Effects of Laboratory Warming on Active Soil Organic Matter and Bacterial Diversity During the Long-Term Decomposition of Forest Litter in Soil Microcosms

Abstract

To investigate the combined impacts of temperature and plant residues on the mineralization capacity of soil organic matter, in addition to the impact on the taxonomic composition and activity of the soil microbiome, a 364-day experiment involving gray forest soil (Greyzemic Phaeozem Albic) was conducted under controlled laboratory conditions. Three substrate treatments were applied, control (C), amendment with aspen leaves (L), and amendment with aspen branches (B), combined with three temperature regimes (2, 12, and 22 °C). The results showed that long-term experimental warming reduced microbial alpha diversity (number of species and richness), increased microbial respiration and decomposition rates, and altered community composition. Over the year, the percentage of mineralization of added carbon was higher for leaves (29.9%–57.8%) than for branches (20.1%–47.6%). The efficiency of organic matter decomposition increased by 1.5- to 2-fold between 2 and 12 °C. Across all treatments, Proteobacteria were found to be the dominant phylum. According to α-diversity analysis, leaves served as the most preferred substrate for enhancing species representation. β-diversity analysis results indicated that temperature was the most significant factor shaping the microbial community’s structure. Our research findings provide new insights into soil organic matter formation and highlight the need for further research on microbial functional genes.

1. Introduction

Soil organic matter (SOM) is a crucial component of terrestrial ecosystems and a major reservoir in the global carbon (C) cycle [1]. The dynamics of SOM regulate green-house gas concentrations, making it a significant factor in climate change [2,3]. Additionally, SOM fulfills a variety of biological, chemical, and physical functions, including improving soil structure and water retention, facilitating cation exchange, and increasing P availability [4,5]. Both the quantity of plant litter input and the rates of organic matter decomposition in soil and vegetation are controlled by climate-related factors, directly influencing soil C stocks [6,7].

A significant portion of plant primary production enters the soil as fresh plant material, forming the detrital carbon pool [8]. The remains of aboveground and belowground biomass are utilized by mesofauna and microfauna communities, epiphytic and soil microorganisms [9]. Through enzymatic breakdown, organic materials are decomposed into individual fragments and particles (Particulate Organic Matter, POM), and large and small biopolymer molecules are formed, which are either stabilized (Mineral-Associated Organic Matter, MAOM) or continue to decompose, mineralizing into carbon dioxide [3,10,11]. The role of microbial necromass in the creation of MAOM is well documented. While microorganisms and plants are the primary sources of MAOM, it is still unclear to what extent each source contributes. Based on the Microbial Efficiency-Matrix Stabilization (MEMS) hypothesis, it is widely accepted that microbial necromass is a major source of carbon in MAOM [12]. However, current research findings do not support this hypothesis, indicating that earlier estimates of microbial-derived MAOM may have been overestimated. Microbial inputs account between 34% and 47% of the MAOM pool according to a stoichiometric approach based on two-pool, isotope-mixing models [13]. In another global dataset, it was also found that the contribution of plant litter materials to MAOM ranged from 53% to 66% [14]. Ecosystem type can affect necromass contribution. In a study, it was shown that microbial necromass contributes approximately 50% of soil organic carbon in croplands and grasslands, versus 35% in forests [15].

Dissolved organic matter (DOM) in soil is a highly mobile and active fraction of soil organic matter (SOM) that consists of water-soluble molecules with relatively strong biological activity. Both the global carbon cycle and biogeochemical processes, including soil formation, mineral weathering, and pollutant transport, depend on DOM [16]. The main sources of soil DOM include plant-based compounds, soil humus, root exudates, and the microbial degradation of SOM [17].

The movement of plant biomass into the soil, the incorporation of plant residues into SOM, and the storage of carbon within the stable SOM pool are natural stages in the soil sequestration of atmospheric C by soils [18]. In forest ecosystems, the organic inputs to soil consist of a variety of residues. Leaves account for roughly 22%–80% of the total litterfall, while other types of litter include branches, bark, and cones [9,19,20]. Substrate quality is one of the main factors that influence litter decomposition rate. An important characteristic reflecting the quality of litter and determining its decomposition rate is the C/N ratio: the higher this ratio, the lower the quality of the litter [21]. However, some researchers argue that although the C/N ratio provides a general measure of the de-composability of organic materials, it is not always suitable for assessing the biological quality of the decomposing materials or decomposition dynamics [9]. More precise indicators include the N, lignin, and polyphenol contents, in addition to the lignin/N ratio [9,22]. Plant litter often possesses a high C/N ratio because it primarily consists of cell wall materials rich in carbohydrates [23]. Cellulose is the main carbohydrate present in primary cell walls and is linked to hemicellulose and embedded in a pectin-rich gel-like matrix. Conversely, lignin is deposited inside and around cellulose and binds to hemi-cellulose in secondary cell walls [24]. Because leaves possess higher nutrient content and a lower lignin concentration than branches, they usually degrade more rapidly, because high lignin concentrations have a major effect on reducing litter decomposition rates [25,26]. Under favorable climatic conditions, the decomposition of plant residues is primarily limited by the quality of the organic material; in comparison, under unfavorable conditions, it is limited by environmental factors [27].

