Presence and role of viruses in anaerobic digestion of food waste under environmental variability

doi:10.21203/rs.3.rs-1974582/v1

Download PDF

Research Article

Presence and role of viruses in anaerobic digestion of food waste under environmental variability

https://doi.org/10.21203/rs.3.rs-1974582/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 04 Aug, 2023

Read the published version in Microbiome →

You are reading this latest preprint version

Background

The interaction among microorganisms in anaerobic digestion of food waste (ADFW) reactors achieves the degradation of organics and the recycling of energy. As an important component of the microorganisms in ADFW, viruses were rarely explored. Furthermore, little is known about how viruses affect methanogenesis.

Results

Here, we recovered thousands of viral sequences from five full-scale ADFW reactors. Gene-sharing networks indicated that ADFW contained a large number of unexplored anaerobic-specific viruses. Moreover, viral communities in different full-scale reactors exhibited both commonalities and heterogeneities. Lab-scale dynamic analysis in typical scenarios further suggested that viruses have similar kinetic characteristics to their prokaryotic hosts. By associating with the putative hosts, the vast majority of bacteria and archaea phyla can be infected by viruses. Viruses may influence prokaryotic ecological niches and thus methanogenesis by infecting key functional microorganisms such as sulfate-reducing bacteria, syntrophic acetate oxidizing bacteria, and methanogens. Metabolic predictions of viruses revealed that viruses can facilitate the metabolism of sulfur and fatty acids by enhancing host metabolic capacity in key steps and participate in typical methanogenic pathways to promote methane production.

Conclusions

Overall, our results expand on the diversity of viruses in the ADFW systems and point out two ways of viral manipulation of ADFW biochemical processes.

Anaerobic digestion

Food waste

Viruses

Ammonium

Long chain fatty acid

Community

Auxiliary metabolic genes

Anaerobic digestion of food waste (ADFW) can convert organic wastes into methane-rich biogas, known as important renewable energy [1]. Microorganisms, both bacteria and archaea, are considered to be the main participants to catalyze different anaerobic digestion steps. Concerning enormous microbial diversity, viruses were rarely considered as an important role in anaerobic digestion systems even though a large number of viruses-like particles (PLPs) have been found to be produced in methanogenic reactors [2]. Besides, the presence of viruses was confirmed to be the main biological factor in shaping the prokaryotic composition and controlling process performance [3]. Nevertheless, studies on the ethology of viruses and the effect of viruses on methanogenesis in typical scenarios of ADFW systems have not been reported.

Recent developments in metagenomics have contributed to advances in viral ecology. Compared with conventional virus isolation, a large number of viral sequences can be identified in metagenomic data [4]. Scientific researches on ocean [5], peatlands [6], soils [7], and wastewater treatment plants [8] have greatly expanded viral datasets and the cognition of global distribution characteristics through the application of metagenomic techniques. The importance of viruses in global biogeochemistry has been gradually revealed. Viruses play a key role in regulating microflora through lysis or lysogen patterns, thereby adjusting the relative abundance of host communities and indirectly reshaping non-target groups by releasing niches [9]. In addition, they also encode the auxiliary metabolic genes (AMGs) in carbon metabolism [10], nitrogen cycle [11] and sulfur metabolism [12], which significantly affect biogeochemical cycles. As well summarized in the review work of Möller et al [13], biogas production and digestate nutrient availability are closely linked to the conversions of carbon, nitrogen, and sulfur in an anaerobic digester. Therefore, the study of viral communities in various ADFW environments may offer some insights into microbial dynamics and digester performance.

Aiming to identify the diversity of viruses in ADFW and expand the understanding of viruses on the ecological role of methanogens, we restored the existing characteristics of viruses in five full-scale ADFW reactors. To further explore the behavior of viruses in ADFW under different stresses appearing in the industrial application, we performed a continuous cultivation experiment by using lab-scale reactors under ammonium-rich and lipid-rich environments. The kinetics relationship between viruses and prokaryotes was revealed using time-series analysis. By predicting hosts and identifying AMGs, the effect of viruses on the functional microorganisms associated with methanogenesis was studied from the perspective of population and gene. On balance, our research performed a comprehensive analysis of the ecological role of viruses and proposed the potential for viral action on methanogenic efficiency in the ADFW systems.

Sample collection and description of incubation reactors

Five samples were collected from five full-scale continuous stirred tank reactors of the mesophilic food waste treatment plant in Shanghai, China (referred to LG). Details about the operation of the full-scale reactors were described in our previous study of LGf [14]. The other 15 samples were taken from 0^th, 44^th, 92^th, 108^th, and 126^th days from three lab-scale continuous reactors. The three mesophilic incubated ADFW reactors were started in 2 L (actual working volume was 1.8 L) continuously tank reactors with a stirring device at 38.0 ± 0.5 °C (BPC Instruments AB, Sweden). The digestors were started-up with LG slurry as inoculum. The original feedstock and food-waste pulp were collected from LG homogenization tank and stored at 4 ℃. One reactor as the control group was only fed with original feedstock which was labeled “V”. Another reactor supplied by extra ammonium as the ammonium-rich group was labeled as “N”. The third reactor, as the lipid-rich group, supplemented with sodium palmitate in the daily feedstock named as “L”. The cultivation period of three reactors lasted 126 days (see details in Description S1 and Table S1).

Analysis of physicochemical properties

Each slurry sample was performed physicochemical properties measurement, with pH was determined by a benchtop pH meter (PHS-3BW, Bante Instruments). All samples were centrifuged at 15000 rpm for 10 minutes to obtain the supernatant for the determination of chemical oxygen demand (COD), total organic carbon (TOC), total nitrogen (TN), and volatile fatty acids (VFAs). COD was analyzed by using the spectrophotometer (DR3900, HACH, USA) according to the manual of the HACH COD reagent kit (20-1500 mg/L COD, HACH, USA). TOC and TN were determined using a total organic carbon analyzer (TOC-V CPH, Shimadzu, Japan). VFAs were determined using gas chromatography equipped with a flame ionization detector (TRACE 1300, Thermo Fisher Scientific, USA).

DNA extraction and metagenomic sequencing

The DNA of each sample was extracted using PowerSoil^TM DNA isolation kit (MoBio Laboratories Inc, CA) confirmed to the operation manual. Extracted DNA was stored at ‒40℃ for metagenomic sequencing. The details of shotgun metagenomic sequencing were described in our previous study [15]. Ultimately, metagenomic sequencing of each sample yielded an average of 15Gb raw reads.

Metagenomic assembly

The paired-end metagenomic raw reads were trimmed using the Read_QC module within metaWRAP (v1.3.2) [16]. The clean reads of each sample were individually assembled using the ASSEMBLY module within the metaWRAP pipeline (choose the metaSPAdes method) [16]. The final assembled contigs shorter than 1000bp were removed.

