Abstract
Mesophyll conductance (gm) determines the diffusion of CO2 from the substomatal cavities to the site of carboxylation in the chloroplasts and represents a critical limiting factor to photosynthesis. In this study, we evaluated the average effect sizes of different environmental constraints on gm in Populus spp., a forest tree model. We collected raw data of 815 A-Ci response curves from 26 datasets to estimate gm, using a single curve-fitting method to alleviate method-related bias. We performed a meta-analysis to assess the effects of different abiotic stresses on gm. We found a significant increase in gm from the bottom to the top of the canopy that was concomitant with the increase of maximum rate of carboxylation and light-saturated photosynthetic rate (Amax). gm was positively associated with increases in soil moisture and nutrient availability, but insensitive to increasing soil copper concentration, and did not vary with atmospheric CO2 concentration. Our results showed that gm was strongly related to Amax and to a lesser extent to stomatal conductance (gs). Also, a negative linear relation was obtained between gm and specific leaf area, which may be used to scale-up gm within the canopy.
Introduction
Carbon assimilation of plants is importantly determined by the diffusion efficiency of CO2 from the atmosphere to the site of carboxylation. The rate of CO2 diffusion is affected by two main diffusion limitations. The first limitation controls the CO2 flux from the atmosphere to the sub-stomatal cavities through the stomata and is characterized by stomatal conductance (gs). The second limitation determines the diffusion of CO2 from the substomatal cavities to the sites of carboxylation in the chloroplasts and is characterized by mesophyll conductance (gm). gm is composed of gaseous and liquid phase resistances (Flexas et al., 2008; Evans et al., 2009; Niinemets et al., 2009). CO2 diffusion inside the leaves is complex, facing a series of structural barriers coupled with biochemical regulations. It has been shown that gm is typically limited by liquid phase conductance both in species with soft mesophytic leaves as well as in species with tough xerophytic leaves (Tomás et al., 2013; Tosens et al., 2012a,b). The liquid phase is a multi-components pathway that involves the mesophyll cell wall thickness and porosity, the plasmalemma, the chloroplast envelope, the chloroplast thickness and the mesophyll surface area exposed to intercellular air spaces per unit of leaf area (Evans et al., 2009; Tosens et al., 2012b; Tomás et al., 2013). After extensive study during the last two decades, gm is now widely accepted as a critical limiting factor to photosynthesis, which has to be considered in characterizing plant carbon gain potentials and responses to future climate change (Evans et al., 2009; Niinemets et al., 2009; Niinemets et al., 2011; Flexas et al., 2016).
Mesophyll conductance has been shown to respond to environmental stress and may govern functional plasticity of photosynthesis and plant fitness under limited resources (Galle et al., 2009; Barbour et al., 2010; Buckley and Warren, 2014; Théroux-Rancourt et al., 2015; Flexas et al., 2016; Shrestha et al., 2018). However, recent findings on the response of gm to abiotic stress are conflicting and inconclusive, demonstrating the complex nature of gm variation (Flexas et al., 2008; Niinemets et al., 2009; Zhou et al., 2014; Shrestha et al., 2018). This suggests that the environmental and species-specific response (acclimation) of gm to growth conditions should be considered to predict plant performance in the field. Among the contrasting environmental responses, growth temperature may (Warren, 2008; Silim et al., 2010) or may not (Dillaway and Kruger, 2010; Benomar et al., 2018) affect gm. Similarly, the increase in soil nitrogen may (Warren 2004; Shrestha et al., 2018; Xu et al., 2020; Zhu et al., 2020) or may not (Bown et al., 2009) stimulate gm. The magnitude of decrease in gm under water stress and low light differed among studies (Warren et al., 2003; Niinemets et al., 2006; Montpied et al., 2009; Bögelein et al., 2012; Tosens et al., 2012a; Zhou et al., 2014; Peguero-Pina et al., 2015; Théroux Rancourt et al., 2015). These discrepancies among studies result in part from (i) the absolute changes in anatomical, morphological (mesophyll structure) and biochemical (aquaporins and carbonic anhydrase) traits controlling gm, as well as from changes in the relative contribution of these traits (Marchi et al., 2008; Tomás et al., 2013), and from (ii) the level of coordination between gm, gs and leaf-specific hydraulic conductivity (KL) (Flexas et al., 2013; Théroux-Rancourt et al., 2014; Xiong et al., 2017). Given the complex interplay between different factors controlling gm, it is important to examine its acclimation at the genus and species level to gain a general insight into the mechanistic basis of changes in gm.
