Carbohydrate starvation stimulates differential expression of rice alpha-amylase genes that is modulated through complicated transcriptional and posttranscriptional processes.

Expression of alpha-amylase genes in cultured rice suspension cells is induced by sucrose starvation. To study the mechanism of sugar metabolite regulation on the expression of individual alpha-amylase genes, DNA fragments specific to each of eight rice alpha-amylase genes were synthesized and used as gene-specific probes. Comparison of the relative abundance of mRNA revealed that expression of the eight alpha-amylase genes in rice cells was differentially regulated by sucrose starvation. Accumulation of all the alpha-amylase mRNAs increased in response to sucrose starvation; however, levels of the alphaAmy3 and alphaAmy8 mRNAs were distinctly higher and constituted 90% of total alpha-amylase mRNAs. RNA gel blot and nuclear run-on transcription analyses demonstrated a positive correlation between the increased transcription rates and the elevated steady-state levels of alpha-amylase mRNAs induced by sucrose starvation. The half-lives of alphaAmy3, alphaAmy7, and alphaAmy8 were prolonged by sucrose-starvation; however, the stability of the three mRNAs seems controlled by different mechanisms. The translation inhibitors cycloheximide and anisomycin preferentially blocked the sucrose-suppressed expression of alphaAmy3 but not that of alphaAmy7 and alphaAmy8. These inhibitors also enhanced the sucrose starvation-induced accumulation of alphaAmy3 mRNA but not that of alphaAmy7 or alphaAmy8 mRNAs. Cycloheximide did not significantly alter the transcription rates of alpha-amylase genes, suggesting that labile proteins may selectively stabilize the alphaAmy7 and alphaAmy8 mRNAs but destabilize the alphaAmy3 mRNA.

Carbon catabolite repression is a fundamental and ubiquitous regulatory system in both prokaryotic and eukaryotic cells (1)(2)(3)(4). In bacteria and yeast, catabolite-regulated gene expression is an essential mechanism for adjusting to changes in nutrient availability (5)(6)(7). Studies using Saccharomyces cerevisiae mutants have revealed many of the components involved in the response to carbon catabolite repression, but it is still unclear how all of these components interact to regulate transcription (2,8). In higher plants, carbon metabolite regulation of gene expression provides a mechanism for maintaining an economical balance between supply (source) and demand (sink) for carbohydrate allocation and utilization within and among various organs and tissues (9). Expression of en-zymes involved in carbohydrate metabolism often is feedbackregulated by the sugar metabolites (9). For example, expression of seven maize photosynthetic genes in mesophyll protoplasts is repressed by sucrose and the mechanism involves transcriptional control (10). Phosphorylation of hexose sugars by hexokinase has been proposed to act as a key signal transmitter in initiating sugar repression responses of photosynthetic genes (11), and of malate synthase and isocitrate lyase genes involved in the glyoxylate cycle (12).
␣-Amylases are major amylolytic enzymes for hydrolysis of stored starch in the endosperm during germination of cereal grains. In germinating cereal grains, gibberellic acid stimulates and abscisic acid represses ␣-amylase gene expression (reviewed in Ref. 13). We have previously reported that expression of ␣-amylase genes in rice is under two different modes of tissue-specific regulation; the genes are activated by hormones in the aleurone of germinating seed and suppressed by sugars in cultured suspension cells (14,15). Later, expression of one ␣-amylase gene in the embryo of germinating rice seed was also reported to be suppressed by sugars (16). Recently, we observed that sugars that accumulate in the embryo and endosperm during germination act as signals and osmotica to regulate the expression of ␣-amylase genes and the metabolic activities in germinating rice seeds (17). To study the mechanism of metabolite regulation of ␣-amylase gene expression in plants, we have used the cultured rice suspension cells as a model system. Previous work shows that, in cultured rice suspension cells, ␣-amylase expression, carbohydrate metabolism, and vacuolar autophagy are coordinately regulated by sucrose levels in the medium (18). Both the transcription rate and mRNA stability of ␣-amylase genes in cells increase in response to sucrose depletion in the culture medium (19). Use of transgenic rice carrying an ␣-amylase gene promoter/␤-glucuronidase gene proved that the regulation of ␣-amylase gene expression by sugars involves a transcriptional control mechanism (20 -22).
