This is the accepted version of the journal article: Li, Tengteng; Yuan, Ye; Mou, Zhijian; [et al.]. «Faster accumulation and greater contribution of glomalin to the soil organic carbon pool than amino sugars do under tropical coastal forest restoration». Global change biology, Vol. 29, issue 2 (Jan. 2023), p. 533-546. DOI 10.1111/gcb.16467 This version is available at https://ddd.uab.cat/record/284321 under the terms of the license 1 Faster accumulation and greater contribution of glomalin to the soil organic carbon pool than 2 amino sugars do under tropical coastal forest restoration 3 Tengteng Li1,2,3†, Ye Yuan1,4†, Zhijian Mou1,4, Yue Li1,2,3, Luhui Kuang1,4, Jing Zhang1,2,3, Wenjia Wu1,2,3, 4 Faming Wang1,2,3, Jun Wang1,2, Hans Lambers5, Jordi Sardans6,7, Josep Peñuelas6,7, Hai Ren1,2, Zhanfeng 5 Liu1,2,3* 1 6 Key Laboratory of Vegetation Restoration and Management of Degraded Ecosystems & CAS 7 Engineering Laboratory for Vegetation Ecosystem Restoration on Islands and Coastal Zones, South 8 China Botanical Garden, Chinese Academy of Sciences, Guangzhou 510650, China 2 9 Guangdong Provincial Key Laboratory of Applied Botany, South China Botanical Garden, Guangzhou 10 510650, China 3 11 Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou), Guangzhou 511458, 12 China 4 13 College of Resources and Environment, University of Chinese Academy of Sciences, Beijing 100049, 14 China 5 15 School of Biological Sciences, University of Western Australia, Perth, WA 6009, Australia 6 16 CSIC, Global Ecology Unit CREAF-CSIC-UAB, Cerdanyola del Valles, Catalonia 08193, Spain; 7 17 CREAF, Cerdanyola del Valles, Catalonia 08193, Spain; 18 † 19 These authors contributed equally to this work. 20 *Corresponding author: 21 Zhanfeng Liu Email: liuzf@scbg.ac.cn; Tel: +86-20-37252592 22 Running title: Afforestation and microbial-derived products 23 Type of article: Primary research article 24 Number of words: 5640 words in the main text 25 Number of figures and tables: 7 figures and 2 tables 26 Supporting information: 2 figures 27 28 29 30 1 31 Abstract 32 Microbial metabolic products play a vital role in maintaining ecosystem multifunctionality, such as soil 33 physical structure and soil organic carbon (SOC) preservation. Afforestation is an effective strategy to 34 restore degraded land and mitigate climate change. Glomalin-related soil proteins and amino sugars are 35 regarded stable microbial-derived C, and their distribution within soil aggregate fractions affects soil 36 structure stability and SOC sequestration. However, the information about how afforestation affects the 37 microbial contribution to SOC pools within aggregates is poorly understood. We assessed the 38 accumulation and contribution of glomalin-related soil proteins (GRSP, produced by arbuscular 39 mycorrhizal fungi) and amino sugars (originating from microbes) within soil aggregate fractions along a 40 restoration chronosequence (Bare land (BL), Eucalyptus exserta plantation (EP), native species mixed 41 forest (MF) and native forest (NF)) in tropical coastal terraces. The concentrations of amino sugars and 42 GRSP increased, whereas their contributions to the SOC pool decreased along the restoration 43 chronosequence. Although microaggregates harbored greater microbial abundances, the concentrations 44 of amino sugars and GRSP were not significantly affected by aggregate sizes. Interestingly, the 45 contributions of amino sugars and GRSP to SOC pools decreased with decreasing aggregate fraction size 46 which might be associated with increasing accumulation of plant-derived carbon. However, the relative 47 change rate of GRSP was consistently greater in all restoration chronosequences than that of amino 48 sugars. The accumulation of GRSP and amino sugars in SOC pools was closely associated with the 49 dynamics of soil fertility and the microbial community. Our findings suggest that glomalin accumulates 50 faster and contributes more to SOC pools during forest restoration than amino sugars did which was 51 greatly affected by soil aggregate size. Afforestation substantially enhanced soil quality, with native 52 forest comprising species sequestering more SOC than the monoculture plantation did. Such information 53 is invaluable for improving our mechanistic understanding of microbial control over SOC preservation 54 during degraded ecosystem restoration. Our findings also show that plantations using arbuscular 55 mycorrhizal plants can be an effective practice to sequester more soil carbon during restoration. 56 Keywords: glomalin-related soil protein, amino sugars, soil aggregates, soil microbial community, 57 afforestation, tropical coastal terrace 58 59 60 2 61 1. INTRODUCTION 62 Forests play an important role in maintaining coastal ecosystem functioning such as biodiversity, typhoon 63 sheltering and carbon sequestration (Snäll et al., 2021). However, about 25-50% of coastal vegetation 64 habitats (such as mangroves, seagrasses, salt-marshes, kelp forests) have become degraded worldwide in 65 the past 50 years due to land-use and climate change, which are threatening their health and causing 66 major CO2 emission (Duarte et al., 2013). Of the natural coastal habitats in China, more than 50% have 67 been lost in the past 60 years (Ma et al., 2014), specifically as a result of severe forest degradation in 68 south China in the early 1950s, affecting ecosystem functioning (Ren et al., 2007). Forest restoration is 69 an effective way to mitigate the degradation and improve its ecosystem services (Canadell & Raupach, 70 2008). Soil carbon (C) cycling is central to the reestablishment of belowground ecosystem structure and 71 functioning during restoration processes; however, most previous studies focused on the aboveground 72 biodiversity and its functioning, and less attention has been paid to the microbial role in mediating soil 73 C preservation during forest restoration (Capellesso et al., 2021; Crouzeilles et al., 2016; Hisano et al., 74 2018). 75 As the largest terrestrial C pool, soils contain more C than the vegetation and atmosphere combined 76 and play a vital role in the terrestrial C cycle and climate change mitigation (Lal et al., 2021). The 77 formation and stabilization of soil organic carbon (SOC) is a complex process of microbial 78 transformation of plant residues via ex vivo (mainly exoenzymatic decomposition, catabolism) 79 modification and in vivo (anabolism) turnover pathways (Liang, 2020). The process is affected by various 80 biotic and abiotic factors (Jackson et al., 2017). For example, plant diversity could increase soil microbial 81 activity and soil C storage (Lange et al., 2015); yet, the stability of SOC is affected by tree species and 82 characteristics through the composition of their aboveground organs and roots (Angst et al., 2019). The 83 physical protection by soil aggregates and the formation of organo-mineral associations are thought to 84 stabilize SOC pools (Liang et al., 2020; Schmidt et al., 2011; Wang et al., 2017). Due to the variation of 85 physicochemical conditions among aggregate fraction sizes (Rillig et al., 2017), greater microbial 86 diversity and potential functions are associated with microaggregates than with macroaggregates (Bach 87 et al., 2018; Navas et al., 2021; Upton et al., 2019) which may result in the accumulation of different 88 microbial-derived C within them (Murugan et al., 2019). Specifically, microbial metabolic products such 89 as amino sugars and glomalin-related soil proteins, have been widely investigated for their effects on 90 SOC preservation (Irving et al., 2021; Joergensen, 2018) due to their relatively greater recalcitrance and 3 91 benefit for soil aggregation (Agnihotri et al., 2022; Buckeridge et al., 2020). However, the information 92 related to their accumulation and contribution to SOC during vegetation restoration is not well 93 understood. 