Enhancement of Biomethane Yield from Spent Mushroom Substrate: Biological Pretreatment with the Chlamydospores of Trichoderma viride

Abstract

Fungal chlamydospores are asexual spores formed by fungi under adverse conditions and could be used in biological pretreatment for biogas projects fed by lignocellulosic substrates. In this study, Trichoderma viride (Tv) chlamydospores were used as the pretreatment agent to enhance the methane yield of spent mushroom substrates (SMSs). Lignocellulosic composition, methanogenesis performance, and anaerobic microbial communities were investigated for different Tv pretreatment durations (0 h, 12 h, 24 h, 48 h, 96 h, and 192 h). The results showed that the optimal Tv pretreatment duration was 24 h, and the cumulative methane yield reached 173.4 mL/gVS, which was 16.8% higher than that of the control. A pretreatment duration longer than 48 h was not conducive to methanogenesis. Sequencing analysis of anaerobic microbial communities showed that the pretreatment duration was directly proportional to the relative abundance of Tv at the beginning of digestion. When the initial Tv abundance was higher than 50%, Trichoderma became the absolute dominant fungus with an abundance higher than 97% in fungal communities in the later stage of digestion. The correlation network among fungi, bacteria, and archaea showed that Tv was directly related to 11 genera, and through these taxa, Tv affected 58% of the taxa in the whole microbial network. Cost accounting showed that Tv pretreatment has a net income of 45.5 CNY/1000 kg SMS, and is a promising technology. This study provides important guidance for the use of fungal chlamydospores in pretreatment and also promotes the understanding of fungi in anaerobic digestion.

1. Introduction

Spent mushroom substrate (SMS) is a by-product of mushroom cultivation, commonly containing mushroom mycelium and semi-degraded lignocellulose. With the increase in population and changes in dietary structure, the production and consumption of mushrooms are increasing rapidly, leading to increasing amounts of SMS. The annual output of SMS has exceeded 12.5 million tons worldwide [1]. However, the large amounts of SMS are not effectively reused as a resource, but instead pose a serious threat to the environment [2]. At present, the traditional disposal methods of SMS mainly include field dispersal, open burning, landfill, or co-composting with animal manure [3]. To improve the resource utilization efficiency of SMS, researchers have developed it into land fertilizer, animal feed, and bioenergy [4].

Anaerobic digestion is a biological method to convert organic substrates into green energy and bio-fertilizer, which can solve both the challenges of waste disposal and greenhouse gas emission reduction. However, when degrading lignocellulosic material such as SMS, the hydrolysis of tightly compacted lignocellulose is a limiting step of speed and efficiency [5]. To overcome this bottleneck, previous research has tested various pretreatment processes, including physical (mechanical, steam explosion, etc.), chemical (alkaline, acid, ozone, etc.), biological (fungal, enzymatic, microbial consortium, etc.), and the combined methods [6,7,8]. Among them, fungal pretreatment is a promising process with the advantages of moderate conditions, low chemical input, and environmental friendliness, although it may involve longer processing times [9].

Fungal pretreatment proceeds with different fungi that can break down lignocellulose by secreting various enzymes. It was considered that white rot fungi are more effective than other fungi [10,11]. Trichoderma species are one kind of white rot fungi, and the pretreatment of Trichoderma sp. increased the methane yields of rice and corn straw, wood pellets, garden waste, and cattle dung by 13–300% [12,13,14,15,16] However, for industrial biogas projects, the culture conditions and the huge injection amount of Trichoderma sp. are difficult to maintain. Compared with currently used mycelia and conidia, the chlamydospores formed under extreme conditions are more suitable for the commercial-scale pretreatment because of their long shelf life and simple germination requirements [17]. In China, the chlamydospores of Trichoderma sp. are already industrialized products as liquid fertilizers for plant growth promotion. More investigation is required to evaluate whether pretreatment using the chlamydospores of Trichoderma sp. could enhance the efficiency of methane production.

The knowledge of microbial structure and ecology in anaerobic digestion has been rapidly promoted with the development of microbiome technologies. Compared with bacteria and archaea, fungi in anaerobic digestion have received little attention. Moreover, the effect of fungi from the pretreatment process on anaerobic microbial communities is still unknown. In the rumen, some anaerobic fungi could stimulate the growth of methanogenic archaea and maintain methanogenic diversity [18,19]. In soil and plant rhizosphere ecosystems, fungi play a leading role in adjusting the evolutionary direction of microbial community structure [20,21]. Thus, the research on in situ and externally sourced fungi, as well as the relationship between these fungi and other microorganisms in anaerobic systems, will benefit the theory development and efficiency promotion of anaerobic digestion.

This study aims to explore the possibility of biological pretreatment using fungal chlamydospores to improve biomethane yield from SMS. The growth of fungi, enzyme activity, and lignocellulose content during the pretreatment period were investigated. The dynamics of acids and biogas production through batch anaerobic digestion were evaluated. The responses of anaerobic microbial communities (fungi, bacteria, and archaea) to pretreatment were analyzed. The results of this study will offer theoretical support for energy recovery from lignocellulose biomass and a deeper understanding of anaerobic fungi.

2. Materials and Methods

2.1. Fungal Chlamydospores

The fungal strain was isolated from the rhizosphere of a vegetable field using Rose Bengal–Agar medium [22]. After morphological observation and internal transcribed spacer (ITS) sequencing, it was identified as Trichoderma viride (Tv), and deposited in the China General Microbiological Culture Collection Center (CGMCC No. 16800). The chlamydospores of Tv were in the form of a spore suspension, which was provided by Jinxing Agricultural Technology Development Co., Ltd. (Rizhao, China). The spore suspension was fermented by a 7-day pilot-scale liquid fermentation with molasses as the carbon source and peptone as the nitrogen source. The formation rate and density of chlamydospores were 90% and 2 × 10⁸ cfu/mL, respectively. The spore suspension was stored at 4 °C before use.

