The Arabidopsis bZIP1 Transcription Factor Is Involved in ... - Cell Press

Molecular Plant

•

Volume 3

•

Number 2

•

Pages 361–373

•

March 2010

RESEARCH ARTICLE

The Arabidopsis bZIP1 Transcription Factor Is Involved in Sugar Signaling, Protein Networking, and DNA Binding Shin Gene Kanga,b, John Pricea,b, Pei-Chi Linb,c, Jong Chan Hongd and Jyan-Chyun Janga,b,c,1 a b c d

Department of Horticulture and Crop Science, The Ohio State University, Columbus, OH 43210, USA Plant Biotechnology Center, The Ohio State University, Columbus, OH 43210, USA Department of Plant Cellular and Molecular Biology, The Ohio State University, Columbus, OH 43210, USA Department of Biochemistry, Gyeongsang National University, Jinju, Gyeongnam 660-701, Korea

ABSTRACT Sugar signaling is a mechanism that plants use to integrate various internal and external cues to achieve nutrient homeostasis, mediate developmental programs, and articulate stress responses. Many bZIP transcription factors are known to be involved in nutrient and/or stress signaling. An Arabidopsis S1-group bZIP gene, AtbZIP1, was identified as a sugar-sensitive gene in a previous gene expression profiling study (Plant Cell. 16, 2128–2150). In this report, we show that the expression of AtbZIP1 is repressed by sugars in a fast, sensitive, and reversible way. The sugar repression of AtbZIP1 is affected by a conserved sugar signaling component, hexokinase. Besides being a sugar-regulated gene, AtbZIP1 can mediate sugar signaling and affect gene expression, plant growth, and development. When carbon nutrients are limited, gain or loss of function of AtbZIP1 causes changes in the rates of early seedling establishment. Results of phenotypic analyses indicate that AtbZIP1 acts as a negative regulator of early seedling growth. Using gain- and loss-of-function plants in a microarray analysis, two sets of putative AtbZIP1-regulated genes have been identified. Among them, sugar-responsive genes are highly over-represented, implicating a role of AtbZIP1 in sugar-mediated gene expression. Using yeast two-hybrid (Y-2-H) screens and bimolecular fluorescence complementation (BiFC) analyses, we are able to recapitulate extensive C/S1 AtbZIP protein interacting network in living cells. Finally, we show that AtbZIP1 can bind ACGT-based motifs in vitro and that the binding characteristics appear to be affected by the heterodimerization between AtbZIP1 and the C-group AtbZIPs, including AtbZIP10 and AtbZIP63. Key words:

bZIP transcription factors; sugar response; protein dimerization; microarray analysis; BiFC; EMSA; ACGT motif.

INTRODUCTION The perception and management of sugar levels in an organism are important for survival. Plants respond to sugar signals through an integrative process to assess their growth, metabolism, and levels of stress, and appropriately control gene expression (Smeekens, 2000; Rolland et al., 2006; BaenaGonzalez and Sheen, 2008). Regulation of cell signaling can occur at many different levels: transcriptional control via ciselements and trans-acting factors are amongst the most important mechanisms. However, despite the fact that at least 10% of Arabidopsis genes are sugar-responsive (Price et al., 2004; Blasing et al., 2005; Li et al., 2006), very few regulatory circuits are known to use such mechanisms to mediate gene expression via sugar signalings. The bZIP transcription factors are characterized by a basic region mediating sequence-specific DNA-binding, and a leucine zipper region required for dimerization. The bZIP is one of the largest transcription factor families in higher plants. In Arabi-

dopsis, there are 72–77 bZIPs (Guo et al., 2005; Riano-Pachon et al., 2007; Correa et al., 2008) known to regulate genes involved in a diverse array of functions in plant growth, development, and environmental responses (Jakoby et al., 2002). On the basis of phylogenetic analyses, the Arabidopsis bZIP factors can be divided into 13 sub-groups (Correa et al., 2008). The S-group is the largest bZIP group in Arabidopsis, among which AtbZIP11/ATB2 and AtbZIP53 have been well characterized (Kim et al., 2004; Satoh et al., 2004; Wiese et al., 2004, 2005; Weltmeier et al., 2006; Hanson et al., 2008; Alonso et al.,

1 To whom correspondence should be addressed. E-mail [email protected], fax 614-292-5379, tel. 614-292-8496.

ª The Author 2010. Published by the Molecular Plant Shanghai Editorial Office in association with Oxford University Press on behalf of CSPP and IPPE, SIBS, CAS. doi: 10.1093/mp/ssp115, Advance Access publication 15 January 2010 Received 15 October 2009; accepted 14 December 2009

362

|

Kang et al.

d

bZIP Network and Sugar Signaling

2009). Interestingly, AtbZIP11 is up-regulated in carbohydrateconsuming sink tissue at the mRNA level, and repressed by sucrose at the translational level (Rook et al., 1998). The AtbZIP11 affects amino acids’ metabolism by regulating expression of ASPARAGINE SYNTHETASE1 (ASN1) and PROLINE DEHYDROGENASE2 (ProDH2) via binding to the G-box in their promoters (Hanson et al., 2008). In addition to carbohydrate balancing, S-group bZIPs are also involved in stress responses (Kusano et al., 1995). For example, AtbZIP1, 2, 11, and 53 specifically activate the G-box containing ASN1 promoter in a protein kinase KIN10-mediated energy/stress signaling cascade (BaenaGonzalez et al., 2007). Phylogenetic analysis of plant bZIPs reveals that both C- and S-group bZIPs share the same proto-C ancestors and participate in energy metabolism and oxidative stress responses (Correa et al., 2008). The C-group consists of four AtbZIPs that share similarities with maize Opaque2. Opaque2 regulates seed storage protein production via interaction with a Dof zinc finger prolamin-box binding factor (PBF) (Vicente-Carbajosa et al., 1997; Hwang et al., 2004). The S-group AtbZIP53 forms heterodimers with the C-group AtbZIP10 or AtbZIP25 in enhancing DNA binding activity and target gene activation. These heterodimers can form ternary complexes with VP1 class transcription factor ABI3 to further activate gene expression (Lara et al., 2003; Alonso et al., 2009). The combinatorial control of gene expression by AtbZIP heterodimerization has also been found in ProDH gene activation, in which the C- and S1-group AtbZIPs are involved (Weltmeier et al., 2006). We have identified 10 sugar-responsive AtbZIP factors in our transcriptome analysis (Price et al., 2004), of which AtbZIP1, AtbZIP11, and AtbZIP53 belong to the S1-group, and AtbZIP9 and AtbZIP63 belong to the C-group. Since AtbZIP1 has the most dramatic sugar response, we use it as a model to explore the roles of AtbZIPs in sugar signaling. We have found that, unlike traditional sugar marker genes with slow responses to high concentrations of sugar (Jang et al., 1997), glucose repression of AtbZIP1 is rapid and sensitive. This repression appears to be hexokinase-dependent. Results of reverse genetic analyses suggest that AtbZIP1 acts as a negative regulator for seedling growth when an exogenous carbon source is absent. A role of AtbZIP1 in sugar signaling is further supported by the identification of over-represented sugar-responsive genes in AtbZIP1 gain- and loss-of-function plants. Using both yeast two-hybrid analysis and bimolecular fluorescence complementation (BiFC), we demonstrate that C- and S1-group AtbZIPs form an extensive protein interacting network. Electrophoretic mobility shift assays (EMSA) indicate that AtbZIP1 can bind both C- and G-box motifs. Heterodimerization of AtbZIP1 with C-group AtbZIPs affects DNA-binding properties.

