Amino Acid Auxotrophs Research Articles

Ubiquitous genome streamlined Acidobacteriota in freshwater environments.

Acidobacteriota are abundant in soil, peatlands, and sediments, but their ecology in freshwater environments remains understudied. UBA12189, an Acidobacteriota genus, is an uncultivated, genome-streamlined lineage with a small genome size found in aquatic environments where detailed genomic analyses are lacking. Here, we analyzed 66 MAGs of UBA12189 (including one complete genome) from freshwater lakes and rivers in Europe, North America, and Asia. UBA12189 has small genome sizes (<1.4 Mbp), low GC content, and a highly diverse pangenome. In freshwater lakes, this bacterial lineage is abundant from the surface waters (epilimnion) down to a 300-m depth (hypolimnion). UBA12189 appears to be free-living from CARD-FISH analysis. When compared to other genome-streamlined bacteria such as Nanopelagicales and Methylopumilus, genome reduction has caused UBA12189 to have a more limited metabolic repertoire in carbon, sulfur, and nitrogen metabolisms, limited numbers of membrane transporters, as well as a higher degree of auxotrophy for various amino acids, vitamins, and reduced sulfur. Despite having reduced genomes, UBA12189 encodes proteorhodopsin, complete biosynthesis pathways for heme and vitamin K2, cbb3-type cytochrome c oxidases, and heme-requiring enzymes. These genes may give a selective advantage during the genome streamlining process. We propose the new genus Acidiparvus, with two new species named "A. lacustris" and "A. fluvialis". Acidiparvus is the first described genome-streamlined lineage under the phylum Acidobacteriota, which is a free-living, slow-growing scavenger in freshwater environments.

A genome-scale metabolic model of a globally disseminated hyperinvasive M1 strain of Streptococcus pyogenes.

Streptococcus pyogenes is responsible for a range of diseases in humans contributing significantly to morbidity and mortality. Among more than 200 serotypes of S. pyogenes, serotype M1 strains hold the greatest clinical relevance due to their high prevalence in severe human infections. To enhance our understanding of pathogenesis and discovery of potential therapeutic approaches, we have developed the first genome-scale metabolic model (GEM) for a serotype M1 S. pyogenes strain, which we name iYH543. The curation of iYH543 involved cross-referencing a draft GEM of S. pyogenes serotype M1 from the AGORA2 database with gene essentiality and autotrophy data obtained from transposon mutagenesis-based and growth screens. We achieved a 92.6% (503/543 genes) accuracy in predicting gene essentiality and a 95% (19/20 amino acids) accuracy in predicting amino acid auxotrophy. Additionally, Biolog Phenotype microarrays were employed to examine the growth phenotypes of S. pyogenes, which further contributed to the refinement of iYH543. Notably, iYH543 demonstrated 88% accuracy (168/190 carbon sources) in predicting growth on various sole carbon sources. Discrepancies observed between iYH543 and the actual behavior of living S. pyogenes highlighted areas of uncertainty in the current understanding of S. pyogenes metabolism. iYH543 offers novel insights and hypotheses that can guide future research efforts and ultimately inform novel therapeutic strategies.IMPORTANCEGenome-scale models (GEMs) play a crucial role in investigating bacterial metabolism, predicting the effects of inhibiting specific metabolic genes and pathways, and aiding in the identification of potential drug targets. Here, we have developed the first GEM for the S. pyogenes highly virulent serotype, M1, which we name iYH543. The iYH543 achieved high accuracy in predicting gene essentiality. We also show that the knowledge obtained by substituting actual measurement values for iYH543 helps us gain insights that connect metabolism and virulence. iYH543 will serve as a useful tool for rational drug design targeting S. pyogenes metabolism and computational screening to investigate the interplay between inhibiting virulence factor synthesis and growth.

Liquorilactobacillus: A Context of the Evolutionary History and Metabolic Adaptation of a Bacterial Genus from Fermentation Liquid Environments.

