The gut microbiome promotes the growth performance of black soldier fly larvae by detoxifying uric acid

This study demonstrates the detrimental effects of exogenous uric acid (UA) on the growth of black soldier fly (BSF) larvae, highlighting the role of gut microbiota in UA degradation. We isolated UA-degrading bacterial strains associated with BSF, including Enterococcus faecalis AHAU24. Genomic analysis identified a UA degradation gene cluster in this strain, providing insights into its potential for UA catabolism. Furthermore, mono-association of germ-free larvae with E. faecalis AHAU24 confirmed that this strain alleviates UA-induced growth retardation and enhances bioconversion efficiency by degrading dietary UA. The black soldier fly (BSF, Hermetia illucens) has recently garnered increasing attention, due to its remarkable power to convert low-value organic waste into high-quality insect biomass [1]. However, many feedstocks contain undesirable or toxic compounds that can adversely affect the growth performance of BSF larvae [2, 3]. For example, poultry manure contains approximately 3% uric acid (UA), which accounts for about 50% of the total nitrogen content in poultry excreta [4]. Notably, it's established that UA is also a major nitrogenous waste product in terrestrial insects [5, 6]. Since UA acts as excreta of both poultry and insects, it is unlikely to be recycled by BSF larvae, raising concerns about its potential toxicity in rearing substrates. Emerging studies reveal that the microbiome, or specific microbial taxa, confer benefits to hosts by degrading a variety of toxins. Certain UA-related microbes have been recently identified in mammals, in which they contribute to the alleviation of UA-induced stress [7]. It's therefore proposed that the microbiome could promote the growth performance of BSF larvae by degrading UA. We aim to (I) evaluate the effect of UA on the growth performance of BSF larvae, (II) identify UA-degrading bacteria associated with larvae, and (III) elucidate the underlying mechanisms by which these bacteria reduce UA toxicity and promote host development. The conserved conversion of nitrogenous waste to UA facilitates water conservation in terrestrial insects [8]. Consistently, UA accumulated predominantly in the Malpighian tubules of BSF larvae, followed by the cuticle and midgut (Figure 1A), indicating that they follow the paradigm of generating UA as the primary end product of nitrogen metabolism [8]. Next, we investigated whether excess UA could adversely impact the growth performance of BSF larvae. Strikingly, an apparent reduction in body weight and width was observed in larvae exposed to increasing doses of UA (Figure 1B,C). Simultaneously, the larval developmental time was prolonged by increasing UA concentrations (Figure 1D), accompanying a reduction in the pupation rate of larvae (Figure 1E). Altogether, our data suggested that UA negatively affected the growth and development of BSF larvae. Given that BSF larvae are highly efficient decomposers, we finally evaluated whether UA could affect bioconversion rates of food. The bioconversion rate plummeted from 14.52% in the control to 8.59% in case of 2% UA (Figure 1F), while the food reduction rate at 2% UA was comparable (Figure S1). In response to toxin-induced stress, genes associated with oxidative stress are often upregulated to mitigate cellular damage [9-11]. Indeed, the expression of genes involved in mitochondrial respiration (COX4), reactive oxygen species generation (Duox), and nitric oxide synthesis (Nos) was significantly upregulated in the larval intestine under UA stress (Figure 1G). In summary, our findings underscore the detrimental impact of UA on BSF larval development and bioconversion efficiency. To study the role of the microbiome in enhancing larval tolerance to UA, germ-free (GF) and mixed-bacteria (MB) associated BSF larvae were generated (Figure S2A), where the axenic state was confirmed by bacterial plating (Figure S2B). Strikingly, GF larvae subjected to UA gained merely 24.6% of the body weight of their siblings under UA-free conditions (Figure 1H,I). Moreover, MB inoculation completely restored the growth retardation observed in UA-treated GF larvae (Figure 1H). Notably, UA-treated MB larvae reached 80.7% of the body weight of their UA-free siblings, suggesting that the microbiome was required to mitigate UA