ACC synthase expression regulates leaf performance and drought tolerance in maize

Authors


(fax +951 827 4434; e-mail drgallie@citrus.ucr.edu).

Summary

Ethylene regulates entry into several types of plant developmental cell death and senescence programs besides mediating plant responses to biotic and abiotic stress. The response of cereals to conditions of drought includes loss of leaf function and premature onset of senescence in older leaves. In this study, ACC synthase (ACS) mutants, affecting the first step in ethylene biosynthesis, were isolated in maize and their effect on leaf function examined. Loss of ZmACS6 expression resulted in delayed leaf senescence under normal growth conditions and inhibited drought-induced senescence. Zmacs6 leaves continued to be photosynthetically active under both conditions indicating that leaf function was maintained. The delayed senescence phenotype associated with loss of ZmACS6 expression was complemented by exogenous ACC. Surprisingly, elevated levels of foliar chlorophyll, Rubisco, and soluble protein as well as improved leaf performance was observed for all Zmasc6 leaves, including young and fully expanded leaves which were far from initiating senescence. These observations suggest that ethylene may serve to regulate leaf performance throughout its lifespan as well as to determine the onset of natural senescence and mediate drought-induced senescence.

Introduction

Photosynthetic capacity increases during leaf expansion and declines with leaf age until it reaches a low level prior to the onset of leaf senescence (Gay and Thomas, 1995). The rate of initiation and execution of a senescence program significantly impacts the ultimate contribution that a leaf makes to a plant. A delay in the onset of senescence in Lolium temulentum by just 2 days has been calculated to increase the amount of carbon fixed by the plant by 11% (Thomas and Howarth, 2000). This is of particular relevance to those crops where yield potential is reduced by adverse environmental conditions, such as drought, that can induce premature leaf senescence in older leaves (Bradford and Hsiao, 1982; Kramer, 1983). Several examples of delayed senescence have been described which can be grouped into two types of staygreen (Thomas and Howarth, 2000; Thomas and Smart, 1993). The first of these is functional in that the photosynthetically active lifespan of a leaf is extended whereas the second is cosmetic in that foliar chlorophyll is retained during the execution of an otherwise normal senescence program.

Several environmental and physiological conditions significantly alter the initiation and execution of leaf senescence. Among these, light plays a significant role in that light deprivation can induce senescence (i.e. dark-induced senescence) (Weaver and Amasino, 2001). In addition to lack of light, insufficient nutrients and water can induce premature leaf senescence (Gan and Amasino, 1997; Rosenow et al., 1983). Several hormones also control entry of leaves into the senescence program. Cytokinin levels decline in organs during senescence and exogenous application of cytokinin can retard senescence (Richmond and Lang, 1957; Van Staden et al., 1988). Expression of the cytokinin-synthesizing gene, isopentenyl transferase, from a senescence-inducible promoter resulted in the autoregulated production of cytokinin and inhibited leaf senescence (Gan and Amasino, 1995). Ethylene has also been implicated in regulating leaf senescence in that defects in ethylene synthesis or perception delay leaf senescence in tomato and Arabidopsis (Abeles et al., 1992; Davies and Grierson, 1989; Grbic and Bleecker, 1995; John et al., 1995; Picton et al., 1993). The effect of a loss of ethylene synthesis or signaling on leaf senescence can be phenocopied by the exogenous application of inhibitors of ethylene biosynthesis or action (Beltrano et al., 1994; Labrana and Araus, 1991). Drought, which can induce senescence in older leaves, promotes increased ethylene production in plants by increasing ACC synthesis and its conversion to ethylene (Apelbaum and Yang, 1981; McKeon et al., 1982; McMichael et al., 1972). Foliar chlorophyll content decreases as a function of the severity of a water stress (Baisak et al., 1994; Levitt, 1980; Thomas and Stoddart, 1980) and inhibition of ethylene synthesis reduces the drought-induced loss of chlorophyll and prevents drought-induced senescence (Beltrano et al., 1997, 1999). As such conclusions can be complicated by possible pleiotropic effects associated with pharmaceutical approaches, the isolation of ethylene mutants in dicot species has been particularly useful in demonstrating the role that ethylene plays in controlling the onset of senescence. However, no such mutants have been identified in monocots, such as cereals, where the control of senescence may contribute significantly to the value of these economically important species. Mutants exhibiting delayed leaf senescence have been described for cereals but the molecular basis for a functional staygreen phenotype has not been identified (Spano et al., 2003; Thomas and Howarth, 2000).

