Effects of black soldier fly larvae on biotransformation and residues of spent mushroom substrate and wet distiller’s grains

Abstract

Black soldier fly larvae (BSFL) could convert a variety of organic wastes, including spent mushroom substrate (SMS) and wet distiller’s grains (WDG). Nevertheless, little is known about the conversion of these wastes by BSFL. Thus, this study investigates the conversion of SMS and WDG in five different proportions by BSFL. This study demonstrates that BSFL can convert SMS, WDG, and their mixtures. It can also encourage the humification of the substrate, increasing the amount of element in the residues. It is evident that there were differences in the carbon and nitrogen element fractionation mode as well as the microbial community present in the residue. The microbial community of the substrate and the physiochemical parameters are intimately related to this. Although the mixture treated with BSFL helps to generate a residue with more humus, it might not be stable.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-024-72959-y.

Introduction

Waste management is the biggest concern in modern society due to the unprecedented increase in the total waste generation caused by the urbanization, population growth, and industrialization¹. Among these wastes, spent mushroom substrate (SMS) and wet distiller’s grains (WDG) are two growing organic wastes with potential applications.

SMS is the residues of edible mushroom after harvesting, which mostly consist of lignocellulosic biomass and residual fungal mycelium². Each kilogram of fresh mushrooms generates around 5 kg of SMS³. There are many ways to recycle SMS, but due to the limitation of economic benefit and technical feasibility, composting is the most commonly used method⁴. However, it also has the problems of long loss time and low product value⁵.

WDG is the residual solids after grain fermentation to produce biofuels (ethanol), it has high nutrition and the protein content is about 36.6%⁶. Due to its high moisture content, it can only be stored for 4–5 days⁷; after drying, it is typically used as feed. However, due to issues like mycotoxin contamination and nutritional instability, its use in animal feed is severely limite⁸.

With the increasing demand for edible mushroom and biofuels, the production of SMS and WDG is expanding every year^9,10. Hence, the exploration and implementation of alternative utilization approaches for SMS and WDG are crucial to reduce waste and maximize their potential benefits.

Currently, the use of insects for waste management offers a sustainable and cost-effective solution that could help address the challenges of waste management by many countries worldwide. The insect, particularly black soldier fly (Hermetia illucens L.) larvae (BSFL), has an excellent treatment effect on a variety of organic wastes^11,12. The treatment of BSFL can inhibit the production of harmful bacteria. BSFL accelerates the decomposition of organic matter and converts it into valuable animal feed and fertilizer^13,14. However, research on the transformation of organic matter and the transformation of nutrient elements during BSFL feeding mainly focuses on food wastes and manure, and there is little research on the mode of action of BSFL on other wastes¹⁵.

The substrate has a major impact on how well BSFL processes organic matter and how much element is left in the residues. For example, after treating chicken, pig, and cow manure with BSFL, the aromatic protein substrate in the residue decreased and it became more humic^16,17. In other study of BSFL processed pig manure, the conversion rates of carbon and nitrogen were 13% and 25% respectively and 53% nitrogen and 73% carbon could be saved in the residue¹⁵. In addition, BSFL could convert 9.51% total organic carbon and 16% total nitrogen in food waste, leaving 66% carbon and 65% nitrogen in the residues¹⁸. At present, there is no comprehensive study on the treatment of different substrates by BSFL.

Sawdust SMS and corn WDG are the most productive in the edible mushroom industry and alcohol production industry¹⁹. Therefore, this study selected them as research objects to explore the transformation of different proportions of SMS and WDG by BSFL and its influence on the residues.

Results

Growth and treatment performance of BSFL

BSFL survived in all five formulations, indicating that BSFL has biotransformed each of the substrate (Fig. 1a). There were significant differences in waste reduction and bioconversion rate of different substrates (Fig. 1b). The waste reduction and bioconversion rate ranged between 31.69% (T5) and 64.34% (T1), 1.46% (T5) and 13.96% (T1) respectively (Fig. 1b). The treatment performance of BSFL is roughly positive related to the amount of WDG added.