Five methods exist to estimate gm: a) chlorophyll fluorescence coupled to gas exchange (Harley et al., 1992), (b) carbon isotope discrimination coupled to gas exchange (initially developed by Evans et al., 1986), c) oxygen isotope discrimination (Barbour et al., 2016), (d) A-Ci curve fitting (Ethier and Livingston, 2004; Sharkey et al., 2007), and (e) 1D modeling of gm from leaf structural characteristics (Evans et al., 2009, Tosens et al., 2012b, Tomas et al., 2013). All of these methods are based on assumptions and each one has its limitations (Flexas et al., 2013; Tosens and Laanisto, 2018). The standard deviation of the estimate of gm may vary from 10% to 40%, which may limit our understanding of gm acclimation to growth conditions, particularly when the variation between treatments or studies is less than the error of estimates (Sun et al., 2014a).
Populus spp., model crops in forestry characterized by high yield potential, have been the subject of numerous studies to understand the physiological response to environmental factors but research is still necessary to make assessment of effects sizes and to draw generalizations (Larocque et al., 2013). A general understanding of the CO2 pathway through mesophyll and how it is affected by environmental factors would be beneficial in the efforts to (i) accurately predict canopy photosynthesis under different environmental conditions, particularly under warmer and drier climate, and improve global carbon assimilation models and (ii) develop and effectively select more resilient and productive cultivars for wood, bioenergy and bioremediation. Substantial data of A-Ci response curves in the literature has been used to estimate photosynthetic parameters, not to estimate gm, and such compiled dataset would provide a basis to make such assessments on the response of gm to the environment.
In this study, we compiled 815 A-Ci response curves from 26 datasets of different poplar species and hybrids (Table 1). Published A-Ci curve-fitting approaches differ broadly regarding the rectangularity of the hyperbola, segmentations of the model of photosynthesis and determination of the transition value of CO2 from carboxylation to electron transport (Harley et al., 1992; Ethier and Livingston, 2004; Manter and Kerrigan 2004; Dubois et al., 2007; Sharkey et al., 2007; Pons et al., 2009; Gu et al., 2010). These approaches led to different fitted values (Miao et al., 2009, Sun et al., 2014a). Although A-Ci curve fitting is unreliable for species with large gm, it can provide results similar to those obtained from direct measurements for species with medium to low gm (Niinemets et al., 2005, 2006; Qiu et al., 2017; Xu et al., 2020). Using the compiled A-Ci response curves, we performed curve fitting using a single method (Ethier and Livingston, 2004) to alleviate the fitting method bias and to obtain uniformed estimates of gm, maximum rate of carboxylation (Vcmax) and rate of electron transport (J). We further collected related variables like leaf nitrogen content, stomatal conductance and specific leaf area (SLA) when data were available. Our main goal was to find trends in the response of mesophyll conductance to prevalent abiotic stressors and to examine the relationship between gm and other leaf physiological and morphological traits. We believe that a meta-analytical approach to analyze the accumulated data on the diffusion of CO2 through the mesophyll diffusion pathway in relation to other photosynthesis-related traits provides key insight into physiological and structural controls on mesophyll conductance and into the environmental plasticity of mesophyll conductance. We aim at contributing to the efforts of improving poplar photosynthetic efficiency in poplar breeding programs, and at improving modelling of global carbon assimilation of biomass and bioenergy crops under climate change.
List of dataset sources used in the meta-analysis
Materials and methods
Data collection
Data were collected by a web search in Web of Science, Scopus, and Google Scholar using the following key words: (“Populus” or “poplar” or “hybrid poplar” or “aspen”) and (“Vcmax” or “maximum rate of electron transport (Jmax)” or “mesophyll conductance”). At this step, abstract of every item was checked to confirm the paper is actually about gm. Then, we looked at the materials and methods section of selected papers where A-Ci response curves of Populus spp. were measured.
To get raw data of A-Ci response curves, we contacted the corresponding authors or co-authors of the targeted studies by e-mail and via ResearchGate. We obtained 23 data sets from published studies and three data sets from unpublished studies (Table 1). Collectively, they provided a total of 815 A-Ci response curves.