Although ethylene may be involved in the final stage of leaf function, it is synthesized throughout leaf development (Beltrano et al., 1994; Hunter et al., 1999). Other than its senescence-related function, little is known about its role in earlier aspects of leaf development. In this report, we describe the isolation of maize knockout mutants deficient in ACC synthase (ACS), the first enzyme in the ethylene biosynthetic pathway. Mutants were deficient in ethylene production and leaves exhibited delayed senescence. Loss of one member, ZmACS6, resulted in a reduction of up to 90% of foliar ethylene and a substantial delay in leaf senescence which could be reversed by the exogenous application of ACC. Zmacs6 leaves retained chlorophyll, leaf protein, and Rubisco and maintained photosynthetic function longer than wild-type leaves. Loss of a different member of the ACS family, Zmacs2, resulted in a reduction of up to 45% of foliar ethylene and an intermediate delay in leaf senescence, establishing a correlation between a reduction in foliar ethylene and the extent of delay of leaf senescence. Drought-induced senescence was delayed in Zmacs6 plants and, in addition to the maintenance of chlorophyll and leaf protein during water stress, Zmacs6 leaves maintained photosynthetic function. Surprisingly, all leaves with reduced ACS expression and ethylene production contained higher levels of chlorophyll and leaf protein as well as a higher rate of CO2 assimilation, including those young and fully expanded leaves that were far from senescing. Our findings suggest that ACS expression regulates the onset of senescence under normal growth conditions and inhibits drought-induced senescence in older leaves. They also indicate that ACS expression serves to control aspects of leaf development that are independent of senescence and include the regulation of leaf physiology and function.

Results

Identification of ACC synthase knockout mutants

Three genes encoding ACC synthase were isolated from the inbred B73 (Gallie and Young, 2004). Two members of the family (i.e. ZmACS2 and ZmACS7) are closely related (95% amino acid identity) whereas the third gene (i.e. ZmACS6) is considerably more divergent (54 and 53% amino acid identity with ZmACS2 and ZmACS7, respectively) (Gallie and Young, 2004). A reverse genetic approach was used to screen for Mu insertions in ACC synthase gene family members (Bensen et al., 1995). Nineteen candidate lines were identified, 13 of which were confirmed by terminal-inverted-repeat (TIR)-PCR to harbor a Mu insertion in one of the three ACC synthase genes. Of these, five lines stably inherited the transposon in the first backcross to B73 that were backcrossed an additional four times to reduce unwanted Mu insertions. Plants were then self-pollinated to generate homozygous null individuals that were identified by PCR. PCR amplification of wild-type lines or heterozygous null mutants with ACCF1 and ACC-1 primers (see Experimental procedures) resulted in three different sized fragments corresponding to the three ACC synthase genes whereas the products of PCR amplification of homozygous insertion mutants lack the corresponding fragment (Figure 1a). The Mu insertion site for each mutant line was determined by sequencing across the Mu/ACC synthase junction using the Mu-TIR primer (Figure 1b). Four of the five insertion lines contained a Mu in ZmACS2: one mutant (i.e. Zmacs2-1) contained an insertion in the third exon whereas the other three (i.e. Zmacs2-2, Zmacs2-3, Zmacs2-4) contained insertions in the fourth exon at unique positions (Figure 1b). The fifth insertion line (i.e. Zmacs6) contained a Mu in ZmACS6 in the second intron near the 3′ splice site. Mu insertions in ZmACS7 were identified in the first generation but were not inherited, suggesting that they were somatic mutants or that expression of ZmACS7 is required for germ line development.

Figure 1.

Isolation of ACC synthase Mu-insertion lines.
(a) PCR analysis was performed on wild-type (i.e. B73) and Zmacs2 and Zmacs6Mu-insertion lines using conserved ACC synthase primers. The PCR products representing each gene family member (which differ because of differences in intron length) are indicated to the left.
(b) The Mu-insertion site was determined by sequencing out from the Mu transposon into the ACC synthase gene.

Quantitative real-time RT-PCR revealed that all three genes are expressed during maize leaf development (Figure 2). Insertion of Mu into ZmACS2 resulted in the loss of most ZmACS2 expression in Zmacs2 leaves (Figure 2a). Similarly, insertion of Mu into ZmACS6 reduced expression from the mutant gene to 0.01% of the wild-type expression level in the oldest leaves and 4–5% in younger leaves (Figure 2b). Residual ZmACS6 expression may have resulted from the removal of Mu through splicing of the second intron in which the transposon resides. ZmACS2 expression increased 2–2.5-fold in Zmacs6 mutant leaves (Figure 2a). A twofold reduction in ZmACS6 expression was observed in the oldest leaves of Zmacs2 plants but was little changed in younger Zmacs2 leaves (Figure 2b). ZmACS7 expression was unchanged in Zmacs2 leaves but was reduced by twofold in the oldest leaves of Zmacs6 leaves (Figure 2c). In contrast, expression from the housekeeping gene, β-tubulin, did not decrease in either mutant and in some cases increased to a small extent (Figure 2d) that correlated with the increase in soluble protein in the mutants (see below).

Figure 2.

Loss of ZmACS2 and ZmACS6 expression in Mu-insertion lines.
Expression from (a) ZmACS2, (b) ZmACS6, (c) ZmACS7, and (d) β-tubulin in the third, sixth, and ninth oldest leaves of B73 wild-type, Zmacs2, and Zmacs6 mutant lines was quantitated by real time RT-PCR analysis at 40 DAP.

