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 2 de 3)

When cross sections of the stem from transgenic tobacco were incubated with GUS staining solution, Agp S1 was found highly expressed in the starch sheath cells located between the cortex and the vascular tissue (Fig. 3). There is also strong staining in the stem guard cells (Fig. 4A). No staining was observed in other cells, even parenchyma cells, which contain starch granules.

In the leaves, only the guard cells were stained (Fig. 4B) and all veins were stained as well (Fig. 4C). No GUS activity was evident in mesophyll cells, even though massive transitory starch is accumulated in these cells. There is also no staining found in leaf trichomes.

Flowers were stained at various stages of growth. For flower buds 2 days before anthesis, there was intense staining in the ovary and anthers (Fig. 5A). At anthesis, however, staining was completely absent in the ovary, and much reduced in the anthers (Fig. 5B). Two days after anthesis, during the fruit set stage, there was significant activity in the placental tissue of the ovary. Seeds were also stained (Fig. 5C).

The overall expression pattern of tomato Agp S1 promoter is quite similar to potato Agp S1 promoter, except

Fig. 4. Gus activity in transgenic tobacco stem epidermis and leaves with Agp S1 promoter-Gus fusion construct: (A) stem guard cell was stained (magnification 400 ); (B) leaf guard cells were stained (magnification 100 ); (C) leaf veins and guard cells were stained (magnification 40 ).

Fig. 5. Transgenic tobacco flowers with Agp S1 promoter-Gus fusion construct: (A) flower buds 2 days before anthesis, the ovary and anthers are stained; (B) flower at anthesis, no staining is evident; (C) flower 2 days after anthesis with staining in the placental region and seeds (magnification 4 ).

that tomato the Agp S1 promoter is also expressed in roots.

3.3. Agp S1 promoter expression in transgenic tomato is similar to transgenic tobacco plants

In the reproductive tissue, Agp S1 is expressed in the flower ovary, fruit inner pericarp and placenta from 25-dayold fruit (Fig. 6). The expression of Agp S1 promoter in vegetative tissue of transgenic tomato is quite similar to transgenic tobacco; GUS staining was evident in the starch sheath cells of the stem (Fig. 7), and the root cap (data not shown). In the leaves, Agp S1 expression occurs exclusively in the guard cells and was conspicuously absent in the leaf mesophyll (data not shown).

3.4. Isolation of tomato AGPase large subunit S3 promoter

The Agp S3 genomic clone sequence is 6409 bp long, containinga 3927 bpupstreampromotersequence(GenBank accession no.: AY858854), and a 2428 bp structural gene (GenBank accession no.: AY858855). The structural gene component contains at least nine exons and eight introns. The first exon contained the putative start codon (Fig. 1C). The sequence of the structural gene is not complete. No further effort was made to get the whole sequence of the structural gene. Examination of the 50-flanking sequences revealed several conserved transcriptional motifs. A CCAAT box, a widespread regulatory sequence found in promoters and enhancers [24,25], is located at 1 bp upstream of the putative transcriptional start site. A TATAA box was not present in the Agp S3 promoter region. There was also no initiator sequence (Inr), an element encompassing the transcription start site in a variety of eukaryotic promoters. There were, however, multiple potential regulatory sequences, GC rich regions, in the vicinity of the CCAAT box. Two potential DPEs were also identified at +105 and +120 bp, respectively. There are also some other regulatory motifs that were found in the promoter region, specifically, pollen maturity responsive element, pollen box at 3711 bp, gibberellic acid responsive element [27], Amy Boxes, at 2997 and 1739 bp, Pyr Box at 1440 bp, as well as ARE Boxes at 3183, 1711, 1730, and at 393 bp.

3.5. Histochemical localization of Agp S3 promoter expression in transgenic plants

In leaves, all guard cells and mesophyll cells stained very strongly (Fig. 8A and B). Veins were also stained. No staining was found in trichomes. In stems, no GUS activity was found at all (Fig. 8C), and there was no GUS activity in

Fig. 6. Transgenic tomato fruit 25 days after anthesis with Agp S1 promoter-Gus fusion construct: (A) the inner pericarp and placenta are stained; (B) untransformed control fruit.

the root (Fig. 8D). In flowers, 2 days before anthesis, there was intense staining in anthers, stigma and the top part of the style (Fig. 9). There was also slight staining in the sepal and the placental tissue of the ovary. No staining in the petal was found. During anthesis, the sepal, the placental tissue of the ovary and the whole ovary stained intensely (Fig. 10). At this stage, the anthers, stigma and the top part of the style maintained strong GUS activity, and the bottom of the style was also strongly stained. Two days after anthesis, there was no staining in the ovary wall, but there was still very slight staining in the placental tissue (Fig. 1). The sepals were only lightly stained at this time. GUS activity still can be seen in the stigma, and both the top and bottom part of the style. No GUS activity was found in the petals throughout the flowering period.