RESULTS Glucose Repression of AtbZIP1 Results of RNA gel blot analysis indicated that glucose repression of AtbZIP1 was sensitive (Figure 1A), rapid (Figure 1B),

Figure 1. Hexokinase-dependent glucose repression of AtbZIP1. Shown are results of RNA gel-blot analyses. (A) AtbZIP1 is repressed by low levels of glucose. (B) Glucose repression of AtbZIP1 is rapid. (C) Glucose repression can be reversed when glucose is removed from the culture medium in the dark condition (D-glc). Compare to the dark condition (D+glc), light (L+glc) appears to enhance glucose repression. Although cycloheximide (CHX) enhances AtbZIP1 expression, it does not block glucose repression. (D) AtbZIP1 is repressed by sugars that can be taken up by the cells and phosphorylated efficiently by hexokinase. (E) ABA alone or together with glucose does not affect glucose repression of AtbZIP1. (F) Glucose repression of AtbZIP1 is compromised in HXK1 knockout (gin2) or antisense-HXK1. Indicated by arrows are the incompletely repressed AtbZIP1 transcripts. Note that the overexpression and antisense HXK plants are in BE background; abi4-1 and aba2-1 are in Col-0 background, and gin2-1 (AtHXK1 KO) is in Ler background. All experiments use 110 mM glucose (+glc) unless indicated otherwise.

and reversible (Figure 1C). Cycloheximide (CHX) appeared to enhance transcript levels. However, it did not interfere with the glucose repression (Figure 1C), consistent with the

Kang et al.

results of global gene expression profiling that glucose repression of AtbZIP1 does not require de novo protein synthesis (Price et al., 2004). Light enhanced glucose repression (comparing L+glc to D+glc), possibly via additional endogenous sugars produced by photosynthesis. The expression of AtbZIP1 was likely affected by the circadian clock, as evidenced by a slight de-repression at 24-h time point (Figure 1B and 1C). Hexokinase plays an important role in sugar signaling in yeasts, humans, and plants (Rolland et al., 2001; Rolland and Sheen, 2005). We used sugar analogs to determine whether hexokinase was involved in sugar repression of AtbZIP1 (Figure 1D). Mannitol did not cause AtbZIP1 repression, ruling out the involvement of osmotic effects. A response to L-glucose would suggest that the repression could be mediated through a cell surface receptor, because L-glucose cannot be transported across cell membrane. By contrast, a response to D-glucose (glucose and mannose) but not glucose analog 3-OMG would suggest a role of hexokinase (HXK) in repression, because 3-OMG is a poor and D-glucose/mannose is an excellent substrate, respectively, for HXK (Xiao et al., 2000). The results showed that sugar transport and sugar phosphorylation were required for glucose repression of AtbZIP1, because L-glucose and 3-OMG failed to repress AtbZIP1 (Figure 1D). These results also support the notion that hexokinase is likely to be involved in the sugar repression of AtbZIP1. ABA plays a role in sugar signaling, and many sugar response mutants are also ABA biosynthetic or response mutants (Leo´n and Sheen, 2003; Rook and Bevan, 2003; Gibson, 2004, 2005). To test whether ABA is involved in sugar repression of AtbZIP1, we determined whether low levels of external glucose could repress AtbZIP1 expression, since our previous studies have shown that glucose below 0.5% (w/v) is unable to cause ABA accumulation (Price et al., 2003). Interestingly, external glucose as low as 0.1% (5.5 mM) could repress AtbZIP1 expression (Figure 1A), implicating that ABA might not play a role in this process. This was further evidenced by the results that when plants treated with exogenous ABA directly in the absence/presence of glucose, ABA had little effect on the glucose repression or de-repression of AtbZIP1 (Figure 1E). To further determine the roles of hexokinase and ABA in sugar repression of AtbZIP1, we employed several sugar response mutants (Jang et al., 1997; Arenas-Huertero et al., 2000; Cheng et al., 2002; Moore et al., 2003). Glucose repression of AtbZIP1 was less complete in hexokinase loss-offunction plants, including antisense-HXK1 and HXK1 knockout (KO) gin2 mutants, consistent with the idea that hexokinase is involved in this process (Figure 1F). In contrast, glucose repression remained effective in ABA signaling mutant abi4-1 (also known as glucose insensitive 6) (Arenas-Huertero et al., 2000) and ABA biosynthetic mutant aba2-1 (also known as glucose insensitive 1) (Cheng et al., 2002), suggesting that ABA is not directly involved in glucose repression of AtbZIP1.

d


|

363

AtbZIP1 Expression In Planta To determine the temporal, spatial, and sugar-regulated expression pattern of AtbZIP1, transgenic plants with a promoter–reporter fusion gene were characterized. In the absence of exogenous sugar, AtbZIP1 was expressed in both the shoot (devoid of cotyledons and hypocotyl) and root systems (Figure 2A). In the presence of either sucrose or glucose, AtbZIP1 expression was restricted to the vasculatures and roots at basal levels. To determine whether the expression of the reporter fusion gene was consistent with the endogenous gene, RNA gel-blot analyses were conducted. Glucose repression of endogenous AtbZIP1 was, as expected in the wild-type (WT), CaMV 35S:GUS control plants, and the two reporter lines (Figure 2B). In the absence of glucose, GUS expression was lower in the PbZIP1(1.3):GUS than in the PbZIP1(1.7):GUS plants, indicating an enhancer effect derived from a 0.4-kb region between 1.3 and 1.7 kb of the promoter. To identify potential elements that could affect promoter activities within the 0.4-kb region, a promoter scan program (Prestridge, 1991) using PLACE database (Higo et al., 1999) was conducted. Compared to the 1.3-kb region (1–1300), the 0.4-kb region contained a unique motif, TAACAAA, that could affect the promoter activity. The TAACAAA element is known as

Figure 2. Sugar Repression of AtbZIP1 in Transgenic Plants with a Promoter–Reporter Fusion Gene PAtbZIP1:GUS. (A) The expression of the reporter gene can be suppressed by exogenous sucrose or glucose. (B) Results of RNA gel-blot analyses showing that the endogenous AtbZIP1 is repressed by glucose normally in the WT and various reporter lines. Consistent with the endogenous AtbZIP1 gene, the PAtbZIP1:GUS reporter gene is repressed by 2% glucose.

364

|

Kang et al.

d


GA-responsive element (GARE) in Arabidopsis (Ogawa et al., 2003) and MYBGA binding site in cereals (Gubler et al., 1995, 1999). This element is important for both GA induction and sugar repression (Morita et al., 1998; Chen et al., 2006b). Missing the GARE/MYBGA could result in the loss of GA induction as well as sugar repression. Nevertheless, both reporter constructs still responded to glucose repression, likely due to the existence of other sugar-responsive elements, such as the pyrimidine box (Morita et al., 1998) in the 1.3-kb region. Constitutive expression of the GUS reporter gene in CaMV 35S:GUS plants indicated the specificity of glucose repression of AtbZIP1.

AtbZIP1 Is Involved in Glucose Response Because AtbZIP1 is a sugar-sensitive transcription factor, reverse genetic analyses were conducted to determine whether AtbZIP1 is involved in plant sugar response. There were no discernible phenotypic differences between WT and the gain- or loss-of-function AtbZIP1 plants in the presence of exogenous glucose (Figure 3A and 3B). The same results were observed when tested by using 4 or 6% glucose (not shown). Due to limited carbohydrate and oil reserve in Arabidopsis seeds, WT seedling developed poorly on MS plates without exogenous sugars. However, whereas the AtbZIP1 KO plants had higher rates of true leaf development, the AtbZIP1 overexpression (OX) plants had lower rates of true leaf development than the WT (Figure 3A and 3B). Results of RNA gel blot analysis indicated that Salk_059343 (Salk_069489 not shown) was indeed an AtbZIP1 KO, whereas ectopic expression using CaMV 35S promoter created ubiquitous expression and glucose insensitivity (Figure 3C). To further determine whether lossof-function AtbZIP1 plants had reduced requirement for exogenous sugar for seedling growth, the dominant-suppression plants (Ohta et al., 2001; Hiratsu et al., 2003, 2004) CaMV 35S:AtbZIP1-SRDX were used to test this hypothesis (Figure 3D). Consistent with the AtbZIP1 KO plants, in addition to higher rates of true leaf development (results not shown), CaMV 35S:AtbZIP1-SRDX seedlings grew faster than the WT on MS plates without exogenous sugars, as evidenced by larger shoots and root systems (Figure 3D). Since the differences in seedling establishment and growth between AtbZIP1 lossof-function and WT plants could be diminished by exogenous glucose, we propose that AtbZIP1 is a negative regulator of seedling growth, and sugar can de-repress this mechanism in the WT plants.

