JXB Advance Access originally published online on November 8, 2004
Journal of Experimental Botany 2005 56(409):179-190; doi:10.1093/jxb/eri018
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
RESEARCH PAPER |
Selective transcriptional down-regulation of anther invertases precedes the failure of pollen development in water-stressed wheat
Institut de Recherche en Biologie Végétale, Université de Montréal, 4101, rue Sherbrooke est, Montreal H1X 2B2, Canada
To whom correspondence should be addressed. Fax: +1 514 872 9406. E-mail: hs.saini{at}umontreal.ca
Received 5 March 2004; Accepted 25 August 2004
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Water deficit during male meiosis in wheat (Triticum aestivum L.) causes pollen sterility. With a view to identifying the internal trigger for this failure, it was found that water stress specifically impairs the activities of vacuolar and cell-wall invertases in anthers prior to the arrest of pollen development. The enzymes are affected only when water deficit occurs around meiosis. Three invertase cDNAs, two encoding the cell-wall (Ivr1, Ivr3) and one the vacuolar (Ivr5) isoform, were isolated from an anther cDNA library. RNA gel-blot analysis using floral organs of well-watered plants revealed that these genes were expressed preferentially, though not exclusively, in anthers. Semi-quantitative RT-PCR demonstrated that transitory water deficit during meiosis selectively down-regulated the transcription of two of the three genes, one encoding the vacuolar (Ivr5) and the other a cell-wall (Ivr1) isoform, without affecting the Ivr3 message. Their expression did not recover upon resumption of watering. Another homologue of Ivr1 was also down-regulated, but only during the post-stress period. The stress effects on invertase transcripts were consistent with those on the developmental profiles of the corresponding enzyme activities. In situ hybridization revealed that the stress-sensitive invertase genes, unlike an insensitive one, were expressed within the microspores. No evidence for an invertase inhibitor under stress was found. Together the results show that the decline in invertase activity is probably regulated primarily at the transcriptional level in a gene- and cell-specific manner.
Key words: Gene expression, invertase, male sterility, pollen development, sugar metabolism, reproduction, water stress
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The sensitivity of a cereal plant towards water deficit varies considerably over its life cycle (Salter and Goode, 1967
). With the exception of seed germination, reproductive development is arguably the most water-stress-sensitive phase in the life of a cereal crop (reviewed in Salter and Goode, 1967; O'Toole and Moya, 1981; Saini and Westgate, 2000). Within the reproductive phase, the sensitivity of male organs increases dramatically from the start of meiosis to the break-up of the tetrad, events that last approximately 24 h in a single anther (collectively referred to here as ‘meiosis’; Fig. 1a). By contrast, the female tissue remains quite insensitive to water stress during the same period (Saini and Aspinall, 1981). Water deficit during meiosis induces pollen sterility, leading to a failure of fertilization and hence grain set (reviewed in Saini, 1997; Saini and Westgate, 2000). The impact of this on cereal yield under rain-fed conditions can be quite dramatic (Salter and Goode, 1967).
|
In wheat, pollen development often fails when a brief episode of even moderately severe water deficit coincides with meiosis (reviewed in Saini, 1997
). The failure is not due to a general desiccation of the reproductive tissue, because the inflorescence, which at this stage is completely enclosed within a tube formed by the sheaths of the two uppermost leaves (Fig. 1A), can maintain normal water status while the vegetative parts suffer substantial water loss (Morgan, 1980; Saini and Aspinall, 1981; Westgate et al., 1996; Saini, 1997). Therefore, it appears that some cellular lesion, which is probably triggered by an as yet unidentified signal from the vegetative organs, inhibits pollen development (Saini, 1997; Saini and Westgate, 2000).
The major events during normal and water-stress-induced abortive development of the male reproductive organs of wheat are summarized in Fig. 1. When water is withheld briefly during meiosis and re-supplied immediately thereafter, the meiotic division is completed but the ensuing pollen development is arrested a few days later (Saini et al., 1984; Lalonde et al., 1997a). However, the post-meiotic stages at which the effects of stress imposed earlier during meiosis become visible (Fig. 1C–E) are themselves insensitive to concurrent water deficit (Saini and Aspinall, 1981). These observations indicate that meiosis harbours a highly stress-sensitive event that is crucial for subsequent male gametophyte development.
Normal pollen grains of wheat and other cereals accumulate large quantities of starch during their final stages of development (Fig. 1D, E) (Saini et al., 1984; Franchi et al., 1996). Stored starch is used to support pollen germination and growth of the pollen tube (Pacini and Franchi, 1988; Clément et al., 1994). The most conspicuous manifestation of metabolic disruption in water-stress-affected wheat pollen grains is their failure to accumulate starch (Saini et al., 1984; Lalonde et al., 1997a). In an effort to locate the point(s) within carbohydrate metabolism where this deficiency is regulated, it was found that water stress during meiosis caused a precipitous depression in the activity of soluble acid invertase in anthers (Dorion et al., 1996). This effect precedes any visible developmental lesion and is not reversed even after the stress is relieved. It is also quite specific, as other enzymes in the pathway leading to starch biosynthesis do not suffer the same fate (Dorion et al., 1996). The decline in invertase activity is accompanied by the accumulation of sucrose, a change in the profile of other sugars, and some spatial redistribution of starch within the anther (Dorion et al., 1996; Lalonde et al., 1997a). A similar pattern of events is also evident in rice (Sheoran and Saini, 1996). Taken together with the fact that invertase is the dominant sucrolytic enzyme in wheat anthers (Dorion et al., 1996), the timing and nature of these events suggest that the blockage of invertase-mediated sucrose utilization by anthers could be a critical trigger in the failure of pollen development under stress.
