1 Introduction

Fungi, bacteria, and actinomycetes are the three primary microbial groups (Lucie et al. 2019; Luan et al. 2020) that decompose organic matters in soil ecosystems and have a profound effect on the greenhouse gas emission and soil organic carbon (SOC) sequestration (Schutter and Dick 2001; Malik et al. 2018). During the carbon substrate decomposition by microbes, part of the substrate is assimilated as body tissue, while the remaining part is released as CO2, thereby providing an energy source for microbes. The decomposition of organic matters can be directly affected by microbial groups, substrate properties, and soil environmental conditions, especially soil pH (Soares and Rousk 2019; Zhang et al. 2020; Fierer et al. 2021).

Globally, soil pH is critical in shaping microbial characteristics such as community structure and microbial activity and controlling C turnover (Shen et al. 2019; Chen et al. 2022). Soil acidification is an increasingly severe problem in intensive farming, leading to SOC loss and low fertility. Generally, lime materials are added to neutralize acidic soil and increase SOC stocks, which can overcome the problems related to soil acidification (Mosharrof et al. 2021; Wang et al. 2023). Understanding the role of soil pH change by lime addition on the C metabolism of different microbial groups is important to guide the agricultural measure management in farmland soil.

Dissolved organic matter (DOM), as the primary source of microbially accessible C fraction for physiological functions and energy, is very crucial to SOC dynamics and greenhouse gas emission (Bolan et al. 2011; Shen and Benner 2019; Gunina and Kuzyakov 2022). DOM consists of labile and recalcitrant compounds, indicating the chemodiversity of structural complexity, surface properties, and molecular weight (He et al. 2016; Ding et al. 2020). Considering DOM chemodiversity, partitioning DOM into hydrophilic and hydrophobic fractions may be a powerful indication for microbial assimilation of the substrate and conduce to elucidate the substrate mineralization in soil (Marhaba et al. 2003).

Previous studies have demonstrated that bacteria and fungi prefer to utilize hydrophilic and hydrophobic fractions, respectively (Xia et al. 2020; Deng et al. 2021); however, the response of microbial utilization on these two fractions to soil pH change remains unclear, especially in microbial species such as actinomycetes, bacteria, and fungi. Actinomycete activity in neutral soil was higher than in acidic soil due to the domination of neutrophilic species in soil actinomycetes (Zenova et al. 2011). Bacteria and fungi are widely present in acidic and neutral soils, while fungi have a higher adaptation ability to soil pH change than bacteria (Fierer 2017; Pokharel et al. 2021). This implies that soil pH affects substrate utilization to varying degrees.

A historical perspective has suggested lower mineralization of organic material in farmland soil to maintain soil C levels (Yuan et al. 2022; Li et al. 2023). The substrate stability effect, namely the lower mineralization of complex compounds over simple compounds, was regarded as the primary mechanism of SOC sequestration (Spaccini et al. 2000; Logue et al. 2016). An overall perspective held the view that the contribution of a substrate stability effect on SOC sequestration was overestimated (Van Hees et al. 2005), as the mineralization rate was independent of substrate complexities in most cases (Kiikkilä et al. 2012). Whether the substrate stability effect widely exists in natural soils or depends on soil environment conditions is largely unknown. The substrate mineralization rate depends on microbial substrate use efficiency (SUE) and is closely related to substrate complexity, microorganism type, and soil properties (Öquist et al. 2017). Bacteria were found to be sensitive to soil pH, with lower SUE in acidic soil than in neutral soil (Zhang et al. 2020). By contrast, fungi have a higher SUE than bacteria, which are less affected by soil pH (Sánchez et al. 2019). Thus, considering that bacteria dominated the decomposition of the hydrophilic fraction, the mineralization could be higher than that of the hydrophobic fraction in acidic soil.