Loss of ZmACS6 or ZmACS2 expression reduces ethylene synthesis

To determine whether the loss of ZmACS6 or ZmACS2 expression in the Zmacs6 and Zmacs2 mutants described above resulted in a decrease in ethylene evolution, ethylene was measured from wild-type and mutant plants. Ethylene evolution from Zmacs2 leaves was approximately 55% of wild type, a level that was similar for all Zmacs2 alleles (Figure 3). Ethylene evolution from Zmacs6 leaves was only 10% of that from wild-type leaves (Figure 3). The level of ethylene production in Zmacs2/Zmacs6 leaves was reduced only slightly from the level in Zmacs6 leaves. Interestingly, the level of ZmACS6 transcript accumulation was lower than for ZmACS2 or ZmACS7 indicating that ZmACS6 is disproportionately responsible for the amount of ethylene produced in leaves and suggesting that ACS expression may be subject to post-transcriptional regulation. These data suggest that loss of ZmACS6 expression results in a greater reduction in the ability of maize leaves to produce ethylene than does the loss of ZmACS2 expression.

Figure 3.

Ethylene production is reduced in Zmacs2 and Zmacs6 leaves. Ethylene production in leaf 4 of B73 wild-type, Zmacs2, Zmacs6, and Zmacs2/Zmacs6 plants. Three replicates were measured and the average and standard deviation reported.

Loss of ZmACS6 expression delays entry into the senescence program

The level of ethylene evolution in maize leaves increased as a function of leaf age up to the point of senescence whereupon it decreased: at 20 days after pollination (DAP) ethylene production was highest in leaf 1 (the oldest surviving leaf) and at 30 and 40 DAP ethylene was highest in leaf 3 and leaves 4–5, respectively (data not shown). As the peak in ethylene evolution in a leaf occurred prior to the onset of any visible signs of senescence, ethylene may promote the onset of senescence. To examine whether the decrease in foliar ethylene production in ACC synthase mutants delays entry of leaves into the senescence program, homozygous Zmacs6 mutants (i.e. Zmacs6/Zmacs6), heterozygous Zmacs6 mutants (i.e. ZmACS6/Zmacs6), and B73 wild-type (i.e. ZmACS6/ZmACS6) plants were field-grown until 50 DAP. At this stage, the oldest wild-type leaves had senesced fully, whereas the ZmACS6/Zmacs6 leaves of the same developmental age had just begun to senesce and Zmacs6/Zmacs6 leaves remained fully green (Figure 4a). These observations suggest that the level of ethylene can regulate the onset of senescence in leaves.

Figure 4.

Zmacs6 mutant leaves exhibit delayed senescence.
(a) Leaf 1 from three representative B73 wild-type (i.e. ZmACS6/ZmACS6), heterozygous (i.e. ZmACS6/Zmacs6), and homozygous (i.e. Zmacs6/Zmacs6) plants at 50 DAP. Ethylene production determined from seedling leaves for each genotype is indicated below and the percentage of ethylene in mutant leaves relative to wild-type leaves (defined as 100%) is indicated. Three replicates were measured and the average reported.
(b) Dark-induced senescence in ZmACS6/ZmACS6, ZmACS6/Zmacs6, and Zmacs6/Zmacs6 leaves following 2 weeks of light deprivation. A representative leaf is shown for each genotype.

Senescence can also be induced following deprivation of light. To determine whether reduced foliar ethylene levels can delay the onset of dark-induced senescence, leaves from adult plants were covered with black sheaths to exclude light for 2 weeks. The leaves from younger plants (i.e. 20 DAP) that were allowed to remain attached to the plant were used to ensure that age-related senescence would not occur during the course of the experiment. Greenhouse-grown maize was also employed to avoid any heating that might occur in the field as a consequence of sheathing. Following light deprivation for 2 weeks, senescence was observed over the entire region of wild-type leaves subject to light deprivation (the region covered by the sheath is indicated by the distinct transition from yellow to green, Figure 4b). The tip of ZmACS6/Zmacs6 leaves, representing the developmentally oldest region, had undergone dark-induced senescence but the rest of the covered region showed significantly less senescence than did wild-type leaves (Figure 4b). In contrast, Zmacs6/Zmacs6 leaves remained fully green (Figure 4b). ZmACS6/Zmacs6 leaves produced 70% of wild-type ethylene and Zmacs6/Zmacs6 leaves produced only 14.6% of wild-type ethylene, in good agreement with the results obtained for Zmacs6/Zmacs6 leaves in Figure 3. These results suggest that ethylene mediates the onset of senescence following light deprivation as it does natural senescence. They also indicate that even loss of one copy of ZmACS6 in the heterozygous mutant is sufficient to reduce ethylene and delay the onset of senescence.

To examine whether exogenous ACC could complement the delayed senescence phenotype of the Zmacs6 mutant, the third, sixth, and ninth oldest leaf from 20 DAP wild-type, Zmacs2, Zmacs6 plants were light-deprived and watered with water only or 100 μm ACC for 7 days. All leaves were fully green at the onset of the experiment and remained attached to the plant. Following the dark treatment, senescence had initiated in wild-type leaves (Figure 5), although it had not progressed to the extent observed following a 2-week dark treatment (see Figure 4b). The extent of dark-induced senescence increased as a function of leaf age such that leaf 3 exhibited more senescence than did leaf 6 or leaf 9 (which were younger), confirming that competency for senescence increases with leaf age. Leaf 3 of Zmacs2 exhibited less extensive senescence than that observed in the corresponding wild-type leaves whereas Zmacs6 leaves exhibited no visible sign of senescence (Figure 5), consistent with the observations in Figure 4. However, senescence in light-deprived Zmacs6 leaves was induced when Zmacs6 leaves were treated with 100 μm ACC for 7 days and to an extent similar to that of wild-type leaves (Figure 5). ACC treatment had no effect on Zmacs6 leaves that were maintained in the light, demonstrating that ACC mediated senescence in 20 DAP leaves only in response to light deprivation.

