Elioraea tepida, sp. nov., a Moderately Thermophilic Aerobic Anoxygenic Phototrophic Bacterium Isolated from the Mat Community of an Alkaline Siliceous Hot Spring in Yellowstone National Park, WY, USA

Strain MS-P2T was isolated from microbial mats associated with Mushroom Spring, an alkaline siliceous hot spring in Yellowstone National Park, WY, USA. The isolate grows chemoheterotrophically by oxygen-dependent respiration, and light stimulates photoheterotrophic growth under strictly oxic conditions. Strain MS-P2T synthesizes bacteriochlorophyll a and the carotenoid spirilloxanthin. However, photoautotrophic growth did not occur under oxic or anoxic conditions, suggesting that this strain should be classified as an aerobic anoxygenic phototrophic bacterium. Strain MS-P2T cells are motile, curved rods about 0.5 to 1.0 μm wide and 1.0 to 1.5 μm long. The optimum growth temperature is 45–50 °C, and the optimum pH for growth is circum-neutral (pH 7.0–7.5). Sequence analysis of the 16S rRNA gene revealed that strain MS-P2T is closely related to Elioraea species, members of the class Alphaproteobacteria, with a sequence identity of 96.58 to 98%. The genome of strain MS-P2T is a single circular DNA molecule of 3,367,643 bp with a mol% guanine-plus-cytosine content of 70.6%. Based on phylogenetic, physiological, biochemical, and genomic characteristics, we propose this bacteriochlorophyll a-containing isolate is a new species belonging to the genus Elioraea, with the suggested name Elioraea tepida. The type-strain is strain MS-P2T (= JCM33060T = ATCC TSD-174T).


Introduction
Aerobic anoxygenic phototrophic bacteria (AAPB) are bacteria that grow chemoheterotrophically by oxygen-dependent respiration, but they also synthesize photosynthetic reaction centers and light-harvesting complexes containing bacteriochlorophyll (BChl) a under oxic conditions [1,2]. Although AAPB can produce a photosynthetic apparatus, AAPB are unable to reduce carbon dioxide; light usually stimulates growth slightly under chemoorganoheterotropic conditions. Many species of AAPB have been found in a variety of habitats, including oceans, freshwater lakes and rivers, and soils [3]. Although most Based on its phenotypic and phylogenetic characteristics, we propose that this isolate represents a new species within the genus Elioraea with the suggested name, Elioraea tepida.

Strain Isolation and Cultivation Conditions
'Chloracidobacterium thermophilum Midnight medium' (CTM medium) [21,22] supplemented with 0.8% (w/v) agar (pH 7.0) was used as the basal medium for the isolation. A microbial mat sample collected from Mushroom Spring in the Lower Geyser Basin of Yellowstone National Park, WY, USA (GPS coordinates: Lat.: 44.5387, Long.: -110.798) was inoculated into sterile liquid CTM medium, and the culture was incubated in a beaker covered with aluminum-foil under continuous illumination from an incandescent light (approx. 20-50 µmol photons m −2 s −1 ) at 50 • C for >10 days. Diluted samples from the enrichment culture were mixed with liquified agar prepared with CTM medium at 45 • C (tempered agar), and the agar mixtures were then poured into Petri dishes. After cooling to solidify the agar, the plates were incubated at 50 • C. Pink colonies that formed in the solidified agar medium within a week were picked up with a sterilized Pasteur pipet, resuspended into 1.0 mL of sterile distilled water, and mixed with tempered agar in CTM medium, cooled to solidify the agar, and incubated at 50 • C. This procedure was repeated until axenic cultures were obtained. Liquid cultivation was also successful in basal CTM medium containing succinate, acetate, and yeast extract (1.0 g L -1 of each). The axenic cultures could be maintained as 60-mL cultures in 100-mL Erlenmeyer flasks covered with aluminum foil, with or without shaking (150 rpm) at 50 • C (see Supplementary Figure S1A).

Microscopy and Analytical Procedures
The size and shape of the cells of strain MS-P2 T were determined by phase-contrast microscopy. Autofluorescence from BChl a of cells of strain MS-P2 T was visualized under an epifluorescence microscope (Nikon Eclipse E600 (NIS-Elements software D), Nikon Xenon power supply XPS-100, monochromatic CMOS camera (Orcaflash 4.0, Hamamatsu), Filterset Excitation: 350-510 nm dichroic mirror:665 nm/Emission: 830 nm LP (SEMROCK, Rochester, NY, USA)).
