Introduction

The supply of CO₂ to the photosynthetic machinery depends on how efficiently it can be transferred from the ambient air to the chloroplasts located inside the leaf mesophyll cells. This efficiency can be quantified as a series of resistances (or the inverse quantity, conductances) caused by the leaf boundary layer, the stomata, as well as leaf internal components in the mesophyll. This last part of the CO₂ pathway, the mesophyll conductance (g_m), accounts for one-third to one-half of the overall CO₂ drawdown from the atmosphere to the chloroplasts (Warren, 2008; Flexas et al., 2012) and therefore constitutes a major controlling factor of the CO₂ concentration available for photosynthesis. Knowledge of the determinants of g_m can therefore support efforts aiming to improve photosynthesis to ensure that global food and bioenergy demand can be met in the future (von Caemmerer & Evans, 2010; Ort et al., 2015). Furthermore, information of how g_m is related to key leaf structural and biochemical traits is important for understanding and modelling carbon uptake from the leaf to the global scale (Niinemets et al., 2009; Sun et al., 2014; Knauer et al., 2019, 2020).

The pathway of CO₂ within plant leaves can be divided into several components, which in combination determine the magnitude of g_m: the intercellular airspaces, the cell wall, the plasma membrane, the cytosol, the chloroplast envelope, and the chloroplast stroma (Niinemets & Reichstein, 2003; Evans et al., 2009). Some of the conductances within these components depend primarily on biophysical characteristics (e.g. surface area of chloroplasts exposed to intercellular airspaces, cell wall thickness and porosity) and are therefore subject to anatomical constraints, whereas CO₂ transfer through other cell compartments such as membranes and the cytosol are primarily the result of biochemical factors, in particular the expression of proteins associated with CO₂ transfer. These include aquaporins (cooporins), proteins that regulate water and CO₂ transfer across membranes (Uehlein et al., 2003), and carbonic anhydrase (CA), which governs the interconversion between CO₂ and bicarbonate in the cytosol and chloroplast stroma (Fabre et al., 2007; Evans et al., 2009). Despite the fact that it is well established that g_m is affected by both anatomical and biochemical leaf traits (Warren, 2008; Flexas et al., 2012, 2018; Gago et al., 2020), their relative contribution across plant functional types (PFTs) has not yet been assessed.

The complexity of the CO₂ diffusion pathway within leaves results in considerable uncertainties regarding the contributions of the individual components to the overall conductance as well as the associated importance of key anatomical and biochemical traits. One possible avenue to elucidate the role of certain leaf traits in determining g_m are gas diffusion models that calculate the component conductances based on biophysical and biochemical principles (Niinemets & Reichstein, 2003; Tomás et al., 2013; Berghuijs et al., 2015; Xiao & Zhu, 2017). However, these models either do not take all relevant mechanisms into account (e.g. biochemistry, location of individual elements of the diffusion pathway) or require parameters that are unknown or only available for a few species, which hinders the interpretation of these models as well as their application across PFTs.

An alternative approach followed by many studies is to use correlation analysis to investigate to what extent g_m measurements are related to leaf anatomical and biochemical traits. However, most studies are restricted to one or a few species of the same PFT and are subject to differences in growth environments, measurement conditions, as well as assumptions and uncertainties inherent in different measurement approaches (Pons et al., 2009). These differences can hamper a direct comparison between individual studies and preclude robust conclusions. In addition, correlations can only provide associative rather than causal relationships between g_m and leaf traits. Despite these limitations, a correlative approach can provide information about key traits covarying with g_m and therefore highlight the trait syndromes responsible for the variation in g_m, especially if relationships emerge across studies, species, and conditions (e.g. Xiong & Flexas, 2018; Ren et al., 2019; Elferjani et al., 2021).

Here, we present the hitherto largest published dataset of g_m measurements compiled from the literature (comprising 563 studies). We performed a comprehensive analysis that aimed to investigate the relationships between g_m and accompanying leaf structural, anatomical, biochemical and physiological traits measured on the same set of plants. The overarching goal of this top-down approach was to identify patterns between g_m and leaf traits that are robust with respect to existing confounding effects of different species and genotypes, growth conditions, or methodological considerations, and that may guide future research priorities. In particular, we asked (1) how much g_m limits photosynthesis across PFTs, (2) to what extent leaf anatomical and biochemical factors can explain variations in g_m across and within PFTs, and (3) how our findings could be used to enhance g_m and photosynthesis.

