Stomatal, mesophyll conductance, and biochemical limitations to photosynthesis during induction

Abstract

The dynamics of leaf photosynthesis in fluctuating light affects carbon gain by plants. Mesophyll conductance (g_m) limits CO₂ assimilation rate (A) under the steady state, but the extent of this limitation under non-steady-state conditions is unknown. In the present study, we aimed to characterize the dynamics of g_m and the limitations to A imposed by gas diffusional and biochemical processes under fluctuating light. The induction responses of A, stomatal conductance (g_s), g_m, and the maximum rate of RuBP carboxylation (V_cmax) or electron transport (J) were investigated in Arabidopsis (Arabidopsis thaliana (L.)) and tobacco (Nicotiana tabacum L.). We first characterized g_m induction after a change from darkness to light. Each limitation to A imposed by g_m, g_s and V_cmax or J was significant during induction, indicating that gas diffusional and biochemical processes limit photosynthesis. Initially, g_s imposed the greatest limitation to A, showing the slowest response under high light after long and short periods of darkness, assuming RuBP-carboxylation limitation. However, if RuBP-regeneration limitation was assumed, then J imposed the greatest limitation. g_m did not vary much following short interruptions to light. The limitation to A imposed by g_m was the smallest of all the limitations for most of the induction phase. This suggests that altering induction kinetics of mesophyll conductance would have little impact on A following a change in light. To enhance the carbon gain by plants under naturally dynamic light environments, attention should therefore be focused on faster stomatal opening or activation of electron transport.

Gas diffusional and biochemical processes impose significant limitations to CO2 assimilation during photosynthetic induction.

Introduction

Under field environments, light intensity fluctuates over seconds to minutes throughout the day due to changes in solar position, cloud cover, or self-shading in the plant canopy, which affects carbon gain via leaf photosynthesis (Pearcy and Way, 2012; Tanaka et al., 2019). The transition from low to high light induces a gradual increase in CO₂ assimilation rate (A), which is termed “photosynthetic induction” (Pearcy, 1990). A is determined by the combination of CO₂ diffusion from the atmosphere to the chloroplast stroma, and CO₂ fixation in the chloroplast stroma. The CO₂ diffusion pathway for photosynthesis consists of resistances through the leaf boundary layer, stomata, intercellular airspaces, and the components of mesophyll cells such as the cell wall, plasma membrane, cytosol, and chloroplast envelope and stroma (Evans et al., 2009). The conductance, that is the reciprocal of resistance, to gas diffusion via stomata (g_s) has been shown to be a limiting factor of A under fluctuating light (Kaiser et al., 2016; Matthews et al., 2019; Papanatsiou et al., 2019; Shimadzu et al., 2019; Kimura et al., 2020; Yamori et al., 2020). Slow activation of electron transport, Calvin–Benson cycle enzymes, especially Rubisco, and sucrose synthesis can also impose a major limitation to A during photosynthetic induction (Stitt and Schreiber, 1988; Yamori et al., 2012, 2016; Carmo-Silva and Salvucci, 2013; Kaiser et al., 2016). Previous studies have highlighted that in addition to g_s, the conductance to gas diffusion from intercellular airspaces to the chloroplast stroma (g_m) imposes a significant limitation to A under steady-state conditions (Evans et al., 1986; Pons et al., 2009; Flexas et al., 2013). However, no studies have elucidated the limitation to A by g_m under non-steady-state conditions following a change in light intensity. Characterizing the dynamics of g_m and how it limits A under fluctuating light is crucial for understanding the physiological mechanisms regulating carbon gain by plants under field conditions.

Short-term responses of g_m to changes in environmental factors such as CO₂ concentration (Mizokami et al., 2019), temperature (Yamori et al., 2006; von Caemmerer and Evans, 2015), light intensity (Tazoe et al., 2009; Yamori et al., 2010), soil water content, and vapor pressure depression (Warren, 2008) have been determined. The response of g_m to light intensity is still understudied as it varies between plant species and the methods used to estimate g_m. It was reported that g_m increased with increasing light intensity in chickpea and several Eucalyptus species (Campany et al., 2016; Xiong et al., 2018; Shrestha et al., 2019) when estimated from concurrent measurements of gas exchange and carbon isotope discrimination (Evans et al., 1986). On the contrary, g_m was independent of light intensity in wheat and tobacco when estimated by the same method (Tazoe et al., 2009; Yamori et al., 2010). Both light-dependent or -independent responses of g_m have been reported for the same plant species (Xiong et al., 2015, 2018; Carriquí et al., 2019) when estimated from concurrent measurements of gas exchange and chlorophyll fluorescence (Harley et al., 1992). Consequently, the nature of the light response of g_m needs to be studied from a new angle.

Four methods have been developed to estimate g_m based on (1) gas exchange measurements (Ethier and Livingston, 2004), the concurrent measurement of gas exchange with chlorophyll fluorescence to estimate the electron transport rate (J); the (2) constant and (3) variable J methods (Harley et al., 1992), or (4) the concurrent measurement of gas exchange with carbon isotope discrimination (Evans et al., 1986). Although the variable J method is most commonly used to analyze the light response of g_m, estimating g_m under fluctuating light is problematic due to potential changes in alternative electron transport, ATP and NADPH production, and leaf absorptance. The concurrent measurement of gas exchange and carbon isotope discrimination using tunable diode laser (TDL) spectroscopy has enabled the dynamic measurement of g_m under changing CO₂ or temperature conditions, although estimated g_m can be more variable when A is low (Tazoe et al., 2011; Evans and von Caemmerer, 2013). Therefore, a TDL system should enable the dynamics of g_m to be estimated following a step change in light intensity.

In the present study, we aimed to characterize the dynamics of g_m and its limitation of A while considering stomatal opening, Rubisco activation, and electron transport after a step change in light. Measurements were made under 2% O₂ conditions with a gas exchange system coupled to a TDL that measured carbon isotope discrimination. Leaves of two model plants, Arabidopsis (Arabidopsis thaliana (L.) Heynh) and tobacco (Nicotiana tabacum L.), which are commonly used for modeling analyses of leaf photosynthesis (Bernacchi et al., 2002; Walker et al., 2013), were examined under several conditions with step changes in light. The limitations to A imposed by g_s, g_m, and the maximum rate of RuBP carboxylation (V_cmax) or the electron transport rate (J) were analyzed based on the biochemical model for C₃ photosynthesis (Farquhar et al., 1980). The dynamic response of g_m and carbon gain during photosynthetic induction was revealed.