Metagenomic binning and taxonomy of prokaryotic

The assembled contigs were binned based on three methods including maxbin2, metabat1, and metabat2, executed in the Binning module. The final bin sets were consolidated predicted bins from three methods using the Bin_refinement module within metaWRAP [16]. The quality of all bins was assessed by CheckM (v1.1.3) [17], including completeness and contamination. In general, bins with low quality that completeness lower than 50% or contamination higher than 10% were discarded. The filtered bins were de-replicated at 95% average nucleotide identity (ANI) to produce metagenome assembled genomes (MAGs) using dRep (v3.2.2) [18]. Taxonomic classification of MAGs was assigned using the GTDB-TK (v1.7.0) [19] based on Genome Taxonomy Database (05-RS95 release). The resulted maximum-likelihood phylogenetic tree was visualized and refined using iTOL (v6.5.8) [20].

Identification of viral contigs and taxonomic assignments

For assembly contigs, three accepted tools were adopted to recover viral contigs, including Virsorter2 [21], DeepVirfinder [22], and CheckV (v0.8.1) [23]. Only contigs longer than 1.5 kb were retained for following criteria: (1) Virsoter2 score ≥ 0.95, (2) DeepVirfinder score ≥ 0.9 and p < 0.05, (3) identified by Virsorter2 and DeepVirfinder score ≥ 0.7 and p < 0.05. Contigs that met any of the above three points were retained. Of these contigs who identified as “Not-determined” by CheckV were removed. The open reading frames (ORFs) of these contigs were called by Prodigal (v2.6.3) [24]. And the protein sequences were annotated to NCBI RefSeq viral protein database using DIAMOND (v2.0.14.152) [25] with the e-value of 1e^-5. Those contigs with no gene hit were further removed. All remaining contigs were grouped into viral operational taxonomic units (vOTUs) if they shared ≥ 95% nucleotide identity using CD-HIT (v4.8.1) [26]. For each vOTU, protein sequences predicted by Prodigal (v2.6.3) [24] were aligned against the NCBI Refseq viral proteins database (https://ftp.ncbi.nlm.nih.gov/refseq/release/viral) using BLASTp (e-value of 1e^-5) [27]. The taxonomic affiliations of each vOTU were predicted by using lowest common ancestor (LCA) algorithm within Taxonkit (v0.10.1) [28] to calculate the lowest common of all hit genes in one vOTU. vOTUs were finally classified using the naming system of the NCBI taxonomy [29]. For vOTUs who had no specific affiliation to any known taxa were classified as “Unknown”.

Virus network construction

Gene-sharing network was used for correlation exploration. All proteins predicted by Prodigal (v2.6.3) [24] were uploaded to vConTACT2 [30] to produce viral clusters (VCs). Diamond was applied for the protein-protein similarity method and protein clusters were generated by using the Markov Cluster Algorithm (MCL). VCs were generated by ClusterONE [31]. The results of the gene-sharing network were visualized in Cytoscape (v3.8.2) [32]. To compare similarity and difference between the data set in our study and other broader habitats, viral sequences from three engineered ecosystems were downloaded from IMG/VR database v3.0: (i) wastewater, (ii) bioreactor, (iii) lab enrichment [33]. For wastewater (n=7,452, n: the number of viral sequences), bioreactor (n=6,513), and lab enrichment (n=1,090), each ecosystem contains both anaerobic (n=5,163, 5,911, 736) and aerobic (n=2,289, 602, 354) subclasses.

Host prediction

Total 169 MAGs were used to predict “in situ” hosts for vOTUs derived from the same sample. The linkages of vOTUs and prokaryotic genomes were performed based on CRISPR spacer matching. Details analysis method consulted Martínez Arbas et al’s research[34], with minor modification. In brief, identification of protospacer-containing contigs workflow was applicated for vOTUs. The final virus-host linkages were defined by the spacer co-occurred with its assigned MAGs and targeted vOTUs at one same sample, which we deemed these interrelations may be more active.

Gene analysis

Gene and protein sequences of each MAG were analyzed using EnrichM (v0.6.4) (https://github.com/geronimp/enrichM) based on Kyoto Encyclopedia of Genes and Genome (KEGG) Orthogroups (KOs) for metabolic reconstruction. For identified viral contigs, CheckV (v0.8.1) [23] was used to remove host contamination. The clean viral contigs were called ORFs sequences by Prodigal (v2.6.3) [24]. The ORFs set was further annotated using EnrichM (v0.6.4) for metabolic pathway analysis.

Relative abundance profiles

Relative abundances of viruses and prokaryotes were expressed in the form of TPM (Transcripts per kilobase of exon model per million mapped reads). To estimate the TPM values of each vOTU and MAG, clean reads from each sample were mapped to a reference database of vOTUs or 169 MAGs using coverM (v0.6.1) (https://github.com/wwood/CoverM) within contig or genome mode. Bwa-mem was chosen as the method for calculating coverage. The low-quality alignments were removed which reads identity ≤ 95% and aligned percent ≤ 75%. The relative abundance of taxa in vOTUs and MAGs was counted by summing up all the relative abundance in the same taxonomy.

Statistical analyses

All statistical analyses were performed in R version 4.0.5. Alpha diversity of viruses and prokaryote communities were calculated using vegan (v2.5-7) package. Principal coordinate analysis (PCoA) was implemented based on Bray–Curtis dissimilarities generated from the relative abundance of viruses and prokaryotes. Analysis of similarity (ANOSIM) was used for verifying different time point samples, which were performed in vegan (v2.5-7) package. Pearson correlations between viruses and their predicted host were calculated in vegan (v2.5-7) by their relative abundance in each sample. Multiple regression matrix (MRM) was performed to test the correlation between microbial communities and abiotic factors in vegan (v2.5-7) and ecodist packages.

Availability of data and materials

All raw reads of metagenomic sequencing in this research have been submitted to NCBI Sequence Read Archive under the project ID PRJNA867664.

Industrial ADFW systems displayed novel and diverse virus pools

The viruses in five industrial ADFW reactors were investigated. After assembly of reads, 5,036 vOTUs were recovered using the identification pipeline with one vOTU greater than 300 kb in length that might be assigned to a huge virus [35]. According to the estimation using CheckV [23], 93 vOTUs with completeness higher than 50% were assessed as complete, high-quality, and medium-quality (Table S2). Exploration of the composition of 5,036 vOTUs found that 46.8% of the viral community could not be attributed to any known family. Siphoviridae (32.7%), Myoviridae (11.9%), and Podoviridae (3.0%) were detected as the dominant classifiable families, accounting for 47.6% of the total composition (Figure S1 and Table S3).

