Characterization of a Maize Wip1 Promoter in Transgenic Plants

The Maize Wip1 gene encodes a wound-induced Bowman-Birk inhibitor (BBI) protein which is a type of serine protease inhibitor, and its expression is induced by wounding or infection, conferring resistance against pathogens and pests. In this study, the maize Wip1 promoter was isolated and its function was analyzed. Different truncated Wip1 promoters were fused upstream of the GUS reporter gene and transformed into Arabidopsis, tobacco and rice plants. We found that (1) several truncated maize Wip1 promoters led to strong GUS activities in both transgenic Arabidopsis and tobacco leaves, whereas low GUS activity was detected in transgenic rice leaves; (2) the Wip1 promoter was not wound-induced in transgenic tobacco leaves, but was induced by wounding in transgenic rice leaves; (3) the truncated Wip1 promoter had different activity in different organs of transgenic tobacco plants; (4) the transgenic plant leaves containing different truncated Wip1 promoters had low GUS transcripts, even though high GUS protein level and GUS activities were observed; (5) there was one transcription start site of Wip1 gene in maize and two transcription start sites of GUS in Wip1::GUS transgenic lines; (6) the adjacent 35S promoter which is present in the transformation vectors enhanced the activity of the truncated Wip1 promoters in transgenic tobacco leaves, but did not influence the disability of truncated Wip1231 promoter to respond to wounding signals. We speculate that an ACAAAA hexamer, several CAA trimers and several elements similar to ACAATTAC octamer in the 5′-untranslated region might contribute to the strong GUS activity in Wip1231 transgenic lines, meanwhile, compared to the 5′-untranslated region from Wip1231 transgenic lines, the additional upstream open reading frames (uORFs) in the 5′-untranslated region from Wip1737 transgenic lines might contribute to the lower level of GUS transcript and GUS activity.


Introduction
It is necessary to use different promoters when multiple genes are cloned in tandem in transgenic plants because multiple copies of a single promoter could lead to the silencing of the transgenes. Promoters may be classified as constitutive, tissue-specific, or inducible based on their expression patterns. The most widely used constitutive promoters include the CaMV 35S promoter [1], the maize ubiquitin promoter [2], and the rice actin promoter [3]. Several other strong promoters have also been isolated and identified from plants or viruses [4][5][6]. Although many promoters have been isolated and tested, only a few have been successfully used in agricultural biotechnology. The constitutive expression of certain genes might be harmful to their host plants, affecting plant growth and development; inducible or tissue-specific promoters would be more useful than constitutive promoters in these cases.
The function of a promoter is normally determined by the combinatorial action of multiple regulatory elements, i.e., enhancers and cis-elements, and by the interactions between regulatory elements and nuclear protein factors [7]. For the CaMV 35S promoter, the region from −46 to +1 upstream of the gene is the basal promoter and the region from −343 to −46 acts as an enhancer [1]. Although the CaMV 35S promoter is presumed to be constitutive, some reports have shown that it is developmentally regulated [8] and can be affected by abiotic stress [9]. Negative cis-regulatory regions determine the tissue-specific activity of a number of promoters [10]. Besides the cis elements in DNA sequence affect transcription, various cis elements in mRNA have the functions to control the translation efficiency. Many reports have shown that the 5'-untranslated region (5'-UTR) and 3'-untranslated region (3'-UTR) can increase the translation efficiency [11]. Upstream open reading frames (uORFs) that are located in the 5'-UTR have been reported to generally downregulate the expression of the major open reading frame (mORF) through ribosomal stalling or reducing initiation efficiency [12]. However, a few uORFs can upregulate the expression of mORF [13]. Additionally, more and more reports showed that the 35S promoter interferes with the tissue specificity, strength [14] or inducibility (our data, unpublished) of its adjacent promoter. The onset and extent of influence depends on the responsive nature of the adjacent promoter and the distance between the 35S promoter and adjacent promoter [15].
The Bowman-Birk inhibitor (BBI) is a type of serine protease inhibitor, and its expression is induced by wounding or infection, conferring resistance against pathogens and pests [16]. It was reported that the BBI confers heavy metal and multiple drug tolerance in yeast [17]. There are seven BBI genes in rice and overexpression of the rice BBI2-3 conferred fungal pathogen resistance in transgenic rice plants [18]. The maize Wip1 gene, encoding a wound-induced BBI protein, has been cloned, and the promoter of the gene might respond to wound signals [19,20]. In contrast to most BBIs, which inhibit trypsin and chymotrypsin proteases, both of the inhibitory domains of maize Wip1 inhibit chymotrypsin [20]; however, these two domains have different evolutionary histories and ecological functions [21].
Here, we isolated the maize Wip1 promoter and characterized its function in transgenic plants, showing that Wip1 promoters showed different properties in dicot and monocot plants, and some truncated promoters under the influence of adjacent 35S promoter can drive high GUS activities even with a low GUS transcriptional level.