AtbZIP1 Affects Sugar-Mediated Gene Expression To determine whether AtbZIP1 could mediate the expression of sugar-responsive genes, an exploratory microarray analysis was conducted. We hypothesized that opposing phenotypes observed in AtbZIP1 KO and OX plants were due to changes of a similar set of genes. Because AtbZIP1 was highly repressed by sugars, we compared gene expression profiles between AtbZIP1 KO (Salk_059343) and WT in the absence of exogenous sugars. Under this condition, the AtbZIP1 was fully de-

Figure 3. AtbZIP1 Acts as a Negative Regulator of Early Seedling Growth in the Absence of Exogenous Glucose (2%). (A, B) Compared to the WT (Col-0), while the AtbZIP1 KO plants have higher rates, the overexpression (OX) plants have lower rates of true leaf development. (C) Results of RNA gel-blot analyses confirm the KO and overexpression of AtbZIP1 in KO and OX plants, respectively. (D) Dominant suppressing AtbZIP1–SRDX seedlings develop faster than the WT (Col-0) on sugar-free MS plates.

repressed in the WT, as evidenced by almost 20-fold higher expression in the WT than that in the KO plants (Supplemental Table 1). This set of comparisons was expected to reveal genes that could be activated or repressed by endogenous levels of AtbZIP1. The second set of analyses was conducted by comparing gene expression profiles between AtbZIP1 OX plants and the WT in the presence of exogenous glucose (2%). Under this condition, whereas the expression of the endogenous AtbZIP1 was severely repressed in the WT (Figures 1 and 2), it was unaffected in the OX plants, due to the ubiquitous and strong expression driven by the CaMV 35S promoter (Figure 3C). This was further confirmed by the results of microarray analysis in which AtbZIP1 was up-regulated nearly 150-fold in OX plants (Supplemental Table 2). This set of comparisons was expected

Kang et al.

to reveal the genes that could be activated or repressed by overexpression of AtbZIP1. Affymetrix Microarray Suite 5.0 was used for data processing and analysis (Price et al., 2004). Comparison analyses (experiment versus baseline arrays) were conducted to identify genes, with increase or decrease calls based on the Wilcoxon’s Signed Rank test. Using a cutoff of >1.5-fold, we identified 256 genes up-regulated in the KO, and 249 genes down-regulated in the OX plants (Supplemental Tables 2 and 3, and Figure 4A). These two pools of genes were designated as putative AtbZIP1-repressible genes. It was noteworthy that nearly 70% of these genes (174/256 and 170/249) were previously identified as glucoseresponsive genes, and 80–90% of them (161/174 and 140/ 170) were glucose-repressible (Price et al., 2004). The Vann diagram was then applied to identify common changes (Figure 4A). A 40% (98 genes) overlap was found between the two pools of (256 and 249) putative AtbZIP1-repressible genes (Supplemental Table 4). Once again, there were 70% (68/98) sugar-responsive genes and as high as 90% (61/68) of them were sugar-repressible (Supplemental Tables 1 and 5). The analysis also identified 260 genes down-regulated in KO plants and 221 genes up-regulated in OX plants (Supplemental Tables 6 and 7). Similar to putative AtbZIP1-repressible genes, about 60% of these genes (148/260 and 128/221) were glucoseresponsive, and about 85% of them (125/148 and 110/128) were glucose-inducible (Price et al., 2004). The Vann diagram identified 46 overlapping genes in these two pools of putative AtbZIP1-inducible genes (Supplemental Table 8 and Figure 4B).

Figure 4. AtbZIP1 can mediate the expression of sugar responsive genes. (A) Microarray analysis reveals 98 putative AtbZIP1 repressible genes by Vann Diagram selection of 256 genes up regulated in AtbZIP1 KO plants and 249 genes down regulated in AtbZIP1 OX plants. (B) Similar analysis was carried out to identify 46 putative AtbZIP1 inducible genes by comparing 260 down regulated genes in KO plants and 221 up regulated genes in OX plants. For the comparison of WT vs. KO, both received sugar-depletion treatment before RNA extraction. For the comparison of WT vs. OX, both were treated with 2% glucose before RNA extraction. Using a list of 2,348 glucose responsive genes identified from previous microarray analyses (Price et al., 2004), numbers of glucose responsive genes (in red) and the way these genes responding to glucose from each pool (in blue) are indicated.

d


|

365

Again, 30 out of 46 genes (65%) were sugar-responsive, and all 30 putative AtbZIP1-inducible genes (100%) were glucoseinducible (Supplemental Table 9 and Figure 4B). The lower percentage of overlap, 20% (46 genes), was presumably due to the requirement of other limiting factors and/or functional redundancy in the activation of so-called AtbZIP1-inducible genes. Nevertheless, the results suggested that similar genes were affected in KO and OX plants. AtbZIP1 is likely involved in sugar signaling because the putative AtbZIP1-repressible genes and the putative AtbZIP1-inducible genes are largely glucose-repressible (61 out of 68) and inducible (30 out of 30), respectively (Figure 4A and 4B).

AtbZIP1 Protein Interacting Network Enhanced transcriptional activation via dimerization is a hallmark of bZIP factors in all eukaryotes (Deppmann et al., 2006; Vinson et al., 2006), including Arabidopsis (Deppmann et al., 2004; Alonso et al., 2009). The yeast two-hybrid (Y-2-H) screen was conducted to identify AtbZIP1 interacting partners. Because transcription factors normally express at low levels, they may be under-represented in standard Y-2-H libraries. To overcome this potential problem, we used a transcription factor library in which all the clones were PCR-amplified to similar titers (Kim et al., 2005). Because the full-length AtbZIP1 protein (1–145 aa) caused self-activation of the reporter gene, deletion analysis was conducted to determine the region that could avoid self-activation activities. Since the potential protein–protein interaction domain (basic leucine zipper region) was located around the 20–70-aa region, a small N-terminal deletion (1–18 aa) was made, and it still caused self-activation (data not shown). Two C-terminal deletions outside the basic leucine zipper region (68–145 and 92–145 aa) were then tested. Both of them were free of self-activation activities (data not shown). Thus, the longer fragment of AtbZIP1 (1–91 aa) was used as a bait in the Y-2-H screen. We identified AtbZIP10 (At4g02640, BZO2H1), AtbZIP44 (At1g75390), and At1g24625 (C2H2 Zinc finger, ZFP7) as positive interactors. Since AtbZIP10 belongs to a small C-group, we tested the interaction between AtbZIP1 and other members of the C-group (AtbZIP9, AtbZIP25, and AtbZIP63). As both AtbZIP1 and AtbZIP44 belonged to the S1 subgroup, we tested whether AtbZIP1 could interact with another S1 member, AtbZIP11. Results showed that AtbZIP1 interacted strongly with AtbZIP10, AtbZIP25, and AtbZIP44, and interacted weakly with AtbZIP11 and AtbZIP63 in the Y-2-H assays (Figure 5A). To validate the results of Y-2-H screens and to confirm the protein–protein interaction in vivo, bimolecular fluorescence complementation (BiFC) (Hu et al., 2002; Walter et al., 2004) was conducted using both maize and Arabidopsis protoplast transient expression systems. In addition to the interactors identified in Y-2-H, another S1 member, AtbZIP53, was also included. While predictions based on electrostatic interactions indicated that both S1-group (AtbZIP11, AtbZIP44, and AtbZIP53) and C-group (AtbZIP9, AtbZIP25, and AtbZIP63) could form homodimers (Deppmann et al.,

366

|

Kang et al.

d


2006), homo-dimerization was found only for AtbZIP9, AtbZIP53, and AtbZIP63 in our BiFC analyses (Figure 5B and 5C). Moreover, the heterodimerization within the C-group was only found between AtbZIP63 and AtbZIP10 or between AtbZIP63 and AtbZIP25. Similarly, heterodimerization within the S1-group was limited to AtbZIP44 and AtbZIP1, or AtbZIP44 and AtbZIP53. Interestingly, the most consistent heterodimerizations were found between the C- and S1-groups. Except for the interactions with AtbZIP9, the four S1-group bZIPs interacted strongly with the C-group AtbZIP10, AtbZIP25, and AtbZIP63, respectively (Figure 5B and 5C). Whereas our results were consistent with a previous report (Ehlert et al., 2006), we have additionally identified the homodimerization of AtbZIP9, AtbZIP53, and AtbZIP63, as well as heterodimerization between AtbZIP1 and AtbZIP44 or AtbZIP63, respectively (Figure 5D). Results of the BiFC analyses appeared to be specific, as no signals could be detected from the negative controls, pA7-AtbZIP1-NYFP + empty vector pA7-CYFP or empty vector pA7-NYFP + pA7-AtbZIP1-CYFP (data not shown).

