The study was conducted at the experimental base of Heilongjiang Bayi Agricultural University. Agaricus bisporus substrate was collected from Daqing Hengrui Edible Mushroom Company, and cattle manure was gathered from Anda’s farm. The physicochemical properties of the raw materials (Detailed information is available in Supplementary Table S1). Besides, a strip-stack composting method was employed, with a trapezoidal pile, 2.5 m × 1.5 m at the bottom, and 1.9 m × 0.9 m at the top, with a height of 1.5 m. Through pre-experimentation, a mixture of Agaricus bisporus mushroom substrate and cattle manure in a 7:3 (w:w) ratio was determined for the CD treatment; the CK treatment consisted solely of Agaricus bisporus mushroom substrate that had been composted. Furthermore, the initial moisture content (MC) was approximately 60%, and pile and ambient temperatures were measured daily during the composting process. By the temperature change on the 10th day, 20th day, the 29th day artificially turn the pile 3 times on the condition that the pile temperature stabilization tends to resemble the surrounding temperature that the compost decomposition is complete. Based on temperature change, the collected samples from the 1st, 3rd, 17th, and 36th days were picked to illustrate the mesophilic, thermophilic, cooling, and maturation phases. A random sampling technique was employed to select five points at 20, 70, and 120 cm from the top of the pile; the resulting samples had an approximate mass of 0.5 kg each. Following mixing, the samples were gathered in two parts, one was stored at −80°C for high-throughput sequencing, the other was air-dried and sieved for the analysis of physicochemical indexes, and the four periods of CK and CD were termed CK1, CK3, CK17, CK36, CD1, CD3, CD17, and CD36.

An automatic temperature recorder was used to measure compost temperature and ambient temperature during the composting process, and the temperatures at the top, middle, and bottom of the pile were recorded at 3:00 p.m. each day. Water content (WC) was determined by drying the samples to constant weight in a 105°C oven; Total organic carbon (TOC) was ascertained by the externally heated potassium dichromate-volumetric method and total nitrogen (TN) by the Kjeldahl method. NH₄⁺-N and NO₃⁻-N were extracted with 2 M KCl and determined by indophenol blue colorimetry and dual wavelength UV spectrophotometry (Varma et al., 2018); Using a conductivity meter and pH meter, the pH and EC were defined by a liquid-to-solid ratio of 10:1.

Based on molecular weight and functional carbon content, HS consists of two main components, humic acid (HA) and fulvic acid (FA). Typically, the TOC of humic components can characterize their concentration. Besides, the extraction and determination of humus was by the method of Zhou et al. (2014). Humus content was ascertained by mixing air-dried samples with an extract of 0.1 M Na₄P₂O₇ mixed with NaOH at a solid–liquid ratio of 1:20 (w/v), shaking at 200 rpm for 24 h, and centrifuging at 12,000 rpm for 15 min at ambient temperature. FA was acquired by filtration through a 0.45 μm filter, and the precipitate was washed several times with 0.05 M HCl to acquire HA by dissolution with 0.05 M NaHCO₃. The methodology utilized to determine total humic acid (HA) and fulvic acid (FA) was the externally heated potassium dichromate-volumetric method. Moreover, the total carbon content of the compost samples was determined and subtracted from the humic and fulvic acid content to obtain the humin content. The humification index was calculated based on the following formula (Wu et al., 2017b):

(Note: C_HA, C_FA, and C_HS, indicate the TOC content of HA, FA, and HS; HA/FA represents Degree of Polymerization, HI represents Humification Index, and HPA denotes the Percentage of HA.)

The identification of humus precursors predominantly entails the analysis of sugar constituents, including polysaccharides, total sugars, reducing sugars, and polyphenols. Besides, total sugars were ascertained by the phenol-sulfuric acid method, reducing sugars were determined by the 3,5-dinitro salicylic acid (DNS) method, with polysaccharide content equal to total sugars minus reducing sugars, and polyphenols were ascertained by the FoLin–CiocaLteu (FC) (Singleton and Rossi, 1965).To each sample, add 800 μL of Na₂CO₃ (7.5% w,v), mix well, and allow it to rest for 2 min. Subsequently, add 1 mL of FC reagent. It was placed at 20°C for 2 h and the absorbance was measured spectrophotometrically at 765 nm.

