Isolation and expression analysis of two tomato ADP-glucosepyrophosphorylase S (large) subunit gene promoters

Isolation and expression analysis of two tomato ADP-glucosepyrophosphorylase S...

(Parte 1 de 3)

Isolation and expression analysis of two tomato ADP-glucose pyrophosphorylase S (large) subunit gene promoters

Jinpeng Xing, Xiangyang Li, Yuying Luo, Thomas J. Gianfagna*, Harry W. Janes

Department of Plant Biology and Pathology, Rutgers University, 59 Dudley Road, New Brunswick, NJ 08901-8520, USA

Received 20 April 2005; received in revised form 7 June 2005 Available online 1 July 2005


ADP-glucose pyrophosphorylase (AGPase, EC2.7.7.27) is a key enzyme in starch synthesis. The enzyme is a heterotetramer with two S (large) and two B (small) subunits. In tomato (Lycopersicon esculentum Mill), there are three S subunit genes. The Agp S1 and Agp S3 genes and their promoters were isolated from a tomato genomic library. The Agp S1 promoter region has a TATAA box at 57 bp and a CCAAT box at the 8 bp position. Tomato Agp S1 promoter is active in starch storage tissues, and in organs that are carbohydrate sinks. Its activity is localized in the guard cells and veins of the leaf, in the root cap and root vascular tissues, in the starch sheath cells of the stem and in the ovary and stamens of the flower. It is also active in pollen, seeds and in the inner pericarp wall and placental tissue of developing tomato fruits. No activity was observed in mesophyll cells, or in the ovaries at anthesis. The Agp S3 promoter has a CCAAT box at the 115 bp position and multiple GC rich regions around the CCAAT box, but it does not contain a TATAA box. Tomato Agp S3 promoter is expressed in the mesophyll cells, guard cells and veins of the leaf. In the flower, Agp S3 is strongly expressed in the sepals, ovary and anthers at anthesis. Deletions in the distil 50 upstream region of Agp S3 indicated that only a 0.5 kb fragment of the Agp S3 promoter is required for complete expression in transgenic plants. Agp S3 contains a minimal promoter with little evidence of functional cis-acting regulatory elements. For AGPase, the different requirements for regulation of enzyme activity between source leaves and sink organs are met in part by the evolution of multiple S subunit genes with very different promoters. # 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Starch; Tomato (Lycopersicon esculentum Mill); ADP-glucose pyrophosphorylase; Plant promoters; Transgenic plant; GUS expression

1. Introduction

ADP-glucose pyrophosphorylase (AGPase, EC2.7.7.27) is the critical enzyme in starch synthesis, both in photosynthetic and non-photosynthetic tissues [1].I t catalyzes the reaction of glucose-1-phosphate and ATP to ADP-glucose, which is the substrate for starch synthesis. Several lines of evidence indicate that alternative pathways to provide activated glucose for starch synthesis do not exist. Maize mutants with only 5–10% AGPase activity have reduced starch levels, with only 20–30% of the starch of wild type [2].I n Arabidopsis, a mutant lacking leaf starch was identified that had no extractable AGPase activity, but other enzymes such as starch synthase and UDP-glucose pyrophosphorylase were unaffected [3]. Transgenic potato plants with reduced levels of AGPase activity had lower starch content in the tubers [4], whereas expression of an AGPase variant from E. coli in potato tubers increased the starch content by an average of 35% [5].

The AGPase enzyme in plants is a heterotetramer of 200– 240 kDa consisting of two S (large) and two B (small) subunits with molecular weights of 54 and 51 kDa, respectively [6]. In a number of plants there are multiple forms of the genes for both subunits. In tomato, there are three S subunit genes (Agp S1, Agp S2, and Agp S3) and one gene for the B subunit (Agp B) [7–9]. All of the cDNAs from these genes are unique and do not hybridize with each other. Multiple cDNAs for each of the subunit genes have been isolated from other plants such as potato, wheat, rice and Arabidopsis. Plant Science 169 (2005) 882–893

* Corresponding author. Tel.: +1 732 932 9711x252; fax: +1 732 932 9441. E-mail address: (T.J. Gianfagna).