中文翻译：

---

比较植物功能类型对叶肉电导的解剖学和生化控制

介绍

CO ₂对光合机器的供应取决于它从环境空气转移到叶肉细胞内叶绿体的效率。这种效率可以量化为由叶片边界层、气孔以及叶肉中的叶片内部成分引起的一系列阻力（或反量，电导）。_{CO 2}途径的最后一部分，即叶肉电导率 ( g _m )，占从大气到叶绿体的 CO ₂总消耗量的三分之一到二分之一（Warren， 2008 年；Flexas等人，  2012 年）因此构成了CO的主要控制因素₂浓度可供光合作用。_{因此，了解gm}的决定因素可以支持旨在改善光合作用的努力，以确保未来能够满足全球粮食和生物能源需求（von Caemmerer 和 Evans，  2010年；Ort等人，  2015 年）。此外，有关g _m如何与关键叶片结构和生化特性相关的信息对于理解和模拟从叶片到全球范围内的碳吸收非常重要（Niinemets等人，  2009 年；Sun等人，  2014 年；Knauer等人，2014 年） 。 ,  2019 ,2020 年）。

CO ₂在植物叶片内的途径可分为几个组成部分，它们共同决定了g _m的大小：细胞间空间、细胞壁、质膜、细胞质、叶绿体包膜和叶绿体基质 (Niinemets & Reichstein，  2003 年；埃文斯等人，  2009 年）。这些成分中的一些电导主要取决于生物物理特性（例如暴露于细胞间空间的叶绿体表面积、细胞壁厚度和孔隙率），因此受到解剖学限制，而 CO ₂通过其他细胞区室（例如膜和胞质溶胶）的转移主要是生化因素的结果，特别是与 CO ₂转移相关的蛋白质的表达。这些包括水通道蛋白（cooporins）、调节水和 CO ₂跨膜转移的蛋白质（Uehlein等人，  2003 年）和碳酸酐酶 (CA)，它控制胞质溶胶和叶绿体基质中 CO ₂和碳酸氢盐之间的相互转化（Fabre等人，  2007 年；埃文斯等人，  2009 年）。尽管事实上g _m受解剖学和生化叶片性状的影响（Warren，  2008 年；Flexas等人，  2012 年，2018 年；Gago等人，  2020 年），它们在植物功能类型 (PFT) 中的相对贡献尚未得到评估。

叶子内CO ₂扩散途径的复杂性导致关于各个成分对整体电导的贡献以及关键解剖学和生化特征的相关重要性的相当大的不确定性。阐明某些叶片性状在确定g _m中的作用的一种可能途径是气体扩散模型，该模型根据生物物理和生化原理计算成分电导（Niinemets 和 Reichstein，  2003 年；Tomás等人，  2013 年；Berghuijs等人，  2015 年） ; 肖 & 朱,  2017). 然而，这些模型要么没有考虑所有相关机制（例如生物化学、扩散途径的各个元素的位置），要么需要未知或仅适用于少数物种的参数，这阻碍了对这些模型的解释以及它们在 PFT 中的应用。

许多研究采用的另一种方法是使用相关性分析来研究gm测量值与叶片解剖学和生化性状相关的_程度。然而，大多数研究仅限于同一 PFT 的一个或几个物种，并且受生长环境、测量条件以及不同测量方法固有的假设和不确定性的影响（Pons等，  2009）。这些差异会妨碍个别研究之间的直接比较，并排除可靠的结论。此外，相关性只能提供g _{m之间的关联关系而不是因果关系}和叶性状。尽管存在这些局限性，但相关方法可以提供与g _m共变的关键性状的信息，因此突出了导致g _m变化的性状综合症，特别是如果跨研究、物种和条件出现关系时（例如 Xiong & Flexas，  2018 年； Ren等人，  2019 年；Elferjani等人，  2021 年）。

在这里，我们展示了迄今为止最大的已发表的g _m测量数据集，这些数据集是根据文献（包括 563 项研究）编制的。我们进行了一项综合分析，旨在研究g _m与在同一组植物上测量的伴随叶片结构、解剖学、生化和生理性状之间的关系。这种自上而下方法的总体目标是确定g _m和叶片性状之间的模式，这些模式对于不同物种和基因型、生长条件或方法学考虑的现有混杂效应而言是稳健的，并且可以指导未来的研究重点。特别地，我们问 (1) 多少g _m限制 PFT 间的光合作用，(2) 叶片解剖学和生化因素在多大程度上可以解释PFT 间和内部g _m的变化，以及 (3) 我们的发现如何用于增强g _m和光合作用。