Absorbance spectra were measured with a Shimadzu model UV-1800 spectrophotometer (Shimadzu Corporation, Kyoto, Japan). Pigments were extracted with acetone-methanol (7:2, v/v), and the absorbance spectra of extracts were measured to assess in a preliminary manner the pigmentation of cells. Extracted pigments were also analyzed by reversedphase, high-performance liquid chromatography (HPLC) as described [23]. Pigments were extracted with acetone-methanol (7:2, v/v) and were filtered with a 0.2-µm polytetrafluoroethylene, single-use filter device (Cytiva, Marlborough, MA, USA) prior to injection into the column (25 cm × 4.6 mm MilliporeSigma™ Supelco™ Discovery™ 5-µm C-18 column; Fisher Scientific, Hampton, NH, USA). The filtered pigment solution was analyzed with an Agilent 1100 HPLC system equipped with a diode-array detector (Santa Clara, CA, USA), as previously described [24]. The identities of pigments were determined by comparison of elution times and in-line absorbance spectra to those of authenticated standards that are maintained in the laboratory.
Gram-staining was performed by using the Favor-G kit (Nissui Pharmaceutical, Tokyo, Japan). Production of catalase was assessed by measuring bubble production in 3% (v/v) H 2 O 2 , and cytochrome c oxidase activity was assessed by using the oxidase reagent (bioMérieux, Marcy l'Etoile, France). The presence of other selected enzymes was tested using the APIZYM system (bioMérieux, Marcy l'Etoile, France). The ability to oxidize a variety of substrates was tested using the Biolog GEN III plate system (Biolog, Hayward, CA, USA). The Biolog Gen III system tests 71 carbon substrates and 23 potentially inhibitory growth conditions in a 96-well microtiter plate format. Cells were resuspended at a recommended concentration in a proprietary "inoculation solution", that was amended to contain 0.1 g/L yeast extract. The cell suspension was then pipetted into each of the 96 wells, which included positive and negative control wells. Each well contained a carbon substrate and Microorganisms 2022, 10, 80 4 of 18 other nutrients and buffer or contained a potential growth-inhibiting condition (e.g., salt, pH, antibiotic, etc.). The plates were incubated at 45 • C for 24 to 48 h, and the OD 590 of each well was recorded at 8-h intervals. Positive wells turned purple in color due to the reduction of a tetrazolium redox dye, indicating growth and/or oxidation of the included substrate. The small amount of yeast extract added to the inoculation medium was insufficient to produce a positive color reaction.
Finally, fatty acid methyl esters, respiratory quinones, and the GC content of the DNA were analyzed according to previously described procedures [25]. Polar lipids were extracted using the Bligh-Dyer method and were analyzed by two-dimensional thin-layer chromatography as previously described [26,27].

Genome Sequencing and Bioinformatic Analyses
Total genomic DNA of strain MS-P2 T was extracted, sequenced, and assembled as previously described [28]. The genome of strain MS-P2 T was sequenced using the PacBio Sequel platform. Sequence assembly was performed using a Canu 1.8 assembler, yielding a high-quality, closed circular genome. The completeness and contamination of the genome were checked using the online version of CheckM implemented in the Kbase software and data platform [29]. The assembled genome was annotated by the RAST annotation system (Rapid Annotation using SEED Technology) [30][31][32] as well as by using the NCBI Prokaryotic Genome Annotation Pipeline (www.ncbi.nlm.nih.gov/genome/annotation_ prok/ (accessed on 15 July 2021)) [33][34][35] to predict the number of coding genes, total RNA genes, and functional genes. The gene locus designations and gene names used in the manuscript are those from the NCBI Prokaryotic Genome Annotation Pipeline. Average nucleotide identity (ANI) values were calculated by the OrthoANI algorithm method as implemented in the OAT software package [36]. Digital DNA-DNA hybridization (dDDH) values were determined with the in-silico genome-to-genome distance calculator (GGDC2.1; http://ggdc.dsmz.de/distcalc2.php (accessed on 26 May 2021)) using the alignment method blast+ [37][38][39]. Genomes of type species belonging to the genus Elioraea (E. tepidiphila TU-7 T [13], E. thermophila YIM 72297 T [14], and E. rosea PF-30 T [19]) were compared with the genome of strain MS-P2 T . The MS-P2 T genome was also compared to the genome of Elioraea sp. strain "Yellowstone" [20].