The overall objective of this study was to examine the effect of soil pH on the microbial utilization of hydrophilic and hydrophobic pools. The 13C-labelled hydrophilic and hydrophobic pools were extracted from 13C-labelled maize biomass and then amended into both acidic and CaO-neutralized Ferralsol. The composition of the microbial community responsible for the decomposition of the hydrophilic and hydrophobic fractions was traced using phospholipid fatty acid (PLFA) analysis, and contents of actinomycetes, bacteria, and fungi PLFAs were quantified, as well as the incorporation of 13C into the above-named biomarkers. We hypothesized that (1) the mineralization of hydrophilic fraction could be higher than that of hydrophobic fraction in acidic soils, and (2) the pH sensitivity of substrate utilization was in an order of actinomycetes > bacteria > fungi. The specific aims of the study were to (1) investigate the impact of soil pH on the mineralization of hydrophilic and hydrophobic fractions and (2) the response of substrate utilization by actinomycetes, bacteria, and fungi to soil pH, and (3) to evaluate the relative contributions of actinomycetes, bacteria, and fungi to the microbial assimilation of hydrophilic and hydrophobic fractions.

2 Materials and methods

2.1 Soil collection and pH adjustment

A loam soil (Ferralsol) with 12% clay, 81% silt, and 7% sand was collected in four replicates from the 0–15 cm layer of arable land in a 20-year cotton-canola rotation located in a subtropical region of China (25°6′10″N, 113°39′59″E). We selected Ferralsol as the target soil sample owing to its distinct characteristics, including its highly weathered development, high microbial activity, prevalence in southern China with intense agricultural management, and ability to support the production of most grains. The soil had pH 5.36 (H2O, w/v 1: 2.5), organic carbon (OC) 15.3 g kg−1, 13C atom percent 1.07%, total dissolved organic carbon (DOC) 26.3 mg kg−1, available nitrogen (AN) 125.3 mg kg−1, Olsen-P 133.6 mg kg−1, and available potassium (AK) 493.5 mg kg−1. After sieving through a mesh < 2 mm together with the removal of visible roots, rocks, and plant residues, a portion of the fresh soil sample was adjusted from acidic to neutral pH (pH 7.31) by adding calcium oxide (CaO) (w/w 0.26%) and thoroughly mixing (Mituniewicz et al. 2016; Lu et al. 2017). After the pH adjustment, the PLFA contents were found to be similar between acidic and neutralized Ferralsol, with the bacteria-derived PLFAs of 14.6 and 15.0 nmol g−1, fungi-derived PLFAs of 2.77 and 3.47 nmol g−1, and actinomycete-derived PLFAs of 4.52 and 5.15 nmol g−1, respectively. The final pH value was 7.31, and DOC was 38.6 mg kg−1, while other parameters were not significantly changed during pH adjustment. Both acidic and neutralized Ferralsol were pre-incubated at 25 °C for 2 weeks in a sealed container with 100% humidity following the soil moisture adjustment to 45% water-holding capacity.

2.2 Extraction of 13C-labelled hydrophilic and hydrophobic fractions

Maize (Zea mays L.) plants were grown in a greenhouse labelling chamber with 13CO2 by pulse labelling (Chowdhury et al. 2014; Ge et al. 2017). After 60 days, the aboveground biomass was harvested and dried at 60 °C. Both 13C-labelled hydrophilic and hydrophobic fractions were extracted from 13C-labelled maize straw (4.12% 13C atom percent) using hot water and hexane, respectively (Miao et al. 2017). Briefly, one portion of the ground maize straw (< 0.25 mm) was extracted with deionized water (1:20 w/v) at 80 °C for 5 h, and another with hexane using Soxhlet extraction (1:20 w/v) for 8 h. The substrates were cooled to room temperature (25 °C), and the hot water- and hexane-extracts were passed through a 0.45-mm fiber membrane, which was denoted as hydrophilic (45.6 mg C g−1 maize straw) and hydrophobic fraction (12.2 mg C g−1 maize straw), respectively. Fine quartz sand (0.25 mm) heated in a muffle furnace at 1000 °C for 24 h was used as a carrier to deliver each fraction into the soil (Sardessai and Bhosle 2002). In brief, 150 g quartz sand was mixed with 500 mL of either hydrophilic or hydrophobic solution and then evaporated into the solid at 45 °C with continuous stirring to ensure a uniform coating on the sand. The hydrophilic and hydrophobic fraction-coated sands had 2.28% and 0.61% C, 3.07% and 3.98% 13C atom percent, and C: N ratios of 6.51 and 20.3, respectively. The C, N, and 13C atom percent of both fractions and 13C-labelled maize straw were measured with an isotope ratio mass spectrometer (MAT 253, Thermo Fisher Scientific, Waltham, MA, USA).