The total data of 72 genotypes were collected from measurements on plants growing in plantations (5 studies), or under controlled conditions (greenhouse or growth chamber setups; 21 studies) with optimal and stressful conditions (Table 1). After compiling all A-Ci curves, the quality of the data was assessed based on the following criteria: 1) only curves with at least 2 points in the saturation region (J region) were retained; 2) only curves with P<0.05 for the p-value of fitted curves using the method of Ethier and Livingston (2004) were retained; Consequently, 65 curves that did not meet these conditions were removed; and 3) based on the literature, gm values in Populus spp. using at least two methods simultaneously never exceeded 1 mol m−2 s−1 (Singsaas et al., 2004; Flexas et al., 2008; Velikova et al., 2011; Tosens et al., 2012a; Théroux-Rancourt et al., 2014; Momayyezi and Guy, 2017; Xu et al., 2020). Then, gm > 1 mol m−2 s−1 were considered as non-available data (94 entries), and Vcmax and J values retained for further analyses. Furthermore, fewer data point and/or large interval in the intracellular concentration of CO2 (Ci μmol mol−1) in the RuBisCO limited phase region are the major source of error and probability of failing gm (reached the set bound during the grid search) frequently encountered during A-Ci curve fitting (Warren, 2006; Miao et al., 2009, Sun et al., 2014a; Moualeu-Ngangue et al., 2016).
Data subsets
To examine the effect of a given abiotic factor on gm, we estimated that a minimum of three studies is necessary to have reliable conclusions, regardless of the genotype used. Then, we could come up with subsets of data that focused on the same variable and performed analyses on them separately (identified in the column ‘treatment’ in Table 1). Our first goal was to examine the effect of variations in these factors on gm, light-saturated photosynthetic rate (Amax), gs, J, Vcmax and in a second step, the relationships between gm and other photosynthetic characteristics (Amax, gs, J, Vcmax). The data subsets included the following environmental factors:
– Canopy level: four studies addressed the photosynthetic activity of leaves at the bottom, middle and top of trees (Niinemets et al., 1998; Merilo et al., 2010; Calfapietra et al., 2005; Benomar, 2012).
– Ambient CO2: we examined the response of trees to elevated ambient CO2 from Calfapietra et al. (2005), Merilo et al. (2010) and Tissue et al. (2010). We considered 370 ppm as the control treatment in the three studies, while the elevated CO2 was 550 ppm of CO2 for the studies of Calfapietra et al. (2005) and Merilo et al. (2010), and 700 ppm for the study of Tissue et al. (2010).
– Copper stress: data sets from studies of Borghi et al. (2007) and Borghi et al. (2008) were used to examine the response of poplar trees to contamination of the substrate with copper (Cu). Treatments were assigned to three levels of Cu: 0 (0 to 0.4 μM), 20 (20 to 25 μM) and 75 (75 to 100 μM).
– Soil nitrogen (N) content: High vs. low soil N content treatments were reported in four studies: Benomar et al. (2018), Ripullone et al. (2003), Calfapietra et al. (2005) and Xu et al. (2020). In Merilo et al. (2010), no effect of nitrogen fertilization was observed by authors due to high background nutrient availability in the plantation site.
– Soil moisture: water status of trees was assessed and data from four studies was classified into two treatments: control (optimal watering) vs. water deficit (Tosens et al., 2012a; Li et al., 2012; Théroux-Rancourt, unpublished; Benomar et al., unpublished).
For Xu et al. (2020), we extracted data from the article (means and standard errors) and generated three replicates assuming a normal distribution using SURVEYSELECT procedure of SAS (SAS Institute, software version 9.4, Cary, NC, USA). The reason is that the authors used the same curve fitting approach (Ethier and Livingston 2004) we used in the meta-analysis (Table 1).
For studies with two or more investigated factors, we considered the different levels of the factor of interest and the control level of the rest of factors to avoid between-factors interaction effects on results. For example, in Calfapietra et al. (2005), trees were subject to different levels of N and CO2. When we focused on the effect of N, we selected trees exposed to ambient CO2 only (control).
Curve analysis
Mesophyll conductance and photosynthetic capacity variables, Vcmax and J, were estimated by fitting A-Ci curve with the non-rectangular hyperbola version of the biochemical model of C3 plants (Farquhar et al., 1980). The model was fitted using non-linear regression techniques (Proc NLIN, SAS) following Dubois et al. (2007) and Sun et al. (2014a).