To investigate the relationship between ADFW vOTUs and the NCBI reference viral genomes, a gene-sharing network was constructed by vConTACT2 [30]. Consequently, 1278 viral clusters (VCs) were grouped at approximately the genus level, of which only 14 (1.10%) VCs were shared with the known viral genomes from NCBI Refseq (version 201) (Figure 1a and Table S4). While a large proportion (64.48%) of VCs clustered by ADFW alone, which suggests that most ADFW viruses may reflect habitat specificity. These findings prompt further research to compare the ADFW viruses in broader habitats. Partial viral sequences in IMG/VR database v3.0 [33] were used to explore the co-occurrence between ADFW viruses and specific ecosystems. A total of 15,055 viral sequences were used to execute the construction of the gene-sharing network. Results showed that the vOTUs in the present work were shared among all other ecosystems except for the aerobic subclass of the bioreactor. There were considerable shared clusters between ADFW and the anaerobic subclass compared with the aerobic subclass (Figure S2).

Further analysis of the relevance among five reactors showed that these reactors were distanced from each other, with only one pair of LG1 and LG5 representing a higher similarity (Figure 1b). As shown in the gene-sharing network (Figure 1a), ADFW vOTUs were clustered at 838 VCs, of which only 26 (3.1%) VCs were detected in five reactors and 61 (7.3%) VCs were unique. However, the vast majority of VCs were shared among more than two reactors (Figure 1c). On the whole, the reactors in the same plant have similar feedstock and operating conditions, which the ADFW viral communities showed commonality, but also variability due to the individuality of each reactor.

The performance during the operation of the lab-scale reactors

During the incubation period, the same feedstock was maintained for all three semi-continuous reactors in Stage I to maintain stable operation. As shown in Figure 2, the organic loading rate (OLR) in Stage I has been increased twice from 0.5 to 0.83 g-VS/L·d, at which point there was a downward trend in COD and TOC. Different feeding strategies were introduced until the 47^th day and continued at 0.83 g-VS/L·d until the 92^th day. Stage III and IV were cultured for 16 days with 1.04 and 1.3 g-VS/L·d, respectively. Due to the addition of NH₄Cl in the feedstock, the TAN in N-group gradually increased and was maintained at more than 6000 mg·N/L. Acetate was observed to accumulate in Stages III and IV in N-group. Corresponding, the specific methane production (SMP) in N-group was consistently lower than that in the V-group with the change in feeding strategy. However, the addition of palmitate (L-group) appeared to have a positive effect on methane production. Although the propionate concentration in L-group was lower than that in V-group, the β-oxidation of lipid may drive the formation of acetate (see details of physicochemical parameters in Table S5).

Dynamics of viruses and prokaryotes under ammonium or palmitate stress

Three lab-scale reactors were conducted to further excavate the behavioral characteristics of the viruses in anaerobic environments with high levels of ammonium or palmitate. From the 15 metagenomes, 13,691 vOTUs were eventually recovered (Table S6), with two-fifths of the viral community that could not be attributed to a known annotation at the family level (Table S7). Siphoviridae was the most dominant family in all samples (35.1%), while Myoviridae (14.9%) and Podoviridae (2.8%) had a high relative abundance (Figure S3), which is similar to that of full-scale reactors. High or medium-quality of binned genomes were demonstrated by the 169 MAGs for accessing the characteristics of prokaryotes (Table S8).

Regarding the dependence of viral populations on different factors, the Pearson correlation was used to reveal the relationships between viral taxa and abiotic factors. Results showed that the majority of taxa can be affected by different physicochemical properties (Figure 3a and Table S9). Significant correlations were observed among TAN and five viral families, especially Tectiviridae and Salasmaviridae which were correlated to COD. As of 126^th day, the relative abundance of Tectiviridae increased up to 2.62-fold (0.60% to 1.57%) and that of Salasmaviridae increased 7.45-fold (0.04% to 0.35%) in the N-group, which indicated that the members of Tectiviridae and Salasmaviridae were tolerant at high ammonium conditions. In addition, four families were significantly correlated with SMP, of which Plectroviridae and Smacoviridae showed a negative correlation. The dual character of viruses illustrated that some taxa are positive for methanogenesis while some are negative. Mantel test was performed to evaluate the driving factors of viral and prokaryote taxonomic dissimilarities (Figure 3b and Table S10). SMP was the stronger correlate of both viral and prokaryote taxonomic dissimilarities, indicating that viruses, like prokaryotes, tend to evolve in the direction of methane production.

To verify the virus-prokaryote dynamics, we examined the correlations between viral and prokaryote community diversities. Specifically, the Shannon diversity (R=0.95, p=2.4×10^-8, Figure 4a) and Simpson diversity (R=0.72, p=0.0023, Figure S4) of prokaryotes and viruses were observed significantly correlated. In addition, the correlation was exhibited in the observed species index of prokaryotes and viruses (R=0.83, p=1.1×10^-4, Figure 4b) (details diversity index see Table S11). PCoA analysis was applied to reveal the clustering patterns of the viral community (ANOSIM, R=0.5674, p=0.002, Figure 4c) and prokaryotic community (ANOSIM, R=0.7422, p=0.001, Figure 4d) in time series, respectively. Results showed that the microbial community structure was significantly different from the original community structure during the incubating process, especially L126. For the different feedstock conditions, the microbial community of the N-group converged after the addition of NH₄Cl in feedstock (92^th, 108^th, and 126^th days). The microbial community of the V-group showed convergence in one direction with increasing organic loading and gathered when the OLR reached more than 1 g-VS/L·d (108^th and 126^th days) (Figure 4d). Meanwhile, compared to V-group, the microbial community of the L-group showed significant differences in the initial stage of palmitate addition, until the 126^th day, it showed a clear distance from all other samples. On the other hand, comparing the communities of viruses and prokaryotes, congruous dynamic changes have been observed during the incubating process. Furthermore, the results of the multiple regression matrix (MRM) showed that viruses explained 85.4% of prokaryotic community composition, much higher than the power of abiotic factors (21.0% for pH, TOC, TAN, COD, SMP) (Table S12).

Consequently, viruses should be participants in methanogenesis, which may regulate functional microorganisms through some mechanisms, enabling synergistic development with prokaryotes under different conditions.