Isolation and Analysis of the Maize Wip1 Promoter
To isolate the maize Wip1 promoter, we performed a BLAST analysis with the mRNA sequence of maize Wip1 (X71396) in www.maizegdb.org [22]. The maize Wip1 sequence was homologous to the genome sequence from 12,274,189 to 12,273,525 on chromosome 8, and one intron was found in the Wip1 coding region. Based on the maize genomic sequence, a 1737 bp fragment (chr8: 12,275,876-12,274,140) upstream of the translation start site of Wip1 was amplified by PCR. The cis-elements in the promoter sequence were analyzed using the PlantCARE [23] and PLACE [24] databases, and the transcription start sites were predicated on the website: http://www.fruitfly.org/seq_tools/promoter.html [25]. Two putative TATA boxes and two corresponding potential transcription start sites were present in the promoter sequence. Six W-boxes and three GCC-boxes were found in the Wip1 promoter. In addition, several CAAT-boxes, which are common cis-acting elements in promoter and enhancer regions, were also found ( Figure 1).

Truncated Wip1 Promoter Has Strong Activity in Transgenic Arabidopsis and Tobacco Leaves
To analyze the function of the Wip1 promoter, several expression vectors (pWip 1737 , pWip 1500 , pWip 1231 , pWip 1191 , pWip 791 , pWip 491 ) were constructed (Figure 2A, Tables 1 and 2). The constructed plasmids were transformed into tobacco, and at least ten independent transgenic tobacco lines for each construct were obtained. The transformations were confirmed by PCR analysis of genomic DNA using primers specific for GUS (data not shown). We performed quantitative analysis of GUS activity using leaf samples from 40-day-old T 3 progeny of plants. The results showed that the Wip 1231 promoter had strong activity, while no or weak activity was detected for the other truncated promoters ( Figure 2B). To confirm the results of the GUS activity analysis, the seedlings were subjected to histochemical staining. Strong GUS staining was observed for the seedlings that were transgenic for the Wip 1231 promoter and 35S promoter, whereas no or light staining was observed for the seedlings that were transgenic for the full-length and other truncated promoters (Figure 3), consistent with the GUS activity analysis results. To investigate whether the truncated Wip 1231 promoter also had strong activity in other plant species, the constructs were transformed into Arabidopsis. As expected, the Wip 1231 promoter also drove high GUS activity in transgenic Arabidopsis plant leaves ( Figure 2C), further confirming the high activity of the truncated Wip 1231 promoter.   Table 1. The sequence of the primers used in the paper.

Primers
Sequence  To find out the potential negative cis-acting elements, detailed 5' and 3' deletions of Wip 1737 , based on Wip 1231 , were performed, involving the generation of an additional 15 deletion constructs (pWip 1231A , pWip 1231C , pWip 931 , pWip 1231a , pWip 1231b , pWip 1231c , pWip 1231Aa , pWip 1231Ab , pWip 1231Ac , pWip 1231Ba , pWip 1231Bb , pWip 1231Bc , pWip 1231Ca , pWip 1231Cb , pWip 1231Cc ; Figure 4A, Tables 1 and 2). These constructs were transformed into Arabidopsis, and the GUS activity in the transgenic plant leaves was analyzed. The results showed that Wip 1231A , Wip 1231C , and Wip 931 , which were of the same 3' end as Wip 1231 , had high activity similar to that of Wip 1231 , while the other Wip1 promoter fragments presented very low levels of activity ( Figure 4C). To further confirm the functions of these three fragments, three constructs (pWip 1231A , pWip 1231C and pWip 931 ) were also transformed into tobacco, respectively. High GUS activities were observed in the transgenic tobacco leaves transformed with each of these three promoters ( Figure 4B).