Results

Sensitivity of mesophyll conductance after a step change in light

The induction response of g_m was observed after changing from overnight darkness to high light of 1,000 µmol m⁻² s⁻¹ in Arabidopsis (Figure 1A) or 1,500 µmol m⁻² s⁻¹ in tobacco (Figure 1C), in air containing 2% O₂ and 400 µmol mol⁻¹ CO₂. In the present study, unreasonable values of g_m (i.e. g_m ≥ 0.8 or g_m ≤ 0) were eliminated prior to the calculation of the corresponding average values at each time point (Figure 1). The induction curves of g_m including or eliminating unreasonable values were shown in Supplemental Figure S1, which reveals that data elimination had only a minor effect on the evaluation of g_m induction. The calculation of g_m depends on the values assumed for various parameters and is sensitive to measurement errors. Thus, we began by examining the sensitivity of apparent g_m to changes in the four parameters in Equation 1, day respiration (R_d), discrimination in the carboxylation reactions by Rubisco and phosphoenolpyruvate carboxylase (b), CO₂ concentration in intercellular airspaces (C_i), and the observed fractionation (Δ_o), under steady and non-steady-state conditions. Changing R_d from -10% to +10% of the true value had little impact on g_m, which varied by only -0.3% to 0.3% in either Arabidopsis (Figure 1B) or tobacco (Figure 1D). By contrast, g_m varied by -36.2% to 92.1% when b, C_i, or Δ_o were varied by -10% to +10% of their true value. The potential error in Δ_o was initially about 1.5‰, and gradually decreased to 0.2‰–0.3‰ during photosynthetic induction in Arabidopsis and tobacco (Supplemental Figure S2). The potential uncertainty in Δ_o, defined as the ratio of potential error in Δ_o to the actual value of Δ_o at each time point, was initially 13% in Arabidopsis and 10% in tobacco and gradually decreased to 1%–2% in both species. This lies within the variation range shown for the sensitivity analysis (maximum 20%). In both species, the estimated sensitivity of g_m to errors in parameters were similar at 10 and 40 min after changing from darkness to high light, representing nonsteady and steady states, respectively.

Figure 1.

Potential variability in the estimation of mesophyll conductance after a step change in light. Mesophyll conductance (g_m) was measured every 100 s (A) for 10 min in darkness followed by 50 min under a PPFD of 1,000-μmol photons m⁻² s⁻¹ in Arabidopsis (blue), and (C) for 10 min in darkness followed by 50 min under a PPFD of 1,500-μmol photons m⁻² s⁻¹ in tobacco (red), respectively. CO₂ concentration and air temperature in the leaf chamber were set to 400 µmol mol⁻¹ and 24°C, respectively. A sensitivity of g_m was evaluated in (B) Arabidopsis and (D) tobacco by changing the value of day respiration rate in the light (R_d), CO₂ concentration in intercellular airspaces (C_i), discrimination in the carboxylation reaction by Rubisco and PEPC (b), and the observed carbon isotope discrimination (Δ_o) using data collected 10 (dark) and 40 min (pale) after the light was turned on. The vertical bars on each plot indicate the standard deviation (A, C) or standard error (B, D) with 6–8 replicate leaves for Arabidopsis and 3–6 leaves for tobacco. The numbers (0, 1,000, and 1,500) in the grey and white boxes at the top of (A) and (C) indicate the light intensity.

In addition, the value assumed for cuticular conductance to gas diffusion (g_cut) would influence the calculation of C_i when g_s is low. In turn, this would affect the estimation of g_m and how it changes during induction. We therefore examined the influence of g_cut on apparent g_m during induction in Arabidopsis and tobacco. When C_i was calculated assuming g_cut of 0, 1, 3, or 5 mmol m⁻² s⁻¹, estimated non-steady-state g_m varied up to 45% in Arabidopsis (Figure 2A) and 86% in tobacco (Figure 2B). Greater g_cut resulted in greater g_m during photosynthetic induction when g_s was small. However, varying the assumed value of g_cut had little impact on the relative importance of g_m during induction in either plant species. Also, varying g_cut had no impact on steady-state estimates of g_m.

Figure 2.

Effects of cuticular conductance on estimated mesophyll conductance values. Variation in mesophyll conductance (g_m) was calculated assuming a cuticular conductance (g_cut) of 0 (dark), 1 (slightly dark), 3 (slightly pale), or 5 (pale) mmol m⁻² s⁻¹ in (A) Arabidopsis (blue) and (B) tobacco (red), respectively, with the dataset shown in Figure 1, A and C. The vertical bars on each plot indicate the standard error with 6–8 replicate leaves for Arabidopsis and 3–6 leaves for tobacco. Grey boxes in each figure indicate the initial period of darkness. The numbers (0, 1,000, and 1,500) in grey and white boxes at the top of (A) and (B) indicate the light intensity.

Dynamics of photosynthetic parameters after a step change in light from overnight darkness to high or low light

The induction responses of A, g_s, and g_m were shown after changing from overnight darkness to high light (1,000 or 1,500 µmol m⁻² s⁻¹) in Arabidopsis and tobacco (Figure 3). According to the biochemical model of C₃ photosynthesis, A can be limited by RuBP carboxylation or regeneration (Farquhar et al., 1980). The responses of V_cmax and J were shown assuming either RuBP-carboxylation or RuBP-regeneration limiting conditions, respectively. The speed of response for each parameter varied (Table 1). The time taken for each parameter to reach 50% and 90% of the maximum value (t₅₀ and t₉₀, respectively) increased in the order of V_cmax, g_m, J, A, and g_s, and those for g_s were significantly longer than those for the other parameters in both plant species (P < 0.05). These results indicate that stomatal conductance was the slowest to respond to a change in light out of all the parameters. There was no significant variation in t₅₀ and t₉₀ values between the two plant species for any of the parameters. The responses of C_i and C_c following the step increase in light were similar for both plant species.

Figure 3.

Response of CO₂ assimilation rate, stomatal and mesophyll conductance, CO₂ concentration in the intercellular airspaces and chloroplast stroma, the maximum rate of RuBP carboxylation, and electron transport after changing from darkness to high light. A CO₂ assimilation rate (A), stomatal (g_s; dark circle) and mesophyll conductance (g_m; pale triangle) to CO₂, and CO₂ concentration in intercellular airspaces (C_i; dark circle) and chloroplast stroma (C_c; pale triangle) were measured every 100 s under the light condition of (A) a PPFD of 0 and 1,000-μmol photons m⁻² s⁻¹ for 10 min and 50 min in Arabidopsis (blue), and (B) a PPFD of 0 and 1,500-μmol photons m⁻² s⁻¹ for 10 and 50 min in tobacco (red), respectively. The maximum rate of RuBP carboxylation (V_cmax; dark circle) and electron transport (J; pale triangle) were also estimated assuming RuBP-carboxylation or regeneration limiting conditions according to the biochemical model for C₃ photosynthesis. CO₂ concentration and air temperature in the leaf chamber were set to 400 µmol mol⁻¹ and 24°C, respectively. The vertical bars on each plot indicate the standard error with 6–8 replicate leaves for Arabidopsis and 3–6 leaves for tobacco. Grey boxes in each figure indicate the initial period of darkness. The numbers (0, 1,000, and 1,500) in the grey and white boxes at the top of (A) and (B) indicate the light intensity.