94 Soil microbial community is a core driver of SOC transformation and is sensitive to land 95 degradation and climate change (Xiao et al., 2016). Phospholipid fatty acids (PLFAs) are widely used as 96 biomarkers to indicate the ‘fingerprint’ profile of the living soil microbial community (Vestal & White, 97 1989; Zelles, 1999). By contrast, amino sugars are primarily derived from the cell walls of dead microbes 98 and are commonly used as biomarkers to quantify the contribution of microbial necromass to SOC 99 accumulation (Liang, 2019). Glucosamine is present in both fungal and bacterial cell walls, whereas 100 muramic acid exclusively occurs in the cell walls of bacteria (Joergensen, 2018). As microbial metabolic 101 products, amino sugars reflect a time-integrated microbial community (Glaser et al., 2004). The ratio of 102 total amino sugars to total PLFAs provides some information on the microbial necromass C- 103 transformation efficiency (Xu et al., 2022). Furthermore, there is no consensus on the contribution of 104 necromass to the SOC pool during vegetation restoration (Guo et al., 2021). A recent study showed that 105 the concentrations of PLFAs and amino sugars respond differently to forest restoration in a subtropical 106 region with divergent contributions of fungi or bacteria to the SOC pool via physical protection pathways 107 (Zhang et al., 2021). Especially microbial communities and biomasses might vary with aggregate sizes 108 due to their difference in microhabitats (Gupta and Germida, 2015), hence influencing the accumulation 109 of amino sugars in soil aggregates. Further exploration is needed to clarify the role of soil aggregates on 110 the accumulation and contribution of amnio sugars in the SOC pool during vegetation restoration 111 (Murugan et al., 2019). 112 Glomalin-related soil proteins (GRSP) are microbial products produced by arbuscular mycorrhizal 113 fungi (AMF) and characterized as hydrophobic sticky and recalcitrant glycoproteins (Wright et al., 1998). 114 GRSP are composed of a broad range of elements (e.g., C/N/H/O/Fe/Al), functional groups (e.g., 115 aromatic- and carboxyl-C), and composite substances such as proteins and carbohydrates (Agnihotri et 116 al., 2022). Recently produced GRSP (EE-GRSP, easily extractable GRSP) are more labile in soil than 117 total GRSP (T-GRSP) (Wright & Upadhyaya, 1996). GRSP generally increase with AMF colonization 118 and biomass (Agnihotri et al., 2021) during vegetation restoration (Qiao et al., 2019) and is affected by 119 land-use change, nutrient availability and tillage (Agnihotri et al., 2022). The turnover of glomalin is 120 slower than that of AMF hyphae (Rillig et al. 2001). Zhang et al. (2017; 2022) found that aromatic and 4 121 alkyl-C in glomalin are more recalcitrant, with benefits for aggregate stability, jointly enhancing SOC 122 persistence in tropical forests. Iron is an important element in the composition of GRSP, converting 123 monomeric GRSP units into a multimeric complex, thus promoting GRSP stabilization. Other metal ions 124 such as Al3+, Ca2+ and Mg2+ probably have similar effects on GRSP. GRSP may contribute to SOC content 125 not only owing to their recalcitrant chemistry, but also due to their stabilizing effect on soil aggregates 126 (Rillig and Mummey, 2006). GRSP sorbed onto organic substances, clays and silt particles, facilitates 127 adsorption between and within microaggregates and AMF hyphae could bind particles or 128 microaggregates, thereby promoting the formation and stabilization of soil macroaggregates (Agnihotri 129 et al., 2022). The distribution of GRSP in soil macroaggregates may influence aggregate stability (Xie et 130 al., 2015). Although it is acknowledged that both GRSP and amino sugars play important roles in 131 mediating SOC sequestration, the information about the dynamics of their relative contribution to the 132 SOC pool during forest restoration is poorly understood. Simultaneous measurements of GRSP and 133 amino sugars make it possible to compare their relative contribution to SOC accumulation and stability, 134 and provide valuable information for developing a restoration strategy in terms of carbon sequestration. 135 To obtain a comprehensive understanding of microbial-driven SOC preservation during vegetation 136 restoration, it is worth investigating how GRSP and amino sugars accumulate and contribute to SOC 137 across soil aggregate fraction sizes during restoration. 138 Coastal vegetation habitats have been degraded extensively in south China and this has caused a 139 series of ecological problems, such as biodiversity loss and soil erosion (Ren et al., 2007). Starting from 140 the 1950s, forest restoration was conducted by planting pioneer plant species and mixtures of native plant 141 species on bare land on tropical coastal terraces (Ren et al., 2007). After 60 years of afforestation, native 142 species mixtures have recovered and restored plant communities, soil biodiversity and soil fertility (Wu 143 et al., 2021). However, less attention was paid to how microbial metabolic products accumulate and 144 contribute to the SOC pool, which is central to the reestablishment of soil structure and functioning 145 during restoration. Such information is valuable for guiding restoration practices in the study area and 146 mitigating climate change . In this study, we aimed to investigate the accumulation dynamics of glomalin 147 and amino sugars within aggregates and to evaluate their relative contribution to the SOC pool following 148 forest restoration. We hypothesized that: (1) greater accumulation of GRSP contrasted with that of amino 149 sugars during forest restoration due to enhanced plant C inputs and microbial transformation (Guo et al., 150 2021; Qiao et al., 2019); (2) the accumulation rate and contribution of GRSP to the SOC pool would be 5 151 greater than those of amino sugars due to the differences in their chemical recalcitrance and propensity 152 to mineral protection (Agnihotri et al., 2022; Rillig et al., 2001). 153 154 2. MATERIALS AND METHODS 155 2.1. Site description and experimental design 156 The present study was carried out at the Xiaoliang Tropical Coastal Ecosystem Research Station, Chinese 157 Academy of Sciences, in Guangdong province of China (21°27′N; 110°54′E). The climate in this region 158 is typical tropical monsoon, with a mean annual temperature of 23°C and mean annual precipitation of 159 1,400-1,700 mm with wet (Apr.-Sept.) and dry (Nov.-Mar.) seasons. The soil is classified as a latosol that 160 originated from granite (Yao et al., 1984). Evergreen broad-leaved seasonal rainforest was the climax 161 vegetation in this region before the 1850s, while soils became severely degraded resulting from massive 162 deforestation and soil erosion by the 1950s and only a small part of native forest (NF) was protected for 163 more than 200 years (Yu et al., 1985; Ren et al., 2007). A Eucalyptus exserta plantation (EP) was 164 established on bare land in the early 1960s; it can associate with both arbuscular and ectomycorrhizal 165 fungi. Forests using mixed native species (MF) were promoted from EP after clear-cutting in 1974 and 166 are dominated by AMF plants including Carallia brachiata, Aphanamixis polystachya, Schefflera 167 octophylla, Carallia brachiata, Symplocos chunii, Acacia auriculaeformis, Photinia benthamiana, and 168 Cinnamomum burmanni, Lygodium japonicum, Ophiopogon japonicus and Nephrolepis cordifolia, and 169 ectomycorrhizal plants including Acacia auriculaeformis and Calamus tetradactylu. NF is dominated by 170 AMF plants inculding Sterculia lanceolata, Cinnamomum camphora, Cryptocarya chinensis, Syzygium 171 levinei, Syzygium hancei, Schefflera octophylla, Auquilaria sinensis (Wu et al., 2021). Bare land (BL) 172 was used as a reference system in this study (Ren et al., 2007; Wang et al., 2017). Four treatments (BL, 173 EP, MF and NF) were a randomized block design in the study and the distance between plots was over 174 50 m; five replicated plots (10 ×10 m) in each treatment (Fig. 1). More detailed information on the 175 study site and forest restoration is included in Yao et al. (1984) and Ren et al. (2007). 6 176 177 Figure 1. Degraded coastal land and forest restorations in the study area. 178 179 2.2. Soil sampling and analyses 180 In May 2019, five soil cores (0-20 cm depth) were randomly sampled and then mixed into one composite 181 sample in each plot. After removing litter, roots and stones by sieving through a 4-mm sieve, fresh soils 182 were fractionated into three aggregate-size classes: > 2.0 mm large macroaggregate, LMA; 0.25-2.0 mm 183 small macroaggregate, SMA; < 0.25 mm microaggregate, MA, by using a dry sieving method (Yuan et 184 al., 2021). Then every aggregate fraction was divided into two parts. One part was air-dried for 185 determining soil physicochemical characteristics, and the other was stored at -20°C for analysing 186 microbial properties. The concentrations of SOC, total nitrogen (TN) and total phosphorus (TP) in each 187 aggregate fraction were determined following the protocols described by Lu (2000). 