The chemical composition of extracted hydrophilic and hydrophobic fractions was measured using pyrolysis-gas-chromatography/mass spectrometry (Py-GC-MS) (Zhou et al. 2010). The molecular apportionment of the hydrophilic fraction was 61.8% N-containing compounds, 4.31% fatty acids, 8.47% saccharides, 8.89% aromatics, 4.47% alcohols, 0.97% hydrocarbon, and 11.1% phenols. In contrast, the hydrophobic fraction contained a relative proportion of fatty acids at 67.9%, saccharides at 6.78%, aromatics at 1.51%, alcohols at 12.4%, and hydrocarbons at 11.4%. Thus, the hydrophilic fraction had a very high abundance of N-containing compounds and phenols, whereas the hydrophobic fraction was enriched with saturated and unsaturated fatty acids.

2.3 Incubation experiment

Six treatments were established following a completely randomized design with two factors: substrate (control, hydrophilic fraction, and hydrophobic fraction) and soil pH (acidic and neutralized Ferralsol), with four replicates for each treatment. Pristine-acidic (pH: 5.36) and neutral-pH-adjusted (pH: 7.31) soil composite samples (equivalent to 50 g dry soil in a vial) were thoroughly mixed with either hydrophilic- or hydrophobic fraction-coated sand at a concentration of 150 mg C kg−1 soil. Adding pure sand was used as a control (1.23 g). Additional pure sand was added to ensure equal sand for all treatments. Each treatment had four replicates. Each vial was put in a 1 L jar that held a vial with 2 mL distilled water. Jars were sealed with a rubber plug and incubated at 25 °C for 60 days in a dark room. Gas samples, including blanks (30 mL), were collected from the headspace of each jar at 6 h and on days 1, 2, 3, 5, 10, 15, 20, 30, 40, 50, and 60 for the quantification of CO2 and 13C-CO2. The jar was opened for 1 h for aeration after each gas sampling. Destructive soil samples were collected on days 3, 10, and 60 for PLFA and 13C-PLFA analysis.

2.4 Analysis and calculation

2.4.1 CO2 analysis and calculation

Mineralization of added organic C fractions was quantified based on CO2 emissions. CO2 was measured by a gas chromatograph (Agilent 7890A; Agilent Technologies, Santa Clara, CA, USA) equipped with a flame ionization detector at 250 °C. The δ13C-CO2 was analysed with an ultrahigh-resolution isotope ratio mass spectrometer (MAT 253, Thermo Fisher Scientific, Waltham, MA, USA). The CO2 evolved from hydrophilic and hydrophobic fractions was calculated as follows (Rochette et al. 1999):

$${\mathrm{F}}_{{\mathrm{CO}}_{2}}\left(\mathrm{\%}\right)=\left({\updelta }^{13}{\mathrm{CO}}_{2,\mathrm{fraction}}-{\updelta }^{13}{\mathrm{CO}}_{2,\mathrm{control}}\right)/\left({\updelta }^{13}{\mathrm{C}}_{\mathrm{fraction}}-{\updelta }^{13}{\mathrm{C}}_{\mathrm{control}}\right)$$
(1)

where δ13CO2, fraction and δ13CO2, control are the δ13CO2 produced (%) from the soil after adding hydrophilic or hydrophobic fraction and control, respectively. The δ13Cfraction and δ13Ccontrol are the δ13C values (%) of the added fraction (hydrophilic or hydrophobic fraction) and control, respectively.