With the aim of further dissecting the regulation of this important event during the induction of pollen sterility by water deficit, the relative stress-sensitivity of the soluble and the wall-bound invertase isoforms, in relation to the timing of stress for maximal effect on pollen fertility, were ascertained first. The possible role of any inhibitor in lowering invertase activity was also ruled out. Using three invertase cDNAs cloned from a wheat anther cDNA library, it was then demonstrated that the effect of water stress is primarily at the transcriptional level and is highly gene-specific. Moreover, in situ hybridization revealed that the effect was also cell-specific.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Plant growth, stress treatment and sampling
Wheat (Triticum aestivum L. cv. Katepwa) plants were grown under controlled environmental conditions as described previously (Dorion et al., 1996
). To ensure precision and uniformity of stress among different batches of plants, pots were filled with equal volumes of potting-mix, each pot was sown with 6 seeds, and 3 weeks after germination, 2 seedlings were retained in each pot so that all plants were of uniform size. Pots were watered to field capacity and then stress was initiated by withholding water for 4 d such that most florets underwent meiosis on the last day of stress (Dorion et al., 1996). Normal watering was resumed at the end of the fourth day, when the average leaf water potential had dropped to –2.3 MPa compared with approximately –0.5 MPa in the controls (Dorion et al., 1996). Anthers were sampled at the conclusion of meiosis just before the stress treatment ended, and at the following developmental stages after the resumption of watering (approximate time after the resumption of watering given in the parentheses): young microspore (1–2 d), vacuolate microspore (3–6 d), first pollen grain mitosis (7–8 d), second pollen grain mitosis (9–10 d) and anthesis (11 d). The actual stages of development were confirmed by microscopic examination of the anthers stained with ferric acetocarmine. Anthers from well-watered plants, as well as other floral tissues were harvested at the same stages. Tissues from multiple batches of plants were pooled to obtain a large enough representative population for each developmental stage. Three random samples were drawn from this pool for various analyses, and each analysis was repeated at least three times.
Estimation of vacuolar and cell wall invertase activities
Anthers (30–50 mg fresh weight) were ground to a fine powder with liquid nitrogen, and then homogenized in 1 ml extraction buffer containing 50 mM HEPES-KOH pH 7.4, 5 mM MgCl2, 1 mM EDTA, 5 mM 1,4-dithiothreitol (DTT), and 2% (w/v) polyvinylpyrrolidone (PVP). The homogenate was centrifuged for 10 min at 4 °C at 16 000 g. Supernatant was desalted on G-25 microspin columns, and 20 µl of it was used for the assay of vacuolar invertase according to Tsai et al. (1970). The reaction mixture (200 µl) containing 20 mM sucrose in 100 mM sodium acetate, pH 4.8, was incubated at 37 °C for 30 min. The reaction was stopped by adding Nelson's copper reagent, and the reducing sugars released were measured by the method of Nelson (1944). Reaction stopped at time zero served as blank.
Cell wall invertase activity was estimated according to the protocol described by Appeldoorn et al. (1997). The procedure was similar to that for vacuolar invertase except that the extraction mixture did not contain PVP or DTT, and the activity was assayed in the pellet instead of the supernatant. The pellet was washed three times with the extraction buffer to remove all soluble invertase, and was then incubated for 18 h at 4 °C in 1 ml of 20 mM MES-KOH buffer, pH 6.0, and 1 M NaCl to dislodge the bound invertase protein from the cell wall. The sample was then centrifuged and the activity was determined as described for vacuolar invertase.
A mixture of 10 µl enzyme extract each from control and stressed anthers was used for the mixed-extract assays.
Isolation of nucleic acids
Total RNA, isolated using the RNeasy kit (Qiagen, Mississauga, ON, Canada) was eluted in DEPC-treated water and was quantified spectrophotometrically. Genomic DNA from 1 g of freshly collected leaves from well-watered plants was isolated using the protocol of Murray and Thompson (1980). Briefly, plant tissue was homogenized in 2.5 vols of 50 mM TRIS-HCl, pH 8, 2% (w/v) hexadecyl-trimethylammonium bromide (CTAB), 50 mM EDTA, 0.7 M NaCl, and 0.1% (v/v) ß–mercaptoethanol. The homogenate was extracted with chloroform-isoamylalcohol, and the DNA in the aqueous phase was precipitated with isopropanol.
Amplification of invertases by RT-PCR
The following primers, based on the known invertase sequences, were used to amplify invertase fragments by RT-PCR (Tymowska-Lalanne and Kreis, 1998; Sturm, 1999): the first primer, 5'-ATGAA(T/C)GA(T/C)CCNAA(T/C)GGNCCT-3', represented a conserved motif common to all invertases (MNDPNGP). The second primer was based on conserved motifs differing in one amino acid (WECPDFY or WECVDFY) specific to cell-wall or vacuolar invertase, respectively. The respective sequences of these primers were 5'-(A/G)TA(A/G)AA(A/G)TCNGG(C/T)TCCCA-3' and 5'-(A/G)TA(A/G)AA(A/G)TC(A/G)CA(C/T)TCCCA-3'. Restriction sites for EcoRI (GAATTC) and BamHI (GGATCC) were added at the 5' ends of primers 1 and 2, respectively, to allow directional cloning of the cDNA fragment into pBluescript. Total RNA from a mixture of anthers between meiosis and anthesis was used for the reverse transcription. The first strand cDNA was synthesized using the RT-PCR kit from Boehringer Manheim using 1 µg of RNA and 2.5 ng µl–1 of oligo-p(dT)15 in a total reaction volume of 20 µl. Five µl of the RT mixture was used as template for PCR in a 100 µl reaction mixture containing 2.5 mM MgCl2, 0.2 mM dNTP, 0.2 µM of each primer, and 2.5 units of Taq DNA polymerase (Roche Diagnostics, Laval, QC, Canada). After initial DNA denaturation at 95 °C for 5 min, 35 cycles of PCR were carried out, each comprising 30 s at 94 °C, 30 s at 58 °C, and 1 min at 72 °C. PCR products were electrophoresed on a 1% (w/v) agarose gel, visualized by staining with ethidium bromide, and purified using the Gene Clean Kit (BIO 101, Vista, CA, USA). The PCR products were digested with EcoRI and BamHI and subcloned into pBluescript phagemid, and transformed into XL-1 Blue MRF' E. coli.