Fig. 1.

(a) Changes in body weight of 10 black soldier fly (BSFL) during experiments with different substrates. (b) Effects of the different substrate on the waste reduction and bioconversion rate of BSFL. Means, standard deviations and results per replicate are displayed. Results with no shared letter are significantly different from each other. All results are given in dry mass. T1: 100% wet distiller’s grains (WDG) + BSFL; T2: 75% WDG + 25% spent mushroom substrate (SMS) + BSFL; T3: 50% WDG + 50% SMS + BSFL; T4: WDG + 75% SMS + BSFL; T5: 100% SMS + BSFL.

The dissolve organic matter content and spectra analysis

In order to explore the degradation of organic matter by BSFL, the content and composition of dissolved organic matter (DOM) from the original substrates, control substrates and residues were determined. In the raw materials, the SMS (28.8%) has a higher DOM content than WDG (13.66%), so the DOM content of 1–5 substrates was gradually increased (Table 1). After treatment with BSFL, the DOM content was significant increase, reaching 79.38%, 60.81%, 57.62%, 60.66%, 54.62% respectively (Table 1).

Table 1.

DOM content in initial substrate, control substrate and residue in different treatments.

	Substrate
	1	2	3	4	5
Initial	28.8	24.62	22.43	18.51	13.66
Control	42.45 ± 0.46	34.16 ± 0.40	37.18 ± 0.50	42.31 ± 0.47	52.4 ± 0.53
Residue	79.38 ± 0.42	60.81 ± 0.55	57.62 ± 0.65	60.66 ± 0.30	54.62 ± 0.37

Note: T1: 100% wet distiller’s grains (WDG) ; 2: 75% WDG + 25% spent mushroom substrate (SMS) ; 3: 50% WDG + 50% SMS; 4: WDG + 75% SMS ; 5: 100% SMS + BSFL.

To investigate the conversion of organic matter by BSFL, the DOM of different samples was analyzed by UV-vis and excitation-emission matrix (EEM). The UV-vis spectra show the 100% WDG of initial and control has obvious shoulder peak at the wavelength of 250–260 nm, the shoulder peak disappears after with BSFL treatment (Fig. 2a). The SUVA₂₅₄, SUVA₂₈₀ and E₂₅₃/E₂₀₃ value of residues increased after with BSFL treatment compared with the control. The SUVA₂₅₄ value of the treatment respectively rose by 0.31, 0.50, 1.20, 0.64 and 0.15. The SUVA₂₈₀ value respectively rose by 0.02, 0, 0.78, 0.51 and 0.01. The E₂₅₃/E₂₀₃ value respectively rose by 0.16, 0.21, 0.05, 0.02 and 0.03.

Fig. 2.

(a) UV-vis spectra of DOM from different sample, (b) P_{i, n} of the five regions of DOM from different sample. (c) The seed germination index of different residues. GR , germination rate; RL, root length; GI, germination index. Means, standard deviations and results per replicate are displayed. Results with no shared letter are significantly different from each other. IS, initial substrate; CK, control substrate without BSFL; T, residue.

The percentage fluorescence response (P_{i, n}) of each region in EEM spectrum presented in Fig. 2b. In the initial substrate, the accumulative P_{i, n} value of region I, region II in WDG (46.76%) is higher than in the SMS (37.54%), the accumulative P_{i, n} value of region III, region V in WDG (31.12%) is lower than in the SMS (45.10%). After treatment with BSFL, the accumulative P_{i, n} value of region I, region II in five residues were 43%, 30%, 32%, 31%, 32% respectively. The accumulative P_{i, n} value of region III, region V were 37%, 55%, 52%, 54%, 54% respectively. The P_{i, n} value of region IV were 20%, 15%, 16%, 16%, 14% respectively (Fig. 2b). The accumulative P_{i, n} value of region I, region II in five treatments respectively reduced by 4%, 9%, 5%, 2% and 1%. The accumulative P_{i, n} value of region III, region V in five treatments respectively increased by 7%, 12%, 8%, 4% and 4%.