2.2. SMS and Inoculum

The SMS was provided by a mushroom farmer cooperative in Jining, China. It consisted of Hypsizygus marmoreus SMS, Auricularia polytricha SMS, and Auricularia auricular SMS in a general proportion of 3:2:1 (v/v). The SMS was ground into 24 meshes, then sufficiently mixed before use. The inoculum was provided by Shandong Sifon Environmental and Bio-energy Co., Ltd. (Jinan, China). It was the effluent of a well-run mesophilic digester with kitchen wet waste as the influent. The inoculum was stored in covered barrels at 14–17 °C and was domesticated at 37 °C for one week before use. The characteristics of the SMS and inoculum are listed in

Table 1. The characteristics of the SMS and inoculum.

Character (Unit)	SMS	Inoculum
TS (%)	93.87 ± 0.00	1.06 ± 0.00
VS (%TS)	90.14 ± 0.10	49.63 ± 0.10
C (%TS)	34.79 ± 0.11	18.38 ± 0.09
N (%TS)	2.04 ± 0.04	2.81 ± 0.08
C/N	17.05	6.55
Hemicellulose (%TS)	19.83 ± 0.25	n.d.
Cellulose (%TS)	29.30 ± 0.24	n.d.
Lignin (%TS)	19.81 ± 0.28	n.d.
Ash (%TS)	9.86 ± 0.10	50.37 ± 0.10
pH	n.d.	8.02 ± 0.03
Formic acid (mg/mL)	n.d.	0.000 ± 0.000
Acetic acid (mg/mL)	n.d.	0.112 ± 0.002
Propionic acid (mg/mL)	n.d.	0.058 ± 0.000
Butyric acid (mg/mL)	n.d.	0.013 ± 0.000

Average ± standard deviation (n = 3); n.d. means not determined.

.

2.3. Experimental Design

The pretreatment was carried out in a 26 °C incubator. The different pretreatment periods were set to the following six treatments, namely Tv0, Tv12, Tv24, Tv48, Tv96, and Tv192, where the numbers represent the pretreatment durations in hours. The spore suspension of Tv chlamydospores was diluted 1000 times with sterile water. Using a 500 mL reagent bottle as digester, diluted spore suspension of 25 mL was mixed into 15 g of SMS to cause the inoculation amount and moisture content to reach 3 × 10⁵ cfu/g and 60%. For each treatment, six replicates were set, three of which received 150 mL of anaerobic inoculum immediately after pretreatment, and the other three were used for the analysis of lignocellulosic composition and enzyme activity.

The working volume of the anaerobic digesters was 200 mL with sterile water. The digesters were filled with nitrogen and sealed with silicone stoppers to ensure an anaerobic environment. The digestion period was 40 days, and the digestion temperature was 37 ± 1 °C. The biogas volume and methane content were measured every day. Approximately 4 mL of samples were taken on the 0th, 1st, 2nd, 4th, 6th, 8th, 10th, 13th, 16th, 20th, 25th, and 30th days from each digester. The samples were centrifuged (5000 rpm/min, 5 min) to divide the supernatant and pellet. The supernatant was used to test pH, organic acids, and total ammonia content (TAN). Solid samples from day 1 and day 20 were used for high-throughput sequencing of fungi, bacteria, and archaea. After 40 days of digestion, when the daily methane production during the last three days was below 1% of the accumulated volume of methane [23], the residues were dried and collected for the analysis of the reduction rates of lignocellulosic components.

2.4. Chemical-Physical Analysis

Total solids (TSs), volatile solids (VSs), total carbon (TC), and total nitrogen (TN) of SMS and inoculum were measured according to standard methods [24]. For TS and VS measurements, samples were dried at 105 °C until a constant weight was achieved, followed by ignition at 550 °C for 4 h to determine the volatile solids. Hemicellulose, cellulose, lignin, and ash were measured using a fiber analyzer (Model 2000, AMKOM, Macedon, NY, USA). The pH value was measured using a pH meter (PB-10, Sartorius, Gottingen, Germany). The concentrations of formic, acetic, lactic and propionic acids were analyzed using a high-performance liquid chromatography (HPLC) system equipped with an ion-exchange column (Aminex HPX-87H; 300 mm × 7.8 mm, BioRad Laboratories, Hercules, CA, USA) and a detector (SPD-M20A, Shimadzu, Kyoto, Japan). For analysis, 5 mmol/L H₂SO₄ was applied as the eluent at a flow rate of 0.6 mL/min and 35 °C. Data analysis was performed using commercial HPLC Software (Chromatography Station for Windows 3.2, Shimadzu, Kyoto, Japan). Calibration standards for HPLC analysis were prepared with known concentrations of formic, acetic, propionic, and butyric acids, and all samples were filtered through 0.22 µm filters prior to injection. The specific operational details were based on [25]. The microphenotypes were observed and captured with an optical microscope (BX53, Olympus, Tokyo, Japan). The enzyme activities of carboxymethyl cellulase (CMCase) and xylanase were measured according to Wen et al. [26]. The reduction rate was expressed as the ratio of the reducing part to the total amount using Equation (1).

$$R\left(\%\right)={\displaystyle \frac{\mathrm{Total} \mathrm{weight}\left(\mathrm{g}\right)-\mathrm{Residue} \mathrm{weight}\left(\mathrm{g}\right)}{\mathrm{Total} \mathrm{weight}\left(\mathrm{g}\right)}}\times 100\%$$

(1)

The daily biogas production was calculated from daily pressure differences in the headspace of digester using Equation (2), where V_biogas is the daily biogas volume (mL), ΔP is the absolute pressure difference (kPa), V_headspace is the volume of the headspace (mL), C is the molar volume (22.41 L/mol at 273.15 K, 101.325 kPa), T is the absolute temperature (K), and R is the universal gas constant 8.314 J/(mol·K).