Among abiotic factors related to climate, temperature is one of the most important [28,29]. Rising temperatures can increase the input of organic residues into the soil but also simultaneously accelerate their microbial decomposition [30,31]. With increasing temperature, C losses from the soil tend to exceed C inputs, since the hydrolytic processes are much more temperature-sensitive than those of photosynthesis [32]. The long-term effects of global warming on SOM remain debated, particularly regarding the extent to which temperature changes will affect SOM availability to microorganisms [33]. Some researchers suggest that after a brief burst period of mineralization affecting the active (labile) pool of SOM, the process as a whole will slow down [29,34,35,36]. Other researchers posit that C mineralization may remain stable or even increase, as the decomposition of the large stable pool of SOM will intensify [37,38]. Regardless, temperature changes will alter the balance between labile and stable SOM components, which may impair SOM functioning [39]. The results of incubation experiments have shown that the degree of availability of substrates from plant residues significantly influences the temperature dependence of soil respiration [36]. In one study, as the quality of organic C in plant residues decreased, hydrolytic activity became more sensitive to temperature fluctuations [40].

Temperature also affects the biotic factors of litter decomposition by altering the structure and activity of soil microbial communities [41,42]. Microbial communities regulate not only soil energy balance but also accelerate the rate of C turnover [43,44,45]. For example, there is evidence of significant changes in soil microbial communities in response to warming after a short-term warming period (up to 12 months) [22,46] and long-term warming period (8 years) [47]; in comparison, other researchers found no significant changes under similar conditions [48]. In our previous study, we showed that temperature affected the succession of cultivated saprotrophic bacteria in soil microcosms during the early stage of forest litter decomposition [49]. The relationship between microorganisms and the temperature sensitivity in plant residues and soil organic matter decomposition also remains unclear [50,51].

Global warming poses a significant threat to soil microbial diversity. The results of a global meta-analysis reveal that bacterial and fungal diversity is greatly reduced by warming, particularly in nutrient-poor soils and during prolonged warming periods [52,53].

In this study, we address current research gaps by investigating the influence of temperature on the decomposition of forest litter residues, the formation of the biologically active SOM pool, and the taxonomic composition and activity of the soil microbiome. Based on our hypothesis, microbial mechanisms regulating soil C dynamics will shift significantly under global warming, which will be reflected in changes in microbial community composition and decomposition activity, ultimately affecting the soil’s capacity to sequester C.

We conducted a 364-day incubation experiment using soil microcosms from a mixed forest in the Moscow region, with the addition of two types of plant residues (fragments of aspen leaves and branches). Microcosms were incubated under aerobic conditions at three temperature regimes characteristic of the southern Moscow region: (1) spring and autumn, (2) summer, and (3) under simulated warming. We measured the dynamics of CO₂ production to assess the decomposition of plant residues and the characteristics of active soil organic matter and examined soil microbial communities by means of 16S rRNA amplicon sequencing. Our study provides a scientific basis for predicting the trends in SOM changes and microbial community in forest soil under future global changes.

2. Materials and Methods

2.1. Soil and Plant Residue Sampling

Soil samples used for incubation experiments were collected in June 2023 from a secondary deciduous forest near Pushchino town, Moscow region (54.8° N, 37.6° E), where the mean annual temperature (MAAT) is 8.9 °C and the mean annual precipitation (MAAP) is 1005 mm. Samples were taken from a depth of 0–20 cm in five replicates. The tree layer vegetation in this area was represented by Norway maple (Acer platanoides), birch (Betula sp.), aspen (Populus tremula), and alder (Alnus). The soil was classified as gray forest loam developed on cover loess, underlain by moraine (Greyzemic Phaeozems Albic). Soil properties were determined using standard methods at the core facility “Physico-chemical methods of soil and ecosystem research” of IPCBPSS RAS, Puschino, Moscow region, Russia, and are presented in

Table 1. Basic parameters of the experimental gray forest soil.

Soil’s Parameter, Dimension	Parameter’s Value
Texture	Loamy a, coarse b
Clay, %	46.4 ± 0.7
Silt, %	44.8 ± 0.5
Sand, %	8.8 ± 0.5
${\mathrm{p}\mathrm{H}}_{{\mathrm{H}}_2\mathrm{O}}$	5.28 ± 0.02
SOC, %	2.53 ± 0.04
TN,%	0.15 ± 0.03
C:N	12.13
$\mathrm{N}\text{-}\mathrm{N}{\mathrm{H}}_4^{+}$, mg/100 g	0.29 ± 0.04
$\mathrm{N}\text{-}\mathrm{N}{\mathrm{O}}_3^{-}$, mg/100 g	0.27 ± 0.01

Note. ^a US Department of Agriculture textural classes, ^b FAO textural classes.

.

In October 2023, two types of plant residues were collected: small branches and pieces of aspen leaves. Plant residues (PR) were dried at 65 °C, ground to a size of <0.5 mm, and analyzed to their determine carbon and nitrogen content using a CHNS 932 analyzer (LECO, St. Joseph, MI, USA). The leaves were characterized by a C:N ratio of 46.4 with total nitrogen (N) content of 0.92 ± 0.02%, and total carbon (C) content of 42.85 ± 0.58%. For branches, the corresponding values were 63.1, 0.74 ± 0.03% and 46.54 ± 0.45%, respectively.

