ABSTRACT
Exposure to elevated tropospheric ozone concentration ([O3]) accelerates leaf senescence in many C3 crops. However, the effects of elevated [O3] on C4 crops including maize (Zea mays L.) are poorly understood in terms of physiological mechanism and genetic variation in sensitivity. Using Free Air gas Concentration Enrichment (FACE), we investigated the photosynthetic response of 18 diverse maize inbred and hybrid lines to season-long exposure to elevated [O3] (~100 nL L−1) in the field. Gas exchange was measured on the leaf subtending the ear throughout the grain filling period. On average over the lifetime of the leaf, elevated [O3] led to reductions in photosynthetic CO2 assimilation of both inbred (-22%) and hybrid (-33%) genotypes. There was significant variation among both inbred and hybrid lines in the sensitivity of photosynthesis to elevated [O3], with some lines showing no change in photosynthesis at elevated [O3]. Based on analysis of inbred line B73, the reduced CO2 assimilation at elevated [O3] was associated with accelerated senescence decreasing photosynthetic capacity, and not altered stomatal limitation. These findings across diverse maize genotypes could advance the development of more ozone tolerant maize, and provide experimental data for parameterization and validation of studies modeling how O3 impacts crop performance.
INTRODUCTION
Tropospheric ozone (O3) is an airborne pollutant that enters leaves through their stomata and generates reactive oxygen species (ROS) upon contact with intercellular leaf surfaces (Heath 1988). Subsequently, these ROS can cause oxidative damage to membranes and other cellular components. At very high concentrations, O3 can elicit a hypersensitive response, but even at lower concentrations O3 can impair cellular function (Krupa & Manning 1988). When exposed to elevated O3 concentrations ([O3]) throughout the growing season, many plants have reduced photosynthetic C assimilation and biomass production as well as increased antioxidant capacity and mitochondrial respiration rates (Fiscus et al. 2005; Ainsworth et al. 2012). As a cumulative result of these physiological responses, crops exhibit significant reductions in economic yield with increasing [O3] (Mills et al. 2007; Ainsworth 2017).
The effects of exposure to chronic, elevated [O3] on photosynthesis and stomatal conductance can vary with leaf age and canopy position (Tjoelker et al. 1995; Morgan et al. 2004; Betzelberger et al. 2010; Feng et al. 2011). Studies that repeatedly measured a cohort of soybean and wheat leaves found that photosynthesis and stomatal conductance were not significantly affected by elevated [O3] in recently mature leaves, but continued exposure over time accelerated the decline in photosynthetic capacity as leaves aged (Morgan et al. 2004; Feng et al. 2011; Emberson et al., 2017). Stomatal conductance can also become uncoupled from photosynthesis as leaves age under O3 stress, often with greater measured reductions in photosynthesis than stomatal conductance (Lombardozzi et al. 2012), resulting in decreased water use efficiency at elevated [O3] (Tjoelker et al. 1995; VanLoocke et al. 2012). Additionally, stomata of plants exposed to elevated [O3] can show delayed opening or closing responses to environmental or hormonal signals (Keller & Häsler 1984; Wilkinson & Davies 2009; Paoletti & Grulke 2010; Wagg et al. 2013).
Much of our understanding of the effects of elevated [O3] on photosynthesis and stomatal function come from study of C3 species (Reich 1987; Paoletti & Grulke 2005; Felzer et al. 2007; Wittig et al. 2007). A few experiments have shown that elevated [O3] can negatively impact photosynthetic properties and biomass production of the model C4 species, maize (Zea mays L.) (Pino et al. 1995; Leitao et al. 2007a; Leitao et al. 2007b; Bagard et al. 2015; Yendrek et al. 2017). However, current understanding of the mechanisms underlying photosynthetic responses to O3 in maize is limited to a few genotypes, and often to juvenile growth stages. The high ratio of photosynthetic CO2 assimilation relative to stomatal conductance and isolation of Rubisco carboxylation in the bundle sheath cells make it possible that C4 photosynthesis will respond distinctly to C3 photosynthesis at elevated [O3]. Analysis of historical U.S. maize yields suggests that reported physiological responses to O3 pollution drive annual yield reductions of approximately 10% from 1980-2011, amounting to $7.2 billion in lost profit per year on average (McGrath et al. 2015). Considering that O3-mediated yield reductions were more prominent in hot and dry years (McGrath et al. 2015) and future climate models predict an increase in growing season temperature with altered frequency and magnitude of precipitation events (Christensen et al. 2007; Walsh et al. 2014), efforts to address the knowledge gap about mechanisms of O3-sensitivity in maize are likely to become increasingly important.