Wip1 Promoters Are not Induced by Wounding and Have Different Activity in Different Organs in Transgenic Tobacco
It has been shown that the expression of Wip1 gene is wound-induced in maize [19,20]. We investigated whether the Wip1 promoter is wound-induced in transgenic tobacco plant leaves that were wounded with three different kinds of wounding methods, respectively. Under three different treatment conditions, to our surprise, wounding did not induce the increase of the activity of Wip1 promoters ( Figure 5), indicating that Wip1 promoters are not wound-induced in transgenic tobacco. Figure 5. GUS activity in transgenic tobacco leaves treated with wounding. The second leaves from apex of 40-day-old T 3 tobacco plants grown in greenhouse were wounded using three different kinds of methods. GUS activities were assessed 24 h after wounding. A-NW and A-W mean non-wounded or wounded samples using the method described in Walker-Simmons, et al. [26]. B-NW and B-W mean non-wounded or wounded samples using the method described in An, et al. [27]. C-NW and C-W mean non-wounded or wounded samples using the method described in Xu, et al. [28]. Data was shown as average ± S.E. of 6 independent transgenic lines. The GUS activities in different organs of the transgenic tobacco lines transformed with different truncated Wip1 promoters or the control 35S promoter were analyzed. The results showed that the Wip 1231 promoter has high activity in the leaf and stem but low activity in the root and seed ( Figure 6).  Transcriptional level of GUS in transgenic tobacco leaves. Data was shown as average ± S.E. of 6 independent transgenic lines. Letters above the columns indicate statistically significant differences at p < 0.05 level. Relative transcript levels were calculated using the 2 −ΔΔCt method [29] with Actin as a housekeeping gene.

Wip1 Promoters Have Multiple GUS Transcription Start Sites in Transgenic Tobacco
To find out the exact 5'-UTRs, we further determined the accurate transcription start sites in maize and transgenic tobacco plants by the 5'-RACE method. The native transcription start site of Wip1 in maize was identified as the nucleotide G which is located at 61 bp upstream of the Wip1 translation initiation site and 3 bp downstream of the predicated one (Figures 1 and 8). For Wip 1737 ::GUS transgenic tobacco line, a strong band and one weak band were amplified, and sequencing results revealed that two transcription start sites are present (Figures 1 and 8). The latter is the same as the transcription start site of Wip1 gene in maize. The former was identified as nucleotide C which is located at 532 bp upstream of the ATG translation initiation site and identical to the predicated one, and this transcription start site was also identified in Wip 1231 ::GUS transgenic tobacco line. So there is a 272 bp-length untranslated region in GUS transcripts in Wip 1231 ::GUS lines. In the 272 bp 5'-UTR there is an ACAAAA element, several CAA trimers and several elements similar to ACAATTAC octamer ( Figure S2). The elements might contribute to the high translation efficiency of GUS transcripts in Wip 1231 transgenic lines. The sequences of 5'-UTR from Wip 1231 and Wip 1737 transgenic lines were comparatively analyzed to find that there were more uORFs in the latter (Figures S3-S5). Maybe the additional uORFs lead to the lower level of GUS transcript and GUS activity in Wip 1737 transgenic lines.

