Metagenomic Insights into Candidatus Scalindua in a Long-term Cultivated Marine Anammox Consortium: The Important Role of Tetrahydrofolate-mediated Carbon Fixation

Abstract

Marine anammox bacteria have been an exciting research area in recent years due to their high effectiveness in treating ammonia-containing saline wastewater. However, their direct implementation in the wastewater industry faces challenges due to slow growth, difficulty obtaining pure cultures, and their tendency to exist as part of an anammox consortium, interacting symbiotically with other bacteria. In the present study, 91 draft genome metagenome-assembled genomes (MAGs) from a long-term-operated reactor were recovered to clarify detailed symbiotic interactions within an anammox consortium. One marine anammox bacterial MAG, identified as Candidatus Scalindua, was successfully recovered and was abundant within the sampled microbial community. A comprehensive metabolic pathway ana­lysis revealed that Ca. Scalindua exhibited the complete anammox pathway and the Wood–Ljungdahl pathway for carbon fixation. The folate biosynthesis pathway in Ca. Scalindua was incomplete, lacking dihydrofolate reductase, a key enzyme for tetrahydrofolate (THF) production. The folate biopterin transporter, essential for transporting folate-related metabolites among coexisting bacteria, was identified exclusively in Ca. Scalindua. In addition, the impact of exogenously supplied THF on microbial activity and carbon uptake rates was investigated in batch experiments using ¹⁴C-labeled bicarbonate. The results obtained revealed that 2‍ ‍mg L^–1 of exogenous THF resulted in a 43% increase in the carbon uptake rate, while anammox activity remained unaffected. The present results suggest that THF is a key intermediate for carbon fixation in Ca. Scalindua and may be essential for their growth.

Since its discovery in wastewater sludge in the early 1990s, anaerobic ammonium oxidation (anammox) has seen significant improvements and attracted considerable research interest as a nitrogen removal process (Kuenen, 2008; Yang et al., 2022). Anammox bacteria, classified into the phylum Planctomycetota within the Planctomycetes-Verrucomicrobia-Chlamydiae (PVC) group, exhibit chemolithotrophic behavior, enabling the direct conversion of ammonium directly into nitrogen and the production of energy-rich hydrazine under anaerobic conditions using nitrite as an electron acceptor and ammonia as an electron donor (Oshiki et al., 2022). Compared with conventional nitrogen removal methods, anammox technology is suitable for various wastewater treatment processes. This is due to the properties of anammox bacteria, including their low energy requirements, high nitrogen removal capacity, high salt tolerance, endurance of high temperature variations, and ability to survive in very low-oxygen environments with minimum external carbon resources (Kindaichi et al., 2011; Gonzalez-Martinez et al., 2018; Trinh et al., 2021). Anammox bacteria are found in diverse ecological environments, from marine to freshwater and soil (Dalsgaard et al., 2005; Kartal et al., 2006; Zhu et al., 2011). Candidatus Scalindua (Kuypers et al., 2003; van de Vossenberg et al., 2013) has been recognized as the predominant anammox bacteria in marine environments, accounting for 30–50% of nitrogen removal therein (Tan et al., 2023).

Metagenomic ana­lyses have recently been conducted to examine the genomic diversity of freshwater anammox bacteria. The findings obtained revealed that inorganic-based freshwater anammox bacteria have a symbiotic relationship with other organic-based bacteria, including Proteobacteria, Actinomycetota, Chloroflexota, Bacteroidota, and the superphylum Patescibacteria (Hosokawa et al., 2021). These symbiotic microorganisms interact, exchanging nutrients, enzymes, substrates, DNA, and metabolites via a cross-feeding phenomenon (Lang and Beatty, 2000; Smith et al., 2019). Additionally, anammox bacteria generate numerous extracellular polymeric substances, which provide organics for heterotrophic bacteria in anammox consortia (Ji et al., 2021). Ongoing research is investigating secondary biosynthesis, cofactors, and vitamins alongside the main metabolic processes necessary for executing vital metabolic processes (Zhao et al., 2018). Information on the metabolic processes and symbiotic relationships of marine anammox bacteria with other coexisting microbial communities is limited in current scientific literature.

Six natural carbon fixation pathways have been identified within bacterial metabolic processes. These include the Calvin–Benson–Bassham, reductive tricarboxylic acid, 3-hydroxy propionate bi, 4-hydroxybutyarate/3-hydroxy propionate, and dicarboxylate/4-hydroxybutyrate cycles; and the acetyl-CoA/Wood–Ljungdahl and reductive glycine pathways (Garritano et al., 2022). The Wood–Ljungdahl pathway is an energy-efficient autotrophic carbon fixation process that operates under anaerobic conditions. Discovered in the early 1980s in Gram-positive (acetogenic bacteria) and methane-producing archaea (methanogenic archaea), this pathway is now recognized in various phylogenetic phyla, including some Proteobacteria, Planctomycetota, Actinomycetota, and Spirochaetota, which often perform folate metabolism (Ragsdale, 2008). Anammox bacteria are chemolithoautotrophic, utilizing organic compounds as electron donors and CO₂ as their primary carbon source to synthesize the cellular biomass (Berg et al., 2010; Martin, 2020). Among the Planctomycetota group, only anammox bacteria exhibit a complete Wood–Ljungdahl carbon fixation pathway (Hügler and Sievert, 2011). In this pathway, two CO₂ molecules from two parallel processes join the pathway to form acetyl-CoA. One CO₂ molecule is reduced to a methyl group using a tetrahydropterin coenzyme, while the other is reduced to carbon monoxide (CO). The methyl group and CO combine with nickel radicals and release coenzyme A to form an acetyl-CoA molecule (Cohen et al., 2020).