Mineralization of input C (Cmin) was calculated by the following equation (Straathof et al. 2014):

$${\mathrm{C}}_{\mathrm{min}}\left(\mathrm{mg\, }{\mathrm{kg}}^{-1}\mathrm{\, soil}\right)=12/44\times {\mathrm{CO}}_{2,\mathrm{treatment}}\times {\mathrm{F}}_{{\mathrm{CO}}_{2}}$$
(2)

where 12 represents the atomic mass of C, 44 represents the molecular weight of CO2, CO2,treatment represents the CO2 production (mg kg−1 soil) of treatment at each sampling time, and 150 represents the input C amount (mg kg−1 soil).

$${\mathrm{C}}_{\mathrm{cum}}\left(\mathrm{mg\, k}{\mathrm{g}}^{-1} \mathrm{\,soil}\right)={\sum }_{\mathrm{i}=1}^{\mathrm{n}}\left[{\mathrm{C}}_{\mathrm{min}}\right]$$
(3)

where Ccum is the cumulative mineralization of input C, Cmin is the mineralization of input C at each sampling time, and n is the gas sampling time.

2.4.2 Phospholipid fatty acids analysis and calculation

Microbial assimilation of the added organic C fractions was quantified based on PLFA determination. PLFA exists in microorganisms, including specific and non-specific fatty acids. The fatty acids of i14:0, i15:0, a15:0, i15:1 ω6c, 15:0 DMA, i16:0, a16:0, 16:1 ω5c, 16:1 ω7c, 16:1 ω9c, a17:0, cy17:0 ω7c, i17:0, 17:1 ω8c, 18:1 ω7c, 18:1 ω5c, cy19:0 ω7c, 20:1 ω9c, 21:1 ω8c, 21:1 ω3c, 22:1 ω3c, and i22:0 were assigned as bacterial biomass; 10Me16:0, 10Me17:0, 10Me18:0, 10Me17:1 ω7c, 10Me18:1 ω7c, and 10Me20:0 as actinomycetes (Frostegard and Baath 1996); and 18:1 ω9c, 18:2 ω6c, and 18:3 ω6c as fungi (Frostegard et al. 1993).

The PLFA was extracted according to the methods described by Yuan et al. (2016). Fatty acid methyl esters were identified based on their retention times using GC-FID (MIDI Inc., Newark, DE, USA), followed by quantification based on the recovery of methyl nonadecanoate fatty acid (19:0) (internal standard) (Wang et al. 2014). The δ13C value of individual PLFA components was determined by a trace GC ultra-gas chromatograph (Thermo Electron Corp., Milan, Italy) with a combustion column attached via GC combustion III to a delta-V advantage isotope ratio mass spectrometer (Thermo Finnigan, Bremen, Germany). The amount of 13C-labelled PLFA was calculated using the following equation (Pan et al. 2016):

$${\mathrm{P}}_{\mathrm{i}}={\mathrm{C}}_{\mathrm{i}}\times \left({\updelta }^{13}{\mathrm{C}}_{\mathrm{c}}-{\updelta }^{13}{\mathrm{C}}_{\mathrm{control}}\right)/\left({\updelta }^{13}{\mathrm{C}}_{\mathrm{fraction}}-{\updelta }^{13}{\mathrm{C}}_{\mathrm{control}}\right)$$
(4)

Where Ci (μg C kg−1 soil) is the C content of the individual PLFA, δ13Cc is the corrected δ13C value of each PLFA in soil with DOM addition (%), δ13Ccontrol is the corresponding δ13C value of individual PLFA in the control treatment (%), and δ13Cfraction is the δ13C value of labelled hydrophilic or hydrophobic fraction (%).

2.4.3 Microbial substrate use efficiency calculation

SUE was calculated to evaluate the ratio of added C fraction mineralization to its microbial assimilation. The total PLFA content was calculated as the sum of microbial PLFAs for bacteria, fungi, and actinomycetes. The SUE of fraction-derived C was calculated as the fraction-derived C incorporated into PLFAs relative to its total consumption (including respiration and incorporation into microbial PLFAs) (Hicks et al. 2019).