In the present work, we carried out a comparative genomic analysis to trace the evolutionary trajectory of the bacterial species that make up the Liquorilactobacillus genus, from the identification of genes and speciation/adaptation mechanisms in their unique characteristics to the identification of the pattern grouping these species. We present phylogenetic relationships between Liquorilactobacillus and related taxa such as Bacillus, basal lactobacilli and Ligilactobacillus, highlighting evolutionary divergences and lifestyle transitions across different taxa. The species of this genus share a core genome of 1023 genes, distributed in all COGs, which made it possible to characterize it as Liquorilactobacillus sensu lato: few amino acid auxotrophy, low genes number for resistance to antibiotics and general and specific cellular reprogramming mechanisms for environmental responses. These species were divided into four clades, with diversity being enhanced mainly by the diversity of genes involved in sugar metabolism. Clade 1 presented lower (< 70%) average amino acid identity with the other clades, with exclusive or absent genes, and greater distance in the genome compared to clades 2, 3 and 4. The data pointed to an ancestor of clades 2, 3 and 4 as being the origin of the genus Ligilactobacillus, while the species of clade 1 being closer to the ancestral Bacillus. All these traits indicated that the species of clade 1 could be soon separated in a distinct genus.

Comparing genome-scale metabolic models of the non-resistant Enterococcus faecalis ATCC 19433 and the multi-resistant Enterococcus faecalis V583

Enterococcus faecalis is a versatile lactic acid bacterium with a large variety of implications for humans. While some strains of this species are pathobionts being resistant against most of the common antibiotics, other strains are regarded as biological protectants or even probiotics. Accordingly, E. faecalis strains largely differ in the size and content of their accessory genome. In this study, we describe the genome-scale metabolic network reconstruction of E. faecalis ATCC 19433, a non-resistant human-associated strain. A comparison of the genome-scale metabolic model (GSM) of E. faecalis ATCC 19433 with a previously published GSM of the multi-resistant pathobiontic E. faecalis V583 reveals high similarities in the central metabolic abilities of these two human associated strains. This is reflected, e.g., in the identical amino acid auxotrophies. The ATCC 19433 strain, however, has a 14.1 % smaller genome than V583 and lacks the multiple antibiotic resistance genes and genes involved in capsule formation. Based on the measured metabolic fluxes at different growth rates, the energy demand at zero growth was calculated to be about 40 % lower for the ATCC 19433 strain compared to V583. Furthermore, the ATCC 19433 strain seems less prone to the depletion of amino acids utilizable for energy metabolism. This might hint at a lower overall energy demand of the ATCC 19433 strain as compared to V583.

Characterizing a panel of amino acid auxotrophs under auxotrophic starvation conditions.

Auxotrophic strains starving for their cognate nutrient, termed auxotrophic starvation, are characterized by a shorter lifespan, higher glucose wasting phenotype, and inability to accomplish cell cycle arrest when compared to a "natural starvation," where a cell is starving for natural environmental growth-limiting nutrients such as phosphate. Since evidence of this physiological response is limited to only a subset of auxotrophs, we evaluated a panel of auxotrophic mutants to determine whether these responses are characteristic of a broader range of amino acid auxotrophs. Based on the starvation survival kinetics, the panel of strains was grouped into three categories-short-lived strains, strains with survival similar to a prototrophic wild type strain, and long-lived strains. Among the short-lived strains, we observed that the tyrosine, asparagine, threonine, and aspartic acid auxotrophs rapidly decline in viability, with all strains unable to arrest cell cycle progression. The three basic amino acid auxotrophs had a survival similar to a prototrophic strain starving in minimal media. The leucine, tryptophan, methionine, and cysteine auxotrophs displayed the longest lifespan. We also demonstrate how the phenomenon of glucose wasting is limited to only a subset of the tested auxotrophs, namely the asparagine, leucine, and lysine auxotrophs. Furthermore, we observed pleiotropic phenotypes associated with a subgroup of auxotrophs, highlighting the importance of considering unintended phenotypic effects when using auxotrophic strains especially in chronological aging experiments.