Secondary metabolites, cofactors, and vitamins are needed to complete autotrophic carbon fixation. Previous studies identified metabolites involved in molybdenum cofactor (MOCO) and folate biosynthesis, which are crucial for the Wood–Ljungdahl carbon fixation pathway (Ya et al., 2022). In addition, tetrahydrofolate (THF), a secondary folate biosynthesis metabolite, is used in the Wood–Ljungdahl carbon fixation pathway as a building block for methyl compounds, including 10-formyl tetrahydrofolate and 5,10-methyl tetrahydrofolate. However, most folate pathways in anammox bacteria appear to be incomplete (Zhao et al., 2018; Zhou et al., 2020). Therefore, other coexisting bacteria are hypothesized to provide THF metabolites to anammox bacteria via cross-feeding interactions in the anammox consortium. The mechanisms by which THF metabolites are transported between anammox and symbiotic bacteria and the impact of THF metabolites on the activity and growth of anammox bacteria currently remain unclear.

Therefore, the present study investigated the metabolic pathways of marine anammox and coexisting bacteria in a marine anammox consortium cultivated stably under laboratory conditions for more than 15 years using a metagenomic approach. Furthermore, the impact of THF supplementation on anammox activity and the Wood–Ljungdahl carbon fixation pathway was investigated through batch experiments using ¹⁴C-labeled bicarbonate.

Materials and Methods

Reactor operation and chemical ana­lysis

Biomass samples were collected from a lab-scale up-flow column reactor (volume of 300‍ ‍mL) fitted with a non-woven fabric sheet and operated for more than 15 years, as previously reported by Kindaichi et al. (2011). The up-flow column reactor was continuously supplied with a synthetic marine medium, slightly modified from that described by Kindaichi et al. (2011). It contained 30‍ ‍g‍ ‍L^–1 SEALIFE (Marine Tech); (NH₄)₂SO₄ (23.5–33.5‍ ‍mg-N L^–1), NaNO₂ (35–43‍ ‍mg-N L^–1), and KHCO₃ (1,000‍ ‍mg L^–1) as nutrient sources; along with MgSO₄·7H₂O (300‍ ‍mg L^–1), KH₂PO₄ (27‍ ‍mg L^–1), CaCl₂·2H₂O (180‍ ‍mg L^–1), and 1‍ ‍mL each of trace element solutions I and II, as described by van de Graaf et al. (1996). Before adding the nutrient solution, the substrate was degassed for 1‍ ‍h with N₂ gas to remove dissolved oxygen (DO). pH was adjusted using H₂SO₄ to control the influent pH at 7.0–7.5. The concentration of NH₄⁺ was measured using Nessler’s method with a UV-visible spectrophotometer (DR 2800; Hach Company) (Kambara et al., 2022). To measure NO₂^– and NO₃^–, samples were prepared by filtering through a membrane with a pore size of 0.20‍ ‍μm (Advance Tec). NO₂^– and NO₃^– concentrations then assessed using ion chromatography (HPLC 10 Avp; Shimadzu).

Sample collection, sequencing, and assembly

Three separate biomass samples were collected from the lab-scale up-flow Ca. Scalindua reactor fed with synthetic marine medium in April 2019 (day 3,876), August 2019 (day 4,044), and June 2020 (day 4,346) (Fig. S1). These samples were collected at different time points to capture potential variations in the microbial community over time. Fujii et al. (2022) provided a detailed description of the DNA extraction, sequencing, and assembly methodology adopted in the present study. Briefly, DNA was extracted using a Fast DNA Spin kit for soil (MP Biomedicals) according to the manufacturer’s instructions and subsequently purified using Agencount AMPure XP magnetic beads (Beckman Coulter). DNA extracted from the three separate samples was used to generate three separate libraries for Illumina short-read sequencing and PacBio long-read. Illumina sequencing libraries were prepared using a TruSeq DNA PCR-free kit, with paired-end sequencing for short reads (2×151 bp) on a HiSeqX Illumina Platform (Illumina). PacBio sequencing libraries (long reads) were prepared using a SMRT bell Express Template prep kit (PacBio) on a PacBio sequel II system. Circular consensus sequence (CCS) reads were generated from sequel data with a Phred quality score >20 (Q20, 99%). Raw paired-end reads from HiSeqX were trimmed using Trimmomatic version 0.39 (SLIDING WINDOW; 4:30, MINLEN: 100) (Bolger et al., 2014). Three different sets of trimmed HiSeqX and PacBio CCS reads were assembled using SPAdes version 3.15.4 with K numbers of 21, 31, 41, 51, 61, 71, 81, 91, 101, 111, and 127 (Bankevich et al., 2012) to produce coverage files for metagenomic binning. Moreover, contigs exceeding 1,500‍ ‍bp were separated using SeqKit version 2.2.0 (Shen et al., 2016). Minimap2 release 2.17-r941 (Li, 2018) was used to align the whole genome and obtain mapping information.

Binning, quality assessment, and annotation

Metagenomic binning was performed through a multidimensional coverage approach with several tools, including MetaBAT2 version 2.15 (Kang et al., 2019), MaxBin2 version 2.2.7 (Wu et al., 2016), MyCC (MyCC_2017.ova, Lin and Liao, 2016), and Vamb version 3.0.2 (Nissen et al., 2021). Optimized binning results were obtained using DasTool version 1.1.5 (Sieber et al., 2018). The completeness and contamination of each metagenome-assembled genome (MAG) were evaluated using CheckM version 1.2.0 (Parks et al., 2015). Relative abundance at the genus level was estimated using CoverM version 0.6.1 for each of the three samples, and the average value is shown in Table 1.

Table 1.

Metagenome-assembled genomes obtained from a long-term marine anammox column reactor.