188 2.3. Soil microbial community composition 189 Microbial community composition in soil was measured using the phospholipid fatty acid (PLFA) 190 method (Bossio & Scow, 1998). The extracted lipids from soil samples, after separation and 191 transformation into free methyl esters, were analyzed using a gas chromatograph (7890B, Agilent 192 Technologies, Wilmington, USA) and identified by MIDI peak identification software (MIDI Inc., 193 Newark, USA). Specific fatty acids were used to represent Gram-positive (GP) bacteria (i15:0, α15:0, 194 i16:0, i17:0 and α17:0), Gram-negative (GN) bacteria (16:1ω9c, 16:1ω7c, 18:1ω7c, cy17:0, cy19:0), 195 saprotrophic fungi (18:2ω6c), arbuscular mycorrhizal fungi (AMF) (16:1ω5c), and actinomycetes 196 (10Me16:0, 10Me17:0 and 10Me18:0). The sum of GP and GN bacteria was represented as the total 7 197 bacterial biomass. Total microbial biomass represented by total PLFAs concentration was calculated as 198 well. The sum of PLFAs representing fungi was divided by the sum of PLFAs representing bacteria to 199 get the ratio of fungal to bacterial PLFA (F:B ratio). The amount of all PLFAs was expressed as nmol g- 1 200 dry soil. 201 2.4. Glomalin-related soil proteins 202 Total extractable GRSP (T-GRSP) and easily extractable GRSP (EE-GRSP) in each aggregate fraction 203 were determined according to Wright & Upadhyaya (1996). In brief, 1.0 g of air-dried soil sample was 204 added to an autoclavable centrifuge tube, with a mixture of 8 ml of 20 mM sodium citrate solution at pH 205 8.0, and vortexed for 30 s. Then, the mixture was autoclaved for 30 min at 121°C before being centrifuged 206 at 4200 g for 5 min, and the supernatant was determined as EE-GRSP. The residue in the tube was mixed 207 with 8 ml of 50 mM sodium citrate at pH 8.0 and then autoclaved at 121°C for 90 min, and centrifuged 208 at 4200 g for 5 min. The supernatant was transferred to another tube. Extractions would be repeated till 209 the supernatants had a pale straw color, indicating that GRSP was completely extracted. The 210 concentrations of GRSP in the two extracts were summed as T-GRSP and determined at 595 nm by an 211 enzyme microplate reader (Multiskan FC, Thermo Fisher Scientific, Waltham, USA) according to the 212 Bradford (1976) method. Both T-GRSP and EE-GRSP were expressed as mg g-1 dry soil. The detailed 213 meaurements are included in supplementary materials and methods. 214 2.5. Soil amino sugars 215 Amino sugars (ASs), including muramic acid (MurN), galactosamine (GalN) and glucosamine (GlcN) 216 were measured as described in Indorf et al. (2011), with minor modification (Mou et al., 2020). In brief, 217 ASs were hydrolyzed, extracted and derivatized with ortho-phthaldialdehyde, determined by high- 218 performance liquid chromatography (Dionex Ultimate 3000, Thermo Fisher Scientific, Waltham, USA). 219 The detailed relevant information and calculations are shown in supplementary materials and methods. 220 The concentrations of amino sugars in soil were expressed as μg g-1 dry soil. 221 2.6. Statistical analyses 222 All data was compiled in Excel software, and analyzed in SPSS 26.0, R 3.5.1 and Graphpad prism 8.0. 223 Before analysis, normal distribution and homogeneity were tested for all data. One-way ANOVA with 224 Duncan's test was utilized to identify significant differences in soil physicochemical properties, microbial 225 community composition, GRSP and amino sugars among forest restorations and aggregate fraction sizes 226 (P < 0.05). Linear regression models were used to display the correlations between the contributions and 8 227 relative changes of soil amino sugars and T-GRSP to SOC across aggregate fraction sizes. Random forest 228 models were used to assess the average importance of all soil abiotic and biotic properties for the 229 concentration of total GRSP (T-GRSP) and total amino sugars (Total ASs) and their contribution to SOC. 230 The percentage increase in the MSE (mean squared error) of variables was used and higher MSE% values 231 indicated greater importance. With the “rfPermute” package, the significance of predictors for the 232 response variables was calculated and with the “A3” package, the significance of the models and cross- 233 validated R2 was estimated with 1000 permutations of each response variable (Brieman, 2001; Jiao et 234 al., 2018). 235 236 3. RESULTS 237 3.1. Soil physiochemical properties 238 Soil physicochemical properties were consistently affected by forest restoration across aggregate fraction 239 sizes (P < 0.05, Table 1). Afforestation on bare land significantly enhanced soil nutrient accumulation, 240 and the concentrations of SOC, TN and TP in EP, MF and NF were much greater than those in BL. In 241 addition, native species mixtures accumulated about three times more total soil nutrients than EP over 242 60 years. Afforestation also changed soil stoichiometry; e.g., soil C:P and N:P ratios were greatly 243 increased and soil C:N ratio was decreased along the restoration chronosequence. Soil pH decreased 244 constantly during the restoration process. The overall effects of aggregate fraction sizes on soil nutrients 245 and stoichiometry were not statistically significant. 246 247 248 249 250 9 251 Table 1 Soil physicochemical characteristics within soil aggregate fractions along a restoration chronosequence. LMA, large macroaggregates; SMA, small macroaggregates; 252 MA, microaggregates; BL, bare land; EP, Eucalyptus plantation; MF, native species mixed forest; NF, native forest. Values are the means (n=5). Different lowercase letters 253 indicate significant differences among the different forest restorations within the same aggregate fraction at P < 0.05. Aggregate SOC TN TP Treatment pH C: N C: P N: P fraction (g kg-1) LMA BL 1.5 ± 0.2 c 0.2 ± 0.0 d 0.0 ± 0.0 d 4.6 ± 0.1 a 12.3 ± 2.9 ab 38.0 ± 7.3 b 3.5 ± 0.7 b EP 7.1 ± 1.3 b 0.4 ± 0.1 c 0.1 ± 0.0 c 4.6 ± 0.0 a 16.7 ± 0.8 a 96.1 ± 18.4 a 5.8 ± 1.1 a MF 19.7 ± 3.7 ab 1.6 ± 0.1 b 0.2 ± 0.0 b 4.4 ± 0.1 b 12.4 ± 2.3 ab 83.1 ± 11.8 a 6.9 ± 0.5 a NF 29.7 ± 2.0 a 3.0 ± 0.1 a 0.5 ± 0.0 a 4.5 ± 0.1 b 9.8 ± 0.5 b 64.2 ± 4.9 a 6.5 ± 0.2 a SMA BL 2.7 ± 0.6 c 0.3 ± 0.1 b 0.1 ± 0.0 c 4.5 ± 0.0 b 11.9 ± 3.6 ns 54.5 ± 15.4 b 7.5 ± 3.4 ns EP 8.6 ± 0.8 b 0.5 ± 0.1 b 0.1 ± 0.0 c 4.7 ± 0.0 a 17.5 ± 1.5 ns 127.4 ± 21.2 a 7.7 ± 1.5 ns MF 24.0 ± 4.4 a 1.8 ± 0.1 a 0.3 ± 0.0 b 4.3 ± 0.1 c 13.1 ± 2.4 ns 85.5 ± 12.8 ab 6.7 ± 0.5 ns NF 23.6 ± 1.5 a 2.2 ± 0.2 a 0.4 ± 0.0 a 4.4 ± 0.0 b 10.8 ± 0.8 ns 65.6 ± 5.5 b 6.1 ± 0.1 ns MA BL 4.7 ± 0.4 c 0.1 ± 0.1 c 0.1 ± 0.0 c 4.5 ± 0.1 a 196.3 ± 143.0 a 48.7 ± 3.6 c 1.4 ± 0.8 b EP 14.8 ± 2.2 b 1.0 ± 0.2 b 0.1 ± 0.0 c 4.7 ± 0.0 a 15.5 ± 1.1 b 130.4 ± 23.8 a 8.6 ± 1.7 a MF 28.2 ± 4.6 a 2.7 ± 0.4 a 0.4 ± 0.0 b 4.0 ± 0.1 b 10.2 ± 0.3 c 80.5 ± 11.2 b 7.8 ± 0.9 a NF 29.7 ± 1.0 a 3.1 ± 0.1 a 0.5 ± 0.0 a 4.1 ± 0.1 b 9.7 ± 0.3 c 62.2 ± 1.9 bc 6.4 ± 0.1 a 254 255 3.2. Soil microbial community composition 256 Soil microbial community biomass was significantly affected by aggregate fractions and afforestation (P < 0.05, Table 2). Microbial biomass in MA was much greater than that 257 in LMA and SMA. Afforestation consistently increased soil microbial biomass compared with bare land and the recovery of the soil microbial community in MF was much 10 258 faster than that in EP. Afforestation significantly decreased the F:B ratio, whereas the GP:GN ratio 259 increased, regardless of aggregate fraction sizes. 