$$V_{\mathit{biogas}}={\displaystyle \frac{∆P\times V_{\mathit{headspace}}\times C}{R\times T}}$$

(2)

The pressure was measured using a pressure tester (82100, AZ-instrument, Taichung, China) with an accuracy of 0.3%. The biogas in the headspace was released under water to prevent gas exchange. The methane content of biogas was measured using a gas chromatograph (GC-2014, Shimadzu, Japan). The specific operation details were based on Zhao et al. [27]. The daily methane yield was calculated per gram of added feedstock VSs using Equation (3)

$$V_{\mathit{methane}}\left(\mathrm{mL}/\mathrm{gVS}\right)={\displaystyle \frac{V_{\mathit{biogas}}\left(\mathrm{mL}\right)\times \mathrm{Methane} \mathrm{content}(\%)}{{\mathit{VS}}_{\mathit{substrates} \mathit{added}}\left(\mathrm{g}\right)}}$$

(3)

TAN was measured using a flow injection analyzer (AA3, SEAL, Coventry, Germany). The concentration of free ammonia (FAN) was calculated according to Equation (4), where T(K) is the temperature (K) [28].

$$\left[\mathrm{F}\mathrm{A}\mathrm{N}\right]\left({\displaystyle \frac{\mathrm{m}\mathrm{g}}{\mathrm{L}}}\right)=\left[\mathrm{T}\mathrm{A}\mathrm{N}\right]\left({\displaystyle \frac{\mathrm{m}\mathrm{g}}{\mathrm{L}}}\right)\times {\left(1+{\displaystyle \frac{{10}^{-\mathrm{p}\mathrm{H}}}{{10}^{-\left(0.09018+\frac{2729.92}{\mathrm{T}\left(\mathrm{k}\right)}\right)}}}\right)}^{-1}$$

(4)

2.5. Microbial Analysis

High-throughput sequencing of fungi, bacteria, and archaea was performed on day 1 and day 20 to investigate the immediate and delayed responses of microbial community to Tv pretreatment. Microbial DNA was extracted by a DNA kit (Omega, Norcross, GA, USA). The quantity and quality of extracted DNAs were measured by spectrophotometric analysis and agarose-gel electrophoresis. For PCR amplification, the primers of ITS1 (5′-CTTGGTCATTTAGAGGAAGTAA-3′) and ITS2 (5′-GCTGCGTTCTTCATCGATGC-3′) were used for fungi; 338F (5′-ACTCCTACGGGAGGCAGCA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′) were for bacteria; and 524F10extF (5′- TGYCAGCCGCCGCGGTAA-3′) and Arch958RmodR (5′-YCCGGCGTTGAVTCCAATT-3′) were used for archaea. The PCR mixture contained 4 μL of 5 × FastPfu Buffer, 2 μL of 2.5 mM dNTPs, 0.8 μL of each primer (5 μM), 0.4 μL of FastPfu Polymerase, and 10 ng of template DNA. And the PCR program was 95 °C for 3 min, 95 °C for 30 s for 30 cycles, 55 °C for 30 s, 72 °C for 45 s, and 72 °C for 10 min. Amplicons were read on an Illumina MiSeq platform (Illumina, San Diego, CA, USA) by MajorBio Co., Ltd. (Shanghai, China). The operational taxonomic units with more than 97% similarity were integrated. Fungal taxonomy was assigned based on Unite (Release 8.0 http://unite.ut.ee/index.php (accessed on 10 February 2025)), and taxonomy of bacteria and archaea was assigned based on Silva (Release 138 http://www.arb-silva.de (accessed on 10 February 2025)). The microbial population and genus levels were analyzed for microbial structure by the Majorbio Cloud Platform (https://cloud.majorbio.com (accessed on 10 February 2025)).

2.6. Statistical Analysis

2.6.1. Kinetic Model

The first-order kinetic model (Equation (5)) and the modified Gompertz model (Equation (6)) were applied to model batch methane production [29]. The first-order kinetic model provides the hydrolysis rate, while the modified Gompertz model offers the maximum gas production rate. Utilizing both models allows for the acquisition of different kinetic parameters and predictions of maximum gas production from distinct perspectives—hydrolysis and gas production.

$$-Kt=ln\left[1-{\displaystyle \frac{M_{\left(t\right)}}{M_{\mathit{max}}}}\right]$$

(5)

$$M_{\left(t\right)}=M_{\left(\mathit{max}\right) }\exp \left\{-\exp \left[{\displaystyle \frac{R_{\mathit{max}}e}{M_{\mathit{max}}}}\left(\lambda -t\right)+1\right]\right\}$$

(6)

where M_(t) is the cumulative methane yield at day t (mL/gVS); M_(max) is the potential maximum methane yield (mL/gVS); K is the hydrolysis rate constant (d⁻¹) and t is the fermentation time (d); R_max is the maximum methane production rate (mL/gVS·d); and λ is the lag phase time (d). The model fitting was performed using Origin 9.3 (OriginLab, Northampton, MA, USA).

2.6.2. Correlation Network Analysis

The relationships among fungi, bacteria, and archaea at the genus level were visualized through a correlation network [30]. The microbial data of six treatments and two sampling times were integrated. Pearson correlation analysis was conducted at the genus level among fungi, bacteria, and archaea. When the p-value is less than 0.05 between genera, the relationship appears in the network. The nodes represent the genera of the microbes. The edges represent significant correlations based on Pearson correlation analysis. Nodes without any edges were omitted. Visualization and modularization were performed using Gephi 0.9.2 (WebAtlas, Paris, France). The simplified network of genera directly connected to Trichoderma was based on the Gephi network and was drawn using PowerPoint 2016 (Microsoft, Redmond, Washington, DC, USA).