MAG ID	GTDB Taxonomy	Completenessa	Contaminationa	Genome size (bp)	Number of contigsb	Abundance (%)
Sca.027	p__Planctomycetota;c__Brocadiae;o__Brocadiales;f__Scalinduaceae;g__SCAELEC01	96.6	3.4	5,099,442	11	23.3
Sca.016	p__Planctomycetota;c__Phycisphaerae;o__Phycisphaerales;f__UBA5793;g__PNC21	97.7	1.1	4,497,157	34	19.4
Sca.078	p__Proteobacteria;c__Gammaproteobacteria;o__QQUJ01;f__QQUJ01;g__QQUJ01	83.1	2.7	3,509,019	12	6.8
Sca.069	p__Proteobacteria;c__Gammaproteobacteria;o__GCA-2729495;f__GCA-2729495;g__GCA-002729875	86.8	1.5	3,415,451	10	2.6
Sca.039	p__Myxococcota;c__Polyangia;o__Haliangiales;f__Haliangiaceae	94.5	5.3	10,004,283	41	2.4
Sca.042	p__Bacteroidota;c__Ignavibacteria;o__Ignavibacteriales;f__Melioribacteraceae;g__IGN16	94.1	2.0	4,524,917	25	2.1
Sca.075	p__Proteobacteria;c__Gammaproteobacteria;o__Xanthomonadales;f__Wenzhouxiangellaceae;g__Wenzhouxiangella	83.6	2.0	2,763,812	14	2.0
Sca.051	p__Planctomycetota;c__B15-G4;o__B15-G4;f__JADJEK01	92.5	1.1	7,268,394	36	1.8
Sca.011	p__Proteobacteria;c__Gammaproteobacteria;o__UBA5335;f__UBA5335;g__Macondimonas	98.5	1.2	2,642,206	2	1.8
Sca.002	p__Actinobacteriota;c__Acidimicrobiia;o__UBA5794;f__UBA5794;g__UBA5794	100	1.3	2,648,439	13	1.7
Sca.012	p__Bacteroidota;c__Bacteroidia;o__Flavobacteriales;f__Flavobacteriaceae;g__Altibacter	98.5	0	3,109,375	2	1.6
Sca.060	p__Chloroflexota;c__Anaerolineae;o__Anaerolineales;f__UBA11579;g__CFX15	90.9	4.6	4,962,996	5	1.0
Sca.045	p__Planctomycetota;c__Planctomycetia;o__Pirellulales;f__Pirellulaceae	93.7	1.1	7,549,545	42	1.0
Sca.028	p__Planctomycetota;c__UBA1135;o__UBA2386;f__GCA-002748355	96.6	0.1	5,900,917	20	1.0
Sca.047	p__Proteobacteria;c__Gammaproteobacteria;o__Pseudomonadales;f__HTCC2089;g__SZUA-309	93.5	2.2	4,969,807	15	0.9
Sca.019	p__Proteobacteria;c__Alphaproteobacteria;o__Rhodobacterales;f__Rhodobacteraceae;g__Roseovarius	97.3	2.8	3,730,243	8	0.9
Sca.017	p__Spirochaetota;c__Leptospirae;o__Leptospirales	97.7	0	3,888,891	12	0.9
Sca.048	p__Proteobacteria;c__Gammaproteobacteria;o__Xanthomonadales;f__Marinicellaceae;g__Marinicella	93.5	1.5	2,935,724	5	0.9
Sca.074	p__Myxococcota_A;c__UBA796;o__UBA796;f__GCA-2862545	83.9	3.9	10,044,473	117	0.7
Sca.040	p__Planctomycetota;c__UBA8742;o__SKZP01	94.3	5.7	5,934,693	10	0.7
Sca.044	p__Proteobacteria;c__Alphaproteobacteria;o__SMXQ01;f__SMXQ01	93.9	2.6	3,077,441	9	0.6
Sca.006	p__Actinobacteriota;c__Actinomycetia;o__Mycobacteriales;f__Mycobacteriaceae;g__Mycobacterium	98.9	0.6	5,467,850	23	0.6
Sca.082	p__Chloroflexota;c__Anaerolineae;o__4572-78;f__J111;g__J095	78.5	3.6	7,059,453	58	0.6
Sca.026	p__Verrucomicrobiota;c__Verrucomicrobiae;o__Opitutales;f__UBA2995;g__UBA2995	96.6	4.7	6,670,232	36	0.5
Sca.054	p__Actinobacteriota;c__Acidimicrobiia;o__Acidimicrobiales;f__Microtrichaceae	92.3	0.9	4,078,026	21	0.5

^a A quality assessment of the assembled genome was conducted using CheckM version 1.2.0.

^b Contigs and 16S rRNA from the assembled genome were evaluated using PROKKA version 1.1.3.

All MAGs represent >70% completeness, <6% contamination, and >0.5% abundance.

Genome annotation and functional gene identification

Genome annotation was conducted using PROKKA version 1.13 (Seemann, 2014) and BlastKOALA (Kanehisa et al., 2016) from the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. A heat map was constructed using a KEGG decoder (Graham et al., 2018) to visualize the percentage of gene possession related to each gene set. The genome tree was constructed using a combination of GTDB-Tk version 2.2.6 (R207) and IQ-TREE version 2.2.2.3. Conserved marker genes were identified using “gtdbtk identify” as the default parameter. Phylogenetic trees were constructed using an auto-optimized surrogate model (LG+F+I+R10) in IQ-TREE (B1000) (Minh et al., 2020). Furthermore, functional genes related to nitrogen metabolism, folate metabolism, and the Wood–Ljungdahl carbon fixation pathway were thoroughly identified using PROKKA and BlastKOALA. The predicted functional genes were taxonomically confirmed using BLAST P against the non-redundant protein sequences (nr) database. Outputs were further filtered based on identity, with more than 80% identified for nitrogen metabolism and 50% for folate metabolism and carbon fixation in the Wood–Ljungdahl pathway.