toxicity to larval growth (Figure 1I). In parallel, larvae challenged with UA exhibited a reduced pupation rate, which was similarly rescued by MB inoculation (Figure 1J). Our analysis also revealed that UA treatment severely reduced the bioconversion rate of GF larvae, and this adverse effect was alleviated by MB supplementation (Figure 1K). To explore whether the microbiome contributes to UA degradation, we measured UA levels in the diet. Unanticipatedly, bacterial colonization alone led to a minor but statistically insignificant reduction in UA levels compared to the untreated control (Figure 1L). However, larvae inoculated with MB dramatically decreased UA levels in the diet (Figure 1L). Given that single GF larvae produced a high amount of UA, these findings suggest that the host synergized with the microbiome to degrade UA more efficiently. Collectively, our data suggest that the microbiome is essential for optimizing the growth of BSF larvae under UA stress. Next, we reasoned that UA treatment could alter the composition of the bacterial community. Principal coordinates analysis revealed a clear separation between the control and 1% UA-treated groups (Figure 2A), although exposure to 1% UA affected neither microbial richness nor evenness (Figure S3A). Of note, the Permutational Multivariate Analysis of Variance (PERMANOVA) test did not reach statistical significance (p = 0.105), potentially due to high within-group variation. Moreover, Enterobacteriaceae and Wohlfahrtiimonadaceae, dominant in the control group, were substantially replaced by Rhizobiaceae, Dysgonomonadaceae, and Rhodobacteraceae in the guts of UA-treated larvae (Figure 2B). Based on PICRUSt2 predictions, the analysis revealed downregulation of enriched pathways, including metabolism of cofactors and vitamins, antimicrobial drug resistance, and protein folding, sorting, and degradation (Figure S3B). The taxonomic and functional alterations detected here provide initial evidence of UA's impact on the microbiota, but further validation with larger sample sizes is warranted to ensure generalizability. The broad taxonomic distribution of UA-degrading bacteria prompted us to isolate resident strains to degrade UA by using a UA-supplemented selective medium. Five bacterial strains were isolated from BSF larval guts, including three Enterococcus faecalis strains, one Staphylococcus aureus strain, and one Acinetobacter baumannii strain (Figure 2C). Notably, culture-based isolation yielded relatively less dominant taxa, likely due to the selective pressure imposed by in vitro UA enrichment. However, all strains we isolated exhibited the ability to grow under UA condition and degrade over 50% of total UA within 48 h of cultivation (Figure S4A), while Lactobacillus plantarum, a non-UA-degrading bacterium, failed (Figure S4A,B). Since E. faecalis, a member of the family Enterococcaceae ranking among the top 20 most prevalent bacterial families, has been extensively utilized as a probiotic in both humans and animals [12-14], it was chosen for further examination. Next, we attempted to identify potential genes involved in UA degradation in E. faecalis AHAU24 by assembling its complete genome. The CGView genome map was generated based on the complete sequence (Figure 2D). The genome comprises a 15,881,950 bp circular chromosome with a 37.86% GC content, containing 3025 open reading frames (ORFs) predicted to encode proteins. As shown in Figure 2D, we identified a conserved set of uric-acid-degrading genes, including xanthine dehydrogenase (xdhC), hydrolase (ygeY), carbamoyl transferase (ygeW), oxidoreductase (ygfK), and amidohydrolase (ssnA), located within a 196,904 bp chromosomal fragment (Scaffold6, genomic location from 82,789 to 96,715 bp). To confirm them, we tested the expression of these genes under UA stress by the quantitative PCR technique. The result showed that UA treatment significantly increased the expression levels of these genes in E. faecalis (Figure S4C). Two routes for UA catabolism, converting it to either xanthine or to short-chain fatty acids (SCFA), were identified among phylogenetically distant bacterial taxa [15]. Our results suggest that the uric-acid-inducible genes in E. faecalis AHAU24 might participate in the conversion of UA to SCFA. To further investigate whether E. faecalis could mitigate UA-induced growth retardation in BSF larvae, we conducted