Table 1.

Comparison of the response speed for CO₂ assimilation rate, stomatal and mesophyll conductance, and the maximum rate of RuBP carboxylation in Arabidopsis and tobacco

Parameters	Arabidopsis		Tobacco
	t 50 (min)	t 90 (min)	t 50 (min)	t 90 (min)
A	9.5 ± 0.8 b	19.7 ± 1.5 b	10.0 ± 1.4 a	19.0 ± 2.1 a
g s	18.8 ± 1.0 c	29.7 ± 1.4 c	18.5 ± 0.9 b	28.9 ± 0.8 b
g m	8.7 ± 1.2 b	14.8 ± 1.8 ab	9.2 ± 2.2 a	17.1 ± 3.3 a
V cmax	4.3 ± 0.5 a	9.7 ± 1.2 a	5.3 ± 1.0 a	11.0 ± 3.0 a
J	8.9 ± 0.8 b	19.2 ± 1.5 b	9.5 ± 1.4 a	18.6 ± 2.1 a

t ₅₀ and t₉₀ are the times when CO₂ assimilation rate (A), stomatal (g_s) and mesophyll (g_m) conductance to CO₂, the maximum rate of RuBP carboxylation (V_cmax), and the electron transport rate (J) reached 50% and 90% of their maximum value, respectively, after the step change in light. Different letters indicate significant variation in t₅₀ or t₉₀ between the five parameters (P < 0.05) according to the Tukey–Kramer test.

We evaluated the relative limitations to A imposed by g_s (L_gs), g_m (L_gm), and V_cmax (L_Vcmax) or J (L_J), defined as the relative change in A for a relative change in the four parameters, during photosynthetic induction with the dataset shown in Figure 3 (Figure 4). Assuming the RuBP-carboxylation limiting condition, L_gs was 60% during the initial 10 min after switching on the light, then decreased gradually to 20%–30%. L_gm decreased slightly before increasing to within the range of 20%–30% during photosynthetic induction in Arabidopsis and tobacco. L_Vcmax increased from 20% to 40%–50% in both plant species after switching on the light. g_s imposed the greatest limitation during the initial 25 min in Arabidopsis and 15 min in tobacco, while V_cmax imposed the greatest limitation during the latter phase in both plant species. g_m imposed the least limitation during photosynthetic induction except during the initial 10 min. If RuBP regeneration was assumed to limit Rubisco, Lg_s and Lg_m were initially 20%–25% and gradually decreased to 2%–3% during induction in a similar manner for both plant species. J imposed much greater limitation to A than g_s and g_m under both steady and nonsteady states.

Figure 4.

Gas diffusional and biochemical limitations of CO₂ assimilation rate after changing from darkness to high light in Arabidopsis and tobacco. The limitations to CO₂ assimilation rate (A) imposed by stomatal conductance (L_gs; dark), mesophyll conductance (L_gm; pale) and biochemical processes were evaluated in (A) Arabidopsis (blue) and (B) tobacco (red), respectively, with the dataset shown in Figure 3. Biochemical limitations assumed either RuBP-carboxylation (upper panels, L_Vcmax) or RuBP-regeneration (bottom panels, L_J) limiting conditions. The vertical bars on each symbol indicate the standard error with 6–8 replicate leaves for Arabidopsis and 3–6 leaves for tobacco. Grey boxes in each figure represent the initial period of darkness. The numbers (0, 1,000, and 1,500) in the grey and white boxes at the top of (A) and (B) indicate the light intensity.

We investigated the responses of g_m after changing from darkness to high (1,500 µmol m⁻² s⁻¹) or low light (200 µmol m⁻² s⁻¹) followed by a reciprocal transition between high and low light, in tobacco. The induction response of g_m is shown after changing from overnight darkness to high light (Figure 5A) or low light (Figure 5C). Subsequently, upon changing from high to low light or low to high light, g_m did not change, while A, g_s, C_i, and C_c changed (Figure 5, A and B, Supplemental Figure S3). The t₅₀ and t₉₀ values for g_m did not differ between the two light conditions (Figure 5E).

Figure 5.

Responses of mesophyll conductance following transitions to low and high light for tobacco. Mesophyll conductance (g_m) was measured every 100 s under the light sequence (A) from darkness (10 min) to a PPFD of 1,500 (50 min) then 200 (20 min) or (C) from darkness (10 min) to 200 (40 min) then 1,500-µmol photons m⁻² s⁻¹ (30 min). CO₂ concentration and air temperature in the chamber were set to 400 µmol mol⁻¹ and 24°C, respectively. Grey and pale-grey boxes in each figure represent the periods of darkness and 200-µmol photons m⁻² s⁻¹, respectively. The numbers (0, 200, and 1,500) in the grey, pale-grey, and white boxes at the top of (A) and (C) indicate the light intensity. The variation in g_m under steady state was compared between the light intensity of a PPFD of 200 (pale-grey) and 1,500 (white)-µmol photons m⁻² s⁻¹ under the light sequence (B) from darkness (10 min) to 1,500 (50 min) then 200 (20 min) or (D) from darkness (10 min) to 200 (40 min) then 1,500-µmol photons m⁻² s⁻¹ (30 min). (E) The time taken for g_m to reach 50% (t₅₀) and 90% (t₉₀) of their maximum value was compared between two light sequences. The vertical bars on each plot and columns in (A), (C), and € indicate the standard error with 3–6 replicate leaves for tobacco.

Dynamics of the photosynthetic parameters after a step change from short or long periods of darkness to high light

To understand how quickly induction relaxed, tobacco leaves were subjected to 50 min of high light followed by darkness for 10 min or 60 min and responses of A, g_s, g_m, V_cmax, and J were analyzed during the second period of high light (Figure 6, A and B). The speed of response for all the parameters was slower as the intervening dark period lengthened. The t₅₀ and t₉₀ values for g_m, V_cmax, and J were significantly shorter under high light following a 60 min period of darkness compared with after a 5 h period of darkness or overnight darkness (Figure 7). The speed of responses for the parameters were similar following 5 h darkness or overnight darkness (Figure 6, A and C) with no significant variation in t₅₀ and t₉₀ for g_m, V_cmax, and J under those light conditions (Figure 7). V_cmax had the fastest response of all of the parameters under any light condition (Table 1 and Figure 7).

