach for understanding the complex interactions and transcriptional networks regulating the expression of photosynthesis-associated genes in response to diverse signals.
A previous study showed that both the I- and G-box elements are required for the activity of CMA5 as a light-responsive enhancer in plants (Martínez-Hernández et al., 2002). An important goal of this work was the definition of additional cis-acting elements other than the I- and G-boxes that are essential for the light-responsiveness of this LRU. A commonly used approach to identify potential regulatory elements in non-coding gene sequences is to search for conserved DNA motifs or ‘phylogenetic footprints’ (PFs). This was accomplished by applying a ‘phylogenetic-structural method’ for the sequence analysis of several rbcS promoters (Argüello-Astorga and Herrera-Estrella, 1996, 1998; this study). IbAM5 was identified as a PF conserved in the promoters of plants belonging to six different taxonomic orders of dicotyledons (Fig. 1C), hence indicating that this DNA motif along with the I- and G-boxes array have persisted in rbcS promoters since the Late Cretaceous (90–65 million years ago), to which can be dated the divergence of the Asterales, Caryophyllales, Solanales, Gentaniales, Malvales, and Fabales (Wikström et al., 2001; Weeks et al., 2007). The sequence conservation of IbAM5 suggests that this element is a potential binding site for transcription factors, a notion which was strongly supported by our experimental results showing that block mutation of this element completely abolished the light-responsive enhancer activity of CMA5 (Fig. 2). Accordingly, a previous work showed that IbAM5 was important for binding of tomato nuclear factors in the context of a broader sequence also containing the I-box motif (Borello et al., 1993), although the ability of IbAM5 itself to bind nuclear proteins had not been reported. Gel-shift assays showed that IbAM5 is specifically recognized by nuclear factors from Arabidopsis (Fig. 3C). An interesting observation is that unlabelled oligonucleotides containing only the I-box core could compete with IbAM5 for the binding of nuclear proteins, hence suggesting that IbAM5 and the I-box probably interact with the same nuclear factor or protein complex. The sensitivity of the nucleoprotein complex to treatment with a protein dissociation agent indicates that protein–protein interactions are important for this IbAM5-binding activity. However, at this point, it cannot be discerned whether the same protein binds to both IbAM5 and the I-box through distinct but adjacent DNA-binding domains, or that each DNA sequence motif binds different proteins forming part of the same multimeric DNA-binding complex.
Interestingly, it was found that the G-box alone is able to compete with the composite I-box element (Im probe) for the binding of nuclear protein complexes, but this IbAM5/I-box sequence did not compete for the G-box binding activity (Fig. 3B). These results are similar to those reported by Borello et al. (1993) using tomato nuclear extracts. The finding that the composite I-box cannot compete for the G-box binding activity of Arabidopsis nuclear extracts could be explained by the existence of a large number of distinct G-box binding factors (GBFs) encoded in the nuclear genome and present in the protein extracts, many of which probably do not interact with the protein(s) that bind the I-box (IBFs). However, the fact that the same DNA-binding activity in nuclear extracts is able to recognize both the I- and the G-boxes suggests that this activity corresponds to a protein complex containing factors that bind concurrently to their cognate elements.
It is expected that multiple DNA–protein contacts inherent to higher order nucleoprotein complexes follow the ‘helical phasing’ rule for composite elements. According to existing models, this rule establishes that the distribution of DNA motifs showing a preferential binding site arrangement that allows transcription factors to be placed on the same surface of DNA, facilitates protein–protein interactions and promotes formation of specific tertiary complexes (DNA–protein–protein–DNA), involved in activation of specific transcription (Makeev et al., 2003). Our results show that helical phasing of the I- and G-boxes is important for achieving the optimal transcription activation observed for CMA5, suggesting that direct protein–protein interaction between the transcription factors binding the associated elements or indirect interaction through the recruitment of auxiliary proteins is necessary for combinatorial regulation. Although CMA5 activity can be restored by re-establishing helical phasing, the flexibility of the protein complex seems to be limited, since increasing the length of the spacer sequence beyond 23 bp completely abolished CMA5 activity (Fig. 4B). The limitations in the length of the spacer sequence between the I- and G-boxes suggest that a preformed stereospecific transcription enhancer complex that includes IBF, GBF, and the protein that binds IbAM5 binds CMA5. Alternatively, these factors may bind to DNA independently and then interact or recruit other proteins. In this context, however, it should be taken into account that helical phasing and distance constraints on this coactivator recruitment mechanism are usually more flexible than those observed for co-operatively binding transcription factors, such as that observed for CMA5 (Makeev et al., 2003; Chiang et al., 2006).