Overview of prokaryotic communities associated with methanogenesis in lab-scale reactors

A total of 169 MAGs containing 159 bacteria and 10 archaea MAGs were used to evaluate the prokaryotic succession patterns under different feeding strategies. To assess microbial community structure, we calculated the relative abundance of prokaryotes taxa at the phylum level based on GTDB-TK annotation (Table S8). Within the bacteria, the dominant members were Firmicutes (59.9%), Bacteroidota (8.2%), Chloroflexota (7.8%), and Cloacimonadota (6.1%), accounting for 82%. For archaeal lineages, members were mainly affiliated with Methanobacterita (5.0%) and Halobacteriota (4.8%) (Figure S5). Evidence proved that some syntrophic acetate oxidizing bacteria (SAOB) can achieve acetate oxidation using the oxidative Wood-Ljungdahl (WL) pathway [36]. On this basis, the interesting findings were observed that the functional genes encoding a complete or nearly-complete oxidative WL pathway can be reconstructed in four high-quality MAGs (L92bin.31, L44bin.114, N92bin.52, N126bin.29) (Figure S6 and Table S13). All four members were classified as Firmicutes, including two known SAOB Syntrophaceticus schinkii (N92bin.52) [37] and Tepidanaerobacter acetatoxydans (L44bin.114) [38], while the other two MAGs (L92bin.31, N126bin.29) were assigned to the DTU022 order.

Regarding the impact of palmitate on methanogenesis, the relative abundance of archaea in the L-group decreased compared with that in V-group (Figure 5 and Table S8). However, the proportion of Methanoculleus members (L126bin.71, V108bin.21, N0bin.115, and L0bin.128) in the total archaea composition increased from 33.4% (0^th day) to 97.1% (126^th day), while a clear shift of L92bin.31 was observed with the relative abundance range from 2.21% (0^th day) to 5.82% (126^th day), eventually accounting for 74.4% of all SAOB abundance. Other archaea members seemed not to be actively working under the pressure of palmitate, as the declining abundance of Methanobacteriaceae and the consistently low abundance of Methanomethylophilaceae. These results denoted that Methanoculleus dominated hydrogenotrophic methanogen became the dominant pathway in the late stages of palmitate stress. In addition, two MAGs (L126bin.41 and L126bin.10) assigned to the Desulfobacterota phylum were considered to be typical sulfate-reducing bacteria (SRB) which have the ability of sulfate reduction coupling to convert long-chain fatty acids (LCFAs) [39,40]. Both of them experienced a significant increase of more than 3-fold in Stage IV. Therefore, the symbiotic flora of hydrogenotrophic methanogens and SRB shared the responsibility for the metabolism of substances under palmitate stress.

We further looked into the influence of ammonium on the composition of the prokaryotic communities (Figure 5 and Table S8). The relative abundance of DTU022 members was more than 3.5-fold that of the V-group in Stage IV, despite the other two SAOB did not show activity at high ammonium concentrations. At present, recent research has confirmed the positive role of SAOB in high ammonium stress [41]. Besides, a similar trend was observed in L0bin.128, which was assigned to Methanimicrococcus sp012518265. As in the present study, L0bin.128 was not detected in the N-group and V-group at the initial stage but started to grow at Stage II. Moreover, the relative abundances of L0bin.128 in the N-group were 6.12, 2.08, and 3.71-fold than that in V-group at 92^th, 108^th, and 126^th days, demonstrating the higher growth rate in the N-group. The genera of Methanimicrococcus have been reported to perform methanogenetic pathways using methanol or methylamine [42]. The addition of ammonium facilitated the methanogenic pathway of Methanimicrococcus sp012518265 using methylamine. The results showed the contribution of SAOB and methylotrophic methanogens to methane production under ammonium stress.

The screening of the broader prokaryotic communities was oriented to the bacteria and archaea that affected methanogenesis. In general, SAOB and SRB in the bacterial domain and methanogen in the archaeal domain were considered to be the significant prokaryotic players during the cultivation period.

Viruses infect functional microorganisms related to methanogenesis

CRISPR-Cas spacers matching was used to perform host prediction for 13,691 vOTUs. We retained 1,631 spacers which assigned at least one MAGs and targeted at least one vOTUs in the same sample. Based on this, we recovered 1,182 vOTUs (8.63%) with their putative host, comprising 0.95% - 28.26% of vOTUs in each sample (Table S14). These predicted hosts spanned a wide range of bacteria and archaea, with 14 bacterial and three archaeal phyla. Within the bacterial domain, 858 vOTUs were linked to Firmicutes, which is higher than that of other bacterial phyla. Besides, Bacteroidota (n=215, n: the number of vOTUs linked to phyla) and DTU030 (n=113) were also the more predicted hosts (Figure 6). Among archaea, a considerable proportion of vOTUs infected Halobacteriota (n=130), and Methanobacteriota (n=29) (Figure 6). Among these, the vast majority of phylum have not been reported on virus infection. Out of the 1,182 vOTUs, most of them were assigned to a narrow host range. However, 321 vOTUs exhibited the potential to infect more than one host. Especially, observed that 68 vOTUs can infect bacteria and archaea simultaneously, suggesting the existence of viruses infected different domains (Figure S7).

To investigate the relationship between viruses and key functional microorganisms. There were 10 and 20 vOTUs identified to infest Desulfomicrobium and Desulfobulbus, respectively. Amazingly, the vast majority of vOTUs (90%, 75%) associated with the two hosts showed a significant positive correlation (p < 0.001, Figure S8a), which indicated that viruses may indirectly regulate LCFAs metabolism by affecting SRB. Further exploring the characteristics of the infested SAOB viruses, we found that each SAOB can be infected by several vOTUs, except N126bin.29 (Figure S8b). Different from the infection characteristics of SRB, only a small proportion of vOTUs showed a significant positive correlation with their SAOB hosts. In fact, out of 35 vOTUs, 8 vOTUs displayed a significant positive correlation (p < 0.01) for L92bin.31, the proportion that was 39% for L44bin.114, yet 0 for N92bin.52. However, if the relative abundance of all viruses infecting the same host was added as a whole to calculate the correlation with their host, the host-virus pair interactions demonstrated stronger (R=0.66, p=0.0073 for L92bin.31, R=0.71, p=0.0029 for L44bin.114, R=0.56, p=0.029 for N92bin.52, Figure S9). Results showed that the viruses infecting SAOB have some more complicated internal mechanisms, which jointly regulate the population structure of SAOB.

Regarding the archaea (Figure S8c), no corresponding vOTU was matched to Methanimicrococcus sp012518265 and one vOTU was associated with the N0bin.115 which belonged to Methanomethylophilaceae. Two MAGs (N92bin.39, V92bin.105) assigned to Methanobacteriaceae were matched to 28 and 1 vOTU respectively. In comparison, Methanoculleus (L126bin.71, V108bin.21, N0bin.115, L0bin.128) were linked to more vOTUs (25, 41, 30, 34), which suggested that Methanoculleus can be infected by multiple viruses. Methanoculleus is a typical hydrogenotrophic methanogen that has strong tolerance under adverse environments [41]. Multiple viral infestations may become a booster of Methanoculleus tolerance in adverse environments. According to the statistics of correlations between archaeal viruses and hosts, most matches were not significant, which may be due to the slow growth rate of methanogenic archaea and susceptibility to external environmental factors. However, there was a particular finding that a large proportion of viruses infected with N92bin.39 (82.14%) and N0bin.115 (53.3%) showed the opposite trend to their hosts (r < 0). Therefore, in the archaeal domain, a fraction of viruses was negative towards their hosts, which may affect the competition of methanogens for dominant flora.