Construction and screening of cDNA library and DNA sequence analysis
Polyadenylated RNA was isolated from total RNA extracted from a pool of anthers representing all stages between meiosis and anthesis using the Clontech mRNA Isolation Kit (Clontech, Palo Alto, CA, USA). Five µg of the poly(A) RNA was used to construct a cDNA library using the UniZap cDNA Synthesis Kit (Stratagene, La Jolla, CA, USA), according to the manufacturer's instructions. The amplified library had a titre of 4x109 pfu ml–1. The library (3x105 pfu) was screened with the radiolabelled DNA probes representing cell-wall and vacuolar invertase, amplified via PCR as described above. The putative positive plaques were subjected to three more rounds of screening at lower plaque densities, and purified by in vivo excision from the phagemids using the ExAssist helper phage/SOLR cell system (Stratagene).
The cDNAs were sequenced on both strands using the DNA Core Services at the University of Calgary, Alberta. The sequences were analysed using the BLAST network services at the National Center for Biotechnology Information and the MacVector/Clustal W version 6.5.3 software.
Southern and northern blot analyses
Genomic DNA (20 µg) digested with EcoRI, BamHI, KpnI, HindIII, and SacI/XhoI, was electrophoretically separated on a 1.2% (w/v) agarose gel. Fifteen micrograms of total RNA was electrophoresed on a MOPS-formaldehyde gel. The DNA was transferred onto Hybond N nylon membranes (Amersham Biosciences, Buckinghamshire, UK) via capillarity in 10x SSC, the membranes were rinsed with 10x SSC, and baked at 80 °C for 2–4 h. The RNA was electrophoretically transferred to positively charged nylon membranes and cross-linked using a UV cross-linker. The membranes were hybridized under high stringency conditions (60–65 °C in 250 mM sodium phosphate buffer, pH 7.0, 7% (w/v) SDS, 1% (w/v) BSA, 1 mM EDTA) with DNA probes prepared by labelling with [
-32P]dCTP using the High Prime Random Prime labelling kit (Roche Diagnostics). The hybridized membranes were washed with 5x SSC plus 0.1% (w/v) SDS for 60 min at 65 °C, 1x SSC plus 0.1% (w/v) SDS for 60 min at 50 °C, and finally rinsed twice with 0.1x SSC plus 0.1% (w/v) SDS for 15 min at 50 °C. The autoradiograms of the blots were obtained by exposing them to XAR-5 Kodak X-ray films at –80 °C.
Semi-quantitative RT-PCR analysis
The primers set CWIF (5'-CATGAGGGGGATCGCGGTGTTGTA-3') and CWIR (5'-ACCCTTGACGGCCTTGTTGCTGAC-3') was used to amplify the cell wall invertase Ivr1, the set CWIIF (5'-GTGGAGGATGGCAGTTGGTGGTGA-3') and CWIIR (5'-GCTCTATTCCTTGATGGCTGA-3') to amplify the cell wall invertase Ivr3, and the set VAF (5'-CAACGACTCCCTCCTCCGCAACT-3') and VAR (5'-TTCTCGTCCAGCTCCACCGTCCTC-3') to amplify vacuolar invertase Ivr5. Primers A (5'-CAGAAGAACGCGACCATCAAGGAC-3') and B (5'-CGACAAAGGCGACGGAAAACC- 3') were included as the internal standards to amplify ADP-glucose pyrophosphorylase (AGPase), a choice based on the following rationale: notwithstanding the inhibition of starch accumulation in water-stress-affected anthers and the role of AGPase as the rate-limiting enzyme for starch synthesis under normal conditions, the activity (Dorion et al., 1996) and gene expression (Lalonde et al., 1997b) of this enzyme in water-stress-affected anthers do not decline much until nearly the end of anther development. Moreover, the AGPase gene expression in well-watered plants is relatively constant at the critical early stages (Lalonde et al., 1997a). This was fortuitous considering that standards that remain absolutely constant are extremely difficult to find for this type of experiment because the effects of water stress greatly confound the considerable changes in gene expression occurring due to the anther's progression through many developmental stages over a short period.
RT-PCR was carried out using the one-step RT-PCR Kit (Qiagen) under the recommended conditions with the following modifications to optimize the procedure: the concentration of each dNTP was 200 µM instead of 400 µM, and the final primer concentration was decreased from 0.6 µM to 0.4 µM. The number of PCR cycles was optimized so that the product formation was terminated before saturation. The internal standard primers A and B were also added to the reaction mixture at a final concentration of 0.4 µM. The amplification cycles were as follows: 25 cycles of 30 s at 94 °C, 30 s at 60 °C and 1 min at 72 °C; a final extension step of 10 min at 72 °C was used to terminate the reaction. The PCR products (5–10 µl) were separated by electrophoresis on a 1% (w/v) agarose gel and then stained with ethidium bromide. Negative controls with no RNA were always included to ensure that product formation was not due to cross-contamination. The identity of the PCR product was confirmed by nucleotide sequencing.
In situ hybridization
For in situ analysis, anthers at VM and PM-1 stages were fixed in FAA (formaldehyde/glacial acetic acid/ethanol; 5/5/63% (v/v)) at 4 °C for 24 h. The samples were then dehydrated in a graded ethanol/3°-butanol series as described by St-Pierre et al. (1999). The samples were infiltrated with Paraplast Plus (Fischer Scientific, Montreal, QC, Canada) as follows: 1:1 Paraplast:100% (v/v) 3°-butanol for 3 h at 62 °C, 100% Paraplast for 5 h at 62 °C and again in 100 Paraplast 16 h at 62 °C. The samples were then embedded in 100% Paraplast and polymerized at 4 °C.
pBluescript plasmids harbouring the full-length Ivr1 and Ivr5 clones were linearized with ApaI and BamHI whereas Ivr3 was linearized with KpnI and XbaI. Sense and antisense RNA transcripts were synthesized by T3 and T7 RNA polymerase with digoxigenin-UTP (Roche Diagnostics) as label. Since transcripts were longer than 700 bp, alkaline degradation was performed to obtain a probe of approximately 400 bp (Jackson, 1992).