Our previous study on metabolite regulation involved measurement at the total ␣-amylase mRNA level rather than dealing with the individual mRNA encoding different ␣-amylase isozymes. To further study the mechanism(s) of metabolite regulation in detail, it is necessary to identify the individual ␣-amylase genes whose expression is regulated by sugars. This report demonstrates that the expression of various ␣-amylase genes in rice suspension cells is regulated in a coordinated way by sucrose present in the medium, but in a distinctly different manner according to the specific ␣-amylase genes. Both transcription rate and mRNA stability contribute to the regulation of the transcript level of the individual ␣-amylase gene.

EXPERIMENTAL PROCEDURES
Plant Material-Suspension cell cultures of rice (Oryza sativa cv. Tainan 5) were propagated as described previously (14). Cells for puri-fication of RNA were collected by filtration through a 400-mesh nylon sieve, blot-dried on paper towels, quick-frozen in liquid N 2 , and stored at Ϫ70°C until use.
Genomic DNA Isolation and DNA Gel Blot Analysis-One gram of rice suspension cells was ground in liquid N 2 with a mortar and pestle and placed in a 2-ml Eppendorf tube. The ground tissue was suspended in 1 ml of urea extraction buffer (containing 7 M urea, 0.3 M NaCl, 50 mM Tris-HCl, pH 8.0, 20 mM EDTA, and 1% sarkosine) and extracted with 1 ml of phenol:chloroform:isoamyl alcohol (25:24:1) at room temperature for 15 min. The mixture was centrifuged at 8000 rpm and 4°C for 10 min. The aqueous phase was mixed with 200 l of 3 M sodium acetate (pH 5.2) and 1 ml of isopropanol. DNA was spooled out, placed in 70% ethanol, and centrifuged at 8000 rpm for 10 s. The DNA pellet was washed with 70% and 100% ethanol and air-dried. The genomic DNA was resuspended in TE buffer and stored at 4°C.
Ten micrograms of genomic DNA was digested with various restriction enzymes and fractionated on a 0.8% agarose gel. The DNA gel blot analysis was performed as described by Sambrook et al. (25) using the ␣-amylase gene-specific DNAs as probes. The ␣-amylase gene-specific DNA fragments used as probes were excised from the plasmid vectors by appropriate restriction enzymes and labeled with [␣-32 P]dATP using the random primer method (26).
Synthesis of cDNA and DNA Slot-blot Analysis-The 32 P-labeled single-stranded cDNA probe was prepared from total cellular RNA using an oligo(dT) primer and avian myeloblastosis virus reverse transcriptase (Promega) according to the manufacturer's instructions. Plasmid DNA was denatured with 0.4 N NaOH at room temperature for 30 min, and neutralized with a 9-fold volume of 6 ϫ SSC (containing 0.1% DNA agarose gel loading dye). The denatured DNA was blotted onto the Magna nylon membrane (MSI) using a suction slot-blotter (Life Technologies, Inc.) and hybridized with the cDNA probe. Hybridization was performed as described for DNA gel blot analysis (25).