2.2. Design of Incubation Experiments

Aerobic microcosm incubation experiments were carried out in 100 mL glass flasks, each containing 10 g of air-dried soil. Microcosms were incubated for 364 days in thermostats under three temperature regimes: 2 °C (representing the mean spring and autumn temperature of the southern Moscow region), 12 °C (representing the mean summer temperature), and 22 °C (representing global warming). To trace the transformation of plant residues in soil, fragments of aspen leaves or thin branches were added at a rate of 0.5% of soil dry weight.

In total, 135 microcosms were randomly established across nine experimental treatments: (1) control soil (C) incubated at 2, 12, and 22 °C, (2) soil with leaves (L) incubated at 2, 12, and 22 °C, and (3) soil with branches (B) incubated at 2, 12, and 22 °C. A total of 27 microcosms (3 substrates × 3 temperatures × 3 replicates) were used for continuous CO₂ measurement throughout the incubation period, and DNA was extracted destructively from 27 microcosms for the sampling time periods of 0, 6, 9, and 12 months (27 × 4 time points). Samples used for molecular analysis were designated as “month-substrate-temperature”. For example, “6L22” indicates samples that were incubated for 6 months with leaves at 22 °C.

Soil moisture was regularly monitored and maintained at 30% of soil weight (70% WHC) throughout the incubation period.

2.3. Measurement of the Intensity of PR Decomposition and Characteristics of Active Soil Organic Matter Based on Microbial Respiration

Microbial respiration (CO₂ production) was used as a proxy for the rate of decomposition of organic amendments and formation of active organic matter in soil.

Before measurement, 27 glass flasks used for continuous CO₂ measurement were sealed with silicone stoppers rather than Parafilm M laboratory film, and zero-time samples were taken and analyzed immediately. The vials were then incubated for 24 h at the appropriate temperatures and the CO₂ content was analyzed. Thereafter, the flasks were opened, ventilated, sealed with Parafilm M, and incubated until the next measurement in thermostats.

The first measurement was performed 24 h following experiment commencement and then daily during the first and second weeks, three times a week during the second and first decades of the fourth month of incubation, and from the second decade of the fourth month of incubation, once every seven or ten days [54]. A total of 67 gas sample collections were carried out throughout the incubation period.

CO₂ was quantified in the gas phase of the flasks using a Cristal Lux 4000 M GC device (Chromatek CJSC, Yoshkar-Ola, Russia). The C-CO₂ flux rate (mg 100 g⁻¹ day⁻¹) was calculated from the difference in CO₂ concentrations over the exposure period. To recalculate the content of CO₂ in the gas phase to the value of release from the soil samples, the volumes of the vial and gas sample, in addition to the soil mass and time of incubation, were taken into account, and the carbon dioxide content was expressed as recalculated to carbon (C-CO₂).

The cumulative amount of C-CO₂ production (mg 100 g⁻¹) was determined by adding the amount of C-CO₂ at each measurement time point to the sum for the previous time points. In the soil variants with PR, the cumulative amount of C-CO₂ produced from organic residues was determined by subtracting the C-CO₂ released from the control variant soil.

The decomposition rate constant of the studied soil samples and organic residues at different incubation periods was calculated using Equation (1):

$$\mathrm{k}=-({\displaystyle \frac{\ln \left(\frac{Mt}{M0}\right)}{t}})$$

(1)

where k is the decomposition constant, days⁻¹; M_t is the carbon content (C_org) in the sample at different observation periods, % of the initial value; M₀ is the initial C_org content in the sample (at the beginning of observations), %; and t is the decomposition duration, days [55].

The amount of biologically active (potentially mineralizable) organic matter (C₀) was calculated based on the amount of C-CO₂ mineralization losses, using a single-component first-order kinetic Equation (2):

C_t = C₀ (1 − exp(−kt))

(2)

where Ct is the proportion of C-CO₂ losses (% of C_org in the sample) over time t (days); C₀ is the content of potentially mineralizable carbon, % of the initial C_org in the sample; and k is the mineralization rate constant, day⁻¹ [54].

By substituting the cumulative amounts of C-CO₂ (C_t, mg 100 g⁻¹) released during the incubation period (t, days) into Equation (2), the content of active organic matter carbon (C₀, mg 100 g⁻¹) in the soil at the beginning of the incubation period was calculated [54].

The mineralization rate index (MRI), which represents the highest rate at which organic matter is converted into inorganic forms by microorganisms (mg C 100 g⁻¹ day⁻¹), was calculated using Equation (3):

MRI = C₀ × k₁

(3)

where C₀ is the amount of biologically active (potentially mineralizable) organic matter and k₁ is the mineralization constant [56].

The temperature coefficient (Q₁₀), which characterizes the temperature sensitivity of substrate decomposition rate by increasing the intensity of CO₂ release with a 10 °C increase in incubation temperature, was calculated using Equation (4):

Q₁₀ = (K₂/K₁) ^{[10/(T₂ − T₁)]}

(4)

where K₂ is the substrate decomposition rate at the upper temperature value T₂ and K₁ is the rate of substrate decomposition at the lower temperature value T₁ [30].