Figure 5.

ACC treatment complements the delayed senescence phenotype of Zmacs6 leaves.
Onset of senescence was examined in the third, sixth, and ninth oldest leaf of adult ZmACS6/ZmACS6 (i.e. B73 wild-type), Zmacs2/Zmacs2, Zmacs6/Zmacs6 plants following 1 week of light deprivation. Light-deprived or light-treated Zmacs6/Zmacs6 leaves were complemented with 100 μm ACC (supplied in water daily during the treatment). A representative leaf is shown for each genotype.

Determination of the level of chlorophyll a and b from leaf 3 confirmed the visual evidence of senescence in that Zmacs6 leaves retained substantially more chlorophyll after light deprivation than did wild-type leaves but did not do so when treated with 100 μm ACC (Figure 6a). Treatment with ACC in itself did not induce premature loss of chlorophyll as chlorophyll was not lost from light-grown Zmacs6 leaves when treated with ACC. Similar results were observed for leaf 6 and leaf 9, although the level of chlorophyll in these younger leaves was higher than in the older leaf 3 samples as was expected (Figure 6a). Similar results were observed for total soluble leaf protein: Zmacs6 leaves retained substantially more protein following light deprivation than did wild-type leaves but did not do so when treated with 100 μm ACC (Figure 6b).

Figure 6.

Senescence-related loss of chlorophyll, soluble protein, and Rubisco is ACC-mediated.
The level of (a) chlorophyll a and b and (b) soluble protein measured in the third, sixth, and ninth oldest leaf of adult ZmACS6/ZmACS6 (i.e. B73 wild type), Zmacs2/Zmacs2, Zmacs6/Zmacs6 plants following 1 week of light deprivation. Light-deprived or light-treated Zmacs6/Zmacs6 leaves were complemented with 100 μm ACC (supplied in water daily during the treatment). Three replicates were measured and the average and standard deviation reported.
(c) The level of Rubisco in the same leaves was determined by Western analysis using rice anti-Rubisco antiserum. Soluble protein from equal fresh weight samples was used.

Western analysis for ribulose bisphosphate carboxylase/oxygenase (Rubisco) demonstrated a substantial loss in B73 leaves following light deprivation (Figure 6c). A greater loss was observed in the oldest leaves (leaf 3) than the youngest (leaf 9). Zmacs6 leaves retained substantially more Rubisco following light deprivation relative to wild-type leaves whereas Zmacs2 leaves retained a moderate level of Rubisco, consistent with the intermediate level of ethylene in this mutant (Figure 6c). Light-deprived Zmacs6 leaves treated with 100 μm ACC lost an amount of Rubisco similar to that of light-deprived wild-type leaves suggesting that ACC complemented the loss of ZmACS6 expression. No loss of Rubisco was observed in ACC-treated Zmacs6 leaves when they remained in the light demonstrating that treatment with ACC alone did not reduce the level of Rubisco. These data demonstrate that the delay in senescence, which involves retention of chlorophyll and leaf protein such as Rubisco, can be complemented by exogenous ACC, suggesting that the delayed senescence in the Zmacs6 mutant is a consequence of a reduction in ACS expression.

ZmACS6 expression regulates leaf function under normal and drought conditions

Drought is known to induce premature onset of leaf senescence. To investigate whether this aspect of the drought response is mediated by ACS expression, homozygous Zmacs6, Zmacs2, and wild-type plants were field-grown under well-watered (8 h twice a week) and water-stressed conditions (4 h per week for a 1-month period that initiated approximately 1 week before pollination and continued for 3 weeks after pollination). During the period of limited water availability, plants exhibited leaf wilting and rolling, visible conformation of water stress.

Senescence of the oldest leaves was evident in wild-type and Zmacs2 plants under well-watered conditions and even more significantly during drought conditions whereas no visible sign of senescence in Zmacs6 leaves was observed (data not shown). To quantitatively support the visible symptoms, the foliar level of chlorophyll a and b was measured. As expected, chlorophyll levels decreased with leaf age (Figure 7a). Under well-watered conditions, the level of chlorophyll in Zmacs6 leaves was up to eightfold higher than in the corresponding leaves of wild-type plants that had initiated senescence. Surprisingly, the level of chlorophyll in all Zmacs6 leaves, including the youngest and those that were fully expanded and exhibited maximum leaf function (see below), was substantially higher than in wild-type plants (Figure 7a). The level of chlorophyll in Zmacs2 leaves was moderately higher than in wild-type plants, consistent with the intermediate level of ethylene in these leaves. These results indicate an inverse correlation between chlorophyll content and ethylene levels in maize leaves: the moderate reduction in ethylene in Zmacs2 plants correlated with a moderate increase in chlorophyll content whereas the large reduction in ethylene in Zmacs6 plants correlated with a substantial increase in chlorophyll content. These results also demonstrate that ethylene regulates the level of chlorophyll in all leaves, including those far from entering into the senescence program and those at maximum leaf function.

