Full Thesis is on Research Gate DOI:10.13140/RG.2.2.14708.63368
CHAPTER 3
CHARACTERISATION OF STOMATAL AND LEAF PHYSIOLOGY IN RICE LEAVES GROWN IN HIGH AND LOW IRRADIANCE
3.1 Introduction
Many comparative studies have demonstrated that leaf development involves responses of plants to environmental cues such as light (Vogelmann and Martin 1993; Strauss-Debenedetti and Berlyn 1994). This developmental plasticity (an ability to adjust to the environment by means of morphological and/or physiological responses) is distinctive to plants (Sultan, 1995). Thus although, for example, photosynthetic responses to changes in light intensity can include short term changes, such as increasing electron transport rate (Bailey et al., 2001), it can also involve more long term changes involving biochemical and morphological acclimation. For example, in high light (HL) conditions many plants generate “sun leaves”, generally characterised by being smaller and thicker than leaves generated under low light (LL) conditions (“shade leaves”). The cellular basis of these changes have been investigated in some plants. For example, Kubinova (1991) showed that barley plants grown in a HL environment produced sun-type leaves which were thicker due to an increase in the number of mesophyll cells and an overall increase in mesophyll volume. However, not all plants show such extreme differences in sun/shade leaf acclimation response. In those that do, the responses are proposed to allow the plant to maintain photosynthesis and minimise stress in a changing environment (Takahashi & Badger, 2011).
The basic concept is that shade leaves acclimate to maximize the light captured while sun leaves need to dissipate excess adsorbed energy to prevent irreversible damage to the photosynthetic machinery. A number of these changes occur at the molecular level, for example modulating PSI and PSII antenna size (Anderson et al., 1995), whereas others alter plant morphology, such as leaf area and thickness (Murchie & Horton 1997). The process thus involves changes at various scales and Figure 3.1 summarizes the functional key differences between sun and shade plants at these various cellular, leaf and plant levels. The study reported in this chapter was performed to highlight/characterise the major differences between sun and shade leaves in rice in our growth conditions, setting the foundations for the experiments reported in Chapter 4 where I investigate how the sensitivity of the response is dependent on leaf developmental stage.
Figure 3.1: Physical and physiological differences between sun and shade plants. Primary characteristics are highlighted at whole plants, leaves and cells levels. Figure adapted from: Plants In Action, http://plantsinaction.science.uq.edu.au/edition1/?q=content/12-1-1-light-interception-and-utilisation |
A specific aspect of the leaf epidermis of particular relevance to photosynthetic performance, which shows responses to altered light environment, is the stomata. Most of this work has been done on stomata in dicots. For example, Arabidopsis stomatal density varies significantly according to the light environment and Fu et al. (2010) reported that stomatal density and index (percentage stomata out of the total number of epidermal cells plus stomata) for HL grown Capsicum increased, leading to higher values compared to the LL grown plants. However, although such changes may lead to altered gas exchange, the relationship can be complicated. For example, Bussis et al. (2006) showed that increased stomatal density was compensated by reduced stomatal aperture, leading to a constant ratio of internal CO2 concentration in the leaf relative to the ambient CO2 concentration. Understanding the control of these changes in stomatal properties is complicated by the observation that site of light signal perception and the site of altered stomatal differentiation may be physically remote. Thus, Coupe et al. (2006) showed that Arabidopsis leaves acquired sun or shade stomatal density via a systemic signalling system, with mature leaves sensing the irradiance level and transmitting (via an as yet unidentified mechanism) the signal to developing leaves. This plasticity of the leaf in changing stomatal characteristics is believed to be an adaptation to improve gas exchange capacity of the leaf in response to altered environmental conditions. Since stomata differentiation generally occurs during relatively early stages of leaf development, systemic signalling allows plants to sense the prevailing environmental conditions and to adjust developing leaves in preparation for the environment they are likely to be exposed to.
Studies on stomatal response in monocot grasses to altered environmental conditions are fewer in number (e.g., Miranda et al., 1981; Weyers and Lawson, 1997; Hubbart et al., 2013). One challenge in comparative studies of stomatal characteristics in grasses is the extent to which sampling stomata from one area of a leaf provides an accurate representation of stomatal properties for the whole leaf. In Arabidopsis the stomata are scattered across the leaf and not clearly zoned while in plants such as rice (used in this study) stomata occur orderly in specific cell rows on both leaf surfaces. This non-random distribution raises a number of questions, such as whether stomata at different position relative to the midrib show different properties. Previous studies of rice stomata have often revealed relatively high variation in parameters of size and density (Miranda et al., 1981; Weyers and Lawson, 1997) and the question arises of the extent to which this reflects real, endogenous variation or variation introduced due to the sampling procedure used (since it is virtually impossible to count all the stomata on one rice leaf). A major aim of the data presented in this chapter was to provide a comprehensive analysis of stomatal characteristics across the entire width of rice leaves at comparable developmental stages. In addition, the variable of HL or LL was introduced to investigate whether any observed response in terms of stomata size or density depended on the position across the leaf surface at which measurements were taken. These results provide the foundation for the experiments reported in Chapter 4 designed to investigate the link between stomatal structural response to altered light regime and the developmental stage of the leaf in which the response is occurring.