Wip 1231 Promoters Were Influenced by the Adjacent 35S Promoter Sequence
In this study, the transformation vectors were constructed from a derivative of the pCAMBIA1300 binary vector that includes a CaMV 35S promoter controlling a hptII selectable marker gene. It has been reported that the CaMV 35S enhancer can influence the expression of a transgene within the same transformation construct [14,30,31]. To investigate whether the 35S enhancer influences the truncated Wip 1231 , we constructed transformation vectors p1300-1231-NOS and p1300-35S-NOS, where the 35S promoter was replaced with a NOS promoter to control the hptII selectable marker gene, and the reporter gene GUS was controlled by the Wip 1231 promoter and 35S promoter, respectively. Then they were transformed into tobacco, and transgenic tobacco plants were obtained. The GUS activity in transgenic plants transformed with p1300-1231-NOS was very low, compared to transgenic lines transformed with pWip 1231 . However, transgenic plants transformed with p1300-35S-NOS still had high GUS activity, which were similar to the transgenic lines transformed with p1300-221 ( Figure 9). The Wip 1231 promoter was still not induced by wounding in transgenic tobacco plants which was transformed with p1300-1231-NOS (Figure 9). The above results showed that adjacent 35S promoter did enhance the expression of GUS reporter gene that was controlled by truncated Wip 1231 , and the truncated Wip 1231 does not respond to wound signals in transgenic tobacco plants. Figure 9. GUS activity in tobacco plant leaves transformed with constructs p1300-35S-NOS or p1300-1231-NOS. The second leaves from apex of 40-day-old T 3 tobacco plants grown in greenhouse were used. NW and W mean non-wounded or wounded samples using the method described in Walker-Simmons, et al. [26]. Data was shown as average ± S.E of n independent transgenic lines, n = 15 for p1300-1231-NOS; n = 13 for p1300-35S-NOS.

Wip1 Promoter Is Wound Inducible in Transgenic Rice Plants
Expression of the Wip1 gene is induced through wounding in maize. However, the GUS gene, driven by Wip1 promoters, was not induced by wounding in transgenic tobacco which is a dicot. This encouraged us to analyze whether Wip1 promoters have the same function in monocot species. Constructs pWip 1737 , pWip 1500 and pWip 1231 were transformed into rice, respectively, and the transgenic rice events were confirmed by PCR analysis of the genomic DNA (data not shown). Low GUS activities were detected in transgenic rice plant leaves containing different Wip1 promoters, compared with that in transgenic rice transformed with the CaMV 35S promoter. When transgenic rice leaves were wounded, significant increase of GUS activity were observed in all the transgenic lines containing different Wip1 promoters, whereas the GUS activity in transgenic plants containing the 35S promoter did not change ( Figure 10). This indicates that the Wip1 promoter is induced by wounding in rice. Figure 10. GUS activity in transgenic rice leaves treated with wounding. The second leaves from apex of 40-day-old rice plants grown in a greenhouse were wounded using the method described in Xu, et al. [28]. NW and W mean non-wounded or wounded samples. Data was shown as average ± S.E. of 6 independent transgenic lines. Asterisks (**) means the significant difference at p < 0.01 level.