Considering that changing the relative position of the elements in CMA5 but maintaining the spacer length had no effect on its regulatory properties, it is possible that interaction of I- and G-box binding factors is not direct, but mediated by other protein(s). It is possible that the size of this protein or protein complex is large enough so that it is able to mediate the interaction of DNA-bound factors even when the relative position of the elements is altered. Such an explanation has also been proposed for other systems (Shen et al., 2004). In addition, the fact that the G-box is a palindromic sequence could facilitate this conformational adaptation. Once more, this apparent structural flexibility of the nucleoprotein complexes is limited by distance constraints.
This work provides important insights into CMA5 functional architecture, defining IbAM5 as a new functional conserved element and establishing the relative importance of the sequence context around the evolutionarily more conserved elements, as well as the distance and the relative position between them. These data indicate that CMA5 is a complex regulatory unit that requires at least three DNA motifs for its enhancer activity. The use of CMA5 as a model for studying combinatorial control of gene expression in plants is useful in defining the precise protein–protein interactions determining enhancer activity of this LRE or to analyse in more detail the ability of this unit to respond to multiple developmental signals and environmental cues.
|
Acknowledgements |
|---|
This project was supported in part by the following grants CONACyT- SEP 43979 and HHMI (12348739) to LHE. LLO is grateful to CONACYT and CONCYTEG for a PhD fellowship.
|
Footnotes |
|---|
* Present address: Department of Molecular and Structural Biochemistry, North Carolina State University, 360 Partners III, Centennial Campus, 851 Main Campus Drive, Raleigh, NC 27606, USA.
Present address: Colegio de Postgraduados, Campus Campeche, Calle Nicaragua 91 Tercer piso entre Tamaulipas y Circuito Baluartes Col. Santa Ana, Campeche, Campeche 24050, Mexico.
División de Biología Molecular, Instituto Potosino de Investigaciones Científicas y Tecnológicas, Camino a la Presa San José 2055, Lomas Cuarta Sección, San Luis Potosí, San Luis Potosí 78216, Mexico.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Acevedo-Hernández GJ, León P, Herrera-Estrella LR. Sugar and ABA responsiveness of a minimal RBCS light-responsive unit is mediated by direct binding of ABI4. The Plant Journal (2005) 43:506–519.[CrossRef][Web of Science][Medline]
Argüello-Astorga GR, Herrera-Estrella LR. Ancestral multipartite units in light-responsive plant promoters have structural features correlating with specific phototransduction pathways. Plant Physiology (1996) 112:1151–1166.[Abstract]
Argüello-Astorga G, Herrera-Estrella L. Evolution of light-regulated plant promoters. Annual Review of Plant Physiology and Plant Molecular Biology (1998) 49:525–555.[CrossRef][Web of Science][Medline]
Borello U, Ceccarelli E, Giuliano G. Constitutive, light-responsive and circadian clock-responsive factors compete for the different I box elements in plant light-regulated promoters. The Plant Journal (1993) 4:611–619.[CrossRef][Web of Science][Medline]
Casal JJ, Yanovsky MJ. Regulation of gene expression by light. International Journal of Developmental Biology (2005) 49:501–511.[CrossRef][Web of Science][Medline]
Chattopadhyay S, Puente P, Deng X-W, Wei N. Combinatorial interaction of light-responsive elements plays a critical role in determining the response characteristics of light-regulated promoters in Arabidopsis. The Plant Journal (1998) 15:69–77.[CrossRef][Web of Science][Medline]
Chiang DY, Nix DA, Shultzaberger RK, Gasch AP, Eisen MB. Flexible promoter architecture requirements for coactivator recruitment. BMC Molecular Biology (2006) 7:16.[CrossRef][Medline]
Clough SJ, Bent AF. Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. The Plant Journal (1998) 16:735–743.[CrossRef][Web of Science][Medline]
Gallagher SR. Quantitation of GUS activity by fluorometry. In: GUS protocols: using the GUS gene as a reporter of gene expression—Gallagher SR, ed. (1992) San Diego, CA: Academic Press Inc. 47–59.