260 11 261 Table 2 Soil microbial community composition within soil aggregate fractions along a restoration chronosequence. LMA, large macroaggregates; SMA, small macroaggregates; 262 MA, microaggregates; BL, bare land; EP, Eucalyptus plantation; MF, native species mixed forest; NF, native forest; AMF, arbuscular mycorrhizal fungi; GP, Gram-positive 263 bacteria; GN, Gram-negative bacteria; F, fungi; B, bacteria and total microbial biomass (total PLFAs). Values are the means (n=5). Different uppercase letters indicate significant 264 differences among the different sizes of aggregate fractions at P < 0.05. Different lowercase letters indicate significant differences among the different forest restorations within 265 the same aggregate fraction at P < 0.05. ns indicates no significant difference among the different forest restorations within the same aggregate fraction at P < 0.05. Aggregate Bacteria Fungi AMF Actinomycetes GP GN Total PLFAs Treatment F: B GP: GN fraction (nmol g ) -1 LMA BL 4.3 ± 0.9 c 0.4 ± 0.1 c 0.1 ± 0.0 c 1.2 ± 0.3 c 2.2 ± 0.4 d 2.2 ± 0.6 c 6.1 ± 1.3 c 0.1 ± 0.0 a 1.1 ± 0.1 b EP 10.0 ± 2.6 b 0.5 ± 0.1 bc 0.4 ± 0.1 b 3.1 ± 0.8 b 6.0 ± 1.5 c 4.0 ± 11 bc 14.0 ± 3.5 b 0.1 ± 0.0 b 1.6 ± 0.1 a MF 14.9 ± 1.9 b 0.7 ± 0.1 ab 0.7 ± 0.1 b 4.3 ± 0.6 b 9.5 ± 1.2 b 5.4 ± 0.7 b 20.6 ± 2.6 b 0.1 ± 0.0 bc 1.8 ± 01 a NF 25.5 ± 2.1 a B 0.9 ± 0.1 a B 1.2 ± 0.1 a B 7.2 ± 0.5 a B 15.9 ± 1.4 a B 9.6 ± 0.7 a B 34.8 ± 2.8 a B 0.0 ± 0.0 c 1.7 ± 0.1 a SMA BL 5.5 ± 0.9 c 0.5 ± 0.0 c 0.1 ± 0.0 c 1.5 ± 02 c 2.9 ± 0.6 b 2.6 ± 0.4 b 7.5 ± 1.1 b 0.1 ± 0.0 a 1.1 ± 0.1 c EP 10.7 ± 1.5 b 0.7 ± 0.0 b 0.4 ± 0.1 b 3.3 ± 0.4 b 6.4 ± 1.0 b 4.3 ± 0.5 b 15.0 ± 2.0 b 0.1 ± 0.0 b 1.5 ± 0.1 b MF 18.9 ± 3.5 a 1.1 ± 0.2 a 0.8 ± 0.2 a 5.8 ± 1.1 a 12.4 ± 2.3 a 6.5 ± 1.2 a 26.6 ± 5.0 a 0.1 ± 0.0 b 1.9 ± 0.1 a NF 21.7 ± 1.7 a B 1.1 ± 0.1 a B 1.0 ± 0.1 a B 6.0 ± 0.5 a B 13.8 ± 1.1 a B 7.8 ± 0.6 a B 29.6 ± 2.3 a B 0.1 ± 0.0 b 1.8 ± 0.0 a MA BL 10.1 ± 2.3 b 0.8 ± 0.1 ns 0.4 ± 0.1 b 2.6 ± 0.6 c 5.4 ± 1.1 b 4.7 ± 1.2 b 13.8 ± 3.0 b 0.1 ± 0.0 a 1.2 ± 0.1 b EP 16.2 ± 3.3 b 1.3 ± 0.3 ns 0.6 ± 0.1 b 5.4 ± 1.1 b 10.4 ± 2.2 b 5.8 ± 1.1 b 23.5 ± 4.9 b 0.1 ± 0.0 a 1.8 ± 0.1 a MF 33.2 ± 6.7 a 1.8 ± 0.6 ns 1.7 ± 0.3 a 10.3 ± 2.2 a 21.0 ± 4.1 a 12.2 ± 2.6 a 46.9 ± 9.9 a 0.1 ± 0.0 b 1.8 ± 0.1 a NF 32.7 ± 1.7 a A 1.3 ± 0.1 ns A 1.6 ± 0.1 a A 9.0 ± 0.5 a A 21.2 ± 1.1 a A 11.5 ± 0.7 a A 44.7 ± 2.3 a A 0.0 ± 0.0 b 1.8 ± 0.0 a 266 12 267 3.3. Soil microbial-derived product concentrations and contribution to SOC 268 Afforestation on bare land greatly enhanced the accumulation of amino sugars and glomalin-related 269 soil proteins (GRSP) (P < 0.05, Fig. 2). On average, the concentration of total amino sugars in EP, MF 270 and NF significantly increased by 102%, 263% and 249% compared with bare land, and the 271 concentration of total glomalin also significantly increased by 268%, 859% and 922%, respectively. 272 Native species mixtures showed a faster accumulation of microbial-derived products than Eucalyptus 273 plantations and reached the level of native forests after 60 years of afforestation. In addition, native 274 species mixtures accumulated more fungal-derived amino sugars in soil with the F-GluN: MurN ratio 275 being much greater than that in EP and NF (Fig. S1). The effects of aggregate fraction sizes on the 276 concentrations of amino sugars and GRSP were overall not significant. 277 278 Figure 2. Concentrations of muramic acid (MurN), galactosamine (GalN), glucosamine (GlcN), total 279 amino sugars (ASs), easily (EE-GRSP) and total (T-GRSP) extractable glomalin-related soil proteins 280 within soil aggregate fractions along a restoration chronosequence. LMA, large macroaggregates; SMA, 281 small macroaggregates; MA microaggregates; BL, bare land; EP, Eucalyptus plantation; MF, native 13 282 species mixed forest; NF, native forest. Different lowercase letters indicate significant differences among 283 the different forest restorations at P < 0.05. Vertical bars denote standard errors of mean values (n = 5). 284 285 The contributions of amino sugars and GRSP to the SOC pool were significantly affected by 286 afforestation, aggregate fractions and their interaction (P < 0.05, Fig. 3). On average, the contribution of 287 soil total amino sugars to the SOC pool in EP, MF and NF significantly decreased by 50%, 59% and 68% 288 compared with bare land, and the contribution of total glomalin to the SOC pool also declined by 16%, 289 12% and 19%, respectively, but this was not significant. Although the contributions of GluN and total 290 amino sugars in LMA were significantly greater than those in SMA and MA, the contributions of specific 291 amino sugars and total amino sugars decreased constantly along the restoration chronosequence. The 292 contribution of GRSP to the SOC pool in LMA was generally greater than that in SMA and MA. The 293 contribution of total GRSP to SOC significantly decreased in LMA, whereas it increased in MA along 294 the restoration chronosequence. The contribution of EE-GRSP to the SOC pool was greater in SMA than 295 that in LMA and MA, and in MA the contribution was much greater in EP than that in other treatments. 296 297 Figure 3. Contributions of muramic acid (MurN), galactosamine (GalN), glucosamine (GlcN), total 298 amino sugars (ASs), easily (EE-GRSP) and total (T-GRSP) extractable glomalin-related soil proteins to 14 299 soil organic carbon (SOC) within soil aggregate fractions along a restoration chronosequence. LMA, 300 large macroaggregates; SMA, small macroaggregates; MA microaggregates; BL, bare land; EP, 301 Eucalyptus plantation; MF, native species mixed forest; NF, native forest. Different uppercase letters 302 indicate significant differences among the different sizes of aggregates fractions at P < 0.05. Different 303 lowercase letters indicate significant differences among the different forest restorations at P < 0.05. 304 Vertical bars denote standard errors of mean values (n = 5). 305 306 3.4. Relative changes in soil microbial-derived products compared with SOC 307 Afforestation led to a faster accumulation of GRSP than of amino sugars in soil, with the relative change 308 rate of GRSP four times greater than that of amino sugars (P < 0.05, Fig. 4). The relative change rates of 309 microbial-derived products in SMA and MA were much greater than those in LMA (P < 0.05, Fig. 4). 310 The relative change rate of microbial-derived products within aggregate fractions did not significantly 311 vary with afforestation types. The relative change in total ASs was positively correlated with the relative 312 change in T-GRSP in SMA and in total (including LMA, SMA and MA) (P < 0.01, Fig. 5). The 313 contribution of total ASs to SOC was positively correlated with the contribution of T-GRSP to SOC in 314 LMA and in total (P < 0.001, Fig. 5). The change ratio of total ASs relative to SOC was positively 315 correlated with the change ratio of T-GRSP relative to SOC in SMA (P < 0.01, Fig. 5) and in total (P < 316 0.05, Fig. 5). 317 318 Figure 4. Relative changes in total amino sugars (ASs) and total extractable glomalin-related soil 319 proteins (T-GRSP) compared with soil organic carbon (SOC) within soil aggregate fractions along a 320 restoration chronosequence. LMA, large macroaggregates; SMA, small macroaggregates; MA 321 microaggregates; EP, Eucalyptus plantation; MF, native species mixed forest; NF, native forest. * 322 indicates significant differences between the relative change rate of T-GRSP and total ASs at P < 0.05. 323 Vertical bars denote standard errors of mean values (n = 5). 15 LMA SMA MA LMA: Y=1.320*X + 4.715, P < 0.001 SMA: Y=0.2871*X + 6.601, NS MA: Y=-0.2255*X + 7.186, NS Total: Y=1.079*X + 4.595, P < 0.001 30 T-GRSP: SOC (%) 20 10 0 0 5 10 15 Total ASs: SOC (%) LMA: Y=0.9785*X + 3.520, NS SMA: Y=5.596*X - 0.7113, P < 0.01 MA: Y=0.6435*X + 6.775, NS Total: Y=2.295*X + 3.596, P < 0.01 40 30 ΔT-GRSP 20 10 0 -10 -2 0 2 4 6 8 ΔTotal ASs LMA: Y=1.120*X + 0.2613, NS SMA: Y=7.022*X - 0.6536, P < 0.01 MA: Y=0.4142*X + 2.020, NS Total: Y=1.735*X + 0.9293, P < 0.05 6 ΔT-GRSP: ΔSOC 4 2 0 0.0 0.5 1.0 1.5 2.0 2.5 ΔTotal ASs: ΔSOC 324 325 Figure 5. The correlations between contributions or relative change rates of total extractable glomalin- 326 related soil proteins (T-GRSP) and amino sugars (ASs) within the large macroaggregates (LMA), small 327 macroaggregates (SMA) and microaggregates (MA) fractions and in total. The solid blue line represents 328 the correlation for all three sizes of aggreagtes; NS indicates no significance at P < 0.05. 329 330 3.5. Predictors of the accumulation and contribution of soil microbial-derived products to SOC 331 Random forest models suggested that soil fertility (e.g., SOC, TN, TP and their ratios) and microbial 16 332 community explained most of the variation of the accumulation of amino sugars in soil (Fig. 6a,c, P < 333 0.001). The contribution of total GRSP to SOC was mainly affected by fungal biomass and SOC (Fig. 334 6b, P < 0.001). The contribution of amino sugars to SOC was mainly affected by SOC, actinomycete 335 biomass, GP biomass and soil C:P ratio (Fig. 6d, P < 0.001). 