A phylogeny based on concatenated proteins was constructed using a set of 49 universally conserved proteins/gene as defined by COG (Clusters of Orthologous Groups) gene families on the Kbase platform using the "SpeciesTree builder version 2.2.0" [40]. This analysis included the genomes of five Elioraea spp. with a set of closely related genomes selected from the public KBase genomes import from RefSeq. Relatedness was determined by alignment similarity to a select subset of 49 COG domains. FastTree2 was used to prepare the phylogenetic tree [40]. A list of the 49 proteins used, as well as other information, can be found online at https://narrative.kbase.us/#catalog/apps/SpeciesTreeBuilder/insert_ set_of_genomes_into_species_tree/release (accessed on 17 December 2021).

Isolation and Initial Cultivation
Strain MS-P2 T was isolated from the microbial mats associated with the main runoff channel at Mushroom Spring in the Lower Geyser Basin of Yellowstone National Park, WY, USA (GPS coordinates: Lat.: 44.5387, Long.: −110.798). The temperature at the sampling site was 52 • C, and the pH was~8.0. In this hot spring, the microbial mats mainly consist of thermophilic cyanobacteria of the genus Synechococcus [41][42][43] and phototrophic Chloroflexota from the genera Roseiflexus, Chloroflexus, "Candidatus Roseilinea", and "Candidatus Chloranaerofilum" [17,18,[42][43][44]. Several other chlorophototrophic organisms, including two Chloracidobacterium species, C. thermophilum and C. aggregatum [21][22][23], "Candidatus Thermochlorobacter aerophilum" [45], and at least four chlorophototrophic members of the Proteobacteria also occur in these mats [18]. Strain MS-P2 T was isolated as described in Section 2.1 of the Materials and Methods using CTM medium solidified with 0.8% (w/v) agar at pH 7.0. Pink colonies appeared within a week and were purified by restreaking. Liquid cultures were pinkish-orange in color, and concentrated cells were bright pink (Supplementary Figure S1A,B).

Phenotypic, Biochemical, and Chemotaxonomic Characterization
Cells of strain MS-P2 T grown in liquid CTM medium were motile, curved rods with dimensions of 0.5 to 1.0 µm (width) by 1.0 to 1.5 µm (length) ( Table 1). Strain MS-P2 T cells stained Gram-negative and were positive for both catalase and cytochrome c oxidase. Division occurred by binary fission (Figure 1A,B). When viewed by epifluorescence microscopy with a filter set specific for BChl a, cells exhibited intense autofluorescence, which suggested that the cells contain a substantial amount of BChl a ( Figure 1B).
Cells of strain MS-P2 T grown in liquid CTM medium were motile, curved rods with dimensions of 0.5 to 1.0 μm (width) by 1.0 to 1.5 μm (length) ( Table 1). Strain MS-P2 T cells stained Gram-negative and were positive for both catalase and cytochrome c oxidase. Division occurred by binary fission ( Figure 1A,B). When viewed by epifluorescence microscopy with a filter set specific for BChl a, cells exhibited intense autofluorescence, which suggested that the cells contain a substantial amount of BChl a ( Figure 1B).   Figure 2. The in vivo spectrum had distinctive absorbance peaks at 800 and 865 nm in the infrared region and a smaller maximum at 590 nm in the visible region, consistent with the presence of BChl a ( Figure 2). The spectrum also suggested that strain MS-P2 T has light-harvesting complex 1 (LH1) but that it might lack light-harvesting complex 2 (LH2; however, see below). The three absorbance maxima between 450 and 550 nm suggest that the isolate also produces carotenoids. Pigments were extracted with acetone-methanol (7:2, v/v) and an absorbance spectrum was also recorded. The presence of BChl a (Q y band absorbance maximum at 770 nm) and carotenoids (absorbance maxima at 467, 494, and 530 nm) ( Figure 2) were also detected in the spectrum of the pigment extract. To confirm this preliminary assessment, extracted pigments were analyzed by reversed-phase HPLC as described [24]. BChl a esterified with phytol, a small amount of bacteriopheophytin a, and the carotenoids spirilloxanthin (major) and 3,4-dehydrorhodopin (minor) were identified by comparison to authentic standards (Supplementary Figure S2).
