Studies of the decomposition, transformation and stabilization of soil organic matter (SOM) have dramatically increased in recent years owing to growing interest in studying the global carbon (C) cycle as it pertains to climate change. While it is readily accepted that the magnitude of the organic C reservoir in soils depends upon microbial involvement, as soil C dynamics are ultimately the consequence of microbial growth and activity, it remains largely unknown how these microorganism-mediated processes lead to soil C stabilization. Here, we define two pathways—ex vivo modification and in vivo turnover—which jointly explain soil C dynamics driven by microbial catabolism and/or anabolism. Accordingly, we use the conceptual framework of the soil ‘microbial carbon pump’ (MCP) to demonstrate how microorganisms are an active player in soil C storage. The MCP couples microbial production of a set of organic compounds to their further stabilization, which we define as the entombing effect. This integration captures the cumulative long-term legacy of microbial assimilation on SOM formation, with mechanisms (whether via physical protection or a lack of activation energy due to chemical composition) that ultimately enable the entombment of microbial-derived C in soils. We propose a need for increased efforts and seek to inspire new studies that utilize the soil MCP as a conceptual guideline for improving mechanistic understandings of the contributions of soil C dynamics to the responses of the terrestrial C cycle under global change.
Soil carbon (C) stabilization has become an important topic in recent years owing to changes in global climate and atmospheric chemistry1,2. Globally, soil contains a large amount of C—twice that in the atmosphere and more than the C in vegetation and the atmosphere combined3,
Microorganisms have two critical, contrasting roles in controlling terrestrial C fluxes: promoting release of C to the atmosphere through their catabolic activities, but also preventing release by stabilizing C into a form that is not easily decomposed10. To date, research has focused on microbial sources of CO2 with less attention paid to the role of microbial anabolism in generating products that can be sequestered11,12. This has stimulated research considering the direct incorporation of microbial residues (cellular components from both living and senesced biomass) into the stable soil C pool13,14,15,16. For example, recent studies showed that fungal and bacterial necromass are the primary C-containing constituents contributing to the stable soil organic matter (SOM) pool17,18. Thus, the main driver of SOM accumulation under some conditions may not be litter decomposition and transformation per se, but microbial growth that leads to deposition of microbial-derived C into the SOM reservoir via biomass turnover and necromass accumulation14,15,19. Accordingly, any attempt to manage soils for long-term C storage will require an understanding of how to manage microbial-derived C in soil.
To this end, we present a conceptual framework, in which we stress the notion of the microbial carbon pump (MCP, a concept developed in marine systems20) to integrate recent insights into our understanding of how microorganisms regulate soil C dynamics. This framework balances the contrasting functions of microorganisms as agents of SOM decomposition and formation, and it highlights the role of long-term microbial assimilation in the production of organic compounds that are stabilized in soils. Here, we focus on the role of microbial metabolism in soil C dynamics, particularly the contributions of microbial anabolism to soil C storage. We first discuss evolving views on SOM and the metabolic controls of soil microorganisms in soil C turnover. We then discuss microbial anabolism and the soil MCP that control microbial necromass accumulation and stabilization. Finally, we propose areas where we see promise to advance the state of our relevant knowledge.
Soil organic matter and microbial metabolic controls
Long-term C storage in terrestrial ecosystems occurs primarily when plant biomass is stabilized in soils as SOM. SOM is key in maintaining ecosystem productivity and sustainability through its physical, chemical and biological soil functions: as a source of plant nutrients, by increasing infiltration and water-holding capacity, and by enhancing soil structure. Changes in SOM quantity and quality are mainly determined by three factors: abiotic environmental and edaphic variables, types of organic input, and biological activity21.
