Introduction

More than 90% of vascular plant species in terrestrial ecosystems form mycorrhizal symbiosis between plant roots and soil fungi (Brundrett & Tedersoo, 2018). Arbuscular mycorrhizal (AM) and ectomycorrhizal (EcM) associations are the two main types of mycorrhizal symbiosis (Brundrett & Tedersoo, 2018), with both being broadly distributed across temperate forests (Phillips et al., 2013). There is a growing understanding that different mycorrhizal types are linked with distinct ecosystem processes, and thus mycorrhizal type can be considered as a key property predicting ecosystem functions (Tedersoo & Bahram, 2019). For example, multiple common garden studies have demonstrated that the leaf litter and roots of plants associated with EcM fungi decomposed more slowly than those associated with AM fungi when incubated under similar environmental conditions (Cornelissen et al., 2001; Midgley et al., 2015; Jacobs et al., 2018). In line with these studies, global-scale data syntheses have shown that the litter of plants associated with EcM fungi tended to have slower decomposition rates compared to that of AM fungi associated plants in temperate forests (See et al., 2019, but not in tropical biomes, Keller & Phillips, 2019). The leaf litter of plants associated with AM and EcM fungi also showed different chemical composition, with those of EcM associated plants exhibiting chemical traits that typically suppress decomposition (Lin et al., 2017; Peng et al., 2022). The mechanisms driving differences between mycorrhizal types in plant tissue chemistry and decomposition rates are still largely unclear (Jo et al., 2019).

Given that AM and EcM associations have very different structures within plant roots, it is possible that their physical organization has distinct impacts on plant chemical construction. Specifically, while EcM fungi are restricted to intercellular spaces, AM fungi penetrate cortical cells and can spread in a cell-to-cell manner (Brundrett, 2002). Because EcM fungi occupy the apoplastic space, the build-up of lignin and/or suberin in root cell walls can contribute to apoplastic barriers that limit unfavorable fungal spread (Melville et al., 1987; Massicotte et al., 1993; Enstone et al., 2002). By contrast, while AM fungi can penetrate primary cell walls (Smith & Read, 2008), they have been observed to be blocked from entering the tangential primary walls of the endodermis, suggesting that plants may depend on mechanisms other than cell wall barriers to prevent AM fungi from entering the stele (Harley & Smith, 1983; Brundrett & Kendrick, 1990; Enstone et al., 2002). In addition, extensive deposits of tannin-like phenolic compounds at the plant–fungus interface (e.g. between the EcM hyphal mantle and root epidermis) have been found in roots colonized by EcM fungi across many plant hosts (Mejstrik, 1972; Edwards & Gessner, 1984; Dunk et al., 2012; Stögmann et al., 2013). Given the well-documented inhibitory effects of lignin and tannins on litter decomposition (Harrison, 1971; Berg, 2000; Sun et al., 2018; See et al., 2019), the above observations of root structures suggest that, in comparison to AM associations, EcM associations are more likely to enhance the chemical recalcitrance of roots by increasing the levels of these phenolic heteropolymers, which could then contribute to the observed divergence of litter decomposability between mycorrhizal types.

On the other hand, genetic and transcriptional evidence suggests that forming AM or EcM associations may similarly influence plant chemical traits, particularly those involved in plant defense response. This is because both AM and EcM associated plants share a common evolutionary trend to dampen plant defense responses via the expansion of small-secreted protein families (Genre et al., 2020). Consequently, irrespective of mycorrhizal type, many defense-related genes and compounds (e.g. chitinase, peroxidase) were unaltered or downregulated in established mycorrhizal symbiosis (Münzenberger et al., 1997; Liu et al., 2003; Tarkka et al., 2013; Giovannetti et al., 2015). Interestingly, a set of phenolic heteropolymers frequently involved in plant defense against fungal invasion (e.g. lignin and condensed tannins, or CT, Scalbert, 1991; Miedes et al., 2014; Chen et al., 2021), also exert strong negative influence on litter decomposition (Berg, 2000; Sun et al., 2018; See et al., 2019). If the common genetic and transcriptional responses between AM and EcM associations translate to plant chemical alterations, then the formation of either mycorrhizal type would have similar effects on the abundance and composition of these defense-related phenolic heteropolymers, which are also important regulators of litter decomposition.