As indicated above, an alteration in stomatal size/density is expected to influence photosynthetic performance. Therefore, in addition to the analysis of stomatal patterning, the analysis of gas exchange in HL and LL leaves are reported in the second part of work reported in this chapter. The photosynthetic rate of sun leaves has been shown to be higher compared to shade leaves, for instance in the work on Alocasia macrorhiza by Sims and Pearcy (1992). Thicker sun leaves that possess more mesophyll cells increase the mesophyll surfaces occupied by chloroplasts, thus securing the area for CO2 dissolution and transport. In terms of photosynthetic machinery, sun leaves have been shown to regularly have higher electron transport rate (Yamori et al., 2010) and larger amounts of Calvin-cycle enzymes, especially Rubisco, compared with shade leaves (Timm et al., 2012). In contrast, shade leaves generally have higher levels of chlorophyll a/b-binding light harvesting complexes that are essential for them to thrive in low light conditions (Leong and Anderson, 1984). With respect to stomata, various plants, including Arabidopsis (Masle et al., 2005) and Leymus chinensis, a perennial grass (Xu and Zhou, 2008) show a positive correlation between stomatal density and conductance, thus providing a potential link of altered irradiance response (via stomatal density) and gas exchange.
Sun and shade leaves display distinct photosynthetic properties. These differences are best revealed by the analysis of light response curves (Fig. 3.2) which provides exemplar curves for sun and shade leaves. These curves reveal information on two important limitations at particular irradiance levels, namely the light and carboxylation-limited phases The light-limited region of the curve provides information on the light compensation point and the quantum yield (light compensation point is the irradiance point when CO2 uptake (from assimilation) equals CO2 evolution (from respiration) while the maximum quantum yield is obtained from the slope of the linear portion of the curve). The second phase of the curve, which is relatively flat, reflects the saturating light concentration at which photosynthesis is limited by the carboxylation rate.
Figure 3.2:
Typical response of photosynthesis to different light concentrations for C3 plant. Dotted lines are extrapolations of initial linear slopes in the light-limited phase of the curve, which also indicates quantum yield (moles of O2 evolved per mole quanta absorbed). The light-compensation point is the irradiance level required to offset respiration so that net exchange of CO2 is zero. The Pmax (light-saturated photosynthesis) of sun leaves is higher than shade leaves in the same species. Pmax is found in the plateau region of the curve, which reflects carboxylation limitation. Figure obtained at: Plants In Action, http://plantsinaction.science.uq.edu.au/edition1/?q=content/12-1-1-light-interception-and-utilisation
Being a C3 plant, rice photosynthesis under a light saturating intensity is essentially limited by one or the combination of three parameters, namely (i) the maximum carboxylation rate by Rubisco (Vcmax); (ii) the regeneration of RuBP via the maximum rate of electron transport (Jmax) and (iii) RuBP regeneration via triose phosphate utilization (TPU) (Farquhar et al., 1980 and Bernachi et al., 2009) (Fig. 3.2). These three conditions can be predicted using the C3 photosynthesis model developed by Farquhar et al. (1980) and that has been updated by Sharkey et al. (2007).
In the study reported here, modelling of rice photosynthesis was performed using an assimilation versus intercellular CO2 (A-Ci) curve fitting tool from Landflux.org which employs equations from Ethier and Livingston (2003) for theoretical (Vcmax) and (Jmax) estimation. The initial phase or slope of the A-Ci curve provides a measure of the activity of Rubisco in the leaf, as influenced by the amount present (Ac) (Fig. 3.3). Furthermore the curve also provides a means to distinguish between the mesophyll and stomatal limitation of the observed photosynthesis. The second phase of the curve reflects the limitations imposed by RuBP regeneration, which is connected to electron transport rate and ATP synthesis (Aj). The levelling off or decline in assimilation rate at higher CO2 concentrations can be attributable to the third phase of the A-Ci curve (which is not included in the curve fitting tool from Landflux.org), namely triose phosphate utilization limitation (TPU) (Fig.3.3). TPU happens when triose phosphates accumulate due to the increase in photosynthesis, causing chloroplasts to be starved of phosphates (Sharkey et al. 1986). TPU in fact is a collection of events, including the drop in phosphate level, decline in ATP level and, eventually, deactivation of Rubisco (Sharkey 1989, 1990). The A-Ci curves obtained in this study serve as a useful and non-destructive tool to quantify these biochemical components, as well as allowing a separation between stomatal and non-stomatal effects on CO2 assimilation in rice constantly growing in either HL or LL conditions.
Figure 3.3:
Modelled rate of net CO2 assimilation (A) in as a function of chloroplastic CO2 partial pressure (CC).
In general the limitations of photosynthesis in well-watered sun and shade leaves are known, thus this part of the study was performed to establish the techniques to enable the characterisation of the role of stomatal development in limiting rice photosynthesis in HL and LL conditions. The overall aim was to set the foundations for the experiments reported in the next chapter aimed at characterising the effect on final leaf physiology when plants were transferred between HL and LL at specific stages of leaf 5 development, i.e., if transfer between different irradiances at different stages of leaf development leads to different stomatal patterns, what is the outcome on the photosynthetic performance of the mature leaves?