In 12-day-old fruit, GUS activity was found in the inner pericarp and placental tissue of the ovary and also slightly in the sepal (Fig. 12). No GUS activity was found in the seeds.

3.6. Expression of the truncated derivatives of the Agp S3 promoter in transgenic tobacco plants

The 0.2 kb construct did not induce expression as expected. No differences in expression were found for the

0.5, 0.9 and 1.5 kb constructs compared to the 3.9 kb promoter region. The 0.2 kb construct may be too short for inducing expression. An element that confers guard cell expression was found at 537 bp that would be missing in the 0.2 kb promoter sequence.

4. Discussion 4.1. Agp S1 is expressed in sink tissues

Plant organs can be classified into sources and sinks according to whether they are importers or exporters of photoassimilates. Starch biosynthesis occurs in both sink and sources. AGPase is one of the major regulatory enzymes in higher plant starch biosynthesis so it is likely that all the cells that undergo starch biosynthesis will express one or more of the AGPase S subunit genes along with a B subunit gene.

Agp S1 promoter expression in leaves and epidermis of green stems are restricted only to the guard cells, no activity was found in leaf mesophyll cells, which are the major starch reservoirs in leaves. It was reported that there are at least three AGPase S (large) subunit genes in tomato, and northern blot analysis indicated that the Agp S3 gene has strong activity in leaves [8]. In contrast, expression of Agp S1 is low in leaves. Since the ratio of guard cells to other cells in leaves is very small, and Agp S1 is only expressed in the guard cells, this may explain the low Agp S1 activity in leaves.

The Agp S1 promoter leads to activity in the starch storage tissue of the root cap and in the vascular cylinder of the root near the tip. This may be important for gravitropic sensitivity. The root cap contains starch cells that are believed to be involved in the transduction of the gravitropic signal. This promoter has significant potential to contribute to studies in gravitropism since it will be possible to target gene expression to the root cap. Despite the presence of starch granules in the cortex there was no expression of Agp S1, however; another S subunitgene Agp S2 is also expressed in the root [8] and may be responsible for starch biosynthesis in this tissue.

The expression of Agp S1 in the inner pericarp and placenta of the tomato fruit is consistent with both the accumulation of starch in these tissues, and northern analysis that identified transcripts of Agp S1 in 10–25-day-old fruit [18]. Although the pericarp is a morphologically uniform tissue of vacuolated parenchyma cells, tissue differentiation between the inner and outer pericarp was established early in fruit development. In contrast to Agp S1, the gene for tomato polygalacturonase (PG) was expressed only in the outer pericarp and only at the onset of fruit ripening [29]. Given the exclusive staining patterns for the Agp S1 and PG promoters, there may be distinct biochemical functions for the inner and outer pericarp.

Our results demonstrate that Agp S1 is expressed in the guard cells of the leaves, the starch sheath cells of the stem, ovaries, root cap and root vascular tissues, suggesting that

Fig. 7. Transgenic tomato stem with Agp S1 promoter-Gus construct: (A) untransformed control stem; (B) starch sheath cells are stained along with cells within the vascular cylinder.

Agp S1 is expressed exclusively in sink tissues. In direct contradiction to the expression analyses obtained for the Agp S1 gene from potato [2], the tomato variant is clearly expressed in several tissues within the root that accumulate starch. Expression analysis of the Agp B gene from Arabidopsis also demonstrated strong expression in the root cap and root vascular tissue similar to our results with Agp S1 [30].

The three highly conserved sequences between the tomato and potato Agp S1 do not appear in sequence searches for any other organisms and their function is not known. We are currently creating truncated derivatives of the S1 promoter that contain different combinations of these elements to assess their function in gene expression.