Taken together, previous observations suggest two potentially contrasting ways in which different mycorrhizal types might influence plant chemical traits related to litter decomposition. In this context, we hypothesize that EcM colonization would enhance plant chemical features that slow decomposition, while AM colonization would show an opposite trend, reflecting different strategies employed by plants associated with EcM and AM fungi to restrict unfavorable fungal spread. Alternatively, due to their shared symbiotic nature, AM and EcM associations may similarly influence plant tissue chemistry by either maintaining or decreasing the abundance of defense-related phenolic heteropolymers, reflecting the common tendency among mycorrhizal plants to suppress plant defense. We focused on the effects of mycorrhizal associations on the chemical composition of roots and leaves, given that mycorrhizal effects can be systemic, affecting both roots and leaves (Cordier et al., 1998; Jung et al., 2012). Determining whether alterations in the chemical composition of roots are also reflected in leaves is important since both forms of litter are significant inputs for organic matter decomposition in temperate forests (Gaudinski et al., 2000).

The chemical features of focus in this study include a set of plant chemical traits directly related to decomposition rates: lignin abundance and molecular composition, CT content, and the ratio of lignin and nitrogen (lignin : N ratio). Higher contents of lignin and CT, as well as higher lignin : N ratios, have been associated with slower root and leaf decomposition rates across a wide range of plant species (Melillo et al., 1982; Berg, 2000; Sun et al., 2018; See et al., 2019). Greater amounts of CT can also slow soil N mineralization (Verkaik et al., 2006; Tharayil et al., 2013). Differences in the proportional abundance of guaiacyl (G), syringyl (S) and p-hydroxyphenyl (H) units of polymeric lignin, often dictated by the species identity (Weng & Chapple, 2010; Dixon & Barros, 2019; Ralph et al., 2019; Vanholme et al., 2019), may further influence litter decomposability: G-rich lignin tends to be more resistant to biodegradation as G units can form aryl–aryl bonds resistant to depolymerization (Talbot et al., 2012), but a higher proportional abundance of S units has also been observed to enhance resistance to fungal decay (Skyba et al., 2013). To better understand the mechanisms underlying the potential lignin modifications by mycorrhizal fungi at the tissue level, we also visualized the autofluorescence characteristic of lignin to determine the spatial distribution of lignin in cross-sections of mycorrhizal fungi-colonized roots and their noncolonized counterparts (Kamiya et al., 2015). This imaging analysis helps to relate changes in lignin abundance to adjustments in lignin-rich structures such as the endodermis and xylem, as well as their associated functions in the presence of mycorrhizal associations.

We tested the proposed hypotheses by characterizing the chemistry of roots and leaves in noncolonized and mycorrhizal fungi-colonized seedlings in various combinations of plant hosts and mycorrhizal fungal species, forming either AM or EcM associations. Two angiosperm and two gymnosperm species, with one in each group forming AMs and the other forming EcM associations, were included to compare the potential effects of plant phylogenetic relationship vs mycorrhizal type in shaping tissue chemistry (Table 1). The plant and fungal species were selected to be representative of temperate forest trees and mycorrhizal fungi, as well as to encompass a large phylogenetic and functional diversity. Our results suggest that the mycorrhiza-associated changes in plant chemical traits that regulate litter decomposition may not be unique to AM or EcM associations; rather, both associations can reduce root and leaf chemical recalcitrance. Further, changes in lignin molecular composition by mycorrhizal symbiosis differed between plant phylogenetic lineages irrespective of mycorrhizal type, highlighting the influence of plant evolutionary history in plant–mycorrhizal interactions.

Table 1. Mycorrhizal colonization rate, seedling height, and nitrogen (N) contents of roots and leaves from mycorrhizal fungi-colonized and noncolonized (NM) seedlings.