AtbZIP1 Binds Both C- and G-Box Motifs The ACGT-based motifs are bZIP consensus binding sites (Deppmann et al., 2006). The G-(CCACGTGG), C-(TGACGTCA), and ABRE-(PyACGTGGC)-box are some of the conserved plant bZIP binding sites (Menkens et al., 1995; Uno et al., 2000; Lara et al., 2003; Song et al., 2008). To determine whether AtbZIP1 could bind ACGT-based motifs, we tested the following three different DNA probes: the Hex (hybrid C/G-box) motif CTGACGTGGC, a C-box motif TGCTGACGTCA, and a G-box motif CCACGTGGCC using EMSA. Intriguingly, while neither AtbZIP1 nor AtbZIP10 could form homodimers (Figure 5B and 5C), AtbZIP1 could bind all three probes (Figure 6A and 6B), presumably as a monomer form. Although AtbZIP10 alone did not bind any of the three probes, it caused a super-shift of the AtbZIP1–DNA complexes at higher concentrations. This implicated an interaction between AtbZIP10 and AtbZIP1–DNA complexes (Figure 6A). By contrast, while AtbZIP63 did not cause any super-shift, it diminished the levels of AtbZIP1– DNA complexes at the concentration of 3 pmol (Figure 6B). This suggests that the AtbZIP1–AtbZIP63 heterodimers have

Figure 5. The AtbZIP1 interacting network.

(A) Yeast two-hybrid screen identifies AtbZIP11, AtbZIP44, AtbZIP10, AtbZIP25, and AtbZIP63 as AtbZIP1 interacting partners. Strength of each interaction is determined by colony growth on normal (-Leu-Trp-His) and high (-Leu-Trp-His-Ade) stringency selection plates, and as indicated by line thickness in the model. (B-C) Results of bimolecular fluorescence complementation (BiFC) analyses indicate that AtbZIP1 interacts extensively with both S-group (AtbZIP44, AtbZIP53) and C-group (AtbZIP9, AtbZIP10, AtbZIP25, AtbZIP63) AtbZIPs in maize (B) and Arabidopsis (C) transient expression system. Symbols of cross-lined circles indicate negative interactions. (D) Model depicting the AtbZIP1 interacting network based on the results of Y-2-H and BiFC. Strength of interaction is indicated by the line thickness. Red lines denote new interactions not being reported previously. Dotted lines indicate interactions found in Ehlert et al. 2006, but not in current study.

Kang et al.

d


|

367

negative effects for binding to these three probes. To determine the DNA binding specificity, we compared the binding between the native and mutant Hex probes (Schindler et al., 1992). The DNA–protein interactions appeared to be specific because no complexes formed when a mutant Hex probe was used (Figure 6C). Some of the negative DNA–protein interactions could be due to unequimolar ratio of the proteins or insufficient length of the DNA probe. To test these possibilities, we performed EMSA using equimolar amounts of proteins and longer probes named strong C- and G-box (Izawa et al., 1993; Foster et al., 1994), respectively. Results showed that only AtbZIP1 at 3 pmol could bind to strong C- and G-box, respectively (Figure 6D). The double bands suggested that there were two forms of DNA–protein complexes, presumably due to extended length of the DNA probes. The effects of AtbZIP10 and AtbZIP63 were also determined in these new binding assays. Unlike earlier results (Figure 6A and 6B), AtbZIP63 seemed to enhance the complex formation (Figure 6E). The AtbZIP10 once again induced super-shift of the DNA–AtbZIP1 complex (Figure 6E), consistent with earlier results using shorter DNA probes (Figure 6A). Together, these results indicate that AtbZIP1 can bind the ACGT-based DNA motif in vitro. While C-group AtbZIP10 or 63 alone could not bind the DNA probes tested, they affected the characteristics of AtbZIP1–DNA interactions, consistent with the notion that protein–protein interaction can influence DNA–protein interaction.

DISCUSSION AtbZIP1 in Sugar Signaling The bZIP transcription factors are involved in a diverse array of functions in plant growth, development, and environmental responses. The Group-S AtbZIPs are generally involved in sugar and stress signaling (Jakoby et al., 2002). The subgroup S1, AtbZIP1, AtbZIP2, AtbZIP11, AtbZIP44, and AtbZIP53 are translationally repressed by sucrose signaling. This repression is dependent on the translation of a conserved upstream open reading frame (uORF) in the 5’ leader of their mRNAs (Wiese et al., 2004, 2005; Hummel et al., 2009; Weltmeier et al., 2009). Interestingly, their transcriptional responses to sugars are variable: while AtbZIP11 is sugar-inducible, AtbZIP1, AtbZIP2, and AtbZIP53 are sugar-repressible (Price et al., 2004). In this report, we have found that the expression of AtbZIP1 is highly sensitive to sugars. The glucose repression of AtbZIP1 is rapid

Figure 6. AtbZIP1 can bind ACGT motif-based C- or G-box.

(A-B) EMSA results showing that AtbZIP1 can bind Hex, C-box, and G-box motif, respectively. The AtbZIP10 (A) and AtbZIP63 (B) can interfere with the binding as evidenced by the super-shift and reduction of the DNA-AtbZIP1 complexes, respectively. (C) AtbZIP1 binding to the Hex motif is specific, as DNA-protein complex cannot be formed with mutant probe. (D) AtbZIP1, but not AtbZIP10 or 25 can bind strong C- or G-box. The maltose binding protein is used as a background negative control. (E) AtbZIP 63 seems to enhance DNA-AtbZIP1 complex formation, and AtbZIP1 causes a super-shift of the complex. Signals in the bottom of each gel are generated by free DNA probes.

368

|

Kang et al.

d


and reversible. While de novo protein synthesis is not required, the repression is affected by hexokinase, as poor hexokinase substrates (Xiao et al., 2000) and hexokinase KO (Moore et al., 2003) and antisense (Jang and Sheen, 1997) mutants are unable to repress AtbZIP1 normally. Currently, it is unknown how hexokinase affects glucose repression of AtbZIP1, although it is likely mediated through hexokinase-dependent signaling processes (Rolland et al., 2006). We have used reverse genetic analyses to determine whether sugar-regulated AtbZIP1 could be involved in sugar signaling. Our results indicate that AtbZIP1 acts as a negative regulator of early seedling growth in the absence of exogenous sugars in the culture medium. This regulation has an important scenario, because sugar repression of AtbZIP1, transcriptionally and/or post-transcriptionally, would alleviate its negative effects on growth and development. Therefore, WT Arabidopsis seedlings grow much better on MS plates in the presence of exogenous sugars. On one hand, this may be due to global metabolic effects of carbon nutrition; on the other hand, it might be due to signaling processes, such as a repression on the negative regulator AtbZIP1. The latter proposition is supported by the fact that there are no discernible differences on growth between the WT and AtbZIP1 OX or KO plants when grown on soil in greenhouse, where all types of plants are fully photoautotrophic and sugars can be produced via photosynthesis. In contrast to AtbZIP1, AtbZIP11 expression is up-regulated by sugars (Price et al., 2004; Thum et al., 2004; Blasing et al., 2005; Li et al., 2006). It has been shown recently that AtbZIP11 specifically activates ASN1 and ProDH2 via binding to G-box elements in their promoters (Baena-Gonzalez et al., 2007; Hanson et al., 2008). ASN1 and ProDH2 encode enzymes that are critical for nitrogen metabolism. Based on this scenario, ASN1 and ProDH2 are presumably up-regulated by sugars. However, because AtbZIP11’s translation is repressed by sucrose via a uORF in the 5’ leader of the mRNA (Rook et al., 1998), ASN1 and ProDH2 are ultimately repressed by sugars. These intricate mechanisms are biologically important, because changing the expression levels of AtbZIP11 can cause pronounced effects on cellular amino acids levels, indicating its role in integrating sugar signals in balancing nitrogen metabolism (Hanson et al., 2008). Similar to what we have reported here for AtbZIP1, overexpression of AtbZIP11 also causes growth defects (Hanson et al., 2008). However, whereas a high percentage of AtbZIP11inducible genes are sugar-repressible, most of the AtbZIP1associated gene expression changes are consistent with their sugar responses (Figure 4A and 4B). That means putative AtbZIP1-inducible genes are sugar-inducible, and AtbZIP1repressible genes are sugar-repressible. While work is in progress to reveal the mechanisms AtbZIP1 uses to regulate its potential target genes, there are multiple possibilities for the discrepancy. One likely reason is that AtbZIP1 is highly sugar-repressible; although there is a sugar-mediated translational repression (Weltmeier et al., 2009), the target gene expression would most likely have the same trends as AtbZIP1’s