Figure 7.

Zmacs6 leaves exhibit enhanced chlorophyll content under normal and drought conditions.
Comparison of the foliar level of chlorophyll a and b (where leaf 1 represents the oldest and leaf 11, the youngest) in B73 wild-type, Zmacs2, and Zmacs6 plants grown under (a) well-watered (non-drought) or (b) drought conditions at 40 DAP.
(c) Comparison of the foliar level of chlorophyll a and b in 40 DAP B73 or Zmacs6 plants grown under well-watered (control) or drought conditions. Drought conditions were imposed by watering for 4 h per week for a 1-month period that initiated approximately 1 week before pollination and continued for 3 weeks after pollination. Three replicates were measured and the average and standard deviation reported.

Imposition of drought conditions reduced foliar levels of chlorophyll in both wild-type and mutant plants but the decrease was substantially greater in wild-type plants (Figure 7b). For example, the level of chlorophyll in leaf 5 of water-stressed wild-type plants decreased 2.5-fold relative to non-drought plants whereas it decreased by only 20% in leaf 5 of water-stressed Zmacs6 plants (Figure 7c). Consequently, reducing ZmACS6 expression resulted in a level of chlorophyll in the oldest leaves that was up to 20-fold higher than in the corresponding leaves of wild-type plants. As observed for non-water-stressed plants, the level of chlorophyll was higher in all Zmacs6 leaves, including the youngest. The chlorophyll content in Zmacs2 leaves also remained moderately higher under drought conditions than in wild-type plants. Thus, loss of ACS expression, and ZmACS6 in particular, delayed drought-induced senescence as measured by loss of chlorophyll content.

As observed for chlorophyll, protein content declined with leaf age (Figure 8a). Under non-drought conditions, the level of protein in Zmacs6 leaves was up to twofold higher than in the corresponding leaves of wild-type plants that had initiated senescence (Figure 8a). As observed for chlorophyll, the level of protein in all Zmacs6 leaves, including the youngest, was substantially higher than in wild-type plants (Figure 8a). Imposition of drought conditions reduces foliar levels of protein in both wild-type and mutant plants but the decrease was substantially greater in wild-type plants (Figure 8b). Moreover, as observed for non-drought plants, the level of protein was higher in all Zmacs6 leaves, including the youngest. Therefore, the loss of ACS expression had a similar effect on chlorophyll and protein, which can serve as biochemical indicators of whether a senescence program has initiated in a leaf.

Figure 8.

Zmacs6 leaves exhibit enhanced protein content under normal and drought conditions.
Comparison of the foliar level of soluble protein (where leaf 1 represents the oldest and leaf 11, the youngest) in B73 wild-type, Zmacs2, and Zmacs6 plants grown under (a) well-watered (non-drought) or (b) drought conditions at 40 DAP.
(c) Comparison of the foliar soluble protein level in 40 DAP B73 or Zmacs6 plants grown under well-watered (control) or drought conditions. Drought conditions were imposed as described in Figure 7. Three replicates were measured and the average and standard deviation reported.

To investigate whether leaf function, for example the ability to transpire and assimilate CO2, was also maintained in leaves of acs plants which would indicate maintenance of leaf performance, the rate of transpiration, stomatal conductance, and rate of CO2 assimilation were measured in every leaf of well-watered Zmacs6 and wild-type plants at 40 DAP when the oldest leaves of wild-type plants had begun to senesce. The youngest leaves of Zmacs6 plants exhibited a higher rate of transpiration (Figure 9a) and stomatal conductance (Figure 9c) than control plants whereas no significant difference was observed in older leaves. In contrast, the rate of CO2 assimilation was substantially higher in all leaves of Zmacs6 plants than in control plants (Figure 9e). Specifically, older leaves of Zmacs6 plants exhibited more than a twofold higher rate of CO2 assimilation than wild-type plants and the rate of CO2 assimilation in younger leaves increased from 50 to 100% (Figure 9e).

Figure 9.

Zmacs6 plants exhibit enhanced leaf performance under normal and drought conditions.
Rate of transpiration (a and b), stomatal conductance (c and d), and rate of CO2 assimilation (e and f) of every leaf (where leaf 1 represents the oldest) from B73 wild-type, Zmacs2, and Zmacs6 plants at 40 DAP grown under well-watered (control) or water-stressed conditions. Drought conditions were imposed as described in Figure 7. Three replicates were measured and the average and standard deviation reported.