Viruses encoded extensive auxiliary metabolic genes

Having demonstrated the profile of viral effects on methanogenic-related functional microorganisms, we next attempted to understand the function of viruses on ADFW systems about the gene profiles. Viral ORFs annotated using EnrichM were considered as the broad AMGs that refer to all metabolic genes [43]. Overall, ADFW viruses tended to encode AMGs for substance metabolism including carbohydrate, energy, lipid, nucleotide, amino acid, etc. (Table S15), demonstrating the high diversity of AMGs in the ADFW systems.

The degradation of the lipid hydrolysis product LCFAs is the rate-limiting step in the ADFW systems, which is usually accomplished by β-oxidation [44]. Long-chain acyl-CoA synthetase (ACSL) was found to be encoded by 3 vOTUs that activate LCFAs to form acyl-CoAs. As the key intermediates in β-oxidation, acyl-CoAs play an important role in downstream metabolism [45]. Consequently, results revealed the crucial role of the virus-encoded AMGs in the key steps of LCFAs degradation. Several AMGs were found to be associated with sulfur metabolism. We identified 13 vOTUs encoded cysH, which has been found in viral sequences from various ecosystems to facilitate host in assimilatory sulfate reduction (ASR) [46–48]. In addition, sulfate adenylyltransferase (sat) within AMGs was observed to be responsible for adenosine 5-phosphosulfate (APS) reduction, which is the final step of sulfide oxidation in dissimilatory sulfate reduction (DSR).

Broadly speaking, 40 AMGs were potentially associated with methane metabolism pathways, of which 37 were labeled as carbon fixation pathways in KEGG. To further investigate the impacts of virus-encoded AMGs on methane production in the ADFW systems, we focused on two major pathways for the conversion of acetate to methane, acetoclastic methanogenesis (AM), and syntrophic acetate oxidation coupled with hydrogenotrophic methanogenesis (SAO-HM) (Figure 7a,b and Table S16). For AM pathway, viral members encoded a near-complete list of genes joined in the conversion of acetate to methane, except the gene encoding phosphate acetyltransferase (Pta) and ACDS that catalyzes the activation of acetyl-P to 5,10-Methyl-THMPT. 5,10-Methyl-THMPT is a key precursor for methanogenesis in both AM and SAO-HM pathways. Although the viruses missed a complete pathway to convert acetate into 5,10-Methyl-THMPT in the AM pathway, acetate transformed to acetyl-P by acetate kinase (AckA) was found in five vOTUs (Figure 7c), revealing the possibility that viruses contributed to the conversion of acetate to important intermediates. On the other hand, HM pathway produces methane with H₂ and CO₂ as substrates, with two steps activated in the transformation of CO₂ to 5-Methyl-THMPT. Two vOTUs could catalyze the first step from CO₂ to formyl-MFR by formylmethanofuran dehydrogenase (fwdC) and activate 5,10-Methylene-THMPT to 5-Methyl-THMPT through 5,10-methylenetetrahydromethanopterin reductase (mer) (Figure 7c). For the shared steps of both methanogenic pathways, viruses encoded the corresponding enzymes to accomplish the reaction from 5,10-Methyl-THMPT to methane. The oxidative WL pathway is generally considered to be constructed by SAOB, however, we found that viruses could assist in most steps of WL pathway. Viruses encoding enzymes were involved in the conversion of acetate to acetyl-CoA. In addition, there were corresponding enzymes to participate in the conversion of CH₃-THF to formate, which was the main step in the WL pathway.

Viruses are being taken into account in the regulation of community composition and elemental cycling in ecosystems [43]. At present, few studies have been conducted to characterize the presence of viruses in AD systems [1,2,49,50]. The presence of abundant microbial populations in ADFW enables complex material metabolism, but the role of viruses in ADFW is still poorly understood. Based on the excavation of metagenomes, our study confirmed that there were a large number of viral sequences in full-scale ADFW (Figure 1a and Table S4). However, the ADFW viruses were mostly not clustered to NCBI viral Refseq, possibly due to the majority of viruses have not been able to obtain the complete genomes (11,607 complete viral genomes in NCBI viral Refseq as of June 2022, https://www.ncbi.nlm.nih.gov/genomes/GenomesGroup.cgi?taxid=10239). In addition, the researches on viruses focus more on natural ecosystems [5,7,51] and human health [52], resulting in a certain bias in the reference database. Unlike natural environments, engineering ecosystems are bio-enhanced facilities with artificial interventions, whose microbial populations exhibit specificity depending on the needs of the facilities. Interestingly, the ADFW viruses were associated with anaerobic environments of engineered systems in the IMG/VR database, indicating the presence of anaerobic-specific viruses in ADFW and the possibility of association with anaerobic processes.

We performed the first attempt to analyze virus-prokaryotic interactions in time series in typical scenarios at the lab-scale reactors. The frequent fluctuations in viral and prokaryotic diversity were observed throughout the cultivation period, which indicated a constant evolution of microbial composition in response to changes in the external environment (Table S11). Meanwhile, the viral community followed the same evolutionary pattern as the prokaryotic community, and this evolution proceeded in a direction that favors methane production (Figure 3b and Table S10). At present, it is not clear whether this dynamic pattern is driven by the external environment or the influence of predator-prey dynamics. Long-standing studies have revealed the adaptive strategies of functional microorganisms under ammonium [53] and palmitate stress [54]. However, the calculation of viral contribution explained 85.4% of the prokaryotic composition (Table S12) which was much higher than the influence of abiotic factors. Therefore, the role of viruses in anaerobic digestion may have been underestimated.

CRISPR-Cas spacer matching announced that viruses can infect the majority of phylum (Figure S7), whose functions involved hydrolytic acidification and methane production in the ADFW systems. The range of virus infection may be narrow or extensive, and even the possibility of cross-domain infection, which were widely occurring in ecosystems [55]. The fact that only a fraction of bacterial and archaeal genomes encodes the CRISPR system [56], making it difficult to match all reliable hosts. However, the CRISPR-based approach to identifying hosts remains genetically reliable at present. Although future validation is needed in combination with other host prediction methods, such as experiments or Hi-C metagenomics [55]. Overall, the present study revealed a comprehensive view of virus-host interactions in the ADFW systems and attempted to interpret the infection mode of functional microorganisms affecting methanogenesis.