336 337 Figure 6. The average predictive importance (mean square error (MSE) increase percentage) for all soil 338 abiotic and biotic properties for concentrations and contributions of total amino sugars (ASs) and total 339 extractable glomalin-related soil proteins (T-GRSP) to soil organic carbon (SOC). TP, total phosphorus; 340 TN, total nitrogen; C:P, SOC:TP; C:N, SOC:TN; AMF, N:P, TN:TP; arbuscular mycorrhizal fungi; GP, 17 341 Gram-positive bacteria; GN, Gram-negative bacteria; F, fungi; B, bacteria and total microbial biomass 342 (total PLFAs). 343 344 4. DISCUSSION 345 4.1. The accumulation of GRSP and amino sugars increased constantly by afforestation 346 In this study, the accumulation of glomalin and amino sugars in soil increased along the afforestation 347 chronosequence which confirmed our first hypothesis (Fig. 2). The enhanced accumulation of microbial- 348 derived products in soil was ascribed to the synchronized increases in microbial biomasses and soil 349 fertility. Vegetation restoration could enhance plant-C inputs and in tandem stimulate soil microbial 350 activities (Hu et al., 2020). Microbial products mainly accumulate in soils via microbial decomposition 351 and turnover (Liang, 2020; Zhang et al., 2021). However, the contribution of GRSP and amino sugars to 352 the SOC pool decreased along the afforestation chronosequence (Fig. 3), suggesting an increase in the 353 contribution of plant-derived C to the SOC pool with forest restoration. This was in line with previous 354 studies that microbial-derived C contribution to SOC declined along a forest restoration chronosequence 355 (Shao et al., 2019); SOC in forest soil may be dominated by particulate organic matter (plant-derived) 356 (Cotrufo et al., 2019) and decompose less owing to unfavorable conditions (e.g., low pH) for bacterial 357 growth in subtropical forest soils compared with cropland (Angst et al., 2021). The contribution of 358 different microorganisms to SOC with succession changes was also observed in other studies (Shao et 359 al., 2017). 360 The overall effects of aggregate fractions on the accumulation of microbial-derived products in soil 361 were not significant. Although microbial biomass tended to increase with decreasing size of aggregate 362 fraction, lower transformation efficiency from living microbial biomass to necromass within 363 microaggregates might inhibit the accumulation of amino sugars and GRSP there (Xu et al., 2022). 364 Furthermore, microaggregates might harbor more diverse and abundant microbial communities relative 365 to macroaggregates (Bach et al., 2018), which favors faster microbial metabolic activities and further 366 promotes the recycling of microbial by-products. Our results suggest that necromass recycling might be 367 a vital mechanism for mediating microbial metabolism and soil C cycling (Cui et al., 2020), and forest 368 restoration may strengthen this effect in microaggregates. Additionally, the accumulation of GRSP and 369 amino sugars was positively correlated with aggregate stability (Tables S1 and S2, Zhang et al., 2022). 370 Regression analysis shows that the accumulation of GRSP and amino sugars during forest 18 371 restoration was synergistic (Fig. 5). AMF could not only delay the turnover of macroaggregates providing 372 more time and space for metabolic interactions between AMF and their associated microbiota (Rillig & 373 Mummey, 2006) but also accelerate the turnover of microaggregates (Morris et al., 2019). 374 Macroaggregates contain larger pore spaces, more AMF hyphae and more GRSP which might facilitate 375 the synergistic accumulation of amino sugars and glomalin (Lovelock et al., 2004). However, the 376 underlying mechanism should be explored via manipulative experiments and microscopic observation in 377 future research. 378 379 4.2. Faster accumulation and greater contribution of GRSP to the SOC pool than of amino sugars 380 during forest restoration 381 Our data support our second hypothesis that GRSP comprised a larger component of SOC than amino 382 sugars did and therefore accumulated faster during forest restoration (Figs 2, 3 and 4). This can be 383 explained as follows. First, GRSP is likely more stable than soil amino sugars (Agnihotri et al., 2022; 384 Rillig et al., 2001). GRSP is composed of > 50% recalcitrant components such as aromatic- and alkyl-C, 385 and has a greater chemical recalcitrance than amino sugars (Agnihotri et al., 2022). GRSP has a higher 386 propensity to form stable aggregates via binding mineral and organic particles than amino sugars do 387 (Gunina & Kuzyakov, 2015). Conversely, amino sugars may establish relatively fewer bonds with 388 minerals than glomalin does. Second, AMF may reduce bacterial biomass and soil amino sugar 389 concentrations due to nutrient deficiency (He et al., 2020), which would intensify the different 390 accumulation rates of GRSP and amino sugars. Third, GRSP concentrations are positively correlated 391 with net primary productivity (NPP) globally and higher plant productivity can provide more available 392 C to AM fungi for glomalin production (Treseder & Turner, 2007). NPP may increase with vegetation 393 restoration on barren land, if AM host plants are more abundant for the fungi that would benefit the 394 accumulation of glomalin (Treseder & Turner, 2007) and promote the microbial-derived C in soil. In the 395 study, Eucalyptus exserta (EP) associates with both arbuscular and ectomycorrhizal fungi, all dominant 396 tree species in NF are arbuscular mycorrhizal, while Acacia auriculaeformis as a dominant species in 397 MF is ectomycorrhizal, and Calamus tetradactylus is ectomycorrhizal as a dominant herb in MF. 398 However, Guo et al. (2021) found that the contribution of amino sugars to SOC in karst soils increased 399 with vegetation restorations and more bacterial-derived C accumulated, which might be associated with 400 the difference in microbial C transformation efficiency driven by pH (Malik et al., 2018). 19 401 The relative change rate of microbial-derived products was mediated by soil aggregate fraction, 402 with values much greater in microaggregates than in macroaggregates (Fig. 4). Minerals within 403 microaggregates have a huge surface area to adsorb microbial-derived C to form stable organo-mineral 404 complexes (Liang, 2020). In addition, Macroaggregates are more vulnerable to disturbances and 405 environmental changes than microaggregates are (Ye et al., 2020), whereas microaggregates maintain a 406 relatively stable micro-environment. The contribution of microbial-derived products to SOC pools 407 decreased with decreasing aggregate fraction size, suggesting that microaggregates maintain a faster 408 accumulation rate or greater sequestration efficiency of microbial-derived products, and greater potential 409 storage capacity relative to macroaggregates. As argued by Six et al. (2002) and Stewart et al. (2008), 410 when SOC was further from its maximum content (C saturation point), its C-sequestration rate is greater. 411 the farther away from its C saturation point, the more efficient C-sequestration rate. In addition, the 412 decreased contribution of microbial-derived products to SOC pools in smaller aggregate fractions might 413 be accompanied by increased plant-derived C accumulation. The decreased microbial-derived product 414 contributions to SOC pools along the afforestation chronosequence further indicates that the role of the 415 soil microbial community shifted from in vivo turnover (reducing C-use efficiency) to ex vivo 416 modification (decomposing and incorporating more plant-C into the stable C pool) (Liang, 2020). 