Coupled with the above experiments, biochemical analysis of pigments was also performed to investigate the relationship of photon capture efficiency and the observed assimilation rates. Chlorophyll a/b ratio is an indicator of the light harvesting capacity of the photosynthetic apparatus and shade leaves generally possess much higher amounts of LHC-II (light-harvesting pigment protein of PSII) than sun leaves and, consequently, their a/b ratios are lower than in sun leaves (Lichtenthaler et al., 1982). Finally, the analysis of isotopic composition (13C/12C ratio) relative to the standard PeeDee Belemnite is used to compare the photosynthetic discrimination of the heavier carbon isotope 13C in HL and LL grown leaves. Generally, as CO2 assimilation increases or stomatal conductance decreases, intercellular CO2 decreases, resulting in decreased discrimination against 13C (Farquhar et al., 1989). These biochemical and isotope discrimination methods facilitate an understanding of the mechanism underpinning any measured change in photosynthetic performance.
3.2 Aims
1. Characterise the pattern of stomatal differentiation in rice leaves grown under different levels of irradiance.
2. Investigate the outcome of irradiance-induced altered stomatal patterning on rice leaf photosynthetic performance.
3.3 Brief methodology
All rice plants were grown hydroponically using growth chamber settings and nutrients solution as described in section 2.1. Plants were grown at HL or LL (section 2.1) until leaf no. 5 was fully expanded. For the HL condition, this was achieved between 21 - 23 days after sowing while for LL-grown plants it was achieved between 25 - 28 days after sowing. Rice was grown at the edge of a hydroponic container (Fig. 2.1) to minimize shading effect among plants. The middle portion of leaf no. 5 blade was used for all measurements.
3.3.1 Stomatal and epidermal measurements
Rice leaves were prepared and photomicrographs obtained using the methods described in sections 2.4, 2.5 and 2.6.1 as required. Measurements of SD, SCA and SPA and IG area were made following the methods described in section 2.6.2. For each parameter five measurements per leaf were taken and averaged to obtain one data point. Eleven replicate leaves were used in total.
3.3.2 Gas exchange measurements and biochemical analyses
Physiological and biochemical measurements were made on the middle section of 3 replicate fully expanded leaf no 5. Gas exchange measurements were carried out to obtain light response curves for both HL and LL-grown rice as described in section 2.3.1. Apparent quantum yield, maximum light saturated rate of photosynthesis and the light compensation point were measured as described in section 2.3.1. Gas exchange measurements were carried out to obtain A-Ci curves as described in section 2.3.1. PAR was maintained at 2000 µmol m–2s–1 with 10% blue light and 90% red light. Stomatal conductance versus intercellular CO2 concentration curves (gs-Ci), were also obtained. Intrinsic water use efficiency (iWUE) was computed by taking the ratio between assimilation and stomatal conductance at ambient (400 ppm) CO2 (Ca) as well as when iWUE was at its maximum (iWUEmax), which was always at the last point of A-Ci and gs-Ci curves. Pigment quantification and isotopic carbon discrimination determination were performed as described in sections 2.3.2 and 2.3.3 respectively using five replicate leaves for each analysis.
3.3.3 Statistical analysis
In order to compare means two-tailed pair wise t-tests, two sample t-tests or one-way ANOVA followed by Tukey’s HSD post-hoc test was carried out as appropriate. Pearson’s Correlation Analyses were performed on various HL and LL parameters. Statistical tests used are indicated in each figure legend. All calculations, statistical analyses and graphs were performed and prepared using Microsoft Excel 2016, Minitab 17 and GraphPad Prism 6. All Diagrams were prepared using Microsoft Power Point 2016.
3.4 Results
3.4.1 Comparison of stomatal size and density
The DIC image of a typical interveinal gap of a mature 5th rice leaf (abaxial surface) is presented in Fig. 3.4 A. Rice veins come in three orders where the midvein is central while the large (LV) and small veins (SV) lie parallel along the proximodistal axis of the leaf (Scarpella et al., 2003). Since LV is physically larger than SV (Smillie et al., 2012) it was easy to differentiate from the transverse section but from the aerial view (DIC images), LV could either have two strips of x-shaped silica bodies on it or a single but relatively wider x-shaped silica bodies strip (Fig 3.4 A).
Each IG comprises epidermal cells aligned in parallel to the veins to form a cell file. Some of these files contained only epidermal cells whereas others had at least one stoma. To further understand the top view from the DIC image, 2D transverse hand-section (Fig. 3.4 B) and 3D SEM images (Fig. 3.4 C) were obtained. Each IG from five leaves was analysed for stomatal counts and measurement after two treatments- plants grown under HL or low irradiance LL. It is worth to note that due to natural vein number variations, certain IG had n ≠ 5 for comparison and this is clearly seen in Fig. 3.5.
Note that all graphs in Fig. 3.5 lack comparison for IG17 and IG18 between the HL and LL treatments because HL leaves always produced fewer small veins than low light leaves, thus decreasing the total number of IGs. The number of large veins was similar between the two treatments (three) but HL leaves in general had fewer small veins (about sixteen) while LL leaves had eighteen IGs (Fig. 3.4 D). At the midpoint of the laminar leaf width for HL and LL leaves was approximately the same (HL = 0.5 ± 0.03 cm; LL = 0.47 ± 0.03 cm), and were not significantly different.