The corrosive and high lipid nature of food waste makes ADFW systems vulnerable to the accumulation of ammonium [41] and LCFAs [57] in actual biogas plants. Under the ammonium and LCFAs stress, functional microorganisms exhibited corresponding dynamics (Figure 4). Ultimately SRB, SAOB, and methanogenic archaea were screened for their active role to investigate their infection patterns (Figure 5). The current life strategies of viral contain kill-the-winner [58] and piggyback-the-winner models [59]. Viruses can regulate the dominant population of the community through predation or inhibit lysis to maintain high host densities [58,59]. Both of these patterns may be related to environmental factors and virus types [60]. In our study, viruses infected with SRB maintained a significant positive correlation with their host abundance while half of the viruses infected with certain methanogenic archaea showed a negative correlation, which indicated that possibly both strategies are present in the ADFW systems (Figure S8). The tight ecological coupling caused by the competition between prokaryotes and the predation of viruses form the microbial dynamics within the ADFW systems. In addition, highly resistant prokaryotes have a more stable intracellular environment under adverse conditions, providing more suitable conditions for viruses to survive and reproduce [61]. This was reflected in SAOB and Methanoculleus, which both show positive effects in typical inhibition scenarios of ADFW [41]. This trend provides a basis for future research on enhancing the adaptability of microorganisms related to SAO-HM pathway through viral action in adverse environments.

Although the exact impact of viruses on the structure of microbial communities remains to be studied, viruses can affect important metabolic functions of their hosts by carrying AMGs [62]. In our study, viruses encoded enzymes for the degradation of LCFAs to key intermediates, which may facilitate the metabolism of lipid in ADFW. Virus-mediated sulfur auxiliary metabolism widely exists in various ecosystems, which contributes to the global biogeochemical cycle [12]. ASR and DSR have been deemed as the main sulfate-reducing pathways, which may relate to the carbon cycle driven by SRB [63]. The involvement of viruses in the ASR and DSR pathway was the first discovery in the ADFW systems, revealing the possibility that viruses participate in carbon and sulfur cycles. It has been suggested that the main pathways for the conversion of acetate to methane include the AM pathway and the SAO-HM pathway. Interestingly, we found that viruses encoded part of the enzymes of the methanogenic pathway (Figure 7). Hence, the importance of viruses in the ADFW systems may have been underestimated and their impact on methanogenesis deserves further study.

Owing to the complex microbial ecological patterns in the ADFW systems, the role of viruses in affecting microbial population and ecology has not been thoroughly explored. In this study, in-depth mining of metagenomic data has successfully revealed the existence of novel and diverse viral populations in practical engineering scenarios. The results have greatly expanded the diversity of viral populations in the ADFW systems. While nearly half of viruses appear to be unknown taxa, gene-share networks suggested anaerobic-specific of ADFW viruses. ADFW systems were proved to be the microbial pool whose community structure may influence by the feed characteristics and physicochemical parameters. Evidence from dynamic analysis in multivariate scenarios at the lab scale proved that viruses and prokaryotes have similar kinetic characteristics. Many of the putative hosts for ADFW systems belonged to key functional microorganisms, whose ecological niche can be regulated through viral infection behavior. In addition, viruses-encoded AMGs, including carbon, sulfur, and fatty acid metabolism, may enhance the metabolism of substances in the prokaryotic hosts during the infection period and thus promote methanogenesis. In conclusion, our results highlight the role of viruses for ecology in the ADFW systems and provide support for improving the efficiency of material transformation in engineering applications.

ADFW: Anaerobic digestion of food waste

AMGs: Auxiliary metabolic genes

COD: Chemical oxygen demand

TOC: Total organic carbon

TN: Total nitrogen

VFAs: Volatile fatty acids

MAGs: Metagenome assembled genomes

ORFs: Open reading frames

vOTUs: Viral operational taxonomic units

LCA: Lowest common ancestor

VCs: Viral clusters

TPM: Transcripts per kilobase of exon model per million mapped reads

PCoA: Principal coordinate analysis

MRM: Multiple regression matrix

OLR: Organic loading rate

SMP: Specific methane production

LCFAs: Long-chain fatty acids

WL: Wood-Ljungdahl

ASR: Assimilatory sulfate reduction

DSR: Dissimilatory sulfate reduction

SRB: Sulfate-reducing bacteria

SAOB: Syntrophic acetate oxidizing bacteria

AM: Acetoclastic methanogenesis

SAO-HM: Syntrophic acetate oxidation coupled with hydrogenotrophic methanogenesis

Acknowledgments

We thanks Paul Wilmes and Susana Martínez Arbas from Luxembourg Centre for Systems Biomedicine, University of Luxembourg for the assistance in CRISPR spacers matching analysis.

Funding

The study was financially supported by National Key R&D Program of China (grant number: 2018YFD1100600) and National Natural Science Foundation of China (52000144).

Author information

Authors and Affiliations

Institute of Waste Treatment and Reclamation, Tongji University, Shanghai, 200092, China

Lu Fan, Wei Peng, Haowen Duan, Fan Lü, Hua Zhang & Pinjing He

Shanghai Institute of Pollution Control and Ecological Security, Shanghai, 200092, China

Wei Peng, Fan Lü, Hua Zhang & Pinjing He

Shanghai Engineering Research Center of Multi-Source Solid Wastes Co-processing and Energy Utilization, Shanghai, 200092, China

Wei Peng, Fan Lü, Hua Zhang & Pinjing He

Contributions

WP designed the experiments and FL provided general concept for experiments. LF and WP conducted all the experiments and collected data. LF analyzed the data and wrote the initial manuscript, while WP, HD, FL, HZ, and PH contributed in manuscript revision.

Corresponding authors

Correspondence to Pinjing He or Wei Peng.

Ethics declarations

Ethics approval and consent to participate

Not applicable

Consent for publication

Not applicable

Competing interests

The authors declare that they have no competing interests.