Historically, studies have focused primarily on relating the magnitude and composition of SOM to non-biological environmental constraints that drive SOM fluctuations1,4, with less emphasis on biological controls over C transformations4,22. There is a huge body of literature on humification and humic substances23, but many aspects of this classical concept of SOM formation are being re-evaluated and displaced as modern analytical tools provide new insights into the chemical nature of SOM22,24,25. Evolving analytical approaches, coupled with studies using physical fractionation as an alternative to traditional alkali extraction, have implied that microbial-derived materials comprise a significant component of SOM26,
Considering the metabolic activities of microorganisms during C transformations, we categorize two major pathways by which microorganisms influence SOM formation: ex vivo (extracellular) modification, in which extracellular enzymes attack and transform plant residues, resulting in deposition of plant-derived C that is not readily assimilated by microorganisms; and in vivo turnover of organic substrates via cell uptake–biosynthesis–growth–death, resulting in deposition of microbial-derived C. Through these two pathways, compounds that are more resistant to further degradation or more readily stabilized are produced by modifying compounds in the original tissues or by forming new compounds through microbial synthesis, such as polymers associated with degradative lignin products and amino sugars. In any event, ex vivo modification implies restructuring or altering molecules by microbial degradative enzymes (that is, purely catabolic processes), while in vivo turnover implies breakdown and resynthesis of molecules, and can suggest a mix of both catabolic and anabolic processes. We regard all C compounds that are deposited to SOM through the in vivo turnover pathway as products of microbial anabolism, as these compounds exclusively originate from constituents of microbial cells.
In the ex vivo modification pathway, transformations of plant materials and residues occur without actual assimilation by microorganisms. Variability in the degree to which plant-derived C is modified depends largely on plant traits31. Additionally, microbial groups may use and modify the structures of plant materials differently, leading to distinct patterns of C use and stabilization, depending on plant and tissue type and which microbial populations are active. For example, microbial communities in forests are better adapted to degrading complex C compounds than microorganisms in grassland32. Yet grassland microorganisms degrade grass litter more effectively than forest litter, while microorganisms in forests do not preferentially degrade forest litter33. Poll et al.34 suggested that fungi preferentially degrade fresh plant litter by releasing a more potent suite of exoenzymes that break down complex materials.
The other key pathway of microorganism-mediated SOM transformation/formation is in vivo turnover, which leads to the deposition of microbial-derived C. Changes in the anabolic capacity of microorganisms and their activity rates directly affect microbial contributions to SOM, regulating the amount of microbial-derived C and the proportion of microbial- versus plant-derived C in the soil C pool. The importance of microbial anabolism is not only embedded in the production of biomass, but also in processing accessible organic compounds and their re-synthesis into the novel forms present in microbial biomass and necromass. For the latter, there is a chance that the C might be reformed into molecules that are relatively more chemically stable or that can be stabilized through associations with soil minerals, such as cell wall fragments, exoenzymes and osmolytes35. The consequences of in vivo turnover are ecosystem specific and dependent on how in situ fungal and bacterial groups grow and assimilate organic substrates. For example, forest soils contain higher microbial biomass than agricultural soils, because fungal groups active in forest ecosystems can contribute more biomass36, probably leading to higher microbial necromass in forests. In forest ecosystems, bacteria contribute more to the soil C pool in broadleaf than in coniferous systems37. Microbial contributions to SOM may shift from fungal to bacterial dominance with increasing land degradation or land-use intensification38. Such changes in microbial residues may affect future C balances as fungal residues are thought to be more persistent in soils than bacterial residues39,40. Significant knowledge gaps regarding microorganism-mediated C preservation in soils include our understanding of the specific compounds involved, their turnover rates, and the nature of the stabilization mechanisms.
Microbial anabolism and the soil microbial carbon pump
Direct microbial contributions to sequestered C were often regarded as minimal, as living microbial biomass makes up <5% of SOM41,
The MCP, a concept first raised by marine researchers20,52, provides a formalized focus for understanding microbial processes in producing persistent organic matter. This draws attention to the microbial transformation of plant-derived C through microbial biomass and ultimately into the stable soil C pool12,19. In both aquatic and terrestrial systems, microorganisms can grow and turn over rapidly during organic matter decomposition, leaving necromass, part of which can be stabilized in the environment. An analogue to the marine MCP20 has been hypothesized to operate in soils11,13,53, but the idea of a soil MCP has received little study. Here, we couple the MCP concept with the ability of the synthesized compounds to be stabilized on mineral surfaces and within soil structures—a phenomenon we define as the ‘entombing effect’.