Figure 3.5 shows the variation in stomatal characteristics (SCA, SPA and SD) between the HL and LL treatments in each IG, starting from the mid-vein towards the leaf margin. Pair-wise comparison between HL and LL values for each stomatal measurement was made for each IG. Considering the HL and LL treatments, it is clear that for both SCA (Fig. 3.5 A) and SPA (Fig. 3.5 B) the LL treatment led to significantly smaller stomata at most IG positions across the leaf as illustrated in Fig. 3.6.
Figure 3.4: Vein number and position in a rice leaf. (A) is a DIC image of an interveinal gap (IG) between a large vein (LV) and a small vein (SV) of a control HL leaf where some of the cell files contain at least one stoma (S) or none (E), as indicated on the left hand side of the figure (Scale bar = 50µm). A transverse hand-section (B) shows relative position of the veins, S, E and bulliform cells (BULL). An SEM micrograph (C) further shows a leaf transverse section with the same labelled structures (D) Provides a schematic diagram to show the relative numbers and positions of veins and IGs for a HL leaf (13 SVs, upper diagram) and LL leaf (15 SVs, lower diagram). |
Figure 3.5: Variation of stomatal properties in each interveinal gap (IG) on the abaxial surface for high light (HL) and low light (LL) grown leaves. The graph shows the mean of (A) stomatal complex area (B) pore area and (C) density for each IG from HL leaves (green lines) and LL leaves (blue lines). Error bars represent standard deviation. Two-tailed pair-wise t-tests between HL and LL in each IG have been performed with significant differences between comparisons being indicated as: p<0.05, *p<0.01 and ***p<0.001 (n=5 for each comparison except where stated on the x-axis). |
Figure 3.6: Selected DIC images of stomatal complexes of a high light (HL) grown leaf at IG1 and low light (LL) grown leaf at IG10. Scale bar equals 10µm. |
However there was a tendency for the mean SCA, SPA and SD values to be slightly higher (or more variable) in IG1 and IGs 14 -17 i.e. at the leaf margins compared to the middle IGs where values were more consistent for both HL and LL leaves. For example, mean SPA values in LL grown leaves were highest in IG1 (about 63µm2) with the lowest value being observed in the last IG18 (about 39µm2). In HL grown leaves SPA had an extremely low value (about 64µm2) in IG12 while SPA was highest (about 80µm2) in the last three IGs (IG 14-16). Thus the middle IGs were least variable part of the leaf.
Comparison of SD between HL and LL leaves (Fig. 3.5 C) did not yield any significant differences between the IGs tested except in IG16 (the last IG for HL) where HL leaves had a significantly higher value of stomatal density compared to other IGs in HL leaves. A relatively higher SD was also observed in the last IG in LL leaves although not as high as in HL leaves. IG areas became progressively smaller towards the leaf margin, yet stomata number per IG did not greatly decrease, resulting in higher frequency per given area.
Table 3.1 shows Pearson’s correlation coefficients that were computed to assess the relationship between stomatal properties in HL and LL leaves. Overall, there was a significant positive relationship between stomatal complex and pore areas in all IGs. Increases in SCA were positively correlated with increases in SPA. Stomatal density showed moderate negative correlation (r= -0.45) between HL and LL leaves.
Table 3.1:
Pearson’s correlation coefficients (r) between the mean values of stomatal properties from Fig. 3.6 up to IG16. Single asterisks (*) indicate correlations which are significant at the p<0.05 confidence limit. HL (high light); LL (low light); SCA (stomatal complex area); SPA (stomatal pore area); SD (stomatal density).
Stomatal properties | HL-SCA | LL-SCA | HL-SPA | LL-SPA | HL-SD |
LL-SCA | *0.59 |
|
|
|
|
HL-SPA | *0.58 | 0.25 |
|
|
|
LL-SPA | 0.43 | *0.80 | 0.30 |
|
|
HL-SD | 0.09 | 0.16 | 0.37 | 0.30 |
|
LL-SD | 0.33 | 0.14 | 0.01 | 0.21 | *-0.45 |
Even though LV number was always three (at least for leaf 5 in this study), SV number (and hence IG number) varied even for the same treatment. This complicates comparison between leaves under different treatments. Thus an alternative approach was taken in order to standardize comparison between the leaves and to see whether fewer sample points could be taken for future leaf comparisons without compromising the analysis, thus decreasing the workload/time required for the analysis of stomatal properties after transfer of leaves from one light environment to another at different stages of leaf development.
In this approach, the IGs starting from MV to the margin were divided into 7 sections as shown in Fig 3.7: (i) MV- IG immediately next to the midvein; (ii) SV1- Middle IG in between MV and first LV; (iii) LV1ab- The average of two IGs bordering with the first LV; (iv) SV2- Middle IG in between the first and second LV; (v) LV2ab- The average of 2 IGs bordering with the second LV; (v) SV3- Middle IG in between the second and third LV and (vii) LV3ab- the average of two IGs bordering with the third LV (Fig. 3.7 A). When the number of IGs did not halve evenly when divided for the middle IG in between MV and LV or LV and LV, the approximate middle IG taken was always the one closer to the MV (Fig. 3.7 B).