Host tree species	Fungal inoculum	Fungus abbr.	Individual seedlings	Colonization (%)	Height (cm)	N (%, root)	N (%, leave)
AM
Juniperus virginiana (gymnosperm)	Noncolonized	NM	9	0a	2.31a ± 0.26	0.95a ± 0.09	1.67a ± 0.12
	Funneliformis mosseae	Fm	9	70.4b ± 6.1	8.51c ± 0.21	1.59c ± 0.14	2.15b ± 0.38
	Gigaspora margarita	Gm	9	66.4b ± 5.3	6.67b ± 0.87	1.38b ± 0.13	1.58a ± 0.12
Liriodendron tulipifera (angiosperm)	Noncolonized	NM	8	0a	2.43a ± 0.29	2.36b ± 0.24
	Funneliformis mosseae	Fm	9	75.4b ± 10.7	6.36b ± 2.15	1.73a ± 0.27
	Gigaspora margarita	Gm	5	74.8b ± 5.0	4.48b ± 0.05	1.64a ± 0.33
EcM
Pinus taeda (gymnosperm)	Noncolonized	NM	11	0a	4.67a ± 0.42	1.19 ± 0.08	1.26a ± 0.29
	Suillus brevipes	Sb	10	44.8c ± 3.8	6.34c ± 0.81	1.08 ± 0.08	1.48ab ± 0.15
	Suillus luteus	Sl	3	6.00b ± 2.8	5.55bc ± 0.88	0.96 ± 0.10	1.63ab ± 0.11
	Sordariales sp.	Ss	5	74.6d ± 23.9	4.88ab ± 1.01	1.04 ± 0.31	1.68b ± 0.41
Quercus macrocarpa (angiosperm)	Noncolonized	NM	8	0a	7.38 ± 1.95	1.15a ± 0.17	1.93 ± 0.28
	Tuber sp.	Ts	6	36.4b ± 20.6	7.12 ± 1.03	1.48b ± 0.19	1.76 ± 0.36

• The values are shown as means (SD) of three trays (blocking factor) of each plant–fungus combination or their noncolonized controls. Three trays were included in the analysis for most plant–fungus combinations, with each tray containing one to four individual seedlings, resulting in a total of 6–11 individual seedlings for each combination (Supporting Information Table S1). However, P. taeda × Sl and P. taeda × Ss pairings only occurred in one tray, for which the means (SD) of individual seedlings are shown. AM, arbuscular mycorrhizal; EcM, ectomycorrhizal. Different letters in the same column indicate significant differences across mycorrhizal treatments within each plant species (post hoc Tukey HSD tests, P < 0.05, see details of statistical results in Table S2).

中文翻译：

---

菌根共生的形成改变了四种温带木本植物根和叶中的酚类杂聚物

介绍

陆地生态系统中超过 90% 的维管植物物种在植物根部和土壤真菌之间形成菌根共生（Brundrett & Tedersoo，  2018）。丛枝菌根（AM）和外生菌根（EcM）关联是菌根共生的两种主要类型（Brundrett＆Tedersoo，  2018），两者都广泛分布在温带森林中（Phillips等，  2013）。人们越来越认识到不同的菌根类型与不同的生态系统过程相关，因此菌根类型可以被视为预测生态系统功能的关键属性（Tedersoo＆Bahram，  2019）。例如，多项常见花园研究表明，在相似的环境条件下培养时，与 EcM 真菌相关的植物的叶凋落物和根部的分解速度比与 AM 真菌相关的植物慢（Cornelissen等人，  2001 年；Midgley等人，  2015 年）雅各布斯等人，  2018）。根据这些研究，全球范围的数据综合表明，与温带森林中与 AM 真菌相关的植物相比，与 EcM 真菌相关的植物凋落物的分解速度往往更慢（参见et al .,  2019，但没有热带生物群落，Keller & Phillips，  2019）。与 AM 和 EcM 真菌相关的植物的叶凋落物也表现出不同的化学成分，其中与 EcM 真菌相关的植物的叶凋落物表现出通常抑制分解的化学特征（Lin等，2017；Peng等，  2022）。植物组织化学和分解速率方面菌根类型之间差异的驱动机制仍不清楚（Jo et al .,  2019）。