own expression, namely target gene expression will be down if AtbZIP1 is repressed by sugars, transcriptionally and/or translationally. Here, we have attempted to identify AtbZIP1regulated genes using a simple and effective microarray analysis. By comparing AtbZIP1 KO and WT plants in the absence of sugars, the effects of AtbZIP1 in gene expression are maximized due to the full de-repression of AtbZIP1 in the WT and the null background of the KO plants. On the other hand, we compared the gene expression profiles between AtbZIP1 OX and WT in the presence of 2% glucose. Although it is an artificial condition, a nearly 150-fold difference in AtbZIP1 expression level has been created, thus maximizing the opportunities to reveal AtbZIP1 target genes. Results of these analyses suggest that AtbZIP1 is indeed involved in sugar-mediated gene expression, because: (1) there is a fair number of overlap (40 and 20%) between the two sets of comparisons; (2) for the genes identified as AtbZIP1-regulated, a high percentage (60– 70%) of them are sugar-responsive; and (3) the sugar response, induction, or repression, from each pool of the genes, is consistent with the predicted action of AtbZIP1 in regulating these potential target genes. However, it is conceivable that some of these genes are indirect targets of AtbZIP1. Work is in progress to globally identify AtbZIP1’s direct targets.

AtbZIP Dimerization Network The dimerization of bZIPs has been used a paradigm to study structure–function relationships in multiple biological systems. The bZIP interacting network has been established from humans by using the protein arrays (Newman and Keating, 2003), to Arabidopsis, primarily based on computational analysis using predicted interface characteristics, electrostatic properties, and coupling energy (Deppmann et al., 2004, 2006; Vinson et al., 2006). Although the predicted models were comprehensive, it was not until recently that high throughput approaches were used to experimentally validate the Arabidopsis bZIP interacting network (Ehlert et al., 2006). Using both conventional yeast two-hybrid (Y-2-H) and a novel Arabidopsis protoplast two-hybrid (P-2-H) system, it was found that S-group preferentially interacted with the C-group AtbZIPs and there was almost no homodimerization and only a few weak interactions within the C- or S1-group (Ehlert et al., 2006). In mammalian systems, bimolecular fluorescence complementation (BiFC) was first used to visualize the interactions between bZIP family proteins, Jun and Fos, in COS-1, NIH3T3, and Hela living cells (Hu et al., 2002). This approach has since been widely used in plant systems to determine protein– protein interaction in sub-cellular domains in vivo (BrachaDrori et al., 2004; Walter et al., 2004; Weinthal and Tzfira, 2009). Taking advantage of the established high throughput protoplast transient expression systems (Sheen, 2001; Yoo et al., 2007), we have mapped the AtbZIP1 interacting network using BiFC. Our results are not only in agreement with Ehlert et al. (2006), but also revealing additional interactions that have not been reported previously. For instance, we have found homodimerization of AtbZIP53 and AtbZIP63,

Kang et al.

respectively, at moderate levels and of AtbZIP9 at a low level. For interactions within each group, AtbZIP1+AtbZIP44, AtbZIP1+AtbZIP53, and AtbZIP25+AtbZIP9 have been revealed. For the heterodimerization between C- and S-group, we have found that AtbZIP1 interacts strongly with AtbZIP63 in both Arabidopsis and maize systems (Figure 5D). This newly identified interaction is likely to have impact on DNA binding and regulation of gene expression, because our EMSA results indicate an effect of AtbZIP63 on DNA–AtbZIP1 complex formation (Figure 6B, 6C, and 6E). The discrepancies between these studies are likely due to the intrinsic sensitivity associated with each method. The results can also be affected by the cells and reagents used. Nevertheless, the combined results suggest that all possible interactions exist between tested C- and S1-group bZIPs. There are moderate levels of heterodimerization, and low levels of homodimerization within each group.

AtbZIP1 Dimerization and DNA-Binding Heterodimerization of AtbZIP10 or AtbZIP25 with a non-bZIP factor, ABI3, can synergistically activate a seed storage protein gene At2S1, via interactions with the ACGT-boxes in the promoter (Lara et al., 2003). However, homo- and heterodimeric bZIPs recognition of ACGT cores has long been demonstrated in higher plants (Armstrong et al., 1992). In fact, AtbZIPs preferentially form heterodimers amongst themselves, and dimerization can affect their abilities in the modulation of gene expression (Deppmann et al., 2004; Ehlert et al., 2006; Weltmeier et al., 2006; Alonso et al., 2009; Weltmeier et al., 2009). While it is possible that bZIP monomers can bind DNA (Metallo and Schepartz, 1997; Weltmeier et al., 2006), it is unclear whether or not they have functions similar to the bZIP dimmers. A recent report has elegantly demonstrated that AtbZIP53 binds a G-box motif in vitro and in vivo. Heterodimerization of AtbZIP53 with AtbZIP10 or 25 enhances DNA binding, and synergistically increases target gene activation. These heterodimers can interact with ABI3 to form ternary complexes that further enhance the expression of target genes (Alonso et al., 2009). Although AtbZIP1 belongs to the C/S1 bZIP transcription factor network, it is unknown whether or not it can bind ACGTbased DNA motif. Using EMSAs, we have found that AtbZIP1 can bind C-box, G-box, and C/G hybrid Hex-box motifs. The DNA–AtbZIP1 interactions appear to be specific because no signals can be detected when mutant DNA probes or MBP are used. It is noteworthy that AtbZIP1 can consistently bind various ACGT-based motifs in our EMSAs. The inclusion of AtbZIP10 or AtbZIP63 in the binding reactions causes changes in DNA–AtbZIP1 characteristics. For example, AtbZIP10 causes super-shifts of DNA–AtbZIP1 complexes in a concentrationdependent manner (Figure 6A and 6E). It is noticeable that at higher concentrations, both AtbZIP10 and AtbZIP63 could reduce the DNA–protein complex formation when C-box, G-box, or hex motif was used (Figure 6A–6C), probably due to the conformational differences between AtbZIP1 homodimer/monomer and AtbZIP1–AtbZIP10 or AtbZIP1–AtbZIP63

d


|

369

heterodimers. By contrast, both AtbZIP10 and AtbZIP63 seem to enhance the DNA–AtbZIP1 complex formation, presumably due to the longer ‘Strong’ C- and G-box motifs that were used. In our EMSAs, no DNA–protein complexes were detected when AtbZIP10 or AtbZIP63 was used alone in the reactions. This appears to be inconsistent with previous reports indicating that AtbZIP10, AtbZIP25, and AtbZIP63 can bind both C- and G-box (Lara et al., 2003; Kaminaka et al., 2006). The discrepancies can be attributed to the variations of the DNA probes used in each assay system. For example, while we use 5#-CCACGTGCC-3’ as G-box probe, 5#-TTCTTACACGTGAACTC-3’ and 5#-TAATCCTACGTGGCTCA-3’ have been used in Lara et al. (2003). It is well documented that the flanking sequences of the bZIP binding sites play pivotal roles in DNA–bZIP interactions (Foster et al., 1994), as well as in structural and functional properties of regulatory protein complexes (Ramirez-Carrozzi and Kerppola, 2001). Other possibilities include the amount of protein and reaction conditions used in the assays. In summary, we have shown here that the expression of AtbZIP1 is highly regulated by sugars. The sugar repression of AtbZIP1 is affected by a conserved sugar signaling component, hexokinase (Jang et al., 1997; Moore et al., 2003; Rolland et al., 2006). Besides being a sugar-regulated gene, AtbZIP1 can mediate sugar signaling and control gene expression, plant growth, and development. When carbon nutrients are limited, gain or loss of function of AtbZIP1 causes profound changes in early seedling establishment. Results of phenotypic analyses indicate that AtbZIP1 acts as a negative regulator of early seedling growth. Using gain- and loss-of-function plants in a microarray analysis, two sets of putative AtbZIP1 target genes have been identified. Sugar-responsive genes are highly over-represented in these potential target genes, implicating a role of AtbZIP1 in sugar-regulated gene expression. Using BiFC, we are able to recapitulate extensive C/S1 AtbZIP protein interacting network in living cells. Finally, we show that AtbZIP1 can bind ACGF-based motifs in vitro, and that the binding characteristics are affected by the heterodimerization of AtbZIP1 and the C-group AtbZIPs.