Heyer R, Schallert K, Siewert C, Kohrs F, Greve J, Maus I, et al. Metaproteome analysis reveals that syntrophy, competition, and phage-host interaction shape microbial communities in biogas plants. Microbiome. 2019,7:69.
Park M-O, Ikenaga H, Watanabe K. Phage diversity in a methanogenic digester. Microbial Ecology. 2007,53:98–103.
Zhang J, Gao Q, Zhang Q, Wang T, Yue H, Wu L, et al. Bacteriophage–prokaryote dynamics and interaction within anaerobic digestion processes across time and space. Microbiome. 2017,5:57.
Simmonds P, Adams MJ, Benkő M, Breitbart M, Brister JR, Carstens EB, et al. Virus taxonomy in the age of metagenomics. Nature Reviews Microbiology. 2017,15:161–8.
Gregory AC, Zayed AA, Conceição-Neto N, Temperton B, Bolduc B, Alberti A, et al. Marine DNA viral macro- and microdiversity from pole to pole. Cell. 2019,177:1109-1123.e14.
ter Horst AM, Santos-Medellín C, Sorensen JW, Zinke LA, Wilson RM, Johnston ER, et al. Minnesota peat viromes reveal terrestrial and aquatic niche partitioning for local and global viral populations. Microbiome. 2021,9:233.
Jin M, Guo X, Zhang R, Qu W, Gao B, Zeng R. Diversities and potential biogeochemical impacts of mangrove soil viruses. Microbiome. 2019,7:58.
Tamaki H, Zhang R, Angly FE, Nakamura S, Hong P-Y, Yasunaga T, et al. Metagenomic analysis of DNA viruses in a wastewater treatment plant in tropical climate. Environmental Microbiology. 2012,14:441–52.
Liu N, Lewis C, Zheng W, Fu ZQ. Phage cocktail therapy: multiple ways to suppress pathogenicity. Trends in Plant Science. 2020,25:315–7.
Pan D, Watson R, Wang D, Tan ZH, Snow DD, Weber KA. Correlation between viral production and carbon mineralization under nitrate-reducing conditions in aquifer sediment. The ISME Journal. 2014,8:1691–703.
Gazitúa MC, Vik DR, Roux S, Gregory AC, Bolduc B, Widner B, et al. Potential virus-mediated nitrogen cycling in oxygen-depleted oceanic waters. The ISME Journal. 2021,15:981–98.
Kieft K, Zhou Z, Anderson RE, Buchan A, Campbell BJ, Hallam SJ, et al. Ecology of inorganic sulfur auxiliary metabolism in widespread bacteriophages. Nature Communications. 2021,12:3503.
Möller K, Müller T. Effects of anaerobic digestion on digestate nutrient availability and crop growth: a review. Engineering in Life Sciences. 2012,12:242–57.
Li C, He P, Hao L, Lü F, Shao L, Zhang H. Diverse acetate-oxidizing syntrophs contributing to biogas production from food waste in full-scale anaerobic digesters in China. Renewable Energy. 2022,193:240–50.
Duan H, He P, Shao L, Lü F. Functional genome-centric view of the CO-driven anaerobic microbiome. The ISME Journal. 2021,15:2906–19.
Uritskiy G v, DiRuggiero J, Taylor J. MetaWRAP—a flexible pipeline for genome-resolved metagenomic data analysis. Microbiome. 2018,6:158.
Parks DH, Imelfort M, Skennerton CT, Hugenholtz P, Tyson GW. CheckM: Assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes. Genome Research. 2015,25:1043–55.
Olm MR, Brown CT, Brooks B, Banfield JF. dRep: a tool for fast and accurate genomic comparisons that enables improved genome recovery from metagenomes through de-replication. The ISME Journal. 2017,11:2864–8.
Chaumeil P-A, Mussig AJ, Hugenholtz P, Parks DH. GTDB-Tk: a toolkit to classify genomes with the genome taxonomy database. Bioinformatics. 2019,36:1925–7.
Letunic I, Bork P. Interactive Tree of life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Research. 2021,49:W293–6.
Guo J, Bolduc B, Zayed AA, Varsani A, Dominguez-Huerta G, Delmont TO, et al. VirSorter2: a multi-classifier, expert-guided approach to detect diverse DNA and RNA viruses. Microbiome. 2021,9:37.
Ren J, Song K, Deng C, Ahlgren NA, Fuhrman JA, Li Y, et al. Identifying viruses from metagenomic data using deep learning. Quantitative Biology. 2020,8:64–77.
Nayfach S, Camargo AP, Schulz F, Eloe-Fadrosh E, Roux S, Kyrpides NC. CheckV assesses the quality and completeness of metagenome-assembled viral genomes. Nature Biotechnology. 2021,39:578–85.
Hyatt D, Chen G-L, LoCascio PF, Land ML, Larimer FW, Hauser LJ. Prodigal: prokaryotic gene recognition and translation initiation site identification. Bioinformatics. 2010,11:119.
Buchfink B, Xie C, Huson DH. Fast and sensitive protein alignment using DIAMOND. Nature Methods. 2015,12:59–60.
Fu L, Niu B, Zhu Z, Wu S, Li W. CD-HIT: accelerated for clustering the next-generation sequencing data. Bioinformatics. 2012,28:3150–2.
Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, et al. BLAST+: architecture and applications. Bioinformatics. 2009,10:421.
Shen W, Ren H. TaxonKit: A practical and efficient NCBI taxonomy toolkit. Journal of Genetics and Genomics. 2021,48:844–50.
Federhen S. The NCBI taxonomy database. Nucleic Acids Research. 2012,40:D136–43.
bin Jang H, Bolduc B, Zablocki O, Kuhn JH, Roux S, Adriaenssens EM, et al. Taxonomic assignment of uncultivated prokaryotic virus genomes is enabled by gene-sharing networks. Nature Biotechnology. 2019,37:632–9.
Nepusz T, Yu H, Paccanaro A. Detecting overlapping protein complexes in protein-protein interaction networks. Nature Methods. 2012,9:471–2.
Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003,13:2498–504.
Roux S, Páez-Espino D, Chen I-MA, Palaniappan K, Ratner A, Chu K, et al. IMG/VR v3: an integrated ecological and evolutionary framework for interrogating genomes of uncultivated viruses. Nucleic Acids Res. 2021,49:D764–75.
Martínez Arbas S, Narayanasamy S, Herold M, Lebrun LA, Hoopmann MR, Li S, et al. Roles of bacteriophages, plasmids and CRISPR immunity in microbial community dynamics revealed using time-series integrated meta-omics. Nature Microbiology. 2021,6:123–35.
Al-Shayeb B, Sachdeva R, Chen L-X, Ward F, Munk P, Devoto A, et al. Clades of huge phages from across Earth’s ecosystems. Nature. 2020,578:425–31.
Dyksma S, Jansen L, Gallert C. Syntrophic acetate oxidation replaces acetoclastic methanogenesis during thermophilic digestion of biowaste. Microbiome. 2020,8:105.
Manzoor S, Müller B, Niazi A, Schnürer A, Bongcam-Rudloff E. Working draft genome sequence of the mesophilic acetate oxidizing bacterium Syntrophaceticus schinkii strain Sp3. Standards in Genomic Sciences. 2015,10:99.
Westerholm M, Roos S, Schnürer A. Tepidanaerobacter acetatoxydans sp. nov., an anaerobic, syntrophic acetate-oxidizing bacterium isolated from two ammonium-enriched mesophilic methanogenic processes. Systematic and Applied Microbiology. 2011,34:260–6.
Sousa DZ, Alves JI, Alves MM, Smidt H, Stams AJM. Effect of sulfate on methanogenic communities that degrade unsaturated and saturated long-chain fatty acids (LCFA). Environmental Microbiology. 2009,11:68–80.
Alves JI, Salvador AF, Castro AR, Zheng Y, Nijsse B, Atashgahi S, et al. Long-chain fatty acids degradation by Desulfomonile species and proposal of “candidatus desulfomonile palmitatoxidans” Frontiers in Microbiology. 2020,11.
Hao L, Fan L, Chapleur O, Guenne A, Bize A, Bureau C, et al. Gradual development of ammonia-induced syntrophic acetate-oxidizing activities under mesophilic and thermophilic conditions quantitatively tracked using multiple isotopic approaches. Water Research. 2021,204:117586.
Thomas CM, Taib N, Gribaldo S, Borrel G. Comparative genomic analysis of Methanimicrococcus blatticola provides insights into host adaptation in archaea and the evolution of methanogenesis. ISME Communications. 2021,1:47.
Breitbart M, Bonnain C, Malki K, Sawaya NA. Phage puppet masters of the marine microbial realm. Nature Microbiology. 2018,3:754–66.
Usman M, Zhao S, Jeon B-H, Salama E-S, Li X. Microbial β-oxidation of synthetic long-chain fatty acids to improve lipid biomethanation. Water Research. 2022,213:118164.
Mashek DG, Li LO, Coleman RA. Long-chain acyl-CoA synthetases and fatty acid channeling. Future Lipidol. 2007,2:465–76.
Shi L-D, Dong X, Liu Z, Yang Y, Lin J-G, Li M, et al. A mixed blessing of viruses in wastewater treatment plants. Water Research. 2022,215:118237.
Gao S-M, Schippers A, Chen N, Yuan Y, Zhang M-M, Li Q, et al. Depth-related variability in viral communities in highly stratified sulfidic mine tailings. Microbiome. 2020,8:89.
Li Z, Pan D, Wei G, Pi W, Zhang C, Wang JH, et al. Deep sea sediments associated with cold seeps are a subsurface reservoir of viral diversity. ISME Journal. 2021,
Rossi A, Morlino MS, Gaspari M, Basile A, Kougias P, Treu L, et al. Analysis of the anaerobic digestion metagenome under environmental stresses stimulating prophage induction. Microbiome. 2022,10:125.
Calusinska M, Marynowska M, Goux X, Lentzen E, Delfosse P. Analysis of dsDNA and RNA viromes in methanogenic digesters reveals novel viral genetic diversity. Environmental Microbiology. 2016,18:1162–75.
Chen L-X, Méheust R, Crits-Christoph A, McMahon KD, Nelson TC, Slater GF, et al. Large freshwater phages with the potential to augment aerobic methane oxidation. Nature Microbiology. 2020,5:1504–15.
Nayfach S, Páez-Espino D, Call L, Low SJ, Sberro H, Ivanova NN, et al. Metagenomic compendium of 189,680 DNA viruses from the human gut microbiome. Nature Microbiology. 2021,6:960–70.
Tian H, Treu L, Konstantopoulos K, Fotidis IA, Angelidaki I. 16s rRNA gene sequencing and radioisotopic analysis reveal the composition of ammonia acclimatized methanogenic consortia. Bioresource Technology. 2019,272:54–62.
Regueiro L, Carballa M, Lema JM. Microbiome response to controlled shifts in ammonium and LCFA levels in co-digestion systems. Journal of Biotechnology. 2016,220:35–44.
Chen Y, Wang Y, Paez-Espino D, Polz MF, Zhang T. Prokaryotic viruses impact functional microorganisms in nutrient removal and carbon cycle in wastewater treatment plants. Nature Communications. 2021,12:5398.
Hille F, Richter H, Wong SP, Bratovič M, Ressel S, Charpentier E. The biology of CRISPR-Cas: backward and forward. Cell. 2018,172:1239–59.
Liu Y, He P, Duan H, Shao L, Lü F. Low calcium dosage favors methanation of long-chain fatty acids. Applied Energy. 2021,285:116421.
Thingstad TF, Våge S, Storesund JE, Sandaa R-A, Giske J. A theoretical analysis of how strain-specific viruses can control microbial species diversity. Proceedings of the National Academy of Sciences. 2014,111:7813–8.
Knowles B, Silveira CB, Bailey BA, Barott K, Cantu VA, Cobián-Güemes AG, et al. Lytic to temperate switching of viral communities. Nature. 2016,531:466–70.
Brum JR, Hurwitz BL, Schofield O, Ducklow HW, Sullivan MB. Seasonal time bombs: dominant temperate viruses affect Southern Ocean microbial dynamics. The ISME Journal. 2016,10:437–49.
Shapiro OH, Kushmaro A, Brenner A. Bacteriophage predation regulates microbial abundance and diversity in a full-scale bioreactor treating industrial wastewater. The ISME Journal. 2010,4:327–36.
Huang D, Yu P, Ye M, Schwarz C, Jiang X, Alvarez PJJ. Enhanced mutualistic symbiosis between soil phages and bacteria with elevated chromium-induced environmental stress. Microbiome. 2021,9:150.
Zhou L, Ou P, Zhao B, Zhang W, Yu K, Xie K, et al. Assimilatory and dissimilatory sulfate reduction in the bacterial diversity of biofoulant from a full-scale biofilm-membrane bioreactor for textile wastewater treatment. Science of The Total Environment. 2021,772:145464.