2.4. DNA Extraction, Quantitative PCR and Illumina 16S rRNA Sequencing

Soil microbial communities were characterized using DNA amplicon sequencing. Total microbial DNA in 0.25 g soil samples across four time points (month 0 and months 6, 9, and 12) was extracted using the DNeasy Power Soil Pro Kit (Qiagen, Hilden, Germany) and a Precellys 24 homogenizer (Bertin Technologies, Montigny-le-Bretonneux, France) at 6500 rpm (40 s). Each sub-sample from the same soil underwent independent extraction. The quantity and quality of the extracted DNA were assessed using a Nanodrop 1000 Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA).

Taxonomic analysis of the bacterial community was conducted using universal primers 341F and 805R [57] specific to the V3-V4 regions of the 16S rRNA gene. PCR was performed as described in [58].

Further library preparation was carried out in accordance with the manufacturer’s instructions in the MiSeq Reagent Kit Preparation Guide (Illumina) (https://support.illumina.com/downloads/16s_metagenomic_sequencing_library_preparation.html) (accessed on 10 June 2024). After obtaining the amplicons, the libraries were cleaned and mixed in equimolar amounts using the SequalPrep™ Normalization Plate Kit (Thermo Fisher Scientific, Waltham, MA, USA). The resulting pool was analyzed by means of capillary electrophoresis and then sequenced at the resource center “Genomic Technologies, Proteomics, and Cell Biology” of the ARRIAM, Russia, using the Illumina MiSeq (2 × 250 bp) system (Illumina, San Diego, CA, USA), according to the manufacturer’s instructions.

2.5. 16S rRNA Gene Amplicon Analyses

Initial data processing, including demultiplexing and adapter removal, was performed using MiSeq System Suite (v.4.1.0) Illumina software.

For subsequent denoising, sequence merging, removal of chimeric reads, recovery of original phylogenetic types (ASV, amplicon sequence variant), and further taxonomic classification of the obtained ASVs, we used the software package phyloseq [58] and the SILVA ribosomal DNA database release 138 [59], with all work being carried out in the R software environment (v4.2.3) [60].

The QIIME2 (v.2024.5) software program was used to illustrate the taxonomic analysis results [61]. Heatmaps, bar graphs, and alpha- and beta-diversity plots were produced using the phyloseq (v1.30.0) package. The R packages tidyverse [62] and vegan [63] were used for statistical analysis and data visualization.

2.6. Data Analysis

The data were tested using the Shapiro–Wilk test and showed normal distribution (p-value < 0.05). Every assay was run in triplicate. Excel for Microsoft Office 10 (v14.0.7248.5000) was used to organize the collected data. Microsoft Excel and SPSS Statistics v. 17.0 software were used to calculate the means and standard deviations (SD) of three separate studies, representing the data that are displayed.

For inter-group comparisons, the datasets were analyzed via single-factor analysis of variance (ANOVA); thereafter, one-way ANOVA and Tukey’s post hoc tests were used (p < 0.05). Statistical processing was conducted with SPSS 25.0 (IBM SPSS, Somers, NY, USA). A p-value of less than 0.05 was deemed statistically significant in all tests.

3. Results

3.1. Dynamics of C-CO₂ Release from Soil Samples

Differences in the dynamics and intensity of the basal soil respiration (C variants) and respiration induced by the incorporation of substrates (L and B treatments) were observed in microcosms incubated with different temperature regimes (Figure 1).

The decomposition of SOM and plant residues in the soil followed classical two-phase dynamics, with rapid and slow rates at the beginning and end of incubation, respectively. The main peak of C-CO₂ release from the soil occurred on the 2nd to 4th day of incubation (Figure 1). At all temperature regimes and time points, the emissions from microcosms with plant residues were significantly higher compared to the control soil and in microcosms with aspen leaves, as compared with those containing branch (p < 0.05). As expected, the higher the incubation temperature, the greater the production of C-CO₂.

For a 2 °C variant, the rate of basal respiration (without PR) was low during the entire period of observation, with a slight increase in intensity (p < 0.05) up to 2.32 ± 0.03 mg C-CO₂ 100g⁻¹ day⁻¹ during the first three days after incubation initiation. The addition of leaves and branches did not significantly change either the dynamics or the intensity of the process (p > 0.05) (Figure 1).

The processes at 12 and 22 °C have entirely different dynamics from those at 2 °C. In 12 °C variations, the addition of leaves caused a gradual increase over the first nine days from 3.27 ± 0.04 to 7.59 ± 0.23 mg C-CO₂ 100 g⁻¹ day⁻¹ (p < 0.05) and then a dramatic decline to 2.44 ±0.21 mg C-CO₂ 100 g⁻¹ day⁻¹ (p < 0.05) over the following fifteen days. When branches were added, activity increased less sharply for the first seven days up to 4.58 ± 0.25 mg C-CO₂ 100 g⁻¹ day⁻¹ and then plateaued at around the same level for the following fourteen days (p > 0.05) before declining (Figure 1). For all substrate options, including the control (C), a sharp increase in activity was observed in the first 48 h, followed by a rapid decline (Figure 1). The activity level depended on the substrate and was 12.05 ± 0.01, 21.28 ± 0.47 and 15.75 ± 0.57 mg C-CO₂ 100 g⁻¹ day⁻¹ for C, L, and B, respectively.