417 418 4.3. Soil fertility and microbial community mediated the accumulation and contribution of 419 microbial-derived products during forest restoration 420 Random forest modeling revealed that the accumulation and contribution of GRSP and amino 421 sugars were mainly affected by soil fertility and soil microbial community during forest restoration in 422 our study (Fig. 6; Agnihotri et al., 2022; Chen et al., 2020). Soil fertility and microbial biomass explained 423 most of the variation of GRSP and amino sugars along the afforestation chronosequence. The 424 contribution of GRSP to SOC was mainly affected by fungal biomass and SOC, and the contribution of 425 amino sugars was affected by SOC, actinomycete biomass, GP biomass and soil C:P ratio. SOC was the 426 major substrate for microbial metabolism; in tandem the microbial metabolites also contributed to the 427 soil C pool. Hence, the accumulation of microbial byproducts (glomalin and amino sugars) and SOC 428 preservation were enhanced synergistically. Especially, glomalin and amino sugars bind with minerals to 429 form mineral-associated organic C (Agnihotri et al., 2022; Liang et al., 2020) which can be stabilized in 430 soils for decades to centuries (Lavallee et al., 2020). The important role of TN and TP in explaining the 20 431 variation of the concentrations of total glomalin and amino sugars indicates that soil fertility plays a vital 432 role in mediating microbial byproduct accumulation, which can be explained by the beneficial effect of 433 vegetation restoration on fungi in P-limited ecosystems. Indeed, our data also support the contention that 434 fungal biomass plays an important role in mediating the contribution of GRSP to SOC. Vegetation 435 restoration offered favorable conditions for the growth and development of mycorrhizal fungi, and 436 further facilitated the release and accumulation of glomalin in soils (Li et al., 2020; Wright & 437 Anderson, 2000). Actinomycetes are aerobic spore-forming Gram-positive bacteria characterized by 438 substrate and aerial mycelium growth (Bhatti et al., 2017), whereas Gram-positive bacteria have a thicker 439 peptidoglycan cell wall (Joergensen, 2018; Liang et al., 2019). Hence, proliferation of Gram-positive 440 bacteria was also an important contributor to microbial-derived byproducts to SOC. The soil C:P ratio 441 may affect the microbial community composition and its activities and thus influence soil C cycling 442 (Shen et al., 2019). Yuan et al. (2021) also demonstrated in the same study area that P addition decreases 443 the contribution of amino sugars to SOC via increasing microbial biomass and enzymes activities. 444 The present results highlight the greater benefits for soil C-accumutation and nutrient fertility 445 associated afforestation with native versus exotic fast-growing species. The capacity to symbiotic 446 establishment (in this case with arbuscular mycorrhizal fungi) and the increased SOC were due to fungi 447 and native tree species. The stability of SOC has even been demonstrated to be affected by tree species 448 (Angst et al., 2019). The results also show a higher P-use efficiency (higher soil C:P) with native than 449 with exotic species. The native forests improve soil C storage capacity and nutrient retention and use- 450 efficiency, maintaining greater soil microbial populations and diversity than fast-growing non-native 451 species. All this suggests greater biodiversity conservation and service provision such as mitigation of 452 climate chage with native reforestation (Wu et al., 2021). 21 453 454 Figure 7. A conceptual diagram illustrating the accumulation and contribution of glomalin and amino 455 sugars to soil organic carbon (SOC) within soil aggregate fractions along a restoration chronosequence. 456 LMA, large macroaggregates; SMA, small macroaggregates; MA, microaggregates; BL, bare land; EP, 457 Eucalyptus plantation; MF, native species mixed forest; NF, native forest. 458 459 5. CONCLUSIONS 460 Our study provides new insight into the accumulation and contribution of glomalin and amino sugars to 461 the SOC pool during forest restoration (Fig. 7). Afforestation of bare land greatly enhanced the 462 accumulation of GRSP and amino sugars, but it decreased their contribution to SOC. The faster 463 accumulation and greater contribution of GRSP to SOC compared with those of amino sugars highlights 464 the important role of AMF in mediating soil C cycling during forest restoration, despite the proportional 465 contribution of GRSP and amino sugars to the SOC pool diminishing with forest restoration. Such 466 information is valuable for improving our mechanistic understanding of the microbial control of SOC 467 preservation during the restoration of degraded ecosystems. Our findings also suggest that favoring 468 arbuscular mycorrhizal plants can be an effective option to sequester more soil C during restoration 469 practices. The importance of the soil C:P ratio in mediating the accumulation of microbial-derived 470 products suggests that appropriate fertilization may also play an important role in mediating soil C 471 sequestration and stabilization, particularly in a P-limited ecosystem. These results together provide 472 important guidance for management practices considering belowground microbial processes and 22 473 functions during coastal restoration, benefiting both aboveground and belowground biodiversity and 474 multifunctionality. 475 476 ACKNOWLEDGMENTS 477 This work was supported by the National Key Research and Development Program of China (No. 478 2021YFC3100405), the National Natural Science Foundation of China (42177289), the Key Special 479 Project for Introduced Talents Team of Southern Marine Science and Engineering Guangdong 480 Laboratory (Guangzhou) (Grant No. GML2019ZD0408), the Youth Innovation Promotion Association 481 CAS (2019340), the Spanish Government projects PID2019-110521GB-I00 and PID2020115770RB-I, 482 Fundación Ramón Areces project CIVP20A6621 and Catalan government project SGR2017-1005. 483 484 CONFLICT OF INTEREST 485 The authors declare to have no conflict of interest. 486 487 REFERENCES 488 Agnihotri, R., Bharti, A., Ramesh, A., Prakash, A. & Sharma, M. P. (2021). Glomalin related protein and 489 C16:1 ω 5 PLFA associated with AM fungi as potential signatures for assessing the soil C 490 sequestration under contrasting soil management practices. European Journal of Soil Biology, 103, 491 103286. https://doi.org/10.1016/j.ejsobi.2021.103286 492 Agnihotri, R., Sharma, M. P., Prakash, A., Ramesh, A., Bhattacharjya, S., Patra, A. K., Manna, M. C., 493 Kurganova, I. & Kuzyakov, Y. (2022). Glycoproteins of arbuscular mycorrhiza for soil carbon 494 sequestration: Review of mechanisms and controls. Science of The Total Environment, 806, 150571. 495 https://doi.org/10.1016/j.scitotenv.2021.150571 496 Angst, G., Mueller, K. E., Eissenstat, D. M., Trumbore, S., Freeman, K. H., Hobbie, S. E., Chorover, J., 497 Oleksyn, J., Reich, P. B. & Mueller, C. W. (2019). Soil organic carbon stability in forests: Distinct 498 effects of tree species identity and traits. Global Change Biology, 25, 1529-1546. 499 https://doi.org/10.1111/gcb.14548 500 Angst, G., Mueller, K. E., Nierop, K., G. J. & Simpson., M. J. (2021). Plant- or microbial-derived? A 501 review on the molecular composition of stabilized soil organic matter. Soil Biology and 502 Biochemistry, 156, 108189. https://doi.org/10.1016/j.soilbio.2021.108189 23 503 Bach, E. M., Williams, R. J., Hargreaves, S. K., Yang, F. & Hofmockel, K. S. (2018). Greatest soil 504 microbial diversity found in micro-habitats. Soil Biology and Biochemistry, 118, 217-226. 505 https://doi.org/10.1016/j.soilbio.2017.12.018 506 Bhatti, A. A., Haq, S. & Bhat, R. A. (2017). Actinomycetes benefaction role in soil and plant health. 507 Microbial Pathogenesis, 111, 458-467. https://doi.org/10.1016/j.micpath.2017.09.036 508 Bossio, D. & Scow, K. (1998). Impacts of carbon and flooding on soil microbial communities: 509 phospholipid fatty acid profiles and substrate utilization patterns. Microbial Ecology, 35, 265-278. 510 https://doi.org/10.1007/s002489900082 511 Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of 512 protein utilizing the principle of protein-dye binding. Analytical Biochemistry, 72, 248-254. 513 https://doi.org/10.1006/abio.1976.9999 514 Breiman, L. (2001). Random forests. Machine Learning, 45, 5-32. 515 https://doi.org/10.1023/A:1010933404324 516 Buckeridge, K. M., La Rosa, A. F., Mason, K. E., Whitaker, J., Mcnamara, N. P., Grant, H. K. & Ostle, 517 N. J. (2020). Sticky dead microbes: Rapid abiotic retention of microbial necromass in soil. Soil 518 Biology and Biochemistry, 149, 107929. https://doi.org/10.1016/j.soilbio.2020.107929 519 Canadell, J. G. & Raupach, M. R. (2008). Managing forests for climate change mitigation. Science, 320, 520 1456-7. https://doi.org/10.1126/science.1155458 521 Capellesso, E. S., Cequinel, A., Marques, R., Sausen, T. L., Bayer, C. & Marques, M. C. M. (2021). Co- 522 benefits in biodiversity conservation and carbon stock during forest regeneration in a preserved 523 tropical landscape. Forest Ecology and Management, 492, 119222. 