No competing interests reported.

Download PDF

Journal Publication

published 04 Aug, 2023

Read the published version in Microbiome →

Editorial decision: Major revision
19 Mar, 2023
Reviews received at journal
16 Mar, 2023
Reviews received at journal
09 Oct, 2022
Reviewers agreed at journal
30 Sep, 2022
Reviewers agreed at journal
29 Sep, 2022
Reviewers invited by journal
22 Aug, 2022
Editor assigned by journal
21 Aug, 2022
Submission checks completed at journal
18 Aug, 2022
First submitted to journal
18 Aug, 2022

You are reading this latest preprint version

Presence and role of viruses in anaerobic digestion of food waste under environmental variability

Status:

Journal Publication

Version 1

Abstract

Background

Results

Conclusions

Figures

Background

Methods

Sample collection and description of incubation reactors

Analysis of physicochemical properties

DNA extraction and metagenomic sequencing

Metagenomic assembly

Metagenomic binning and taxonomy of prokaryotic

Identification of viral contigs and taxonomic assignments

Virus network construction

Host prediction

Gene analysis

Relative abundance profiles

Statistical analyses

Availability of data and materials

Results

Industrial ADFW systems displayed novel and diverse virus pools

The performance during the operation of the lab-scale reactors

Dynamics of viruses and prokaryotes under ammonium or palmitate stress

Overview of prokaryotic communities associated with methanogenesis in lab-scale reactors

Viruses infect functional microorganisms related to methanogenesis

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1