A schematic diagram of this conceptual framework displays the relevant pathways and consequences of microbial growth, metabolism and death (Fig. 1). In this diagram, microbial controls on terrestrial C cycles, driven by catabolism and/or anabolism, are emphasized to demonstrate the microorganism-mediated C transformation process—where the soil MCP, particularly via in vivo turnover, strengthens the entombing effect on soil C storage. Accordingly, the efficiency of the soil MCP, as affected by factors including internal features (for example, microbial physiology, chemical stability and physical interaction) and external constraints (for example, edaphic variables and global change drivers), will play a significant role in a soil's capacity to retain persistent organic C.
Hypothesis-driven perspectives and future directions
By synthesizing knowledge on microbial C cycling in soils, we illustrate that the microbial entombing effect can drive the generation and sequestration of persistent soil C. A conceptual framework encompassing the soil MCP not only provides a theoretical structure for studying microbial anabolism in soil C storage, but also enables us to generate testable hypotheses regarding ecosystem responses to external disturbance and the determinants of SOM chemistry, as well as other promising future research directions.
The balance between microbial priming and entombing regulates the stable soil C pool. The stable soil C pool and its dynamics are an enigma that has puzzled scientists for decades. We are aware that small changes in the global balance between C inputs to and outputs from soil can alter atmospheric CO2 concentrations. However, the heterogeneity of SOM makes detecting changes in pool sizes difficult, and also obscures our understanding of ‘unseen’ mechanistic controls on those pools.
As an ironic twist, soil microorganisms that primarily decompose SOM can also drive C sequestration by producing stable or stabilized SOM. We suggest that the magnitude of soil C storage is largely controlled by the balance between microbial catabolic activity, which releases C as CO2, and anabolic activity, which contributes senesced microbial biomass. Therefore, these two functions must be considered simultaneously when investigating stable soil C. Following addition of new external C, CO2 evolution may increase dramatically (for example, by up to 400%54) by stimulating microbial decomposition of existing stabilized SOM, a phenomenon known as the priming effect55. In contrast to the loss of primed C, the same microorganisms also synthesize new organic compounds as they build biomass and ultimately part of their necromass will be stabilized, a phenomenon defined above as the entombing effect. Here, we hypothesize that litter input quality regulates stable soil C pool dynamics by driving shifts in microbial community activity and composition (Fig. 2). In our conceptual framework with a particular focus on microbial control over C loss/gain in the stable soil C pool, the dynamics of the stable soil C pool are determined by the balance between microbial priming and microbial entombing (Fig. 2a). More specifically, we hypothesize that fungal dominance, driven by low-quality litter inputs36, will lead to not only a higher rate of CO2 release by decomposing relatively stable C, but also greater accumulation of microbial-derived C by incorporating more fungal necromass into the stable C pool. Both effects will determine the change in pool size, so the net effect on the stable soil C pool by microbial priming and entombing can be negative (C loss), positive (C gain) or zero (no change) at different temporal scales. We propose that fluctuation of stable soil C pool size, under three different priming-entombing scenarios, can be predicted by mathematical simulation56 (Fig. 2b).
This conceptual framework is simplified and does not consider all possible C-flow channels, some of which should be included for particular systems. For example, leaching is a channel for C loss, which may translocate C deeper in the soil profile and then move it with below-ground water flow. In general, although the priming and entombing channels can be differentiated at a conceptual level, it is often difficult to discriminate between them. Consequently, while individual studies have shown that these channels all occur (probably simultaneously), we are currently unable to rank the relative importance of each acting channel and how this might vary under different conditions. In order to elucidate the mechanisms and improve our ability to predict C cycling, this hypothesis calls for more studies to investigate these processes—particularly the entombing effect.