In addition to being the most widely produced food crop, maize is a model C4 species with considerable resources for analysis of genotype to phenotype associations. This includes the maize nested association mapping (NAM) population, which was developed to encompass much of the existing diversity in a relatively small number of inbred lines (Yu et al. 2008). Substantial phenotypic diversity for many agronomic traits exists within the parental lines of the NAM (Flint-Garcia et al. 2005). In order to exploit these resources to locate genetic factors associated with complex traits such as O3 tolerance, there is a need to first identify lines that respond differently to elevated [O3] for key physiological traits. For understanding and modeling the impacts of O3 on crops, photosynthesis and stomatal conductance are two key traits (Reich 1987; Emberson et al. 2000; Sitch et al. 2007). In this study, we investigated the effects of season-long growth at elevated [O3] on maize photosynthetic gas exchange. Maize was exposed to elevated [O3] in the field using Free Air gas Concentration Enrichment (FACE). In 2013 and 2014, we studied the inbred cultivar B73. We hypothesized that growth at elevated [O3] would reduce photosynthetic capacity and daily C gain as leaves progressed more rapidly through senescence. In 2015, we investigated the response of photosynthesis and stomatal conductance to elevated [O3] in 10 diverse inbred and 8 diverse hybrid lines in order to test for genetic variation in O3 response.
MATERIALS AND METHODS
Field site and experimental conditions
Maize (Zea mays) inbred and hybrid lines (Table 1) were studied at the FACE research facility in Champaign, IL (www.igb.illinois.edu/soyface/) in 2013, 2014 and 2015. This facility is located on a 32 ha farm. Maize is grown on half of the land each year, and rotated annually with soybean. Each year, the portion of the field growing maize was fertilized with N (180 lbs per acre) and treated with pre-emergent and post-emergent herbicides at the recommended rates. Additional weeding was done inside the FACE rings as needed. Maize inbred and hybrid lines were planted with a precision planter in rows spaced 0.76 m apart and 3.35 m in length, at a density of 8 plants m−1 in replicated blocks (n=4). Each block had one ambient plot and one elevated [O3] plot, which were octagonal in shape and of 20-m diameter. Weather conditions were monitored at the FACE facility, and Supplemental Figure 1 shows the seasonal time courses of: (1) daily maximum, (2) daily minimum temperatures, and (3) daily total precipitation for the 2013, 2014 and 2015 growing seasons.
Air enriched with O3 was delivered to the experimental rings with FACE technology, as described in Yendrek et al. (2017). The target fumigation set-point in the elevated [O3] rings was 100 nL L−1, which was imposed for ~8 h per d throughout the growing season except when leaves were wet or wind speeds were too low to ensure accurate fumigation (i.e., <0.5 m s−1). Over the three years, when the fumigation was on, the averaged 1-min [O3] within the treatment rings was within 10% of the 100 nL L−1 set point for 53.7% of the time and within 20% of the set point for 78% of the time. Season-long 8 hr average [O3] in ambient rings was 40.8 nL L−1 in 2013, 40.1 nL L−1 in 2014, and 40.0 nL L−1 in 2015, and season-long elevated [O3] was 71.0 nL L−1 in 2013, 70.8 nL L−1 in 2014 and 64.0 nL L−1 in 2015.
Leaf-level gas exchange measurements
In situ photosynthetic gas exchange of the leaf subtending the ear was measured using portable photosynthesis systems (LI-6400, LICOR Biosciences, Lincoln, NE) in a modified version of a previously published protocol (Leakey et al. 2006). The net rate of photosynthetic CO2 assimilation (A), stomatal conductance (gs), the ratio of intercellular [CO2] (ci) to atmospheric [CO2] (ca), and instantaneous water use efficiency (iWUE = A/gs) were measured. Reproductive development (time to silking and anthesis) was monitored in all experimental rows, and all dates of gas exchange measurements are reported relative to the date of anthesis.