2.6.3. LEfSe Analysis

Significant differences in microbial genera between high-dose and low-dose Trichoderma were identified by linear discriminant analysis (LDA) effect size (LEfSe) analysis (http://huttenhower.sph.harvard.edu/lefse/ (accessed on 10 February 2025)). The non-parametric Kruskal–Wallis sum-rank test was used to determine significant differences (p < 0.05) in microbial abundance. Each diagram of evolutionary branching tree was drawn using the Majorbio Cloud Platform (https://cloud.majorbio.com (accessed on 10 February 2025)), and the composite graph was created using PowerPoint 2016 (Microsoft, Redmond, Washington, DC, USA).

2.6.4. Data Analysis

Data processing and statistical analysis were performed using Excel 2016 (Microsoft, Redmond, Washington, DC, USA), SPSS 25 (IBM, Armonk, New York, NY, USA), and Origin 9.3 (OriginLab, Northampton, MA, USA). All results are expressed as the mean ± standard deviation (n ≥ 3). Variance analysis was performed using Duncan’s multiple range test and p < 0.05 was set as significant.

3. Results

3.1. Phenotypes, Enzyme Activities, and SMS Composition Changes During Pretreatment

In the pretreatment, fungi secrete enzymes and degrade the substrate to guarantee nutrient growth and reproductive growth. For anaerobic digestion, the beneficial aspect is that it can open the compact and complex lignocellulosic structure, while the disadvantage is that it consumes some nutrients that should belong to anaerobic bacteria. It is necessary to choose an appropriate pretreatment time to make the pretreatment play a beneficial role.

The phenotype changes in SMS and Tv after pretreatment for different durations were monitored (Figure 1a). SMS and Tv both exhibited evident phenotypic changes, which was conducive to the control of the pretreatment process. For SMS, white mycelia were visualized in Tv24, which had gone through 24 h if pretreatment. And the white deepened in Tv48. The color turned to green in Tv96, and deepened in Tv192. The color change was caused by the different growth states of Tv. Visualization through a microscope showed that the Tv went through mycelial growth in Tv12 and Tv24, and achieved the most prosperity in Tv48. Then, Tv started mycelia autolysis and sporulation in Tv96. Finally, only newly generated chlamydospores were seen in Tv192, which covered the SMS with green mycelia.

The mycelial growth and sporulation of Tv were based on nutrient intake by producing enzymes and degrading SMS. The enzyme activities of CMCase and xylanase were investigated during pretreatment (Figure 1b). The CMCase activity reflects the function of endoglucanase, which cuts cellulose chains into cello-oligosaccharides. In this study, the CMCase activity increased first and then decreased as the duration of pretreatment increased, and reached the peak in Tv48. The xylanases cut hemicellulose into xylo-oligosaccharides and xylose. The trend of xylanase activity was similar to that of CMCase, but the peak value appeared later at Tv96. Both enzyme activities were at a high level in Tv48 and Tv96. Combining phenotypic characteristics, the peak of enzyme activities was observed in the exuberance period of mycelial growth and the early stage of sporulation.

With the growth and reproduction of Tv, SMS was partially degraded during pretreatment (Figure 1c). The weight loss of TS rapidly reached 8.0% of the total TS in the first 48 h of pretreatment; then, the loss slowed down, reaching 13.2% in Tv192. The trends of weight loss rates of cellulose and lignin were similar to those of TS. But, the loss of hemicellulose was uniform over the whole pretreatment process. The loss rate of hemicellulose reached 27.9% in Tv192, which might lead to insufficient substrate availability in anaerobic digestion. Notably, the trend of weight loss was inconsistent with that of enzyme activities. This may be due to the limited flowability of enzymes in solid-state fermentation in which the enzymes produced in the later stage of pretreatment had not interacted with SMS. If so, these enzymes could continue to play a degradation role when dissolved in anaerobic systems and met with SMS.

In summary, pretreatment durations of 24 h and 48 h might be beneficial for subsequent anaerobic digestion, because Tv had started to produce enzymes actively and the SMS weight loss remained limited in Tv24 and Tv48. The visible phenotype is that the SMS is covered with white mycelia, which is easy to identify in practice.

3.2. Methanogenic Performance of SMS After Tv Chlamydospores Pretreatment

A 40-day batch anaerobic digestion experiment was conducted on SMS after chlamydospore pretreatment of varying durations. The methane production and SMS consumption are shown in Figure 2. The digestion process was stable for all treatments. With the longer pretreatment duration, the methane yield of SMS increased first and then decreased. Compared with Tv0, the methane yields of Tv 12, Tv24, and Tv48 were higher by 8.6%, 16.8%, and 7.9%, while those of Tv96 and Tv192 were lower by 2.6% and 14.4%. The highest cumulative methane yield was achieved at Tv24, the initial stage of Tv mycelium development, and when the white mycelium was visible on the substrate surface.

The daily methane yields of Tv12 (17.4 mL/gVS·d), Tv24 (17.5 mL/gVS·d), and Tv48 (15.8 mL/gVS·d) were significantly higher than those of Tv0 (10.3 mL/gVS·d), Tv96 (11.0 mL/gVS·d), and Tv192 (11.6 mL/gVS·d). This indicates that the pretreated SMS with a duration of less than 48 h contained abundant substances that were readily convertible into methane. In contrast, it is hypothesized that, with longer pretreatment durations exceeding 48 h, these readily convertible substances may have been consumed by Tv, rather than by anaerobic microbes. Additionally, the concentrations of easily digestible substances, such as acetic acid, propionic acid, and butyric acid, were measured and will be presented in the following section, further supporting this hypothesis regarding the availability of rapidly convertible substrates for methane production.

With a longer pretreatment duration than 48 h, the distribution of the methane content was more centralized, indicating that multifarious easily degradable substances had been consumed, and the complicated lignocellulosic structure was left. In summary, the SMS substrate changed from diversification to simplification. According to the results of first-order hydrolysis and modified Gompertz models (

Table 2. Model fitting parameters of the first-order kinetic model and the modified Gompertz model in the digestion of Tv pretreated SMS.