Batch experiments

Batch experiments were performed to investigate the effects of exogenously added THF on anammox activity and carbon uptake, as summarized in Table S1. A 0.5-mL biomass sample (0.35‍ ‍mg protein vial^–1) and 4‍ ‍mL of the synthetic marine medium were transferred into 10-mL vials. Regarding THF samples, 0.5‍ ‍mL of THF was added, while for control samples, 0.5‍ ‍mL of the synthetic marine medium (excluding [NH₄]₂SO₄, NaNO₂, and KHCO₃) was used. The vials were degassed with N₂ gas at 0.04‍ ‍MPa for 3‍ ‍min to remove DO. Samples were then pressurized with N₂ gas at 0.04‍ ‍MPa for 10‍ ‍s to prevent air contamination before adding ¹⁴C. ¹⁴C-labeled bicarbonate (Sodium [¹⁴C] bicarbonate, specific activity 2.146 GBq mmol^–1) (Revvity) was added to vials at a final concentration of 37 kBq vial^–1. Thereafter, vials were sealed with a butyl rubber plug and aluminum seal and then incubated at 28°C for 24 h. An additional experiment involved pasteurizing the anammox biomass at 70°C for 15‍ ‍min to test for adsorption phenomena. Biomass concentrations were measured as protein using a TaKaRa BCA Protein Assay kit (Takara Bio). After the incubation, all biomass samples were fixed by adding 2.5‍ ‍mL of 12% paraformaldehyde to the 5-mL culture, yielding a final concentration of 4% paraformaldehyde. The uptake of ¹⁴C was confirmed by liquid scintillation counting. The ¹⁴C content was calculated from 1‍ ‍mL of the total culture sample (¹⁴C-containing medium plus biomass) and 1‍ ‍mL of the washed ¹⁴C-containing biomass sample (obtained by resuspending in 1‍ ‍mL of pure water after centrifugation at 10,000×g for 8‍ ‍min following three washes with phosphate-buffered saline, pH=7.4). Each 1-mL sample was mixed with 9‍ ‍mL of scintillation cocktail (Clear-sol I; Nacalai Tesque), and radioactivity was measured using an LSC-6100 liquid scintillation counter (ALOKA). Since the variation between the two samples was unequal, Welch’s t-test was performed using R version 4.3.1 (R CoreTeam, 2023).

Nucleotide sequence accession number

Metagenomic sequence data were deposited in the DDBJ database under the DDBJ/EMBL/GenBank accession number PRJDB17738.

Results and Discussion

Genome reconstruction and general genome features

In total, 0.77 billion reads were generated collectively from the sequencing process (Tables S2 and S3). After removing low-quality bases by trimming, 0.39 billion high-quality reads were generated from HiSeq X short-read sequencing. A total of 9,333 contigs over 1,500 bp were obtained from these sequence reads and used in the multiple-dimensional coverage binning approach, generating 91 MAGs. The 25 most abundant MAGs (23.3%) were then listed based on a criterion of <6% contamination and >70% completeness (Table 1). The total abundance of the 25 MAGs accounted for 76.3%. Fig. S2 shows a phylogenetic tree of the 25 recovered genome MAGs based on protein sequences. These 25 MAGs were associated with Planctomycetota (6 MAGs), Proteobacteria (8 MAGs), Myxococcota (2 MAGs), Bacteroidota (2 MAGs), Actinomycetota (3 MAGs), Chloroflexota (2 MAGs), Spirochaetota (1 MAG), and Verrucomicrobiota (1 MAG). The most abundant MAG (23.3%) was closely related to Ca. Scalindua (GCA_004282745.1). The unclassified Planctomycetota genome, with the second highest abundance (19.4%), was closely related to the uncultured, genus-level lineage PNC21 in Phycisphaerales. The third most abundant MAG belonged to the phylum Proteobacteria. The abundance of the phyla Chloroflexota, Verrucomicrobiota, and Bacteroidota, which showed close symbiosis with freshwater and marine anammox bacteria in previous studies (Ali et al., 2020; Oshiki et al., 2022; Oba et al., 2024), was extremely low in the present study.

Metabolic ana­lysis

The most highlighted MAG in this study, Ca. Scalindua (Sca.027 MAG), clearly showed high completeness of the TCA cycle, gluconeogenesis, and glycolysis metabolism (Fig. S3). Moreover, concerning nitrogen metabolism, the Sca.027 MAG showed all related genes for anammox metabolism, including hydrazine dehydrogenase (hdh), with 100% identity with Ca. Scalindua sp. (Accession Number KAA3603839.1), and hydrazine synthase (hzs), with 92.4% identity with Ca. Scalindua rubra (Accession Number ODQ59245.1) (Fig. 1A). Sca.016, the second most abundant genome affiliated with the UBA5793 family in the phylum Planctomycetota, was predicted to possess a complete assimilatory nitrate reduction pathway, including narB and nasB (Fig. 1A).

Fig. 1.

A heatmap showing the presence and absence of functional genes associated with nitrogen metabolism, carbon fixation, and folate metabolism. Gene presence was confirmed using a BLAST P reference dataset, with nitrogen metabolism-related genes considered to be present at >80% identity and carbon fixation and folate metabolism genes at >50% identity. (A) Nitrogen metabolism-related functional genes. narGHI and napAB, nitrate reductase; narB, ferredoxin-nitrate reductase; nasAB, assimilatory nitrate reductase catalytic subunit; nasBDE, nitrite reductase (NADCP)H large subunit; nirBD, nitrite reductase (NADH) larger subunit; nxrAB, nitrite oxidation; nirK/nirS, nitrite reduction; norBC, nitric oxide reduction; nosZ, nitrous oxide reduction; nifDKH, nitrogenase molybdenum-iron protein alpha chain; hao, hydroxylamine oxidoreductase; amoABC, Pmmo, ammonia oxidation; hdh, hydrazine dehydrogenase; hzs, hydrazine synthase subunit. (B) Wood–Lundahl carbon fixation pathway-related functional genes. fdhA, formate dehydrogenase alpha subunit; fhs, formate THF ligase; fchA, methyl-THF cyclohydrolase; fold, methylene-THF dehydrogenase (NADP⁺); Metf, methylene-THF reductase (NADH); acsA, carbon monoxide dehydrogenase; acsB, acetyl-CoA synthase; and acsE, 5-methyl-THF corrinoid/iron-sulfur protein methyltransferase. (C) Folate biosynthesis-related functional genes. folA, Dihydrofolate reductase; folC, folylpolyglutamate synthase; and FBT, folate-biopterin transporter. After the ana­lysis, nitrogen metabolism-related functional genes were sorted at >80%, and carbon fixation and folate metabolism-related functional genes were sorted at >50%.