In the control soil (C), a steady-state condition, when the rate of CO₂ production and loss from the microcosm stabilized, was reached after one month of incubation. In contrast, in the treatments with plant residues, this steady state was achieved after approximately two months. Notably, throughout the entire incubation period, including the final stage (months 7–12) the CO₂ release remained approximately twice as high in the L and B treatments compared to the control. A stimulating effect of elevated temperature was also detected.

The trajectory of cumulative CO₂ emission curves clearly demonstrated the combined effects of temperature and plant residue type (Figure 2). While all cumulative curves exhibited a similar shape, the treatments with leaf and branch additions incubated at 22 °C did not reach a plateau even after 364 days. Temperature has a major impact on the total amount of C released. For example, leaf addition led to a total release of 174.23 mg C at 2 °C, which increased to 298.6 mg and 481.74 mg at 12 and 22 °C, respectively (

Table 2. Total cumulative C-CO₂ production value, mg C-CO₂ 100 g⁻¹, for 364-day incubation under different temperature regimes.

Variant	2 °C	12 °C	22 °C
Control (C)	61.36 ± 1.28	104.54 ± 1.88	207.52 ± 1.71
Leaves (L)	174.23 ± 0.05	298.60 ± 4.33	481.74 ± 3.62
Branches (B)	129.56 ± 0.96	255.65 ± 0.74	446.00 ± 1.72

).

During the first 30 days of incubation, 24 to 44% of the C-CO₂ released over the year of the experiment was accounted for by the soil. Cumulative C-CO₂ production values decreased in the following temperature series in all variants: 22 °C > 12 °C > 2 °C. In turn, based on the total amount of C-CO₂ released, the ranking of the organic materials tested was as follows: aspen leaves > aspen branches > SOM.

3.2. Decomposition of Plant Residues and Formation of the Active SOM Pool

During the incubation period, the substrates decomposed, and the intensity and kinetic parameters of the process depended not only on the type of substrate but also on the temperature. Over the annual incubation period, 2 to 7% of SOM was mineralized in the soil without the addition of litter material (

Table 3. Decomposition of soil organic matter and plant residues during annual incubation depending on temperature.

Temperature	Variant	Corg, % of Initial	k1, Day−1
2 °C	Control	2.6 ± 0.1	0.019 ± 0.000
	Leaves	29.9 ± 0.1	0.013 ± 0.000
	Branches	20.1 ± 0.2	0.006 ± 0.000
12 °C	Control	3.7 ± 0.1	0.021 ± 0.000
	Leaves	44.7 ± 1.1	0.017 ± 0.001
	Branches	32.1 ± 0.4	0.014 ± 0.000
22 °C	Control	7.4 ± 0.1	0.011 ± 0.000
	Leaves	57.8 ± 1.7	0.025 ± 0.001
	Branches	47.6 ± 0.6	0.015 ± 0.001

). The added plant residues decomposed at a greater rate than the SOM, and leaf organic matter decomposed 1.2–1.5 times more than that of small branches. Increasing the temperature from 2 to 12 °C accelerated the decomposition of SOM, leaves, and branches by 1.4, 1.5, and 1.6 times, respectively. Increasing the temperature to 22 °C led to a further increase in the decomposition intensity of these organic substrates by 2.0, 1.3, and 1.5 times, respectively.

The amount of organic matter carbon (C_org) in samples with leaf addition was significantly higher (p ˂ 0.05) than in samples with branch addition under all temperature regimes. The maximum values of this decomposition efficiency were estimated at 22 °C, accounting for 57.8% and 47.6% of the added amount with leaves and branches, respectively (Table 3).

The computed mineralization rate constants verified that the organic matter broke down more quickly at higher incubation temperatures. The decomposition rate constant value was higher in leaf litter samples, reaching maximum values (0.025 ± 0.001 day⁻¹) at 22 °C, and lowest in branch litter samples under all temperature regimes (Table 3). Two important factors must be taken into account: First, compared to SOM, plant residues had lower decomposition rate constants. Second, unlike plant residues, the SOM decomposition constant did not exhibit a strong dependence on temperature.

Increasing temperature and the presence of decomposable material in the soil positively influenced the size of the biologically active SOM pool (

Table 4. The influence of decomposable materials on the size of the biologically active SOM pool.

Temperature	Variant	C0, mg/100 g	k1, Day−1	Corg, % of Initial	MRI, mg C 100 g−1 Day−1
2 °C	Control (C2)	64.8 ± 1.1	0.019 ± 0.000	2.6	1.24
	Leaves (L2)	192.8 ± 0.8	0.015 ± 0.000	6.5	2.89
	Branches (B2)	149.6 ± 1.5	0.010 ± 0.000	5.0	1.56
12 °C	Control (C12)	94.1 ± 1.8	0.021 ± 0.000	3.7	1.94
	Leaves (L12)	286.0 ± 3.0	0.018 ± 0.000	9.7	5.11
	Branches (B12)	244.1 ± 0.2	0.016 ± 0.000	8.2	3.90
22 °C	Control (C22)	186.1 ± 2.7	0.011 ± 0.000	7.4	2.13
	Leaves (L22)	423.5 ± 5.8	0.019 ± 0.000	14.3	7.86
	Branches (B22)	405.3 ± 1.6	0.014 ± 0.001	13.5	5.47

). The C₀ content in the soil without plant residues increased from 64 to 186 mg 100 g⁻¹ within a temperature range of 2 °C to 22 °C. The amount of active soil organic carbon (C₀) was higher in amended variants and increased in tandem with the temperature growth. At all temperature regimes, the content of biologically active organic matter (C₀) and its proportion in C_org were higher in the leaf litter treatment compared with other variants (Table 4). The samples with leaf addition exhibited the greatest C₀ content, reaching 423.54 mg C 100 g soil ⁻¹. Overall, the addition of plant residues contributed to a 2.2–3.0-fold increase in the content of active organic matter in the soil. The content of C₀ in soil without plant residues was 2.6%–7.4% of C_org; in comparison, in the soil with plant residues, it reached 5.0%–14.3% (Table 4).