Taxonomic and environmental distribution of bacterial amino acid auxotrophies

Many microorganisms are auxotrophic—unable to synthesize the compounds they require for growth. With this work, we quantify the prevalence of amino acid auxotrophies across a broad diversity of bacteria and habitats. We predicted the amino acid biosynthetic capabilities of 26,277 unique bacterial genomes spanning 12 phyla using a metabolic pathway model validated with empirical data. Amino acid auxotrophy is widespread across bacterial phyla, but we conservatively estimate that the majority of taxa (78.4%) are able to synthesize all amino acids. Our estimates indicate that amino acid auxotrophies are more prevalent among obligate intracellular parasites and in free-living taxa with genomic attributes characteristic of ‘streamlined’ life history strategies. We predicted the amino acid biosynthetic capabilities of bacterial communities found in 12 unique habitats to investigate environmental associations with auxotrophy, using data compiled from 3813 samples spanning major aquatic, terrestrial, and engineered environments. Auxotrophic taxa were more abundant in host-associated environments (including the human oral cavity and gut) and in fermented food products, with auxotrophic taxa being relatively rare in soil and aquatic systems. Overall, this work contributes to a more complete understanding of amino acid auxotrophy across the bacterial tree of life and the ecological contexts in which auxotrophy can be a successful strategy.

Amino acid auxotrophies in human gut bacteria are linked to higher microbiome diversity and long-term stability

Amino acid auxotrophies are prevalent among bacteria. They can govern ecological dynamics in microbial communities and indicate metabolic cross-feeding interactions among coexisting genotypes. Despite the ecological importance of auxotrophies, their distribution and impact on the diversity and function of the human gut microbiome remain poorly understood. This study performed the first systematic analysis of the distribution of amino acid auxotrophies in the human gut microbiome using a combined metabolomic, metagenomic, and metabolic modeling approach. Results showed that amino acid auxotrophies are ubiquitous in the colon microbiome, with tryptophan auxotrophy being the most common. Auxotrophy frequencies were higher for those amino acids that are also essential to the human host. Moreover, a higher overall abundance of auxotrophies was associated with greater microbiome diversity and stability, and the distribution of auxotrophs was found to be related to the human host’s metabolome, including trimethylamine oxide, small aromatic acids, and secondary bile acids. Thus, our results suggest that amino acid auxotrophies are important factors contributing to microbiome ecology and host-microbiome metabolic interactions.

Genome-based metabolic and phylogenomic analysis of three Terrisporobacter species.

Acetogenic bacteria are of high interest for biotechnological applications as industrial platform organisms, however, acetogenic strains from the genus Terrisporobacter have hitherto been neglected. To date, three published type strains of the genus Terrisporobacter are only covered by draft genome sequences, and the genes and pathway responsible for acetogenesis have not been analyzed. Here, we report complete genome sequences of the bacterial type strains Terrisporobacter petrolearius JCM 19845T, Terrisporobacter mayombei DSM 6539T and Terrisporobacter glycolicus DSM 1288T. Functional annotation, KEGG pathway module reconstructions and screening for virulence factors were performed. Various species-specific vitamin, cofactor and amino acid auxotrophies were identified and a model for acetogenesis of Terrisporobacter was constructed. The complete genomes harbored a gene cluster for the reductive proline-dependent branch of the Stickland reaction located on an approximately 21 kb plasmid, which is exclusively found in the Terrisporobacter genus. Phylogenomic analysis of available Terrisporobacter genomes suggested a reclassification of most isolates as T. glycolicus into T. petrolearius.