2.4.4 Statistical analysis

All the data were expressed as the mean ± standard error of four replicates. A two-way ANOVA was conducted to assess the effects of two factors (soil pH with 2-factor levels and substrate type with 3-factor levels: control, hydrophilic fraction, and hydrophobic fraction) on the substrate mineralization and incorporation of added fraction-derived C into actinomycetes, bacteria, and fungi. Residuals of the model were checked for normality and homogeneity, and if conditions were met, the Tukey test was completed (p < 0.05) separately for each sampling time. Principal component analysis (PCA) was conducted based on the 13C-PLFAs data using CANOCO 5.0 for Windows (Microcomputer Power, Ithaca, NY, USA). Figures were created using Origin 8.0 software.

3 Results

3.1 Mineralization of hydrophilic and hydrophobic fractions

The mineralization rates of 13C-hydrophilic fraction were initially significantly higher in the acidic than the neutralized Ferralsol on days 0.5, 1, and 2 (p < 0.05), with CO2 emission reaching 16.8–257.3 mg kg−1 d−1 in the acidic Ferralsol and 2.81–111.5 mg kg−1 d−1 in the neutralized Ferralsol. After this initial period, no significant difference in the mineralization rate of 13C-hydrophilic fraction between soils was detected (Fig. 1b). The cumulative mineralization of 13C-hydrophilic fraction over 60 days was significantly higher (p < 0.05) in the acidic (190 mg CO2 kg−1) than in the neutralized Ferralsol (123 mg CO2 kg−1) (Fig. 1c).

Fig. 1
figure 1

The cumulative CO2 efflux (a), mineralization rate (b), and cumulative mineralization (c) of 13C-labelled hydrophilic (Hydrophilic) and hydrophobic (Hydrophobic) fractions in the acidic and neutralized Ferralsol. Data are expressed as mean ± standard error (n = 4). Different lowercase letters represent significant differences between the treatments in chart (a) or (c) on day 60 (p < 0.05)

Full size image

Higher mineralization rates of 13C-hydrophobic fraction were also initially found in the acidic than in the neutralized Ferralsol within the first 2 days (p < 0.05), with CO2 emission reaching 9.37–43.2 mg kg−1 d−1 in the acidic Ferralsol and 1.16–15.0 mg kg−1 d−1 in the neutralized Ferralsol. However, no significant difference in the mineralization rate of hydrophobic fraction in the acidic and neutralized Ferralsol was found after 2 days of the incubation (Fig. 1b).

3.2 Microbial assimilation of hydrophilic and hydrophobic fractions

The ratio of actinomycetes PLFAs-13C derived from 13C-hydrophilic fraction to the total actinomycetes PLFAs-C was higher in the neutralized than the acidic Ferralsol on days 3, 10, and 60 (Fig. 2a) (p < 0.05). A higher ratio of bacteria PLFAs-13C to total bacteria PLFAs-C in the neutralized than the acidic Ferralsol was detected only on day 3, whereas no significant difference was found in the acidic and the neutralized Ferralsol on days 10 and 60 (Fig. 2b). The ratio of fungi PLFAs-13C to total fungi PLFAs-C in the acidic and the neutralized Ferralsol remained similar (Fig. 2c) (p > 0.05).

The ratio of actinomycetes PLFAs-13C derived from 13C-hydrophobic fraction to the total actinomycetes PLFAs-C was lower in the acidic than the neutralized Ferralsol on day 60 (Fig. 2a) (p < 0.05); in contrast, no significant difference was found on days 3 and 10 (p > 0.05). In contrast to the actinomycetes, a higher ratio of bacteria PLFAs-13C derived from 13C-hydrophobic fraction in the neutralized than the acidic Ferralsol was found at the early incubation stage (day 3) (Fig. 2b). In comparison, pH did not affect the ratio of fungi PLFAs-13C derived from 13C-hydrophobic fraction on days 3, 10, and 60 (Fig. 2c) (p > 0.05).