鉴于 AM 和 EcM 关联在植物根部内具有非常不同的结构，因此它们的物理组织可能对植物化学结构具有不同的影响。具体来说，虽然 EcM 真菌仅限于细胞间隙，但 AM 真菌穿透皮质细胞并可以以细胞间的方式传播（Brundrett，  2002）。由于 EcM 真菌占据质外体空间，根细胞壁中木质素和/或木栓质的积聚可能有助于限制不利的真菌传播的质外体屏障（Melville等人，  1987；Massicotte等人，  1993；Enstone等人） .,  2002 )。相比之下，虽然 AM 真菌可以穿透初生细胞壁（Smith & Read，  2008），但观察到它们无法进入内皮层的切向初生壁，这表明植物可能依赖于细胞壁屏障以外的机制来阻止AM 真菌进入石碑（Harley & Smith，  1983；Brundrett & Kendrick，  1990；Enstone等人，  2002）。此外，在许多植物宿主中被 EcM 真菌定植的根中发现了植物-真菌界面（例如 EcM 菌丝套和根表皮之间）大量沉积的类单宁酚类化合物（Mejstrik，  1972；Edwards & Gessner，  1984；邓克等人，  2012；斯托格曼等人，2013）。鉴于木质素和单宁对凋落物分解的抑制作用已有充分记录（Harrison，  1971；Berg，2000；Sun等人，  2018；See等人，  2019），上述对根结构的观察表明，与AM 关联、EcM 关联更有可能通过增加这些酚类杂聚物的水平来增强根的化学不顺应性，这可能有助于观察到的菌根类型之间凋落物分解性的差异。

另一方面，遗传和转录证据表明形成 AM 或 EcM 关联可能类似地影响植物化学性状，特别是那些涉及植物防御反应的化学性状。这是因为 AM 和 EcM 相关植物具有共同的进化趋势，即通过扩展小分泌蛋白家族来抑制植物防御反应（Genre等，  2020）。因此，无论菌根类型如何，许多防御相关基因和化合物（例如几丁质酶、过氧化物酶）在已建立的菌根共生中没有改变或下调（Münzenberger等，  1997；Liu等，  2003；Tarkka等，  2013； Giovannetti等人，  2015）。有趣的是，一组经常参与植物防御真菌入侵的酚类杂聚物（例如木质素和缩合单宁，或 CT，Scalbert，  1991；Miedes等，  2014；Chen等，  2021），也对植物产生强烈的负面影响。凋落物分解（Berg，  2000；Sun et al .，  2018；See et al .，  2019）。如果AM和EcM关联之间共同的遗传和转录反应转化为植物化学变化，那么任何一种菌根类型的形成都会对这些与防御相关的酚类杂聚物的丰度和组成产生类似的影响，这些酚类杂聚物也是凋落物分解的重要调节剂。

综上所述，之前的观察结果表明，不同的菌根类型可能会影响与凋落物分解相关的植物化学性状，有两种可能相反的方式。在这种情况下，我们假设 EcM 定植会增强植物的化学特征，从而减缓分解，而 AM 定植将表现出相反的趋势，反映出与 EcM 和 AM 真菌相关的植物采用不同的策略来限制不利的真菌传播。或者，由于它们共有的共生性质，AM 和 EcM 关联可能通过维持或减少与防御相关的酚类杂聚物的丰度来类似地影响植物组织化学，反映了菌根植物抑制植物防御的共同趋势。鉴于菌根效应可能是系统性的，同时影响根和叶，我们重点关注菌根关联对根和叶化学成分的影响（Cordier等，  1998；Jung等，2012）。确定根化学成分的变化是否也反映在叶子中非常重要，因为这两种形式的凋落物都是温带森林有机物分解的重要输入（Gaudinski等，  2000）。

本研究重点关注的化学特征包括一组与分解速率直接相关的植物化学性状：木质素丰度和分子组成、CT含量以及木质素与氮的比率（木质素：N比率）。较高的木质素和 CT 含量，以及较高的木质素：氮比，与多种植物物种较慢的根和叶分解速率有关（Melillo等人，  1982 年；Berg，  2000 年；Sun等人，  2018；参见等人，  2019）。大量的 CT 还可以减缓土壤氮矿化（Verkaik等人，  2006 年；Tharayil等人，  2013 年）。聚合木质素的愈创木基 (G)、紫丁香基 (S) 和对羟基苯基(H) 单元的比例丰度差异，通常由物种特性决定（Weng & Chapple，  2010；Dixon & Barros，  2019；Ralph等人，2019） 。 ，  2019；Vanholme等，  2019），可能会进一步影响凋落物的分解性：富含 G 的木质素往往更能抵抗生物降解，因为 G 单元可以形成抗解聚的芳基-芳基键（Talbot等，  2012），但是还观察到较高比例的 S 单元丰度可以增强对真菌腐烂的抵抗力（Skyba等，  2013）。为了更好地了解菌根真菌在组织水平上潜在木质素修饰的机制，我们还可视化了木质素的自发荧光特征，以确定木质素在菌根真菌定植根及其非定植根横截面中的空间分布（Kamiya等人）等，  2015）。这种成像分析有助于将木质素丰度的变化与富含木质素的结构（例如内皮层和木质部）的调整以及它们在存在菌根关联的情况下的相关功能联系起来。