Figure 6.

Responses of CO₂ assimilation rate, stomatal and mesophyll conductance, CO₂ concentration in intercellular airspaces and chloroplast stroma, the maximum rate of RuBP carboxylation, and electron transport after short or long periods of darkness. CO₂ assimilation rate (A), stomatal (g_s; dark circle), and mesophyll conductance (g_m; pale triangle) to CO₂, and CO₂ concentration in the intercellular airspaces (C_i; dark circle) and chloroplast stroma (C_c; pale triangle) were measured every 100 s under the light condition of (A) a PPFD of 0, 1,500, 0, and 1,500-μmol photons m⁻² s⁻¹ for 10, 50, 10, and 30 min, and (B) 0, 1,500, 0, and 1,500-μmol photons m⁻² s⁻¹ for 10, 50, 60, and 30 min preceded by overnight-darkness, or (C) 2 h of sunlight followed by 5 h of darkness, then 0 and 1,500-μmol photons m⁻² s⁻¹ for 10 and 50 min, respectively. The maximum rate of RuBP carboxylation (V_cmax; dark circle) and electron transport (J; pale triangle) were also estimated assuming RuBP-carboxylation or RuBP-regeneration limiting conditions. CO₂ concentration and air temperature in the chamber were set to 400 µmol mol⁻¹ and 24°C, respectively. The vertical bars on each plot indicate the standard error with 3–6 replicate leaves for tobacco. Grey boxes in each figure represent the periods of darkness. The numbers (0 and 1,500) in the grey and white boxes at the top of (A), (B), and (C) indicate the light intensity.

Figure 7.

Comparison of response speeds for mesophyll conductance and the maximum rate of RuBP carboxylation in tobacco after changing from short or long periods of darkness to high light. The time taken for mesophyll conductance (g_m) (A), and the maximum rate of RuBP carboxylation (V_cmax) (B) and electron transport (J) (C) to reach 50% (t₅₀) and 90% (t₉₀) of their maximum value was compared between inductions under high light after darkness for 60 min or 5 h following illumination for more than 1 h, or overnight darkness. The vertical bars on each plot indicate the standard error with 3–6 replicate leaves for tobacco. Different letters on each column indicate significant variation in t₅₀ or t₉₀ of three parameters at P < 0.05 according to Tukey–Kramer test.

Finally, we evaluated L_gs, L_gm, and L_vcmax or L_J with the dataset shown in Figure 6 (Figure 8). The dynamics of gas diffusional (L_gs and L_gm) and biochemical (L_vcmax and L_J) limitations were similar to those shown in Figure 4 after changing from overnight darkness to high light. Following a 10 min dark interruption after 50 min in high light, all the limitations scarcely changed under high light, and L_vcmax was much larger than L_gs and L_gm, assuming the RuBP-carboxylation limiting condition (Figure 8A). However, those limitations were affected by longer dark interruptions (Figure 8, B and C). During induction subsequent to darkness for 60 min, 5 h, or overnight, initially L_gs was dominant but then decreased as L_vcmax increased, while L_gm changed only slightly. Stomata imposed the greatest limitation during the early phase of photosynthetic induction, while V_cmax imposed the greatest limitation during the latter phase under any of the light sequences, except for high light after 10 min darkness. Assuming the RuBP-regeneration limiting condition, 10 min and 60 min dark interruptions induced little or no changes in any of the limitations during induction, while 5 h dark interruption induced substantial changes in each limitation. Electron transport rate imposed the greatest limitation to A under both nonsteady and steady states under any of the light sequences.

Figure 8.

Gas diffusional and biochemical limitations of CO₂ assimilation rate after changing from short or long periods of darkness to high light. The limitations to CO₂ assimilation rate (A) imposed by stomatal conductance (L_gs; dark), mesophyll conductance (L_gm; pale), and biochemical processes were evaluated in tobacco with the dataset shown in Figure 6. Biochemical limitations assumed either RuBP carboxylation (upper panels, L_Vcmax) or RuBP regeneration (bottom panels, L_J) under the light condition of: A, 0, 1,500, 0, and 1,500-μmol photons m⁻² s⁻¹ for 10, 50, 10, and 30 min; B, 0, 1,500, 0, and 1,500-μmol photons m⁻² s⁻¹ for 10, 50, 60, and 30 min preceded by overnight darkness; or C, 2 h of sunlight followed by 5 h of darkness, then 0 and 1,500-μmol photons m⁻² s⁻¹ for 10 and 50 min, respectively. The vertical bars on each symbol indicate the standard error with 3–6 replicate leaves for tobacco. Grey boxes in each figure represent the periods of darkness. The numbers (0 and 1,500) in the grey and white boxes at the top of (A), (B), and (C) indicate the light intensity.

Discussion

Photosynthetic induction has been investigated in relation to stomatal opening, activation of electron transport, or the enzymes of the Calvin–Benson cycle, especially Rubisco and sucrose synthesis (Stitt and Schreiber, 1988; Tanaka et al., 2019; Yamori et al., 2020). Although g_m can impose a major limitation to A during the steady state, little is known about the dynamics of g_m and its limitation of A under fluctuating light. In the present study, we aimed to characterize the dynamics of g_m and the limitations to A imposed by gas diffusional and biochemical processes during photosynthetic induction to provide novel insight into the physiological mechanisms regulating carbon gain by plants under changing light conditions.

Light response of mesophyll conductance during steady and nonsteady states

The light response of g_m under steady state is controversial since it varies between plant species and the method used to estimate g_m. In tobacco, steady-state g_m did not vary with light intensity when calculated from concurrent measurements of gas exchange and carbon isotope discrimination (Yamori et al., 2010), but increased with increasing light intensity when calculated using the variable J method (Carriquí et al., 2019). In the present study, there was no clear variation in g_m under steady state after changing from high (1,500 µmol m⁻² s⁻¹) to low (200 µmol m⁻² s⁻¹) light or low to high light in tobacco (Figure 5, A and C). This result confirms that g_m estimated by concurrent measurements of gas exchange and carbon isotope discrimination was similar under 200 and 1,500 µmol m⁻² s⁻¹ in tobacco.