rescue experiments by monoassociating GF larvae with E. faecalis, Escherichia coli, or L. plantarum. The results showed that monoassociation with the UA-degrading bacteria, either E. faecalis or E. coli, led to an apparent increase in larval body weight (Figure 2E), whereas L. plantarum conferred much less benefit. Both pupation and bioconversion rates of E. faecalis or E. coli monoassociation larvae were also significantly higher than those of either GF or L. plantarum-associated siblings (Figure 2F,G). Notably, the loads of E. faecalis, E. coli, and L. plantarum in larval guts were comparable after 5 days (Figure S5A). Our results furthermore showed that E. faecalis or E. coli monoassociation was effective in reducing the UA content in both the diet and larvae (Figure 2H and Figure S5B), underlying the alleviation of larval growth arrest under UA stress. Further genomic analysis revealed that E. coli K12 (GenBank accession number: ON394478) harbors a homologous uric acid gene cluster, with the key gene ygfK exhibiting 98% sequence identity. Due to the genetic intractability of E. faecalis, we turned to the more tractable E. coli that harbors the same uric-acid-inducible genes. The ygfK mutant of E. coli almost failed to degrade UA, whereas the wild-type (WT) strain nearly consumed all UA within 48 h (Figure 2I). As anticipated, larvae associated with the ygfK mutant exhibited significantly lower weight gain compared to their WT-associated counterparts (Figure 2J). Consistently, the deletion of ygfK significantly reduced the pupation rate in colonized larvae challenged with UA (Figure 2K). Indeed, the ygfK mutant was deficient in degrading UA in the diet (Figure 2L). Furthermore, we observed that WT strain supplementation mitigated the transcriptional expression of COX4, Duox, and Nos under UA stress, but ygfK mutant did not have an influence on these gene expressions compared to GF larvae (Figure S6). While our current findings provide valuable insights into the potential role of UA-degrading genes, further investigation through cross-species validation, gene complementation studies, or in vivo models would help strengthen the evidence for their causal involvement and enhance the generalizability of these observations. Given that chicken manure generally contains high levels of UA, we finally investigated whether E. faecalis could promote the growth of larvae employed for chicken manure management. To minimize interference from resident UA-degrading bacteria, most of the resident bacteria in the manure were killed by microwave heating for 2 min prior to inoculation with E. faecalis. Our data showed that an increase in body weight was observed following E. faecalis inoculation (Figure S7A). More importantly, the bioconversion rate of larvae was increased in group with E. faecalis supplementation (Figure S7B). Correspondingly, the lowest UA content was observed in chicken manure supplemented with E. faecalis (Figure S7C). One potential limitation of this study is that certain bacteria, including heat-resistant or spore-forming strains, may survive microwave heat-killing and influence experimental outcomes. However, the large inoculum of UA-degrading bacteria likely outcompeted any surviving microbes, so future studies should verify inactivation efficiency and consider complementary sterilization methods if complete elimination is necessary. Here, we report that the gut microbiome is required to promote BSF development by breaking down UA. These findings agree with previous studies on insect herbivores, in which gut bacteria enable the breakdown of plant toxic compounds, such as glucosinolates in Brassicaceae or alkaloids in solanaceous plants [16]. Related UA-degrading genes are present in the genomes of E. faecalis and E. coli, which underlie the mechanism by which these bacteria promote BSF growth performance. However, many other bacteria widely exist in the BSF gut, and additional UA-degrading pathways need to be explored in the future. Understanding the molecular mechanisms provides insights into optimizing BSF-rearing systems by leveraging microbial contributions to boost productivity and waste recycling efficiency [17, 18]. While aerobic Luria-Bertani (LB)-based approach successfully identified facultative anaerobes like Enterococcus as UA degraders, this method inherently selects for oxygen-tolerant, fast-growing bacteria, potentially