0168-9452/$ – see front matter # 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.plantsci.2005.06.004

AGPase activity may be regulated allosterically, transcriptionally, and/or post-transcriptionally. AGPase is activated by 3-PGA and inhibited by Pi with the ratio of the two effectors playing a major role in modulating AGPase activity [10,1]. In source tissues, during photosynthetic carbon fixation,whichproduces3-PGAandconsumesPiduringATP synthesis, the ratio of 3-PGA/Pi is high, activating AGPase and increasing starch biosynthesis. While in sink tissues, allosteric control may not be the main regulatory mechanism, since the precursor for starch synthesis is imported sucrose. In this case, the steady supply of sucrose to the sink may induce a sustainable increase in the transcription of the AGPase subunit genes and/or an increase in mRNA stability. For example, in rice endosperm, AGPase B subunit mRNA was detected 3 days after pollination (dap) and reached its maximum at 14 dap. These changes were closely correlated to the AGPase protein level, suggesting that AGPase gene expression isregulated atthe transcription level in sink tissues [12]. Similar results were found in other plants like potato, maize, and bean. A third level of regulation of AGPase activity is the light or sucrose induced formation of an interchain disulfide bond between the small subunits, which leads to activation of the enzyme [13]. There is also evidence that sugar may be an important regulatory factor for AGPase geneexpression[14–16].Forexample,AGPasesubunitgenes in Arabidopsis were found to be significantly and differentially regulated by sugar and light/dark exposures [17].

Our previous research [18] and the research of [19] also indicated that sucrose is an important regulatory factor for AGPase, and that the transcription of Agp S1 was directly correlated to the sucrose content of sink tissue.

In source tissues, such as potato leaves, the AGPase B subunit protein, detected by using western blots, was not well correlated to its mRNA level. This suggests that the expression of Agp B subunit gene in source tissue was regulated post-transcriptionally [20]. The research of [16] suggested that in source tissues the primary mode of regulation of AGPase is allosteric and post-transcriptional, whereas in sink tissue, the predominant level of AGPase regulation is transcriptional. In Arabidopsis three cDNAs for the large subunit (designated ApL1, ApL2 and ApL3) were obtained and their kinetic and regulatory properties studied by expressing the ApL genes with the small subunit gene in E. coli. They found that ApL1 is regulated allosterically [21], whereas [17] found that ApL3 is regulated at the transcriptional level.

In tomato, the AGPase activity in mature leaves is very high, in young leaves and fruits moderate and in roots low [7]. In fruit, Agp S1 and Agp B are expressed very highly compared to Agp S2 and Agp S3, whereas in leaves, Agp S3 and Agp B are highly expressed compared to Agp S1 and Agp S2 [8]. In roots, only Agp S1 and Agp S2 were expressed.

There is less information concerning the spatial expression of the S subunit genes within leaves, roots and fruit. Agp S1 from potato is active in the guard cells of the leaf, the starch sheath of the stem and in the flower anthers and ovary.

No activity was detected in the root or in the mesophyll of the leaves [2]. Given our finding that the tomato S1 gene was expressed in roots [8] in contrast to the results for the potato gene, and that there are no reports on the spatial expression of Agp S3, we isolated the tomato Agp S1 promoter, and we report for the first time the expression pattern in roots and a fleshy fruit. We also isolated the tomato Agp S3 gene (equivalent to the Arabidopsis ApL1) and determined the expression pattern of the promoter and truncated versions in transgenic tobacco plants.

2. Materials and methods 2.1. Screening of the tomato genomic library

A tomato genomic library was constructed in l-phage.

Plaques were transferred to nitrocellulose filters, and the DNA denatured and neutralized. Tomato Agp S1 and Agp S3 cDNA were labeled with [a-32P]dCTP and used as probes. Hybridization was performed at 65 8Ci n5 SSC, 0.1% SDS, 0.2 mg/ml bovine serum albumin, 0.2 mg/ml Ficoll, and 0.2 mg/ml denatured herring sperm DNA. Filters were washed in 2 SSC, 0.1% SDS at room temperature once for 20 min, and in 0.1 SSC, 0.1% SDS at 65 8C twice for 20 min. The filters were exposed between intensifying screens at 70 8C for 24 h to Kodak XOMAT films. Several hybridizing phage clones were obtained. An 8.1-kb long insert clone for Agp S1 and a 6.4 kb long insert clone for Agp S3 were selected and purified by three rounds of hybridizations.