524 https://doi.org/10.1016/j.foreco.2021.119222 525 Chen, G. P., Ma, S. H., Tian, D., Xiao, W., Jiang, L., Xing, A. J., Zou, A. L., Zhou, L. H., Shen, H. H., 526 Zheng, C. Y., Ji, C. J., He, H. B., Zhu, B., Liu, L. L. & Fang, J. Y. (2020). Patterns and determinants 527 of soil microbial residues from tropical to boreal forests. Soil Biology and Biochemistry, 151, 528 108059. https://doi.org/10.1016/j.soilbio.2020.108059 529 Cotrufo, M. F., Ranalli, M. G., Haddix, M. L. Six, J. & Lugato, E. (2019). Soil carbon storage informed 530 by particulate and mineral-associated organic matter. Nature Geoscience, 12, 989-994. 531 https://doi.org/10.1038/s41561-019-0484-6 532 Crouzeilles, R., Ferreira, M. S. & Curran, M. (2016). Forest restoration: a global dataset for biodiversity 24 533 and vegetation structure. Ecology, 97, 2167. https://doi.org/10.1002/ecy.1474 534 Cui, J., Zhu, Z. K., Xu, X. L., Liu, S. L., Jones, D. L., Kuzyakov, Y., Shibistova, O., Wu, J. S. & Ge, T. 535 (2020). Carbon and nitrogen recycling from microbial necromass to cope with C: N stoichiometric 536 imbalance by priming. Soil Biology and Biochemistry, 142, 107720. 537 https://doi.org/10.1016/j.soilbio.2020.107720 538 Duarte, C. M., Losada, I. J., Hendriks, I. E., Mazarrasa, I. & Marbà, N. (2013). The role of coastal plant 539 communities for climate change mitigation and adaptation. Nature Climate Change, 3, 961-968. 540 https://doi.org/10.1038/nclimate1970 541 Glaser, B., Turrión, M.-B., Alef, K. (2004). Amino sugars and muramic acid-biomarkers for soil 542 microbial community structure analysis. Soil Biology and Biochemistry, 36, 399-407. 543 https://doi.org/10.1016/j.soilbio.2003.10.013 544 Gunina, A. & Kuzyakov, Y. (2015). Sugars in soil and sweets for microorganisms: Review of origin, 545 content, composition and fate. Soil Biology and Biochemistry, 90, 87-100. 546 https://doi.org/10.1016/j.soilbio.2015.07.021 547 Guo, Z. M., Zhang, X. Y., Dungait, J. A. J., Green, S. M., Wen, X. F. & Quine, T. A. (2021). Contribution 548 of soil microbial necromass to SOC stocks during vegetation recovery in a subtropical karst 549 ecosystem. Science of The Total Environment, 761, 143945. 550 https://doi.org/10.1016/j.scitotenv.2020.143945 551 Gupta V. V. S. R. & Germida, J. J. (2015). Soil aggregation: Influence on microbial biomass and 552 implications for biological processes. Soil Biology and Biochemistry, 80, A3-A9. 553 https://doi.org/10.1016/j.soilbio.2014.09.002 554 He, X. X., Li, X. B., Liu, T., Yang, X. J., Cao, J. B., Tao, L. B., Wang, X. L., Liu, Z. F., Yao, Q., Li, Y. T., 555 Zou, X. M., Shao, Y. H., Li, J. X., Zhang, W. X. & Fu, S. (2020). Earthworms negate the adverse 556 effect of arbuscular mycorrhizae on living bacterial biomass and bacterial necromass accumulation 557 in a subtropical soil. Soil Biology and Biochemistry, 151, 108052. 558 https://doi.org/10.1016/j.soilbio.2020.108052 559 Hisano, M., Searle, E. B. & Chen, H. Y. H. (2018). Biodiversity as a solution to mitigate climate change 560 impacts on the functioning of forest ecosystems. Biological Reviews, 93, 439-456. 561 https://doi.org/10.1111/brv.12351 562 Hu, P. L., Xiao, J., Zhang, W., Xiao, L. M., Yang, R., Xiao, D., Zhao, J. & Wang, K. L. (2020). Response 25 563 of soil microbial communities to natural and managed vegetation restoration in a subtropical karst 564 region. Catena, 195, 104849. https://doi.org/10.1016/j.catena.2020.104849 565 Indorf, C., Dyckmans, J., Khan, K. S. & Joergensen, R.G. (2011). Optimisation of amino sugar 566 quantification by HPLC in soil and plant hydrolysates. Biology and Fertility of Soils, 47, 387-396. 567 https://doi.org/10.1007/s00374-011-0545-5 568 Irving, T. B., Alptekin, B., Kleven, B. & Ane, J. M. (2021). A critical review of 25 years of glomalin 569 research: A better mechanical understanding and robust quantification techniques are required. New 570 Phytologist, 232, 1572-1581. https://doi.org/10.1111/nph.17713 571 Jackson, R. B., Lajtha, K., Crow, S. E., Hugelius, G., Kramer, M. G. & Piñeiro, G. (2017). The ecology 572 of soil carbon: Pools, vulnerabilities, and biotic and abiotic controls. Annual Review of Ecology, 573 Evolution, and Systematics, 48, 419-445. https://doi.org/10.1146/annurev-ecolsys-112414-054234 574 Jiao, S., Chen, W. M., Wang, J. L., Du, N. N., Li, Q. P. & Wei, G. H. (2018). Soil microbiomes with 575 distinct assemblies through vertical soil profiles drive the cycling of multiple nutrients in reforested 576 ecosystems. Microbiome, 6, 146. https://doi.org/10.1186/s40168-018-0526-0 577 Joergensen, R. G. (2018). Amino sugars as specific indices for fungal and bacterial residues in soil. 578 Biology and Fertility of Soils, 54, 559-568. https://doi.org/10.1007/s00374-018-1288-3 579 Lal, R., Monger, C., Nave, L. & Smith, P. (2021). The role of soil in regulation of climate. Philosophical 580 Transactions of the Royal Society B: Biological Sciences, 376, 20210084. 581 https://doi.org/10.1098/rstb.2021.0084 582 Lange, M., Eisenhauer, N., Sierra, C. A., Bessler, H., Engels, C., Griffiths, R. I., Mellado-Vázquez, P. 583 G., Malik, A. A., Roy, J., Scheu, S., Steinbeiss, S., Thomson, B. C., Trumbore, S. E. & Gleixner, 584 G. (2015). Plant diversity increases soil microbial activity and soil carbon storage. Nature 585 Communications, 6, 6707. https://doi.org/10.1038/ncomms7707 586 Lavallee, J. M., Soong, J. L. & Cotrufo, M. F. (2020). Conceptualizing soil organic matter into particulate 587 and mineral‐associated forms to address global change in the 21st century. Global Change Biology, 588 26, 261-273. https://doi.org/10.1111/gcb.14859 589 Liang, C. (2020). Soil microbial carbon pump: Mechanism and appraisal. Soil Ecology Letters, 2, 241- 590 254. https://doi.org/10.1007/s42832-020-0052-4 591 Liang, C., Amelung, W., Lehmann, J. & Kästner, M. (2019). Quantitative assessment of microbial 592 necromass contribution to soil organic matter. Global Change Biology, 25, 3578-3590. 26 593 https://doi.org/10.1111/gcb.14781 594 Liang, C., Kästner, M. & Joergensen, R. G. (2020). Microbial necromass on the rise: The growing focus 595 on its role in soil organic matter development. Soil Biology and Biochemistry, 150, 108000. 596 https://doi.org/10.1016/j.soilbio.2020.108000 597 Lovelock, C. E., Wright, S. F. & Nichols, K. A. (2004). Using glomalin as an indicator for arbuscular 598 mycorrhizal hyphal growth: An example from a tropical rain forest soil. Soil Biology and 599 Biochemistry, 36, 1009-1012. https://doi.org/10.1016/j.soilbio.2004.02.010 600 Lu, R. K. (2000). Analytical methods for soils and agricultural chemistry. Beijing: Scientific and 601 Technology Press. 602 Ma, Z., Melville, D. S., Liu, J., Chen, Y., Yang, H., Ren, W., Zhang, Z., Piersma, T. & Li, B. (2014). 603 Rethinking china's new great wall. Science, 346, 6212. https://doi.org/10.1126/science.1257258 604 Malik, A. A., Puissant, J., Buckeridge, K. M., Goodall, T., Jehmlich, N., Chowdhury, S., Gweon, H. S., 605 Peyton, J. M., Mason, K. E., van Agtmaal, M., Blaud, A., Clark, I. M., Whitaker, J., Pywell, R. F., 606 Ostle, N., Gleixner, G. & Griffiths, R. I. (2018). Land use driven change in soil pH affects microbial 607 carbon cycling processes. Nature Communications, 9, 3591. https://doi.org/10.1038/s41467-018- 608 05980-1 609 Morris, E. K., Morris, D. J. P., Vogt, S., Gleber, S. C., Bigalke, M., Wilcke, W. & Rillig, M. C. (2019). 610 Visualizing the dynamics of soil aggregation as affected by arbuscular mycorrhizal fungi. The ISME 611 Journal, 13, 1639-1646. https://doi.org/10.1038/s41396-019-0369-0 612 Mou, Z. J., Kuang, L. H., Yan, B. Y., Zhang, X. Y., Wang, Y. Q. & Liu, Z. F. (2020). Influences of sample 613 storage and grinding on the extraction of soil amino sugars. Soil Ecology Letters, 20, 157-163. 614 https://doi.org/10.1007/s42832-020-0031-9 615 Murugan, R., Djukic, I., Keiblinger, K., Zehetner, F., Bierbaumer, M., Zechmeister-Bolternstern, S. & 616 Joergernsen, R. G. (2019). Spatial distribution of microbial biomass and residues across soil 617 aggregate fractions at different elevations in the Central Austrian Alps. Geoderma, 339, 1-8. 618 https://doi.org/10.1016/j.geoderma.2018.12.018 619 Navas, M., Martín Lammerding, D., Hontoria, C., Ulcuango, K. & Mariscal Sancho, I. (2021). The 620 distinct responses of bacteria and fungi in different‐sized soil aggregates under different 621 management practices. European Journal of Soil Science, 72, 1177-1189. 622 https://doi.org/10.1111/ejss.12997 27 623 Qiao, L. L., Li, Y. Z., Song, Y. H, Zhai, J. Y, Wu, Y., Chen, W. J., Liu, G. B. & Xue, S. (2019). Effects of 624 vegetation restoration on the distribution of nutrients, glomalin-related soil protein, and enzyme 625 activity in soil aggregates on the loess plateau, china. Forests, 10, 796. 