Strain-level genetic diversity of Methylophaga nitratireducenticrescens confers plasticity to denitrification capacity in a methylotrophic marine denitrifying biofilm

Background The biofilm of a methanol-fed, fluidized denitrification system treating a marine effluent is composed of multi-species microorganisms, among which Hyphomicrobium nitrativorans NL23 and Methylophaga nitratireducenticrescens JAM1 are the principal bacteria involved in the denitrifying activities. Strain NL23 can carry complete nitrate (NO\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}${}_{3}^{-}$\end{document}3−) reduction to N2, whereas strain JAM1 can perform 3 out of the 4 reduction steps. A small proportion of other denitrifiers exists in the biofilm, suggesting the potential plasticity of the biofilm in adapting to environmental changes. Here, we report the acclimation of the denitrifying biofilm from continuous operating mode to batch operating mode, and the isolation and characterization from the acclimated biofilm of a new denitrifying bacterial strain, named GP59. Methods The denitrifying biofilm was batch-cultured under anoxic conditions. The acclimated biofilm was plated on Methylophaga specific medium to isolate denitrifying Methylophaga isolates. Planktonic cultures of strains GP59 and JAM1 were performed, and the growth and the dynamics of NO\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}${}_{3}^{-}$\end{document}3−, nitrite (NO\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}${}_{2}^{-}$\end{document}2−) and N2O were determined. The genomes of strains GP59 and JAM1 were sequenced and compared. The transcriptomes of strains GP59 and JAM1 were derived from anoxic cultures. Results During batch cultures of the biofilm, we observed the disappearance of H. nitrativorans NL23 without affecting the denitrification performance. From the acclimated biofilm, we isolated strain GP59 that can perform, like H. nitrativorans NL23, the complete denitrification pathway. The GP59 cell concentration in the acclimated biofilm was 2–3 orders of magnitude higher than M. nitratireducenticrescens JAM1 and H. nitrativorans NL23. Genome analyses revealed that strain GP59 belongs to the species M. nitratireducenticrescens. The GP59 genome shares more than 85% of its coding sequences with those of strain JAM1. Based on transcriptomic analyses of anoxic cultures, most of these common genes in strain GP59 were expressed at similar level than their counterparts in strain JAM1. In contrast to strain JAM1, strain GP59 cannot reduce NO\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}${}_{3}^{-}$\end{document}3− under oxic culture conditions, and has a 24-h lag time before growth and NO\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}${}_{3}^{-}$\end{document}3− reduction start to occur in anoxic cultures, suggesting that both strains regulate differently the expression of their denitrification genes. Strain GP59 has the ability to reduce NO\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}${}_{2}^{-}$\end{document}2− as it carries a gene encoding a NirK-type NO\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}${}_{2}^{-}$\end{document}2− reductase. Based on the CRISPR sequences, strain GP59 did not emerge from strain JAM1 during the biofilm batch cultures but rather was present in the original biofilm and was enriched during this process. Discussion These results reinforce the unique trait of the species M. nitratireducenticrescens among the Methylophaga genus as facultative anaerobic bacterium. These findings also showed the plasticity of denitrifying population of the biofilm in adapting to anoxic marine environments of the bioreactor.


219
220 NirK Phylogenetic analysis 221 The deduced amino acid sequence of the GP59 nirK was compared with public protein 222 databases by BLASTP at the NCBI web site to retrieve the closest NirK sequences. These 223 sequences were than aligned with Cobalt (33). Evolutionary analysis was conducted using   (Fig. 2a). The growth yields reached about 1.2 OD 600nm with 42.8 mM 301 NO 3exposure and did not increase significantly in cultures exposed with higher NO 3 − 302 concentrations. Strain JAM1 cultured under the same conditions showed no lag phase, and 303 growth yields were 3-to 6-fold lower than those of the GP59 cultures (Fig. 2a). The maximum 304 specific growth rate (µmax) and the half-saturation constants of NO 3 − for growth (Ks) were 305 calculated ( Table 2). The µmax for strain GP59 is higher (3.3 times) than that of strain JAM1. To 306 assess the affinity of strain GP59 toward NO 3 − for growth, the μmax/Ks ratio was calculated (38) 307 (Table 2). This ratio (1.2 µM -1 NO 3 − h -1 ) is not different than the one calculated for strain JAM1 308 at 1.3 µM -1 NO 3 − h -1 (19). These results concur with the genome sequences (see below) with near 309 100% identity between the two strains in gene clusters encoding the two Nar systems and the 310 NO 3 − transporters (NarK).

311
As observed with growth, a 24h-lag period was observed in the GP59 cultures before NO 3 − 312 started to be consumed, whereas NO 3 − consumption appeared in the JAM1 cultures within 24 h 313 (Fig. 2b). The NO 3 − reduction rates increased linearly with the increase of NO 3 − concentrations 314 in the GP59 cultures (Fig. 3a). The specific NO 3 − reduction rates (rates normalized by the 315 biomass) averaged around 1.5 to 2 mM NO 3 − h -1 OD -1 and showed no significant changes at any 316 NO 3 − concentrations tested (Fig. 3b). In the JAM1 cultures, the NO 3 − reduction rates reached a 317 plateau at 24 mM NO 3 − (Fig. 3a), and, as observed with the GP59 cultures, showed no significant 318 changes in the specific NO 3 − reduction rates (averaged around 4 to 6 mM NO 3 − h -1 OD -1 ) ( Fig.   319 3b). Interestingly, these specific rates are 2.4 to 5.4-fold higher than those of strain GP59, 320 suggesting that the JAM1 cells have a higher dynamism of NO 3 − processing (e.g. NO 3intake 321 and reduction) than the GP59 cells. Close examination of the respective genomes and 322 transcriptomes (see below) did not reveal specific gene(s) that would explain these differences.

323
N 2 O was detected in the GP59 anoxic cultures and reached maximum accumulation (ca. 324 0.5 µmol N 2 O vial -1 or 0.04% N-input) when NO 2 − peaked in the medium, and then decreased in 325 concentration afterwards (Fig. 2b). This result concurs with the presence of a complete 326 denitrification pathway in strain GP59 as revealed by its genome sequence (see below).

333
Both strains cannot grow on methylamine and fructose. They showed similar growth 334 profile related to NaCl concentrations (optimal growth at 1 to 5% NaCl; weaker growth at 8%; 335 no growth at 0 and 10%), pH (optimal growth at pH 8; weaker growth at pH 7, 9 and 10; no 336 growth at pH 6 and 11) and temperature (optimal growth at 30°C; variability in culture replicates

427
The CRISPR region in GP59 and JAM1 genomes comprises 5 associated CRISPR genes 428 (Fig. 5a). Except for the spacer sequences, the nucleotide sequence of the CRISPR genes and the 429 repeat sequences are identical in both genomes (Fig. 5a, c)  Manuscript to be reviewed