Bacteriophage-mediated lysis supports robust growth of amino acid auxotrophs

Microbial communities host many auxotrophs—organisms unable to synthesize one or more metabolites required for their growth. Auxotrophy is thought to confer an evolutionary advantage, yet auxotrophs must rely on other organisms that produce the metabolites they require. The mechanisms of metabolite provisioning by “producers” remain unknown. In particular, it is unclear how metabolites such as amino acids and cofactors, which are found inside the cell, are released by producers to become available to auxotrophs. Here, we explore metabolite secretion and cell lysis as two distinct possible mechanisms that result in the release of intracellular metabolites from producer cells. We measured the extent to which secretion or lysis of Escherichia coli and Bacteroides thetaiotaomicron amino acid producers can support the growth of engineered Escherichia coli amino acid auxotrophs. We found that cell-free supernatants and mechanically lysed cells provide minimal levels of amino acids to auxotrophs. In contrast, bacteriophage lysates of the same producer bacteria can support as many as 47 auxotroph cells per lysed producer cell. Each phage lysate released distinct levels of different amino acids, suggesting that in a microbial community the collective lysis of many different hosts by multiple phages could contribute to the availability of an array of intracellular metabolites for use by auxotrophs. Based on these results, we speculate that viral lysis could be a dominant mechanism of provisioning of intracellular metabolites that shapes microbial community structure.

A set of vectors and strains for chromosomal integration in fission yeast

The expression of heterologous genes is an important technique in yeast genetics. In fission yeast, the leu1 and ura4 genes have been used mainly as selectable markers for heterologous expression. To expand the repertoire of selection markers available for heterologous expression of genes, here we developed new host-vector systems employing lys1 and arg3. By employing genome editing with the CRISPR/Cas9 system, we isolated several alleles of lys1 and arg3, each having a critical mutation in the ORF region. In parallel, we developed a set of vectors that complement the amino acid auxotrophy of lys1 and arg3 mutants when integrated into each locus. Using these vectors in combination with the previously developed integration vector pDUAL, we successfully observed the localization of three proteins in a cell simultaneously by fusing them with different fluorescent proteins. Thus, these vectors enable combinatorial expression of heterologous genes, which addresses increasingly diverse experimental challenges.

How reliable are Chinese hamster ovary (CHO) cell genome-scale metabolic models?

Genome-scale metabolic models (GEMs) possess the power to revolutionize bioprocess and cell line engineering workflows thanks to their ability to predict and understand whole-cell metabolism in silico. Despite this potential, it is currently unclear how accurately GEMs can capture both intracellular metabolic states and extracellular phenotypes. Here, we investigate this knowledge gap to determine the reliability of current Chinese hamster ovary (CHO) cell metabolic models. We introduce a new GEM, iCHO2441, and create CHO-S and CHO-K1 specific GEMs. These are compared against iCHO1766, iCHO2048, and iCHO2291. Model predictions are assessed via comparison with experimentally measured growth rates, gene essentialities, amino acid auxotrophies, and 13 C intracellular reaction rates. Our results highlight that all CHO cell models are able to capture extracellular phenotypes and intracellular fluxes, with the updated GEM outperforming the original CHO cell GEM. Cell line-specific models were able to better capture extracellular phenotypes but failed to improve intracellular reaction rate predictions in this case. Ultimately, this work provides an updated CHO cell GEM to the community and lays a foundation for the development and assessment of next-generation flux analysis techniques, highlighting areas for model improvements.

Extending the Range of SLIM-Labeling Applications: From Human Cell Lines in Culture to Caenorhabditis elegans Whole-Organism Labeling.

