Enhancement of leaf photosynthetic capacity through increased stomatal density in Arabidopsis


Author for correspondence:

Ikuko Hara-Nishimura

Tel: +81 75 753 4142

Email: ihnishi@gr.bot.kyoto-u.ac.jp


  • Photosynthetic rate is determined by CO2 fixation and CO2 entry into the plant through pores in the leaf epidermis called stomata. However, the effect of increased stomatal density on photosynthetic rate remains unclear. This work investigated the effect of alteration of stomatal density on leaf photosynthetic capacity in Arabidopsis thaliana.
  • Stomatal density was modulated by overexpressing or silencing STOMAGEN, a positive regulator of stomatal development. Leaf photosynthetic capacity and plant growth were examined in transgenic plants.
  • Increased stomatal density in STOMAGEN-overexpressing plants enhanced the photosynthetic rate by 30% compared to wild-type plants. Transgenic plants showed increased stomatal conductance under ambient CO2 conditions and did not show alterations in the maximum rate of carboxylation, indicating that the enhancement of photosynthetic rate was caused by gas diffusion changes. A leaf photosynthesis-intercellular CO2 concentration response curve showed that photosynthetic rate was increased under high CO2 conditions in association with increased stomatal density. STOMAGEN overexpression did not alter whole plant biomass, whereas its silencing caused biomass reduction.
  • Our results indicate that increased stomatal density enhanced leaf photosynthetic capacity by modulating gas diffusion. Stomatal density may be a target trait for plant engineering to improve photosynthetic capacity.


The process of photosynthesis in leaves is fundamental to biomass production and the global carbon cycle (Field et al., 1998). In the face of global warming and rising CO2 concentration, the manipulation of leaf CO2 assimilation rate could be useful for plant science in general and for crop production in the future (Long et al., 2006). Much attention has been paid to the manipulation of leaf photosynthetic capacity in the past several decades (Sheehy et al., 2008). The biochemical processes associated with photosynthesis, including the carboxylation reaction catalyzed by RuBPCase (Rubisco), have been important targets of these approaches. The genetic modification of the carboxylation process, however, has not consistently led to the enhancement of photosynthesis (Suzuki et al., 2007; Ishikawa et al., 2011; Zurbriggen et al., 2012).

Along with the biochemical processes associated with photosynthesis, gas diffusion through stomata is an attractive target for engineering photosynthetic capacity because it is a well-characterized process in terms of physiology (Farquhar & Sharkey, 1982), physics (Parlange & Waggoner, 1970), and multicellular patterning of the epidermis (Lau & Bergmann, 2012). Stomata are valves formed by a pair of guard cells that control the exchange of gases between the leaf and the atmosphere. Stomatal density, which is defined as the number of stomata per unit leaf area, was theoretically shown to be correlated with stomatal conductance (gs) (Franks & Beerling, 2009). Indeed, previous reports using Arabidopsis thaliana mutants with altered stomatal density showed that stomatal density variation affects gas exchange (Schluter et al., 2003; Yoo et al., 2010; Doheny-Adams et al., 2012). However, it is not known whether stomatal density is correlated with the leaf CO2 assimilation rate (A) through regulation of the gas exchange. Schluter et al. (2003) and Bussis et al. (2006) reported that the increased stomatal density in the Arabidopsis mutant STOMATAL DENSITY AND DISTRIBUTION (sdd1) had no significant effect on A under constant growth conditions. Bussis et al. (2006) described that increased stomatal density was compensated by reduced stomatal aperture, which is controlled to maintain a constant ratio of internal CO2 concentration (Ci) to ambient CO2 concentration (Ca) in plants (Santrucek & Sage, 1996; Bussis et al., 2006). To our knowledge, there are no reports showing a positive effect of stomatal density on photosynthetic rate in plants maintained under constant growth conditions.