Strain MS-P2 T was able to grow chemoheterotrophically under aerobic conditions in the dark but was unable to grow photoautotrophically under oxic or anoxic conditions in the light. Cells grown in light grew slightly faster than cells grown under the same aerobic chemoheterotrophic conditions in the dark.
The growth temperature range of strain MS-P2 T was examined by measuring growth at 35, 45, 50, 55, and 60 • C by measuring the OD 660 of cultures incubated in 60 mL of liquid CTM medium containing succinate, acetate, and yeast extract (contained 1.0 g L −1 , respectively) in 100-mL aluminum-foil-capped flasks, with shaking at 150 rpm. The optimal growth of strain MS-P2 T occurred between 45 and 50 • C ( Figure 3A). The strain also grew at 40 • C but did not grow above 55 • C or below 35 • C. Thus, the growth temperature range can be described as~40 • C and higher but less than 55 • C (Table 1).  The pH range for growth was tested from pH 4.0 to 10.5 at 50 • C in the CTM medium by using appropriate buffering agents: 10 mM 2-(N-morpholino)ethanesulfonic acid (MES), 10 mM (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 10 mM Bicine, and 50 mM N-cyclohexyl-3-aminopropanesulfonic acid (CAPS). Strain MS-P2 T grew at pH values between 6.0 and 10.0 but was unable to grow at pH 5.5 or at pH 10.5. The optimum pH for growth was pH 7.0 to 7.5 ( Figure 3B; Table 1).

Genomic Features
The assembled genome of MS-P2 T comprises a single circular contig with a total length of 3,367,643 bp exhibiting 100% completeness with no contamination. No plasmids are present. The mol % G + C content calculated from the genome sequence is 70.6%. The genome encoded 3083 protein-coding sequences (CDS), a single rRNA operon, and 46 tRNAs ( Table 3). The genome contains a complete set of genes for the synthesis of bacteriochlorophyll a, carotenoids of the spirilloxanthin series, and the photosynthetic apparatus (e.g., pufABLM, puhA). The presence of acsF and bchE allows BChl synthesis to occur under both oxic and anoxic conditions, respectively [46]. Surprisingly, although the absorbance spectrum of cells suggests the absence of LH2 antenna complexes, the genome encods a pucBAC operon that could potentially produce such LH2 complexes. Three terminal oxidases (cytochrome aa 3 , bb 3 , and bd oxidases) are encoded in the genome. Genes (soxBCDYZ) for oxidation of thiosulfate are present, but consistent with the absence of photoautotrophic growth, genes for phosphoribulokinase, and ribulose-1,5-bisphosphate carboxylase-oxygenase (RuBisCO), and other enzymes for CO 2 fixation are absent. Genes encoding nitrogenase and nitrate and nitrite reductase are missing, but genes for urease (ureABC) and its assembly (ureDEFGJ) are present (Table 3). Finally, E. tepida lacks genes for the synthesis of vitamin B 12 and methionine synthase, but the genome encodes a methionine transporter. Figure 4 shows the pairwise relationships based on the calculated average nucleotide identity (ANI) for all five Elioraea spp. strains for which genome sequence information was available. These data show that E. tepidiphila and E. thermophila, with a pairwise ANI value of 85.08%, are slightly more similar than any other pair of strains, but that all strains are otherwise similarly and distantly related with pairwise ANI values ranging from 73 to 78%. Because ANI values above 95-96% are expected when two strains belong to the same species, strain MS-P2 T belongs to the genus Elioraea but is obviously different from the other three type-species of this genus. Strain MS-P2 T is also distinct from another recently sequenced isolate, Elioraea sp. strain "Yellowstone", from Octopus Spring [20]. Genomic relatedness by digital DNA-DNA hybridization (dDDH) indicates that genomic similarities between strain MS-P2 T and E. rosea PF-30 T , Elioraea sp. strain "Yellowstone", E. tepidiphila TU-7 T , and E. thermophila YIM 72297 T were only 21.90%, 20.50%, 19.80%, and 18.70%, respectively ( Table 3). The low pairwise dDDH values (19 to 22%) are far below the threshold values for species-level relatedness (70% dDDH) [47], and thus these genome-wide comparisons strongly support the proposal that strain MS-P2 T represents a new species within the genus Elioraea.  [13,14], and GenBank entry NZ_ARKI00000000. 2 Data are from reference [14]. 3 Data are from reference [19]. 4 Data are from reference [20]. 5 Values are the results from comparisons with the E. tepida MS-P2 T genome. 6 Values are the results from comparisons with the E. tepida MS-P2 T 16S rRNA gene sequence.