In order to evaluate the availability of residues as fertilizer, GI value used as an evaluation index, which calculated from the seed germination rate (GR) and root length (RL). The GR value of five residues ranged from 90 to 112%, with no significant difference, while the RL and GI value were significantly different (Fig. 2c). The RL value of the five residues were 121.03%, 25.12%, 76.17%, 83.00% and 89.01% respectively, and the GI value were 117.12%, 24.32%, 79.03%, 86.14% and 92.30% respectively.

The above results show that the increase degree of DOM content by BSFL is positively correlates with the treatment performance. Adding a small amount of SMS to WDG when treated by BSFL is helpful to the formation of humus, but the residue may be unstable.

Physicochemical parameters analysis

In order to clarify the impact of BSFL treatment on substrate elements, the content of total kjeldahl nitrogen (TKN), total organic carbon (TOC), total phosphorus (TP) and total potassium (TK) in initial substrates, control substrates and residues were determined and their total content were calculated. In the initial substrate, WDG has higher content of TKN and SMS has higher content of TP and TK (Table 2). Substrates with BSFL substantially decreased the total content of TOC, TKN, TP and TK in the residues. The contents of TOC, TKN, TP, and TK in the residue increased (Table 2).

Table 2.

The TOC, TKN, TP and TK content in initial substrate, control substrate and residue in different treatments.

Subsrate	TOC(%)			TKN(%)			TP(%)			TK(%)
	Initial	Control	Residue	Initial	Control	Residue	Initial	Control	Residue	Initial	Control	Residue
1	45.13	43.15	42.82 ± 0.97ab	1.73	1.71	3.36 ± 0.14a	0.24	0.25	0.68 ± 0.04b	0.18	0.18	0.36 ± 0.02e
2	44.79a	43.78	45.29 ± 2.96a	1.62	1.57	2.45 ± 0.11b	0.33	0.33	0.61 ± 0.05c	0.37	0.37	0.63 ± 0.03d
3	43.44	41.61	40.48 ± 0.69b	1.43	1.45	2.10 ± 0.06c	0.41	0.42	0.65 ± 0.04bc	0.53	0.54.	0.68 ± 0.03c
4	40.10	37.58	40.49 ± 1.10b	1.33	1.32	1.81 ± 0.02d	0.50	0.53	0.64 ± 0.02bc	0.74	0.76	0.85 ± 0.01b
5	40.75	40.75	37.40 ± 0.11c	1.21	1.25	1.77 ± 0.10d	0.59	0.60	0.79 ± 0.02a	0.91	0.93	1.18 ± 0.06a

Note: TOC, total organic carbon; TKN, total Kjeldahl nitrogen; TP, total phosphorus; TK, total potassium.

The loss of different substrates varied significantly (Fig. 3). With the increase of SMS, the loss of TOC gradually decreased, the reduction in the T1 was 57.22%, and T5 was 21.72%. The TKN reduction between T1 and controls has no significant difference. The TKN reduction of T1, T2 and T3 (24–26%) were significantly higher than T4 and T5 (1–6%). There was no significant difference of TP reduction between theT1 and the control. The TP reduction of other four treatments was were 11.97%, 20.71%, 16.20% and 6.58%, respectively. The TK reductions of five treatments were 25.28%, 19.62%, 35.34%, 21.46%, and 8.45% respectively.

Fig. 3.

The reduction of TOC (a), TKN (b), TP (c) and TK (d) in different treatments by BSFL. The changes of δ13C (e) and δ15N (f) in different treatments by BSFL. The mean and standard deviation of three replicates are displayed, the results without shared letters are significantly different from each other. TOC, total organic carbon; TKN, total Kjeldahl nitrogen; TP, total phosphorus; TK, total potassium.