We previously reported a correlation between stomatal density and A in Glycine max varieties having different genetic backgrounds (Tanaka et al., 2008). However, this report was not sufficient to conclude that A is influenced by stomatal density alteration, because of a comparison of cultivars with different genetic backgrounds. To investigate the effect of stomatal density on photosynthesis, an ideal experimental tool capable of providing a wide range of stomatal densities in the same genetic background is necessary. STOMAGEN/EPFL9 (At4 g12970) encodes a precursor of the cysteine-rich peptide stomagen that functions as a cell-to-cell signalling factor for stomatal differentiation in Arabidopsis (Hunt et al., 2010; Kondo et al., 2010; Sugano et al., 2010). The STOMAGEN expression level is positively correlated with stomatal density. The remarkable feature of STOMAGEN is that it is possible to modulate stomatal density as desired by overexpressing or silencing the gene (Hunt et al., 2010; Sugano et al., 2010). Alternatively, stomatal density can be increased by dipping plants into a solution containing stomagen peptide (Kondo et al., 2010; Sugano et al., 2010). This quantitative control of stomatal density by the STOMAGEN gene and the stomagen peptide provides a powerful tool for modulating gas exchange in the leaf. In the present study, we aimed to demonstrate the direct effect of alterations in stomatal density on leaf photosynthesis and related traits in Arabidopsis using STOMAGEN.

Materials and Methods

Plant materials and growth conditions

Arabidopsis thaliana (L.) Heynh, Columbia-0 (CS60000) was used as the wild-type (WT). The generation of STOMAGEN (At4 g12970) overexpressing (ST-OX) and amiRNA mediated silencing (ST-RNAi) lines was described previously (Sugano et al., 2010). T3 generations were used for Fig. 1. We selected representatives of the lines (STOMAGEN-OX10 and STOMAGEN-RNAi10; Sugano et al., 2010) and established homozygous lines of these lines, which were designated as ST-OX10-3 and ST-RNAi10-3, respectively. T4 generations of ST-OX10-3 and ST-RNAi10-3 plants were used for physiological experiments in Figs 2-6 and Supporting Information figures. Plants were initially grown on Murashige and Skoog (MS) medium plates (0.5×MS salts, 1% sucrose, 0.5% MES-KOH, pH 5.7 and 0.5% Gellan gum) and transplanted to each experimental condition. The growth conditions in MS plates were 22°C and a light intensity (photosynthetic photon flux density, PPFD) of 100 μml m−2 s−1. CO2 concentration was set to 380 ppm.

Figure 1.

Stomatal density, size, and water loss in the leaves of STOMAGEN transgenic lines and in leaves exogenously treated with stomagen peptide in Arabidopsis thaliana. Abaxial leaf epidermis (a), stomatal density (b), guard cell length (c), stomatal index (d) and water loss (e) of wild-type (WT), STOMAGEN-overexpressing (ST-OX), and -silencing (ST-RNAi) lines. Data from A. thaliana leaves exogenously treated with 10 μM stomagen peptide are shown as WT (10 μM). Stomatal density in the primary leaf was determined for each leaf after measurement of water loss. At least five plants were used for the experiment. Bars, SE (n = 4). ST-OX10-3, ST-RNAi10-3 and WT were used for Figs 2–6. Values marked with different letters are significantly different among genotypes/treatment at P < 0.05.

Figure 2.

Whole plant transpiration and water use efficiency of WT, ST-OX and ST-RNAi in A. thaliana. (a) Daily transpiration rate per unit leaf area averaged for 6 successive days. (b) Water use efficiency (WUE) as determined by the sum of daily transpiration and biomass production over 6 d. Plants were grown under a light intensity of PPFD = 100 μmol m−2 s−1. Two replicates were included for all the trials. Bars, SE (n = 2). Values marked with different letters are significantly different among genotypes at P < 0.05. ns, not significant.

Figure 3.

Stomatal density has a positive effect on A. thaliana leaf photosynthetic rate under the light-saturated condition. (a) Stomatal density of the abaxial epidermis of mature leaves of WT, ST-OX, and ST-RNAi. Stomatal conductance (gs) (b), leaf CO2 assimilation rate (A) (c) and intracellular CO2 concentration (Ci) (d) of WT, ST-OX, and ST-RNAi plants measured using a GFS-3000 gas exchange system under the light-saturated condition (PPFD = 1200 μmol m−2 s−1). Bars, SE (n = 4). Values marked with different letters are significantly different among genotypes at P < 0.05.

Figure 4.