Fig. 2
figure 2

The portion of 13C-PLFAs in total pools for actinomycetes (a), bacteria (b), and fungi (c) PLFA-C. Different lowercase letters represent significant differences between the treatments in chart (a) or (b) or (c) (p < 0.05). Hydrophilic: 13C-labelled hydrophilic fraction. Hydrophobic: 13C-labelled hydrophobic fraction. 3d: on day 3. 10d, on day 10. 60d, on day 60. Data are expressed as mean ± standard error (n = 4)

Full size image

3.3 Microbial substrate use efficiency of hydrophilic and hydrophobic fractions

The SUE of the fraction was affected by substrate type, soil pH, and incubation time to varying degrees (Fig. 3). The SUE of the hydrophilic fraction was 1.9, 1.2, and 0.9 times larger in the neutralized than in the acidic Ferralsol on days 3, 10, and 60 (p < 0.05), respectively. In contrast, the SUE of the hydrophobic fraction was 1.3 times larger in the neutralized than in the acidic Ferralsol on day 3 (p < 0.05) but not significantly different on days 10 and 60 (Fig. 3).

Fig. 3
figure 3

Microbial substrate use efficiency of 13C-labelled hydrophilic (Hydrophilic) and hydrophobic (Hydrophobic) fraction in acidic and neutralized Ferralsol on day 3, 10, and 60. Different lowercase letters represent significant differences between the treatments (p < 0.05). Data are expressed as mean ± standard error (n = 4)

Full size image

3.4 Substrate-derived C allocation into microbial groups

Regarding the hydrophilic fraction, the portion of bacterial PLFAs-13C in total microbial PLFAs-13C was significantly higher in the acidic Ferralsol (60–68%) than the neutralized Ferralsol (57–64%) on days 3, 10, and 60 (p < 0.05). In contrast, the portion of actinomycetes PLFAs-13C in total microbial PLFAs-13C was significantly higher in the neutralized Ferralsol (14–19%) than the acidic Ferralsol (10–12%) on days 3, 10, and 60 (p < 0.05). For the hydrophobic fraction, on day 60, a higher portion of actinomycetes PLFAs-13C in total microbial PLFAs-13C (p < 0.05) was found in the neutralized Ferralsol (15%) than the acidic Ferralsol (9.0%) (Fig. 4).

The microbial utilization of 13C-hydrophilic fraction followed the order: bacteria (57–68%) > actinomycetes (10–19%) > fungi (4.2–5.5%), and that of hydrophobic fraction ranked as fungi (24–38%) > bacteria (21–29%) > actinomycetes (3.2–15%) during the incubation period (Fig. 4).

Fig. 4
figure 4

The portion of actinomycetes, bacterial, or fungal 13C-PLFAs in total 13C-PLFAs (actinomycetes + bacteria + fungi + universal PLFAs) for 13C-labelled hydrophilic (Hydrophilic) and hydrophobic (Hydrophobic) fractions in the acidic and neutralized Ferralsol on day 3, 10, and 60. Data are expressed as mean ± standard error (n = 4)

Full size image

4 Discussion

4.1 Soil acidity increased the mineralization of hydrophilic but not hydrophobic fraction

Our results demonstrated that the mineralization of hydrophilic fraction was significantly higher than that of hydrophobic fraction in the acidic Ferralsol (p < 0.05), but that of hydrophilic and hydrophobic fraction during the incubation remained similar in the neutralized Ferralsol, which supported the first hypothesis that the substrate stability effect as evident from the mineralization rate depends on soil pH. Since hydrophobic fraction has higher molecular diversity than hydrophilic fraction (Wang et al. 2019; Almeida et al. 2023), structurally simple substrates were mineralized at a higher degree than complex compounds under low soil pH. This could be attributed to the following reasons: first, the mineralization rate of the hydrophilic fraction was significantly higher in the acidic than the neutralized Ferralsol in the initial 2 days of incubation (Fig. 1b), which could be attributed to its greater bioavailability and utilization by bacteria (Fig. 2). Previous studies have shown that early-stage substrate mineralization accounted for a higher proportion over the entire stage, especially for simple compounds, mainly due to the rapid increase of bacteria activity (Bai et al. 2013; Deng et al. 2021). Second, the SUE of the hydrophilic fraction was significantly higher in the neutralized than the acidic Ferralsol (Fig. 3), indicating that microorganisms could use more organic C for microbial biomass synthesis. Moreover, it implied that bacteria and actinomycetes were responsible for the hydrophilic fraction mineralization in acidic soil because of their preference for the utilization of simple substrates (Fig. 4) and the decline of SUE with a pH decrease (Jones et al. 2019).