In order to explore the loss patterns of carbon and nitrogen elements during BSFL treatment, the isotope content of three samples were measured. According to the type of carbon metabolism of plant origin, WDG is C₄ category and SMS is C₃ category. After treatment with BSFL, the δ¹³C value of the residues in T1, T2 and T4 were not significantly different from those in the control. The δ¹³C value of the T3 and T5 were significantly increase compared to the control, with the increases being 1.48‰ and 0.31‰ respectively (Fig. 3e).

The δ¹⁵N value of the residues in T3 was not significantly different from its control, while the others treatment were significantly different from their control. Among them, the δ¹⁵N value of T1 decreases, and the decrease value is 0.285‰. The δ¹⁵N value of T2, T3 and T5 increase, and the increased value are 0.157‰, 0.481‰ and 0.635‰ (Fig. 3f), with significant differences. In summary, BSFL will significantly consume TKN, TOC, TP and TK in the substrate, significantly increase their content. Moreover, it significantly change the fractionation pattern of carbon and nitrogen elements in some substrate.

Bacterial community of different residues

In order to investigate the treatment of different substrates by BSFL from a microbial perspective, the bacterial communities of the residues were measured. The Chao1 index, Shannon index and unique OTUs of T5 residues was the highest exhibiting a significant difference from other treatment (Fig. 4, Table S1). It showed that the microorganisms in the residues of T5 are the most abundant. According to PCoA, the residues under different treatment clustered well (R = 0.9037, P = 0.001). The two principal coordinates, PCoA1 and PCoA2 axis explained 76.34% and 11.98% of the variation in the bacterial community, respectively. On the PCoA axis, the distance between the bacterial communities of different treatment was obvious, showed that there were significant differences in the bacterial community of residues of different treatment.

Fig. 4.

(a) Chao1 richness and (b) Shannon index for different residues (n = 3), the results without shared letters are significantly different from each other. (c) Venn diagram displaying unique and shared OTUs of different residues (d) Principal coordinates analysis (PCoA) using relative abundance of OTUs based on the Bray-Curtis distances.

At phylum level (Fig. 5a), Bacteroidetes (13.9–56%), Proteobacteria (42-66%), Actinobacteria (2–11%) and Firmicutes (1.93–4.79%) were the most dominant phyla in all residues, with cumulative percentages ranging from 82.6 to 99.8%. There are significant differences in the bacterial community of different residues. Bacteroidetes of T1 (50.28%) and T2 (56.05%) residues were higher than those other treatments. The residues of T3 (65.97%) and T4 (71.99%) has higher Proteobacteria, Actinobacteria (10.37%) has the highest proportion in T5 residues.

Fig. 5.

(a) Relative abundance of top 4 phylum, (b) Relative abundance of top 20 genus, (c) Cladogram plot generated by LEfSe analysis displaying the taxonomic biomarkers of different residue and the specific biomarkers were shown in the bar chart at genus level (P < 0.05, LDA score ≥ 4.5), (d) Level 3 KEGG function predictions in terms of the relative abundances of the main functions related to amino acid metabolism, (e) carbohydrate metabolism. The mean of three replicates are displayed. The mean of three replicates is displayed.

At genus level (Fig. 5b), the top three genera in relative abundance in all residues were Dyella (68.1%), Sphingobacterium (50%) and Klebsiella (40.4%). There were significant differences in the relative abundance of genera under different treatment. LEfSe analysis (P < 0.05, LDA score ≥ 4.5) was used to further describe the biomarkers in different residues (Fig. 5c). For example, a significant enrichment of phylum Bacteroidetes in T1 was mainly due to the increase of Dysgonomonas (17.28%). Notably, the biomarkers of T3 were Dyella, accounting for 31.37%.