Photosynthetic rates of mature A. thaliana leaves of WT, ST-OX and ST-RNAi in relation to light intensity. (a) Photosynthetic response to light intensity was measured using a GFS-3000 gas exchange system. Differences in A between WT and ST-OX were significant under light conditions PPFD = 300 μmol m−2 s−1 and above (P < 0.05). Differences in A between WT and ST-RNAi were not significant under any light conditions. (b) Response of gs to light intensity. Differences in gs between WT and ST-RNAi were not significant under any light conditions. Differences in gs between WT and ST-OX were significant under all the light conditions (P < 0.05). Bars indicate SE (n = 4).

Figure 5.

A-Ci curve in mature A. thaliana leaves of WT, ST-OX and ST-RNAi. (a) Photosynthetic response to different Ci values was measured using a GFS-3000 gas exchange system. The response curves were plotted in the full range. (b) The response curve in the relatively low range of Ci is magnified, and the maximal rate of carboxylation (Vcmax, average ± SE) was calculated for each plant. Bars, SE (n = 4). Differences in A between WT and ST-OX were significant at 300, 500 and 750 ppm (P < 0.05). Differences in A between WT and ST-RNAi were not significant under any of the CO2 conditions used. Note that Vcmax was slightly enhanced in ST-OX, but there was no significance between the three lines..

Figure 6.

Effects of alteration of stomatal density on biomass production in A. thaliana. Shoot biomass production was measured in plants grown for 32 d. At least 20 plants were used for the experiment. Plants were grown under the light intensity of PPFD = 300 μmol m−2 s−1. Bars, SE (n > 20). Values marked with different letters are significantly different among genotypes at P < 0.05.

Leaf excision experiment

The gravimetric measurement of water loss after leaf excision is a rapid method to evaluate transpiration rate (Mustilli et al., 2002; Merlot et al., 2007). Two independent lines (T3 or T4) for ST-OX, ST-RNAi and the WT, and at least five plants for each line were used for the experiment. At 7 d after germination (DAG), WT whole plants were dipped in liquid medium (0.5×MS salts, 1% sucrose and 0.5% MES-KOH, pH 5.7) containing 10 μM stomagen peptide. Chemically synthesized stomagen was refolded using a method described previously (Sugano et al., 2010). A sample of 10 ml of refolded stomagen peptide (100 μM) was dialysed three times in 1 l MS liquid medium (0.5×MS salts, 1% sucrose and 0.5% MES-KOH, pH 5.7) before usage. Whole plants of ST-OX and ST-RNAi lines were also dipped in the liquid medium without stomagen peptide to achieve uniform growth conditions. After dipping for 2 d (9 DAG), plants were re-transplanted to the MS plates. At 17 DAG, the primary leaves were excised for each line (n = 4) and immediately placed on an analytical balance (XS105 DualRange, Mettler Toledo Japan, Tokyo, Japan). The initial fresh weight (FW) and the weight after 5 min (W) were recorded. Water loss after 5 min was calculated as (FW–W)/FW × 100 (%). After the water loss measurement, stomatal density in the same leaves was determined following the method of (Sugano et al., 2010) with minor modifications. Leaves were fixed in 70% ethanol overnight and cleared in chloral hydrate solution (chloral hydrate: water: glycerol = 8:2:1, w/v/w) for over 9 h. The chloral hydrate was then removed and stomata were stained with safranin-O (200 μg ml−1). Digital images of the abaxial side of leaves were obtained using an optical microscope (Axioskop 2 plus system, Zeiss) equipped with a CCD camera (model VB-7010, Keyence, Osaka, Japan) under ×200 magnification. Four square areas (a total of 0.5 mm−2) were analysed in each leaf for the quantification of stomatal density, epidermal cell density and guard cell length. Stomatal index (SI) was determined as the ratio of stomata to stomata plus other epidermal cells.