Fig. 1B shows the key enzymes of the Wood–Ljungdahl carbon fixation pathways for all 25 MAGs, including the formate dehydrogenase alpha subunit (fdh), formate THF ligase (fhs), methyl THF cyclohydrolase (fchA), methyl THF dehydrogenase (folD), and methyl THF reductase (metf) for the methyl branch and carbon monoxide dihydrofolate (acsA), acetyl-CoA synthase (acsB), and 5-methyl-THF corrinoid/iron-sulfur protein methyltransferase (acsE) for the carbonyl branch. However, only the genome of the Ca. Scalindua-related MAG, Sca.027, clearly showed a complete Wood–Ljungdahl carbon fixation pathway, encoding all specific key enzymes with high identity (BLAST P). We confirmed that the Sca.027 MAG lacked the complete set of genes required for the five other carbon fixation pathways based on BlastKOALA results. Since Ca. Scalindua is recognized as an anaerobic autotrophic bacterium, it primarily utilizes the Wood–Ljungdahl pathway for carbon fixation (Ragsdale and Pierce, 2008). Conversely, all other MAGs exhibit an incomplete carbon fixation pathway characterized by heterotrophic bacterial behavior.

Formate dehydrogenases, metabolite products of MOCO biosynthesis, and THF from the complete folate biosynthesis pathway are involved in the Wood–Ljungdahl pathway of anammox bacteria (Zhuang et al., 2014). The present study primarily focused on investigating marine anammox folate biosynthesis, the effects of THF metabolites on the anammox carbon fixation pathway, and marine anammox bacteria growth potential. Notably, marine anammox bacteria (MAG Sca.027) exhibited an incomplete folate biosynthesis pathway (Fig. 1C), lacking the dihydrofolate reductase (folA) enzyme essential for THF production. Dihydrofolate reductase is one of the critical enzymes for converting dihydrofolate to THF and is involved in purine and thymidylate synthesis (Lin and Gerson, 2014).

The Sca.078 (Proteobacteria), Sca.042 (Bacteroidota), Sca.060 (Chloroflexota), Sca.026 (Verrucomicrobiota), and Sca.006 (Actinobacteriota) MAGs exhibited a complete folate biosynthesis pathway (i.e., possessing folA and folC) with high sequence identities based on a BLAST P ana­lysis. Previous studies proposed that cross-feeding transfers may be crucial for facilitating the exchange of deficient enzymes or secondary metabolites between anammox bacteria and other symbiotic microorganisms (Zhou et al., 2020). Precisely defining the cross-feeding mechanism and identifying suitable transporters that facilitate the transfer of THF metabolites between species is necessary. Therefore, we conducted a comprehensive ana­lysis to examine the presence of folate transporters within MAGs. The folate biopterin transporter (FBT) in the Sca.027 MAG (Ca. Scalindua) was identified using PROKKA and confirmed via a BLAST P ana­lysis as an MFS transporter with 100% identity (KAA3592863.1) (Fig. 1C). FBT transporters are proton-coupled symporters that move folates and biopterin across membranes, primarily in bacteria and other lower organisms (Fig. 2) (Eudes et al., 2010). However, further research is needed to identify more specific roles for and the mechanisms of these transporters. Furthermore, we compared the recovered Ca. Scalindua MAG (Sca.027) with 30 other Ca. Scalindua MAGs (Table S4). Our ana­lysis revealed that all Ca. Scalindua MAGs lacked folA, the genome encoding dihydrofolate reductase responsible for metabolizing folate to DHF and THF. It is possible that anammox bacteria obtain the necessary secondary metabolites to complete carbon fixation through mechanisms such as uptake from dead cells, cell fusion, cell-cell interactions, or bacterial membrane vesicles (Venturi and Bez, 2021; Toyofuku et al., 2023). Furthermore, a key observation in the ana­lysis of Wood–Ljungdahl metabolism is that the THF generated during the conversion from 5-methyl-THF to methyl-CfeSP undergoes a re-reaction with folate, thereby completing the Wood–Ljungdahl carbon fixation pathway. This catalytic process is facilitated by the 5-methyl-THF corrinoid/iron-sulfur protein methyltransferase (acsE) enzyme, present only in MAG Sca.027.

Fig. 2.

Overview of predicted metabolic pathways. The pathway containing secondary metabolism is related to the Wood–Ljungdahl carbon fixation pathway in marine anammox bacteria. The black solid line indicates a complete pathway, and the gray dashed line represents an incomplete pathway in the folate biosynthesis pathway.

The impact of exogenous THF on carbon fixation

The impact of exogenous THF on laboratory-cultivated marine anammox bacterial consortia was investigated through batch experiments. The effects of THF on carbon fixation were assessed by quantifying the carbon uptake rate. Additionally, consequential effects on anammox bacterial activity were exami­ned. The average ammonia removal rates in control and THF-supplemented samples were 1.86±0.17 and 1.95±0.16‍ ‍μmol-N mg protein^–1 h^–1 (Fig. 3A), and carbon uptake rates were 0.054±0.02 and 0.077±0.01‍ ‍μmol-C mg protein^–1 h^–1, respectively (Fig. 3B). Pasteurized samples showed a significantly low carbon uptake rate due to the inactivation of enzymes and microbial activity.