METHODS Molecular Cloning and RNA Gel-Blot Analysis RNA extraction and RNA gel-blot analyses were conducted as described (Price et al., 2004). Seven-day-old light-grown seedlings were dark-adapted for 24 h in sugar-free MS liquid medium before applications of sugar/hormone as specified. The full-length cDNA cloned in pBluescriptKS(+) was labeled with a 32P dATP using gene-specific primers by PCR reactions. The binary construct of PAtbZIP1:GUS, CaMV 35S:AtbZIP1-GFP, and CaMV 35S:AtbZIP1-SRDX were made using the Gateway (Invitrogen) vectors listed in http://www.psb.ugent.be/gateway/ index.php. The maltose binding protein-tagged AtbZIP1 construct (MBP–AtbZIP1) was made by inserting the AtbZIP1 fulllength cDNA into pMAL–c2X (N8076) vector (New England Biolabs). His(6X)–AtbZIPs constructs were made using the

370

|

Kang et al.

d


Univector Plasmid Fusion (UPF) System as described (http:// signal.salk.edu/UPS_Protocols_Vectors.pdf). For BiFC analyses, each full-length cDNA was inserted into pA7–NYFP and pA7–CYFP vector (Chen et al., 2006a), respectively.

Mutant and Transgenic Plants The AtbZIP1 KO plants were obtained from the Arabidopsis Biological Resource Center (ABRC, Columbus, Ohio). The two independent KO lines, Salk_059343 and Salk_069489C, were confirmed by RNA gel blot analyses. The CaMV 35S:AtbZIP1– GFP plants were generated using Agrobacterium-mediated transformation by infiltration methods (Bechtold and Pelletier, 1998).

Microarray Analysis Microarray analysis was conducted essentially as described (Price et al., 2004). Since the goal of the microarray analysis was to determine whether AtbZIP1, under endogenous and overexpression high levels, could affect the expression of sugar-responsive genes, a single biological repeat was used for this exploratory purpose. Due to the use of Affymetrix platform, technical repeats (e.g. dye swapping) could be avoided. Six-day-old seedlings were grown in liquid MS medium with 2% sucrose on a platform shaker at 120 rpm, at 24C under constant fluorescent white light (80 lmol m 2 s 1). Seedlings were then washed five times with sterile distilled water and incubated in MS liquid without sugar in the dark for 24 h for sugar depletion. A mock solution or glucose (final 2%) was then added to the culture for 3 h before samples were collected. RNA extraction was carried out using qiagen plant RNA isolation kits. The RNA quality control, quantification, labeling, Affymetrix GeneChip processing, array data processing, and analyses were as described (Price et al., 2004). Affymetrix Microarray Suite 5.0 was used for data processing and analysis. Comparison analyses (experiment versus baseline arrays) were conducted to identify genes with increase or decrease calls based on the Wilcoxon’s Signed Rank test. The Venn diagram (http://bbc.botany.utoronto.ca/ntools/cgi-bin/ ntools_venn_selector.cgi) was then applied to identify common changes in all comparisons.

Yeast Two-Hybrid Screen and Bimolecular Fluorescence Complementation (BiFC) The Gateway system (Invitrogen) was used to produce constructs for yeast two-hybrid screen. Yeast transformations were performed using Saccharomyces cerevisiae strain PJ69-4A (James et al., 1996). Because the full-length AtbZIP1 exerted self-activating activities on the reporter gene, the non-selfactivating N-terminal fragment (1–91 aa) of AtbZIP1 was cloned into pDEST32 vector, containing the yeast GAL4 DNA binding domain, as a bait. The Y-2-H library screen was conducted using a synthetic Arabidopsis transcription factor library (Welchen et al., 2009) in the pDEST22 vector, containing GAL4 activation domain, as a prey. Positive co-transformation was selected on leucine and tryptophan dropout plates for pu-

tative interaction. To verify the interactions identified in initial screens, individual candidate cDNAs were re-transformed with the AtbZIP1, swapping between the bait and prey vector and the growth assay was repeated. A synthetic complete medium lacking histidine and adenine was used to determine protein– protein interactions. AtbZIP10 (At1g07350) and AtbZIP44 (At4g02640) were identified from the library screen. Genes in the same family, including AtbZIP11, AtbZIP53, AtbZIP9, AtbZIP25, and AtbZIP63, were further tested using candidate interaction approach by individual co-expression with the bait AtbZIP1. The BiFC analyses were conducted using protoplast transient expression system. Candidate interacting proteins were cloned reciprocally in pA7–NYFP and pA7–CYFP vectors (Chen et al., 2006a). The pair-wise constructs were then cotransformed into protoplasts. Positive interaction was determined by the YFP signals displayed in multiple cells using fluorescence microscopy. A Nikon Eclipse E600 fluorescence microscope with filter sets was used for microscopy. YFP fluorescence was visualized with a filter set consisting of an excitation filter of 450–490 nm, a dichroic mirror of 510 nm, and a barrier filter of 520–560 nm. Images were captured by a SPOT RT Slider multi-mode camera and Advanced SPOT Software (Diagnostics Instruments, Sterling Heights, MI). For protoplasts, 10-ll samples were loaded to hemacytometer for observation.

Electrophoretic Mobility Gel-Shift Assay (EMSA) The EMSA was conducted essentially as described (Harter et al., 1994). The recombinant proteins, including MBP–AtbZIP1, His(6X)–AtbZIP10, His(6X)–AtbZIP25, and His(6X)–AtbZIP63, were produced in E. coli and purified using standard procedures. DNA probes were labeled with c32P ATP using DNA polynucleotide kinase (New England Biolabs). DNA–protein binding reaction was conducted in a buffer containing 7 mM HEPES, 3 mM Tris pH 7.85, 40 mM KCl, 0.6 mM EDTA, 0.6 mM DTT, 7.5% glycerol, 0.8 lg poly (dIdC), and 1.5 mM spermidine. The binding reaction was conducted on ice with described amount of protein and DNA probe (10 000 cpm) for 30 min and then separated on 4% native PAGE gel before drying and exposing to the X-ray film.

SUPPLEMENTARY DATA Supplementary Data are available at Molecular Plant Online.

FUNDING This work was supported by The National Science Foundation (IOB0543751 to J.C.J.).

ACKNOWLEDGMENTS We thank the Arabidopsis Biological Resource Center (Columbus, Ohio) for providing DNA clones and seeds, Dr Biao Ding for

Kang et al.

microscopy facility, Dr Steven St Martin for microarray design and data analysis, Cyrus Hah for protoplast transient expression analysis, Drs John Finer and Michelle Jones for critical reading of the manuscript, and Joe Takayama for greenhouse support. No conflict of interest declared.