Photosynthetic measurements

Representative ST-OX10-3, ST-RNAi10-3 (T4) and WT plants at 23 DAG were transplanted to Jiffy 7 peat pellets surrounded by vermiculite. Mature leaves of 9-wk-old plants were used for measuring photosynthesis with the GFS-3000 gas exchange system (Walz, Effeltrich, Germany). Measurements were performed on four plants per line at 22°C and a relative humidity of 60%. The light-saturated photosynthetic rate was determined at PPFD = 1200 μmol m−2 s−1 with [CO2] = 380 ppm. To generate a light response curve, photosynthetic measurements were conducted at PPFD intensities of 1200, 800, 600, 400, 300, 200, 150, 100, 50 and 0 μmol m−2 s−1. A CO2 response curve was generated by measuring the photosynthetic rate at [CO2] of 2000, 1000, 750, 500, 300, 200 and 100 ppm. A–Ci curves were analysed using the equations of Farquhar et al. (1980) with a linear two-segment model as described (Long & Bernacchi, 2003);

display math

(Vcmax, maximum rate of carboxylation; Γ*, photorespiratory compensation point; O, partial pressure of oxygen; Rd, mitochondrial respiration; Kc and Ko, Michaelis constant of Rubisco for carbon dioxide and oxygen, respectively). Temperature-dependent kinetic parameters of Rubisco (Kc and Ko) were used as described (Sharkey et al., 2007). Vcmax for each plant was derived by fitting the Farquhar model of RuBP saturated segment, where Ci is below 400 ppm (Long & Bernacchi, 2003). The stomatal density of the mature leaves was determined by Suzuki's Universal Method of Printing (SUMP). Replicas of the abaxial side of leaves were pressed onto a SUMP plate treated with 10 μl SUMP liquid and held in place until the liquid became solid. Microscopic observations were conducted as described above. SPAD values were determined for four successive leaf positions in each plant with the chlorophyll meter (SPAD-502, Konica Minolta, Osaka, Japan). The light response curve for the electron transport rate of the same plants was determined using the pulse-amplitude modulation system (MINI-PAM, Walz).

Gravimetric evaluation of whole plant water use

Twenty-four pellets of Jiffy 7 peat were placed on a 32 × 23 cm plastic tray, and the remaining space was filled with 125 g vermiculite. The initial weight of the tray with soil was recorded. Twenty-four plants of representative ST-OX, ST-RNAi and WT were transplanted to the Jiffy 7 peat pellets at 23 DAG. The weight of the tray was recorded every 3 d until 35 DAG and then every day until 41 DAG. The soil was watered and the soil water content was determined as the difference between the weight after watering and the initial weight of the tray. Irrigation was conducted every 3 d to keep the soil water content above 800 g. Images of the plant population were obtained to calculate the plant coverage every 3 d. To standardize the water use, leaf area was nondestructively estimated as the projection area using the colour threshold tool of ImageJ software (http://rsb.info.nih.gov/ij/). Shoot biomass was determined by harvesting 12 plants at 12 d and 18 d after transplanting (35 and 41 DAG, respectively). Between the first and second harvests, the soil surface was covered with polyethylene to prevent evaporation. Transpiration for the remaining 12 plants was determined every day by the gravimetric method and averaged across the successive 6 d. Water use efficiency (WUE) was calculated from the dry matter accumulation and cumulative transpiration between the first and second harvest. Two replicates were included for all the trials.

Evaluation of whole plant growth

Twenty-four pellets of Jiffy 7 peat were placed on a 32 × 23 cm plastic tray, and the remaining space was filled with 125 g vermiculite. Seeds of ST-OX, ST-RNAi and WT were directly sown in the Jiffy 7 pellets and placed under 300 μmol m−2s−1 white light in a phytotron (Nihon Ika Sophisticated environmental control room, model LPH-0.5P-SH; Nihon Ika, Osaka, Japan) under short-day conditions (8 h light : 16 h dark). The plastic trays were rotated every day during the incubation period. The aerial parts of plants harvested at 32 d were scanned to take two photos using Image Scanner (GT-9700; Epson, Tokyo, Japan). First, the intact aerial part of plants was scanned. Secondly, the leaves of the plants were artificially flattened by making one small section, uncurling the leaf and gently pressing it against the scanner. Projection areas were selected and measured using ImageJ software. After the scanning, the aboveground parts were dried at 80°C for 2 d and biomass was determined by gravimetric measurements.