Discussion
Arabidopsis and tobacco have long been used as simple and high-efficiency transformation systems in which to analyze gene function. In addition to protein-coding genes, the promoters from rice or maize have been studied in one of these two systems [32][33][34]. In this study, we used both systems to study the function of maize Wip1 promoter. To our surprise, expression of GUS driven by truncated maize Wip1 fragments was not induced by wound in transgenic tobacco plants ( Figure 5). This may be due to the fact that the maize Wip1 promoter comes from a monocot, and its function may be different in monocot and dicot species. Although it has been reported that the wound-inducible PinII promoter from potato was able to drive gene expression in response to wound signals in rice [28], many plant promoters exhibit different functionality in dicots and monocots. For example, the widely used maize ubiquitin promoter conferred lower expression of a reporter gene in the tobacco (dicot) protoplasts than in maize (monocot) protoplasts [2]. Several dicot promoters also have been shown to have lower activities than monocot promoters when transformed into monocot species [35].
Few strong promoters have been identified to date, and it remains important for researchers in plant biotechnology to identify promoters that are as strong as or stronger than the CaMV 35S promoter. In transgenic plants, high GUS activities were observed in transgenic lines containing truncated Wip1 promoters even though low GUS transcriptional levels were detected. The above results indicate that truncated Wip1 promoters are not strong, and the reason for the high GUS activity driven by the truncated Wip1 promoters might be due to the increased translation of the GUS gene. It has been shown that the 35S-GUS and 35S-ABI4-GUS lines had at least 50 fold differences in transcript level, however more than 300 fold difference in GUS activity was observed [36], indicating that transcriptional level is not correlated with the protein level in some cases. Many factors, i.e., promoter activity, mRNA stability, protein translation efficiency, can affect the expression level of foreign genes in transgenic plants. The high GUS activity driven by truncated Wip1 promoters indicate that Wip1 promoters are also useful in plant biotechnology, even though these promoters led to low transcriptional level.
Normally, a promoter shows different properties in monocot and dicot plants. It has been shown that the CaMV 35S promoter can drive high levels of transgene expression in dicot plants, whereas its activity is relatively lower in monocot plants [37]. In transgenic rice, lower activity of the Wip 1231 promoter compared to the 35S promoter was observed, and wound treatment significantly increased GUS activities in transgenic rice containing different Wip1 promoters. These results are contrary to the results in transgenic tobacco, indicating that the Wip1 promoter also showed different properties in monocot and dicot plants; this will limit the usage of the Wip1 promoter as a general (cross-species) promoter in transgenic crops.
5'-RACE results revealed that only one native transcription start site was present in the maize Wip1 gene, and a new one was identified in both Wip 1737 ::GUS and Wip 1231 ::GUS transgenic tobacco lines. We analyzed the 272 bp 5'-UTR of GUS mRNA from Wip 1231 ::GUS transgenic tobacco lines to find that there is an ACAAAA hexamer, several separate CAA trimers and several elements similar to ACAATTAC octamer ( Figure S2). The ACAAAA hexamer exists in many 5'-UTRs of plant genes and has been speculated to enhance translation [38]. The poly(CAA) region and ACAATTAC octamer have been confirmed to enhance translation [39][40][41]. The features of the sequence may contribute to the property of the 5'-UTR for high efficient translation. It has been reported that uORF plays an important role in the regulation of gene expression by different mechanisms, for example, by ribosomal stalling, reducing initiation efficiency [12]. uORF can downregulate gene expression by accelerating mRNA degradation or upregulating gene expression by reforming the 5'-UTR through the product of the uORF [42]. Six upstream ATGs (uATGs) were found in the 272 bp UTR. Among the six uATGs, the first three are in two very short uORFs, the fourth and fifth are in the reading frame of the GUS gene and will produce GUS fusion protein, and the sixth uATG should not have the opportunity to initiate translation ( Figure S3). We suggest that the two uORFs may have little possibility to influence the mORF translation. Strong GUS activity in transgenic plants transformed with pWip 1231 construct and related pWip 1231A , pWip 1231C and pWip 931 might be due to that they all have the same 3' end and produce the same 5'-UTR containing two uATGs which produce GUS fusion proteins, and strong translation may initiate from the two uATGs.
14 uATGs were found in the 532 bp UTR of GUS mRNA from Wip 1737 ::GUS transgenic tobacco lines. These uATGs form more uORFs than the six uATGs in 272 bp UTR of GUS mRNA from Wip 1231 ::GUS transgenic lines. The fourth and fifth uATGs in the 272 bp UTR which can produce GUS fusion protein, are in frame with a termination code in the 532 bp UTR and form a uORF. Interestingly, the uORF also exists in the 361 bp UTR of GUS mRNA from Wip 1231a ::GUS transgenic tobacco lines if the lines have the same transcription start site as Wip 1231 ::GUS lines. We suggest that the uORF may lead to the low GUS activity of transgenic plants transformed by pWip 1737 and pWip 1231a . We also suggest that the uORF might accelerate the degradation of mRNA from Wip 1731 ::GUS and Wip 1500 ::GUS transgenic tobacco lines, because the levels of mRNA in these lines were lower than that in Wip 1231 ::GUS transgenic tobacco lines. We cannot exclude other factors that contribute to the attained results.
A few reports have shown that the CaMV 35S enhancer has the ability to alter the expression of nearby genes by affecting nearby promoters [14,30,31]. In the T-DNA region of transformation plasmids used for the promoter analysis, the selectable marker gene hptII was controlled with double enhanced CaMV 35S promoter. We confirmed that the 35S promoter influenced the GUS expression in Wip 1231 ::GUS transgenic plants, because low GUS activity was observed when the 35S promoter was replaced with the NOS promoter. A problem may appear if the 35S promoter exists nearby an inducible or tissue-specific promoter that drives target transgene expression, but the problem might be negligible when the adjacent promoter is a strong constitutive one. The disadvantage of the 35S promoter can be overcome by using enhancer-blocking insulators to disturb the interaction between the 35S promoter and adjacent promoters [43]. Every coin has two sides; maybe the property of the 35S promoter can be used to develop a "strong constitutive" promoter controlling transgene expression in plants, such as the Wip 1231 and several other truncated Wip1 promoters studied in the paper, although these truncated Wip1 promoters drive low transcriptional level of transgene.