Fig. 3.

Effects of tetrahydrofolate (THF) on anammox activity (A) and carbon uptake (B).

These results demonstrated that the carbon uptake rate was significantly higher in THF-supplemented samples than in control samples, while no significant differences were observed in ammonia removal rates between these two conditions. It is important to note that biomass yield efficiency was calculated as 0.029‍ ‍μmol-C μmol-N^–1 in the control, which was consistent with previous findings (Awata et al., 2013). On the other hand, biomass yield efficiency with the THF addition (2‍ ‍mg L^–1) was 0.040‍ ‍μmol-C μmol-N^–1, which showed a 38% increase. Further confirmation that exogenously introduced components enhance biomass yield in Ca. Scalindua is needed (Awata et al., 2015; Mojiri et al., 2018). Exogenous THF may exert complex effects on the Wood–Ljungdahl carbon fixation pathway. In cases where Ca. Scalindua is deficient in THF, supplementation with THF may simulate the methyl branch of the carbon fixation pathway by providing the necessary cofactors (Fig. 2). This may lead to increased acetyl-CoA production and an enhanced carbon uptake capacity. Furthermore, FBT facilitates the translocation of THF across the biological membrane. However, since THF is also produced as a byproduct of the Wood–Ljungdahl carbon fixation pathway, exogenous THF may not always be required for effective fixation.

Conclusion

The present study performed a comprehensive metagenomic ana­lysis and evaluated the effects of exogenous THF supplementation in a marine anammox consortium that has been cultivated under controlled laboratory conditions for more than 15 years. The metagenomic approach successfully recovered the predominant Ca. Scalindua metagenomic MAG, which contained FBTs, but lacked folate biosynthesis genes. In addition, some coexisting bacteria within the anammox consortium possessed THF. Batch experiments with exogenous THF and ¹⁴C-labeled bicarbonate showed an increase of 43% in the carbon uptake rate. These results support the important role of THF in promoting enzymatic activity during carbon processing without disturbing anammox activity. Future research is needed on other compounds that may enhance anammox activity, accelerate growth rates, and reveal gene expression. Additionally, detailed biochemical and structural studies are essential to fully elucidate the mechanisms of FBTs.

Citation

Nawarathna, T. N. T. K., Fujii, N., Yamamoto, K., Kuroda, K., Narihiro, T., Ozaki, N., et al. (2025) Metagenomic Insights into Candidatus Scalindua in a Long-term Cultivated Marine Anammox Consortium: The Important Role of Tetrahydrofolate-mediated Carbon Fixation. Microbes Environ 40: ME25007.

https://doi.org/10.1264/jsme2.ME25007

Acknowledgements

This work was supported by JSPS KAKENHI Grant Numbers JP20KK0244 and JP21K19866. The ¹⁴C experiment was conducted at the Department of Radioisotope Science and Safety, Integrated Experimental Support/Research Division, Natural Science Center for Basic Research and Development, Hiroshima University.