The induction response of g_m was observed after changing from overnight darkness to high or low light in Arabidopsis and tobacco in air containing 2% O₂ (Figures 3, 5). Previously, Kaiser et al., (2017) reported the dynamics of g_m using the variable J method, but the authors pointed out uncertainties associated with parameters relating to the estimation of electron transport rate (alternative electron transport, ATP and NADPH production, and leaf absorptance) and R_d after a step change in light. The evaluation of g_m during photosynthetic induction would be affected by uncertainty in parameter values assumed for the calculation of g_m (Gu and Sun, 2014) and by ignoring g_cut when g_s is low (Mizokami et al., 2015). In the present study, g_m was rather insensitive to the assumed value of R_d in Arabidopsis and tobacco (Figure 1, B and D), indicating that small changes in R_d can be ignored in the evaluation of g_m after a step change in light. The sensitivity of g_m to changes in the four parameters, R_d, b, C_i, and Δ_o, was similar between the steady and nonsteady states. Varying each of these parameters by ±10% alters estimated g_m values from -36.2% to 92.1% in both states (Figure 1, B and D). It should be noted that g_m appears more variable during the early phase rather than the later phase of induction because Δ_o tended to be more uncertain owing to low A (Supplemental Figure S2 and Supplemental Data Set S1). Although variation in g_cut caused a large variation in the estimation of g_m initially when g_s was very low, it resulted in only minor variation in the induction curves of g_m in both plant species (Figure 2). This suggests that g_cut would not be a major factor affecting the dynamics of g_m. These results support the conclusion that the change observed in apparent g_m during induction is unlikely to be attributable to measurement artifacts in the present study.

There was no significant variation in the induction speed of g_m between the light transitions from darkness to low (200 µmol m⁻² s⁻¹) or high light (1,500 µmol m⁻² s⁻¹) in tobacco (Figure 5E). This observation suggests that g_m has a similar speed of response regardless of light intensity. Following a 10 min dark interruption, g_m reached a high value immediately after switching on the light (Figure 6A). Moreover, t₅₀ and t₉₀ values for g_m were shorter under high light after a 60 min dark interruption following high light than after a 5 h dark interruption (Figure 7A). These results suggest that once g_m is induced, g_m recovers more quickly following a short interrupting period of darkness.

What is the physiological mechanism underlying the induction response of g_m after a step change in light? It was shown that g_m is closely related to the surface area of chloroplasts exposed to intercellular airspaces per unit leaf area (S_c; Evans et al., 1994). The thickness of the mesophyll cell wall is known to contribute significant resistance to internal CO₂ diffusion (Terashima et al., 2011; Peguero-Pina et al., 2012). A simulation analysis indicated that the light response of g_m would be affected by the three-dimensional structure of a leaf in which there are cells with different characteristics and light penetration is spatially variable (Théroux-Rancourt and Gilbert, 2017). In addition, the CO₂ permeability of the plasma membrane and chloroplast envelope are major components of the internal resistance (Evans et al., 2009). Carriquí et al. (2019) reported in tobacco that while g_m varied under different light intensities, morphological factors did not, concluding that the response of g_m to light was likely due to changes in biochemical factors such as the activity of aquaporins and carbonic anhydrase. It should be noted that Carriquí et al. (2019) investigated variation in morphological factors and g_m under different light intensities after adaptation to high light (1,500 µmol photons m⁻² s⁻¹) in which the morphological arrangement might be fully induced. Thus, in the present study, the induction response of g_m after changing from darkness to low or high light could be attributable to changes in biochemical and/or morphological factors.

The chloroplast avoidance response induced by blue-light irradiation decreased S_c and then g_m in Arabidopsis (Tholen et al., 2008), which may possibly suppress the induction of g_m under high light after darkness. However, the rapid reduction of g_m by blue-light irradiation that was completed within 2–3 min was insensitive to the addition of cytochalasin (Loreto et al., 2009), suggesting chloroplast movement probably does not contribute to the dynamics of g_m following a step change in light. It has been shown that aquaporins function to regulate g_m by affecting the CO₂ permeability of the plasma membrane in Arabidopsis (Heckwolf et al., 2011; Uehlein et al., 2012), tobacco (Flexas et al., 2006; Uehlein et al., 2008), and rice (Hanba et al., 2004; Xu et al., 2019), respectively. Changes in membrane permeability associated with aquaporins are regulated by phosphorylation, heteromerization of isoforms, Ca²⁺ and proton concentrations, pressure, osmotic solute concentration, internal or external factors such as nutrient, temperature and reactive oxygen species, and subcellular trafficking (Chaumont et al., 2005), which might be associated with the regulation of the g_m dynamics.

Gas diffusional and biochemical limitations during photosynthetic induction

According to the biochemical model for C₃ photosynthesis, A is limited by RuBP carboxylation, RuBP regeneration, or triose phosphate utilization (TPU, starch, and sucrose synthesis), depending upon CO₂ and O₂ partial pressures, light intensity, and temperature (Farquhar et al., 1980, Sharkey, 1985). A few studies have evaluated the responses of V_cmax and J during induction under ambient CO₂ and O₂ conditions based on dynamic A–C_i analysis. The responses of A to C_i were generated by varying ambient CO₂ concentrations at different times during photosynthetic induction (Soleh et al., 2016; Taylor and Long, 2017; Salter et al., 2019; De Souza et al., 2019; Acevedo-siaca et al., 2020). These previous studies evaluated the stomatal and nonstomatal limitations during induction (Soleh, et al., 2016; Taylor and Long, 2017; Deans et al., 2019), and reported the primary limitation to A was imposed by either g_s, V_cmax, or J, depending on the plant species or the measurement conditions. However, V_cmax and J estimated by a dynamic A–C_i analysis is not only affected by biochemical factors, but also by g_m, meaning that the estimated limitations depend on what happens to g_m. Under 21% O₂, one could expect a lag between the oxygenase reaction by Rubisco and the release of CO₂ from photorespiration. This lag could vary during photosynthetic induction and is evident as a CO₂ burst in darkness following a period in the light (Vines et al., 1983). The lag could result in the overestimation of V_cmax. In the present study, we evaluated the induction response of V_cmax and J under 2% O₂ condition knowing the dynamic changes in g_m, and assumed A was under RuBP-carboxylation or RuBP-regeneration limiting conditions (Figures 3, 6). As photorespiration is virtually eliminated under 2% O₂, photorespiration is unlikely to have influenced our estimation of V_cmax during induction. The induction response of V_cmax that we observed after changing from darkness to high light in Arabidopsis and tobacco (Figure 3) is similar to that reported for several plant species, while the response of J was much slower in the present study than that shown in Taylor and Long (2017). Notably, the present study revealed substantial limitations to A by g_s, g_m, and V_cmax or J during photosynthetic induction (Figures 4, 8), indicating that speeding up both gas diffusional and biochemical processes could achieve faster photosynthetic induction in plants after a step change in light. In particular, g_s responded more slowly than g_m and V_cmax to changing light, which resulted in much greater L_gs than L_gm and L_Vcmax during photosynthetic induction assuming the RuBP-carboxylation limiting condition (Table 1, Figures 4, 8). On the other hand, L_J was much greater than L_gs and L_gm during the induction assuming the RuBP-regeneration limiting condition. Taken together, faster stomatal opening and activation of electron transport could improve carbon gain during photosynthetic induction following a transition from darkness to high light (Yamori, 2016; McAusland et al., 2016; Lawson and Vialet-Chabrand, 2019; Matthews et al., 2019; Papanatsiou et al., 2019; Kimura et al., 2020; Yamori et al., 2020).