The effect of reducing ethylene on the maintenance of leaf function under drought conditions was also investigated. The rate of transpiration (Figure 9b) and stomatal conductance (Figure 9d) were significantly reduced in wild-type leaves when subjected to conditions of drought whereas they remained largely unaffected in Zmacs6 leaves, resulting in a substantially higher rate of transpiration and increased stomatal conductance relative to the wild type. In addition, drought treatment resulted in a significant decrease in the rate of CO2 assimilation in wild-type leaves but not in Zmacs6 leaves, resulting in up to a 2.5-fold increase in CO2 assimilation in younger Zmacs6 leaves and up to a sixfold increase in older Zmacs6 leaves relative to the wild type (Figure 9f). These results indicate that ethylene controls leaf function during conditions of drought and a loss of ACS expression results in a delay of leaf senescence in older leaves while maintaining leaf function in all leaves thus providing greater tolerance to drought.

Discussion

In this study, we show that ethylene plays a significant role in regulating leaf function and lifespan. ZmACS6 expression is largely responsible for directing natural or dark-induced leaf senescence as the loss of ZmACS6 expression resulted in a delayed onset of senescence under both conditions. The loss of ZmACS6 expression resulted in an 85–90% reduction in ethylene production and a delayed entry into the senescence program as determined by loss of chlorophyll and protein which could be complemented by exogenous ACC. Loss of even one copy of ZmACS6 resulted in a 30% reduction in ethylene production which was sufficient to delay partially the onset of dark-induced senescence. Loss of ZmACS2 expression resulted in a 45% reduction in ethylene production and a moderate delay in senescence, suggesting that the regulation of leaf lifespan may be controlled by more than one ACS gene family member. Maintenance of chlorophyll and protein in Zmacs6 leaves was accompanied by the maintenance of normal rates of transpiration, CO2 assimilation, and stomatal conductance demonstrating that the leaves retain physiological and photosynthetic function when equivalent-aged leaves in wild-type plants have fully senesced. The observation that ethylene controls the onset of leaf senescence in maize is consistent with its role in dicot species (Abeles et al., 1992; Davies and Grierson, 1989; Grbic and Bleecker, 1995; John et al., 1995; Picton et al., 1993) and demonstrates conservation of function in a cereal species such as maize.

ACS expression, and in particular expression from ZmACS6, was also responsible for drought-induced senescence observed in maize that was subjected to water stress. Zmacs6 leaves exhibited a delay in drought-induced senescence, retained near-normal levels of chlorophyll and protein, and maintained physiological and biochemical function. These results suggest that ACS expression regulates the onset of senescence under normal growth or during conditions of drought. Thus the induction of senescence under drought conditions may be simply an accelerated onset of the natural ethylene-mediated senescence program.

The level of ZmACS6 transcript accumulation in leaves was lower than that for ZmACS2 or ZmACS7 demonstrating that expression from ZmACS6 is disproportionately responsible for ethylene production in leaves. The level of protein or its activity does not always correspond to the level of transcript accumulation as a result of post-transcriptional regulatory mechanisms that can determine the efficiency of translation, protein stability, or regulation of enzyme activity. An elicitor-induced increase in ACS activity was observed in parsley and tomato cells even when transcription was inhibited, indicating that ACS is, in fact, regulated post-transcriptionally (Chappell et al., 1984; Felix et al., 1991). ACC synthase protein is present in low abundance due, at least in part, to protein instability that is regulated through its C-terminal domain possibly through changes in phosphorylation (Chae et al., 2003; Spanu et al., 1994; Tatsuki and Mori, 2001; Vogel et al., 1998). The putative phosphorylation site in the C-terminal domain is conserved in ZmACS6 (Gallie and Young, 2004) suggesting that it too may be regulated post-transcriptionally. ZmACS2 and ZmACS7 have diverged substantially from ZmACS6 (54 and 53% amino acid identity with ZmACS6, respectively), whereas ZmACS2 and ZmACS7 are highly similar to one another (95% amino acid identity) (Gallie and Young, 2004). This difference may affect protein stability or enzyme activity and may account for the disproportionate contribution to ethylene production in leaves by ZmACS6.

Although the role of ethylene as a regulator of senescence is well known for dicot species, its role in regulating leaf function for those leaves that have not entered the senescence program has not been reported. Loss of ZmACS6 expression not only delayed the onset of natural and drought-induced senescence, but also affected the development and function of all leaves, including the youngest and those that were fully expanded and exhibited maximum leaf function. A higher level of chlorophyll and protein and a higher rate of CO2 assimilation was observed for all leaves of Zmacs6 plants grown under normal field conditions. Increased rates of transpiration and stomatal conductance were observed in the youngest leaves of Zmacs6 plants that, together with the higher level of chlorophyll in all leaves, may account for the higher rate of CO2 assimilation relative to wild-type plants. This suggests that ethylene may control the chlorophyll content in leaves by serving as a negative regulator. Although such a role would be independent of its function in promoting senescence, in both cases, ethylene would function to reduce foliar chlorophyll content. Moreover, the substantially higher rate of CO2 assimilation observed in all leaves of Zmacs6 plants suggests that ethylene negatively controls leaf performance. Among the many physiological and biochemical responses to water stress, a decrease in stomatal conductance and rate of transpiration were observed. However, loss of ZmACS6 expression blocked these stress responses. Thus ethylene may serve to control aspects of leaf development, physiology, and performance under drought conditions as it does under normal growth conditions. Prolonged exposure to water stress not only induced senescence in older leaves, but also reduced the chlorophyll and protein content of all leaves, including the youngest leaves that were far from initiating senescence. Loss of ZmACS6 expression prevented this drought-induced reduction in foliar chlorophyll and protein levels, suggesting that ethylene is responsible for mediating this aspect of the drought response. Moreover, although leaf photosynthetic activity was reduced in wild-type leaves following the imposition of drought conditions, it was maintained in Zmacs6 leaves, suggesting that ethylene may be involved in this response. These results suggest that ACS expression is important in maize in determining the level of leaf function throughout its lifespan and in response to conditions of water stress.