References

•  Ali,  M.,  Shaw,  D.R.,  Albertsen,  M., and  Saikaly,  P.E. (2020) Comparative genome-centric ana­lysis of freshwater and marine anammox cultures suggests functional redundancy in nitrogen removal processes. Front Microbiol 11: 1637.
•  Awata,  T.,  Oshiki,  M.,  Kindaichi,  T.,  Ozaki,  N.,  Ohashi,  A., and  Okabe,  S. (2013) Physiological characterization of an anaerobic ammonium-oxidizing bacterium belonging to the “Candidatus Scalindua” group. Appl Environ Microbiol 79: 4145–4148.
•  Awata,  T.,  Kindaichi,  T.,  Ozaki,  N., and  Ohashi,  A. (2015) Biomass yield efficiency of the marine anammox bacterium, “Candidatus Scalindua sp.,” is affected by salinity. Microbes Environ 30: 86–91.
•  Bankevich,  A.,  Nurk,  S.,  Antipov,  D.,  Gurevich,  A.A.,  Dvorkin,  M.,  Kulikov,  A.S., et al. (2012) SPAdes: A new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol 19: 455–477.
•  Berg,  I.A.,  Kockelkorn,  D.,  Ramos-Vera,  W.H.,  Say,  R.F.,  Zarzycki,  J.,  Hügler,  M., et al. (2010) Autotrophic carbon fixation in archaea. Nat Rev Microbiol 8: 447–460.
•  Bolger,  A.M.,  Lohse,  M., and  Usadel,  B. (2014) Trimmomatic: A flexible trimmer for Illumina sequence data. Bioinformatics 30: 2114–2120.
•  Cohen,  S.E.,  Can,  M.,  Wittenborn,  E.C.,  Hendrickson,  R.A.,  Ragsdale,  S.W., and  Drennan,  C.L. (2020) Crystallographic characterization of the carbonylated A-cluster in carbon monoxide dehydrogenase/acetyl-CoA synthase. ACS Catal 10: 9741–9746.
•  Dalsgaard,  T.,  Thamdrup,  B., and  Canfield,  D.E. (2005) Anaerobic ammonium oxidation (anammox) in the marine environment. Res Microbiol 156: 457–464.
•  Eudes,  A.,  Kunji,  E.R.S.,  Noiriel,  A.,  Klaus,  S.M.J.,  Vickers,  T.J.,  Beverley,  S.M., et al. (2010) Identification of transport-critical residues in a folate transporter from the folate-biopterin transporter (FBT) family. J Biol Chem 285: 2867–2875.
•  Fujii,  N.,  Kuroda,  K.,  Narihiro,  T.,  Aoi,  Y.,  Ozaki,  N.,  Ohashi,  A., and  Kindaichi,  T. (2022) Metabolic potential of the superphylum Patescibacteria reconstructed from activated sludge samples from a municipal wastewater treatment plant. Microbes Environ 37: ME22012.
•  Garritano,  A.N.,  Song,  W., and  Thomas,  T. (2022) Carbon fixation pathways across the bacterial and archaeal tree of life. PNAS Nexus 1: pgac226.
•  Gonzalez-Martinez,  A.,  Muñoz-Palazon,  B.,  Rodriguez-Sanchez,  A., and  Gonzalez-Lopez,  J. (2018) New concepts in anammox processes for wastewater nitrogen removal: Recent advances and future prospects. FEMS Microbiol Lett 365: fny031.
•  Graham,  E.D.,  Heidelberg,  J.F., and  Tully,  B.J. (2018) Potential for primary productivity in a globally distributed bacterial phototroph. ISME J 12: 1861–1866.
•  Hosokawa,  S.,  Kuroda,  K.,  Narihiro,  T.,  Aoi,  Y.,  Ozaki,  N.,  Ohashi,  A., and  Kindaichi,  T. (2021) Cometabolism of the superphylum Patescibacteria with anammox bacteria in a long-term freshwater anammox column reactor. Water 13: 208.
•  Hügler,  M., and  Sievert,  S.M. (2011) Beyond the Calvin cycle: Autotrophic carbon fixation in the ocean. Ann Rev Mar Sci 3: 261–289.
•  Ji,  X.-M.,  Zheng,  C.,  Wang,  Y.-L., and  Jin,  R.-C. (2021) Decoding the interspecies interaction in anammox process with inorganic feeding through metagenomic and metatranscriptomic ana­lysis. J Cleaner Prod 288: 125691.
•  Kambara,  H.,  Shinno,  T.,  Matsuura,  N.,  Matsushita,  S.,  Aoi,  Y.,  Kindaichi,  T., et al. (2022) Environmental factors affecting the community of methane-oxidizing bacteria. Microbes Environ 37: ME21074.
•  Kanehisa,  M.,  Sato,  Y., and  Morishima,  K. (2016) BlastKOALA and GhostKOALA: KEGG tools for functional characterization of genome and metagenome sequences. J Mol Biol 428: 726–731.
•  Kang,  D.D.,  Li,  F.,  Kirton,  E.,  Thomas,  A.,  Egan,  R.,  An,  H., and  Wang,  Z. (2019) MetaBAT 2: An adaptive binning algorithm for robust and efficient genome reconstruction from metagenome assemblies. PeerJ 7: e7359.
•  Kartal,  B.,  Koleva,  M.,  Arsov,  R.,  van der Star,  W.,  Jetten,  M.S.M., and  Strous,  M. (2006) Adaptation of a freshwater anammox population to high salinity wastewater. J Biotechnol 126: 546–553.
•  Kindaichi,  T.,  Awata,  T.,  Suzuki,  Y.,  Tanabe,  K.,  Hatamoto,  M.,  Ozaki,  N., and  Ohashi,  A. (2011) Enrichment using an up-flow column reactor and community structure of marine anammox bacteria from coastal sediment. Microbes Environ 26: 67–73.
•  Kuenen,  J.G. (2008) Anammox bacteria: From discovery to application. Nat Rev Microbiol 6: 320–326.
•  Kuypers,  M.M.M.,  Sliekers,  A.O.,  Lavik,  G.,  Schmid,  M.,  Jørgensen,  B.B.,  Kuenen,  J.G., et al. (2003) Anaerobic ammonium oxidation by anammox bacteria in the Black Sea. Nature 422: 608–611.
•  Lang,  A.S., and  Beatty,  J.T. (2000) Genetic ana­lysis of a bacterial genetic exchange element: The gene transfer agent of Rhodobacter capsulatus. Proc Natl Acad Sci U S A 97: 859–864.
•  Li,  H. (2018) Minimap2: Pairwise alignment for nucleotide sequences. Bioinformatics 34: 3094–3100.
•  Lin,  H.-H., and  Liao,  Y.-C. (2016) Accurate binning of metagenomic contigs via automated clustering sequences using information of genomic signatures and marker genes. Sci Rep 6: 24175.
• Lin, Y., and Gerson, S.L. (2014) Clinical trials using LV-P140K-MGMT for gliomas. In Gene Therapy of Cancer. Elsevier, pp. 379–391.