4.2. Structure of the Agp S3 promoter

Agp S3 promoter has no TATAA box or Inr elements, which are generally found in eukaryotic promoters;

J. Xing et al./Plant Science 169 (2005) 882–893 889 Fig. 8. GUS staining of transgenic tobacco with Agp S3-gus construct in leaf, stem and root (A) whole leaf; (B) leaf segment; (C) stem; (D) root.

however, the promoters for most of the other nuclear genes associated with photosynthesis also do not contain TATAA boxes [31]. It was reported that relatively simple TATAA- less promoters, containing one or two additional activator sites, absolutely require an intact CCAAT box, while very strong TATAA-containing promoters, with several sites for powerful activators, are somewhat less critically dependent on the CCAAT box [32–34]. Agp S3 promoter does have a CCAAT box at position 112 bp upstream of the mRNA start site, and this box along with some additional upstream elements is essential for Agp S3 expression. Usually simple promoters without the TATA box and Inr elements have weak transcriptional activity, since TATA box and Inr sequences are very important for recruiting the RNA polymerase I transcriptional complex [35,36], but actually Agp S3 is very strongly expressed in leaf mesophyll cells. The Agp S3 promoter has multiple GC rich regions between 125 and 8 bp and these GC rich regions may help in stimulating Agp S3 expression [37–39]. Two DPE sites were also found at +105 and +120 bp, respectively, but these two DPE boxes may not contribute to the expression of Agp S3 promoter, since usually DPEs cooperate with Inr sequences [40]. The functions of the other elements in the Agp S3 promoter are also unknown. The Agp S3 promoter contains little evidence for upstream regulatory elements since there were no differences in expression between the 0.5 kb and larger promoter constructs. This is in contrast to the Agp S1 promoter, which contains at least four sucrose responsive elements between 3 kb and the transcription start site [18].

4.3. Agp S3 promoter is expressed in source tissues

The results from our GUS expression studies are consistent with northern analysis that indicated strong activity of Agp S3 in leaves [8], and demonstrate for the first time that Agp S3 is the predominant subunit isoform of AGPase expressed in the mesophyll tissue. Agp S3

Fig. 9. Agp S3 promoter expression in the flower 2 days before anthesis.

Fig. 10. Agp S3 promoter expression in the flower at anthesis: (A) view from front; (B) view from back.

transcription is not stimulated by sucrose, in contrast to the other subunit genes [18]. It is likely that Agp S3 is equivalent to the ApL1 gene of Arabidopsis since northern analysis indicated that ApL1 is the isoform most strongly expressed in leaves and its transcription is insensitive to sucrose [17]. Nevertheless, after a search of the Arabidopsis genome sequence, we were unable to find any similarity in the

50upstream region of the putative ApL1 gene with the Agp S3 promoter.

The guard cells are unusual since they strongly express both the S1 and S3 isoforms simultaneously. Guard cells are also unusual in that they hydrolyze starch during the day and resynthesize it at night, exactly the opposite of mesophyll cells. For the guard cells, at the beginning of the illumination period, starch is hydrolyzed and metabolized to malate, which serves in part as a counter ion to the K+ ions that are pumped into the guard cell. During the middle part of the day, however; the Calvin–Benson cycle in the guard cell chloroplast becomes active and sucrose replaces K+ ions as an osmoticum [41]. The S3 isoform of AGPase might function to regulate the amount of triose-P leaving the guard cell chloroplast for sucrose synthesis and respiration. However, photosynthetic activity is not likely to be the only source of sucrose in the guard cells. At the end of the day and at night sucrose is taken up from the guard cell apoplast [42]. There is a monosaccaride H+ ion symport in the guard cell plasma membrane that is transcriptionally upregulated during the dark period and may play a role in the reabsorption of sugars [43]. As the content of sucrose increases, the S1 isoform of the AGPase may increase through transcriptional activation to convert some of the sugars to starch. Upon illumination, starch is used again to produce malate during K+ ion uptake. Both isoforms of the AGPase are required in the guard cells because they behave as both source and sink tissue during the diurnal cycle.

In the flower, Agp S3 was expressed in the sepal, ovary and anthers at anthesis. Almost no expression was apparent

Fig. 1. Agp S3 promoter expression in the flower 2 days after anthesis: (A) top part of the flower; (B) bottom part of the flower.

Fig. 12. Agp S3 promoter expression in the fruit (12 DAA).

before and after anthesis. This is in contrast to the expression of Agp S1, which was high before and after anthesis, but with little or no expression at anthesis. Both flowers and developing fruit are capable of photosynthesis. In many cases this is confined to refixing the CO2 lost due to respiration. Flowers contain high levels of soluble sugars and maintain high respiration rates [4]; therefore it is not surprising that some of the sugars are conserved as starch. In one study [45], the photosynthesis of apple flower sepals were comparable to leaves, and flower photosynthesis contributed as much as 15–3% to flower carbohydrate balance during flowering and fruit set. The flower sepals most likely import carbohydrate prior to expansion. At anthesis they are capable of positive net photosynthesis and may synthesize starch during a period of considerable demand for photosynthates for growth, but limited sink strength. At fruit set and during fruit development, sepals rapidly senesce and no longer express Agp S3.