The combined effect of temperature and the quality of plant additives on the mineralization process is characterized by the mineralization rate index (MRI). The ranges of MRI values varied from 1.24 to 7.86 mg 100 g⁻¹ dry weight. The mineralization rate of organic matter from all tested substrates was highest at 22 °C, with the greatest values observed in the samples amended with leaves (Table 4).

The temperature coefficient (Q10) characterizes the sensitivity of the substrate de-composition process to a 10 °C increase in temperature. SOM was more temperature-sensitive than plant residues, especially in the medium-to-high temperature range (12–22 °C) (

Table 5. Temperature coefficient Q₁₀ of decomposition of soil organic matter and plant residues.

Temperature Range	Control	Leaves	Branches
	Based on decomposition rate efficiency
2–12 °C	1.42	1.49	1.60
12–22 °C	2.00	1.29	1.48
	Based on C-CO2 production during first month of decomposition
2–12 °C	1.87	3.12	10.31
12–22 °C	1.72	1.38	1.89
	Based on C-CO2 production during 10th month of decomposition
2–12 °C	1.48	1.71	1.94
12–22 °C	3.17	2.21	2.20

).

Aspen branches with a C/N ratio of 63 showed higher temperature sensitivity than the leaves of this tree, which possess a C/N ratio of 46. Within the low-to-medium temperature range, the temperature sensitivity of plant residue decomposition was more pronounced than in the 12–22 °C interval.

In the early stages of decomposition, samples containing plant residues showed the strongest temperature sensitivity in the 2–12 °C temperature range. The samples with branch addition (B) had the highest temperature coefficient values, whereas the control samples (C) had the lowest values. Experimental samples showed a considerable decrease in temperature sensitivity when the temperature was raised to 22 °C, with the exception of those with leaf addition, where it slightly, but significantly (p < 0.05), increased from 1.26 to 1.47 (Table 5).

At the late stage of decomposition, an inverse relationship was observed. The temperature sensitivity of leaf and branch decomposition was higher in the range of 12–22 °C. On average, across the three criteria evaluated, the temperature coefficient of soil organic matter decomposition was 1.94, that of aspen leaves was 1.87, and that of aspen branches was 2.90.

3.3. Effect of Plant Residues on the Soil Bacterial Communities over Time at Different Incubation Temperatures

Proteobacteria stood as the dominant phylum in all soil samples, with relative abundance reaching up to 58%, and primarily represented by the classes Alpha- and Betaproteobacteria. Other dominant groups included Acidobacteria (up to 18%), Gemmatimonadetes (up to 14%), and Actinobacteria (up to 10%) (Figure 3).

Shifts in bacterial community structure were observed at different stages of plant residue decomposition (Figure 3). In the control soil, the phylum Acidobacteria was initially dominant; however, after 6 months of incubation at temperatures of 2 °C and 12 °C (6C2 and 6C12 sample), Betaproteobacteria became the dominant group, whereas Alphaproteobacteria were dominant at 22 °C (6C22 sample).

When plant residues were added, Betaproteobacteria were dominant at 2 °C across all incubation times. However, with increasing temperature (12 and 22 °C), Alphaproteobacteria, Acidobacteria, and Gemmatimonadetes became dominant.

At the genus level, the majority of the bacteria from the dominant phylum Proteobacteria were aerobic (Figure 4). Among them, the relative abundance of Massilia, Gemmatimonas, and Gp1 bacterial groups was higher than 10%.

Massilia was evidently the dominant group, averaging 42% of all of the genera present. In some cases, especially at a temperature of 2 °C, this value reached up to 67%. Gemmatimonas was also well represented, making up 30% of the entire community, and unlike Massilia, these bacteria were distributed approximately equally across all experimental conditions. Another dominant group was Gp1, which accounted for up to 29% of the total genetic diversity in the substrate-amended samples.

Spartobacteria were highly abundant in the control soil (9%); however, their relative abundance decreased with the addition of substrates, especially for treatments with input of leaves at 2 °C. The genera Gp16 (11%), Gaiella (7%), Phenylobacterium (6%), Mucilaginibacter (6%), Gp3 (4%), Caulobacter (4%), and Spingomonas (4%) were also characterized by relatively high abundance (Figure 4). The WPS-2 family was temperature-sensitive and dominated (up to 12%) in control samples but only at 22 °C.

3.4. Diversity of Soil Bacterial Communities

To assess functional diversity, the Observed and Shannon indices were calculated for the different sample groups. We compared the bacterial alpha diversity at different incubation temperatures and varying substrate addition. Both Observed and Shannon indices increased in substrate-enriched samples. The highest numbers of Observed species (OTU) (950) and the Shannon index (5.3) were found in soil samples amended with leaves (Figure 5A,C). In the control soil samples, the number of OTUs was 730, and the Shannon index was 4.2.