Isolation of Total RNA and RNA Gel Blot Analysis-Total RNA was purified from rice suspension cells by the method of Verwoerd et al. (27). RNA gel blot analysis was performed as described by Chao et al. (28). The 1.4-kb ␣Amy8-C and Act1 cDNA inserts used as probes were excised from the plasmid vectors by restriction enzyme EcoRI and labeled with [␣-32 P]dCTP using the random primer method (26). Because the nucleotide sequence identity among the coding regions of rice ␣-amylase genes ranges between 75 and 95% 2 and the coding regions of rice actin genes are highly conserved (23), the ␣Amy8-C and Act1 cDNA containing coding region probably hybridize to the mRNAs of all ␣-amylase and actin genes. The ␣-amylase gene-specific DNA probes were prepared by PCR and labeled with [␣-32 P]dCTP using the random primer method. In each experiment the hybridized ␣-amylase DNA probe on the Magna nylon membrane was stripped off by incubating twice (30 min each) in washing solution containing 0.1 ϫ SSC and 0.1% SDS at 90°C. The same membrane was rehybridized with different ␣-amylase gene-specific DNA probes or the actin cDNA probe.
Isolation of Nuclei and Nuclear Run-on Transcription Analysis-Nuclei were isolated from rice suspension cells as described by Watson and Thompson (29). Nuclear run-on transcription analysis was performed as described previously (19).

Synthesis of ␣-Amylase
Gene-specific DNA Fragments-The rice ␣-amylase isozymes are encoded by nine genes (30). Nomenclature and accession numbers of the rice ␣-amylase genes in GenBank™ are shown in Table I. Comparison of nucleotide sequences of the 3Ј-untranslated regions (3Ј-UTR) of ␣-amylase genes shows low identity except between RAmy3B and RAmy3C ( Fig. 1). There is also some similarity between the 3Ј-UTRs of ␣Amy7(Ramy1A) and ␣Amy10(RAmy1C); however, several gaps are distributed between their homologous regions. Primers locating at the 5Ј and 3Ј regions in the 3Ј-UTR of each of the eight ␣-amylase genes were designed ( Fig. 1) so that amplification of DNA fragments flanked by the paired primers would yield gene-specific DNA fragments. Replicated gel blots of genomic DNA digested with various restriction enzymes were hybridized with the 32 P-labeled ␣-amylase gene-specific probes. Each probe hybridized specifically to the ␣-amylase gene from which it was derived ( Fig. 2A), demonstrating that the eight gene-specific probes were able to detect the individual ␣-amylase genes. To demonstrate over what range these probes can discriminate the mRNA of one gene from another, eight pieces of parallel prepared membranes containing three dilution series of each gene-specific DNA were hybridized with the individual probes. The result shows that only RAmy1B and RAmy1C probes slightly cross-hybridized with some of the other ␣-amylase gene-specific DNAs (Fig. 2B). No cross-hybridization was detected among the rest of the probes.
Differential Expression of ␣-Amylase Genes upon Sucrose Starvation-The steady-state levels of ␣-amylase mRNA in rice suspension cells before and after sucrose starvation were examined by RNA gel blot analysis. Amount of rRNA was used as RNA amount loading control in the RNA gel blot analysis, although previously we have shown that the total transcription rate in cells provided with sucrose was twice that in cells starved for sucrose (19). Accumulation of ␣-amylase mRNAs was very low or undetectable in cells provided with sucrose (Fig. 3, lane 2). In contrast, in sucrose-starved cells, accumulation of the ␣-amylase mRNAs increased (Fig. 3, lane 1). The magnitude of increase in mRNA concentrations varied with ␣-amylase genes and was particularly dramatic for ␣Amy3 and ␣Amy8.
To compare the relative abundance of mRNA of the eight ␣-amylase genes, an excess of rice rRNA gene, actin cDNA, and ␣-amylase gene-specific DNAs was spotted onto a membrane and hybridized with the 32 P-labeled, single-stranded cDNA probe transcribed from the total mRNA of sucrose-starved cells. The relative mRNA levels, corresponding to different genes in a given population of RNA, were then compared. The cDNA of the rRNA was also synthesized, probably due to the presence of the repeated short stretch of the poly(A) sequence in the rRNA (31,32). In sucrose-starved cells, the amounts of 1 The abbreviations used are: kb, kilobase pair(s); PCR, polymerase chain reaction; UTR, untranslated region.
␣Amy8 M59352 a -indicates that the gene has no synonym.