REFERENCES Alonso, R., Onate-Sanchez, L., Weltmeier, F., Ehlert, A., Diaz, I., Dietrich, K., Vicente-Carbajosa, J., and Droge-Laser, W. (2009). A pivotal role of the basic leucine zipper transcription factor bZIP53 in the regulation of Arabidopsis seed maturation gene expression based on heterodimerization and protein complex formation. Plant Cell. Arenas-Huertero, F., Arroyo, A., Zhou, L., Sheen, J., and Leon, P. (2000). Analysis of Arabidopsis glucose insensitive mutants, gin5 and gin6, reveals a central role of the plant hormone ABA in the regulation of plant vegetative development by sugar. Genes Dev. 14, 2085–2096. Armstrong, G.A., Weisshaar, B., and Hahlbrock, K. (1992). Homodimeric and heterodimeric leucine zipper proteins and nuclear factors from parsley recognize diverse promoter elements with ACGT cores. Plant Cell. 4, 525–537. Baena-Gonzalez, E., and Sheen, J. (2008). Convergent energy and stress signaling. Trends Plant Sci. 13, 474–482. Baena-Gonzalez, E., Rolland, F., Thevelein, J.M., and Sheen, J. (2007). A central integrator of transcription networks in plant stress and energy signalling. Nature. 448, 938–942. Bechtold, N., and Pelletier, G. (1998). In planta Agrobacterium-mediated transformation of adult Arabidopsis thaliana plants by vacuum infiltration. Methods Mol. Biol. 82, 259–266. Blasing, O.E., Gibon, Y., Gunther, M., Hohne, M., Morcuende, R., Osuna, D., Thimm, O., Usadel, B., Scheible, W.R., and Stitt, M. (2005). Sugars and circadian regulation make major contributions to the global regulation of diurnal gene expression in Arabidopsis. Plant Cell. 17, 3257–3281. Bracha-Drori, K., Shichrur, K., Katz, A., Oliva, M., Angelovici, R., Yalovsky, S., and Ohad, N. (2004). Detection of protein–protein interactions in plants using bimolecular fluorescence complementation. Plant J. 40, 419–427. Chen, S., Tao, L., Zheng, L.-R., Vega-Sanchez, M., and Wang, G.L. (2006a). A highly efficient transient protoplast system for analyzing defense gene expression and protein–protein interactions in rice. Mol. Plant Pathol. 7, 417–427. Chen, P.W., Chiang, C.M., Tseng, T.H., and Yu, S.M. (2006b). Interaction between rice MYBGA and the gibberellin response element controls tissue-specific sugar sensitivity of alpha-amylase genes. Plant Cell. 18, 2326–2340. Cheng, W.H., et al. (2002). A unique short-chain dehydrogenase/reductase in Arabidopsis glucose signaling and abscisic acid biosynthesis and functions. Plant Cell. 14, 2723–2743. Correa, L.G., Riano-Pachon, D.M., Schrago, C.G., dos Santos, R.V., Mueller-Roeber, B., and Vincentz, M. (2008). The role of bZIP transcription factors in green plant evolution: adaptive features emerging from four founder genes. PLoS One. 3, e2944. Deppmann, C.D., Acharya, A., Rishi, V., Wobbes, B., Smeekens, S., Taparowsky, E.J., and Vinson, C. (2004). Dimerization specificity

d


|

371

of all 67 B-ZIP motifs in Arabidopsis thaliana: a comparison to Homo sapiens B-ZIP motifs. Nucleic Acids Res. 32, 3435–3445. Deppmann, C.D., Alvania, R.S., and Taparowsky, E.J. (2006). Crossspecies annotation of basic leucine zipper factor interactions: insight into the evolution of closed interaction networks. Mol. Biol. Evol. 23, 1480–1492. Ehlert, A., Weltmeier, F., Wang, X., Mayer, C.S., Smeekens, S., Vicente-Carbajosa, J., and Droge-Laser, W. (2006). Two-hybrid protein–protein interaction analysis in Arabidopsis protoplasts: establishment of a heterodimerization map of group C and group S bZIP transcription factors. Plant J. 46, 890–900. Foster, R., Izawa, T., and Chua, N.H. (1994). Plant bZIP proteins gather at ACGT elements. FASEB J. 8, 192–200. Gibson, S.I. (2004). Sugar and phytohormone response pathways: navigating a signalling network. J. Exp. Bot. 55, 253–264. Gibson, S.I. (2005). Control of plant development and gene expression by sugar signaling. Curr. Opin. Plant Biol. 8, 93–102. Gubler, F., Kalla, R., Roberts, J.K., and Jacobsen, J.V. (1995). Gibberellin-regulated expression of a myb gene in barley aleurone cells: evidence for Myb transactivation of a high-pI alpha-amylase gene promoter. Plant Cell. 7, 1879–1891. Gubler, F., Raventos, D., Keys, M., Watts, R., Mundy, J., and Jacobsen, J.V. (1999). Target genes and regulatory domains of the GAMYB transcriptional activator in cereal aleurone. Plant J. 17, 1–9. Guo, A., He, K., Liu, D., Bai, S., Gu, X., Wei, L., and Luo, J. (2005). DATF: a database of Arabidopsis transcription factors. Bioinformatics. 21, 2568–2569. Hanson, J., Hanssen, M., Wiese, A., Hendriks, M.M., and Smeekens, S. (2008). The sucrose regulated transcription factor bZIP11 affects amino acid metabolism by regulating the expression of ASPARAGINE SYNTHETASE1 and PROLINE DEHYDROGENASE2. Plant J. 53, 935–949. Harter, K., Kircher, S., Frohnmeyer, H., Krenz, M., Nagy, F., and Schafer, E. (1994). Light-regulated modification and nuclear translocation of cytosolic G-box binding factors in parsley. Plant Cell. 6, 545–559. Higo, K., Ugawa, Y., Iwamoto, M., and Korenaga, T. (1999). Plant cisacting regulatory DNA elements (PLACE) database: 1999. Nucleic Acids Res. 27, 297–300. Hiratsu, K., Matsui, K., Koyama, T., and Ohme-Takagi, M. (2003). Dominant repression of target genes by chimeric repressors that include the EAR motif, a repression domain, in Arabidopsis. Plant J. 34, 733–739. Hiratsu, K., Mitsuda, N., Matsui, K., and Ohme-Takagi, M. (2004). Identification of the minimal repression domain of SUPERMAN shows that the DLELRL hexapeptide is both necessary and sufficient for repression of transcription in Arabidopsis. Biochem. Biophys. Res. Commun. 321, 172–178. Hu, C.D., Chinenov, Y., and Kerppola, T.K. (2002). Visualization of interactions among bZIP and Rel family proteins in living cells using bimolecular fluorescence complementation. Mol. Cell. 9, 789–798. Hummel, M., Rahmani, F., Smeekens, S., and Hanson, J. (2009). Sucrose-mediated translational control. Ann. Bot. (Lond.). 104, 1–7. Hwang, Y.S., Ciceri, P., Parsons, R.L., Moose, S.P., Schmidt, R.J., and Huang, N. (2004). The maize O2 and PBF proteins act additively

372

|

Kang et al.

d


to promote transcription from storage protein gene promoters in rice endosperm cells. Plant Cell Physiol. 45, 1509–1518. Izawa, T., Foster, R., and Chua, N.H. (1993). Plant bZIP protein DNA binding specificity. J. Mol. Biol. 230, 1131–1144. Jakoby, M., Weisshaar, B., Droge-Laser, W., Vicente-Carbajosa, J., Tiedemann, J., Kroj, T., and Parcy, F. (2002). bZIP transcription factors in Arabidopsis. Trends Plant Sci. 7, 106–111. James, P., Halladay, J., and Craig, E.A. (1996). Genomic libraries and a host strain designed for highly efficient two-hybrid selection in yeast. Genetics. 144, 1425–1436. Jang, J.-C., and Sheen, J. (1997). Sugar sensing in higher plants. Trends Plant Sci. 2, 208–214. Jang, J.C., Leon, P., Zhou, L., and Sheen, J. (1997). Hexokinase as a sugar sensor in higher plants. Plant Cell. 9, 5–19. Kaminaka, H., et al. (2006). bZIP10-LSD1 antagonism modulates basal defense and cell death in Arabidopsis following infection. EMBO J. 25, 4400–4411. Kim, H.J., Shin, S.Y., Song, Y.H., Kang, J.H., Kim, H.S., Song, N.Y., Son, G.H., and Hong, J.C. (2005). Construction of Arabidopsistranscription factor ORFeome for genome wide study of protein-protein interactions of plant transcription factors. Abstract #25416th International Conference on Arabidopsis Research (Madison: Wisconsin). Kim, T.H., Kim, B.H., Yahalom, A., Chamovitz, D.A., and von Arnim, A.G. (2004). Translational regulation via 5’ mRNA leader sequences revealed by mutational analysis of the Arabidopsis translation initiation factor subunit eIF3h. Plant Cell. 16, 3341–3356. Kusano, T., Berberich, T., Harada, M., Suzuki, N., and Sugawara, K. (1995). A maize DNA-binding factor with a bZIP motif is induced by low temperature. Mol. Gen. Genet. 248, 507–517. Lara, P., Onate-Sanchez, L., Abraham, Z., Ferrandiz, C., Diaz, I., Carbonero, P., and Vicente-Carbajosa, J. (2003). Synergistic activation of seed storage protein gene expression in Arabidopsis by ABI3 and two bZIPs related to OPAQUE2. J. Biol. Chem. 278, 21003–21011. Leo´n, P., and Sheen, J. (2003). Sugar and hormone connections. Trends Plant Sci. 8, 110–116. Li, Y., Lee, K.K., Walsh, S., Smith, C., Hadingham, S., Sorefan, K., Cawley, G., and Bevan, M.W. (2006). Establishing glucose- and ABA-regulated transcription networks in Arabidopsis by microarray analysis and promoter classification using a Relevance Vector Machine. Genome Res. 16, 414–427. Menkens, A.E., Schindler, U., and Cashmore, A.R. (1995). The G-box: a ubiquitous regulatory DNA element in plants bound by the GBF family of bZIP proteins. Trends Biochem. Sci. 20, 506–510. Metallo, S.J., and Schepartz, A. (1997). Certain bZIP peptides bind DNA sequentially as monomers and dimerize on the DNA. Nat. Struct. Biol. 4, 115–117. Moore, B., Zhou, L., Rolland, F., Hall, Q., Cheng, W.H., Liu, Y.X., Hwang, I., Jones, T., and Sheen, J. (2003). Role of the Arabidopsis glucose sensor HXK1 in nutrient, light, and hormonal signaling. Science. 300, 332–336. Morita, A., Umemura, T., Kuroyanagi, M., Futsuhara, Y., Perata, P., and Yamaguchi, J. (1998). Functional dissection of a sugarrepressed alpha-amylase gene (RAmy1 A) promoter in rice embryos. FEBS Lett. 423, 81–85.