Statistical analysis

Bartlett's test for equal variance was followed by one-way ANOVA or Kruskal–Wallis one-way ANOVA for the data shown in Figs 1(b–e) and 2-6, and Figs S1(b), S2, S3, and S4(b). Differences were assessed by Tukey's multiple comparison test and the threshold of significance was set to P = 0.05 for all the analyses. Data shown in Figs 4 and 5 were analysed by Tukey's multiple comparison test for A at each light intensity or CO2 concentration. Statistical analyses were performed using SigmaPlot v12 (SYSTAT, San Jose, CA, USA).


Stomatal density, size and water loss in the leaves of STOMAGEN transgenic lines and in leaves exogenously treated with stomagen peptide

Stomatal density and size were analysed in leaves of STOMAGEN transgenic lines and in leaves exogenously treated with stomagen peptide (Fig. 1a–d). STOMAGEN-overexpressing (ST-OX) and silencing (ST-RNAi) lines showed significant variation of stomatal density, ranging from 92 to 1065 mm−2 at the abaxial side of the primary leaf, compared to 286 mm−2 in the wild-type (WT) control (Fig. 1a,b). Direct dipping of plants into a stomagen peptide-containing solution also increased stomatal density (Fig. 1b). Stomatal index showed significant variation in parallel with stomatal density (Fig. 1d). The stomatal densities detected in our study, which showed a wide variation, were higher than those reported previously in a study that used the sdd1 mutant to analyse photosynthetic rates (Schluter et al., 2003). Stomatal density and rate of water loss from the same leaves were measured during the first 5 min after excision to examine the relation between stomatal density and transpiration. Stomatal density showed a positive relationship with the rate of water loss (Fig. 1b,e). Taken together, our results demonstrate that STOMAGEN is a powerful tool to manipulate stomatal density in Arabidopsis.

Effects of stomatal density alterations on whole plant water use

A recent study used leaf temperature analysis to show changes in leaf transpiration rates in STOMAGEN-modified plants (Doheny-Adams et al., 2012). To assess this effect quantitatively, we performed gravimetric measurements of whole plant transpiration rate and water use efficiency (WUE). Transpiration rate was significantly enhanced by 82% in ST-OX compared with WT (Fig. 2a). WUE, calculated from biomass production and cumulative transpiration in plants grown under a PPFD of 100 μmol m−2 s−1, was not statistically different between ST-OX, ST-RNAi and the WT control (Figs 2b, S1). Other report shows that the GT-2 LIKE 1 (GTL1) loss-of-function mutant characterized by decreased stomatal density decreased transpiration rate and improved WUE (Yoo et al., 2010). Our data indicate positive correlation between stomatal density and transpiration.

Relation between stomatal density, gas exchange, and photosynthetic capacity in STOMAGEN transgenic lines

The effects of STOMAGEN overexpression and silencing on the rate of photosynthesis were examined. Mature leaves of 9-wk-old ST-OX and ST-RNAi plants showed significantly increased and decreased stomatal density, respectively, compared with the leaves of WT plants (Fig. 3a). Light-saturated leaf photosynthetic rate was measured at equal leaf positions under a PPFD of 1200 μmol m−2 s−1 using a Walz GFS-3000 gas exchange system. ST-OX lines showed a 72% increase in stomatal conductance (gs) compared with the WT control (Fig. 3b). Surprisingly, the light-saturated leaf CO2 assimilation rate (A) in ST-OX plants was 30% higher than in WT plants (Fig. 3c). Consistently, intercellular CO2 concentration (Ci) in ST-OX was increased in parallel with gs (Fig. 3d). The values of gs, Ci and A were positively changed in parallel with stomatal density (Fig. 3a–d).

Photosynthetic light response curves in the leaves of STOMAGEN transgenic lines

In order to evaluate the effect of light intensity on A in STOMAGEN transgenic lines, light response curves of A and gs were generated in ST-OX and ST-RNAi plants and compared to those of WT plants (Fig. 4). Under relatively low light conditions, there was no difference in A, although there were significant differences in gs. It is indicated that photochemical efficiency under laboratory conditions was not affected even in the presence of significant changes in stomatal density or gs (Fig. 4). At approximately PPFD = 300 μmol m−2 s−1 and above, differences in A were detected in parallel with stomatal density changes, which were also shown in Fig. 3(c) under the light condition PPFD = 1200 μmol m−2 s−1. These results clearly indicate that stomatal density affects the light response of photosynthesis under high light conditions.