The simple light isotope metabolic-labeling technique relies on the in vivo biosynthesis of amino acids from U-[12C]-labeled molecules provided as the sole carbon source. The incorporation of the resulting U-[12C]-amino acids into proteins presents several key advantages for mass-spectrometry-based proteomics analysis, as it results in more intense monoisotopic ions, with a better signal-to-noise ratio in bottom-up analysis. In our initial studies, we developed the simple light isotope metabolic (SLIM)-labeling strategy using prototrophic eukaryotic microorganisms, the yeasts Candida albicans and Saccharomyces cerevisiae, as well as strains with genetic markers that lead to amino-acid auxotrophy. To extend the range of SLIM-labeling applications, we evaluated (i) the incorporation of U-[12C]-glucose into proteins of human cells grown in a complex RPMI-based medium containing the labeled molecule, considering that human cell lines require a large number of essential amino-acids to support their growth, and (ii) an indirect labeling strategy in which the nematode Caenorhabditis elegans grown on plates was fed U-[12C]-labeled bacteria (Escherichia coli) and the worm proteome analyzed for 12C incorporation into proteins. In both cases, we were able to demonstrate efficient incorporation of 12C into the newly synthesized proteins, opening the way for original approaches in quantitative proteomics.

Cell Type-Specific Metabolic Response to Amino Acid Starvation Dictates the Role of Sestrin2 in Regulation of mTORC1

Targeting cancer metabolism has become one of the strategies for a rational anti-tumor therapy. However, cellular plasticity, driven by a major regulator of cellular growth and metabolism, mTORC1, often leads toward treatment resistance. Sestrin2, a stress-inducible protein, has been described as an mTORC1 inhibitor upon various types of stress signals. Immune assays and online measurements of cellular bioenergetics were employed to investigate the nature of Sestrin2 regulation, and finally, by silencing the SESN2 gene, to identify the role of induced Sestrin2 upon a single amino acid deprivation in cancer cells of various origins. Our data suggest that a complex interplay of either oxidative, energetic, nutritional stress, or in combination, play a role in Sestrin2 regulation upon single amino acid deprivation. Therefore, cellular metabolic background and sequential metabolic response dictate Sestrin2 expression in the absence of an amino acid. While deprivations of essential amino acids uniformly induce Sestrin2 levels, non-essential amino acids regulate Sestrin2 differently, drawing a characteristic Sestrin2 expression fingerprint, which could serve as a first indication of the underlying cellular vulnerability. Finally, we show that canonical GCN2-ATF4-mediated Sestrin2 induction leads to mTORC1 inhibition only in amino acid auxotroph cells, where the amino acid cannot be replenished by metabolic reprogramming.

Communities of Niche-optimized Strains (CoNoS) – Design and creation of stable, genome-reduced co-cultures

Current bioprocesses for production of value-added compounds are mainly based on pure cultures that are composed of rationally engineered strains of model organisms with versatile metabolic capacities. However, in the comparably well-defined environment of a bioreactor, metabolic flexibility provided by various highly abundant biosynthetic enzymes is much less required and results in suboptimal use of carbon and energy sources for compound production. In nature, non-model organisms have frequently evolved in communities where genome-reduced, auxotrophic strains cross-feed each other, suggesting that there must be a significant advantage compared to growth without cooperation. To prove this, we started to create and study synthetic communities of niche-optimized strains (CoNoS) that consists of two strains of the same species Corynebacterium glutamicum that are mutually dependent on one amino acid. We used both the wild-type and the genome-reduced C1* chassis for introducing selected amino acid auxotrophies, each based on complete deletion of all required biosynthetic genes. The best candidate strains were used to establish several stably growing CoNoS that were further characterized and optimized by metabolic modelling, microfluidic experiments and rational metabolic engineering to improve amino acid production and exchange. Finally, the engineered CoNoS consisting of an l-leucine and l-arginine auxotroph showed a specific growth rate equivalent to 83% of the wild type in monoculture, making it the fastest co-culture of two auxotrophic C. glutamicum strains to date. Overall, our results are a first promising step towards establishing improved biobased production of value-added compounds using the CoNoS approach.

Dynamic proteome allocation regulates the profile of interaction of auxotrophic bacterial consortia.