Experimental procedures

Isolation of maize ACC synthase genes

To isolate ACC synthase genes from maize, primers were designed to regions highly conserved among monocot and dicot species. PCR reaction conditions established using maize genomic DNA with primers ACCF1 (CCAGATGGGCCTCGCCGAGAAC) and ACC1 (GTTGGCGTAGCAGACGCGGAACCA) revealed the presence of three, different-sized fragments that were confirmed by sequencing to be maize ACC synthase homologs.

Genomic clones for each of these genes were isolated by radiolabeling the three PCR fragments with dCTP using the Prime-a-Gene labeling system (Promega, Madison, WI, USA). The probes were used to screen an EMBL3 B73 maize genomic library (Stratagene, La Jolla, CA, USA) as described (Sambrook et al., 1989). Briefly, hybridization was carried out for 18 h at 30°C in buffer containing 5X SSPE, 5X Denhardt's, 50% formamide and 1% SDS. Blots were washed at 45°C in 1X SSPE and 0.1X SSPE containing 0.1% SDS and exposed to film. Putative positive plaques were screened by PCR using the above primers. PCR reactions were performed using HotStarTaq (Qiagen, Valencia, CA, USA) in 1X buffer (200 μm of each dNTP, 3 μm MgCl2, 0.25 μm forward and reverse primer), 1.25 U HotStarTaq and 1 μl primary phage dilution (1/600 total in SM buffer) as a template in a total reaction volume of 25 μl. PCR was performed as follows: 95°C/15 min (one cycle); 95°C/1 min, 62°C/1 min, 72°C/2 min (35 cycles); 72°C/5 min (one cycle). Successful amplification was determined following resolution on a 1% agarose gel following by ethidium bromide staining. Correct amplification was confirmed by restriction analysis and the sequence determined.

Identification of ACS knockout mutants

Terminal-inverted-repeat (TIR)-PCR (Bensen et al., 1995) was performed on pools of DNA collected from maize containing Mu using one primer from the target gene (four separate primers specific to the maize ACC synthase genes were used: ACCF1, CCAGATGGGCCTCGCCGAGAAC; ACC-1, GTTGGCGTAGCAGACGCGGAACCA; ACC-C, CAGTTATGTGAGGGCACACCCTACAGCCA; ACC-D, CATCGAATGCCACAGCTCGAACAACTTC) and one primer from the Mu TIR region [AAGCCAACGCCA(A/T)CGCCTC(C/T)ATTTCGT]. Candidate lines were screened by PCR following germination using HotStarTaq (Qiagen). Reactions contained 1X buffer, 200 μm of each dNTP, 3 mm MgCl2, 0.25 μm ACC synthase-specific primer (ACCF1, ACC-1, ACC-C, or ACC-D), 0.25 μmMu-specific primer (MuTIR), 0.25 μl HotStarTaq and 1.5 μl nucleic acid in a 25-μl reaction volume. PCR conditions were: 95°C/15 min (one cycle); 95°/1 min, 62°C/1 min, 72C/2 min (35 cycles); 72°C/5 min (one cycle). Of 13 candidate lines identified to harbor a Mu insertion in one of the ACC synthase genes, five were stably inherited in the first backcross to B73. These lines were backcrossed an additional four times to B72 to reduce unwanted Mu insertions.

Seed backcrossed five times was self-pollinated to generate homozygous null individuals that were identified by PCR using the ACCF1 and ACC-1 primers. PCR of wild-type lines or heterozygous null mutants in amplification of three different-sized fragments corresponding to the three ACC synthase genes whereas PCR amplification of homozygous null mutants lack the fragment corresponding to the mutant gene. The Mu insertion site was determined by sequencing across the Mu/ACC synthase junction using the Mu-TIR primer.

DNA isolation

DNA was extracted from 1 cm2 of seedling leaf, quick-frozen in liquid nitrogen, and ground to a fine powder. A 600-μl volume of extraction buffer [100 mm Tris (pH 8.0), 50 mm EDTA, 200 mm NaCl, 1% SDS, 10 μl ml−1β-mercaptoethanol] was added and the sample mixed. The sample was extracted with 700 μl phenol/chloroform (1:1) and centrifuged for 10 min at 14,400 g. DNA was precipitated and resuspended in 600 μl H2O.

Real-time RT-PCR

Total nucleic acid was isolated from developing kernels using the protocol described above for DNA isolation. Following the initial precipitation and resuspension in TE, total RNA was further purified by two rounds of LiCl2 precipitation according to methods described by Sambrook et al. (1989). Fifty micrograms of total RNA was treated with RQ1 DNase (Promega) to ensure that no contaminating DNA was present. Two micrograms of total RNA was used directly for cDNA synthesis using the Omniscript RT kit (Qiagen) with oligo-dT(20) as the primer.