	CH4 Yield (mL/gVS)	First-Order Model				Modified Gompertz Model
		R2	K (d−1)	Mmax (mL/gVS)	DIFF *	R2	Rmax (mL/gVS·d)	λ (d)	Mmax (mL/gVS)	DIFF *
Tv0	148.5 ± 11.6 b	0.9938	0.70 ± 0.02	163.4 ± 3.3	10.0%	0.9952	8.3 ± 0.2	0.04 ± 0.25	146.9 ± 1.8	−1.0%
Tv12	161.2 ± 18.9 ab	0.9993	0.88 ± 0.01	167.2 ± 0.9	3.7%	0.9895	9.1 ± 0.3	−0.98 ± 0.36	157.6 ± 2.7	−2.3%
Tv24	173.4 ± 22.6 a	0.9986	0.89 ± 0.01	179.8 ± 1.3	3.7%	0.9892	9.8 ± 0.4	−1.06 ± 0.37	170.1 ± 2.9	−1.9%
Tv48	160.2 ± 12.8 ab	0.9991	0.86 ± 0.01	166.1 ± 1.0	3.7%	0.9870	8.6 ± 0.4	−1.22 ± 0.41	157.0 ± 3.1	−2.0%
Tv96	144.6 ± 8.5 b	0.9952	0.70 ± 0.02	157.7 ± 2.8	9.0%	0.9928	8.0 ± 0.2	0.01 ± 0.30	141.6 ± 2.2	−2.0%
Tv192	127.0 ± 8.6 c	0.9950	0.65 ± 0.02	141.1 ± 2.7	11.1%	0.9927	6.6 ± 0.2	−0.07 ± 0.32	125.8 ± 2.0	−1.0%

^a–c different letters represent significant differences among the treatments (p < 0.05). * difference between M_max and CH₄ yield.

), the K and Rmax of Tv12, Tv24, and Tv48 were higher than those of the other treatments, indicating that pretreatment shorter than 48 h was beneficial for improving the hydrolysis and methanogenesis rates in the anaerobic digestion of SMS.

The trends of weight loss in anaerobic digestion were consistent with those of methane yield. With a longer pretreatment duration, the weight loss rates of TS, hemicellulose, and cellulose initially increased and then decreased. The weight loss peak was achieved at Tv24, where 45.0% of TS, 70.2% of hemicellulose, 41.7% of cellulose, and 13.6% of lignin in SMS were converted into a methane yield of 173.4 ± 22.6 mL/gVS, which was also the maximum yield among all treatments. The reduction of lignin in anaerobic digestion hardly changed with the duration of pretreatment. Lignin is hardly degradable by anaerobic bacteria, and the slight reduction might be due to the dissolution of a small part in a weak alkaline environment [31].

In summary, SMS pretreated with Tv for 24 h achieved a higher hydrolysis rate and methanogenic rate in the anaerobic process, resulting in higher degradation rates of hemicellulose and cellulose, and significantly enhanced the methane yield by 16.8% over that of the control. However, pretreatment above 48 h was unfavorable for the methanogenesis of SMS.

3.3. Dynamic Changes of pH, VFAs, TAN, and FAN During Anaerobic Digestion

pH, VFAs, and ammonia are vital indicators for the dynamic balance between acidification and methanogenesis in anaerobic systems. As shown in Figure 3, the dynamic change in pH showed a trend of first suddenly decreasing and then gradually increasing. The pH value was negatively correlated with VFA concentration. However, the lowest pH and highest VFA concentration among all treatments were 7.37 and 1.005 g/L for Tv12 that caused no disturbance to the fermentation system.

Ammonia is a direct nitrogen source for microbes, a buffering agent for pH adjustment, and an inhibitor when present in excessive amounts. In this study, the TAN concentrations ranged from 1.048 g/L to 1.526 g/L. According to the study referenced [32], ammonia’s inhibitory effects typically manifest at higher concentrations, ranging from 3.4 to 5.77 g/L. This indicates that the TAN concentrations tested in our study were not sufficient to cause excessive inhibition. In Tv96 and Tv192, the TAN levels were slightly lower, which suggests that Tv might have consumed nitrogen from SMS during pretreatment. The FAN in this study did not reach the normal inhibition concentration [30], and the trend was highly similar to that of pH.

In summary, the dynamic changes in pH, VFAs, TAN, and FAN showed that the processes of methanogenesis were relatively stable without inhibitory stress, and the methane yield could be considered as methanogenic potential. A pretreatment duration of less than 48 h could enhance acidification and then achieve efficient methane production.

3.4. Effect of Tv Chlamydospores. Pretreatment on Microbial Community Structures

Generally, research on anaerobic digestion focuses on bacteria and archaea, while relatively little is known about fungi. However, fungi can induce trackable changes in microbial systems, such as those in soil and ruminants [18,19,20]. To explore the effect of Tv and its pretreatment on the anaerobic microbial ecosystem, communities of fungi, bacteria, and archaea were investigated on days 1 and day 20 of anaerobic digestion by high-throughput sequencing.

3.4.1. Alpha Diversity Analysis of Fungal, Bacterial, and Archaeal Communities

Alpha diversity analysis was conducted and is summarized in

Table 3. The alpha diversity indexes of microbial communities on the 1st and 20th days of anaerobic digestion.