In 2013 and 2014, the diurnal time course of leaf photosynthetic gas exchange was assessed with measurements taken across all replicate plots every few hours throughout the day at three or four development stages over the growing season of the B73 inbred genotype. In 2015, in situ gas exchange measurements on 10 inbred and 8 hybrid lines were collected at midday approximately every 7 d starting when the plants reached anthesis and ending when the ear leaf of >50% plants in an individual row had fully senesced. After senescence of a genotype in a given treatment, we recorded zero for A and gs for all subsequent weekly measurements until both ambient and elevated plots were fully senesced. In all years, two or three plants were measured as sub-samples within each replicate plot at either ambient [O3] or elevated [O3].
Prior to a set of measurements being performed over a 1-2 h period of a day in all replicate plots, temperature and relative humidity (RH) in the leaf chamber cuvette were set to match prevailing ambient conditions as measured by an on-site weather station. A linear quantum sensor (AccuPAR LP-80; Decagon Devices, Pullman, WA) was used to measure the average light intensity at the position in the canopy for the leaf subtending the ear, which was used to set the PPFD incident on the leaf in the gas exchange cuvette. In 2015, a constant PPFD was used for all weekly midday measurements based on the measured PPFD at anthesis. For inbred cultivars, PPFD was set at 1,800 μmol m−2 s−1 and for hybrid cultivars PPFD was set at 450 μmol m−2 s−1. Keeping light constant over the entire measurement period enabled comparison of changes in the response of A and gs to leaf aging in ambient and elevated [O3], and eliminated variation in those parameters due to PPFD. On all dates, four instruments were used simultaneously. All environmental conditions were held constant across instruments.
The response of A to ci was measured on the leaf subtending the ear in 2013 and 2014 to examine the effects of elevated [O3] on the maximum apparent rate of phosphoenolpyruvate carboxylase activity (Vpmax) and on CO2-saturated photosynthetic rate (Vmax). Two to three leaves per ring were measured. Leaves were excised pre-dawn, immediately re-cut under water, and measured in a field laboratory. This approach minimized short-term decreases in water potential, chloroplast inorganic phosphate concentration, and maximum PSII efficiency that can occur in the field, and also enabled more leaves to be measured on multiple portable photosynthesis systems over a short period of time (Leakey et al. 2006). Measurements were initiated at ambient ci, then the reference [CO2] in the leaf cuvette was stepped down to 25μmol mol−1, before it was increased stepwise to 1200 μmol mol−1 while keeping PPFD constant at 2,000 μmol m−2 s−1. The initial slope of the A/ci curve (ci < 60 μmol mol−1) was used to calculate Vpmax according to von Caemmerer (2000). A four parameter nonrectangular hyperbolic function was used to estimate Vmax as the horizontal asymptote of the A/ci curve. Stomatal limitation to A (l) was estimated from A/ci curves using the approach described by Farquhar and Sharkey (1982), and modified by Markelz et al. (2011). Mean values of Vpmax and Vmax from A/ci curves were used in combination with in situ values of ci measured at midday to estimate the range of l and mean l for recently mature leaves (DOY 219 in 2013, DOY 213 in 2014) and senescing leaves (DOY 238 in 2013, DOY 245 in 2014) in ambient and elevated [O3].
Statistical analysis
Diurnal gas exchange parameters (A, gs) measured in 2013 and 2014 were analyzed with a repeated mixed model ANOVA with the Kenwood-Rogers option and compound symmetry covariance structure (SAS, version 9.4, Cary, NC). All analysis was done on the ring mean. Days were analyzed independently, time of day was a repeated measure and block was a random effect. A mixed model ANOVA with the Kenwood-Rogers specification was used to analyze Vpmax and Vmax in 2013 and 2014 (SAS, Version 9.4, Cary, NC). O3 treatment and DOY were fixed effects in the model, block was a random effect, and years were analyzed independently. Statistical differences in least squared mean estimates were determined by linear contrasts with a threshold p<0.05.