Samples from 1, 3, 17, and 36d were selected for high-throughput sequencing. DNA was extracted using a soil DNA kit (Phygene, Fujian, China) following the instructions, and before high-throughput sequencing on the Illumina MiSeq300 platform manufactured by Beijing Allwegene Technology, the extracted DNA underwent purification and amplification. Moreover, the fungal ITS sequence was determined by amplifying the ITS1 region with primers ITS1-F (CTTGGTCATTTAGAGGAAGTAA) and ITS2 (TGCGTTCTTCATCGATGC). The bacterial 16S sequences were determined by amplifying the V3-V4 region with primers 338F and 806R, and the raw data was retained after downloading in Fastq format. To preserve and share high-throughput sequencing data, the NCBI Data Center established the Sequence Read Archive (SRA) database.¹ The raw data were processed through sequence splicing to remove low-quality sequences, followed by OTU clustering and species identification. In addition to species annotation results and alpha and beta diversity derived from OTU clustering outcomes, the analysis also incorporated variations in microbial community structure.

GraphPad Prism 9 was employed to analyze the physicochemical parameters, concentrations of humus precursors, and humus components. The IBM SPSS Statistics 22 software was utilized to perform every statistical analysis. The OTU table’s basic statistics, including the results of data normalization and the combination of species from distinct phyla, are calculated using the “dplyr” package of R.Alpha diversity was calculated utilizing the ‘vegan’ package and analyzed and plotted with the ‘EasyStat’ package. The beta diversity was analyzed and plotted using the R package “phyloseq” and “ggplot,” respectively. Computation of dissimilarity based on Bray-Curtis distances between samples to assess distinctions in microbial community structure. The default selection for the species composition analysis consisted of the 10 most abundant OTUs. The data was cleaned utilizing ‘dplyr’ and visualized using ‘ggplot’. Any remaining species were integrated and labeled as ‘other’, and the resulting graphs were displayed as stacked histograms. Procrustes analysis was employed to determine whether there was a substantial whole correlation between microbial community structure and environmental factors. Furthermore, network analysis was performed using “ggClusterNet” with Gephi for calculations and mapping, with default relationship thresholds of r > 0.8 and p < 0.05. To discover and explore network relationships, the 200 microbes with the highest abundance were chosen. The colors filled in on the network graphs represent taxonomic information at the phylum level. Gephi is utilized to visualize and compute topological attributes, including the number of nodes, edges, average degree, average path length, network diameter, as well as network density, to estimate the complexity of the network. Moreover, the stability of the network is evaluated through robustness analysis. Core microbes were identified by adopting the “plot(g)” package of R. The top 20 genera with the highest abundance in the network were selected and defined as core microbes. STAMP was employed to analyze the distinction between groups, and nonparametric tests were integrated with horizontal histogram visualization. Furthermore, p-value correction was performed by default utilizing “fdr” with a threshold of 0.05.RDA was performed utilizing Canoco (Version 5.0) to investigate the correlation between fundamental microbes and environmental variables. Structural equation modeling using AMOS (IBM; SPSS AMOS 20.0.0) was employed to assess the direct and indirect influences on the humus of core microbes and precursors.

The changes in pile temperature and ambient temperature during the composting process are shown (Figure 1). The pile temperature of all treatments increased rapidly in the first 3 d. The pile temperature of CD treatment reached 54°C in the 3rd d and rose to the highest temperature (56°C) in the 5th d. The whole high-temperature stage (>50°C) lasted for a total of 20 d. The temperature of the heap reached a peak in the 3rd d (47°C) after the CK treatment and was maintained for 4 days but the temperature never exceeded 50°C. The temperature of the compost pile after the CK treatment reached a peak in the 3rd d (47°C) and remained at the highest temperature for 4 days but the temperature never exceeded 50°C. The growth and development of microorganisms may have been hampered by the lack of effective carbon and nitrogen sources, which contributed to the low pile temperature. The heat generated by microbial decomposition is the main factor contributing to the temperature change of the heap, and the addition of cow dung to Agaricus bisporus pomace compost can increase the temperature and microbial activity of the heap, thus accelerating the composting process.