The highest microbial diversity indices were observed at an incubation temperature of 12 °C (Figure 5B,D).

The beta diversity provides insights into the functional differences between microbial communities. To assess it, a dissimilarity matrix was calculated based on Bray–Curtis distances and visualized using Principal Coordinate Analysis (PCoA) (Figure 6). Axis 1 primarily captured variation in soil temperature, whereas axis 2 captured substrate variation. The interpretation rate of the x-axis (PCo1) accounted for 41% of the variability among the samples; in comparison, the y-axis (PCo2) explained 16.2%, and both cumulatively explained 57.2% of the variance in sample composition (Figure 6).

The results of PCA demonstrate that the bacterial communities were effectively separated at different incubation temperatures, and the difference was significant under the analysis of similarities test (p < 0.05). The clusters grouping values with the same temperature included all substrate and incubation time variants. These clear groupings among samples belonging to the same substrate-enriched experimental group indicate distinct functional profiles across the conditions. The maximum diversity was reached at a temperature of 12 °C in almost all sample types, with a decline resulting from both an increase and a decrease.

The added substrates also influenced the differences in the structure of soil bacterial communities (p < 0.5) but to a lesser extent than temperature.

4. Discussion

4.1. Effect of Temperature on Soil Organic Matter During Decomposition of Forest Litter

Soil microorganisms are the main drivers of biogeochemical cycles and are essential for soil organic matter formation and stabilization [64]. Microbial respiration and its temperature sensitivity are key proxies for the global C cycle. The rate and cumulative values of CO₂ emission characterize both the functional state of the microbial community and its ability to decompose organic substrates, in addition to the availability of substrates for decomposition. Temperature significantly affects the microbial decomposition of organic matter, with higher temperatures generally leading to an increased decomposition rate [65]. In one study, it was demonstrated that annual soil temperature influenced the decomposition rate of leaf litter and fine roots by 95% and 86%, respectively [66].

Temperature and organic matter quality both affect the decomposition of soil organic matter, and temperature sensitivity (Q₁₀) is dependent on C quality [39]. Based on recent findings, the decomposition of labile OM is less sensitive to temperature than that of stable organic matter [66]. In a meta-analysis, researchers found that warming did not significantly increase the decomposition of SOM globally; however, it did reduce decomposition in warmer, drier areas while simultaneously inducing slight increases in colder regions [67].

An incubation experiment was conducted on gray forest soil from a mixed forest in the Moscow region for 364 days (a model of an annual cycle). Microcosms were incubated at three temperature regimes: the average temperature model for the spring and autumn months in southern Moscow Oblast, the average summer temperature model, and the global warming model. To investigate the agents of plant residue transformation into soil, fragments of fallen leaves and thin branches were introduced.

One important measure of the soil’s ability to provide nutrients and support the carbon cycle is the amount of organic matter that is vulnerable to breakdown, or potentially mineralizable organic matter. In samples that included leaf addition, it was found that the amounts of potentially mineralizable organic matter, active organic matter content, and the mineralization index all achieved their maximum levels. All values were significantly higher compared to the samples with branch addition (p < 0.05) and control samples (p < 0.05).

The Q₁₀ temperature coefficient was highest in the samples with the addition of branches. The temperature sensitivity of substrate decomposition rate, as measured by the increase in CO₂ release intensity, increased within the temperature range of 2–12 °C. We concluded that the temperature sensitivity of substrate decomposition rate, as measured by the rise in CO₂ emission, increased within the temperature range of 2–12 °C and decreased with a further rise up to 22 °C. The maximum values of the temperature coefficient Q₁₀ were observed in samples with the addition of branches; in comparison, the minimum values were found in the control samples. The decomposition rate constant was significantly higher in samples with leaves, and the minimum rate was observed in samples with branches. Similar patterns have been observed in other studies [30], wherein the addition of plant residues increased Q₁₀ in the temperature range of 5–15 °C, whereas a decrease in this parameter was found in the temperature range of 15–25 °C.

As the temperature increased, Q₁₀ decreased by 2–2.2 times in the control soil and in soil amended with branches; however, this indicator remained at the same level with the addition of leaves. Concurrently, adding branches increased temperature sensitivity by a factor of two compared to the control, and almost threefold when adding leaves, but only within the temperature range of 12 to 22 degrees. This finding is due to the different composition of SOM and plant residues, in addition to the presence of both microbial biomass and organic matter stabilized by the mineral fraction in the SOM. The highest values of the temperature coefficient Q₁₀ were characteristic of aspen branches, which had a higher C/N ratio than aspen leaves. It should be noted that the temperature coefficient Q₁₀, calculated from soil respiration, primarily characterizes the apparent temperature sensitivity of organic matter, rather than the intrinsic temperature sensitivity [33].