For phylogenetic assessment of strain MS-P2 T , the complete 16S rRNA gene sequence (1494 bp, NCBI acc. no. MZ358392) was retrieved from the sequenced whole genome. This sequence was 100% identical to the partial 16S rRNA gene sequence derived from an amplified PCR product (NCBI acc. no. MN600983) for this isolate, and the complete sequence was used in phylogenetic calculations. Pairwise nucleotide sequence similarity values for the 16S rRNA gene were calculated with the robust global sequence alignment algorithms in the EzTaxon server (https://www.ezbiocloud.net/ (accessed on 26 May 2021)) [48]. Phylogenetic trees were constructed with the MEGA 7 program [49] using the neighbor-joining, maximum parsimony, and maximum likelihood methods with Kimura's two-parameter model [50] (Figure 5 and Supplementary Figure S4A,B). The maximum likelihood tree based on 16S rRNA gene sequences shows that strain MS-P2 T is a member of the genus Elioraea near the root of the Acetobacteraceae together with E. tepidiphila, E. thermophila, E. rosea, and Elioraea sp. strain "Yellowstone" (Figure 5). These five isolates form a mono-phyletic group that is an early-diverging clade within the family Acetobacteraceae (identity values of 90-92% for other members of this family). Similar results were obtained for phylogenetic trees calculated by the neighbor-joining and maximum parsimony methods (Supplementary Figure S4A,B). Surprisingly, strain MS-P2 T is closest in all cases to E. rosea with a pairwise sequence identity value of 97.99%, which is below the threshold value for novel species demarcation (proposed 16S rRNA gene sequence similarity threshold value of <98.6% [51]). Sequence identities to E. tepidiphila TU-7 T (=DSM 17972 T ), E. thermophila YIM 72297 T , and Elioraea sp. strain "Yellowstone" were 97.26%, 96.58%, and 97.52%, respectively (Table 3). These findings support the proposal that strain MS-P2 T is a novel species within the genus Elioraea. Moreover, our findings support the proposal by Hördt et al. that the genus Elioraea should be placed in the family Acetobacteraceae [15,16]. For phylogenetic assessment of strain MS-P2 T , the complete 16S rRNA gene sequence (1494 bp, NCBI acc. no. MZ358392) was retrieved from the sequenced whole genome. This sequence was 100% identical to the partial 16S rRNA gene sequence derived from an amplified PCR product (NCBI acc. no. MN600983) for this isolate, and the complete sequence was used in phylogenetic calculations. Pairwise nucleotide sequence similarity values for the 16S rRNA gene were calculated with the robust global sequence alignment algorithms in the EzTaxon server (https://www.ezbiocloud.net/ (26 May 2021)) [48]. Phylogenetic trees were constructed with the MEGA 7 program [49] using the neighbor-joining, maximum parsimony, and maximum likelihood methods with Kimura's two-parameter model [50] (Figure 5 and Supplementary Figure S4A,B). The maximum likelihood tree based on 16S rRNA gene sequences shows that strain MS-P2 T is a member of the genus Elioraea near the root of the Acetobacteraceae together with E. tepidiphila, E. thermophila, E. rosea, and Elioraea sp. strain "Yellowstone" (Figure 5). These five isolates form a monophyletic group that is an early-diverging clade within the family Acetobacteraceae (identity values of 90-92% for other members of this family). Similar results were obtained for phylogenetic trees To assess the relationships among the five Elioraea strains currently available further and to assess the relationship of the genus Elioraea to other members of the Alphaproteobacteria, a phylogenetic analysis was performed that is based upon an alignment of a concatenation of 49 universally conserved proteins ( Figure 6). As found in the ANI analysis described above, strain MS-P2 T was most closely related to the mesophilic strain E. rosea; and E. tepidiphila and Elioraea sp. "Yellowstone" were the second pair of more closely related strains. The five Elioraea spp. strains collectively form a monophyletic clade near the base of the strains forming the family Acetobacteraceae ( Figure 6). oorganisms 2022, 9, x FOR PEER REVIEW 12 of 19 Figure 5. Phylogenetic tree based on 16S rRNA gene sequences constructed by the maximum-likelihood method based on the Kimura 2-parameter model [50] showing the phylogenetic position of strain MS-P2 T within the order Rhodospirillales. Robustness of the maximum-likelihood tree was tested by bootstrapping (100 resamplings, values > 50 are given at the nodes). Bacillus subtilis IAM 12118 T was used as an outgroup. The scale bar represents 0.10 substitution per site. Phylogenetic analyses were conducted in MEGA7 [49]. Trees made with neighbor-joining and maximum parsimony methods are very similar and can be seen in Supplementary Figure S4A,B.  [50] showing the phylogenetic position of strain MS-P2 T within the order Rhodospirillales. Robustness of the maximum-likelihood tree was tested by bootstrapping (100 resamplings, values > 50 are given at the nodes). Bacillus subtilis IAM 12118 T was used as an outgroup. The scale bar represents 0.10 substitution per site. Phylogenetic analyses were conducted in MEGA7 [49]. Trees made with neighbor-joining and maximum parsimony methods are very similar and can be seen in Supplementary Figure S4A,B. and to assess the relationship of the genus Elioraea to other members of the Alphaproteobacteria, a phylogenetic analysis was performed that is based upon an alignment of a concatenation of 49 universally conserved proteins ( Figure 6). As found in the ANI analysis described above, strain MS-P2 T was most closely related to the mesophilic strain E. rosea; and E. tepidiphila and Elioraea sp. "Yellowstone" were the second pair of more closely related strains. The five Elioraea spp. strains collectively form a monophyletic clade near the base of the strains forming the family Acetobacteraceae ( Figure 6). It is interesting and surprising that strain MS-P2 T isolated from Mushroom Spring is distinctly different from Elioraea sp. "Yellowstone" isolated from nearby Octopus Spring. These two hot springs are separated by only about a quarter of a mile in the Lower Geyser It is interesting and surprising that strain MS-P2 T isolated from Mushroom Spring is distinctly different from Elioraea sp. "Yellowstone" isolated from nearby Octopus Spring. These two hot springs are separated by only about a quarter of a mile in the Lower Geyser Basin of Yellowstone National Park, and they are chemically very similar and have associated mat communities that are also quite similar [42]. However, Elioraea sp. "Yellowstone" from Octopus Spring is more similar to E. tepidiphila and not to E. tepida or E. rosea.
Phenotypic, fatty acid, and genotypic characteristics of strain MS-P2 T and the three validly described species (E. tepidiphila, E. thermophila, E. rosea) [13,14,19] are summarized in Tables 1-3, respectively. These organisms generally share several common phenotypic traits, such as cell shape, swimming motility (except E. thermophila), moderate thermophily (except E. rosea), use of ubiquinone-10 as the respiratory quinone, production of catalase (except E. rosea, but katG is present in its genome), and the presence of cytochrome c oxidase. Similarities in cellular fatty acid and polar lipid composition also suggest that these organisms are closely related (Table 2). However, there are also important differences among these four isolates as follows: (1) The moderately thermophilic species were isolated from circum-neutral to slightly alkaline hot spring microbial mats, while the mesophilic E. rosea strain was isolated from floodwaters from a paddy field in South Korea. (2) E. thermophila is unique among the strains in having both phosphoribulokinase and type-1 ribulose 1,5-bisphosphate carboxylase-oxygenase, suggesting that this strain fixes CO 2 by the Calvin-Benson-Bassham cycle. This provides an explanation for the unique ability among Elioraea spp. of E. thermophila to grow photoautotrophically [14]. (3) All strains including MS-P2 T produced BChl a, carotenoids, and a similar photosynthetic apparatus, although both E. tepidiphila and Elioraea sp. strain "Yellowstone" were initially reported to be non-pigmented [13,14,19]. (4) Like E. rosea and E. tepidiphilia, strain MS-P2 T showed optimal growth between pH 7.0-7.5, but E. tepidiphila prefers slightly more alkaline conditions (pH 8.0-8.5). (5) Strain MS-P2 T lacked C17:0, C16:0 2-OH, C18:0 2-OH, and 19:0 cyclo ω8 (∆11:12) cellular fatty acids that were detected in other Elioraea spp. and uniquely possesses a significant amount of C15:0 iso, a fatty acid not found in any other Elioraea spp. (Table 2) (6) Strain MS-P2 T oxidizes a wide variety of sugars (e.g., fructose, fucose, galactose, mannose, melibiose, and psicose), while in general other Elioraea spp. cannot. (7) Strain MS-P2 T is the only Elioraea spp. strain that can hydrolyze gelatin. (8) Strain MS-P2 T could only oxidize proline, but E. tepidiphila can oxidize glutamate and glutamine as well as proline. Based on these phenotypic and genotypic differences and others (Tables 1-3), in addition to the phylogenetic and genomic differences among the strains described above, we propose strain MS-P2 T to be a new species belonging to the genus Elioraea with the suggested name Elioraea tepida (See description in Section 3.5).