TPU can limit A under steady-state conditions of high CO₂ and/or low O₂ (Sharkey, 1985). TPU can also limit A under non-steady-state conditions under ambient CO₂ and O₂ (Stitt and Schreiber, 1988). In the present study, gas exchange measurements were conducted in 2% O₂ to minimize the effect of photorespiration on carbon isotope discrimination, which might increase the likelihood for TPU limitation of A during the photosynthetic induction compared to that under 21% O₂. If so, the limitations imposed by V_cmax or J may have been overestimated during induction.

In the present study, the induction responses of g_s and V_cmax were shown after changing from darkness for 10 min to high light (Figure 6A). This suggests that a short period of darkness (even 10 min) would result in stomatal closure and Rubisco inactivation, which would then again require photosynthetic induction upon re-illumination. In addition, a rapid change in RuBP regeneration was reported to limit photosynthetic induction under high light after a short period of low light or darkness (Kobza and Edwards, 1987; Sassenrath-Cole and Pearcy, 1992), which would also apply in the present study. Although L_Vcmax was much larger than L_gs and L_gm when RuBP-carboxylation limiting conditions were assumed, V_cmax rapidly reached steady state under high light after 10 min darkness, while g_s and g_m slowly increased (Figures 6, 8). On the other hand, L_J was more than 95% throughout induction, while g_s and g_m showed slower responses than J. L_gm was smallest of all the limitations during most of the induction phase regardless of which limiting conditions were assumed following a step change in light, suggesting that the induction of mesophyll conductance would not greatly affect A when light intensity varies. These results suggest that the induction speed of A under high light following a short period of darkness would be mainly limited by stomatal opening and/or electron transport.

Conclusion

We characterized the dynamics of mesophyll conductance (g_m) after changing from darkness to high or low light in Arabidopsis and tobacco. The induction speed of g_m was similar irrespective of the light intensity during photosynthetic induction. Once g_m was fully induced, the response speed of g_m was faster the shorter the period of darkness. During the induction of photosynthesis, CO₂ assimilation rate (A) was mainly limited by stomatal conductance (g_s), the maximum rate of RuBP carboxylation (V_cmax), or electron transport (J), whereas the limitation associated with g_m varied little and was less important. The most effective targets for increasing carbon gain by plants in dynamic light conditions are therefore likely to be faster stomatal opening and activation of electron transport.

Materials and methods

Plant materials and cultivation

Arabidopsis (Arabidopsis thaliana (L.) Heynh, “Columbia-0 (CS60000)”) and tobacco (Nicotiana tabacum L. “Petit Havana (N,N)”) were used as plant materials in the present study. Arabidopsis plants were sown and grown on soil under an air temperature of 22°C and a photosynthetic photon flux density (PPFD) of 200 µmol photons m⁻² s⁻¹. The day/night length was set to 8/16h. Tobacco plants were sown on April 16, 2019 and transplanted to 4L pots containing Green Wizard potting mix with slow release fertilizer (Osmocote Exact, Scotts, NSW, Australia) on May 3, 2019. Tobacco plants were grown under sunlight in a greenhouse with the day/night temperature of 25°C/20°C. All the plants were watered and fertilized as needed. Gas exchange measurements were made 53 and 54 d after sowing for Arabidopsis, and 45 to 52 d after sowing for tobacco plants.

Dynamic analysis of mesophyll conductance

The gas exchange and carbon isotope discrimination measurements were simultaneously conducted as described by Evans and von Caemmerer (2013). Prior to the measurement, plants were kept under dark conditions overnight or for 5 h after sunlight illumination for 2 h. In addition, the whole plant was kept under dark conditions during gas exchange measurements except for the leaves clamped into the chamber. We set a flow rate of 200 µmol s⁻¹, a CO₂ concentration of 400 µmol mol⁻¹, 2% O₂, and an air temperature at 24°C in the leaf chamber. The 2% O₂ gas was supplied to LI-6400 (LI-COR, Lincoln, NE, USA) as described by Evans and von Caemmerer (2013) to minimize the effect of photorespiration on the estimation of g_m. Various light sequences were generated in the leaf chamber as follows; a PPFD of 0 and 1,000-μmol photons m⁻² s⁻¹ for 10 min and 50 min (LS1); 0, 1,500, and 200-μmol photons m⁻² s⁻¹ for 10, 50, and 20 min (LS2); 0, 200, and 1,500-μmol photons m⁻² s⁻¹ for 10, 40, and 30 min (LS3); 0, 1,500, 0, and 1,500-μmol photons m⁻² s⁻¹ for 10, 50, 10, and 30 min (LS4); and 0, 1,500, 0, and 1,500-μmol photons m⁻² s⁻¹ for 10, 50, 60, and 30 min (LS5). The leaf to air vapor pressure difference decreased from 1.76 to 1.04 kPa in Arabidopsis and from 2.30 to 0.95 kPa in tobacco, respectively, during induction. The water content of air entering the chambers was set by flowing it through Nafion tubing (Perma Pure LLC, Toms River, NJ, USA, MH-110-12P-4) surrounded by water circulating from a temperature-controlled water bath. Gas exchange measurements were coupled to a tunable diode laser (TDL, TGA100, Campbell Scientific, Inc., Logan, UT, USA) for measurement of the carbon isotope composition. The ¹²CO₂ and ¹³CO₂ composition of five gases were each measured for 20 s in a repeating 100 s cycle: CO₂-zero gas, then reference and sample gases from two LI-6400s. Gas exchange measurements were made to obtain CO₂ assimilation rate (A), stomatal conductance to CO₂ (g_s), and CO₂ concentration in intercellular airspaces (C_i) every 100 s with eight leaves of Arabidopsis under LS1, and four to six leaves of tobacco under LS2, LS3, LS4, LS5, and LS6. The carbon isotope discrimination, g_m, and CO₂ concentration in the chloroplast stroma (C_c) were calculated as described by Evans and von Caemmerer (2013) and Busch et al. (2020). g_m was calculated by using the following equation

$${\mathit{g}}_{\mathit{m}}=\frac{1+\mathit{t}}{1-\mathit{t}}\mathit{ }(\mathit{b}-{\mathit{a}}_{\mathit{i}}-\frac{\mathit{e}{\mathit{R}}_{\mathit{d}}}{\mathit{A}}) \frac{\mathit{A}}{{\mathit{C}}_{\mathit{a}}}/({\Delta }_{\mathit{i}}-{\Delta }_{\mathit{o}}-{\Delta }_{\mathit{e}}-{\Delta }_{\mathit{f}})$$ (1)