To determine the effect of cattle manure addition on the humification process in the compost of Agaricus bisporus mushroom substrate, the fluctuations of humification indices and humic acid component concentrations (HA/FA, HI, and PHA) in the compost were analyzed. Besides, the contents of HA and HS indicated an increasing trend during composting (Figure 2A). Upon completion of the composting procedure, the contents of HA and HS in the composted material CD were higher (49.52 g·kg⁻¹, 245.16 g·kg⁻¹) than CK (28.33 g·kg⁻¹,169.04 g·kg⁻¹). On the contrary, the concentration of FA indicated a decreasing trend, with a 59.25% reduction in CD and a 15.82% reduction in CK. The humification process pertains to the metabolic activity of microbes on organic compounds. Throughout this cyclical process, interconversion between HA and FA is possible (Zhao et al., 2017a). Fulvic acid has a distinct structure characterized by low molecular weight and a high concentration of acidic functional groups. It is a comparatively mobile and active component of HS in compost, as opposed to HA (Tan, 2014). Due to its instability, fulvic acid is frequently transformed into more advanced aromatized humic substances, including HA (Zhou et al., 2014).In addition, microbes can utilize fulvic acid, a readily accessible organic compound, as an energy source (Zhang et al., 2018), thus resulting in an increase in HA concentration and a decrease in FA content.

The humification index, which accurately measures the degree of humification and maturity of the final compost, is calculated as the ratio of the HS, HA, and TOC indicators (Wei et al., 2014). HA/FA is widely used to describe the relative chelation rates of humic acid and fulvic acid and is frequently employed as a measure of the degree of compost humification (Zhao et al., 2017b). HI and PHA increases are indicative of an increase in the structural complexity of HS (Wu et al., 2017b). Under the experimental conditions, the composting process increased the levels of HA/FA, HI, and PHA in both treatments (Figure 2B). Throughout the composting process, the levels of HA/FA, HI, as well as PHA were substantially higher in the CD in comparison to the CK. These results indicate that the addition of cattle manure can remarkably enhance the humic acid conversion efficiency during the composting process and substantially promote the more mature humic acid components HA.

Studies have demonstrated that the decomposition of organic matter during composting produces intermediate products including amino acids, polyphenols, polysaccharides, and reducing sugars that are strikingly similar to humus formation (Wu et al., 2020). The polyphenol content of the two treatments illustrated a decreasing trend as demonstrated in Figure 3A. The CD treatment exhibited a higher polyphenol content than the CK treatment at all time points. From 1d to 36d, the CD treatment witnessed a 531 mg/kg decrease in polyphenol content, whereas the CK treatment witnessed a 312 mg/kg decrease. Furthermore, the rate of decrease in the CD treatment was significantly greater than that of the CK. Under the action of polyphenol oxidase, polyphenols form benzoquinones, which subsequently undergo condensation reactions with intermediates including amino acids, polysaccharides, and reducing sugars (Duan et al., 2019b), which promotes the synthesis of humic substances, leading to a continuous decrease in polyphenol content. Substantial distinctions were observed between reducing sugars and polysaccharides during the initial 3 days of composting. The concentration of polysaccharides decreased slightly, and yet the concentration of reducing sugars increased during this period. This phenomenon could potentially be attributed to the degradation of polysaccharides to reducing sugars by microorganisms in response to the elevated compost temperature. There was an overall decreasing trend in polysaccharide concentration (Figure 3B), with a faster decrease in polysaccharide content in the CD treatment in comparison to CK, which was ascribed to the metabolic activities of microbes being stimulated by the addition of cattle manure. The two treatments illustrated opposite trends in total sugars, with the CK treatment initially decreasing and subsequently slowly increasing after the 17th day, while CD indicated an increasing trend in total sugars (Figure 3C). Reducing sugar functions as a carbon source for microorganisms and is utilizable directly by them. The concentration of reduced sugar indicated an increasing trend followed by a decreasing trend (Figure 3D). Both treatments reached their highest reducing sugar content on the 3rd day, with CD at 50.23% and CK at 45.31%. Nonetheless, the reduced sugar content of CD was constantly higher than that of CK. In the initial period, as the temperature increased, a gradual inactivation of certain mesophilic microbes accompanied a reduction in carbohydrate consumption, resulting in a slight increase in the concentration of reducing sugars on the 3rd day, which were subsequently directly utilized by the microbes, resulting in a rapid decrease. Consequently, we inferred that the reduction of polysaccharides and reducing sugars is related to microbial utilization and HS formation. Microbial metabolism is stimulated, the degradation of organic matter (including proteins and cellulose), and the synthesis of humus precursors are accelerated; the polymerization of additional carbohydrate components HS is encouraged.