Although both sink and source tissue synthesize starch, the factors affecting the availability of glucose-1-phosphate and ATP for AGPase activity are different. In sink tissue, glucose-1-phosphate is derived from imported sucrose and ATP from respiration. Starch accumulation in sink organs such as fruit, roots or tubers occurs when there is a steady stable supply of these substrates. Sucrose activates the transcription of Agp S1 and must be present to maintain transcript levels over time [18]. Transcriptional control is a sufficiently sensitive mechanism to balance AGPase activity with substrate availability. However, in source leaves, starch accumulates transiently and functions to conserve fixed carbon that is not currently being used for growth, energy or transport to sink organs. Since photosynthetic activity can change rapidly in response to environmental factors, more flexible and dynamic mechanisms for control are warranted. Agp S3 contains a simple promoter with little evidence for functional cis-acting elements. The primary mechanism for control of enzyme activity is allosteric and a function of chloroplast PGA/Pi, which acts as a sensor for photosynthetic activity. In this way, AGPase activity can be rapidly balanced with substrate availability. 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.


[1] J. Preiss, N. Robinson, S. Spilatro, K. McNamara, Starch synthesis and its regulation in health. Regulation of carbon partitioning in photosynthetic tissue, in: Proceedings of the Eighth Annual Symposium in Plant Physiology, 1–12 January 1985, University of California, Rockville, Md.: American Society of Plant Physiologists, 1985, p. 1–26. [2] C.Y. Tsai, O.E. Nelson, Starch-deficient maize mutant lacking adenosine diphosphate glucose pyrophosphorylase activity, Science 151 (1966) 341–343.

[3] T.-P. Lin, T. Caspar, C.R. Somerville, J. Preiss, A starch deficient mutant of Arabidopsis thaliana with low ADP-glucose pyrophosphorylase activity lacks one of the two subunits of the enzyme, Plant Physiol. 8 (198) 1175–1181. [4] B. Muller-Rober, U. Sonnewald, L. Willmitzer, Antisense inhibition of the ADP glucose pyrophosphorylase in transgenic potatoes leads to sugar-storing tubers and influences tuber formation and expression of tuber storage proteins, EMBO J. 1 (1992) 1229–1238. [5] D.M. Stark, K.P. Timmerman, G.F. Barry, J. Preiss, G.M. Kishore,

Regulation of the amount of starch in plant tissues by ADP-glucose pyrophosphorylase, Science 258 (1992) 287–292. [6] M.K. Morell, M. Bloom, V. Knowles, J. Preiss, Subunit structure of spinach leaf ADP-glucose pyrophosphorylase, Plant Physiol. 85 (1987) 182–187. [7] B. Chen, H.W. Janes, Multiple forms of ADP-glucose pyrophosphorylase from tomato fruit, Plant Physiol. 113 (1997) 141–235. [8] B. Chen, H.W. Janes, T.J. Gianfagna, PCR cloning and characterization of multiple ADP-glucose pyrophosphorylase cDNAs from tomato, Plant Sci. 136 (1998) 59–67. [9] S. Park, W. Chung, Molecular cloning and organ-specific expression of three isoforms of tomato ADP-glucose pyrophosphorylase gene, Gene 206 (1998) 215–221. [10] L.A. Kleczkowski, P. Villand, J. Press, O.A. Olsen, Kinetic mechanism and regulation of ADP glucose pyrophosphorylase from barley leaves, J. Biol. Chem. 268 (1993) 6228–6233. [1] L.A. Kleczkowski, P. Villand, E. Luthi, O.A. Olsen, J. Press, Insensitivity of barley endosperm ADP glucose pyrophosphorylase to 3- phosphoglycerate and orthophosphate regulation, Plant Physiol. 101 (1993) 179–186. [12] J.M. Anderson, R. Larsen, D. Laudencia, W.T. Kim, D. Morrow, T.W.

Okita, J. Preiss, Molecular characterization of the gene encoding a rice endosperm-specific ADP-glucose pyrophosphorylase subunit and its developmental pattern of transcription, Gene 97 (1991) 199–205. [13] A. Tiessen, J. Hendriks, M. Stitt, A. Branscheid, Y. Gibon, E. Farre, P.

Geigenberger, Starch synthesis in potato tubers is regulated by posttranslational redox modification of ADP-glucose pyrophosphorylase: a novel regulatory mechanism linking starch synthesis to the sucrose supply, Plant Cell 14 (2002) 2191–2213. [14] B.T. Muller-Rober, J. Kossmann, L.C. Hannah, L. Willmitzer, U.

(Parte 2 de 3)