␣Amy3 and ␣Amy8 mRNAs were distinctly higher than those of other ␣-amylase genes (Fig. 4). The ␣Amy3 mRNA was the most abundant (Fig. 4, slot 9), with ␣Amy8 mRNA the next most abundant (Fig. 4, slot 10). Together the results shown in Figs. 3 and 4 suggest that the ␣-amylase genes are subject to differential metabolic regulation at the transcriptional and/or post-transcriptional level. Differential Activation of ␣-Amylase Gene Transcription upon Sucrose Starvation-To define the mechanism that differentially regulates the expression of ␣-amylase genes, we compared the transcription rates of individual ␣-amylase genes. The nuclear run-on transcription analyses were performed with nuclei isolated from cells grown in the absence or presence of sucrose. In cells provided with sucrose, the transcription rates of all the ␣-amylase genes were undetectable (Fig. 5, upper panel). In contrast, in cells starved for sucrose for 12 h, the transcription rates of all the ␣-amylase genes increased, with ␣Amy3 the highest (Fig. 5, upper panel, slot 7) and ␣Amy8 the next highest (Fig. 5, upper panel, slot 8). In cells starved for 24 h (Fig. 5, lower panel), the transcription rates of all the tested genes decreased significantly; however, the tran-scription rates of ␣Amy3 and ␣Amy8 were still distinctly higher than that of the other ␣-amylase genes. Comparison of Figs. 4 and 5 reveals that there is a positive correlation between the transcription rates and the steady-state levels of ␣-amylase mRNAs in cells starved for sucrose.
Increase in Stability of ␣-Amylase mRNA upon Sucrose Starvation-To examine the effect of sucrose starvation on the half-life of individual ␣-amylase mRNA, degradation of the ␣Amy3, ␣Amy7, and ␣Amy8 mRNAs in vivo was monitored by RNA gel blot analysis following the inhibition of transcription with actinomycin D. Addition of 10 g ml Ϫ1 actinomycin D to the medium has been shown previously to inhibit total RNA transcription by more than 95% over a 12-h time course (19). The levels of individual ␣-amylase mRNAs were low in cells grown in the sucrose-containing medium (Fig. 6A, lane 1) and increased significantly after cells had been shifted to sucrosefree medium for 24, 36, and 45 h (Fig. 6A, lanes 2-4). The levels of individual ␣-amylase mRNAs that accumulated after cells have been starved for 24 h and then pretreated with actinomycin D in the absence of sucrose for 12 h (Fig. 6A, lane 5  the medium still lacking sucrose but containing actinomycin D (Fig. 6A, lanes 14 -19). In contrast, the levels of ␣-amylase mRNA decreased with time during the 9-h incubation in the medium containing both sucrose and actinomycin D (Fig. 6A,  lanes 6 -12). The level of actin mRNA was high in cells grown in the presence of sucrose (Fig. 6A, lane 1), but was low in the absence of sucrose (Fig. 6A, lanes 2-4). In the medium containing actinomycin D, the levels of actin mRNA remained low regardless of the presence (Fig. 6A, lanes 5-12) or absence (Fig.  6A, lanes 13-19) of sucrose. The result of actin gene expression suggests that transcription of total mRNA was almost completely inhibited by actinomycin D; otherwise, accumulation of actin mRNA would increase in medium containing both sucrose and actinomycin D.