Newman, J.R., and Keating, A.E. (2003). Comprehensive identification of human bZIP interactions with coiled-coil arrays. Science. 300, 2097–2101. Ogawa, M., Hanada, A., Yamauchi, Y., Kuwahara, A., Kamiya, Y., and Yamaguchi, S. (2003). Gibberellin biosynthesis and response during Arabidopsis seed germination. Plant Cell. 15, 1591–1604. Ohta, M., Matsui, K., Hiratsu, K., Shinshi, H., and Ohme-Takagi, M. (2001). Repression domains of class II ERF transcriptional repressors share an essential motif for active repression. Plant Cell. 13, 1959–1968. Prestridge, D.S. (1991). SIGNAL SCAN: a computer program that scans DNA sequences for eukaryotic transcriptional elements. Comput. Appl. Biosci. 7, 203–206. Price, J., Laxmi, A., St Martin, S.K., and Jang, J.C. (2004). Global transcription profiling reveals multiple sugar signal transduction mechanisms in Arabidopsis. Plant Cell. 16, 2128–2150. Price, J., Li, T.C., Kang, S.G., Na, J.K., and Jang, J.-C. (2003). Mechanisms of glucose signaling during seed germination of Arabidopsis thaliana. Plant Physiol. 132, 1–15. Ramirez-Carrozzi, V.R., and Kerppola, T.K. (2001). Dynamics of Fos– Jun–NFAT1 complexes. Proc. Natl Acad. Sci. U S A. 98, 4893–4898. Riano-Pachon, D.M., Ruzicic, S., Dreyer, I., and Mueller-Roeber, B. (2007). PlnTFDB: an integrative plant transcription factor database. BMC Bioinformatics. 8, 42. Rolland, F., and Sheen, J. (2005). Sugar sensing and signalling networks in plants. Biochem. Soc. Trans. 33, 269–271. Rolland, F., Baena-Gonzalez, E., and Sheen, J. (2006). Sugar sensing and signaling in plants: conserved and novel mechanisms. Annu. Rev. Plant Biol. 57, 675–709. Rolland, F., Winderickx, J., and Thevelein, J.M. (2001). Glucose-sensing mechanisms in eukaryotic cells. Trends Biochem. Sci. 26, 310–317. Rook, F., and Bevan, M.W. (2003). Genetic approaches to understanding sugar-response pathways. J. Exp. Bot. 54, 495–501. Rook, F., Gerrits, N., Kortstee, A., Kampen, M.v, Borrias, M., and Weisbeek, P. (1998). Sucrose-specific signalling represses translation of the Arabidopsis ATB2 bZIP transcription factor gene. Plant J. 15, 253–263. Satoh, R., Fujita, Y., Nakashima, K., Shinozaki, K., and YamaguchiShinozaki, K. (2004). A novel subgroup of bZIP proteins functions as transcriptional activators in hypoosmolarity-responsive expression of the ProDH gene in Arabidopsis. Plant Cell Physiol. 45, 309–317. Schindler, U., Beckmann, H., and Cashmore, A.R. (1992). TGA1 and G-box binding factors: two distinct classes of Arabidopsis leucine zipper proteins compete for the G-box-like element TGACGTGG. Plant Cell. 4, 1309–1319. Sheen, J. (2001). Signal transduction in maize and Arabidopsis mesophyll protoplasts. Plant Physiol. 127, 1466–1475. Smeekens, S. (2000). Sugar-induced signal transduction in plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 51, 49–81. Song, Y.H., et al. (2008). DNA-binding study identifies C-box and hybrid C/G-box or C/A-box motifs as high-affinity binding sites for STF1 and LONG HYPOCOTYL5 proteins. Plant Physiol. 146, 1862–1877. Thum, K.E., Shin, M.J., Palenchar, P.M., Kouranov, A., and Coruzzi, G.M. (2004). Genome-wide investigation of light and

Kang et al.

carbon signaling interactions in Arabidopsis. Genome Biol. 5, R10. Uno, Y., Furihata, T., Abe, H., Yoshida, R., Shinozaki, K., and Yamaguchi-Shinozaki, K. (2000). Arabidopsis basic leucine zipper transcription factors involved in an abscisic acid-dependent signal transduction pathway under drought and high-salinity conditions. Proc. Natl Acad. Sci. U S A. 97, 11632–11637.

d


|

373

characteristics and responses to the Arabidopsis Cytc-2 gene, encoding an isoform of cytochrome c. J. Exper. Bot. 60, 829–845. Weltmeier, F., Ehlert, A., Mayer, C.S., Dietrich, K., Wang, X., Schutze, K., Alonso, R., Harter, K., Vicente-Carbajosa, J., and Droge-Laser, W. (2006). Combinatorial control of Arabidopsis proline dehydrogenase transcription by specific heterodimerisation of bZIP transcription factors. EMBO J. 25, 3133–3143.

Vicente-Carbajosa, J., Moose, S.P., Parsons, R.L., and Schmidt, R.J. (1997). A maize zinc-finger protein binds the prolamin box in zein gene promoters and interacts with the basic leucine zipper transcriptional activator Opaque2. Proc. Natl Acad. Sci. U S A. 94, 7685–7690.

Weltmeier, F., et al. (2009). Expression patterns within the Arabidopsis C/S1 bZIP transcription factor network: availability of heterodimerization partners controls gene expression during stress response and development. Plant Mol. Biol. 69, 107–119.

Vinson, C., Acharya, A., and Taparowsky, E.J. (2006). Deciphering BZIP transcription factor interactions in vitro and in vivo. Biochim. Biophys. Acta. 1759, 4–12.

Wiese, A., Elzinga, N., Wobbes, B., and Smeekens, S. (2004). A conserved upstream open reading frame mediates sucrose-induced repression of translation. Plant Cell. 16, 1717–1729.

Walter, M., et al. (2004). Visualization of protein interactions in living plant cells using bimolecular fluorescence complementation. Plant J. 40, 428–438.

Wiese, A., Elzinga, N., Wobbes, B., and Smeekens, S. (2005). Sucrose-induced translational repression of plant bZIP-type transcription factors. Biochem. Soc. Trans. 33, 272–275.

Weinthal, D., and Tzfira, T. (2009). Imaging protein–protein interactions in plant cells by bimolecular fluorescence complementation assay. Trends Plant Sci. 14, 59–63.

Xiao, W., Sheen, J., and Jang, J.C. (2000). The role of hexokinase in plant sugar signal transduction and growth and development. Plant Mol. Biol. 44, 451–461.

Welchen, E., Viola, I.L., Kim, H.J., Prendes, L.P., Comelli, R.N., Hong, J.C., and Gonzalez, D.H. (2009). A segment containing a G-box and an ACGT motif confers differential expression

Yoo, S.D., Cho, Y.H., and Sheen, J. (2007). Arabidopsis mesophyll protoplasts: a versatile cell system for transient gene expression analysis. Nat. Protoc. 2, 1565–1572.