Effects of Co-existing Heterotrophs on Physiology of and Nitrogen Metabolism in Autotrophic Nitrite-oxidizing Candidatus Nitrotoga

Interactions between autotrophic nitrifiers and heterotrophs have attracted considerable attention in microbial ecology. However, the mechanisms by which heterotrophs affect the physiological activity of and nitrogen metabolism in autotrophic nitrite oxidizers remain unclear. We herein focused on nitrite-oxidizing Candidatus Nitrotoga and compared an axenic culture including only Ca. Nitrotoga with a co-culture of both Ca. Nitrotoga and Acidovorax in physiological experiments and transcriptomics. In the co-culture with Acidovorax, nitrite consumption by Ca. Nitrotoga was promoted, and some genes relevant to nitrogen metabolism in Ca. Nitrotoga were highly expressed. These results provide insights into the mechanisms by which co-existing heterotrophs affect autotrophic nitrifiers.


Nitrite consumption of AM1P in low cell density (axenic culture vs coculture)
AM1P and AM1P + NB1 cultures were incubated in 96-well microtiter plates in the dark and static conditions.Cell suspensions in 150 µL of nitrite (2.1 mM) inorganic medium were inoculated into each well.In both axenic culture (only AM1P cells) and coculture (AM1P + NB1 cells), cell densities of AM1P were prepared with 10 2 -10 4 cells mL -1 .Cell densities of NB1 were equivalent to those of AM1P in coculture.On days 20 and 46, nitrite concentrations were measured in each well.Cell growth of AM1P was observed with fluorescence in situ hybridization (FISH) microscopic analysis.

Measurement of nitrite and nitrate concentrations
Nitrite concentration was determined using Griess reagent (Griess-Romijn van Eck, 1966).Absorbance at a wavelength of 550 nm was measured using a Powerscan HT multiplate reader (DS Pharma Biomedical, Osaka, Japan).Additionally, nitrite and nitrate concentrations were measured using an IC-2010 ion chromatography system (Tosoh Co. Tokyo, Japan).
Growth activity of AM1P, Culture for RNA sequencing (axenic culture vs coculture) In the first trial, axenic culture (AM1P; 1.0×10 5 cell mL -1 ) and cocultures suspended in different NB1 cell densities (AM1P; 1.0×10 5 cell mL -1 , NB1; 10 4 -10 7 cells mL -1 ) were cultivated in the inorganic medium containing 2.0 mM nitrite for approximately two weeks.In the second trial, axenic culture (AM1P; 5.0×10 5 cells mL -1 ) and coculture (AM1P; 5.0×10 5 cells mL -1 and NB1; 5.0×10 4 cells mL -1 ) were cultivated in the inorganic medium containing 3.6 mM nitrite as pre-cultivation at 16°C in the static and dark condition.When the nitrite was completely consumed, the pre-cultured samples were transferred to the fresh inorganic medium containing 3.6 mM nitrite, in which AM1P cells were prepared as 5.0×10 5 cells mL -1 .Nitrite concentration and cell density were measured every 2 or 3 days with ion chromatography and microscopy, respectively.
For RNA sequencing, AM1P cultures were incubated under starvation condition during three days.For pre-cultivation, axenic culture (AM1P; 5.0×10 5 cells mL -1 ) and coculture (AM1P; 5.0×10 5 cells mL -1 and NB1; 5.0×10 5 cells mL -1 ) were cultivated in the inorganic medium containing 4 mM nitrite as 1.3 L volume at 16°C in the static and dark condition.When the nitrite was completely consumed, a part of the pre-cultured samples was transferred to the fresh inorganic medium containing 4 mM nitrite for subsequent cultivation.Three replicate samples of axenic culture (only AM1P) and coculture (AM1P + NB1) were collected in the exponential phase in less than 3 mM of nitrite concentration on days 4 and 3, respectively (Fig. S3).The samples were harvested by filtering (pore size 0.22 µm) at 4°C.The pellets on the filters were suspended in the fresh inorganic medium.The suspensions were centrifuged (2,900 × g, 10 min) to remove the supernatant.Finally, the pellets were preserved at −80°C.

Microscopic observation and estimation of cell density
Cell densities of AM1P and NB1 were estimated with direct counting under a fluorescent microscopy (Axioskop 2 plus: Carl Zeiss, Oberkochen, Germany).Cells of AM1P and NB1 were specifically detected based on FISH method and nucleotide acid staining.FISH was conducted according to a previous protocol (Amann et al., 1990).As a Ca.Nitrotoga-specific FISH probe, NTG840 (5' CTA AGG AAG TCT CCT CCC 3') was used (Alawi et al., 2007).As nucleotide acid staining, SYTOX green (Invitrogen) was used.

RNA extraction and transcriptome sequencing
Total RNA of the cell pellets preserved at −80°C was extracted using RNeasy mini kit (Qiagen).
The paired-end transcriptome sequence reads were aligned to open reading frames (ORFs) deposited at Genbank for Ca.Nitrotoga sp.AM1P (GenBank: AP019547.1)reference genome and Acidovorax sp.NB1 (GenBank: BJHU00000000) reference genome using Bowtie2 (version 2.4.1).Based on the mapping files, read counts including CDS annotation region of AM1P and NB1 were generated using featureCounts.The mapped read count was normalized as Relative log expression (RLE) to compare between samples using DESeq2.Based on student t-test, log2 fold change between AM1P genes under two culture conditions was calculated.For NB1, normalization based on transcript per million (TPM) was conducted using the following formula.
Yt : read count mapped to transcript t; Lt: length of transcript t (bp); Tt: read number per 1,000 bp of transcript t; TPMt: Gene expression level of transcript t per total transcript 1 million.

Figure legends
Fig. S1 Number of wells showing consumption of more than 0.21 mM nitrite.