2.2. Sub-cloning l-DNA was extracted according to the technical bulletin of Promega (Purification of Lambda DNA with WizardTM Lambda Preps DNA Purification System). Selected lphages were picked up from fresh plates and put in phage

(10 g Bacto Tryptone, 5 g Bacto Yeast Extract, 10 g NaCl in 1 l) medium supplemented with 10 ml of 20% maltose and

10 mlof 1 M MgSO4 at 37 8C overnight and then mixedwith 2 ml phage buffer containing l-phage and cultured at 37 8C for 20 min. LB (10 ml) was added and the cultures kept at 37 8C for 5 h until lysis occurred. The lysate was centrifuged at 8000 g for 10 min and the supernatant was collected. Nuclease mixture (40 ml) (containing 0.25 mg/ml RNAase A, 0.25 mg/ml DNAase I, 150 mM NaCl, 50%glycerol) was added to the supernatant and incubated at 37 8C for 15 min followed by the addition of 4 ml phage precipitant (3% polyethylene glycol, 3.3 M NaCl) with incubation on ice for 30 min. The mixture was centrifuged at 10,0 g for 10 min, and the supernatant was discarded. The pellet was resuspended with 500 ml phage buffer. Purification resin (1 ml) was added and transferred to a minicolumn. The column was centrifuged at 4000 g for 30 s, washed with 80% isopropanol, and centrifuged at 12,0 g for 2 min. TE buffer (100 mM Tris–HCl, 1 mM EDTA, pH 7.5) preheated to 80 8C (100 ml) was added and centrifuged at 12,0 g for 1 min to collect l-DNA. Purified l-DNAwas cut with Bam HI, Eco RI, Eco RV and Hind I. The cut fragments were cloned to bluescript plasmid. A physical map for the Agp S1 subunit gene was constructed.

2.3. PCR

50 ml PCR reaction solution was prepared containing

2.4. Sequencing

AgpS1 and S3 clones were sequenced according to the technical manual of Promega Silver SequenceTM DNA Sequencing System. Briefly, after sequencing PCR, each sample will have four sequencing PCR reactions with ddATP/dNTP, ddGTP/dNTP, ddTTP/dNTP, ddCTP/dNTP, respectively. The PCR products were run on 0.4-m thick 4% polyacrylamide gel (19:1, acrylamide:bisacrylamide) at 2000 V for about 2.5 h. The gel was fixed in 10% glacial acetic acid for 20 min and washed with water three times for 2 min each. The gel was stained in 0.1% AgNO3,0 .05% formaldehyde for 30 min and then rinsed for 6 s. The gel was developed in a pre-chilled solution containing 3%

rinsed twice for 2 min each. The sequence was read from the gel. Sequence analysis was performed by using Genetics Computer Group (GCG), the WISCONSIN PACKAGE, version 9.1.

2.5. Construction of promoter-GUS fusion

The3178 bpAgpS1promoterregionwasfusedtotheGUS reporter gene of the binary vector PBI101.1 [23]. The GUS genewasligatedtotheAgpS1promoterat+40 bp.Constructs used for transforming tobacco plants with Agp S3 promoter containing the b-GUS gene were created. 0.2, 0.5, 0.9 and 1.5 KbAgpS3truncatedpromoterconstructswerealsomade. AlltheAgpS3promoterconstructsusedPBI101asthevector. The GUS gene was ligated at +3 bp to the Agp S3 promoter.

2.6. Plant transformation

The above constructs were transformed to Agrobacterium

LBA 4404 by electroporation. Tobacco cv. Samsun N or tomato cv. Laura leaf disks were precultured for 2 days on feeding plates containing MS media plus 5 mM BA and 0.1 mM NAA. For the tomato cultures, feeder layer cells (1.5 ml) from 7-day-old tobacco suspension cells were plated after 1 day culture in the dark at 25 8C with a sterile filter paper on the surface. Leaf disks were co-cultured with Agrobacterium for 2 days on the feeding media. The plant tissues were then transferred to selection media (feeding media plus 200 mg/ml kanamycin and 200 mg/ml augmentin) to regenerate transgenic plants. Shoots were cut from the selection media and moved to rooting media (50% MS media plus 200 mg/ml kanamycin and 200 mg/ml augmentin). Two weeks later, the young plants were transferred to the greenhouse. For controls, plants were regenerated from tissue culture on media without antibiotics.

