Understanding why soil organic carbon (SOC) persists—or is lost—under global change ultimately depends on which biological processes constrain pathways of carbon stabilization. In this issue of Global Change Biology, Du et al. (2025) show that soil food webs, long treated as secondary to decomposition, play an active and context-dependent role in regulating microbial carbon cycling along successional gradients. Predicting SOC persistence has long been central to ecosystem–climate feedback research. Over the past decade, microbial residues have emerged as major precursors of long-lived SOC, particularly mineral-associated organic matter, as formalized in the Microbial Efficiency–Matrix Stabilization (MEMS) framework (Cotrufo et al. 2013). Subsequent work has reinforced the importance of microbial anabolism and physicochemical protection in shaping SOC persistence (Liang et al. 2017). However, most SOC theory remains largely microbe-centered. Soil fauna are typically subsumed within microbial biomass dynamics rather than treated as regulators in their own right. At the global scale, soil fauna—particularly nematodes—mediate substantial carbon fluxes through grazing and trophic interactions. This influence underscores their potential to shape soil carbon cycling in ways not captured by microbial-centric frameworks (van den Hoogen et al. 2019; Neher 2010). It is against this background that Du et al. (2025) integrated replicated field sampling across two long-term successional chronosequences on the eastern Qinghai–Tibet Plateau. These included a glacier-retreat primary succession and a post-disturbance secondary succession, both examined via complementary molecular, biochemical and food-web approaches. Soils (0–10 cm) from six stages in each sequence (n = 60 composite samples) were analysed for SOC and microbial necromass using amino-sugar biomarkers. Nematode trophic structure was quantified using 18S rDNA metabarcoding, while microbial functional potential was assessed by shotgun metagenomics targeting growth pathways, carbon-degrading CAZymes and carbon-fixation genes. Finally, these data were integrated using partial least-squares path modelling (PLS-PM), redundancy analysis and mixed-effects models. This analytical framework explicitly partitioned trophic versus edaphic controls on microbial carbon processing. Rather than emphasizing variation in decomposition rates, their analysis shifts attention from carbon loss to carbon retention. In doing so, it brings food-web structure into closer alignment with microbial metabolic allocation and necromass formation. A central outcome of Du et al. (2025) is that trophic regulation of microbial carbon metabolism depends strongly on successional stage and associated nutrient constraints. In primary successional soils, microbial growth is constrained. These soils are typically characterized by low organic inputs, nitrogen limitation, and weak pedogenic development, so changes in food-web structure have pronounced consequences. Across the primary chronosequence, SOC followed an S-shaped trajectory (from ~12 g kg−1 at early stages to ~6 g kg−1 at mid-succession, then recovering to ~16 g kg−1 at late stages). In parallel, amino-sugar concentrations increased four-fold from 14.8 to 61.8 mg kg−1, a trend tightly correlated with SOC (R2 = 0.56, p 64%) were progressively replaced by predators (> 57% in late stages). This transition coincided with declines in microbial growth genes (−11.5%) and carbon-fixation genes (−13.5%), alongside a marked increase in microbial necromass accumulation. Together, these patterns indicate that increasing trophic complexity redirects microbial carbon flow away from rapid mineralization and towards turnover-driven residue formation. This pattern is consistent with food-web–mediated pathways of microbial residue formation reported elsewhere (Kou et al. 2023). Secondary successional systems present a different picture. These systems develop on soils with residual organic matter and more advanced pedogenesis, often under phosphorus rather than nitrogen limitation. Consequently, they show stronger baseline control of microbial function by soil physicochemical properties. Here, SOC ranged from ~57 to ~34 g kg−1 before partially recovering in late stages. Amino-sugar concentrations remained comparatively stable across stages, although they maintained a strong SOC relationship (R2 = 0.43, p < 0.001). Microbial growth genes declined by 14.0% and carbon-fixation genes by 9.2%. By contrast, shifts in omnivorous nematodes (from ~44% to ~9%) were more closely associated with changes in carbon-fixation and CAZyme profiles than with decomposition per se. Faunal effects therefore persist, but are expressed primarily through modulation of anabolic allocation rather than through direct suppression of decomposition. Similar SOC trajectories along successional gradients may thus arise from mechanistically distinct stabilization pathways, rather than reflecting a single underlying process. Most existing SOC frameworks treat long-term carbon storage primarily as a function of microbial processing efficiency and mineral protection (Cotrufo et al. 2013; Liang et al. 2017), with trophic interactions folded into microbial biomass dynamics, if considered at all. The synthesis by Du et al. (2025) challenges this simplification by showing that food-web structure shapes how microbes allocate carbon among growth, enzyme production and turnover. PLS-PM revealed that nematode functional groups explained more variance in amino-sugar accumulation than abiotic factors in both successions. By comparison, soil pH alone accounted for up to ~29% of the variation in microbial gene profiles. From this perspective, soil fauna regulate not only the magnitude of microbial carbon processing but also the dominant pathways through which microbial carbon persists in soil. These results elevate soil fauna from secondary modifiers to integral regulators of microbial carbon persistence pathways. This view is consistent with recent conceptual advances highlighting faunal regulation of labile versus stabilized soil organic matter pools and the organization of carbon and energy channels within soil micro-food webs (Angst et al. 2024; van Bommel et al. 2024). Microbial necromass formation—widely recognized as a dominant precursor of persistent SOC—thus reflects not only microbial growth efficiency but also trophically mediated trade-offs. These trade-offs occur between turnover-driven residue formation (in N-limited primary succession) and anabolic, carbon-fixation-linked necromass production (in P-limited secondary succession). In this way, Du et al. (2025) extend the MEMS framework from a microbe-centred model to a food-web–aware perspective without displacing its core assumptions (Tao et al. 2023). Looking forward, explicit representation of trophic interactions in Earth system and land-surface models may improve predictions of SOC dynamics, particularly along successional or disturbance gradients where biological organization changes systematically. However, Du et al. (2025) also highlight important limitations. Their inferences are based on space-for-time chronosequences, shallow (0–10 cm) soils and genomic potential rather than realized metabolic activity. These constraints limit direct causal inference. Future work should therefore combine nematode manipulations with isotopic tracing (e.g., 13C-SIP), metatranscriptomics and deeper soil profiling. Such integrated approaches are needed to directly quantify how trophic regulation alters microbial carbon flows into stable mineral-associated pools under contrasting nutrient regimes. At the same time, extending these findings across broader biomes and disturbance regimes will be necessary to assess their generality. Disentangling trophic effects from co-varying edaphic constraints remains a central challenge, underscoring the need for experimental approaches that manipulate food-web structure under controlled nutrient conditions. Such trophic considerations may be particularly relevant in restoration and afforestation contexts, where long-term recovery of soil biodiversity has been shown to shape soil functioning and carbon outcomes (Wu et al. 2021). Overall, Du et al. (2025) make a strong case that trophic interactions represent a consequential, yet long underappreciated, component of soil carbon stabilization under global change. Wenjia Wu: writing – original draft; writing – review and editing. Zhanfeng Liu: writing – original draft; writing – review and editing. This work was supported by the National Natural Science Foundation of China (Grant Nos. 32301565 and 42177289). This work was supported by National Natural Science Foundation of China, 32301565, 42177289. The authors declare no conflicts of interest. This article is a invited commentary on Du et al., https://doi.org/10.1111/gcb.70642. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.
Wu et al. (Sun,) studied this question.