A and gs of the flag leaf measured in 2015 were fit with a quadratic equation: y = yo + αx +βx2, where y was A or gs, and x was days after anthesis (Proc Nlin, SAS). Other models (exponential decay, 3-parameter Weibull function) were tested, but did not fit all of the data for each genotype. Therefore, a simple quadratic equation was used. In order to test if there were differences in the decline in A or gs over time in ambient and elevated [O3], a single quadratic model was first fit to the A and gs data for each genotype, then models were fit to each genotype and treatment combination. An F statistic was used to test if the model with genotype and treatment produced a significantly better fit to the data, i.e., if there were significant differences in the response of A or gs over time in ambient and elevated [O3]. Additionally the quadratic model was fit to each genotype within each ring in order to obtain parameter estimates for yo, α and β. Parameter estimates as well as initial values of A and gs measured shortly after anthesis were analyzed with a mixed model ANOVA with the Kenwood-Rogers specification. O3 treatment and genotype were fixed effects in the model and block was a random effect. Statistical differences in least squared mean estimates between ambient and elevated [O3] for each inbred or hybrid line were determined by linear contrasts with a threshold p<0.05.
RESULTS
Elevated [O3] accelerates loss of photosynthetic capacity in inbred line B73
In 2013 and 2014, daily time courses of leaf photosynthetic gas exchange were measured in situ as the leaf subtending the ear aged in ambient and elevated [O3] (Figs. 1, 2). When the leaf was recently fully expanded, shortly before or after anthesis, there was no significant effect of elevated [O3] on A (Figs. 1d, 2e, 2f). However, as the leaf aged, growth at elevated [O3] accelerated the decline in diurnal C gain, and later in the season A was significantly reduced by elevated [O3] (Figs. 1e, 1f, 2g, 2h). Significant differences in A between ambient and elevated [O3] were apparent earlier in the growing season than differences in gs (Figs. 1e versus 1h, 2g versus 2k). But, there were no consistent differences in the ratio of intercellular [CO2] to atmospheric [CO2] (ci/ca) or instantaneous water use efficiency (iWUE = A/gs) in ambient and elevated [O3] in either year (Fig. 1j-o, Fig. 2m-t). On only one sampling date out of seven, which was characterized by low PPFD, elevated [O3] led to significantly greater ci/ca and significantly lower iWUE (Fig. 1k, n).
Declines in in situ A over the leaf aging process were associated with decreased photosynthetic capacity, measured as Vpmax (Fig. 3a, b; Table 2) and Vmax (Fig. 3c, d; Table 2). Declines in Vpmax from when the leaf was recently fully expanded to until ~30 days later were greater in elevated [O3] (-64% in 2013 and −64% in 2014) than in ambient [O3] (-51% in 2013 and −33% in 2014; Fig. 3a, b). Likewise, declines in Vmax from when the leaf was recently fully expanded to until ~30days later were greater in elevated [O3] (-40% in 2013 and −52% in 2014) than in ambient [O3] (-25% in 2013 and −33% in 2014; Fig. 3c, d). Although an O3 treatment effect was not significant in the mixed ANOVA model (Table 2), pair-wise comparisons of the means showed that Vpmax was significantly lower in elevated [O3] on the final sampling date in 2014 (Fig. 3b) and Vmax was significantly lower in elevated [O3] on the final sampling date in both 2013 and 2014 (Fig. 3c, d).
A/ci curves were also used to assess stomatal vs. biochemical limitations to A in aging leaves grown at ambient and elevated [O3]. Average A/ci curves in ambient and elevated [O3] were plotted for the leaf subtending the ear shortly after anthesis, and then ~3-4 weeks later (Fig. 4). Shortly after anthesis, l was minimal in both ambient and elevated [O3] in 2013 (2-3%), and mean ci was well above the inflection point of the A/ci curve (Fig. 4a). In 2014, there was a wider range of observed ci shortly after anthesis, but mean l was still low (5-6%) in both ambient and elevated [O3], and the mean ci was above the inflection point (Fig 4b). As leaves aged, l increased in both ambient and elevated [O3] to 20-30% (Fig. 4c, d). Both increased l and decreased photosynthetic capacity result in lower A in aging leaves, but elevated [O3] did not appear to change l.