2.7. Histochemical localization of GUS activity

Histochemical staining for GUS activity was performed

EDTA, 0.05 mM K3FeCN6 and K4FeCN6, 0.1 mg/ml X glucuronide. Plant tissues were incubated in GUS staining solution for 6 h or overnight at 37 8C and washed with water.

Fig. 1. (A) The alignment of Agp S1 promoter and S1 gene. (B) The similar regions between tomato and potato Agp S1 promoter. The colored boxes and their relative positions denote high sequence similarity between tomato and potato Agp S1 promoter (E1) S180 to +16 bp in tomato, 190 to +16 bp in potato; (E2) S888 to 632 bp in tomato, 804 to 544 bp in potato; (E3) S994 to 937 bp in tomato, 917 to 860 bp in potato. (C) The alignment of Agp S3 gene and promoter.

Photosynthetic tissues were bleached with 75% ethanol for 3 h, followed by 95% ethanol for 3 h and 100% ethanol overnight. After bleaching, tissue was washed in water for 1 h. Photographs were taken under a compound microscope and/or dissecting-microscope.

2.8. Plant materials

Transformed tomato and tobacco plants were grown in a greenhouse at 25/20 8C (d/n). Supplemental lighting was provided from high-pressure sodium lamps at an intensity of 80 mmol/(m2/s) for 16 h from 05:0 to 21:0 if the natural light intensity was less than 800 mmol/(m2/s).

3. Results 3.1. Isolation of tomato Agp S1 promoter

A lambda phage clone with 8120 base pairs (bp) of tomato DNA that hybridized with the Agp S1 cDNA was isolated and sequenced from a tomato genomic library. Based on the comparison with the Agp S1 cDNA isolated from tomato fruit [8], this clone contains a 3178 bp upstream promoter region (GenBank accession no.: AY858853), and 4306 bp structural gene (GenBank accession no.: AY858852)( Fig. 1A). There are 14 introns and 15 exons in the Agp S1 structural gene. One TATAA box, a common eukaryotic regulatory sequence, at position 5 bp upstream from the 50end of the transcription initiation site was found (Fig. 1A), and there is also a CCAAT box [24,25] at 8 bp position. No initiator sequence was identified. One member of a large family of AT rich nuclear factors [26], NAT2, at 266 bp was also identified. A downstream promoter element (DPE) at the 200 bp position was identified. There is also a Pyr box [27] at 2422 bp, an OSE motif at 40 bp and three anaerobic responsive element (ARE) boxes in the Agp S1 promoter [28].

ThereishighsimilaritybetweenthetomatoAgpS1andthe potato Agp S1 genes [2]; however, there are only a few sequences in the promoter region that these two genes have in common. A sequence comparison search revealed three regions with high sequence identity between the two promoters (Fig. 1B). The first region is at 180 to +16 bp (relative to the transcription start site) in tomato and at 190 to +16 bp in potato, is 8.4% identical. The second region is from 8 to 632 bp in tomato and 804 to 544 bp in potato,is87%identical,andthethirdregion,whichhas94.8% sequence identity, is from 994 to 937 bp in tomato and 917 to 860 bp in potato.

Fig. 2. Gus activity in transgenic tobacco root with Agp S1 promoter-Gus fusion construct: (A) the root cap, root meristem and root vascular tissues were all stained; (B) root vascular tissue, both phloem and xylem were stained (magnification 100 ).

Fig. 3. Gus activity in transgenic tobacco plant stem with Agp S1 promoter- Gus fusion construct: (A) untransformed control plant; (B) transgenic plant, showing that the starch sheath cells were stained (magnification 100 ).

3.2. Histochemical localization of Agp S1 promoter expression in transgenic tobacco plants

Tomato Agp S1 is highly expressed in roots (Fig. 2A).

Most activity was found in the root cap, root meristem and both the phloem and xylem of the root vascular tissues (Fig. 2B). No staining was seen in epidermal cells, or cortex cells, even though they often contain starch granules. The potato Agp S1 promoter was not expressed in roots [2].

(Parte 1 de 3)