**1. Introduction**

*Chloropseudomonas ethylica* was originally described as a motile green sulfur bacterium capable of utilizing ethanol [1]. However, no pure cultures of green sulfur bacteria were previously known that would utilize ethanol or that were motile. Strains of *Cp. ethylica* were isolated from the Krujalnik

Estuary, near Odesa, Russia (strain 2-K) and from Lake Saksky in Crimea, Russia [2], which in hindsight is a remarkable coincidence considering the approximately 200-mile distance between these habitats. However, doubts were raised about the purity of the strains when two di fferent cell morphologies were observed in growing cultures [2,3]. Subsequently, the colorless, motile, ethanol-oxidizing, and sulfur-reducing bacterium *Desulfuromonas acetoxidans* strains DSM 1675 and DSM 1676 were isolated from either the 2-K or the N2 syntrophic mixtures [4]. A pure culture of *Ds. acetoxidans* DSM 684<sup>T</sup> was also isolated from South Orkney Island, Antarctica [4]. The green component was either described as a species of *Chlorobium* [2] or *Prosthecochloris* [3]. The amino acid sequences of cytochrome C-555, alternatively known as c5, from the *Cp. ethylica* 2-K mixture was clearly related to those of *Prosthecochloris* species and not to *Chlorobium* [5], whereas the cytochrome C-551.5, alternatively known as c7 [6] came from the *Desulfuromonas* component [7]. It is unclear whether there is a specificity to the connection between the two species in the *Cp. ethylica* mixture. Agar shake cultures with ethanol, sulfide, and bicarbonate show large green colonies (faster-growing when involved in sulfur cycling) and small green colonies. The large ones were verified to be in mixed culture and the smaller ones as a pure culture [8] and our unpublished observations.

Mutualistic relationships that involve close cell–cell interactions are most studied between bacterial and eukaryotic interactions, e.g., between nitrogen-fixing *Rhizobium* species and legumes, or bacterial pathogens and eukaryotic hosts. Symbiotic interactions amongs<sup>t</sup> archaea and bacteria can be found in microbial mats where nutrient exchange and waste removal roles are crucial, in anaerobic methane-oxidizing communities of marine environments or in human digestive systems [9–13]. These interactions have only recently been studied in more detail and appear to be more common than historically expected. Larger bacterial and archaeal consortia that are formed through cell–cell interactions of two or more microorganisms have been observed to form a high degree of interdependence between taxonomically unrelated species [14–16]. Phototrophic consortia of this sort were first reported over a century ago [17]. The nature of the *Cp. ethylica* syntrophy appears to be centered around a closed sulfur cycle [18], similar to what has been shown in syntrophic cocultures of *Chlorobium phaeovibrioides* and *Desulfuromonas acetoxidans*, where acetate is oxidized by *Ds. acetoxidans*, with sulfur as an electron acceptor [19]. The process leads to the recycling of the sulfide that can then be used for anoxygenic photosynthesis by *Cb. phaeovibrioides*. Although these syntrophic cocultures appear to be more widespread than commonly expected and the nutrient cycle and mutual benefits are clear in some cases, very little is known about the physical interaction and formation of these cell–cell interactions and the specific components of the electron chemistry involved to establish such a mutually beneficial nutrient cycle. Studies with larger phototrophic consortia, such as *Chlorochromatium aggregatum*, have shown that the green sulfur bacteria involved in this complex are likely preadapted to a symbiotic lifestyle, and specific ultrastructures (periplasmic extruding tubules) can be formed between the central bacterium and the epibiont [14,20]. *Cp. ethylica* was not described to form larger aggregates and appears to form a simpler model of syntrophy. However, the formation of such larger consortia, even in *Chlorochromatium aggregatum*, is dependent on the cultivation strategy. To gain further insight into the possible physical interactions and the electron transfer mechanism involved in *Cp. ethylica*, we set out to determine the genomes of the syntrophic mixtures.

We now report the simultaneous determination of the genome sequences of the green and colorless components of the *Cp. ethylica* N3 mixture. Although the genome sequence of *Desulfuromonas acetoxidans* DSM 684<sup>T</sup> was previously determined [21], we have now also established the genome sequences of the *Desulfuromonas* strains 2-K (DSM 1675) and N2 (DSM 1676), previously isolated from two of the *Cp. ethylica* mixtures. We attempted to simultaneously determine the sequences of both organisms from strain N2, but obtained only the genome belonging to the green sulfur bacterial genome, likely due to slightly di fferent culture conditions that may have led to a lower abundance of the colorless component in the mixture.