Kyoto Encyclopedia of Genes and Genomes (KEGG) database prediction results display metabolism (47.3–49.9%) in all treatment is the most abundant metabolic pathway in pathway level 1 (Figure S2). Carbohydrate metabolism (10%) and amino acid metabolism (10%) were the main pathways of metabolism-related pathways level 2, and showed significant differences among different treatment (Figure S3). Arginine and proline metabolism were most active, this is followed by glycine, serine and threonine metabolism in the amino acid metabolism (Fig. 5d). The abundance of D- arginine and D- ornithine metabolic increased with the increase of SMS. Pyruvate metabolism, amino sugar and nucleotide sugar metabolism, glycolysis/gluconeogenesis and citrate cycle more active in the carbohydrate metabolism (Fig. 5e).

Relationship between physicochemical parameters and bacterial community in residues

To explore the potential impact between bacterial community and physicochemical parameters, the spearman correlation analysis was carried out the microorganisms of the first 20 genera and physicochemical parameters (Figure S4, Fig. 6, Table S2). The redundancy analysis (RDA) shows that the two axes explained 63.44% and 13.73% of microbial differences, respectively. DOM, TOC and TKN were significantly positively correlated, while TK and TP were significantly positively correlated. RDA indicated that the TK was the prime driver of bacterial variation. Rhodanobacter (B7), TM7a (B14), Parapedobacter (B16), Devosia (B17) were positively correlates with TK and TP. Sphingobacterium (B2), Klebsiella (B3), Dysgonomonas (B5) and Ochrobactrum (B11) were positively correlates with DOM, TKN and TOC.

Fig. 6.

Redundancy analyses of the correlation between relative the bacterial and physiochemical properties. B2: Sphingobacterium; B3:Klebsiella; B5: Dysgonomonas; B7: Rhodanobacter; B11: Ochrobactrum; B14: TM7a; B16: Parapedobacter; B17: Devosia.

Discussions

In order to better evaluate the conversion of BSFL to different substrates, it is necessary to evaluate the conversion efficiency and by-products. Our study comprehensively analyzed the bioconversion rate, waste reduction, change in organic matter, physicochemical parameters and microbial communities at different ratios, enhancing the understanding of the biological transformation process of the BSFL and the influence of substrate on residues.

Growth and treatment performance of BSFL

In this study, the waste reduction and bioconversion rate of 100%WDG by BSFL were 63.34% and 13.96%, which are similar to that of food waste suitable for the growth of BSFL²⁰, indicating WDG are also suitable for the feeding of BSFL (Fig. 1). The bioconversion rate of 100% SMS by BSFL is 1.46%, which is similar to that of cow and both are at low levels, the reason may be the discomfort of BSFL to lignocellulosic²¹.

There is no significant difference in bioconversion rate between T2 and T1, which indicates that adding a small amount of SMS to WDG has no significant effect on the treatment of waste by BSFL. And, adding an appropriate amount of lignocellulosic to high-nutrient substances can to increase the feasibility the residues as fertilizer²². This is conducive to increase the BSFL’s treatment of SMS and increasing the value of residues. In addition, although there are many studies on the treatment performance of BSFL on different substrates²⁰, but the influence of different substrates on the whole BSFL system is still unknown.

Changes in the organic matter

DOM represents the most active organic component, and its evaluation could obtain the internal mechanism of BSFL on organic matter transformation^16,17. This study results show that BSFL can helps to increase the DOM content, the increase of each treatment compared with the control were 36.93%, 26.65%, 20.45%, 18.35% and 2.22% respectively (Table 1). The UV-vis spectra of DOM showed that BSFL treatment increased the SUVA₂₅₄, SUVA₂₈₀ and E₂₅₃/E₂₀₃ value (Fig. 2a), this finding was in accordance with^16,17. It shows that BSFL treatment can increase the content of DOM in the residues, and improve the aromaticity and molecular weight of the substances in it. The increased DOM content could promote the growth of the plant²³.