Briefly, the net assimilation rate (An) is given as:
with
where Vcmax is the maximum rate of carboxylation (μmol m−2 s−1), O is the partial atmospheric pressure of O2 (mmol mol−1), Γ* is the CO2 compensation point in the absence of mitochondrial respiration, Rday is mitochondrial respiration in the light (μmol CO2 m−2 s−1), Cc is the chloroplast CO2 (μmol mol−1), Ci is the intercellular air space concentration of CO2 (μmol mol−1), Kc (μmol mol−1) and Ko (mmol mol−1) are the Michaelis–Menten constants of RuBisCO for CO2 and O2, respectively, J is the rate of electron transport (μmol m−2 s−1). The values at 25 °C used for Kc, Ko and Γ* were 272 μmol mol−1, 166 mmol mol−1 and 37.4 μmol mol−1, respectively (Sharkey et al., 2007) and their temperature dependencies were as in Sharkey et al. (2007).
In four data sets, measurements were carried out under a temperature that was different from the reference (25 °C). In this case, Vcmax and J were normalized to 25 °C using the model of Kattge and Knorr (2007), which integrates the acclimation to growth temperature. However, the actual values of Vcmax and J were more often significant compared to normalized values, and this was true using both ANOVA and regression.
Statistical analyses
Data analysis assessing the effect of the environmental factors on gm and the relationship between gm and the other traits were carried out using SAS software (SAS Institute, software version 9.4, Cary, NC, USA).
When at least three studies focused on one factor (nitrogen, CO2, canopy level, copper), the effect of treatments on light-saturated photosynthetic rate (Amax), gm, and gs was assessed, separately for each response variable, through mixed model analyses of variance using the primary data (Riley et al., 2010; Mengersen et al., 2013). “Treatment” was the fixed effect while “study” and “genotype” nested within study were the random effects. The number of replicates was not necessarily balanced across treatments. The assumptions of normality of the residuals and homogeneity of variance were verified, and a log-transformation was made when necessary.
Results
The number of studies on mesophyll conductance has rapidly increased since 2000, and more remarkably since 2013 (Fig. 1), suggesting a growing interest among plant ecophysiologists to understand the role of gm in photosynthesis. This pattern was very similar to the increase of publication number on mesophyll conductance in Populus spp. (Fig. 2).
Cumulative number of published studies for mesophyll conductance (gm) between the years 2000 to 2020. Number of publications were determined using keywords (e.g. gm) through database search available at the Web of Science Core Collection. (https://clarivate.com/webofsciencegroup/solutions/web-of-science-core-collection/).
Cumulative number of published studies for mesophyll conductance (gm) in Populus spp. between the years 2001 to 2020. Number of publications were determined using keywords (e.g. Populus) through database search available at the Web of Science Core Collection. (https://clarivate.com/webofsciencegroup/solutions/web-of-science-core-collection/).
Canopy level
Light-saturated photosynthetic rate at an ambient CO2 concentration (380-400 μmol mol−1), Amax, significantly increased from 7.1±0.44 μmol m−2 s−1 on average at the bottom leaves to 13.0±0.45 μmol m−2 s−1 at the mid-canopy, to 16.2±0.53 μmol m−2 s−1 at the upper canopy (Fig. 3a). Similar to Amax, gm had an ascending pattern, from the bottom (0.12±0.01 mol CO2 m−2 s−1) to the top of the canopy (0.24±0.02 mol m−2 s−1) (Fig. 3c). Stomatal conductance (gs) was the lowest at the bottom canopy (0.17±0.01 mol H2O m−2 s−1) and then increased to 0.36±0.02 mol H2O m−2 s−1 at the mid and the upper canopy (Fig. 3b). The gm/gs ratio was significantly greater at the upper canopy (1.17±0.1), compared to the mid-canopy leaves (0.88±0.1) and was not different everywhere else (Fig. 3d). While, Vcmax increased similarly to Amax and gm from the bottom to the top of the canopy (Fig. 3e); however, SLA had an opposite trend (Fig. 3f).
Effect of the leaf position in the canopy (Bottom: Bot, Middle: Mid, Upper: Upp) on light-saturated photosynthetic rate (Amax, a); stomatal conductance (gs, b); mesophyll conductance (gm, c); gm/gs ratio (d); maximum rate of carboxylation (Vcmax, e) and specific leaf area (SLA, f). In gm/gs ratio, gs for water (mol H2O m−2 s−1) was divided by 1.6 to obtain gs in mol CO2 m−2 s−1. Means having the same letters are not significantly different at α = 0.05.