	Fungi			Bacteria			Archaea
	Chao	Heip	Shannon	Chao	Heip	Shannon	Chao	Heip	Shannon
Inoculum	46	0.466	3.09	365	0.077	3.19	46	0.012	0.25
Day_1
Tv0_1	144	0.362	3.96	433	0.096	3.53	61	0.015	0.56
Tv12_1	44	0.682	3.41	422	0.081	3.35	29	0.032	0.63
Tv24_1	23	0.390	2.26	394	0.084	3.31	57	0.017	0.60
Tv48_1	29	0.246	2.06	372	0.091	3.38	47	0.017	0.56
Tv96_1	5	0.005	0.02	428	0.083	3.33	46	0.015	0.47
Tv192_1	5	0.042	0.16	370	0.090	3.33	52	0.010	0.38
Day_20
Tv0_20	23	0.186	1.59	571	0.115	3.89	80	0.031	1.17
Tv12_20	109	0.407	3.81	548	0.115	4.04	74	0.018	0.48
Tv24_20	107	0.125	2.65	555	0.126	4.15	63	0.019	0.69
Tv48_20	31	0.000	0.01	555	0.090	3.79	82	0.047	1.52
Tv96_20	7	0.000	0.00	547	0.086	3.75	86	0.017	0.87
Tv192_20	7	0.005	0.02	521	0.081	3.63	73	0.019	0.74

. The Chao, Heip, and Shannon indexes reflect the richness, evenness, and diversity of each community, respectively. Compared with bacteria and archaea, the pretreatment of Tv caused greater differences among fungal communities as the richness, evenness, and diversity of the fungal community declined drastically with the pretreatment duration longer than 48 h, whether in the beginning or the later stage. This demonstrated the exclusivity of Tv within the fungal community. A previous study also reported this kind of phenomenon in the soil microbial system [31]. For the bacterial community, different Tv pretreatment durations had little effect at the beginning of fermentation, but the evenness and diversity of bacteria of Tv48, Tv96, and Tv192 in the later stage of fermentation were lower than those in Tv0, Tv12, and Tv24. In anaerobic digestion, the functions of bacteria are redundant, as various bacteria carry out the same metabolic processes [33]. Lower evenness and diversity indicate that an individual taxon of bacteria occupies the dominant niche. Two possible explanations for this phenomenon are: (1) Tv or its metabolites induce and regulate the bacterial community; (2) the over-consumed SMS leaves monotonous nutrients to create a metabolic pattern with dominant groups. For the archaeal community, the effect of Tv pretreatment was observed at the beginning of digestion as the Heip and Shannon indices decreased over the duration of pretreatment, but the effect did not continue in the later stage.

3.4.2. Beta Diversity and Composition Analysis of Fungal, Bacterial, and Archaeal Communities

Beta diversity analysis of hierarchical clustering showed that the differences between fungal communities were the largest, which was consistent with the results of the α-indexes (Figure 4). The differences in fungal communities were mainly caused by the changes in Trichoderma, which was also the genus of the pretreatment strain Tv. The abundance of Trichoderma in the inoculum was 0.8%. As the duration of Tv pretreatment increased, the abundance of Trichoderma on the first day of anaerobic digestion increased from 2.9% (Tv0), 10.7% (Tv12), 39.4% (Tv24), and 50.0% (Tv48) to 99.7% (Tv96) and 96.7% (Tv192). This indicated that the longer the pretreatment duration, the greater the Tv dose entering the anaerobic digesters. On fermentation day 20, the later stage of digestion, the abundance of Trichoderma exhibited a two-level differentiation. Trichoderma did not dominate the fungal communities in Tv0 (5.7%), Tv12 (14.2%), and Tv24 (5.9%), which had a short pretreatment duration and low-dose Tv. However, it had an absolute advantage in Tv48 (99.9%), Tv96 (99.9%), and Tv192 (99.7%), which had a long pretreatment duration and high-dose Tv. When the initial abundance of Trichoderma was higher than 50%, it exhibited absolute exclusivity in the fungal communities of anaerobic digestion.

When Trichoderma was not dominant, the dominant species in fungal communities were from the Penicillium, Aspergillus, and Apiotrichum genera. The relative abundances of Penicillium, Aspergillus, and Apiotrichum species were 72.1%, 31.3%, and 57.9% in Tv0, Tv12, and Tv24 on digestion day 20. For Penicillium and Aspergillus species, they are ubiquitous in the environment and participate in carbon and nitrogen cycles. Apiotrichum species are yeast-like fungi, and are usually isolated from human and animal sources. Physiological tests showed that they could metabolize various sugars [34]. Therefore, their function seemed to be similar to hydrolytic bacteria that degraded sugar and provided acids in anaerobic digestion. Moreover, individual species of Aspergillus could participate in formaldehyde assimilation in the xylose monophosphate pathway, which benefits the degradation of lignin monomer vanillin [35].

The impact of Tv pretreatment on bacterial communities was slight on the first day of digestion and greater on day 20. The relative abundances of norank_f_W27, Fermentimonas, and Syntrophomonas, the dominant bacterial taxa, were lower in Tv0, Tv12, and Tv24 than in Tv48, Tv96, and Tv192 on day 20 of digestion. This indicates that a high dose of Tv made the dominant bacterial taxa more advantageous. Similarly, genera with an abundance lower than 5% in Tv0, Tv12, and Tv24 occupied a larger proportion than in Tv48, Tv96, and Tv192. However, the microbial structure of bacterial communities did not change in that the dominant bacteria were dominant throughout the digestion duration. Over 80% of bacteria were from the phyla Firmicutes, Bacteroidota, and Cloacimonadota, and 54.5% to 75.8% of bacteria were from the top 15 bacterial genera. Notably, changes in bacteria over the course of digestion were observed. For example, the abundance of Sphaerochaeta increased from day 1 to day 20 in all treatments, while the abundance of Acholeplasma decreased over time. They both function in gluconeogenesis and fatty acid biosynthesis, and the trade-off mechanism between bacteria with similar functions is worth further investigation.