Analysis of transcript abundance was accomplished using the QuantiTect SYBR Green PCR kit (Qiagen). Reactions contained 1X buffer, 0.5 μl of the reverse transcription reaction (equivalent to 50 ng total RNA) and 0.25 μm (final concentration) forward and reverse primers (see Table 1) in a total reaction volume of 25 μl.

Table 1.  Primers used for Maize ACC Synthase RT-PCR
GeneForward primer (5′-3′)Reverse primer (5′-3′)
ZmACS2ATCGCGTACAGCCTCTCCAAGGAGATAGTCTTTTGTCAACCATCCCATAGA
ZmACS7ATCGCGTACAGCCTCTCCAAGGACAACGTCTCTGTCACTCTGTGTAATGT
ZmACS6AGCTGTGGAAGAAGGTGGTCTTCGAGGTAGTACGTGACCGTGGTTTCTATGA

Reactions were carried out using an ABI PRISM 7700 sequence detection system (Applied Biosystems, Foster City, CA, USA) under the following conditions: 95°C/15 min (one cycle); 95°C/30 sec, 62°C/30 sec, 72°C/2 min (50 cycles); 72°C/5 min (one cycle). Each gene was analyzed a minimum of four times and the average and standard deviation reported. All the primer combinations were initially run and visualized on an agarose gel to confirm the presence of a single product of the correct size. All amplification products were subcloned into the pGEM-T Easy vector system (Promega) to use for generation of standard curves to facilitate conversion of expression data to a copy/μg RNA basis.

Ethylene determination

Ethylene was measured from seedling leaves or from the terminal 15 cm of leaves of plants 20, 30, or 40 DAP. Leaves were harvested at the indicated times and allowed to recover for 2 h prior to collecting ethylene. Leaves were placed into glass vials and capped with a rubber septum. Following a 3–4-h incubation, 0.9 ml of headspace was sampled from each vial and the ethylene content measured using a 6850 series gas chromatography system (Hewlett-Packard, Palo Alto, CA, USA) equipped with a HP Plot alumina-based capillary column (Agilent Technologies, Palo Alto, CA, USA). Tissue fresh weight was measured for each sample. Three replicates were measured and the average and standard deviation reported.

Western blot analysis

Leaves were ground in liquid nitrogen to a fine powder. One milliliter of extraction buffer [20 mm HEPES (pH 7.6), 100 mm KCl, 10% glycerol, 1 mm PMSF] was added to approximately 0.1 g frozen powder and mixed thoroughly. Cell debris was pelleted by centrifugation at 12,000 g for 10 min and the protein concentration determined as described (Bradford, 1976).

Antiserum raised against the large subunit of rice Rubisco was obtained from Dr Tadahiko Mae (Tohoku University, Sendai, Japan). Protein extracts were resolved using standard SDS-PAGE and the protein transferred to 0.22 μm nitrocellulose membrane by electroblotting. Following transfer, the membranes were blocked in 5% milk, 0.01% thimerosal in TPBS (0.1% TWEEN 20, 13.7 mm NaCl, 0.27 mm KCl, 1 mm Na2HPO4, 0.14 mm KH2PO4) followed by incubation with primary antibodies diluted typically 1:1000–1:2000 in TPBS with 1% milk for 1.5 h. The blots were then washed twice with TPBS and incubated with goat anti-rabbit horseradish peroxidase-conjugated antibodies (Southern Biotechnology Associates, Inc., Birmingham, AL, USA) diluted to 1:5000–1:10 000 for 1 h. The blots were washed twice with TPBS and the signal detected typically between 1 and 15 min using chemiluminescence (Amersham Corp., Piscataway, NJ, USA).

Chlorophyll quantitation

Leaves were ground in liquid nitrogen to a fine powder. Approximately 0.1 g was removed to a 1.5-ml tube and 1 ml of acetone added. Chlorophyll was extracted from a weighed sample five times with 0.8 ml of 80% acetone. Individual extractions were combined and the final volume adjusted to 15 ml with additional 80% acetone. Chlorophyll a and b was determined spectrophotometrically as described (Wellburn, 1994).

Measurement of photosynthesis

Plants were grown in the field under normal and drought-stress conditions. Normal plants were watered for 8 h twice a week. For drought-stressed plants, water was limited to approximately 4 h per week for a 1-month period starting approximately 1 week before pollination and continuing through 3 weeks after pollination. During the period of limited water availability, drought-stressed plants showed visible signs of leaf wilting and rolling. Transpiration, stomatal conductance and CO2 assimilation were determined with a portable TPS-1 Photosynthesis System (PP Systems, Amesbury, MA, USA). Each leaf on a plant was measured at 40 DAP. Values represent a mean of six determinations.

Acknowledgements

The authors thank Patricia Springer for use of light microscope and Dr Tadahiko Mae for the rice Rubisco antiserum. This work was supported by grant NRICGP 2002-00743 from the United States Department of Agriculture.

Ancillary