The amounts of mRNA shown in Fig. 6A were quantified, and Fig. 6B shows the log plots of the ␣-amylase mRNA levels as a function of time. The half-life of mRNA was calculated from the slope of the line in the log plot of the data. The relative half-lives of ␣Amy3 and ␣Amy8, which were 82 and 98 min, respectively, in cells provided with sucrose, increased to 6 h in cells starved for sucrose (Fig. 6B). Both the half-lives of ␣Amy3 and ␣Amy8 mRNAs were substantially shorter than that of ␣Amy7 mRNA, regardless of the presence or absence of sucrose in the medium. The relative half-life of ␣Amy7 was almost 5 h in cells provided with sucrose and increased to 23 h in cells starved for sucrose. These results suggest that the ␣-amylase mRNAs are more stable in cells starved for sucrose. Rice suspension cells were cultured in sucrose-containing (ϩS) or sucrose-free (ϪS) medium for 2 days, and total RNA was isolated. RNA gel blot analysis was performed using the ␣-amylase gene-specific DNAs, or the ␣Amy8-C (␣Amy) and actin cDNA as probes. The equivalence of RNA loading among lanes was demonstrated by ethidium bromide staining of rRNA.
FIG. 4. Slot-blot analysis demonstrating relative mRNA abundance of eight ␣-amylase genes in cultured rice suspension cells. Rice cells were transferred from sucrose-containing medium to sucrose-free medium for 2 days. Cells were collected, and total RNA was isolated. Synthesis of 32 P-labeled cDNA probes and slot-blot analysis was performed. Plasmid DNAs containing the genes indicated were applied in 5-g aliquots and hybridized with the cDNA probes. pBS, pBluescript vector containing no ␣-amylase cDNA.
Diverse Effect of Translation Inhibitors on the Accumulation of ␣-Amylase mRNA-To examine whether the increase in mRNA levels of various ␣-amylase genes requires a prior synthesis of other gene products, cells were treated with translation inhibitors and accumulation of mRNA was monitored with RNA gel blot analysis. Levels of ␣-amylase mRNAs were very low or almost undetectable in cells provided with sucrose (Fig.  7A, lanes 1 and 2), but increased greatly in cells starved for sucrose for 2 days (Fig. 7A, lanes 8 and 9). Cycloheximide, an inhibitor of translocase (33), with concentrations ranging from 20 to 300 M inhibited total protein synthesis by 90 -98%. Cycloheximide inhibited the accumulation of ␣Amy7 and ␣Amy8 mRNAs regardless of the presence (Fig. 7A, lanes 3-7) or absence (Fig. 7A, lanes 10 -14) of sucrose. In contrast, cycloheximide did not inhibit but instead effectively enhanced the accumulation of ␣Amy3 mRNA. In the absence of sucrose, cycloheximide significantly increased the level of ␣Amy3 mRNA independent of its dosage (Fig. 7A, lanes 10 -14). In the presence of sucrose, cycloheximide increased the level of ␣Amy3 mRNA in a dose-dependent manner (Fig. 7A, lanes  3-7). Accumulation of actin mRNA was high in the presence of sucrose (Fig. 7A, lanes 1 and 2) and became low in the absence of sucrose (Fig. 7A, lanes 8 and 9). Cycloheximide also inhibited the accumulation of actin mRNA either in the presence (Fig.  7A, lanes 3-7) or absence (Fig. 7A, lanes 10 -14) of sucrose but to a lesser extent as compared with the ␣Amy7 and ␣Amy8 mRNAs.
Anisomycin, an inhibitor of transpeptidase (33), with concentration of 300 M only 50% effective in inhibiting total protein synthesis, but the trends are exactly in the same direction as what was found when cycloheximide was used. Anisomycin promoted ␣Amy3 mRNA accumulation but suppressed ␣Amy7 and ␣Amy8 mRNA accumulation under both plus and minus sucrose conditions (Fig. 7B). Anisomycin also suppressed the actin mRNA accumulation under both plus and minus sucrose conditions. The suppression of ␣Amy8 mRNA accumulation in sucrose-starved cells by anisomycin (Fig. 7B, lane 4) was not as complete as that by cycloheximide (Fig. 7A, lanes 10 -14), probably due to the less efficient inhibition of protein synthesis by anisomycin under our experimental condition.