The Q₁₀ value depends on substrate availability and soil properties. Increased substrate availability typically increases Q₁₀ values, with this effect being particularly pronounced in soils characterized by low substrate availability [68,69]. Some researchers note the influence of a substrate on temperature coefficient values. For example, in one study [70], it was suggested that Q₁₀ should increase when transitioning from a higher-quality substrate (straw) to a lower-quality one (needles). However, the authors’ main assumption that straw decomposes more easily than coniferous litter was only confirmed for samples at 15 and 25 °C, but not for 5 °C. They concluded that the wide range of microorganisms, both bacteria and fungi, involved in the decomposition of wheat straw [71] were more sensitive to low temperatures than the fungi primarily responsible for the decomposition of coniferous litter [72]. In another study, it was found that the addition of litter to the soil from coniferous and broadleaf forests reduced Q₁₀ [26].

We concluded that litter addition reduces Q₁₀, indicating a decrease in the response of SOM decomposition to rising temperatures. Moreover, the input of labile organic matter can slow down the degradation of SOM, potentially leading to carbon sequestration [73]. Conversely, the input of easily decomposable substrate can cause the priming effect [74] and an increase in carbon emissions. The slowing down of organic matter decomposition at low soil temperature creates the illusion of carbon sequestration in the soil. However, the potentially mineralizable carbon that has been preserved due to unfavorable temperatures remains unstabilized and can be lost if the temperature rises [36].

4.2. Effect of Temperature on Microbial Communities During Forest Litter Decomposition

Limitations in growth resources (e.g., carbon substrates) may change bacterial utilization strategy for a particular substrate or resource and microbial interactions. Thus, the ability to adapt and recover from environmental changes is a critical process in ecosystem functioning [75]. We observed a shift in community composition, demonstrating that bacterial community abundance was impacted by warming conditions and C dynamics. Based on our findings, temperature was the most important factor influencing the structure and diversity of the bacterial community, regardless of the presence and composition of plant residues.

Bacterial diversity was highly dependent on the temperature regime, with different taxa exhibiting varying sensitivities and adaptation capabilities. In the control soil, the phylum Acidobacteria was dominant; however, after 6 months of incubation at 2 °C and 12 °C, the class Betaproteobacteria became dominant, whereas at 22 °C, the class Alphaproteobacteria emerged as dominant. In another study [76], Actinobacteria and Firmicutes were the most sensitive to temperature changes, whereas Proteobacteria exhibited resistance to temperature variation.

Analysis of community α-diversity revealed that leaves are the most preferred substrate for increasing species richness: Shannon and OTU index values reached their maximum specifically in samples with leaf addition. These findings are similar to the results of a previous study, in which the authors concluded that temperature significantly affects the alpha-diversity of the soil bacterial community [77]. However, the results of a recent meta-analysis showed that rising temperatures have a minor impact on the diversity and taxonomic structure of soil microbial communities in field warming experiments [78]. Based on β-diversity data, temperature was the most important factor influencing the structure of the microbial community, with the optimum being 12 °C.

The most abundant bacteria in the soil of our study were Massilia, with the maximum number observed at a temperature of 2 °C. These bacteria are recognized for their ability to colonize the surface and root systems of plants, and some species can dissolve phosphorus compounds in the soil, accelerating plant growth, while preferring low temperatures [79]. The abundance of Spartobacteria, Niastella, Mucilaginibacter, and WPS-2 also depended on the temperature regime. Gp1, Gp3, Gp6, and Gp16 genera, which are common in forest ecosystems and are typical of gray forest soil [80], were also abundant. Among the dominant bacteria were those of the genus Mucilaginibacter, which are known as denitrifiers and are sensitive to substrate addition, with them being more frequently found in the experimental variants with the leaves and branches. Moreover, Mucilaginibacter preferred lower temperatures, and their numbers decreased at 22 °C. A similar pattern is observed for bacteria of the genus Caulobacter, which are also denitrifiers and prefer rich in substrate soils [80].

The substrate was another important factor influencing the composition of the microbial community. The results of some studies indicate that the diversity of soil microorganisms depends on the composition and amount of litter [81,82]. However, some authors are of the opinion that the influence of the substrate should only be assessed by taking into account the temperature regime [83]. Based on our findings, we concluded that leaves were the most nutritious substrate: samples with their addition showed increased diversity compared to samples with the addition of branches and the control. For example, bacteria of the genus Niastella, which were found in large numbers in the samples from the experiment with the addition of substrates (leaves and branches), were absent in the control samples. This phenomenon may be explained by the findings that such bacteria prefer readily available substrates and are most often isolated from the soil and rhizosphere of agricultural ecosystems [82,84]. A similar pattern was observed for the genus Devosia.

5. Conclusions

The crucial role of microorganisms in soil C dynamics is widely recognized; however, their responses to climate warming remain difficult to predict. In this study, we analyzed the effect of elevated temperatures on the active organic matter content and bacterial community diversity in forest soil microcosms amended with plant litter. We found that rising temperature increased the content of active organic C, and the bacterial community structure was highly temperature-dependent, further highlighting the role of microorganisms in soil C mineralization during warming. Slow plant litter decomposition at low temperatures creates apparent carbon sequestration in the soil. Overall, this study provides valuable insights for a deeper understanding of the interactions between soil microorganisms and active organic matter in forest soils under climate warming conditions. However, this inference is limited by the laboratory experiments not representing real environmental conditions, the characteristics of labile SOM based on microbial respiration, and 16S fingerprinting of soil microbial communities. Future work combining laboratory and field experiments and multi-omics analyses will be essential to confirm the functional roles of potential keystone bacterial taxa in carbon cycling processes.