Distribution and Ecological Considerations
Ward et al. [52] reported a 16S rRNA sequence, designated "Type O", in the microbial mats associated with Octopus Spring, Yellowstone National Park, WY, USA, in 1992. However, because the type species of the genus Elioraea was not described until 2008 [13], the Type O sequence (Elioraea spp.) initially could not be attributed to any specific organism. Later, using 16S rRNA amplicon sequencing and metagenomic sequencing of the upper green euphotic layer, a survey study of the mats of nearby Mushroom Spring provided the first evidence that members of the genus Elioraea were present in these hot-spring mat communities [17,53]. The 16S rRNA amplicon, which at that time was most similar to the 16S rRNA sequence of E. tepidiphila, could also be associated with a bin of sequences derived from the metagenome, which represented about 50% of the genome of the organism [17,53]. From the partial genomic information included in the Elioraea bin (Bin 22, OTU-46), it was apparent that the organisms associated with these sequences would likely have the capacity to synthesize BChl a and to produce bacterial reaction centers [17,18,53]. Strain MS-P2 T , described herein, and strain "Yellowstone" [20] were subsequently isolated. Interestingly, the latter strain was reported to be non-pigmented [20], as was initially the case for E. tepidiphila [13], although the genomes of both suggested that they should produce BChl a and carotenoids like other Elioraea sp. strains [14,19]; this study. It is possible that these two closely related strains may only synthesize BChl a under specific growth conditions. The physiology of E. tepida strain MS-P2 T described here and the distribution data from the 16S rRNA amplicon sequencing and metagenomic analyses do not necessarily agree [17,53]. The isolate clearly requires oxygen for growth and does not exhibit sensitivity to oxygen that might suggest it to be a microaerophile. However, 16S rRNA amplicon sequencing suggests that members of the genus Elioraea are not present or are rare in the uppermost green layer of the mat, where members of Synechococcus (Cyanobacteria) and Roseiflexus (Chloroflexota) spp. are the predominant organisms [7,17,18]. Instead, amplicon sequencing showed that sequences from Elioraea spp. are found in the undermat [17,53]. This observation indicates that strain MS-P2 T probably occurs naturally near but below the surface of the mat, i.e., in the upper portion of the undermat that also includes microaerophiles like Chloracidobacterium thermophilum. As previously noted, the genome encodes cytochrome bdand bb 3 -type terminal oxidases, which characteristically have a higher affinity for oxygen than aa 3 -type oxidases. The ability of E. tepida to oxidize thiosulfate might also help to explain its location below the surface of the mats, away from the highest concentrations of oxygen. Sulfate reduction occurs away from the surface of the mat deeper in the anoxic zone and mostly at night [54]. Because Elioraea spp. are found in similar mat communities of both Mushroom and Octopus Springs, and because the two strains are similar but not very closely related, it is likely that these two strains play a specific role in the mat ecophysiology. However, at this time it is unclear what that role might be, and it is likewise uncertain whether Elioraea spp. can form specific associations with other microbes in these mat communities. Given the overall complexity of the chlorophototrophs in the Mushroom Spring mats, with at least 18 types of phototrophs identified to date [7,18], as well as dozens of ecotypes of the major mat inhabitants [55], deciphering the interactions among this panoply of phototrophs and the major chemoheterotrophs in the mat community will likely keep microbiologists busy for many years.