The EEM spectra of DOM showed that BSFL consumes proteinaceous substances (region I, region II) and formed a humic substrate (region III, region IV) (Figure S1). Among them, the accumulative P_{i, n} value of region III, region IV of IS2 are similar to those in pig manure and chicken manure, while T2 is higher than that in pig manure treated by BSFL and lower than that in chicken manure^16,17. It indicates that the treatment performance of BSFL has an impact on waste transformation. In this study, the treatment performance of T2 by BSFL is similar to that of T1, but had a higher accumulative P_{i, n} value of region III, region IV (55% > 37%). It suggesting that lignocellulose plays an important role in humus formation²⁴. During composting, P_{i, n} value of region IV increased first and then decreased²⁵. In this study, P_{i, n} values of each treatment in region IV were lower than those in the control, suggesting that the residue after BSFL treatment is transitioning to a steady state.

The phytotoxicity assay results show that the residues obtained from T1 is stable and can be used as fertilizer, while other treatment still require post-treatment, especially the T2 residues. It is possible that T2 contains a large number of unstable substances²¹. However, it is still valuable to add an appropriate amount of fibrous substances when the BSFL treats high nitrogen substances²².

Changes in the physicochemical parameters

Due to the consumption of BSFL, the total content of TOC, TKN, TP and TK in most treatments is reduced, while the content is increased (Fig. 3; Table 2). This is beneficial to the subsequent application of residue as fertilizer¹⁴. Compared with the open composting, BSFL bio-treatment of food waste could reduce the emission of greenhouse gas (especially CH₄ and N₂O) and NH3¹⁸. The loss of TOC (21.72–57.22%) in five treatment is much greater than that of TKN (1.12–25.25%) in this study (Fig. 3a and b), this implies higher CO₂ emissions and lower NH₃ emissions. And the reduction of δ¹⁵N of T1 is significantly lower than that of other treatments (Fig. 3f), indicating that the ammonia volatilization amount of WDG treated by BSFL may be less. Substrate will significantly affect carbon and nitrogen emissions when BSFL treating waste. The carbon and nitrogen emissions from pig manure are 15% and 20%¹⁵, while those from wheat bran and flour are 24% and 0.7%²⁶. The δ¹³C value of the T3 and T5 were significantly higher than control. This may be due to the partial loss of lignin by the BSFL, which leads to the decrease of lignin content²⁷, suggesting the potential of BSFL in lignin degradation.

The loss of TKN, TP and TK is not completely consistent with the treatment performance, this shows that the content of elements in the substrate significantly affects the absorption of elements by BSFL, this consistent with the existing research¹⁴. The residual of pig manure with BSFL had 9% less phosphorus and potassium than the manure without BSFL¹⁵. While BSFL feeding on wheat bran and flour can reduce 28% phosphorus and 14% potassium in the substrate²⁶.

Changes in the bacterial community

With the feeding of BSFL, the bacteria in the substrate will colonize in the intestine of BSFL and the intestinal bacteria will reshapes the microbial community in residue²⁸. The microbial abundance in T5 residue was significantly higher than that in other treatments. It may be that BSFL feeds less on 100%SMS, leaving the microbes in the gut with less influence on the substrate and residues retaining more diversity²⁹.

Bacterial species at phylum level in five treatments are consistent with those reported in BSFL gut or residues bacterial^30,31. The abundance at each phylum level differs greatly due to the difference of substrate and BSFL treatment performance. T3–T5 with higher lignocellulosic content. Therefore, Proteobacteria and actinomyces, which can effectively degrade fiber materials, were highly enriched in these three treatment^32,33. In addition, the microorganisms from these two phyla make important contributions to carbohydrate metabolism and amino acid metabolism respectively, which promote the formation of humic acid³⁴. This is also consistent with the high humic content of T3–T5 in this study (Fig. 5a). The high abundance of Bacteroidetes in T1 and T2 is consistent with the use of macromolecular substances such as protein and starch³³.