Photosynthetic response to elevated [O3] in diverse inbred and hybrid lines
In 2015, A of the leaf subtending the ear was measured approximately weekly in 10 diverse inbred lines and 8 hybrid lines during the grain filling period. Initial values of A measured shortly after anthesis differed among lines, but not between ambient and elevated [O3], and the interaction between inbred line and [O3] was only marginally significant (Table 3). Similarly, initial measurements of gs varied among inbred lines, but not between ambient and elevated [O3] (Table 3). For 8 of the 10 inbred lines, there was significant acceleration in the decline in A as the leaf aged under elevated [O3], similar to what was observed in previous years in B73 (Fig. 5). However, for two inbred lines (M37W and CML333), there was no significant difference in the quadratic fit to the photosynthetic data between ambient and elevated [O3] (Fig. 5). There were highly significant effects of [O3], inbred line and their interaction on the parameters describing the quadratic equation modeling the decline in A over time in the inbred lines (Table 3). This reflected variation in the estimates of A at anthesis (yo), the time to the start of a decline in A (α) and the rate of decline in A (β). The significant effect of [O3] on yo is not consistent with the lack of an effect on initial rates of A, but some of the data were best fit by convex functions and some by concave functions which affected accurate prediction of yo in the inbred lines (Fig. 5). Average α was 0.30 in ambient [O3] and −0.69 in elevated [O3], indicating that a decline in A started earlier in elevated [O3] (Fig. 5, Supplemental Table 1). Average β was −0.016 in ambient [O3] and −0.00025 in elevated [O3] indicating that the rate of decline in A was greater in ambient [O3] than elevated [O3] on average across inbred lines (Supplemental Table 1).
Measurements of gs in the aging flag leaf in inbred lines differed from the response of A over time. For 7 of the 10 genotypes, there was no evidence for accelerated decline in gs over time (Fig. 6). Of the remaining 3 genotypes, MS71 and C123 displayed more rapid decline in gs over time, while NC338 had initially higher gs in elevated [O3]. The influence of this small set of genotypes meant that there were significant effects of [O3], inbred line and their interaction on the shape of the quadratic fit to the decline in gs over time, especially on parameters α and β (Table 3; Supplemental Table 1). Apparent differences in the response of A and gs to elevated [O3] over the grain filling period did not impact the iWUE, which was not different in ambient and elevated [O3] (Supplemental Fig. 2).
All 8 of the hybrid lines showed acceleration in the decline of A in the leaf subtending the ear over the grain filling period (Fig. 7). Unlike the inbreds, the hybrids had similar initial values of A and gs shortly after anthesis (Table 3). These measurements were taken at a lower light level than the inbred measurements, which was representative of the light measured in the canopy at the leaf subtending the ear in hybrids vs. inbreds, but may explain why there was less variation among hybrid lines. There was a significant effect of [O3] and a significant [O3] x hybrid line interaction on the parameters describing the decline in A over time (Table 3; Supplemental Table 1). Consistent with the inbred lines, α was positive in ambient [O3] (1.06) and negative in elevated [O3] (-0.033), indicating a more immediate decline in A in elevated [O3], and β was more negative in ambient [O3] (-0.02) than elevated [O3] (-0.006) indicating more rapid decline in A in ambient [O3] (Supplemental Table 1). Only 3 of the 8 hybrid lines showed significant changes in gs at elevated [O3] over the grain filling period (Fig. 8), and consistent with the inbred lines, the apparent differences in the response of A and gs to elevated [O3] over time did not impact iWUE (Supplemental Fig. 3). Notably, the hybrids of MS71 and C123 crossed with B73 were among this sensitive group, matching the greater than average sensitivity of these genotypes as inbreds.