Microbial ecosystems are composed of multiple species in constant metabolic exchange. A pervasive interaction in microbial communities is metabolic cross-feeding and occurs when the metabolic burden of producing costly metabolites is distributed between community members, in some cases for the benefit of all interacting partners. In particular, amino acid auxotrophies generate obligate metabolic inter-dependencies in mixed populations and have been shown to produce a dynamic profile of interaction that depends upon nutrient availability. However, identifying the key components that determine the pair-wise interaction profile remains a challenging problem, partly because metabolic exchange has consequences on multiple levels, from allocating proteomic resources at a cellular level to modulating the structure, function and stability of microbial communities. To evaluate how ppGpp-mediated resource allocation drives the population-level profile of interaction, here we postulate a multi-scale mathematical model that incorporates dynamics of proteome partition into a population dynamics model. We compare our computational results with experimental data obtained from co-cultures of auxotrophic Escherichia coli K12 strains under a range of amino acid concentrations and population structures. We conclude by arguing that the stringent response promotes cooperation by inhibiting the growth of fast-growing strains and promoting the synthesis of metabolites essential for other community members.

Effect of Non-Lethal Selection on Spontaneous Revertants of Frameshift Mutations: The Escherichia coli hisF Case

Microorganisms possess the potential to adapt to fluctuations in environmental parameters, and their evolution is driven by the continuous generation of mutations. The reversion of auxotrophic mutations has been widely studied; however, little is known about the reversion of frameshift mutations resulting in amino acid auxotrophy and on the structure and functioning of the protein encoded by the revertant mutated gene. The aims of this work were to analyze the appearance of reverse mutations over time and under different selective pressures and to investigate revertant enzymes’ three-dimensional structures and their correlation with a different growth ability. Escherichia coli FB182 strain, carrying the hisF892 single nucleotide deletion resulting in histidine auxotrophy, was subjected to different selective pressures, and revertant mutants were isolated and characterized. The obtained results allowed us to identify different indels of different lengths located in different positions in the hisF gene, and relations with the incubation time and the selective pressure applied were observed. Moreover, the structure of the different mutant proteins was consistent with the respective revertant ability to grow in absence of histidine, highlighting a correlation between the mutations and the catalytic activity of the mutated HisF enzyme.

The arginine deaminase system plays distinct roles in Borrelia burgdorferi and Borrelia hermsii.

Borrelia species are amino acid auxotrophs that utilize di- and tri- peptides obtained through their oligopeptide transport system to supply amino acids for replicative growth during their enzootic cycles. However, Borrelia species from both the Lyme disease (LD) and relapsing fever (RF) groups harbor an amino acid transport and catabolism system, the Arginine Deiminase System (ADI), that could potentially augment intracellular L-arginine required for growth. RF spirochetes contain a “complete”, four gene ADI (arcA, B, D, and C) while LD spirochetes harbor arcA, B, and sometimes D but lack arcC (encoding carbamate kinase). In this study, we evaluated the role of the ADI system in bacterial survival and virulence and discovered important differences in RF and LD ADIs. Both in vitro and in a murine model of infection, B. hermsii cells significantly reduced extracellular L-arginine levels and that reduction was dependent on arginine deiminase expression. Conversely, B. burgdorferi did not reduce the concentration of L-arginine during in vitro growth experiments nor during infection of the mammalian host, suggesting a fundamental difference in the ability to directly utilize L-arginine compared to B. hermsii. Further experiments using a panel of mutants generated in both B. burgdorferi and B. hermsii, identified important differences in growth characteristics and ADI transcription and protein expression. We also found that the ADI system plays a key role in blood and spleen colonization in RF spirochetes. In this study we have identified divergent metabolic strategies in two closely related human pathogens, that ultimately impacts the host-pathogen interface during infection.

Reactor control system in bacterial co-culture based on fluorescent proteins using an Arduino-based home-made device.