Ambient CO2
Increased air CO2 had no effect on average Amax (14.43 ±0.60 μmol m−2 s−1), gm (0.21±0.02 mol m−2 s−1) and gm/gs (1.09 ±0.1) (Fig. 4a, 4b and 4c). However, average gs was higher (0.40 ±0.03 mol H2O m−2 s−1) under ‘‘Ambient’’, compared to ‘‘Elevated’’ CO2(0.32±0.02 mol H2O m−2 s−1) (Fig. 4b).
Effect of the ambient air CO2 concentration on light-saturated photosynthetic rate (Amax, a); stomatal conductance (gs, b); mesophyll conductance (gm, c) and gm/gs ratio (d). In gm/gs ratio, gs for water (mol H2O m−2 s−1) was divided by 1.6 to obtain gs in mol CO2 m−2 s−1. Means having the same letters are not significantly different at α = 0.05.
Copper stress
Amax was not affected when Cu soil concentration increased from 0 to 20 or 75 μM (9.67±0.95 μmol m−2 s−1) (Fig. 5a). It should be noted that at the highest Cu level (75 μM), Amax ranged from 4 to 15 μmol m−2 s−1. Average gs significantly decreased under medium (20 μM, 0.17±0.02 mol H2O m−2 s−1), and high Cu treatment (75 μM, 0.18±0.03 mol m−2 s−1), compared to control treatment (Fig. 5b). Increasing Cu concentration in soil did not affect gm (Fig. 5c). The gm/gs ratio was greater under 20 and 75 μM Cu, compared to control (Fig. 5d).
Effect of the soil Copper (Cu) concentration on light-saturated photosynthetic rate (Amax, a); stomatal conductance (gs, b); mesophyll conductance (gm, c) and gm/gs ratio (d). In gm/gs ratio, gs for water (mol H2O m−2 s−1) was divided by 1.6 to obtain gs in mol CO2 m−2 s−1. Means having the same letters are not significantly different at α = 0.05.
Soil nitrogen
Amax was significantly greater (16.1 ±0.61 μmol m−2 s−1) under the high soil nitrogen (HN, 250 kg N ha−1 y−1 in field study or 20 mM for pot study), compared to the low nitrogen treatment (LN, 12.9±0.65 μmol m−2 s−1) (Fig. 6a). The high supply of nitrogen increased gs (from 0.29±0.03 at LN to 0.36±0.03 mol m−2 s−1 at HN) and gm (from 0.19±0.02 to 0.23±0.02 to mol m−2 s−1), but had no effect on gm/gs ratio (1.38±0.16 on average) (Fig. 6b, 6c and 6d).
Effect of the soil nitrogen content (high nitogen: HN, low nitrogen: LN) on light-saturated photosynthetic rate (Amax, a); stomatal conductance (gs, b); mesophyll conductance (gm, c) and gm/gs ratio (d). In gm/gs ratio, gs for water (mol H2O m−2 s−1) was divided by 1.6 to obtain gs in mol CO2 m−2 s−1 Means having the same letters are not significantly different at α = 0.05.
Soil moisture
Average Amax decreased by drought (range of leaf predawn water potential under water deficit, Ψleaf = −0.7 to −0.8, soil water content = 10%), dropping from 17.0±0.7 μmol m−2 s−1 to 14.8± 0.8 μmol m−2 s−1 on average with minimum value (3.8 μmol m−2 s−1) much lower than in watered trees (8.9 μmol m−2 s−1) (Fig. 7a). As expected, soil moisture deficit remarkably altered gs, decreasing its average value from 0.33±0.02 mol m−2 s−1 in control trees to 0.20±0.03 mol m−2 s−1 under drought conditions (Fig. 7b). Drought had the same effect on gm, but to a lesser extent than gs. gm decreased from 0.27±0.02 to 0.19±0.02 mol m−2 s−1 under soil moisture deficit (Fig. 7c). In addition, gm/gs ratio increased by 37% when plants were subject to a drought (Fig. 7d).
Effect of the soil moisture on light-saturated photosynthetic rate (Amax, a); stomatal conductance (gs, b); mesophyll conductance (gm, c) and gm/gs ratio (d). In gm/gs ratio, gs for water (mol H2O m−2 s−1) was divided by 1.6 to obtain gs in mol CO2 m−2 s−1. Means having the same letters are not significantly different at α = 0.05.