DISCUSSION
Tropospheric [O3] continues to increase in many of the world’s crop growing regions as a result of inconsistent regulation of precursor pollutants around the globe and short- and long-distance transport of pollutants (Lin et al. 2017). Early studies of maize reported that it was significantly more tolerant to increasing [O3] than C3 crops like soybean or peanut (Heck et al. 1983). However, more recent experimental studies using open top chambers reported that maize leaf physiology is sensitive to chronic O3 exposure in the cultivar Chambord (Leitao et al. 2007a; Leitao et al. 2007b). Furthermore, multiple regression analysis of county-level yield data suggests that there is significant yield loss to O3 in the Midwest U.S. growing region (McGrath et al. 2015). Here, we examined how season-long fumigation with elevated [O3] using FACE technology in the primary area of maize production affected gas exchange of the leaf subtending the ear. We concentrated on this leaf and time period because previous research has established that most of the photosynthate used for grain filling in maize is provided by mid-canopy leaves after anthesis (Borrás et al. 2004), and those leaves are the last on the plant to senesce (Valentinuz & Tollenaar 2004). Diurnal measurements of B73 showed that ear leaf C assimilation was not affected by elevated [O3] around the time of anthesis when the leaf was recently mature; however, loss of photosynthetic capacity of that leaf was accelerated by elevated [O3] (Figs. 1 & 2).
The seasonal pattern of elevated [O3] effects on photosynthetic carbon gain over the diurnal period was reflected in measurements of A at midday (Figs. 1 & 2). Therefore, screening of midday A was used to explore genotypic variation in sensitivity to elevated [O3] among diverse hybrid and inbred lines of maize. On average, the sum of midday CO2 assimilation by the leaf subtending the ear over the period from anthesis to complete leaf senescence was reduced by 22% in inbred lines and 33% in hybrid lines (Table 4). However, sensitivity to elevated [O3] in inbred lines ranged from no loss in photosynthetic CO2 assimilation (CML333, M37W) to 59% lower CO2 assimilation (C123; Fig. 5, Table 4). Likewise, sensitivity to elevated [O3] in hybrids ranged from 13% lower photosynthetic CO2 assimilation (B73 x M37W) to 44% lower CO2 assimilation (B73 x MS71; Fig. 7, Table 4). All ozone sensitive lines showed the same developmentally dependent pattern of response as B73, with similar A in young leaves at ambient and elevated [O3], but more rapid decline in A over time (Fig. 5, 7). Thus, acceleration of senescence rather than decreased initial investment in photosynthetic capacity is a key response of maize to elevated [O3] at the leaf level. But, there is genetic variation in response that might be exploited to breed maize with improved tolerance to ozone pollution, and possibly even broader spectrum oxidative stress tolerance.
Accelerated loss of photosynthetic capacity in elevated [O3] has been documented in FACE experiments with soybean (Morgan et al. 2004; Betzelberger et al. 2010), wheat (Feng et al. 2011; 2016), rice (Pang et al. 2009) and now maize. In rice and wheat, accelerated senescence of the flag leaf at elevated [O3] shortened the grain filling duration and reduced seed yields (Gelang et al. 2000; Pang et al. 2009; Feng et al. 2011). Additionally, genetic variation in the response of seed yield to elevated [O3] was attributed in part to variation in leaf senescence (Pang et al. 2011; Feng et al. 2011) as well as to variation in detoxification and antioxidant capacity (Feng et al. 2010; 2016). Acceleration of senescence in the ear leaf of maize is associated with maize yield loss under a number of environmental stresses (Wolfe et al. 1988; Bänzinger et al. 1999), and greater yield potential of newer maize varieties has been associated with maintenance of photosynthetic capacity of the ear leaf for a longer period of time following anthesis (Ding et al. 2005). Thus, the acceleration of senescence reported here under elevated [O3] for both inbred and hybrid maize has the potential to contribute to grain yield losses.
We found that growth at elevated [O3] both accelerated the speed at which flag leaves lost photosynthetic capacity (represented by the β parameter in our quadratic model) and decreased the length of time that leaves maintained optimum photosynthetic capacity (α term in quadratic model). It would be interesting to test the response of maize lines with delayed leaf senescence or stay-green phenotypes for O3 response. There are a variety of genetic and physiological mechanisms that enable a stay-green phenotype (Thomas & Howarth 2000), and there may be mechanisms that are more promising under O3 stress than others. For example, stay-green lines with altered chlorophyll catabolism may be less O3-tolerant as the build-up of phytotoxic degradation products might compete with O3-induced reactive oxygen species for detoxification by the anti-oxidant system. However, stay-green lines with perennial tendencies or lines where senescence is initiated later, but progresses at the normal rate and with typical catabolism may have the potential to retain higher rates of photosynthesis later in the growing season, and therefore yield better in higher [O3].