A–Ci curves in the leaves of STOMAGEN transgenic lines

In order to determine whether the observed differences in A were attributable to CO2 gas diffusion or carboxylation capacity, changes in A in response to different Ci values were measured under the saturated light condition (PPFD = 1200 μmol m−2 s−1) (Fig. 5a). The A–Ci curve represents the carbon assimilation in mesophyll cells for a given carbon supply through stomata. Maximum carboxylation rate (Vcmax), which was estimated by Farquhar model (Farquhar et al., 1980) (see 'Materials and Methods' for details), was slightly enhanced in ST-OX, although it failed to show significant variance (Fig. 5b). This result suggests that the observed differences in A under ambient CO2 conditions were not dominantly attributable to carboxylation capacity.

Unexpectedly, ST-OX plants grown under high CO2 concentrations showed a significant increase in A (Fig. 5a). It is possible that the CO2 assimilation in mesophyll cells is altered in ST-OX plants, although no difference was observed in the SPAD value, which is an index of the chlorophyll content per unit leaf area (Fig. S2). Determination of the electron transport rate (ETR) in photosystem II using the mini-PAM photosynthetic yield analyzer showed a tendency towards increased ETR in ST-OX under high light conditions, but the difference was not statistically significant (Fig. S3).

Effects of stomatal density alteration on plant growth

In order to determine whether plant growth was affected by changes in single leaf photosynthetic rate, leaf area and whole plant biomass of ST-OX and ST-RNAi plants were compared to those of WT plants under the condition PPFD = 300 μmol m−2 s−1 (see Fig. S4a and 'Materials and Methods' for details). ST-OX showed no significant differences from WT in leaf area and whole plant biomass, whereas ST-RNAi showed growth inhibition (Figs 6, S4). These results indicate that increasing stomatal density does not directly enhance biomass production at PPFD = 300 μmol m−2 s−1.


The present study analysed the relationship between photosynthetic capacity and stomatal density using STOMAGEN transgenic lines in which stomatal density was altered. Photosynthetic capacity increased by 30% in ST-OX compared to WT (Figs 3c, 4a). To our knowledge, this is the first report showing a clear enhancement of A caused by changes in stomatal density in plants grown under constant light conditions. A previous study that used the sdd1 mutant showed that A was increased only when low-light-adapted plants were transferred to high light conditions (Schluter et al., 2003). Furthermore, stomatal aperture was shown to compensate for the effect of stomatal density alterations on photosynthesis in sdd1 (Bussis et al., 2006). The clear enhancement of A observed in ST-OX in the present study indicates that the contribution of stomatal density to photosynthetic rate cannot be negligible, even if compensation by stomatal aperture occurred in ST-OX.

The light photosynthetic response curve of ST-OX showed that increased stomatal density had positive effects on photosynthetic rate under light conditions PPFD = 300 μmol m−2 s−1and above, although the upregulation of gs in ST-OX was observed to be independent of light conditions (Fig. 4). This is consistent with previous reports. The stomatal opening mutant syp121, which decreased stomatal conductance in Arabidopsis, reduced CO2 assimilation only under high light intensity conditions (Eisenach et al., 2012). Drought-related decreases in stomatal conductance affected photosynthesis only at high light intensities (Hummel et al., 2010).

In the present study, we examined the mechanisms underlying the enhancement of A in ST-OX grown under ambient CO2 conditions. Photosynthetic rate is determined by two processes, carboxylation activity (demand function) and CO2 gas diffusion (supply function) (Farquhar & Sharkey, 1982). Changes in light-saturated gs under ambient CO2 conditions, which represent the gas diffusion capacity, occurred in parallel with stomatal density changes (Fig. 3a,b). The enhancement of A also occurred in parallel with the enhancements of Ci and gs (Fig. 3b–d). Furthermore, the Vcmax, which represents carboxylation capacity, did not show any differences (Fig. 5b). Taken together, our data suggest that increased stomatal density affects photosynthetic capacity by modulating the CO2 gas diffusion process rather than carboxylation activity. Our results are consistent with a recent report with stomatal movement mutant slac1 of Oryza sativa (Kusumi et al., 2012), which showed that gas diffusion is the dominant process for photosynthesis under well-watered conditions.