Increase in Transcription Rates Is Essential for Increase in
Level of ␣-Amylase mRNA upon Sucrose Starvation-The availability of gene-specific probes corresponding to each of the eight ␣-amylase genes has enabled us to examine the abundance of mRNA encoding specific ␣-amylase isozymes. Although expression of the eight ␣-amylase genes increased in response to sucrose starvation (Fig. 3), levels of ␣Amy3 and ␣Amy8 mRNAs were significantly higher than those of other ␣-amylase genes and constituted approximately 90% of total ␣-amylase mRNA in cells starved for sucrose (Fig. 4). Therefore, expression of the ␣-amylase genes is coordinately up-regulated but in a distinct manner by sucrose starvation. A positive correlation between FIG. 5. Activation of ␣-amylase gene transcription by sucrose starvation. Rice suspension cells were cultured in sucrose-containing or sucrose-free medium for 12 or 24 h, and nuclei were isolated. In vitro run-on transcription reactions were carried out, and the 32 P-labeled RNAs were hybridized to 5 g each of rice rRNA gene, actin cDNA, and ␣-amylase gene-specific DNAs immobilized on a membrane. pBS, pBluescript vector containing no ␣-amylase cDNA. Upper panel, nuclei isolated from cells cultured with (ϩS) or without (ϪS) sucrose for 12 h. Lower panel, nuclei isolated from cells cultured without sucrose for 24 h.
FIG. 6. Increase in ␣-amylase mRNA half-life by sucrose starvation. A, RNA gel blot analysis using ␣Amy3, ␣Amy7, and Amy8 gene-specific DNAs and actin cDNA as probes. Rice cells were grown in sucrose-containing (ϩS) medium for 4 days (RNA in lane 1) and transferred to sucrose-free (ϪS) medium for 24 h (RNA in lane 2). Actinomycin D was added to the medium to a final concentration of 10 g/ml. Cells were incubated in the sucrose-free medium containing actinomycin D for another 12 h and then divided into two halves. Half the cells were transferred to a medium containing both sucrose and actinomycin D (RNAs in lanes [5][6][7][8][9][10][11][12]. The other half were transferred to a medium lacking sucrose but containing actinomycin D (RNAs in lanes [13][14][15][16][17][18][19]. Cells were collected after 0.5-9 h, and RNAs were purified. Five micrograms of total RNA were loaded in each lane. the transcription rates and the steady-state mRNA levels suggests that the transcriptional regulation plays an important role in the differential expression of ␣-amylase genes. Alteration in transcriptional activity suggests a model in which sugar-regulated transcription factors interact with cisacting elements in the ␣-amylase gene promoters. Regions upstream from the TATA box of ␣Amy3 (RAmy3D) (22) and ␣Amy8 (21) share conserved sequences (34) and possess the elements necessary for sugar repression. However, none of these sequence elements has been demonstrated to play a role in the regulation of rice ␣-amylase gene expression. Metabolic regulation of ␣-amylase gene expression by glucose has also been reported in other eukaryotes such as Drosophila melanogaster and Aspergillus oryzae. No significant sequence similarity is found between the Aspergillus (35) and the Drosophila (36) ␣-amylase gene promoters. However, some conserved sequences are found between the promoter region of either the Drosophila or the Aspergillus ␣-amylase gene with that of ␣Amy3.
Increase in mRNA Stability Enhances the Level of ␣-Amylase mRNA upon Sucrose Starvation-Under conditions in which transcription was inhibited, the half-lives of ␣-amylase mRNAs were longer in cells starved for sucrose than in cells provided with sucrose (Fig. 6), suggesting that increased mRNA stability contributed to an increase in ␣-amylase mRNA level in sucrosestarved cells. The ␣Amy7 mRNA seems inherently more stable because its half-life was 3-4 times longer than those of the ␣Amy3 and ␣Amy8 mRNAs, regardless of whether or not the cells were sucrose-starved. Compared with the ␣Amy7 mRNA, the ␣Amy3 and ␣Amy8 mRNAs were transcribed more rapidly in cells in response to sucrose starvation (Fig. 5) and were degraded more rapidly (Fig. 6) if cells were provided with sucrose. These results suggest that expression of ␣Amy3 and ␣Amy8 is under a stricter regulation and may encode isozymes essential for responses to changes of carbohydrate availability.