Relationship between CO2 diffusion and photosynthetic activity
Amax was strongly correlated to both gs and gm (P=0.001) and to Vcmax (P=0.001) over all the studies (Fig. 8a, 8b and 8c). Based on the collected data, gm was significantly correlated to gs (P = 0.04). However, the relationship was not linear. gm was the highest (0.4-0.5 mmol m−2 s−1) when gs values were intermediate (0.2-0.4 mol m−2 s−1), and lowest at high gs values (Fig. 8e).
Relationship between light-saturated photosynthetic rate (Amax), stomatal conductance (gs), mesophyll conductance (gm), maximum rate of carboxylation (Vcmax), electron transport rate (J), gm/gs ratio, specific leaf area (SLA) and per area leaf Nitrogen concentration (Narea). In gm/gs ratio, gs for water (mol H2O m−2 s−1) was divided by 1.6 to obtain gs in mol CO2 m−2 s−1.
We found a significant negative correlation between SLA and gm (P=0.001) (Fig. 8g) based on the collected data from studies that measured SLA (n=12). Leaf nitrogen content reported by three studies showed a significant correlation between gm and N content per area (Narea) (Fig. 8f). gm increased with Narea until a saturation point (~ 0.25 mol m−2 s−1).
Discussion
Canopy level
The scaling up of photosynthesis from leaves to the canopy and stands (using the model of Farquhar et al., 1980) requires a deep understanding of within-canopy variations in leaf morpho-physiology and the main drivers of foliage acclimation to the dynamic gradient of environmental conditions (light, temperature, vapor pressure deficit (VPD) and soil moisture) (Niinemets et al., 2006; Buckley and Warren, 2014; Niinemets et al., 2015). Unfortunately, pieces of knowledge regarding the variation of gm within the canopy and its mechanistic basis are scarce, in particular for Populus spp. This situation may explain why most global carbon cycle models remain ‘‘gm-lacking’’, with possible consequences, such as overestimating the fertilization effect of CO2 on global gross primary production and underestimation of water-use efficiency (WUE) and canopy gross photosynthesis under future climate (Sun et al., 2014b; Knauer et al., 2019). The steep and parallel increase of gm, Amax and Vcmax from the bottom to the top of the canopy found here for Populus spp. is in agreement with the findings of Niinemets et al. (2006) for Quercus ilex L., Montpied et al. (2009) for Fagus sylvatica L., and Warren et al. (2003) for Pseudotsuga menziesii (Mirbel) Franco. A decrease of gm from the bottom to the top of the canopy was also reported (Bögelein et al., 2012; Cano et al., 2013). In absence of water deficit, gm limitation is greater under high light (top of the canopy), compared to shade conditions (Niinemets et al., 2009; Cano et al., 2013).
We observed a significant inverse relationship between gm and specific leaf area (SLA), comparable to previous studies (Montpied et al., 2009; Niinemets et al., 2006; Tosens et al., 2012b). This suggests that the increase in leaf thickness (lower SLA), e.g. in developing leaves and in leaves grown under higher light, may be associated with increased gm (Tosens et al., 2012b). Tosens et al. (2012b) for leaves grown under low and high soil moisture, and Muir et al. (2014) for four species of Solanum spp. showed a positive relationship between SLA and gm. In Tosens et al. (2012b), this relationship reflected increased density of leaves grown under lower water availability. This evidence collectively demonstrates the complex nature of the relationship between SLA and gm, reflecting the circumstance that SLA is an inverse of the product of leaf thickness and density that can respond differently to environmental drivers (Niinemets, 1999; Poorter et al., 2009). The profile of gm within the canopy observed here may be partially attributable to the morphological acclimation of Populus spp. foliage to light availability within the canopy. Moreover, this inverse-relationship between SLA and gm was used as an empirical model to estimate a maximum attainable gm at different canopy layers for C3 plant and was implemented in Community Land Model (CLM.4.5) (Sun et al., 2014b; Knauer et al., 2019).
The change in morphological traits and their role in the acclimation of gm to a vertical gradient of environmental conditions within the canopy need additional investigations. For instance, shade acclimation of leaf morphology is associated with a lower surface area of chloroplasts exposed to intercellular air spaces (Sc/S) and thicker chloroplasts (Hanba et al., 2002; Niinemets et al., 2006; Tosens et al., 2012b; Peguero-Pina et al., 2015). Species-specific leaf development patterns (i.e., evergreen sclerophyllous vs. deciduous broadleaves) affect limitations to gas diffusion, thus determining the carbon balance of leaves (Marchi et al., 2007). However, light acclimation may be species-specific and altered by water and soil nitrogen, and leaf ontogeny (Niinemets et al., 2006; Tazoe et al., 2009; Peguero-Pina et al., 2015; Shrestha et al., 2018). It is still unclear whether gm profile within the canopy is the result of the change in SLA.