#### **2. Materials and Methods**

#### *2.1. Cultures and DNA Preparation*

Cultures of *Chloropseudomonas ethylica* N3 and N2 were originally obtained by one of us (Terry E. Meyer) directly from the laboratory of E.N. Kondrat'eva, and had been grown in our laboratory over the years and kept in a lyophilized state. Cultures were grown according to [8] and harvested by centrifugation. DNA extracted from decades-old frozen cultures of the N3 and N2 syntrophic mixtures were the source for the genomic analysis presented here. Genomic DNA was extracted using the GeneJET DNA purification kit (Thermo Scientific, Waltham, MA, USA). The quantity and purity of DNA, determined using Qubit and Nanodrop instruments, showed an A260/280 ratio of 1.75.

Genomic DNA for *Prosthecochloris* sp. DSM 1685, *Desulfuromonas acetoxidans* DSM 1675 and DSM 1676 was obtained from DSMZ (Deutsche Sammlung von Mikroorganismen und Zellkulturen, GmbH). The quantity and purity of DNA, determined using Qubit and Nanodrop instruments, showed A260/280 ratios between 1.8 and 1.9.

#### *2.2. DNA Sequencing, Assembly, and Annotation*

The DNA libraries were prepared with the Nextera DNA Flex Library Prep Kit (Illumina, Inc., San Diego, CA, USA). All five genomes were sequenced, using 500 μL of a 1.8 pM library, with an Illumina MiniSeq instrument, using paired-end sequencing (2 × 150 bp). Quality control of the reads was performed using FASTQC within BaseSpace (Illumina, version 1.0.0), using a k-mer size of 5 and contamination filtering. The data for each of the genomes of *Chloropseudomonas ethylica* N3 and N2, *Prosthecochloris* sp. DSM 1685, and *Desulfuromonas acetoxidans* DSM 1675 and DSM 1676 was assembled *de novo* using Unicycler within PATRIC (Pathosystems Resource Integration Center) [22]. The genome sequences were annotated using RAST (Rapid Annotations using Subsystem Technology; version 2.0) [23].

Average percentage nucleotide identity (ANIb) between the whole genomes was calculated using JSpecies [24]. A whole-genome-based phylogenetic tree was generated applying the CodonTree method within PATRIC [22], which used PGFams as homology groups. Moreover, 519 PGFams were found among these selected genomes using the CodonTree analysis to construct the *Prosthecochloris* tree, while 598 unique PGFams were identified for the *Desulfuromonas* tree. The aligned proteins and coding DNA from single-copy genes were used for RAxML analysis [25,26] for the trees. The support values for the phylogenetic tree were generated using 100 rounds of the "Rapid bootstrapping" option of RaxML. Tree visualization was performed with iTOL [27].

This Whole-Genome Shotgun project has been deposited at DDBJ/ENA/GenBank, and the accession numbers for all of the sequenced genomes are listed in Table 1.

## *2.3. Metagenomic Binning*

The sequencing reads of *Cp. ethylica* N3 were used to perform a metagenomic binning using the Metagenomic Binning service within PATRIC [22]. Paired-end reads were used as input, and default parameters were used. Sets of contig bins were constructed, with hits against contigs that have less than fourfold coverage or are less than 400 bp in length being removed. The contig pool was split into bins using reference genomes. Quality control of each bin was performed using checkM [28]. Each bin was automatically annotated using RAST within PATRIC [22], and consistency checks of the annotation were performed, producing a coarse score (percentage of roles that are correctly present or absent) and a fine score (percentage of roles that are correctly absent or present in the correct number). Identified genomes were ranked based on their coarse score, fine score, and completeness.

## *2.4. Synteny Analysis*

For synteny analysis, comparative genome regions were generated using global PATRIC PGFam families to determine a set of genes that match a focus gene. All genomes were used in the search and compared to the reference genome. The gene set was compared to the focus gene using BLAST and sorted by BLAST scores within PATRIC [22]. The *Prosthecochloris ethylica* N3 TadZ/CpaE (PGFam\_00109911) and agglutination protein (PGFam\_02064367) were used as focus genes to analyze synteny of the Tad pili and adhesion protein gene clusters, respectively. For the *Desulfuromonas acetoxidans* synteny comparisons, the Type IV major assembly protein PilA (PGFam\_00056426) and cytochrome C-551.5 (PGFam\_10701576) were used as focus genes.