Studies in a number of C3 species have indicated that biochemical limitation, and not stomatal limitation, are responsible for the sustained effects of elevated [O3] on A (Farage et al. 1991; Morgan et al. 2004; Paoletti & Grulke 2005; Feng et al. 2011). In C3 species, accelerated decline in A in aging leaves was associated with decreased maximum carboxylation capacity and electron transport capacity (Zheng et al. 2002; Morgan et al. 2004; Feng et al. 2011). We found a similar response in the maize inbred line B73 with the decline in A at elevated [O3] associated with decreased photosynthetic capacity (Vpmax and Vmax), and not with any change in limitation imposed by stomata. Stomatal limitation increased from 2-5% in the ear leaf when it was recently mature to 20-30% when it was 4-5 weeks older, which is consistent with the loss of stomatal control in aging leaves previously described for other species (Reich 1984). This increase in stomatal limitation occurred in both ambient and elevated [O3], and so there does not appear to be evidence for an uncoupling of A and gs in elevated [O3] that has been described in other studies (Paoletti & Grulke 2005). Across the diverse inbred and hybrid lines grown at elevated [O3], both A and gs declined with ear leaf age (Figs. 5-8), although the reduction in A was often greater than gs. Still iWUE remained relatively constant (Supplemental Fig. 2, 3), providing further evidence from diverse maize lines that growth at elevated [O3] does not result in uncoupling of A and gs.
In addition to testing for cultivar variation in response to elevated [O3], we were also able to identify significant variation in A and gs in recently mature flag leaves of inbred maize lines (Table 3). The hybrid lines, which all contained B73 as the female parent, did not show variation in A and gs (Table 3). Previous studies have also reported that inbred lines show greater variation in photosynthetic traits compared to hybrid lines (Albergoni et al. 1983; Ahmadzadeh et al. 2004), but that hybrid lines maintain higher rates of A over a leaf lifespan compared to inbred lines (Ahmadzadeh et al. 2004). We found similar results in our studies of diverse maize lines, however, it should be duly noted that due to the recurrence of B73 as the female parent, there was more limited genetic diversity in the hybrids used in this study. Additionally, the light environment at the ear leaf at anthesis was different in inbred and hybrid canopies. A and gs were monitored at higher PPFD in the inbred lines (1,800 μmol m−2 s−1) compared to the hybrid lines (450 μmol m−2 s−1), which may have also contributed to the greater variation observed amongst inbred lines.
Our study showed that accelerated loss of photosynthetic capacity was the basis for photosynthetic sensitivity to elevated [O3] in both inbred and hybrid maize lines. Drought and high temperature stress also negatively impact photosynthesis in maize (Bänzinger et al. 1999; Neiff et al. 2016), and those stresses often co-occur with high O3 events, and will likely increase in the future (Cook et al. 2014). Further experimental studies are needed to understand the interactions of rising [O3] with intensifying drought and heat stress, but our work suggests that a key trait for robust improvement of maize response to elevated [O3] is maintenance of photosynthetic capacity during the grain filling period. By testing a diverse panel of maize genotypes under field conditions in the world’s primary area of production, this study provides a foundation on which to investigate the genetic variation in maize oxidative stress tolerance, and possibly develop more stress tolerant germplasm. Under elevated [O3], both inbred and hybrid maize lines were shown to have reduced carbon gain of the leaf subtending the ear, which heavily influences yield. In contrast to many previous studies of crop responses to atmospheric change using FACE technology (e.g. Markelz et al. 2011; Gillespie et al. 2012), this study describes the average and variance in treatment effects across diverse genotypes, and so should provide experimental data that is a stronger basis for parameterization or validation of studies attempting to model regional crop performance.
ACKNOWLEDGMENTS
This work was supported by the National Science Foundation (grant no. PGR–1238030) and the USDA ARS. We thank Don Ort for directing the SoyFACE facility, Alvaro Sanz for assistance with gas exchange measurements and Kannan Puthuval, Brad Dalsing and Chad Lantz for management of the FACE rings.