Ex vivo modification versus in vivo turnover controls the chemical fate of soil C. Decomposition drives ecosystem C cycling, and plays a central role in shaping the composition and spatial distribution of below-ground C. SOM composition and complexity have the potential to significantly alter soil characteristics and functions such as organo-mineral interactions, environmental sustainability, and soil C stabilization. Thus, it is critical to identify the factors linking the quality of organic matter inputs, residue decomposition and SOM storage, and to understand the origin and consequences of SOM chemistry. However, it remains contentious how initial litter chemistry changes during decomposition and what roles the dual microbial pathways (ex vivo modification and in vivo turnover) play in controlling these changes. One traditional opinion is that the differences in initial litter chemistry will eventually converge towards a set of common compounds that are more resistant to decay29,57. Contrary to this idea, another viewpoint believes that the initial differences in litter chemistry will persist into the late stages of decomposition when external C inputs are incorporated into SOM58,
We hypothesize that microbial anabolism supports the chemical convergence such that over the course of decomposition, chemically unique inputs become more similar as they are assimilated into microbial biomass (Fig. 3a,b). In this case, the varying compositional chemistry of different external litter inputs would tend to converge and the distinct chemistries of the initial litter types would become indistinguishable after intensive microbial turnover via the in vivo turnover pathway. Further, we hypothesize that the relative contributions from in vivo turnover and ex vivo modification by microorganisms ultimately will determine the C chemical structure and complexity of SOM produced during the course of decomposition (Fig. 3b-d), enabling the chemistry of the same litter input to diverge. Specifically, greater dominance of ex vivo modification relative to in vivo turnover will result in greater C chemical complexity of SOM.
Our hypotheses on the determinants of soil C chemistry are difficult to test in field or lab studies. Natural ecosystems are open systems with continuous inputs of plant (and animal) litters, which integrate with pre-existing compounds and potentially dilute the impacts of microorganism-mediated decomposition on SOM chemistry. Although stable C isotopes have the power to disentangle the newly added C from already existing C, isotopic techniques that add labelled litters in the laboratory are not fit for long-term studies of C dynamics. Another issue we need to consider is the variance in cell C chemistry across microbial species. In our conceptual framework, we assume that the differences in C chemistry between fungal and bacterial cells are insignificant compared with the variations among diverse plant litters, which consequently leads to chemical convergence after continuous substrate assimilation by microorganisms. This convergence is plausible when microbial products become an increasingly dominant portion of the remaining mass of a particular litter cohort as decomposition proceeds since many biomass constituents are similar among different taxa57.
Despite its importance to soil C storage, the incorporation of microbial-derived components into the stable soil C pool has not received enough attention yet. Recent research has made it clear that microbial-derived C is ubiquitous and relatively stable against decomposition when it becomes physically protected. At present, our understanding of the differentiation in C allocation and microbial processing through ex vivo modification and in vivo turnover is far from satisfactory and contributes to the uncertainties in quantifying and predicting C disposition and dynamics under global change. Hence, studies on the stabilization of microbial-derived C in soils are indispensable for advancing knowledge of soil C dynamics and stability, improving the structure of global C cycling models to reduce predictive uncertainties, and helping to inform strategies to maximize C sequestration and craft climate policy. In particular, the soil MCP conceptualizes a sequestration mechanism during microbial in vivo turnover—that is, microbial generation of new soil C via accumulating anabolism-induced necromass, part of which can persist in soils. Together, the soil MCP and its related entombing effects, which are mechanistically connected with the terrestrial C cycle and global climate, serve as conceptual models for guiding the multidisciplinary approaches required to integrate empirical, theoretical and predictive perspectives that will be necessary for understanding the contributions and importance of microbial necromass in the formation and stabilization of SOM, as well as its resilience and vulnerability to global change.
How to cite this article: Liang, C., Schimel, J. P. & Jastrow, J. D. The importance of anabolism in microbial control over soil carbon storage. Nat. Microbiol. 2, 17105 (2017).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
We would like to thank X. Zhang, T. Balser, J. Tiedje, E. DeLucia, J. Kao-Kniffin and M. Kästner for their help with the evolution of ideas and concepts, along with the career development of C.L. We would like to thank K. Wickings and H. Gan for valuable inputs during preparation of the manuscript, and J. Lehmann for constructive comments and suggestions to improve the manuscript at a later stage. Particularly, we would like to thank X. Zhu for enhancing the visual quality of the figures. This work was supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (No. XDB15010303), the National Natural Science Foundation of China (No. 41471218), the National Key Research and Development Program of China (No. 2016YFA0600802), and the US Department of Energy, Office of Science, Office of Biological and Environmental Research. The grants or other support to C.L. from the National Thousand Young Talents Program of China and the Alexander von Humboldt Foundation of Germany are also acknowledged with gratitude.