Pre-hybridization and hybridization were done as described by St-Pierre et al. (1999). After hybridization, the slides were treated with RNase A (50 µg ml–1) at 37 °C for 30 min, followed by 1 h washes at room temperature in 2x SSC, 1x SSC, and 0.1x SSC, respectively. The slides were then washed for 10 min at room temperature in buffer I (50 mM TRIS-HCl, pH 7.5, 75 mM NaCl, 0.15% (v/v) Triton 100X), and again for 30 min in buffer I containing 0.2% (w/v) BSA. The sections were incubated in buffer I containing 1% (w/v) BSA and a 1:200 dilution of anti-digoxigenin-alkaline phosphatase conjugated Fab fragments, and were then washed at room temperature twice for 15 min each in buffer I and twice more for 10 min each in buffer 2 (100 mM TRIS-HCl, pH 9.5, 100 mM NaCl, 10 mM MgCl2). Signal was developed for 16 h in buffer II containing 18.75 mg ml–1 Nitro blue tetrazolium chloride (NBT) and 9.4 mg ml–1 5-bromo-4-chloro-3-indolyl phosphate, toluidine salt (BCIP). Reaction was stopped by immersing slides in distilled water. The sections were dehydrated through a gradient series of ethanol, mounted with Permount, and observed with a Leitz Orthoplan light microscope. Pictures were taken using a Nikon COOLPIX 995 digital camera. Sections stained with toluidine blue-O were included for comparison of structure with in situ message localization.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Water deficit during meiosis irreversibly impairs invertase activity in anthers
Vacuolar (soluble-acid) and cell-wall-bound invertase activities in anthers of well-watered plants increased steadily and substantially from meiosis to anthesis (Fig. 2). A transient episode of water deficit during meiosis caused an immediate decline in the activities of both isoforms (Fig. 2); the effect is more clearly visible in Figs 3 and 4. Although the activities in stress-affected anthers recovered partially upon re-watering, they remained substantially below those in the well-watered plants (Fig. 2).
|
|
|
The sensitivity of invertase to water stress was focused around meiosis (Fig. 3). Whereas the water deficit coincident with meiosis caused a substantial decline in the activities of both invertase isoforms (45–50%), the effect was much smaller if the stress was imposed just 1–2 d later at the young microspore stage. Water stress had no significant effect when imposed later during pollen development.
A mixed-extract assay was carried out to determine if an inhibitor of invertase could be responsible for the decline in the enzyme activity under water stress during meiosis. Although, for both forms of the enzyme, the activity in a mixture containing equal amounts of extracts from well-watered and stressed anthers was slightly lower than the theoretical average of activities for the two extracts (Fig. 4), the difference was far too small to indicate the presence of any invertase inhibitor.
Isolation of three anther-expressed invertase cDNAs
The cDNAs for vacuolar and cell-wall forms of invertase expressed in anthers were amplified via RT-PCR using degenerate primers for the two types of genes. The primer set for vacuolar and cell-wall invertase yielded a 550 and a 500 bp fragment, respectively. These PCR products were purified from agarose gels, cloned into pBluescript SK(+) at the EcoRI-BamHI site and used to transform XL1-Blue MRF' E. coli cells. Sequence analysis of plasmids from transformed colonies revealed three distinct groups of clones. One representative clone from each of these groups was further characterized. Clone 17 shared 72% identity with cell wall invertase from various species, including Avena sativa and tomato (accession numbers X73601 and X91390). Clone 34 shared between 66–69% identity with cell wall invertases from carrot, tobacco, and Arabidopsis thaliana (accession numbers X69321, X81834, and X70691). Clone 45, representing the third group, shared approximately 50% identity with vacuolar invertase genes from tulip and maize (accession numbers X95651 and U16123). The three PCR products did not cross-hybridize to one another (data not shown), and therefore, were presumed to have non-homologous sequences representing three distinct genes.
An anther cDNA library was screened with the above PCR products (i.e. clones 17, 34 or 45). After four rounds of screening, three full-length or near full-length cDNAs were isolated and were designated Ivr3, Ivr1, and Ivr5, respectively. Ivr1 and Ivr3 cDNAs (accession numbers AF030420 and AF030421) shared high homology with several known cell-wall invertase genes, whereas Ivr5 (accession number AF069309) showed high homology to the vacuolar form of invertases (see details below).
The details of Ivr1, a full-length cell-wall invertase cDNA encoding a protein with a calculated molecular mass of 66.24 kDa, have already been published (Minhas and Saini, 1998).
Ivr3 represents a partial cDNA containing an open reading frame (ORF) starting at position 1 (not an ATG start codon) and ending at position 1594. This partial ORF encodes a polypeptide of 531 residues with a calculated molecular mass of 59.64 kDa and an isoelectric point of 8.6, characteristic of cell wall invertases (Albersheim, 1975; Godt and Roitsch, 1997). The deduced protein sequence of Ivr3 shared 51% identity with cell wall invertase from tobacco (accession number X81834), and contained the conserved WECPDFY motif specific to cell-wall invertases (Sturm and Chrispeels, 1990; Sturm, 1999).
The 1840 bp Ivr5 cDNA lacked the ATG start codon. The longest ORF ended with a stop codon at position 1528 and encoded a protein of 509 amino acids. The deduced protein had a calculated molecular mass of 57 kDa and an isoelectric point of 5.9, typical of vacuolar invertases (Albersheim, 1975). The protein encoded by Ivr5 shared high homology with several vacuolar invertases, including that from maize (78% identity; accession number U16123). The homology with cell wall invertases was lower; for example, 39–40% identity with cell wall invertases from A. thaliana and Vicia faba (accession numbers X70691 and Z49831). Ivr5 cDNA contained the motif WECIDFY instead of the usual WECVDFY signature characteristic of most vacuolar invertases. This motif has, however, also been observed in vacuolar invertases of other plant species (Sturm, 1999).