Plasmid Construction
The plasmid pBI221 was digested with HindIII and EcoRI. The CaMV 35S promoter-GUS-NOS fragment was purified and ligated into plasmid pCAMBIA1300 that had been digested with the same enzymes to construct the plasmid pCAMBIA1300-221. The amplified 1737 bp full-length promoter sequence was digested with restriction enzymes PstI and BamHI and inserted upstream of the GUS gene in plasmid pCAMBIA1300-221, replacing the CaMV 35S promoter to construct a new plasmid, pWip 1737 ( Figure S6). The full-length Wip1 promoter was replaced with other truncated Wip1 promoters to construct the other vectors. The details for the primers used to amplify the truncated promoters are shown in Figure 1 and Tables 1 and 2. To construct vectors where 35S promoter was replaced with NOS promoter to control hptII selectable gene, NOS promoter was amplified using Gateway binary vector pGWB454 (AB294466.1) as template with primer F-NOS (5'-CCAACATGGTGGCATCATGAGCGGAGAATTAAG-3') and primer R-NOS (5'-CTCGAGAGATCCGGTGCAGATTATTTG-3'). The underlined letters indicate BstXI and XhoI restriction sites, respectively. The vector pWip 1231 was digested with BstXI and XhoI to produce three fragments, a 10 kb fragments containing vector bone, a 1.1 kb fragment containing hptII gene and a 0.8 kb fragment containing 35S promoter. The 10 and 1.1 kb fragments were gel purified separately and the 0.8 kb one was discarded. The 10 kb purified fragment ligated with the amplified NOS promoter fragment was digested by the same endonucleases to form the medium vector that was digested by XhoI. Then the purified product ligated with the 1.1 kb purified fragment in correct orientation to form the vector p1300-1231-NOS. The new control vector p1300-35S-NOS derived from pCAMBIA1300-221 was constructed similarly. The restriction endonuclease sites were in marked in detail in supplementary Figure S1.

Plant Transformation
For the tobacco plant transformations, the plant expression plasmids were transferred into competent Agrobacterium tumefaciens (strain LBA4404) cells by freeze-thaw treatment. The transformed Agrobacterium were selected on YEB-agar plates containing 100 mg/L kanamycin and 100 mg/L streptomycin. Recombinant Agrobacterium were infiltrated into the young tobacco (Nicotiana tobacum) leaves according to the described method [45]. For the Arabidopsis transformations, vector constructs were transformed into Agrobacterium tumefaciens strain GV3101. Arabidopsis ecotype Columbia Col-0 was transformed using the floral dip method [46]. The obtained tobacco and Arabidopsis T 0 seeds were selected on MS medium plates containing 30 mg/L Hygromycin B to eliminate the non-transgenic plants.
For the rice transformation, vector constructs were transformed into Agrobacterium tumefaciens strain EHA105. Three-week-old calli derived from mature seed (Rryza sativa L. cv. Kita-ake) were used for the transformation. The transformation was performed according to the described method [47].