我们通过表征植物宿主和菌根真菌物种的各种组合中的非定殖和菌根真菌定殖幼苗的根和叶的化学特性来测试所提出的假设，形成AM或EcM关联。包括两种被子植物和两种裸子植物物种，每组中一种形成 AM，另一种形成 EcM 关联，​​以比较植物系统发育关系与菌根类型在塑造组织化学方面的潜在影响（表 1）。选择的植物和真菌物种是温带森林树木和菌根真菌的代表，并且包含大量的系统发育和功能多样性。我们的结果表明，调节凋落物分解的植物化学性状的菌根相关变化可能并非 AM 或 EcM 关联所独有。相反，这两种关联都可以减少根和叶的化学抗逆性。此外，无论菌根类型如何，菌根共生引起的木质素分子组成的变化在植物系统发育谱系之间存在差异，凸显了植物进化史对植物-菌根相互作用的影响。

表 1.菌根真菌定植和未定植 (NM) 幼苗的菌根定植率、幼苗高度以及根和叶的氮 (N) 含量。

寄主树种	真菌接种物	真菌缩写。	单株苗	定植率 (%)	身高（厘米）	N（%，根）	N（%，离开）
是
弗吉尼亚杜松（裸子植物）	非殖民化	纳米	9	0个	2.31 ±  0.26	0.95 ±  0.09	1.67 ±  0.12
	苔藓漏斗形虫	调频	9	70.4b ±  6.1	8.51 ℃  ±0.21	1.59 ℃  ±0.14	2.15b ±  0.38
	玛格丽特大孢霉	GM	9	66.4b ±  5.3	6.67b ±  0.87	1.38b ±  0.13	1.58 ±  0.12
鹅掌楸（被子植物）	非殖民化	纳米	8	0个	2.43 ±  0.29	2.36b ±  0.24
	苔藓漏斗形虫	调频	9	75.4b ±  10.7	6.36b ±  2.15	1.73 ±  0.27
	玛格丽特大孢霉	GM	5	74.8b ±  5.0	4.48b ±  0.05	1.64 ±  0.33
ECM
火炬松（裸子植物）	非殖民化	纳米	11	0个	4.67 ±  0.42	1.19±0.08	1.26 ±  0.29
	短叶牛肝菌	锑	10	44.8 ℃  ±3.8	6.34 ℃  ±0.81	1.08±0.08	1.48 ab  ±0.15
	牛肝菌	斯尔	3	6.00b ±  2.8	公元前5.55  ± 0.88	0.96±0.10	1.63 ab  ±0.11
	索达利亚目sp.	SS	5	74.6天 ±23.9	4.88 ab  ± 1.01	1.04±0.31	1.68b ±  0.41
大果栎（被子植物）	非殖民化	纳米	8	0个	7.38±1.95	1.15 ±  0.17	1.93±0.28
	块茎属植物。	TS	6	36.4b ±  20.6	7.12±1.03	1.48b ±  0.19	1.76±0.36

• 这些值显示为每个植物-真菌组合或其非定植对照的三个托盘（封闭因子）的平均值（SD）。大多数植物-真菌组合的分析中包括三个托盘，每个托盘包含一到四个单独的幼苗，每个组合总共有 6-11 个单独的幼苗（支持信息表 S1）。然而，火炬松 × Sl 和火炬松 × Ss 配对仅发生在一个托盘中，其中显示了单个幼苗的平均值（SD）。 AM，丛枝菌根； EcM，外生菌根。同一列中的不同字母表示各植物物种内菌根处理之间存在显着差异（事后Tukey HSD检验，P  <0.05，统计结果详情见表S2）。