where t is the ternary correction factor (Farquhar and Cernusak, 2012), b is the discrimination in the carboxylation reaction by Rubisco and PEPC (29‰), a_i is the fractionation factor for dissolution and diffusion through water (1.8‰), e is the fractionation factor for day respiration, R_d is the day respiration, Δ_i, Δ_o, Δ_e, and Δ_f are the fractionations assuming C_i = C_c in the absence of any respiratory fractionation (e = 0), the observed fractionation, the fractionation associated with respiration, and the fractionation associated with photorespiration, respectively. For a more detailed explanation of Equation 1, see Evans and von Caemmerer (2013) and Busch et al. (2020). We measured the respiration rate during the initial dark period for each leaf and used leaf temperature response functions described by Bernacchi et al. (2001) to estimate R_d and the CO₂ compensation point in the absence of R_d (Γ*). The linear relationship between Γ* and O₂ partial pressure was assumed to estimate Γ* under 2% O₂ (Brooks and Farquhar, 1985).

Sensitivity analysis in the estimation of mesophyll conductance

Sensitivity of the g_m estimation to changes in R_d, C_i, b, and Δ_o was evaluated by changing these parameters from -10%, -5%, +5%, or +10% of the true values. Sensitivity analysis was conducted on the data collected at 10 and 40 min after the light intensity was changed from darkness to 1,000 or 1,500 µmol photons m⁻² s⁻¹ in Arabidopsis and tobacco, respectively, with the dataset shown in Figure 1, A and C. In addition, potential errors in Δ_o depending on the measurement stability of the TDL system were calculated from the following equation

$${\text{Potential} \text{error} \text{in} \Delta }_{\mathit{o}}=2 · \text{standard} \text{error} \text{of} {\delta }^{13}{\mathit{C}}_{\text{ref}} · \xi$$ (2)

where δ¹³C_ref was the carbon isotope composition of the reference gases from two LI-6400s. ξ was defined as C_ref/(C_ref - C_sam), in which C_ref and C_sam are the CO₂ concentrations of reference and sample gases. Subsequently, the potential variation range of Δ_o was calculated as the ratio of errors in Δ_o to the actual value of Δ_o at each time point.

It was reported that C_i can be overestimated when g_s is low if cuticular conductance to water vapor (g_cut) is ignored. This could result in the underestimation of g_m (Mizokami et al., 2015). In the present study, we simulated the variability in g_m with the change in g_cut. C_i is calculated from the following equation according to Boyer et al. (1997)

$${\mathit{C}}_{\mathit{i}}=\frac{({\mathit{g}}_{\text{sc}}-{\mathit{E}}_{\mathit{s}}/2){ \mathit{C}}_{\mathit{a}}-\mathit{A}}{{\mathit{g}}_{\text{sc}}+{\mathit{E}}_{\mathit{s}}/2}$$ (3)

where g_sc is the stomatal conductance to CO₂, E_s is the transpiration rate via stomata, and C_a is the CO₂ concentration in the leaf chamber. g_sc and E_s are defined as the following equations

$${\mathit{g}}_{\text{sc}}=\frac{({\mathit{g}}_{\text{lw}}-{\mathit{g}}_{\text{cut}})}{1.6}$$ (4)

$${\mathit{E}}_{\mathit{s}}={ \mathit{E}}_{\mathit{l}}-{\mathit{g}}_{\text{cut}} ({\mathit{W}}_{\mathit{l}}-{ \mathit{W}}_{\mathit{a}})$$ (5)

where g_lw is the total stomatal conductance to water vapor, E_l is the total transpiration rate, W_l and W_a are the mole fraction of water vapor in the leaf and the leaf chamber, respectively. C_i and g_m were calculated by assuming g_cut of 0, 1, 3, or 5 mmol m⁻² s⁻¹ in Arabidopsis and tobacco with the dataset shown in Figure 1, A and C.

Analysis of gas diffusional and biochemical limitations of the photosynthetic induction

According to the biochemical model of photosynthesis developed by Farquhar et al. (1980), A under RuBP-carboxylation (A_c) or RuBP-regeneration (A_r) limiting condition are described as below

$${\mathit{A}}_{\mathit{c}}=\frac{ {\mathit{V}\mathit{c}}_{\text{max}} (\mathit{C}-{\Gamma }^{*})}{\mathit{C}+{\mathit{K}}_{\mathit{c}} (1+\mathit{O}/{\mathit{K}}_{\mathit{o}})}-{\mathit{R}}_{\mathit{d}}$$ (6)

$${\mathit{A}}_{\mathit{r}}=\frac{\mathit{J} (\mathit{C}-{\Gamma }^{*})}{4\mathit{C}+8{\Gamma }^{*}}-{\mathit{R}}_{\mathit{d}}$$ (7)

where V_cmax is the maximum rate of RuBP carboxylation, C and O are the CO₂ and O₂ concentration, and K_c and K_o are the Michaelis constants for CO₂ and O₂, respectively. J is the rate of whole chain linear electron transport. Based on equations (6) and (7), we calculated V_cmax and J using the following equations

$${\mathit{V}}_{\text{cmax}}\mathit{ }=\mathit{ }\frac{({ \mathit{A}}_{\mathit{c}}\mathit{ }+\mathit{ }{\mathit{R}}_{\mathit{d}})({\mathit{C}}_{\mathit{c}}+{\mathit{K}}_{\mathit{c}} (1+\mathit{O}/{\mathit{K}}_{\mathit{o}}\left)\right)}{{\mathit{C}}_{\mathit{c}}\mathit{ }-\mathit{ }{\Gamma }^{*}}$$ (8)

$$\mathit{J}\mathit{ }=\mathit{ }\frac{({ \mathit{A}}_{\mathit{r}}\mathit{ }+\mathit{ }{\mathit{R}}_{\mathit{d}})(4{\mathit{C}}_{\mathit{c}}+{8\Gamma }^{*})}{{\mathit{C}}_{\mathit{c}}\mathit{ }-\mathit{ }{\Gamma }^{*}}$$ (9)

Here, the observed A at each measurement was substituted for A_c and A_r under a PPFD of 1,000 µmol m⁻² s⁻¹ in Arabidopsis or 1,500 µmol m⁻² s⁻¹ in tobacco in Equations 6 and 7, respectively. K_c and K_o are calculated by using leaf temperature response functions described by Bernacchi et al. (2002). We evaluated the dynamics of V_cmax and J under each light condition assuming that A would be limited either by RuBP carboxylation or regeneration throughout the measurements in the present study.