4.2 Substrate utilization depends on the microbial group, substrate type, and soil pH

The two-way ANOVA showed that pH affected the substrate utilization by microorganisms to different degrees (Table 1). The pH sensitivity of microbial substrate utilization was in the order: actinomycetes > bacteria > fungi. The prolonged impact of pH on actinomycetes substrate utilization (Fig. 2a) and the short-term effect on bacteria (Fig. 2b) indicated that the actinomycete’s function depends more on soil environmental conditions than bacteria. This could be ascribed to the difference in microbial mobility and nutrient acquisition responding to the size of microorganisms, further regulating their substrate utilization (Luan et al. 2020). Given the small body size (generally less than one μm), most bacteria (e.g., Acidobacteria, Alphaproteobacteria, Bacteroides, Deltaproteobacteria, Verrucomicrobia) can move with the soil solution, thus enabling better access to nutrients and substrates (Lucie et al. 2019). Actinomycetes and fungi capture nutrients only through mycelium from their surrounding environment since their movement is restricted by elongation (actinomycetes: 4–5 μm; fungi: 10–200 μm) (Bie et al. 2012; Luan et al. 2020). In the present study, pH did not change fungal substrate utilization due to fungal adaptation to edaphic properties (Xiao et al. 2017; Deveau et al. 2018).

Table 1 Analysis of variance for the effects of soil pH and C type on microbial substrate utilization
Full size table

This study highlighted utilization characteristics of substrates with various complexity by bacteria, actinomycetes, and fungi. Bacteria dominated in hydrophilic fraction utilization, whereas fungi were not specialized in it and only accounted for one-tenth of the hydrophilic fraction utilization compared to bacteria (Fig. 4). In contrast, utilization of the hydrophobic fraction by bacteria was slightly lower than that by fungi (Fig. 4). This indicated that bacteria predominated the microbial utilization of structure-simple substrate, whereas bacteria and fungi controlled the fate of structure-complex substrates. In addition, actinomycetes showed a distinct capacity for substrate utilization among microbial groups, being moderate in the utilization of simple compounds, whereas complex compounds were poorly utilized (Fig. 4). Therefore, bacteria and actinomycetes are important in the decomposition of structure-simple organic materials, whereas fungi and bacteria drive the decomposition of organic substrates with complex structure.

The utilization of organic compounds depends on microbial adhesion to substrates, which is closely related to the hydrophilicity or hydrophobicity of the surface interface between the microorganism and its substrate (Fickers et al. 2005). Our study found that the preference of microorganisms for substrate utilization differed between the microbial groups, which was also previously found by Xia et al. (2020). Most bacteria and actinomycetes have hydrophilic surface properties (Maataoui et al. 2014), explaining their preference for the hydrophilic fraction utilization. The strong preference of actinomycetes for the hydrophilic fraction in the neutralized Ferralsol (Figs. 2a and 4) could be attributed to (1) improved enzyme activity from actinomycetes due to pH rise in a pure culture medium experiment (Basilio et al. 2003) and (2) predominance of neutrophilic species within the soil actinomycetes (Zenova et al. 2011). In contrast, fungi have the advantage of taking up hydrophobic fraction due to the strong hydrophobicity of fungal surfaces, which is related to the production of hydrophobins and surfactants (Lang 2002; Bayry et al. 2012).

4.3 Implications and limitations

We used calcium oxide to neutralize the pH of an acidic Ferralsol, while other edaphic factors remained similar. This study demonstrated that the fate of substrates having various qualities relies on soil pH, which has critical implications for understanding soil C cycling and guiding farmland soil C management (Kunhikrishnan et al. 2016). Simple compounds were more readily mineralized in acidic soil compared to neutralized soil, which could be one of the reasons for the lower OC content in acidic soils than neutral soils (despite the properties of parent material and composition of clay minerals being the same in the present experiment) (Wang et al. 2016; Malik et al. 2018). In contrast, lower mineralization of hydrophilic and hydrophobic compounds and higher incorporation of C derived from these fractions into microbial biomass in neutral soil may contribute to SOC sequestration. Second, the substrate stability effect could be affected by edaphic and non-edaphic properties. Whether these edaphic and non-edaphic properties are positive or negative remains largely unknown.