Giuliano G, Pichersky E, Malik VS, Timko MP, Scolnik PA, Cashmore AR. An evolutionarily conserved protein binding sequence upstream of a plant light-regulated gene. Proceedings of the National Academy of Sciences, USA (1988) 85:7089–7093.
Gray JC, Sullivan JA, Wang J-H, Jerome CA, MacLean D. Coordination of plastid and nuclear gene expression. Philosophical Transactions of the Royal Society of London B-Biological Sciences (2003) 358:135–145.[CrossRef]
Jiao Y, Lau OS, Deng XW. Light-regulated transcriptional networks in higher plants. Nature Reviews Genetics (2007) 8:217–230.[CrossRef][Web of Science][Medline]
Koussevitzky S, Nott A, Mockler TC, Hong F, Sachetto-Martins G, Surpin M, Lim J, Mittler R, Chory J. Signals from chloroplasts converge to regulate nuclear gene expression. Science (2007) 316:715–719.
Lorrain S, Genoud T, Fankhauser C. Let there be light in the nucleus! Current Opinion in Plant Biology (2006) 9:509–514.[CrossRef][Web of Science][Medline]
Makeev VJ, Lifanov AP, Nazina AG, Papatsenko DA. Distance preferences in the arrangement of binding motifs and hierarchical levels in organization of transcription regulatory information. Nucleic Acids Research (2003) 31:6016–6026.
Martínez-Hernández A, López-Ochoa L, Argüello-Astorga G, Herrera-Estrella L. Functional properties and regulatory complexity of a minimal RBCS light-responsive unit activated by phytochrome, cryptochrome, and plastid signals. Plant Physiology (2002) 128:1223–1233.
Meier I, Callan KL, Fleming AJ, Gruissem W. Organ-specific differential regulation of a promoter subfamily for the ribulose-1,5-bisphosphate carboxylase/oxygenase small subunit genes in tomato. Plant Physiology (1995) 107:1105–1118.[Abstract]
Oelmüller R, Levitan I, Bergfeld R, Rajasekhar VK, Mohr H. Expression of nuclear genes as affected by treatments acting on the plastids. Planta (1986) 168:482–492.[CrossRef][Web of Science]
Puente P, Wei N, Deng XW. Combinatorial interplay of promoter elements constitutes the minimal determinants for light and developmental control of gene expression in Arabidopsis. EMBO Journal (1996) 15:3732–3743.[Web of Science][Medline]
Shen QJ, Casaretto JA, Zhang P, Ho T-HD. Functional definition of ABA-response complexes: the promoter units necessary and sufficient for ABA induction of gene expression in barley (Hordeum vulgare L.). Plant Molecular Biology (2004) 54:111–124.[CrossRef][Web of Science][Medline]
Terzaghi WB, Cashmore AR. Light-regulated transcription. Annual Review of Plant Physiology and Plant Molecular Biology (1995) 46:445–474.[CrossRef][Web of Science]
Weeks KE, Chuzhanova NA, Donnison IS, Scott IM. Evolutionary hierarchies of conserved blocks in 5'-noncoding sequences of dicot rbcS genes. BMC Evolutionary Biology (2007) 7:51.[CrossRef][Medline]
Wikström N, Savolainen V, Chase MW. Evolution of the angiosperms: calibrating the family tree. Proceedings of the Royal Society of London B-Biological Sciences (2001) 268:2211–2220.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||