The archaeal community of the inoculum consisted of 14 genera, 11 of which were related to methanogenesis. Methanosarcina dominated the inoculum (97.0%) and all the archaeal communities from digestion day 1 to day 20. Methanosarcina species can metabolize various substrates in methanogenic metabolic pathways, including H₂/CO₂, acetate, methanol, and methylamine [36]. Methanoculleus and Methanobacterium occupied 1.7% and 0.3% of the inoculum archaeal community. They are both hydrogenotrophic methanogens with resistance to ammonia or acid inhibition [33]. On day 1 of digestion, the relative abundance of Methanosarcina increased along with the pretreatment duration, while the abundance of Methanosaeta, Methanomassiliicoccus, Methanomethylophilaceae, and Methanospirillum decreased. Additionally, changes in methanogens were observed throughout digestion, with the relative abundance of Methanomassiliicoccus being higher on day 20 compared with day 1 and the inoculum.

In summary, the dose of Tv entering the digesters increased over the pretreatment duration. When the relative abundance of Trichoderma was more than 50% at the beginning, it would be the absolutely dominant taxon in the fungal community in anaerobic digestion. The Tv pretreatment with different durations had an impact on archaea at the beginning of digestion, while it impacted bacteria at the later stage of digestion. However, Tv pretreatment only changed the abundance of taxa, and had little effect on bacterial and archaeal community structures.

3.4.3. LEfSe Analysis of Fungal, Bacterial, and Archaeal Communities

To further investigate the effect of Tv treatment on anaerobic microorganisms, the results of high-throughput sequencing were divided into two groups: (1) The low-dose Tv group included Tv0, Tv12, and Tv24; (2) The high-dose Tv group included Tv48, Tv96, and Tv192. Also, the two groups were Trichoderma-non-dominant and Trichoderma-dominant according to the composition of fungal communities on day 20 of digestion. The marker taxa that made significant contributions to the differentiation between the two groups were analyzed using the LEfSe method (Figure 5).

On day 1 of digestion, differential taxa between the high-dose Tv and low-dose Tv groups were found in fungal, bacterial, and archaeal communities. However, differential taxa were only detected in fungal and bacterial communities on day 20 of digestion. The marker taxa that were enriched in the low-dose Tv group were more diverse than those in the high-dose Tv group. For example, 26 and 20 bacterial genera were enriched in the low-dose Tv group, respectively, on day 1 and day 20, while 14 and 12 were enriched in the high-dose Tv group. This may be related to the self-protection and aggression of Trichoderma [31].

The taxa that were identified as marker taxa both on day 1 and on day 20 deserve more attention. A stable relationship between these taxa and Tv dose might exist. For example, the fungal genus Cladosporium was enriched in the low-dose Tv group. Cladosporium could play a role in lignocellulose degradation, because it is commonly found in compost or the rumen and enriched in treatment with more crude fiber [37,38]. Moreover, individual species of Cladosporium are able to grow under a single carbon source of alkali lignin [39]. For bacteria, Fermentimonas and Vagococcus were enriched in the high-dose Tv group, while HN-HF0106, norank__f__Peptococcaceae, Ruminiclostridium, and Thiopseudomonas were enriched in the low-dose Tv group. Ruminiclostridium are typical cellulolytic bacteria, and acetate is their metabolic end-product [40]. The Peptococcaceae include sulfate-reducing bacteria that have been reported to participate in the degradation of monoaromatic hydrocarbons [41]. Thiopseudomonas includes some denitrifying bacteria, and some of them can degrade surfactants [42,43].

In summary, low-dose Tv tended to enrich various taxa that function in degradation, hydrolysis, and acidogenic metabolism in anaerobic digestion. This may partially explain the relatively high methane yield from the low-dose Tv group.

3.4.4. Correlation Network Analysis of Fungal, Bacterial, and Archaeal Communities

The correlation network among fungal, bacterial, and archaeal genera was constructed based on the results of Pearson correlation analysis (Figure 6). The network consisted of 427 nodes, including 148 fungal genera, 265 bacterial genera, and 14 archaeal genera. And, the network consisted of 14,822 edges, which included 12,572 positive correlations (red lines) and 2250 negative correlations (green lines).

Modular analysis was performed on the correlation network and eight modules were identified (Figure 6a). The eight modules were named M1–M8 according to the number of taxa they contained. The biggest module was M1, which contained 151 taxa including 10 fungi, 132 bacteria, and 9 archaea at the genus level. The smallest module was M8, with only four bacterial genera. The relationship between M1, M2, M4, and M5 was closer, as they overlapped with each other, while M3, M6, M7, and M8 were individual modules.

Trichoderma, the genus of Tv, was in module M1. Direct relationships of Trichoderma with 11 taxa were found (Figure 6b,c). The 11 taxa were from 5 modules, namely M1, M2, M3, M4, and M6, included 6 fungal taxa, 4 bacterial taxa, and 1 archaeal taxon. Trichoderma was negatively correlated with all six fungal taxa. For bacteria, Pseudomonas and norank_f_Peptococcaceae were negatively correlated with Trichoderma, while Defluviitoga, norank_f_Dysgonomonadaceae had a positive correlation. Species of Defluviitoga are probably functional in glucose metabolism and acid synthesis. Species from the Peptococcaceae family can play a role in polysaccharide hydrolysis in anaerobic digestion [44]. The only archaeal taxon that was directly correlated with Trichoderma was Methanolobus. It is a type of methanogen with various methanogenic metabolic pathways. Acetate, methanol, methylamine, and H₂/CO₂ can all be substrates for Methanolobus [33]. To our knowledge, this is the first report that Trichoderma and Methanolobus are correlated. It is worth further verification and exploration of the mechanism. The 11 Tv-correlated taxa were further correlated with 235 taxa covering all modules except for M8 (Figure 6b). These Tv-one-step-correlated taxa covered 58% of the total taxa in the correlation network.