Ivr3 and Ivr5 cDNAs contained untranslated 3' regions of 141 and 312 bp, respectively, each ending with a poly(A) tail. The consensus polyadenylation signal AATAAA, which is usually located 10–40 bp upstream of the poly(A) tail, was missing in both cDNAs. This could have been replaced by the sequence AATAAG located 39 and 50 bp upstream of the poly(A) tail in Ivr3 and Ivr5, respectively. Such deviations from consensus sequence have been shown for several other plant mRNAs (Heidecker and Messing, 1986). The comparison of Ivr3 and Ivr5 lengths with the sizes of the mRNAs they hybridized with (Fig. 5), indicates that these cDNAs represent over 80% and 70% of the respective full clones.
|
Southern blot of genomic DNA probed with Ivr1, Ivr3, or Ivr5 showed that each of the cDNAs hybridized to only a few restriction fragments, but the hybridization pattern was distinct for each probe (data not presented). This indicated the presence of at least three invertase genes in wheat, each with a single or a few copies.
Invertase cDNAs have distinct developmental expression patterns in tissues of well-watered plants
Expression of the invertase cDNAs in anthers of well-watered wheat plants was determined in various floral and vegetative tissues at different developmental stages using northern analysis of total RNA (Fig. 5). Ivr1, Ivr3, and Ivr5 recognized messages of 2.0, 2.1, and 2.4 kb, respectively, in all tissues that expressed these genes (Fig. 5). Predictably, all three genes were expressed in anthers, the source material for the cDNAs (Fig. 5A). However, only Ivr5 was expressed in the pistil (Fig. 5B) whereas Ivr3 and Ivr5 were expressed in glumes (Fig. 5C). None of these messages were present in the leaf (results not shown).
The developmental pattern of expression of these genes was also distinct in each tissue examined. The expression of Ivr1 in anthers, the only tissue to express this gene, was undetectable at the early developmental stages (meiosis, young microspore) but increased dramatically during mid-development and then fell sharply towards anther maturity (Fig. 5A). The Ivr3 message was extremely faint at the early stages of anther development, increased during mid-development, and was maximally expressed towards maturity (Fig. 5A). Ivr3 was also expressed at a constitutively low level in the glumes (Fig. 5C). The vacuolar invertase, Ivr5, was the only gene expressed in all three floral tissues examined. Its expression pattern in anthers and pistil was similar; the message was undetectable during the early stages, was expressed at low levels during mid-development and peaked sharply towards maturity (Fig. 5A, B). Conversely, in glumes, the expression of Ivr5 was maximal during meiosis and declined gradually thereafter throughout the development (Fig. 5C).
Overall, despite easily measurable enzyme activities in anthers (Figs 2, 3), the invertase gene expression was rather low and difficult to detect through northern hybridization, especially during the early developmental stages (Fig. 3A).
Water deficit impairs invertase expression in a highly gene-specific manner
RNA gel-blot analysis of well-watered anthers was intended to establish a baseline against which the effects of water deficit on invertase gene expression could be measured, but it failed to detect the expression of any of the invertase genes at the most critical meiotic stage (Fig. 5A). The low sensitivity of this technique, despite several attempts, also made it difficult to visualize the developmental pattern of gene expression during the early stages. Therefore, the more sensitive semi-quantitative RT-PCR (SQRT-PCR) was used to determine if impaired transcription was responsible for the observed stress-induced decline in invertase activity.
The SQRT-PCR analysis, using total RNA and primers specific for Ivr1, Ivr3, and Ivr5, focused on the three earliest stages of anther development (Fig. 6). Co-amplification of AGPase served as the internal standard because it had been shown earlier that water stress had very little effect on the developmental expression pattern of this gene in anthers (Lalonde et al., 1997b; see also details in the materials and methods). This enabled sample normalization and comparison of transcript abundance across multiple samples. The internal AGPase control co-amplified very well with Ivr3 and Ivr5 primers, yielding a product of 404 bp (lower band in Fig. 6B, C). For unknown reasons, AGPase primers did not give a product when used together with Ivr1 primers (Fig. 6A), but did give the right product when the competing Ivr1 primers were excluded (data not shown). Nevertheless, because PCR with all three primers was done simultaneously and using aliquots of the same RNA extracts, amplification of internal standards in Fig. 6A and B are reasonable evidence of reference constancy.
|
Amplification with Ivr3 and Ivr5 primers gave the expected 566 and 597 bp products (Fig. 6B, C), which when sequenced showed 100% identity with the relevant regions of the respective invertase cDNAs (accession numbers AF030421 and AF069309). Ivr1 primers, on the other hand, yielded two products: the anticipated 577 bp product plus another longer fragment of 683 bp (Fig. 6A). While the sequence of 577 bp product shared 100% identity with the corresponding region of Ivr1 (accession number AF030420), the 683 bp product had a more divergent sequence. The latter contained two segments (407 and 121 bp) that shared 83% identity with stretches of nucleotides in Ivr1, starting at positions 724 and 1128, respectively, but were separated by a 130 bp non-homologous segment. The sequence of this product also contained the WECPDF motif characteristic of cell-wall invertases. Therefore, it was concluded that this amplification product probably represents another distinct cell-wall invertase gene expressed in anthers. Hence, the short (577 bp) and the long (683 bp) products amplified with Ivr1 primers were designated as Ivr1S and Ivr1L, respectively.
In well-watered plants, Ivr1S transcript declined while Ivr1L increased as anthers developed (Fig. 6A). An episode of water deficit during meiosis caused a substantial and immediate decline in Ivr1S transcript, and the repression lingered even after the plants were rewatered. On the other hand, the Ivr1L transcript did not suffer any decline during stress and may even have increased slightly. However, during the post-stress phase, it became progressively less abundant in anthers that had been stressed during meiosis. By contrast, Ivr3 transcript was not affected at all by water deficit (Fig. 6B). The level of Ivr5 transcript was down-regulated by water stress during meiosis, and this effect also persisted after rewatering (Fig. 6C). Put together, these results show that water deficit during meiosis causes an immediate down-regulation of one cell-wall invertase gene, suppresses another related cell-wall invertase gene during the post-stress phase, and has no effect on a third cell-wall gene (Fig. 6A, B). The stress also down-regulates the only vacuolar invertase gene whose expression was detected in the anthers (Fig. 6C).