overlooking anaerobic or slow-growing taxa. Future work should adopt anaerobic culturing conditions or metagenomic analyses to capture a broader diversity of UA-metabolizing microbes. More importantly, the potential synergism among UA-degrading bacteria, as well as their interactions with the broader bacterial community, remains largely unexplored. Understanding these microbial interactions would offer valuable insights into how bacterial consortia collectively contribute to UA metabolism, thereby boosting the overall efficiency of BSF farming. This study systematically investigated the effects of UA on the growth and physiological responses of BSF larvae, as well as the role of intestinal microbiota in UA degradation. Our findings demonstrate that gut bacteria play a pivotal role in mitigating UA-induced stress in BSF larvae. Notably, E. faecalis, a UA-degrading bacterium isolated from the BSF gut, confers host resistance to UA toxicity by efficiently degrading UA. Through genomic and functional analysis, we identified specific UA-degrading genes in E. faecalis that are essential for alleviating UA-related stress. Overall, these findings offer novel insights into host–microbiome interactions in response to UA toxicity, shedding light on potential microbial management strategies to mitigate UA-induced stress. Xiaowen Ji: Methodology; validation; funding acquisition; investigation; writing—original draft. Shaojie Zhou: Methodology; validation; formal analysis; investigation; data curation. Wenwen Chen: Validation; formal analysis. Bin Cao: Validation; formal analysis. Jingjing Zhuang: Validation; formal analysis. Sheng Zhang: Validation; formal analysis. Bingshuo Han: Validation; formal analysis. Benping Yang: Validation; formal analysis. Shuxiang Zhang: Resources; funding acquisition. Yanfeng Xue: Resources. Jibin Zhang: Supervision; project administration. Yinglao Zhang: Supervision; project administration. Wei Liu: Supervision; project administration; funding acquisition; writing—original draft; writing—review & editing. All authors have read the final manuscript and approved it for publication. The authors would like to thank all members of Dr. Liu's lab for discussion. This work was supported by the National Natural Science Foundation of China (32470044), Anhui Provincial Natural Science Foundation (2308085MC74), the Foundation of Anhui Province Key Laboratory of Resource Insect Biology and Innovative Utilization (FKLRIB202401), National Key Laboratory of Agricultural Microbiology (AMLKF 202309, AMLKF202407), Talents in Anhui Agricultural University (RC342201), and China College Students' “Internet+” Innovation and Entrepreneurship Competition (S202410364149). We apologize for not being able to cite additional work owing to space limitations. The authors declare no conflicts of interest. The animal care and all experiments were performed in accordance with the guidelines approved by the Ethics Review Committee for Animal Experimentation of Anhui Agricultural University (Permit number: KJLL2025038). All the raw sequencing data and the genome assembly generated for this research are available in the NCBI BioProject database under accession numbers PRJNA1203020 (https://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA1203020) and PRJNA1202998 (https://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA1202998). The data and scripts used are saved in GitHub https://github.com/jxw-123/BSF-UA. Supplementary materials (methods, figures, tables, graphical abstract, slides, videos, Chinese translated version, and update materials) may be found in the online DOI or iMeta Science http://www.imeta.science/imetaomics/. Figure S1: The impact of uric acid on the food reduction rate. Figure S2: Construction and validation of larval systems in different states. Figure S3: Effects of uric acid on microbial diversity and function. Figure S4: Growth and metabolic characteristics of isolated uric acid-degrading strains. Figure S5: Confirmation of gut bacterial colonization and observation of consequent changes in larval uric acid levels. Figure S6: Effects of ygfK knockout in E. coli on intestinal oxidative gene expression in BSF larvae. Figure S7: Effects of E. faecalis AHAU24 on BSF larval growth, conversion rate, and substrate uric acid content in a chicken manure rearing system. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.