626 https://doi.org/10.3390/f10090796 627 Ren, H., Li, Z. A., Shen, W. J., Yu, Z. Y., Peng, S. L., Liao, C. H., Ding, M. M. & Wu, J. G. (2007). 628 Changes in biodiversity and ecosystem function during the restoration of a tropical forest in south 629 China. Science in China Series C: Life Sciences, 50, 277-284. https://doi.org/10.1007/s11427-007- 630 0028-y 631 Rillig, M. C. & Mummey, D. L. (2006). Mycorrhizas and soil structure. New Phytologist, 171, 41-53. 632 https://doi.org/10.1111/j.1469-8137.2006. 01750.x 633 Rillig, M. C., Muller, L. A. & Lehmann, A. (2017). Soil aggregates as massively concurrent evolutionary 634 incubators. The ISME Journal, 11, 1943-1948. https://doi.org/10.1038/ismej.2017.56 635 Rillig, M. C., Wright, S. F., Nichols, K. A., Schmidt, W. F. & Torn, M. S. (2001). Large contribution of 636 arbuscular mycorrhizal fungi to soil carbon pools in tropical forest soils. Plant and Soil, 233, 167- 637 177. https://doi.org/10.1023/A:1010364221169 638 Schmidt, M. W. I., Torn, M. S., Abiven, S., Dittmar, T., Guggenberger, G., Janssens, I. A., Kleber, M., 639 Kögel-Knabner, I., Lehmann, J., Manning, D. A. C., Nannipieri, P., Rasse, D. P., Weiner, S. & 640 Trumbore, S. E. (2011). Persistence of soil organic matter as an ecosystem property. Nature, 478, 641 49-56. https://doi.org/10.1038/nature10386 642 Shao, P. S., Liang, C., Lynch, L., Xie, H. T. & Bao, X. L. (2019). Reforestation accelerates soil organic 643 carbon accumulation: Evidence from microbial biomarkers. Soil Biology and Biochemistry, 131, 644 182-190. https://doi.org/10.1016/j.soilbio.2019.01.012 645 Shao, S., Zhao, Y., Zhang, W., Hu, G. Q., Xie, H. T., Yan, J. H., Han, S. J., He, H. B. & Zhang, X. D. 646 (2017). Linkage of microbial residue dynamics with soil organic carbon accumulation during 647 subtropical forest succession. Soil Biology and Biochemistry, 114, 114-120. 648 https://doi.org/10.1016/j.soilbio.2017.07.007 649 Shen, F. F., Wu, J. P., Fan, H. B., Liu, W. F., Guo, X. M., Duan, H. L., Hu, L., Lei, X. M. & Wei, X. H. 650 (2019). Soil N/P and C/P ratio regulate the responses of soil microbial community composition and 651 enzyme activities in a long-term nitrogen loaded Chinese fir forest. Plant and Soil, 436, 91-107. 652 https://doi.org/10.1007/s11104-018-03912-y 28 653 Six, J., Conant, R. T., Paul, E. A. & Paustian, K. (2002). Stabilization mechanisms of soil organic matter: 654 Implications for C-saturation of soils. Plant and Soil, 241, 155-176. 655 https://doi.org/10.1023/A:1016125726789 656 Snäll, T., Triviño, M., Mair, L., Bengtsson, J. & Moen, J. (2021). High rates of short-term dynamics of 657 forest ecosystem services. Nature Sustainability, 4, 951-957. https://doi.org/10.1038/s41893-021- 658 00764-w 659 Stewart, C. E., Plante, A. F., Paustian, K., Conant, R. T. & Six, J. (2008). Soil carbon saturation: Linking 660 concept and measurable carbon pools. Soil Science Society America Journal, 72, 379-392. 661 https://doi.org/10.2136/sssaj2007.0104 662 Treseder, K. K. & Turner, K. M. (2007). Glomalin in ecosystems. Soil Science Society of America Journal, 663 71, 1257-1266. https://doi.org/10.2136/sssaj2006.0377 664 Upton, R. N., Bach, E. M. & Hofmockel, K. S. (2019). Spatio-temporal microbial community dynamics 665 within soil aggregates. Soil Biology and Biochemistry, 132, 58-68. 666 https://doi.org/10.1016/j.soilbio.2019.01.016 667 Vestal, J. R. & White, D. C. (1989). Lipid analysis in microbial ecology: quantitative approaches to the 668 study of microbial communities. Bioscience, 39, 535-541. https://doi.org/10.2307/1310976 669 Wang, W. J., Zhong, Z. L., Wang, Q., Wang, H. M., Fu, Y. J. & He, X. Y. (2017). Glomalin contributed 670 more to carbon, nutrients in deeper soils, and differently associated with climates and soil properties 671 in vertical profiles. Scientific Reports, 7, 13003. https://doi.org/10.1038/s41598-017-12731-7 672 Wright, S. F., Upadhyaya, A. & Buyer, J. S. (1998). Comparison of N-linked oligosaccharides of glomalin 673 from arbuscular mycorrhizal fungi and soils by capillary electrophoresis. Soil Biology and 674 Biochemistry, 30, 1853-1857. 10.1016/S0038-0717(98)00047-9 675 Wright, S.F. & Upadhyaya, A. (1996). Extraction of an abundant and unusual protein from soil and 676 comparison with hyphal protein of arbuscular mycorrhizal fungi. Soil Science, 161, 575-586. 677 http://dx.doi.org/10.1097/00010694-199609000-00003 678 Wu, W. J., Kuang, L. H., Li, Y., He, L. F., Mou, Z. J., Wang, F. M., Zhang, J., Wang, J., Li, Z. A., Lambers, 679 H., Sardans, J., Peñuelas, J., Geisen, S. & Liu, Z. F. (2021). Faster recovery of soil biodiversity in 680 native species mixture than in Eucalyptus monoculture after 60 years afforestation in tropical 681 degraded coastal terraces. Global Change Biology, 27, 5329-5340. 682 https://doi.org/10.1111/gcb.15774 29 683 Xiao, L., Liu, G. B., Zhang, J. Y. & Xue, S. (2016). Long-term effects of vegetational restoration on soil 684 microbial communities on the Loess Plateau of China. Restoration Ecology, 24, 794-804. 685 https://doi.org/10.1111/rec.12374 686 Xie, H. T., Li, J. W., Zhang, B., Wang, L F.., Wang, J. K., He, H. B. & Zhang, X. D. (2015). Long-term 687 manure amendments reduced soil aggregate stability via redistribution of the glomalin-related soil 688 protein in macroaggregates. Scientific Reports, 5, 14687. 10.1038/srep14687 689 Xu, Y. D., Gao, X. D., Liu, Y. L., Li, S. Y., Liang, C., Lal, R. & Wang, J. K. (2022). Differential 690 accumulation patterns of microbial necromass induced by maize root vs. Shoot residue addition in 691 agricultural Alfisols. Soil Biology and Biochemistry, 164, 108474. 692 https://doi.org/10.1016/j.soilbio.2021.108474 693 Yao, Q., Zheng, L., & Peng, S. (1984). The erosion of weathering crust in granite tableland and its 694 remediation at Xiaoliang. Tropical and Subtropical Forest Ecosystem, 2, 91-106. 695 Ye, Y. Y., Xiao, S. S., Liu, S. J., Zhang, W., Zhao, J., Chen, H. S., Guggenberger, G. & Wang, K. L. 696 (2020). Tillage induces rapid loss of organic carbon in large macroaggregates of calcareous soils. 697 Soil and Tillage Research, 199, 104549. https://doi.org/10.1016/j.still.2019.104549 698 Yu, Z., & Pi, Y. (1985). The path and its effects of rehabilitating vegetation on the tropical littoral eroded 699 land in Guangdong province. Tropical and Subtropical Forest Ecosystem Research, 3, 97-108. 700 Yuan, Y., Li, Y., Mou, Z. J., Kuang, L. H., Wu, W. J., Zhang, J., Wang, F. M., Hui, D. F., Peñuelas, J., 701 Sardans, J., Lambers, H., Wang, J., Kuang, Y. W., Li, Z. A. & Liu, Z. F. (2021). Phosphorus addition 702 decreases microbial residual contribution to soil organic carbon pool in a tropical coastal forest. 703 Global Change Biology, 27, 454-466. https://doi.org/10.1111/gcb.15407 704 Zelles, L. (1999). Fatty acid patterns of phospholipids and lipopolysaccharides in the characterisation of 705 microbial communities in soil: A review. Biology and Fertility of Soils, 29, 111-129, 706 https://doi.org/10.1007/s003740050533 707 Zhang, C., Zhang, X. Y., Zou, H. T., Kou, L., Yang, Y., Wen, X. F., Li, S. G., Wang, H. M. & Sun, X. M. 708 (2017). Contrasting effects of ammonium and nitrate additions on the biomass of soil microbial 709 communities and enzyme activities in subtropical China. Biogeosciences, 14, 4815-4827. 710 https://doi.org/10.5194/bg-14-4815-2017 711 Zhang, J., Li, J., Ma, L. L., He, X. H., Liu, Z. F., Wang, F. M., Chu, G. W. & Tang, X. L. (2022). 712 Accumulation of glomalin-related soil protein benefits soil carbon sequestration with tropical 30 713 coastal forest restoration. Land Degradation & Development. https://doi.org/10.1002/ldr.4192 714 Zhang, J., Tang, X. L., Zhong, S. Y., Yin, G. C., Gao, Y. F. & He, X. H. (2017). Recalcitrant carbon 715 components in glomalin-related soil protein facilitate soil organic carbon preservation in tropical 716 forests. Scientific Reports, 7, 2391. https://doi.org/10.1038/s41598-017-02486-6 717 Zhang, S., Deng, Q., Wang, Y. P., Chen, J., Yu, M. X., Fang, X., He, H. B., Chen, J. L., Xu, P. P., Wang, 718 S. H. & Yan, J. H. (2021). Linkage of microbial living communities and residues to soil organic 719 carbon accumulation along a forest restoration gradient in southern China. Forest Ecosystems, 8, 720 57. https://doi.org/10.1186/s40663-021-00334-8 31