Expression of water-stress-sensitive genes is cell specific
The stamens used for in situ hybridization were from well-watered plants and at the developmental stage at which the invertase gene in question was maximally expressed (Fig. 5). Figure 7A–C shows anther sections stained with toluidine blue-O to reveal the general structure, Fig. 7D–F represent hybridization with the non-reacting sense RNA probes (control), and Fig. 7G–I are the corresponding sections hybridized with the antisense probes. The three invertase genes were differentially expressed in different cell types within the stamen. Ivr1, a cell-wall invertase, was strongly expressed in the pollen grains and tapetum, and very slightly in the cells associated with the vascular tissue within the filament (Fig. 7G compared with 7D control). Ivr3, the second cell-wall invertase gene, was not expressed in the pollen grains but was highly expressed in the tapetum and the vascular tissue (Fig. 7H compared with 7E control). The vacuolar invertase gene Ivr5 was expressed copiously in the pollen grains, tapetum and the filament cells (Fig. 7I; compared with Fig. 7F control). The most significant observation from this analysis was that the two genes that had been found to be down-regulated by water stress (Ivr1 and Ivr5) were expressed inside microspores or pollen grains whereas the gene that was insensitive to stress (Ivr3) was not (cf. Figs 6 and 7).
|
Extremely low expression of invertase genes at the critical meiotic stages and rapid developmental changes in expression make it nearly impossible to find a control and a water-stress-affected anther that are so perfectly matched in development that the differences in expression between them could be ascribed solely to the effect of stress. Therefore, it was concluded that in situ hybridization would not yield a valid comparison of spatial pattern of expression between stressed and control anthers. Nevertheless, the results in Fig. 7 do demonstrate where the genes, whose sensitivity to water stress had already been demonstrated (Fig. 6), would be normally expressed.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Impairment of invertase is suspected to be a key causal step in the induction of pollen sterility in wheat by water deficit (Dorion et al., 1996
). With the aim of determining the molecular basis for this metabolic dysfunction, it has been shown here that water deficit during pollen mother cell meiosis causes a transcriptional down-regulation of invertase gene expression in stamens. The effect is highly selective: not only does the stress not affect other closely related enzymes (Dorion et al., 1996), it discriminates even among different genes encoding the same class of invertases (Fig. 6); only the invertases expressed within the microspores/pollen grains are repressed (Fig. 7); and water stress affects invertase activity mainly when it coincides with meiosis (Fig. 3). The suppression of invertase by water stress occurs well before any visible disruption in pollen development, suggesting a potential causal role for this lesion in the male reproductive failure (Dorion et al., 1996).
The risk of decline in grain yield because of extreme sensitivity of cereal microsporogenesis to water deficit during meiosis has been well documented since the 1930s (reviewed in: Salter and Goode, 1967; O'Toole and Moya, 1981; Saini, 1997; Saini and Westgate, 2000). Although research over the last two decades has described the cellular and metabolic processes associated with this failure (Saini, 1997; Saini and Westgate, 2000), the primary trigger for the induction of pollen sterility by water deficit remains to be identified. A significant advance towards this goal occurred with the finding that, in wheat and rice anthers affected by meiotic stage water deficit, a precipitous decline in the activity of soluble acid (vacuolar) invertase precedes any other sign of developmental failure (Dorion et al., 1996; Sheoran and Saini, 1996). The present work demonstrates that water deficit also affects cell-wall bound invertase in the same manner—both activities decline in response to a brief, transient episode of water deficit during meiosis (Figs 2, 3, 4) and both remain well below the control levels even after the plants are rewatered (Fig. 2).
To dissect the molecular regulation of this malfunction further, three invertase cDNAs that are preferentially expressed in wheat anthers were isolated. It has been shown here through SQRT-PCR that water deficit affects invertase activity by down-regulating the transcription of two of these genes, one encoding a cell-wall form (Ivr1) and the other a vacuolar form (Ivr5) of the enzyme (Fig. 6). The third gene, which also codes for another cell-wall invertase, remains unaffected. Moreover, SQRT-PCR analysis revealed that an additional invertase gene, which was not picked up during library screening, is probably expressed in anthers. In spite of its high homology to Ivr1, the expression of this cell-wall invertase gene is not repressed (is slightly enhanced) immediately by water stress during meiosis, but is down-regulated later during the post-stress period (Fig. 6A). Thus, the inhibition of invertase gene expression by water deficit is selective to the extreme degree of discriminating between different members of a family encoding the same isoform. This high selectivity apparently also extends to the spatial localization of invertase transcripts, as the two stress-sensitive genes were found to be expressed within the microspores/pollen grains whereas the insensitive one was not (Fig. 7).
The developmental profile of cell-wall invertase activity in anthers of well-watered plants (Fig. 2) would represent a net measurement of the products of three or possibly more genes (Fig. 4). However, while the activity increased steadily during development (Fig. 2), the sum of the three transcripts remained relatively constant (Fig. 6A, B). This indicates that the three transcripts may be differentially translated or their proteins may be differentially active. On the other hand, the developmental pattern of the activity of vacuolar form of invertase (Fig. 2) was in general accord with the profile of the corresponding Ivr5 transcript, at least for the stages at which both measurements were made (Fig. 6C).
There was a better concordance between the effects of water deficit on enzyme activity and transcript levels. The net suppression of the two transcripts that were sensitive (Ivr1S and Ivr1L) and the activity profile of the cell-wall-bound invertase were mutually consistent (cf. Figs. 6A and 2). The same was also true for the vacuolar invertase activity and the corresponding Ivr5 transcript (cf. Fig. 6C and 2). As no evidence for an invertase inhibitor under water stress was found (Fig. 4), the results indicate that the decline in invertase activity in response to water stress is probably regulated primarily at the transcriptional level.