Using this approach, it is clear in Fig. 3.7 C and 3.7 D that HL treated leaves had significantly bigger stomatal complex and pore areas (except for the pore area in MV) compared to LL leaves. When values for SCA and SPA were compared within each treatment (HL or LL), there was no significant difference for HL treated leaves with respect to position across the leaf. For LL treated leaves, comparison of SCA indicated some significant differences, with the MV and SV3 positions having higher and lower SCA, respectively. In addition, SPA also showed the same trend as SCA, with SV2 and SV3 positions having significantly SPA than the MV position.
On the other hand when stomatal density data were analysed using this approach (Fig. 3.7 E), although there was a tendency for a higher stomatal density in the HL treated leaves, the large variation in values means that the two data sets cannot be distinguished.
Figure 3.7:
Variation in stomatal properties across leaves from midvein towards margin in seven sections of selected IGs. (A) Schematic diagram of a leaf to show regions taken for average measurements. Green bands represent IGs and black band is the leaf margin. The middle IG is selected as the position between MV-LV or LV-LV and values from the two IGs bordering an LV are averaged and denoted as ‘ab’. When the number of IGs are odd (B), the middle IG taken is always the one closer to the MV. The graphs show the mean values for (C) stomatal complex area (D) pore area and (E) stomatal density, where green lines show values for HL while blue lines are for LL samples. Two-tailed pair-wise t-tests between HL and LL values in each section have been performed with significant differences between comparisons being indicated: p<0.05, *p<0.01 and ***p<0.001 (n=5). Within each treatment, one-way ANOVA followed by Tukey-Kramer′s post-hoc test is performed. Means not sharing the same letter are significantly different (p<0.05). Error bars represent standard deviation. Key: IG, interveinal gap); LV, large vein; MV, midvein and SV, small vein.
3.4.2 Comparison of assimilation rate and stomatal conductance in HL and LL grown plants
To investigate the extent to which HL and LL treatments altered the light-saturated rate of photosynthesis, apparent quantum yield and light compensation point a light versus assimilation rate response curve was performed (Fig 3.8).
Figure 3.8:Assimilation (A)-light (PAR) response curves of rice leaf no. 5 grown in either a high light (red line) or low light (blue line) environment in Error bars represent standard error of mean for the assimilation where n=3. |
Leaves grown in either HL or LL had a saturated assimilation rate from an irradiance of 1600 µmol m-2 s-1 (Fig. 3.8). Thus, for the all the subsequent A-Ci experiments PAR was set to 2000 µmol m-2 s-1. This was important to eliminate photosynthetic limitation caused by insufficient light quantity as well as to activate Rubisco while studying A-Ci response curves.
The light (and CO2) saturated rate of photosynthesis was 52% higher in HL grown leaves (43.9 ± 2.4 µmol m-2 s-1) compared to LL leaves (28.8 ± 1.4 µmol m-2 s-1). LL-grown leaves had a slightly higher apparent quantum yield of 0.1 ± 0.007 compared to HL leaves ( 0.07 ± 0.0002 ± mol/mol) but there was no difference in LCP. The later is unusual but may be due to the fact that saturating CO2 was used in this experiment.
Assimilation versus intercellular CO2 (A-Ci) response curves for HL and LL-grown plants are shown in Fig 3.9 A and B respectively. In the first phase of the curve, the measured data fitted well with the predicted Ac model (blue dashed lines) for HL leaves (Fig. 3.10 A) but underestimated for LL grown leaves. For the second part of the curve the Aj model (green dashed lines) fitted HL measurements reasonably well but overestimated values for the data points for LL grown leaves. The more parametre values (such as leaf absorptance, mesophyll conductance and day respiration) that could be altered to specifically suit rice, the better curve fitting and prediction for both Ac and Aj models could be.
Figure 3.9: Assimilation versus intercellular CO2 (A-Ci) response curves for high light (HL, A) and low light (LL, B) grown leaves. Data are pooled from 3 plants for each treatment where each red dot is the average observation for the given A rate and Ci concentration with measurements made at the current CO2 concentration (Ca= 400ppm) marked on the curves. Ac (dashed blue lines) are CO2 assimilation rate limited by the amount and activity of Rubisco (enzyme limited/RuBP saturated) while Aj (dashed brown lines) are CO2 assimilation rate limited by RuBP regeneration (light limited/RuBP limited). Error bars show standard error of mean. |
HL leaves had a significantly higher assimilation rate compared to LL leaves at both saturating (1600 ppm) and ambient (400 ppm) CO2 (Fig 3.9). At ambient CO2 HL grown leaves had an assimilation rate of 31 µmol CO2 m-2 s-1 which was 22% higher than that of LL grown leaves (Fig 3.10 A). This difference was attributable to a significantly higher Vcmax and Jmax in HL grown leaves compared to LL grown leaves (Fig. 3.10 B and C respectively) suggesting that HL grown leaves had higher amounts of Rubisco for carboxylation.
HL leaves had a significantly lower stomatal conductance (0.34 mol H₂O m-2 s-1) than LL leaves (0.8 mol H₂O m-2 s-1) indicating that they were transpiring at a slower rate than LL leaves (Fig 3.10 D). This is consistent with a higher water use efficiency in HL compared to LL leaves (Fig 3.11 A and B). The leaf temperature of HL and LL leaves was not significantly different at ambient CO2 but was slightly higher in HL leaves at saturating CO2 (Fig 3.10 E). The stomatal limitation to photosynthesis (ls) was also significantly higher in HL than LL leaves (Fig. 3.10 F).