The limitations to A imposed by the stomatal (L_gs) and mesophyll (L_gm) conductance, and V_cmax (L_Vcmax) or J (L_J) were examined using the following equations as described in Grassi and Magnani (2005)

$${\mathit{L}}_{{\mathit{g}}_{\mathit{s}}}=\frac{{\mathit{g}}_{\text{tot}}/{\mathit{g}}_{\text{sc}} · {\partial \mathit{A}}_{\mathit{c}} (\text{or} {\partial \mathit{A}}_{\mathit{r}})/{\partial \mathit{C}}_{\mathit{c}}}{{\mathit{g}}_{\text{tot}}+{\partial \mathit{A}}_{\mathit{c}} (\text{or} {\partial \mathit{A}}_{\mathit{r}})/{\partial \mathit{C}}_{\mathit{c}}} ·100$$ (10)

$${\mathit{L}}_{{\mathit{g}}_{\mathit{m}}}=\frac{{\mathit{g}}_{\text{tot}}/{\mathit{g}}_{\mathit{m}} · {\partial \mathit{A}}_{\mathit{c}} (\text{or} {\partial \mathit{A}}_{\mathit{r}})/{\partial \mathit{C}}_{\mathit{c}}}{{\mathit{g}}_{\text{tot}}+{\partial \mathit{A}}_{\mathit{c}} (\text{or} {\partial \mathit{A}}_{\mathit{r}})/{\partial \mathit{C}}_{\mathit{c}}} ·100$$ (11)

$${\mathit{L}}_{{\mathit{V}}_{\text{cmax}}\mathit{ }(\text{or} \mathit{J})}=\frac{{\mathit{g}}_{\text{tot}}}{{\mathit{g}}_{\text{tot}}+{\partial \mathit{A}}_{\mathit{c}} (\text{or} {\partial \mathit{A}}_{\mathit{r}})/{\partial \mathit{C}}_{\mathit{c}}} ·100$$ (12)

where g_tot was the total conductance to CO₂ from the leaf surface to chloroplast stroma assuming g_cut of 0 mmol m⁻² s⁻¹. ∂A_c/∂C_c and ∂A_r/∂C_c, the partial differential of Equations 6 and 7 to C_c, were calculated using the following equations

$${\partial \mathit{A}}_{\mathit{c}}/{\partial \mathit{C}}_{\mathit{c}}=\frac{{\mathit{V}}_{\text{cmax}} ({\Gamma }^{*}+{\mathit{K}}_{\mathit{c}} (1+{\mathit{O}/\mathit{K}}_{\mathit{c}}\left)\right)}{{({\mathit{C}}_{\mathit{c}}+{\mathit{K}}_{\mathit{c}} (1+{\mathit{O}/\mathit{K}}_{\mathit{c}}\left)\right)}^2}$$ (13)

$${\partial \mathit{A}}_{\mathit{r}}/{\partial \mathit{C}}_{\mathit{c}}=\frac{3 · \mathit{J} · {\Gamma }^{*}}{{4({\mathit{C}}_{\mathit{c}}+{2\Gamma }^{*})}^2}$$ (14)

Curve fitting and statistical analysis

The plot sequences of four parameters, A, g_s, g_m, V_cmax, and J, were fitted to a Boltzmann sigmoidal function of time (f_(t)) as described in Supplemental Figure S4

$${\mathit{f}}_{\left(\mathit{t}\right)}={ \mathit{X}}_{\text{min}}+\frac{{\mathit{X}}_{\text{max}}-{\mathit{X}}_{\text{min}}}{1+\hspace{0.17em}\text{exp}\hspace{0.17em} ({\mathit{t}}_{50}-\mathit{t})/\mathit{d}\mathit{t}}$$ (15)

where X_min and X_max are the minimum and maximum values for each parameter, respectively, and t₅₀ is the time when each parameter reached 50% of the maximum value after switching on the light, dt is (X_max − X_min)/(4 × slope at the inflection point) for each parameter. In the present study, g_m, V_cmax, and J under the dark condition were assumed to be zero for the curve fitting since these parameters were unmeasurable under such conditions. The time when each parameter reached 90% of the maximum value (t₉₀) was also calculated from the curve-fitted function. Curve-fitting was performed with the dataset under a PPFD of 1,500 µmol m⁻² s⁻¹ shown in Figures 3, 6, B and C, using curve fitting tool in the SciPy optimize module of Python (Python Software Foundation, Delaware, USA).

Unreasonable values of g_m (i.e. g_m ≥ 0.8 or g_m ≤ 0) were eliminated prior to the calculation of its average value and the other parameters, and statistical analysis. Unreasonable values of C_c, Vc_max, and J (≤0) were also eliminated. The time series dataset of gas exchange parameters shown in Figure 3 is provided in Supplemental Data Set S1. It includes the average value of A, g_s, g_m, C_i, C_c, Vc_max, and J, and the replication number and standard deviation (column, “SD”) of those parameters at each timepoint.

We evaluated the significance of the variations in t₅₀ and t₉₀ among parameters or t₅₀ and t₉₀ of g_m among light sequences according to the Tukey–Kramer test. The variation in g_m under the steady state was compared between the light intensity of 200 and 1,500 µmol photons m⁻² s⁻¹ using a boxplot analysis. The values of g_m in the last 10 min under 1,500 and 200 µmol photons m⁻² s⁻¹ in LS2 and LS3 were compared in a boxplot analysis. Statistical analysis was conducted using R software version 3. 6. 1 (The R Foundation for Statistical Computing, Vienna, Austria).

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. Response of mesophyll conductance after changing from darkness to high light.

Supplemental Figure S2. Variations in parameters related to carbon isotope discrimination.

Supplemental Figure S3. Response of CO₂ assimilation rate, stomatal conductance, and CO₂ concentration in the intercellular airspaces and chloroplast stroma, following transitions to low and high light for tobacco.

Supplemental Figure S4. Example of curve-fitting time sequence data for gas exchange parameters through time.

Supplemental Data Set S1. Time series dataset of CO₂ assimilation rate, stomatal and mesophyll conductance, CO₂ concentration in the intercellular airspaces and chloroplast stroma, the maximum rate of RuBP carboxylation, and electron transport after changing from darkness to high light.

 

K.S., W.Y., and J.R.E. conceived and designed this work. K.S. performed all the gas exchange experiments and data analyses with the support by M.G. K.S. wrote the manuscript with input from co-authors and all the authors contributed extensively to its finalization.

The authors responsible for distribution of materials integral to the findings presented in this article are in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plphys) are: K.S. (sakoda@g.ecc.u-tokyo.ac.jp), W.Y. (yamori@g.ecc.u-tokyo.ac.jp).