Physicochemical parameters exert an influence on the production of humic precursors, subsequently impacting microbial activity (Wu et al., 2017b).RDA was used to analyze the connection between physicochemical indicators, humus formation, bacterial (Figure 8A), and fungal (Figure 8B) activity during the composting process. Additionally, CCA was performed for non-restrictive sorting and the contribution of each environmental factor to the microbial community was calculated using anova, and the cca for significance testing (Figures 8C,D). Setting a threshold of p < 0.05 screened 10 indicators for correlation with microbes, with TS, N, T, pH, and HA/FA having the greatest effect on bacteria and fungi (p < 0.001). Overall, the indicators contributed more to the bacteria, and yet the difference was not remarkable. Polysaccharide is the primary precursor that affects the degree of humification, which is strikingly similar to HA/FA. The three most critical environmental factors affecting HA/FA are NO₃⁻, pH, and N. Among these factors, NO₃⁻ and pH are positively correlated with HA/FA, while N is negatively correlated with HA/FA. Additionally, T, EC, and NH₄⁺ are negatively correlated with HA/FA.

Structural equation modeling (SEM) was employed to validate the response of environmental factors, core microbes, as well as precursors to the humification index HA/FA and to evaluate the effect of cattle manure addition on the humification process (Figure 9). The correlation strength between HA/FA in RDA and humus formation led to the identification of three environmental parameters that influence humus formation: NO₃⁻, pH, and NH₄⁺. Additionally, polyphenols and polysaccharides were identified as essential precursors affecting humus formation, while core bacteria and core fungi were identified as drivers. Besides, SEM illustrated that the inclusion of cattle manure modified the correlation between physicochemical parameters and core microbes during the composting process. Particularly, the correlation between pH, EC, and the correlation between core bacteria and CD was reversed from positive to negative in CD relative to CK. Meanwhile, the correlation between NO₃⁻ and core bacteria was strengthened yet weakened with core fungus. Additionally, the positive and negative correlations shifted. In SEM, a positive correlation signifies that two variables contribute to one another’s improvement; conversely, a negative correlation suggests that the variables are utilized or formed in opposition to one another. After the addition of cattle manure, the function of the core bacteria shifted, and the correlation with polyphenols and polysaccharides changed from a positive correlation to a negative correlation, implying that the function shifted from facilitating the production of precursors to transforming them directly into precursors. This is because polyphenols and polysaccharides are used by microbes as the predominantly energy and carbon source to facilitate the conversion of FA to HA (Wu et al., 2017b). Furthermore, the influence of CD core microbes on humus precursors (p < 0.001), which in turn relates to the extent of compost humification through polyphenols and polysaccharides, was increased in comparison to CK. Concurrently, the addition of cattle manure enhanced the direct impact of core bacteria on HA/FA. Confirming the previous assumption that bacteria are more directly involved in the formation of humus components, fungi continue to form HS through the transformation of precursors. The SEM analysis revealed a highly substantial negative correlation between polysaccharides and HA/FA in both treatments, which suggests that the primary precursors for the polymerization-mediated formation of HA are polysaccharides. Additionally, the results suggest that core bacteria play a more remarkable role than fungi in precursor production and humus formation. The SEM analysis indicated a highly considerable negative correlation between polysaccharides and HA/FA in both treatments, demonstrating that polysaccharides are the main precursors for the formation of HA through polymerization. Additionally, they concluded that core bacteria exert a more substantial influence on precursor production and humus formation than fungi.