Our results showing higher gs and gm/gs at the top of the canopy are in disagreement with the findings of Montpied et al. (2009) and Bögelein et al. (2012), suggesting a species- and environment-specific gradient of gm/gs. Temperature and VPD responses of gm and gs are different (Cano et al., 2013), resulting in different diurnal patterns of gm and gs. Then, the gradient of gm/gs ratio along the canopy may drive the WUE at the canopy level and the midday depression of photosynthetic rate regardless of the level of isohydry of clones (Cano et al., 2013; Buckley and Warren, 2014; Stangl et al., 2019).
Ambient CO2
The response of photosynthetic capacity and diffusion of CO2 to free air CO2 enrichment (FACE) considerably differed between species and experimental setups. The decrease in Amax and gs in response to elevated CO2 showed in our meta-analysis is in agreement with numerous studies on Populus spp. and other species (Ainsworth and Rogers, 2007; Medlyn et al., 2013; DaMatta et al., 2016), but is in disagreement with the findings of some other studies, e.g., Sigurdsson et al. (2001) and Uddling et al. (2009). For gm, the effect of growth CO2 changed among studies and some species having an intrinsic low gm are more likely to respond to elevated CO2 than species with high intrinsic gm (Niinemets et al., 2011). However, several studies have reported that gm may decrease or be unresponsive to CO2 enrichment (Singsaas et al., 2004; Zhu et al., 2012; Kitao et al., 2015; Mizokami et al., 2019). This suggests that the increase of Amax under elevated CO2 cannot be attributed solely to gm variation (Singsaas et al., 2004). The absence of gm response to elevated CO2 complicates the research on mechanisms underlying this variation. Unlike gm, researchers proposed some hypotheses like least-cost theory, nitrogen limitation and resources investment to explain the decrease of Amax, Vcmax and gs under elevated CO2 (Leakey et al., 2009; Smith and Keenan, 2020).
Copper stress
Similar to our findings, gm remained unchanged in the herbaceous plant Silene paradoxa L., exposed to high Cu concentration, although gs decreased significantly (Bazihizina et al., 2015). In other cases of exposure to other heavy metals, like nickel (Ni), Velikova et al. (2011) reported a significant decrease in chloroplast CO2 content and mesophyll conductance in black poplar (P. nigra L.) exposed to 200 μM Ni under a hydroponic setup (compared to control = 30 μM Ni). This reduction of gm might be attributed to an alteration of leaf structure by toxic effect of high concentrations of heavy metals in mesophyll cells Velikova et al. (2011). Hermle et al. (2007), reported an acceleration of senescence and necrosis of mesophyll cells in P. tremula L. leaves exposed to Cu, Zn, Cd and Pb at 640, 3000, 10 and 90 mg per kg of soil, respectively, and a decrease of chloroplast size from the early stages of exposure. The study of Hermle et al. (2007) also reported the thickening of cell walls and change of their chemical composition in damaged mesophyll cells, which might have affected permeability of cell walls and diffusion of CO2 through them. Mercury (Hg) (HgCl2 form) altered CO2 diffusion through aquaporins, a membrane channel of CO2 diffusion, in faba bean (Vicia faba L.) (Terashima and Ono, 2002) and significantly reduced gm in P. trichocarpa Torr. & Gray. HgCl2 may also decrease gm indirectly by disrupting carbonic anhydrase activity, as reported by Momayyezi and Guy (2018), and demonstrated that carbonic anhydrase activity is strongly associated with gm variation in P. trichocarpa Torr. & Gray (Momayyezi and Guy, 2017).
Soil Nitrogen
The increase of Amax by the enhancement of Vcmax in response to more available soil nitrogen has been established in the literature. However, the possible contribution of gm to this augmentation remains unexplored for several species. Our results showed a concomitant increase of gm with a higher supply of N. A positive correlation between the level of expression of AQPs genes (PIPs and TIPs) and gm has been reported (Hanba et al., 2004; Flexas et al., 2006; Kaldenhoff et al., 2008; Perez-Martin et al. 2014), although it is still unclear whether this is a direct effect or a pleiotropic effect reflecting simultaneous increase in Amax, gm and gs (Flexas et al., 2012). Recent studies have demonstrated that an increase in gm has coincided with an increase in the amount of AQPs after fertilization (Miyazawa et al., 2008b; Zhu et al., 2020).