What are the implications of this for the regulation of water-stress-induced male sterility? Wheat and other cereal pollen grains accumulate large quantities of starch (Franchi et al., 1996; Lalonde et al., 1997a), which is later used to support germination and pollen tube growth (Pacini and Franchi, 1988; Clément et al., 1994). The most conspicuous feature of water-stress-affected wheat pollen grains is a near-complete absence of starch (Saini et al., 1984; Lalonde et al., 1997a). Yet, the enzymes of starch biosynthesis in anthers are not affected by water deficit (Dorion et al., 1996; Sheoran and Saini, 1996; Lalonde et al., 1997b). This points to a metabolic failure upstream of starch formation. In wheat, sucrose is the primary sugar imported into sinks, where it must be cleaved by either invertase or sucrose synthase before its entry into further metabolism. In pollens and anthers of wheat and several other species, invertase is by far the dominant enzyme of sucrose cleavage, particularly during early development (reviewed in Saini and Westgate, 2000). Both cell-wall and vacuolar types of invertase play crucial roles in carbohydrate utilization; the former regulates sucrose unloading whereas the latter is more important in the cell metabolism (Roitsch, 1999; Sturm, 1999). Therefore, a down-regulation of invertase gene expression is certain to block the provision of hexoses to sustain pollen development, a process that involves the synthesis and storage of large quantities of carbohydrates (Pacini and Franchi, 1988; Goldberg et al., 1993; Clément et al., 1994). Thus, starvation resulting from the observed dramatic downturn in the activities of both forms of invertase (Figs 2, 3, 4) could be a primary cause of pollen death under water stress. The fact that this happens rapidly after stress and prior to any other detected abnormality indicates that this may well be the trigger for the arrest of pollen development and consequent male sterility (Dorion et al., 1996). A recent report that antisense inhibition of Nin88, a cell-wall invertase that, like Ivr1, is expressed in pollen grains and tapetum, induces pollen sterility in tomato, lends further support to this hypothesis (Goetz et al., 2001).
Finally, wheat floral organs maintain high internal water status even in the face of substantial leaf drying during water deficit (see detailed review in Saini and Aspinall, 1981; Morgan and King, 1984; Westgate et al., 1996; Saini and Westgate, 2000). Thus, the effects of water stress on invertase within stamens must be remotely regulated by some signal from water-deficient parts of the plant. Much research has focused on abscisic acid as a sporocidal signal but its role remains inconclusive (reviewed in Saini, 1997; Saini and Westgate, 2000). Alternatively, sugars could be involved in this signalling since sugar flux to anthers can be affected upon inhibition of photosynthesis under water stress (Boyer and McPherson, 1975; Hanson and Hitz, 1982) and sugars modulate the expression of various genes, including invertase (reviewed in Koch, 1996; Smeekens and Rook, 1997; Roitsch, 1999). As abscisic acid, water stress, and glucose all enhance invertase gene expression in vegetative tissues (Trouverie et al., 2004), an interaction between abscisic acid and sugars in modulating invertase activity in anthers is possible. Hence, pollen development in water-stressed wheat plants offers an interesting model in which the role of sugars and other signals in regulating source–sink relations and carbon allocation could be examined at the whole plant level.
|
Acknowledgements |
|---|
This research was supported by a Discovery Grant to HSS from the Natural Sciences and Engineering Research Council of Canada. The authors are thankful to Dr Rajinder Dhindsa of McGill University, Canada for his valuable comments on the manuscript and to Ms Nadia Araar for her help with tissue sampling.
|
Footnotes |
|---|
* Present address: Central Potato Research Institute, Shimla, Himachal Pradesh, India 171 001.
Present address: Collège Bois-de-Boulogne, 10555, Ave de Bois-de-Boulogne, Montreal H4N 1L4, Canada.
Present address: Department of Biology, 112 Science Place, University of Saskatchewan, Saskatoon S7N 5E2, Canada.
Abbreviations: AGPase, ADP-glucose pyrophosphorylase; DTT, 1,4-dithiothreitol; ORF, open reading frame; PVP, polyvinylpyrrolidone; SQRT-PCR, semi quantitative reverse transcriptase polymerase chain reaction.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Albersheim P. 1975. The walls of growing plant cells. Scientific American 232, 80–95.[Web of Science][Medline]
Appeldoorn NJG, Debruijn SM, Kootgronsveld EAM, Visser RGF, Vreugdenhil D, Vanderplas LHW. 1997. Developmental changes of enzymes involved in conversion of sucrose to hexose-phosphate during early tuberization of potato. Planta 202, 220–226.[CrossRef]
Boyer JS, McPherson HG. 1975. Physiology of water deficits in cereal crops. Advances in Agronomy 27, 1–23.
Clément C, Chavant L, Burrus M, Audran JC. 1994. Anther starch variations in Lilium during pollen development. Sexual Plant Reproduction 7, 347–356.
Dorion S, Lalonde S, Saini HS. 1996. Induction of male sterility in wheat by meiotic-stage water deficit is preceded by a decline in invertase activity and changes in carbohydrate metabolism in anthers. Plant Physiology 111, 137–145.[Abstract]
Franchi GG, Bellani L, Nepi M, Pacini E. 1996. Types of carbohydrate reserves in pollen: localization, systematic distribution and ecophysiological significance. Flora 191, 143–159.
Godt DE, Roitsch T. 1997. Regulation and tissue-specific distribution of mRNAs for the three extracellular invertase isoenzymes of tomato suggests an important function in establishing and maintaining sink metabolism. Plant Physiology 115, 273–282.[Abstract]
Goetz M, Godt EE, Guivarc'h A, Kahmann U, Chriqui D, Roitsch T. 2001. Induction of male sterility in plants by metabolic engineering of the carbohydrate supply. Proceedings of the National Academy of Sciences, USA 98, 6522–6527.
Goldberg RB, Beals TP, Sanders PM. 1993. Anther development: basic principles and practicle applications. The Plant Cell 5, 1217–1229.
Hanson AD, Hitz WD. 1982. Metabolic responses of mesophytes to plant water deficits. Annual Review of Plant Physiology 33, 163–203.[Web of Science]
Heidecker G, Messing J. 1986. Structural analysis of plant genes. Annual Review of Plant Physiology 37, 439–466.[CrossRef][Web of Science]
Jackson D. 1992. In situ hybridization in plants. In: Gurr SJ, McPherson MJ, Bowles DJ, eds. Molecular plant pathology: a practical approach. Oxford, UK: Oxford University Press, 163–174.