At the genus level, the relative abundance of the top 30 genera of bacterial community differed significantly among different residues. Among them, Sphingobacterium and Dysgonomonas are the core microbes in the gut of BSFL, and they can significantly promote BSF development by increasing larval and pupal body weight gain³⁰. In this study, their relative abundance of different treatment was consistent with BSFL bioconversion rate and body weight results, especially in the T1, which had 13.96% and 2.45 g (Figs. 1 and 5b).

The prediction results of bacterial functions in the residues showed that carbohydrate and amino acid metabolic pathways have the highest abundance in all treatment (Figure S4), which is consistent with previous studies, but there are some differences in KEGG pathway level 3^35,36. This may be due to differences in rearing substrate and environment. In this study, the main metabolic pathway of carbohydrate metabolism is glucose metabolism, glycolytic–pyruvate–Citrate cycle (Fig. 5d). Among pyruvate metabolism and citrate cycle are positively correlated with humus precursors and humification index³¹, this is consistent with the result that BSFL can promote the humification. In addition, it is worth noting that the T5 had the highest abundance, which may relate to its high microbial diversity.

Main influence factors in residues

In terms of the correlation between physicochemical parameters and bacterial community composition, TK has the greatest impact on bacterial community (Fig. 6). It is similar to the report that TK, PH and TP are the most critical environmental factors³⁵. Dysgonomonas is positively correlated with TOC and DOM (Fig. 5b), which may degrade lignocellulose to improve the level of TOC and DOM^37,38. Rhodanobacter is positively correlated with TKN and negatively correlated with TP (Fig. 5b), which is consistent with the function of nitrogen and phosphorus removal³⁹.

Conclusions

This study show that BSFL can convert SMS, WDG and their mixtures. Among them, 100% WDG is suitable for the growth of BSFL, the waste reduction and bioconversion rate are 64.34% and 13.96%. The 25% SMS in WDG does not reduce the biological conversion rate. The UV-vis and EEM spectra showed that BSFL consume most of the proteinaceous substances as well as formed a humic substrate, and finally increase the aromatic degree of residues. Adding 25% SMS in WDG can make the residues have a higher degree of humification. BSFL increasing the amount of TOC, TKN, TP, and TK in the residues. And the fractionation methods and microbial communities of carbon and nitrogen in different residues were significantly different.TK plays an important role in microbial community change.

This study proposes an environmentally friendly and sustainable solution for the WDG and SMS, accomplishes the recycling of resources. In addition, studies have shown that suitable biowaste formulations can reduce gas pollution and increase the value of subsequent residues without affecting the production performance of BSFL. However, the impact of different substrates on the value of the BSF system still needs to be evaluated, especially the value of the larvae’s subsequent application in aquaculture and livestock farming.

Methods

Experimental materials and design

The BSF eggs collected from the Bijie site, Guizhou province, bred at the Key Laboratory of Institute of Entomology in Guizhou University, Guiyang, China. For the first five days, larvae fed a diet consisting of 75% water-based wheat bran. On the six days, larvae were removed from the wheat bran. The dried SMS obtained from the Institute of Edible Mushroom of Guizhou University, and crushed to a particular size of 2 mm. The WDG purchased from a local ethanol manufacturer of Guizhou province. Five treatments were carried out by adding different amounts WDG and SMS, namely T1 (100% WDG), T2 (75% WDG + 25% SMS), T3 (50% WDG + 50% SMS), T4 (WDG + 75% SMS), T5 (100% SMS). The substrate without BSFL that were incubated for 15 days were regarded as the control (CK).