It is unclear what mechanism directs the ␣-amylase mRNA to be degraded faster in cells provided with sucrose than in cells starved for sucrose. Regulated mRNA turnover has been described in detail in several vertebrate system (37), but the common regulatory mechanisms for mRNA turnover have not been found. Little is known about the processes responsible for the alteration of mRNA stability in plants, although mRNA stability can be affected by environmental conditions such as temperature (38,39) and illumination levels (40,41). The sucrose-induced acceleration of ␣-amylase mRNA degradation indicates that the trans-acting factor(s) of the mRNA decay system is likely subject to regulation. This may be ␣-amylase transcript-specific, since transcripts of the three ␣-amylase genes are affected similarly but the actin transcripts are not affected (Fig. 6). The differential stabilities among ␣-amylase transcripts, on the other hand, may be promoted by transcriptspecific sequence elements and/or secondary structures. Many unstable transcripts in mammalian cells contain sequences within their 3Ј-UTRs that are AU-rich and contain one or more copies of the motif AUUUA (42). Unlike unstable mammalian transcripts, the 3Ј-UTRs of ␣-amylase mRNAs are not particularly AU-rich and do not contain multiple AUUUA sequences (Fig. 1). However, short AU-rich sequences are found throughout the 3Ј-UTRs of the ␣-amylase genes.
Stabilities of Different ␣-Amylase mRNAs Are Selectively Regulated by a Labile Protein-Interestingly, although the ␣-amylase mRNA seems to be generally stabilized by starvation of cells (Fig. 6), the mechanism controlling the stability of different ␣-amylase mRNAs varied. Both the translation inhibitors cycloheximide and anisomycin enhanced the accumulation of ␣Amy3 mRNA regardless of whether or not the cells were provided with sucrose (Fig. 7). Such an effect of translation inhibitors was specific to ␣Amy3 mRNA, as the accumulation of ␣Amy7 and ␣Amy8 mRNA was suppressed by the inhibitors, even in cells starved for sucrose. Nuclear run-on transcription analyses demonstrated that cycloheximide did not significantly alter the transcription rates of ␣Amy3, ␣Amy7, and ␣Amy8 regardless of whether or not the cells were provided with sucrose (data not shown). Together, the results suggest that labile proteins are involved in the destabilization of ␣Amy3 mRNA and the stabilization of ␣Amy7 and ␣Amy8 mRNAs.
In conclusion, the differential response of ␣-amylase genes to sugars provides a potential mechanism for altering the pattern of enzyme accumulation in response to changing carbohydrate status and sugar import. Although transcription of the ␣-amylase genes is induced by sucrose starvation, the dramatic increase in their steady-state mRNA levels requires stabilization of the mRNA as well. The combined processes may lead to more rapid and more marked shifts in the expression of ␣-am-FIG. 7. Effect of translation inhibitors on the expression of ␣-amylase genes. Rice suspension cells were exposed to various concentrations of cycloheximide (A) or 300 M anisomycin in sucrose-free or sucrose-containing medium (B). At 12 h post-inhibitor treatment, cells were collected and total RNA was purified. RNA was subject to RNA gel blot analysis using the ␣Amy3, ␣Amy7, and Amy8 gene-specific DNAs or the ␣Amy8-C (␣Amy) and actin cDNA as probes. Five micrograms of total RNA were loaded in each lane. D indicates dimethyl sulfoxide (Me 2 SO), which was used to dissolve cycloheximide. ϩ and Ϫ indicate presence or absence of sucrose or anisomycin. CHX, cycloheximide; AN, anisomycin. ylase genes. Both processes may share similar upstream signaling pathways and contribute differently and in gene-specific manners to the regulation of the pool size of transcripts of individual ␣-amylase genes.