The regulation of photosynthesis by gas diffusion implies that STOMAGEN transgenic lines should not show any difference in A–Ci curves under high Ci values; however, high CO2 concentrations caused an unexpected increase in A in ST-OX plants (Fig. 5a). The enhancement of A under high CO2 concentrations could be associated with the degree of heterogeneous gas diffusion, which is often referred to as ‘stomatal patchiness’. Severe stomatal patchiness results in the underestimation of A due to the lack of uniform Ci in the leaf (Laisk, 1983; Terashima, 1992). In a previous study, the decrease in stomatal density was shown to cause heterogeneous gas diffusion and underestimation of A (Bussis et al., 2006). In the present study, on the other hand, the increased stomatal density in ST-OX may have resulted in more uniform gas diffusion and greater A than in ST-RNAi and WT. Alternatively, overexpression of STOMAGEN could have enhanced mesophyll activity compared to that of WT under high CO2 conditions, which is consistent with the observed slight tendency towards increased ETR in ST-OX (Fig. S3). However, the variation in ETR was not statistically significant. Regardless of the underlying mechanism, our results indicate an additional advantage of increased stomatal density for photosynthesis under high CO2 conditions, which may be an important effect in particular for crop breeding in the predicted higher CO2 environments of the future (Leakey et al., 2009).

Increased stomatal density in ST-OX did not directly improve biomass production at the light intensity of PPFD = 300 μmol m−2 s−1, in which ST-OX showed an increase in A (Figs 4, 6). There are two possible explanations for this result. The significant upregulation of the transpiration rate in ST-OX plants suggests that water stress could occur in STOMAGEN-overexpressing lines (Fig. 2a). Even under well-watered conditions, water transport capacity can limit plant growth (Hsiao et al., 1976). Alternatively, because an increased number of stomata is associated with a high energy expense due to the metabolic cost of stomatal movements (Franks et al., 2009), the additional stomata in ST-OX could deplete the photosynthetic products and inhibit plant growth at PPFD = 300 μmol m−2 s−1. Nonetheless, the biomasses at PPFD = 300 μmol m−2 s−1 (Fig. 6) were compared with those at PPFD = 100 (Fig. S1b), ST-OX showed a greater biomass change than other genotypes, which corresponds to the results of the photosynthetic light response curves (Fig. 4a). This suggests that ST-OX plants may show improved growth in the field, where light intensity is usually higher than PPFD = 300 μmol m−2 s−1.


We demonstrated that increased stomatal density positively changed CO2 gas exchange and enhanced photosynthesis rate by 30% under constant growth conditions. Our results indicate that stomatal density positively affects leaf photosynthetic capacity by modulating the gas diffusion process. Stomatal density may be a target trait for plant engineering to improve photosynthetic capacity in the leaves.

Recent advances in our understanding of the molecular processes associated with stomatal development have enabled the alteration of stomatal density not only by GM, but also by the direct exogenous application of signalling peptides (Sugano et al., 2010; Lee et al., 2012). Our finding that the carboxylation process is not dominantly affected by stomatal density suggests that a combination of alteration of stomatal density and modifications of the biochemical processes of photosynthesis could be effective for improving crop production.


We are grateful to Dr Dominique Bergmann and Mr Graham Dow of Stanford University for reading the manuscript, to Dr Tatsuhiko Shiraiwa of Kyoto University for his help on the SPAD analysis, and to Dr Hiroshi Yamamoto and Dr Toshiharu Shikanai of Kyoto University for their help on the ETR analysis. This work was supported by Specially Promoted Research of Grants-in-Aid to I.H-.N. (no. 22000014) and a research fellowship to S.S.S. (no. 221386) from the Japan Society for the Promotion of Science (JSPS), and by Core Research for Evolutional Science & Technology (CREST) of Japan Science and Technology Agency (JST).