•  Martin,  W.F. (2020) Older than genes: The acetyl CoA pathway and origins. Front Microbiol 11: 817.
•  Minh,  B.Q.,  Schmidt,  H.A.,  Chernomor,  O.,  Schrempf,  D.,  Woodhams,  M.D.,  von Haeseler,  A., and  Lanfear,  R. (2020) IQ-TREE 2: New models and efficient methods for phylogenetic inference in the genomic era. Mol Biol Evol 37: 1530–1534.
•  Mojiri,  A.,  Nishimoto,  K.,  Awata,  T.,  Aoi,  Y.,  Ozaki,  N.,  Ohashi,  A., and  Kindaichi,  T. (2018) Effects of salts on the activity and growth of “candidatus scalindua sp.”, a marine anammox bacterium. Microbes Environ 33: 336–339.
•  Nissen,  J.N.,  Johansen,  J.,  Allesøe,  R.L.,  Sønderby,  C.K.,  Armenteros,  J.J.A.,  Grønbech,  C.H., et al. (2021) Improved metagenome binning and assembly using deep variational autoencoders. Nat Biotechnol 39: 555–560.
•  Oba,  K.,  Suenaga,  T.,  Yasuda,  S.,  Kuroiwa,  M.,  Hori,  T.,  Lackner,  S., and  Terada,  A. (2024) Quest for nitrous oxide-reducing bacteria present in an anammox biofilm fed with nitrous oxide. Microbes Environ 39: ME23106.
•  Oshiki,  M.,  Takaki,  Y.,  Hirai,  M.,  Nunoura,  T.,  Kamigaito,  A., and  Okabe,  S. (2022) Metagenomic ana­lysis of five phylogenetically distant anammox bacterial enrichment cultures. Microbes Environ 37: ME22017.
•  Parks,  D.H.,  Imelfort,  M.,  Skennerton,  C.T.,  Hugenholtz,  P., and  Tyson,  G.W. (2015) CheckM: Assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes. Genome Res 25: 1043–1055.
• R Core Team (2023) R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. URL https://www.R-project.org/
•  Ragsdale,  S.W. (2008) Enzymology of the Wood–Ljungdahl pathway of acetogenesis. Ann N Y Acad Sci 1125: 129–136.
•  Ragsdale,  S.W., and  Pierce,  E. (2008) Acetogenesis and the Wood–Ljungdahl pathway of CO2 fixation. Biochim Biophys Acta 1784: 1873–1898.
•  Seemann,  T. (2014) Prokka: Rapid prokaryotic genome annotation. Bioinformatics 30: 2068–2069.
•  Shen,  W.,  Le,  S.,  Li,  Y., and  Hu,  F. (2016) SeqKit: A cross-platform and ultrafast toolkit for FASTA/Q file manipulation. PLoS One 11: e0163962.
•  Sieber,  C.M.K.,  Probst,  A.J.,  Sharrar,  A.,  Thomas,  B.C.,  Hess,  M.,  Tringe,  S.G., and  Banfield,  J.F. (2018) Recovery of genomes from metagenomes via a dereplication, aggregation and scoring strategy. Nat Microbiol 3: 836–843.
•  Smith,  N.W.,  Shorten,  P.R.,  Altermann,  E.,  Roy,  N.C., and  McNabb,  W.C. (2019) The classification and evolution of bacterial cross-feeding. Front Ecol Evol 7: 153.
•  Tan,  C.,  Zhang,  W.,  Wei,  Y.,  Zhao,  N., and  Li,  J. (2023) Insights into nitrogen removal and microbial response of marine anammox bacteria-based consortia treating saline wastewater: From high to moderate and low salinities. Bioresour Technol 382: 129220.
•  Toyofuku,  M.,  Schild,  S.,  Kaparakis-Liaskos,  M., and  Eberl,  L. (2023) Composition and functions of bacterial membrane vesicles. Nat Rev Microbiol 21: 415–430.
•  Trinh,  H.P.,  Lee,  S.-H.,  Jeong,  G.,  Yoon,  H., and  Park,  H.-D. (2021) Recent developments of the mainstream anammox processes: Challenges and opportunities. J Environ Chem Eng 9: 105583.
•  van de Graaf,  A.A.,  de Bruijn,  P.,  Robertson,  L.A.,  Jetten,  M.S.M., and  Kuenen,  J.G. (1996) Autotrophic growth of anaerobic ammonium-oxidizing micro-organisms in a fluidized bed reactor. Microbiology 142: 2187–2196.
•  van de Vossenberg,  J.,  Woebken,  D.,  Maalcke,  W.J.,  Wessels,  H.J.C.T.,  Dutilh,  B.E.,  Kartal,  B., et al. (2013) The metagenome of the marine anammox bacterium “Candidatus Scalindua profunda” illustrates the versatility of this globally important nitrogen cycle bacterium. Environ Microbiol 15: 1275–1289.
•  Venturi,  V., and  Bez,  C. (2021) A call to arms for cell–cell interactions between bacteria in the plant microbiome. Trends Plant Sci 26: 1126–1132.
•  Wu,  Y.-W.,  Simmons,  B.A., and  Singer,  S.W. (2016) MaxBin 2.0: An automated binning algorithm to recover genomes from multiple metagenomic datasets. Bioinformatics 32: 605–607.
•  Ya,  T.,  Liu,  J.,  Zhang,  M.,  Wang,  Y.,  Huang,  Y.,  Hai,  R., et al. (2022) Metagenomic insights into the symbiotic relationship in anammox consortia at reduced temperature. Water Res 225: 119184.
•  Yang,  Y.,  Lu,  Z.,  Azari,  M.,  Kartal,  B.,  Du,  H.,  Cai,  M., et al. (2022) Discovery of a new genus of anaerobic ammonium oxidizing bacteria with a mechanism for oxygen tolerance. Water Res 226: 119165.
•  Zhao,  Y.,  Liu,  S.,  Jiang,  B.,  Feng,  Y.,  Zhu,  T.,  Tao,  H., et al. (2018) Genome-centered metagenomics ana­lysis reveals the symbiotic organisms possessing ability to cross-feed with anammox bacteria in anammox consortia. Environ Sci Technol 52: 11285–11296.
•  Zhou,  Y.-Y.,  Shao,  W.-L.,  Liu,  Y.D.,  Li,  X.,  Shan,  X.-Y.,  Jin,  X.-B., et al. (2020) Genome-based ana­lysis to understanding rapid resuscitation of cryopreserved anammox consortia via sequential supernatant addition. Sci Total Environ 744: 140785.
•  Zhu,  G.,  Wang,  S.,  Wang,  Y.,  Wang,  C.,  Risgaard-Petersen,  N.,  Jetten,  M.S.M., and  Yin,  C. (2011) Anaerobic ammonia oxidation in a fertilized paddy soil. ISME J 5: 1905–1912.
•  Zhuang,  W.-Q.,  Yi,  S.,  Bill,  M.,  Brisson,  V.L.,  Feng,  X.,  Men,  Y., et al. (2014) Incomplete Wood–Ljungdahl pathway facilitates one-carbon metabolism in organohalide-respiring Dehalococcoides mccartyi. Proc Natl Acad Sci U S A 111: 6419–6424.