Koch KE. 1996. Carbohydrate-modulated gene expression in plants. Annual Review of Plant Physiology and Plant Molecular Biology 47, 509–540.[CrossRef][Web of Science]
Lalonde S, Beebe D, Saini HS. 1997a. Early signs of disruption of wheat anther development associated with the induction of male sterility by meiotic-stage water deficit. Sexual Plant Reproduction 10, 40–48.[CrossRef]
Lalonde S, Morse D, Saini HS. 1997b. Expression of a wheat ADP-glucose pyrophosphorylase gene during development of normal and water-stress-affected anthers. Plant Molecular Biology 34, 445–453.[CrossRef][Web of Science][Medline]
Minhas JS, Saini HS. 1998. Cloning and characterization of a cDNA (Accession No. AF030420) from wheat anthers, that encodes the cell-wall form of invertase (PGR98-206). Plant Physiology 118, 1535.
Morgan JM. 1980. Possible role of abscisic acid in reducing seed set in water-stressed wheat plants. Nature 289, 655–657.
Morgan JM, King RW. 1984. Association between loss of leaf turgor, abscisic acid levels and seed set in two wheat cultivars. Australian Journal of Plant Physiology 11, 143–150.
Murray M, Thompson WF. 1980. Rapid isolation of high molecular weight plant DNA. Nucleic Acids Research 8, 4321–4325.
Nelson N. 1944. A photometric adaptation of the Somogyi method for the determination of glucose. Journal of Biological Chemistry 153, 375–380.
O'Toole JC, Moya TB. 1981. Water deficit and yield in upland rice. Field Crops Research 4, 247–259.[CrossRef]
Pacini E, Franchi GG. 1988. Amylogenesis and amylosis during pollen grain development. In: Cresti M, Gori P, Pacini E, eds. Sexual reproduction in higher plants. Berlin: Springer-Verlag, 181–186.
Roitsch T. 1999. Source-sink regulation by sugar and stress. Current Opinion in Plant Biology 2, 198–206.[CrossRef][Web of Science][Medline]
Saini HS. 1997. Effects of water stress on male gametophyte development in plants. Sexual Plant Reproduction 10, 67–73.[CrossRef]
Saini HS, Aspinall D. 1981. Effect of water deficit on sporogenesis in wheat (Triticum aestivum L.). Annals of Botany 48, 623–633.
Saini HS, Sedgley M, Aspinall D. 1984. Developmental anatomy in wheat of male sterility induced by heat stress, water deficit or abscisic acid. Australian Journal of Plant Physiology 11, 243–253.
Saini HS, Westgate ME. 2000. Reproductive development in grain crops during drought. Advances in Agronomy 68, 59–96.[Web of Science]
Salter PJ, Goode JE. 1967. Crop responses to water at different stages of growth. Research Review No. 2. Farnham Royal, England: Commonwealth Agricultural Bureaux.
Sheoran IS, Saini HS. 1996. Drought-induced male sterility in rice: changes in carbohydrate levels and enzyme activities associated with the inhibition of starch accumulation in pollen. Sexual Plant Reproduction 9, 161–169.[CrossRef][Web of Science]
Smeekens S, Rook F. 1997. Sugar sensing and sugar-mediated signal transduction in plants. Plant Physiology 115, 7–13.[Web of Science][Medline]
St-Pierre B, Vazquez-Flota FA, Luca VD. 1999. Multicellular compartmentation of Catharanthus roseus alkaloid biosynthesis predicts intercellular translocation of a pathway intermediate. The Plant Cell 11, 887–900.
Sturm A. 1999. Invertases. Primary structure, functions, and roles in plant development and sucrose partitioning. Plant Physiology 121, 1–7.
Sturm A, Chrispeels MJ. 1990. cDNA cloning of carrot extracellular ß-fructosidase and its expression in response to wounding and bacterial infection. The Plant Cell 2, 1107–1119.
Trouverie J, Chateau-Joubert S, Thevenot C, Jacquemot MP, Prioul JL. 2004. Regulation of vacular invertase by abscisic acid or glucose in leaves and roots from maize plantlets. Planta 219, 894–905.[CrossRef][Web of Science][Medline]
Tsai CY, Salamini F, Nelson OE. 1970. Enzymes of carbohydrate metabolism in the developing endosperm of maize. Plant Physiology 46, 299–306.
Tymowska-Lalanne Z, Kreis M. 1998. The plant invertases: physiology, biochemistry and molecular biology. Advances in Botanical Research 28, 71–117.
Westgate ME, Passioura JB, Munns R. 1996. Water status and ABA content of floral organs in drought stressed wheat. Australian Journal of Plant Physiology 23, 763–772.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. K. Proels and T. Roitsch Extracellular invertase LIN6 of tomato: a pivotal enzyme for integration of metabolic, hormonal, and stress signals is regulated by a diurnal rhythm J. Exp. Bot., April 1, 2009; 60(6): 1555 - 1567. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. N. Oliver, E. S. Dennis, and R. Dolferus ABA Regulates Apoplastic Sugar Transport and is a Potential Signal for Cold-Induced Pollen Sterility in Rice Plant Cell Physiol., September 1, 2007; 48(9): 1319 - 1330. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Horst, T. Welham, S. Kelly, T. Kaneko, S. Sato, S. Tabata, M. Parniske, and T. L. Wang TILLING Mutants of Lotus japonicus Reveal That Nitrogen Assimilation and Fixation Can Occur in the Absence of Nodule-Enhanced Sucrose Synthase Plant Physiology, June 1, 2007; 144(2): 806 - 820. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Chen, X. Zhao, Z. Shao, Z. Wei, Y. Wang, L. Zhu, J. Zhao, M. Sun, R. He, and G. He Rice UDP-Glucose Pyrophosphorylase1 Is Essential for Pollen Callose Deposition and Its Cosuppression Results in a New Type of Thermosensitive Genic Male Sterility PLANT CELL, March 1, 2007; 19(3): 847 - 861. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Boyer and J. E. McLaughlin Functional reversion to identify controlling genes in multigenic responses: analysis of floral abortion J. Exp. Bot., January 1, 2007; 58(2): 267 - 277. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||