Various substrate utilization strategies were found among different microbial groups (Malik et al. 2020; Wang et al. 2021), and this study also demonstrated various substrate utilization strategies within the same microbial group (Figs. 5 and 6). The utilization of hydrophobic fraction in 10Me18:0-related actinomycete PLFAs was 1–3 and 0.6–4 times higher than 10Me16:0-related actinomycete in the acidic and the neutralized Ferralsol, respectively (Fig. 6a). Similarly, the hydrophilic fraction utilization in 18:1ω9c-related fungi was 3–4 and 29–30 times higher than that in 18:2ω6c-related fungal PLFAs (Fig. 6e and f). This suggested that a subgroup of actinomycetes or fungi communities can utilize both simple and complex compounds. Additionally, the substrate utilization into cy19:0 ω7c-related bacterial PLFAs increased gradually with time (Fig. 6c), whereas 18:1ω7c-related bacteria peaked on day 3, followed by a decrease for both fractions (Fig. 6d). This indicated that bacteria played a leading role at the early or late stages of substrate utilization, which depended on the bacterial species. This study revealed that although common substrate utilization can occur within the same microbial group, some individual species are harboring particular organic C metabolization strategies. Future studies could focus on specific microbial species’ utilization profiles to uncover keystone microbiota for greater SUE to drive enhanced soil C storage.

Fig. 5
figure 5

PCA of the microbial utilization of substrates (hydrophilic and hydrophobic fractions) in the acidic and neutralized Ferralsol. The individual PLFA marked with green, blue, and red represents actinomycetes, bacteria, and fungi, respectively. The specific individual PLFA with symbol listed as follows: a1, cy19:0 ω7c; a2, 18:1 ω7c; a3, i16:0; a4, i15:0; a5, a15:0; a6, 16:1 ω5c; a7, 16:1 ω7c; a8, 16:1 ω9c; a9, a17:0; a10, cy17:0 ω7c; a11, i17:0; a12, 18:1 ω5c; a13, 15:0 DMA; a14, i14:0; a15, i15:1 ω6c; a16, a16:0; a17, i22:0; a18, 17:1 ω8c; a19, 20:1 ω9c; a20, 21:1 ω8c; a21, 21:1 ω3c; a22, 22:1 ω3c. b1, 18:3 ω6c; b2, 18:2 ω6c; b3, 18:1ω9c. c1, 10Me16:0; c2, 10Me17:0; c3, 10Me18:0; c4, 10Me17:1 ω7c; c5, 10Me18:1 ω7c; c6, 10Me20:0

Full size image
Fig. 6
figure 6

Labelled C in individual PLFA in the acidic and neutralized Ferralsol on day 3, 10, and 60. Different lowercase letters represent significant differences between the treatments in each chart ((a) to (f)) (p < 0.05). Hydrophilic: 13C-labelled hydrophilic fraction. Hydrophobic: 13C-labelled hydrophobic fraction. Data are expressed as mean ± standard error (n = 4)

Full size image

5 Conclusions

Greater mineralization of hydrophilic than hydrophobic fraction was found in the acidic Ferralsol but not the neutralized one, suggesting that the substrate stability effect (namely higher mineralization of simple vs. complex compounds) exists and is controlled by soil pH. The pH sensitivity of microbial substrate utilization followed the order actinomycetes > bacteria > fungi, with fungi being tolerant to the changes in soil pH. Bacteria and actinomycetes dominated the decomposition of hydrophilic fraction, and fungi and bacteria utilized hydrophobic organic molecules. The preference of microorganisms for substrate utilization differed by microbial group, with a substantial selection of actinomycetes for the hydrophilic fraction in the neutralized rather than the acidic Ferralsol; bacteria preferred to utilize hydrophilic fraction, and fungi favored hydrophobic fraction. This study demonstrated that liming Ferralsol results in increased SUE and can potentially result in long-term improvements in soil C storage, which could be used to reveal further the fate of the individual fractions of plant residues in the soil and the management of SOC storage.