Co-culture, fermentation with more than two microbial strains, is a potential flexible method for optimizing the metabolic conversion process in bio-production. However, maintaining an ideal population throughout the fermentation process remains a challenge. In this study, we developed a proportional control system for controlling the population ratio of Escherichia coli strains to a set value during continuous co-culture. Two E. coli strains were distinguished by expressing different fluorescent proteins, and their population ratio was determined by culture fluorescence. Furthermore, different types of amino acid auxotrophs were provided to each strain, and among these, growth was controlled by the amino acid concentrations in the feed medium. An Arduino-based device was developed using light-emitting diodes and cadmium sulfide light sensors for the in-line monitoring of culture fluorescence. Two E. coli strains of methionine auxotroph green fluorescent protein (GFP) expressing (met-GFP) strain and arginine auxotroph red fluorescent protein (RFP) expressing (arg-RFP) strain were co-cultured using a jar-fermenter. The amounts of methionine and arginine in the feed medium were altered to guide the population ratio to a set value. During the continuous culture, the population ratio between the met-GFP and arg-RFP strains was successfully maintained at approximately the setpoint values. This study demonstrated the development of a home-made device for controlling the reactor of E. coli based on fluorescent proteins using inexpensive parts.

Drawing up a collaborative contract: Amino acid cross-feeding between interspecies bacterial pairs.

Synthetic microbial communities have the potential to enable new platforms for bioproduction of biofuels and biopharmaceuticals. However, using engineered communities is often assumed to be difficult because of anticipated challenges in establishing and controlling community composition. Cross-feeding between microbial auxotrophs has the potential to facilitate coculture growth and stability through a mutualistic ecological interaction. We assessed cross-feeding between 13 Escherichia coli amino acid auxotrophs paired with a leucine auxotroph of Bacillus megaterium. We developed a minimal medium capable of supporting the growth of both bacteria and used the media to study coculture growth of the 13 interspecies pairs of auxotrophs in batch and continuous culture, as well as on semi-solid media. In batch culture, 8 of 13 pairs of auxotrophs were observed to grow in coculture. We developed a new metric to quantify the impact of cross-feeding on coculture growth. Six pairs also showed long-term stability in continuous culture, where coculture growth at different dilution rates highlighted differences in cross-feeding amongst the pairs. Finally, we found that cross-feeding-dependent growth on semi-solid media is highly stringent and enables identification of the most efficient pairs. These results demonstrate that cross-feeding is a viable approach for controlling community composition within diverse synthetic communities.

Metabolic patterns reveal enhanced anammox activity at low nitrogen conditions in the integrated I-ABR.

Substrate concentrations greatly influence bacterial growth and metabolism. However, optimal nitrogen concentrations for anammox bacteria in nitrogen-limited environments remain unclear. Here, we observed enhanced nitrogen metabolism and anabolism of anammox bacteria at low nitrogen conditions. Efficient nitrogen removal was achieved at ammonium and nitrite influent concentration of 30mg/L under HRT of 1hr, with an average nitrogen removal rate (NRR) of 0.73kgN/(m3 ·day) in I-ABR composed of four compartments. The highest anammox activity of 6.25mmolN/ (gVSS·hr) was observed in the fourth compartment (C4) with the lowest substrate levels (ammonium and nitrite of 11.6mg/L and 7mg/L). This could be resulted from the highest expression level of genes involved in nitrogen metabolism in C4, which was 1.49-1.67 times higher than that in other compartments. Besides, the second compartment (C2) exhibited the most active anabolism at ammonium and nitrite of 17mg/L and 13mg/L, respectively, which contributed to the most active amino acid synthesis and thus the highest EPS (1.35 times higher) in C2. This enhanced amino acid auxotrophy between anammox bacteria with heterotrophs, and consequently, heterotrophs thrived and competed for nitrite. These results hint at the potential application of anammox process in micro-polluted water. PRACTITIONER POINTS: High nitrogen removal and efficient biomass retention at low nitrogen concentrations under short HRT was achieved in I-ABR. Optimal concentrations for anammox nitrogen removal and anabolism were discussed under low nitrogen concentrations. More active anabolism contributed to enhanced amino acid synthesis and thus higher EPS contents. Low substrate levels led to enhanced expression of genes involved in nitrogen metabolism and thus high anammox activity.