Fungi were distributed in all modules except for M8. There were 3141 correlation edges between fungal taxa and bacterial taxa, of which 3074 edges were positive and 67 edges were negative. There were 83 correlation edges between fungal taxa and archaeal taxa, of which 77 edges were positive and 6 edges were negative. Considering the negative correlation between Trichoderma and other fungal taxa, the taxa that were negatively correlated with fungal taxa are worth paying attention to, because they probably favor the environment that is abundant in Trichoderma. For example, six fungal genera were negatively correlated with the dominant methanogen genus of Methanosarina. There were 38 fungal genera negatively correlated with the bacterial family of Propionibacteriaceae, indicating that fungi in anaerobic digestion may interfere with the synthesis of propionic acid and acetic acid from monosaccharide degradation. Six fungal genera were negatively correlated with the bacterial family Christensenellaceae. Species of Christensenellaceae are commonly reported in studies of intestinal flora, and are generally positively related to body mass index (BMI), inflammation, and fat malabsorption [45]. Therefore it is speculated that fungi may have a regulatory effect on fat metabolism in anaerobic digestion.

Since methanogens are necessary for methane production, the position of methanogenic archaea in the network was marked (Figure 6a,b). The results showed methanogens were in five modules out of the total eight modules. Coincidentally, these modules with methanogens specifically contained the taxa directly related to Trichoderma. Moreover, module M1 where Trichoderma was located contained nine methanogenic archaea, which was the most. The relationship between Trichoderma and methanogens is worth further study.

In summary, Trichoderma was found in the core methanogenic module M1 of anaerobic digestion. Trichoderma was directly related to 11 taxa, and through these 11 taxa, 58% of the taxa in the microbial community were affected. Fungal taxa may influence propionic acid metabolism and fat metabolism in anaerobic digestion.

3.5. Production Recommendation and Future Prospects

The pretreatment of Tv chlamydospores could enhance the degradation of lignocellulose and promote the methane yield of SMS, but the nutrient consumption for Tv growth was large when the pretreatment duration was prolonged. This suggests that the control of pretreatment duration is worth paying great attention to in practice. The results showed that the optimum Tv chlamydospore pretreatment duration for methane production was 24 h, resulting in a 16.8% increment in methane yield. In addition, the phenomenon of evident white mycelium coverage could provide an indication for the determination of the optimum pretreatment duration in actual production. The increased cost of Tv pretreatment based on this study was 66.7 CNY per 1000 kg SMS (20 CNY per 500 mL Tv chlamydospore solution). The increased methane yield was 22.4 m³ per 1000 kg SMS. If the unit price of methane is 5 CNY/m³ (data come from field survey in 2022, Jinan, China), the increased income from selling methane is 112.2 CNY per 1000 kg SMS. In summary, a net income of 45.5 CNY can be achieved per 1000 kg SMS by Tv chlamydospore pretreatment. Semi-continuous digestion is needed to verify the enhancement effect. In addition, the results showed that the pretreatment process was also the process of chlamydospore expansion. If the pretreated SMS of Tv192 can be recirculated as a new pretreatment agent, the input type of chlamydospore will be changed from continuous to one-time, which is very significant for cost-saving. It is worthy of further investigation and verification.

Tv could survive through a 20-day anaerobic digestion, even becoming absolutely dominant in the fungal community when the pretreatment duration is longer than 48 h. It may benefit the utilization of digestate as an input for subsequent agricultural applications. Some fungal species are potential pathogens of plants. For example, species of Cladosporium and Aspergillus can induce germination inhibition, blights, and blemishes in maize plants [46,47]. Most of these phytopathogenic fungi lost their dominant position in anaerobic digestion after Tv pretreatment. However, as the newly dominant fungi of the digestate, Tv itself is a plant growth-promoting biological agent, which is often used in organic fertilizer. A germination experiment showed that the digestate after Tv pretreatment had a good effect on the germination of brassica campestris (

Table 4. Germination characteristics after application of Tv-pretreated SMS digestate.

	Germi- Nation Rate (%)	Plant Height (mm)	Root Length (mm)	Fresh Weight (mg)	Dry Weight (mg)	Index of Seeding Quality
BL	66.7 ± 7.2 a	17.0 ± 2.9 d	23.9 ± 7.9 c	48.5 ± 5.3 d	3.0 ± 0.3 d	0.200
Tv0	60.4 ± 9.1 a	18.3 ± 1.8 cd	43.3 ± 9.8 b	70.6 ± 5.9 c	4.5 ± 0.5 c	0.547
Tv12	58.2 ± 13.3 a	19.8 ± 2.6 bc	44.3 ± 8.6 ab	73.2 ± 6.4 bc	4.8 ± 0.5 bc	0.610
Tv24	56.8 ± 6.4 a	20.3 ± 1.9 bc	46.8 ± 7.7 ab	75.6 ± 7.8 abc	4.8 ± 0.6 bc	0.636
Tv48	61.5 ± 4.5 a	20.9 ± 2.8 ab	47.8 ± 6.8 ab	76.2 ± 3.7 ab	5.1 ± 0.7 ab	0.772
Tv96	56.6 ± 5.7 a	21.8 ± 2.5 a	49.2 ± 8.7 a	77.9 ± 8.9 a	5.4 ± 0.5 a	0.828
Tv192	54.9 ± 9.1 a	22.7 ± 3.4 ab	50.6 ± 6.9 ab	79.5 ± 4.4 ab	5.5 ± 0.6 a	0.739

Average ± standard deviation. ^a–d different letters represent significant differences among the treatments (p < 0.05).

). For a circular farm, some benefits from the planting sector in addition to the biogas sector could be drawn through Tv pretreatment. Further study and accounting are needed before practice.

4. Conclusions

SMS was pretreated with Tv chlamydospores for various durations and then digested, resulting in different methane yields. The highest methane yield of 173.4 mL/gVS was achieved with 24 h of pretreatment. Pretreatments lasting longer than 48 h were unfavorable for the methanogenesis of SMS. Tv shows exclusivity in the fungal community. When the initial abundance of Tv exceeds 50%, Tv would be the dominant fungal taxa during the later digestion stage. Meanwhile, Tv has a limited impact on bacterial and archaeal communities. Tv chlamydospore pretreatment has promising application prospects from the perspective of cost accounting and operation convenience.