Soil moisture
Although many studies showed a decline of gm in response to soil water deficit (Flexas et al., 2009; Galle et al., 2009; Tosens et al., 2012a), it remains unclear if this limitation is happening within the mesophyll environment or occurs as a result of a stomatal limitation, which decreases intercellular CO2 (Ci). Théroux-Rancourt et al. (2015) showed that, in hybrid poplar, gm remained unchanged (~ 0.3 mol m−2 s−1) following soil drying (Ψleaf ~ −0.4 to −1.2 MPa), while gs decreased, until a threshold of gs ~ 0.15 mol m−2 s−1 from, which gm decreased significantly as well. In a trial on Quercus robur L. and Fraxinus angustifolia Vahl grown in field, Grassi and Magnani (2005) reported a concomitant decrease of both gs and gm in a dry year (Ψsoil ~ −1.7 MPa), compared with a wetter year (Ψsoil ~ −0.2 MPa). In P. tremula L., gm significantly declined when Ψleaf of saplings dropped from −0.3 to −0.7 MPa due to applied osmotic stress (Tosens et al., 2012a). Simultaneously, drought stress induced decrease in SLA accompanied with an increase in the cell wall thickness and a decrease in the chloroplast surface area exposed to intercellular air space per unit leaf area (Tosens et al., 2012a). Other studies have shown that biochemical changes induced by drought stress, like deactivation of aquaporins, could decrease CO2 diffusion to carboxylation sites in the chloroplast (Miyazawa et al., 2008a).
Adaptation to the local environment might be a key driver of gm variation among taxa, similarly to other morpho-physiological traits. Interspecific and intraspecific differences in gm from mesic versus xeric environments (Quercus spp. and Eucalyptus spp.) were reported by Zhou et al. (2014). Their study showed that gm, as well as gs, Vcmax and J of species from drier regions were less sensitive to water deficit which maintains transpiration and photosynthesis activity at higher rates under drought, compared to species from the mesic environment. Marchi et al. (2008) observed that structural protection of mesophyll cells had a priority over functional efficiency of photochemical mechanisms in Olea europaea L. (evergreen sclerophyllous) but not in Prunus persica L. (deciduous broadleaf), depending on age-related variation in mesophyll anatomy.
Conclusion and Future Directions
The present review shows that gm in Populus spp. varies predictably along light gradients and that it responds to changes in soil moisture and nutrient availability, but is not affected by metal concentration and increasing atmospheric CO2 concentration. Although metabolic processes noticeably influence the response of gm to environmental changes, physical constraints through leaf development and ageing need to be considered in scaling photosynthesis from leaf to canopy, and in breeding programs for high WUE. Because fast-growing Populus spp. trees are important players in combating climate change, mitigating carbon emissions to some extent, comparisons of genotypes with different adaptations to changing environments and breeding for novel genotype-climate associations are urgently needed. This study shows that the variability of gm in different experimental conditions offers a potential indicator for improving Populus spp. productivity and resilience. However, more research is yet needed, also combined with anatomical studies, to better understand the sources of variation of CO2 diffusion through the mesophyll and their consequences on carbon assimilation and growth.
Moreover, determination of the efficiency and optimal age for early selection of fast-growing poplar clones require an understanding of the genetic control and age-based genetic correlations for traits related to gm and growth. For that, a detailed evaluation of the genotypic control of the variances and clonal heritability of gm is needed. Finally, the identification of molecular bases of the regulation of gm is necessary to further refine a multi-criteria early selection approach of poplar clones dedicated to the future forestry capable of ensuring better productivity and increased resistance to environmental stresses (frost, drought, water logging, heavy metals, heat waves, etc.).
Conflict of interests
The authors declare that there is no conflict of interest.
Data Availability
The datasets generated for this study are available on request to the corresponding author.
Acknowledgements
This research was supported by the University of Québec in Abitibi-Témiscamingue (UQAT) as startup funds to ML. The authors acknowledge researchers who kindly provided data used in the meta-analysis.
Footnotes
Highlight Our meta-analysis on the variation of mesophyll conductance under abiotic stresses shows a noticeable response to light gradient, soil moisture and nitrogen availability and a significant relationship with specific leaf area