Pigment analysis revealed that HL grown leaves had higher chlorophyll a/b ratio compared to LL grown leaves (Fig. 3.10 G), indicating the presence of a larger number of PSII reaction centres and less light harvesting complexes consistent with the higher Jmax in HL leaves (Fig. 3.10 C). Discrimination against the heavier carbon isotope (13C) was also assessed to validate and understand the relatively higher A400 (also higher in iWUE) rate in HL leaves than LL leaves that could be due to stomatal pore (entry) or/and enzymatic preference (Rubisco). LL leaves clearly had a significantly more negative, thus lower, 13C/12C ratio (about -33.5‰) than HL leaves (-32.5‰) (Fig. 3.11 B), suggesting the effect of high stomatal conductance and/or increased Rubisco discrimination against 13C.
Figure 3.10:
Photosynthetic features of high light (HL) and low light (LL) grown rice leaves at ambient CO2 (400 ppm) extracted or calculated from the A-Ci curves in Fig. 3.10. (A) Assimilation rate (A400); (B) Rubisco carboxylation rate (Vcmax400); (C) electron transport rate (Jmax400); D stomatal conductance (gs400); (E) leaf temperature at 400 and 1600 ppm CO2; (F) percentage stomatal limitation (ls) to photosynthesis; and (G) chlorophyll a/b ratio. Two samples t-tests were performed to compare means with significant differences between comparisons indicated as: p<0.05; *p< 0.01 and *** p<0.001 (n=3) with error bars to represent standard error of mean.
Figure 3.11: Water use efficiency (A) and carbon isotope ratio (δ13C) is presented on a per mill (‰) basis (B). Two samples t-tests were performed to compare means. Significant differences between comparisons are indicated as ***p<0.001 (n=3). Error bars indicate standard error of mean. |
3.5 Discussion
Results from the first section of this study showed that HL grown rice consistently produce leaf 5 with larger stomatal dimensions (complex and pore areas) across the entire leaf width (Fig. 3.5 A-D). This finding agrees with Hubbart et al. (2012) who also reported significantly bigger (longer) stomata on average on the abaxial surface of rice grown under HL conditions but a nonlinear change in stomatal size from the midrib outward. This is probably because they used a different method of sampling where stomata were chosen at four different distances from the midrib compared to the more detailed approach used here using individual IGs. The results from this study suggests that stomatal size measured in any IGs is in general not statistically different from any other, though care must be taken at both the extreme edges of the leaf and in the IGs immediately adjacent to the midvein.
In a related plant, Muhlenbergia cuspidate, which is a perennial C4 grass (Steuter, 1987), individual stomatal characteristics that define size (such as pore length and width) did not necessarily change proportionately under HL or LL. Smith and Martin (1987) reported that when this grass was grown under HL or LL conditions abaxial stomatal length did not change but stomatal width did alter significantly. Meanwhile in C3 dicot plants, for example Eucalyptus globulus and tobacco (Nicotiana tabacum), high and low irradiance levels have a clear opposite effect. Thus James and Bell (2000) showed that HL conditions experienced by E. globulus resulted in longer stomatal pores compared to LL conditions, but in tobacco there was no significant difference in stomatal pore length between the two light conditions (Thomas et al., 2003). The varying results from different plant groups suggest that the effect of growth in high or low irradiance conditions on stomatal size is species-specific. Stomata, which are vital in photosynthetic gas exchange, are one of the components that plants can modify in order to adapt in a given light condition. Certain plants which lack the ability to alter stomatal size may employ other ways, such as modifying stomatal density and leaf thickness, so that leaf physiological performance can be optimized for the prevailing light conditions during leaf development.
In this study, stomatal density in rice generally did not vary between HL and LL conditions in any of the IG analysed (except in the last IG of HL leaves where stomatal density was significantly higher compared to the same IG position of LL (Fig. 3.5 C). This is one of the noticeable outcomes resulting from comparing stomata ‘IG to IG’ between different treatments since different light conditions result in different IG number (Fig. 3.4 D). For example, the last IG of HL (IG 16) corresponds to IG16 of the LL that actually is the 3rd last IG for the LL leaf. This partially equivalent IG comparison could be a source for discrepancies for data analysis. Therefore, a simpler system is proposed by basing the comparison on IG’s relative distance to the midvein or large vein (Fig. 3.7 A and 3.7 B). It is a more standardized way of picking IGs to be used for stomatal counting and measurement. This method still shows that there is a clear difference in stomatal dimensions between HL and LL (Fig. 3.7 C and 3.7 D). On average, there is a tendency for HL leaves to have higher stomatal density in the selected IGs compared to LL across the leaf but this is statistically not significant (Fig. 3.7 E). IG gap width and epidermal cell file number (other than stomata) vary widely across the leaf and seem not to follow any particular pattern in response to the light conditions given.
The system reported here for selection of the leaf region for comparative analysis of stomatal properties in rice could be useful in other grass species in order to study variation in stomatal characteristics since grass leaves are normally partitioned into IGs and orderly arranged in cell files. Additionally it might also work in other non-grass species, such as Commelina communis, which have parallel venation and heterogeneous stomatal characteristics of density and size across the leaf (Smith et al., 1989).