Wounding of Transgenic Plant Leaves
The second leaves from apex of 40-day-old T 3 tobacco plants grown in greenhouse were wounded using three different kinds of methods. The first wound treatment was performed according to Walker-Simmons, et al. [26] wherein the leaves were crushed across the midvein by a hemostat, with a second wounding 20 h later, and the leave samples were harvested 24 h after the first wounding. The second wound treatment was performed according to An, et al. [27] wherein the leaves were cut into about 1 cm 2 sections and the leaf slices were put in Murashige and Skoog (MS) liquid medium containing 3% sucrose (pH 5.8). The leaf slices were wounded by making several small holes in each section with forceps, and then were placed in tissue culture incubator at 28 °C for 24 h under light (3000 Lux), then the samples were harvested for the experiment. The third wound treatment was performed according to the method described by Xu, et al. [28] wherein the leaves were cut three times perpendicularly to the midvein by a scissor along both edges of the leaf blade without damaging the midvein. The first leaf under the wounding one was cut off before wounding as unwounded sample, frozen in liquid nitrogen and stored at −80 °C before extracting crude protein.
For the wounding of rice leaves, the second leaves from apex of 40-day-old T 2 rice plants grown in greenhouse were wound according to the method described by Xu, et al. [28] wherein the leaves were cut at intervals of 1 cm perpendicularly to veins by a scissor along both edges of the leaf blade without damaging the midvein, and the leave samples were harvested 24 h after wounding. The first leaf under the wounding one was cut off before wounding as unwounded sample, frozen in liquid nitrogen and stored at −80 °C before extracting crude protein.

Analysis of RNA Level by qRT-PCR
The second leaf from the apex of 40-day-old T 3 transgenic tobacco was sampled. Total RNA was extracted using EasyPure™ Plant RNA Kit (TransGen, Beijing, China). The RNA was reverse transcribed to cDNA using TransScript ® II First-Strand cDNA Synthesis SuperMix Kit (TransGen, Beijing, China), and qRT-PCR was performed using the TransStart ® Green qPCR SuperMix Kit from Beijing TransGen Biotech Co. Ltd. (Beijing, China). The primers used for qRT-PCR were designed according to the sequences of the GUS gene from vector pBI221 (AF502128.1) and tobacco Actin gene (U60495.1). The sense primer for GUS gene is S-GUS: 5'-CCAACTCCTACCGTACCTC-3', and the antisense primer is A-GUS: 5'-TCGAAACCAATGCCTAAA-3'. The sense primer for Actin gene is S-ACTIN: 5'-AAGGGATGCGAGGATGGA-3', and the antisense primer is A-ACTIN: 5'-CAAGGAAATCACCGCTTTGG-3'. The instrument of ABI 7300 was used. The reaction volume is 20 µL. The amplification program was: 95 °C 2 min for pre-denature, then 40 cycles of 95 °C 10 s for denature, 55 °C 30 s for annealing, 72 °C 31 s for elongation were followed. The fluorescence data was collected at the end of elongation and analyzed by the 2 −∆∆Ct method [29].

Analysis of the Transcription Start Sites by 5' Rapid Amplification of cDNA Ends (5'-RACE)
According to the sequences of maize Wip1 gene (X71396.1) and GUS gene from vector pBI221 (AF502128.1), two specific primers, wip1-R: 5'-CAAAAAGGACTGCGACCCCGTCTG-3' and GUS-R: 5'-GTTCTGCGACGCTCACACCGATACC-3', were designed. wip1-R was used to amplify the 5' end of Wip1 mRNA in maize. GUS-R was used to amplify the 5' end of GUS mRNA in transgenic tobacco. Total RNA was extracted from leaves of 10-day-old maize inbred line Z31 and 30-day-old transgenic tobacco plants using EasyPure™ Plant RNA Kit (TransGen, Beijing, China). The generation of RACE-ready cDNA and rapid amplification of cDNA end were performed strictly according to the user manual of SMARTer™ RACE cDNA Amplification Kit (Clontech, Mountain View, CA, USA).

GUS Activity Assay and Histochemical Staining
T 2 transgenic tobacco seeds were germinated on MS medium plates with 30 mg/L Hygromycin B to eliminate the non-transgenic plants. The surviving 14-day-old seedlings were immersed in GUS staining solution [48] and incubated overnight at 37 °C. Then, the samples were de-pigmented with 70% ethanol at 37 °C until the chlorophyll had completely disappeared.

Conclusions
We used standard methods to study the function of the maize Wip1 promoter and found that several truncated Wip1 promoters led to high GUS activity in transgenic plants, even though the GUS transcriptional level was relatively low. The high GUS activity may be ascribed to high translation efficiency. The activity of the truncated Wip1 promoter fragments were influenced by the adjacent 35S promoter, however they may also be used as strong constitutive promoters to drive transgenes in plant biotechnology when a 35S promoter exists nearby. This may be a way to develop new "strong constitutive" promoters.