Determination of BSFL growth and treatment performance

150 g of substrates (humidity 70%) were individually put into each plastic container (22 × 14 × 15 cm). Then, 0.2 g of larvae (10 larvae weigh approximately 0.004 g) were added into each container in the treatment and mixed with the substrate. The rearing conditions maintained at 28 ± 2 ℃, 60–70% relative humidity, and a photocopier of L12:D12. When the pupation rate of larvae in three treatments exceeded 25%, the experiment was ended. All the larvae and prepupae were separated from the mixtures. Some larvae, prepupae and residues were dried and weighed. The initial substrate, the control substrate and the residues were gathered, and then the samples were stored at -80℃ for subsequent examination. The waste reduction and bioconversion rate of BSFL were calculated as Gold et al.²¹. The formula is as follows, all weights are on a dry weight.

DOM extraction and spectra analysis

The solutions were extracted from 10 substrates and 5 residues according to the method of Wang et al.¹⁷. DOM solution was obtained after the solution was filtered through 0.45 μm membrane. The organic carbon content of the DOM solution was analyzed by an automated TOC analyzer and diluted to 10 mg/L by ultra-pure water.

The diluted solutions was used to detect the UV-1452 spectrophotometer, the range was set as 200–600 nm. Meanwhile, the SUVA₂₅₄, SUVA₂₈₀ and E₂₅₃/E₂₀₃ calculated in accordance with Wang et al.⁴⁰. The dilute the solution to UV-vis 260 < 0.02 with ultrapure water, and measure the excitation emission matrix (EEM) fluorescence spectrum with a fluorescence spectrophotometer (Hitachi F-4600, Japan). The parameters of emission wavelength and excitation wavelength were set according to the Wang et al.⁴⁰. According to the study of Chen et al.⁴¹, the EEM spectrum was divided into five regions, and the fluorescence integral of each region was calculated. Determination of seed germination index (GI) with radish seeds according to the method of Wang et al.⁴².

Physicochemical parameters analysis

The TKN, TOC, TP and TK of ten substrates and five residues were determined using the trace kjeldahl method, potassium dichromate method, antimony molybdenum by resistance method and flame photometry method³⁵. The δ¹³C/δ¹² C, δ¹⁵N/¹⁴N ratio of these samples was analyzed using a gas isotope ratio mass spectrometer (MAT253, Thermo Fisher, Waltham, MA, USA).

DNA extraction

The DNA from 5 residues was extracted using the CTAB, the quality of DNA extraction was measured by agarose gel electrophoresis, and the DNA was quantified by ultraviolet spectrophotometer. The V3–V4 regions of the 16 S rRNA genes were amplified using the primers 341 F (CCTACGGGNGGCWGCAG) and 805R (GACTACHVGGGTATCTAATCC). The total DNA was eluted in 50 µL of Elution buffer and stored at − 80 °C until measurement in the PCR by LC-Bio Technology Co., Ltd, Hang Zhou, Zhejiang Province, China.

Data statistical analysis

The statistical analysis was performed with SPSS 22.0, the homogeneity of variance was tested by Levene test. Univariate analysis of variance (Duncan test) were used to determine the differences among samples. Statistical significance was established at P < 0.05. The integration of each region of the EEM spectrum is calculated using Matlab. The PICRUSt2 method was used to predict potential functional profiles of bacterial communities in residues based on the KEGG database⁴³. Venn diagram, principal coordinates analysis (PCoA) diagrams and redundancy analysis (RDA) diagrams were drawn in R 3.5.3. Origin 2019 were used for generation of other plots.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

M. W and T. L conceived the project and designed and interpreted all of the experiments. M. W visualized the data and wrote the manuscript. S. K, H L and J.G sorted out and revised the first draft. T.W provided experimental materials. T.Y is responsible for supervising the whole project. J. G provide financial support.

Funding

This work was funded by two funds, namely Guizhou Provincial Science and Technology Projects (Qiankehe Pingtai Rencai-GCC [2022]029 − 1) and Lancang-Mekong Cooperation Special Fund Projects (JiaoWaiSiYa [2020]619).

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author onreasonable request.

Declarations

Competing interests

The authors declare no competing interests.