When all stomatal characteristics in this study are taken together, only a few could be used as a proxy for another. For example, correlation analysis suggested that in the same HL or LL leaf stomatal complex area has a significant positive relationship with pore area (Table 3.1), thus measurement of one characteristic is just as good to predict the other. Bearing in mind the way in which stomata are structured in grasses, this is not surprising. Stomatal density did not correlate clearly with any size parameter, which is slightly surprising bearing in mind the proposed linkage of stomatal size and density (Franks and Beerling, 2009), although this study examined stomatal size over a much larger range of species.
The analysis shows that stomatal density in HL leaves is negatively correlated with stomatal density in LL leaves. This suggests when stomatal density values of either HL or LL leaf are pooled from all IGs and compared with each other, HL and LL leaves do have high and low stomatal density respectively. Thus, the results agree to some extent with those reported by Coupe et al. (2005) using Arabidopsis, but contrast with Hubbart et al. (2012) who also analysed rice. Clearly the sampling and analytical process chosen can influence the conclusions drawn from analysis of stomata.
The plasticity of rice stomatal size in response to either HL or LL conditions raised a question regarding its effect on leaf gas exchange capacity. Measurement of stomatal conductance to water vapour under ambient CO2 concentration (gs400) which is expressed in mol H2O m-2 s-1 was employed to answer this. It is clear from Figure 3.10 D that LL leaves, despite having relatively smaller stomata (Fig. 3.6), still have the ability to regulate aperture opening more than HL leaves thus resulting in higher gs values. The results are surprising because many investigations using rice, as well as different plants, show that sun leaves have higher gs values than shade leaves or show no difference. Experiments using the same rice variety IR64 by Narawatthana (2013) showed that at ambient CO2, the gs of sun and shade leaves did not differ. In other plant species such as English oak tree (Quercus robur, Gross et al., 2008), Ligustrum vulgare (Guidi et al., 2008) and rice plants of different variety (Oryza sativa cv. F50, Restrepo and Garces, 2013), the sun leaves always had higher gs than shade leaves. Many factors could contribute to this, for instance Narawatthana (2013) during measurement used a light intensity of 1000 µmol m-² and Restrepo and Garces (2013) used 800 µmol m-², but in this study it was 2000 µmol m-². In other species, like English oak and Ligustrum vulgare, direct comparison is fundamentally not equivalent because these two species are dicot plants that have morphologically different stomata. Therefore, the difference in gs values might be attributable to the mechanics of different stomatal type. Nevertheless, high gs observed LL leaves is a common protective measure in many shade tolerant species to prevent damaging temperature at the expense of water use efficiency (Chazdon and Pearcy, 1991; Schymanski et al., 2013).
Equation 1.1 in Chapter 1 is useful to explain mechanistically how higher gs in a LL grown leaf can be achieved, which is through its smaller pore area and shallower pore depth that leading to a shorter diffusion path distance. This gs formula was not used in this study because it is only valid when the measuring conditions that promote maximal stomatal opening (high light, high relative humidity and very low CO2 concentration) are all in place. Small stomata also create more room for extra stomata (Franks and Beerling, 2009) allowing high gs to be attained, i.e., higher SD. Based on this formula, essentially three stomatal properties control gs, namely stomatal density, stomatal pore area and stomatal pore depth. By this definition, since stomatal density was not different between the two treatments, HL leaves should have higher gs than LL leaves simply because they had larger pore area. However the opposite was measured here where LL leaves had a much higher gs than HL leaves. To explain this, it is important to note that the stomatal pore area formula (Equation 2.1) in this study is the maximum area possible for the given aperture length and guard cells width. It may be that there is a mechanism that enables the stomatal aperture to open differentially, thus affecting the gs value even when stomatal density does not change.
To put things into perspective, we can compare the theoretical gs values between the treatments using the equation 1.1. By pooling and averaging all the data from all IGs in each treatment, stomatal pore area for HL leaves is 76 µm² and for LL leaves is 52 µm². Stomatal density for both treatments is 235 mm-² and pore depth is assumed to be the same as guard cell width (taken after Dow et al., 2013), which is approximately 5 µm. By substituting these values into the Equation 1.1, the maximum theoretical values gs for HL and LL leaves are 4.0 and 3.0 mol H₂O m-²s-¹ respectively. From Fig. 3.10 D, the measured or operational gs values for HL and LL leaves under ambient oxygen and CO2 (Ca = 400ppm) conditions are about 0.3 and 0.8 mol H₂O m-² s-¹ respectively. This agrees with a number of authors (Williams et al., 2004; Dow et al., 2013) that predicted conductances are often much higher than observation. For HL leaves, the measured gs is only about 20% of the calculated gs. This means, when stomatal density and pore depth are not variable, the pore area was only 8% open, thus accounting for the low measured gs. For LL leaves, the measured gs is about 27% of the calculated gs meaning the pore area was about 26% open, thus resulting in the higher measured gs than HL leaves. This differential stomatal opening makes it possible for leaves to have variable gs values even when other stomatal parameters stay the same. Another possible reason why LL leaves had high gs was due to the shorter distance to travel for water within the leaf as a result of altered vein number and spacing. The higher SV number in LL leaves (Fig. 3.4 D) causes the individual IGs to become narrower. More veins and narrower IGs mean that the leaves have more ‘pipes’ that provide water and narrow IGs means a smaller tissue area that needs to be filled with water, potentially allowing a greater rate of flow of water to the substomatal chambers.
But why would LL leaves need to transpire a lot of water relative to HL leaves? Since LL leaves have been shown to have relatively lower Vcmax and Jmax (Fig. 3.10 B and C respectively), there is no pressing demand to have more CO2 into the chloroplasts to satisfy high assimilation rate. Moreover, LL leaves also have relatively lower stomatal limitation (Fig. 3.10 F) thus there is no real need to open the stomata widely just to lose water unnecessarily. One possible explanation is that it is a mechanism employed by LL leaves to prevent damage to the leaves. LL ‘shade’ grown leaves, have several distinctive characteristics such as thin leaves and low total carotenoid content (Narawatthana, 2013). Such an anatomy and physiology are unsuitable for high light conditions. However, when the gas exchange measurements were being made, the amount of light used throughout the experiment was 2000 µmol m-²s-1. For LL plants (which were grown in 250 µmol m-²s-1 irradiance), this means the leaves were receiving an excess of light that might result in heat build-up in the leaves. If measuring photosynthesis carried out in the growth environment, the extremely high gs may not occur.
The excess light condition was more intense in LL than HL leaves due to thinness of the leaves and the low level of total carotenoids (for photoprotection, Yang et al., 2002). The fate of electrons from the excited chlorophyll molecules in PSII can be used to drive photosynthesis, re-emitted as fluorescence or re-emitted as heat (Murchie and Lawson, 2013), and since Jmax is relatively lower in LL leaves thus the other options left are fluorescence and heat re-emittances for excess energy dissipation not used for photochemistry.
Biochemical analyses were also made to validate certain physiological parameters such as Vcmax, Jmax and iWUE. Since HL leaves must deal with higher irradiance level than LL leaves, it is expected for them to have a high chlorophyll a/b ratio (Kitajima and Hogan, 2003). It is often generalized that lower chlorophyll a/b ratio in LL grown leaves is due to higher chlorophyll-b content in the LHCII (McDonald, 2003; Lichtenthaler and Babani 2004). This shows that for LL acclimated rice larger investment is necessary into making bigger LHCII antenna size increases in to capture more photons to drive photochemistry in PSII. Since relatively lower Vcmax is commonly correlated with lower Rubisco content (Adachi et al., 2014), this implies lower overall energy consumption needs for carboxylation thus reflected in relatively lower Jmax as well. In terms of electron usage, LL leaves are more efficient than HL leaves since lower Jmax means greater CO2 fixation per electron can be achieved. The simple A/gs ratio to obtain iWUE has shown that HL leaves are more efficient than LL leaves (Fig. 3.11 A). This means CO2 at the site of carboxylation experience fractionation which should be high in HL leaves with relatively higher assimilation rate. This is validated through a more positive value of isotopic carbon discrimination (δ13C) in HL leaves compared to LL leaves. The abundance of intercellular CO2 (also means relatively higher Ci/Ca ratio) coupled with the nature of LL leaves that are known to have a relatively lower Rubisco content (Narawatthana, 2013), thus reduced carbon fixation which allows the Rubisco to discriminate more against the heavier 13C isotope. The more positive carbon isotope ratio in HL leaves is thus indicative of a higher water use efficiency (Franks et al., 2015) because a relatively small amount of water is transpired for the high assimilation rates measured compared to LL leaves.
In this section I have shown that the size of stomata in rice leaves is responsive to the light environment. Stomatal plasticity in rice, particularly dimension, has been shown to respond to light where high and low irradiances produce small and big stomata respectively. Stomatal density is not statistically different in either light condition. A detailed analysis of stomatal properties in each interveinal gap showed that they did not vary significantly, although care has to be taken when considering stomata measured in the interveinal gaps toward the leaf margin. An analysis based on this leaf area alone might lead to spurious conclusions.
Due to the nature of HL and LL leaves that produce inconsistent numbers of interveinal gaps, a simplified method is proposed so that stomatal characteristics across a leaf width can be quantified in a standardized manner. This approach will be used in the subsequent studies of this thesis (Chapter 4) that require analysing stomatal characteristics. The differences in stomatal size in HL and LL leaves are translated into a clearly different gs and assimilation patterns at different CO2 concentrations. The relatively high stomatal limitation in HL leaves suggest that big stomata may co-limit assimilation with non-stomatal properties (such as Rubisco and mesophyll conductance) thus can be a point to further improve general photosynthesis capacity without compromising water use efficiency. Predicted gs values are often much higher than the measured gs values, thus it is risky employ them unless under certain environmental conditions that promote maximum stomatal aperture opening. Small stomata in LL leaves have higher gs values than the larger stomata in HL leaves, suggesting a mechanism of differential stomatal opening to promote cooling through high transpiration in response to heat build-up in LL leaves which occurs to dissipate excess energy from the high intensity light in the measuring chamber.
Attribution 4.0 International — CC BY 4.0 - Creative Commons
Comments