The fecal samples from two bat genera were obtained from the Wuhan Institute of Virology, Chinese Academy of Sciences, which were collected from different far-apart sites in Yunnan and Guangxi provinces in October 2013 and July 2011, respectively. The specific latitude and longitude are E102°04′39″, N25°09′10″ (Yunnan) and E106°49′20″, N22°20′54″ (Guangxi) ( Table S1 ). In addition, the species of bats have been identified as Rousettus leschenaultia and Taphozous perforates by blasting the cytochrome b gene sequence (Huang et al., 2022c). Sample collection, transportation and storage were almost like what was delineate antecedently (Ge et al., 2012; Huang et al., 2022a). Briefly, the clean plastic sheets (around 4 square meters) were placed under known bat roosting sites in their natural habitat (usually caves) at approximately six o’clock in the afternoon, and the fecal samples were collected from the sheets in the next morning (approximately 6:00 am) and placed into sterile tubes.

The samples were transported to the laboratory and stored at –80°C until use. Fecal samples obtained from the same bat species were mixed in the laboratory to minimize variations within the same species. The isolation of bacteria from the animal feces was conducted as described previously (Bai et al., 2016; Wang et al., 2018). Briefly, approximately 1 gram of the fecal sample was added in 1 mL of 0.85% (w/v) NaCl solution by lightly whirlpool oscillation, and 150 µL of the suspension was spread on brain-heart infusion (BHI) agar. After incubation at 28°C for 7 days under aerobic, emerging colonies of different morphology were selected, purified, and preliminarily identified by PCR amplification and sequencing of 16S rRNA gene sequence analysis. As previously described (Wang et al., 2018), the nearly full-length 16S rRNA gene fragments (1510 bp) of the four strains were obtained from TA clones using the pEASY-T3 cloning kit (Transgenes) after amplification with two bacterial universal primers, 27F and 1492R (Jin et al., 2013). The 16S rRNA gene sequences of our isolates were deposited in GenBank and calculated 16S rRNA gene sequence similarities with public 16S rRNA gene sequences of type strains using EzBioCloud (https://www.ezbiocloud.net/) 16S based ID service (Yoon et al., 2017a). Among these isolates, four circular, convex, smooth colonies (designated HY006^T, HY008, HY1745, and HY1793^T), undergoing an identical procedure as other colonies, were identified as potential novel species of genus Ornithinimicrobium and preserved at –80°C in BHI broth supplemented with glycerol (20%, v/v) for further analysis. Strains HY006^T and HY1793^T were isolated from feces of Rousettus leschenaultia whereas HY008 and HY1745 were isolated from feces of Taphozous perforates. According to the Rules 15, 18a, and 18b from Chapter 3 of International Code of Nomenclature of Prokaryotes (Charles et al., 2019), strains HY006^T and HY1793^T were designated as the type strain, both of them were deposited in the China General Microbiological Culture Collection Center (CGMCC) and the Japan Collection of Microorganisms (JCM) from which they would be available.

The assembled sequences were aligned with representative 16S rRNA gene sequences of phylogenetically related species using ClustalW software (Thompson et al., 1994). Phylogenetic trees were reconstructed by the type strain of the family Ornithinimicrobiaceae with three algorithms, neighbor-joining (Saitou and Nei, 1987), maximum-likelihood (Guindon and Gascuel, 2003) and maximum-parsimony (Kolaczkowski and Thornton, 2004), using MEGA version X (Kumar et al., 2018) based on bootstrap analysis with 1000 replications, as previously described (Huang et al., 2019; Zhu et al., 2019). Arthrobacter globiformis JCM 078085^T (=NBRC 12137^T) was used as the outgroup.

The genomic DNA of four strains was extracted using the Wizard Genomic DNA Purification Kit (Promega) according to the manufacturer’s instruction. Strains HY006^T and HY1793^T were sequenced by single molecule real-time (SMRT) technology (McCarthy, 2010) on the Pacific Biosciences (PacBio) sequencing platform, which was assembled using de novo and analyzed using the Hierarchical Genome Assembly Process (HGAP4) application (Castanera et al., 2019) to obtain its whole genome sequence without gaps. Meanwhile, the draft genome sequences of strains HY008 and HY1745 were sequenced on the Illumina HiSeq TM2000 platform and assembled using VELVET (Zerbino and Birney, 2008). The number of tRNAs, rRNAs, and protein-coding DNA sequences (CDSs) were predicted using GeneMarkS+4.2 (Berlin et al., 2015). Gene calling and annotation were performed with the Rapid Annotation using Subsystem Technology (RAST) server (https://rast.nmpdr.org/) (Aziz et al., 2008) and the SEED viewer framework (Overbeek et al., 2014). In addition, the antibiotic resistance genes of strains are obtained by comparison in the comprehensive antibiotic resistance database (https://card.mcmaster.ca/analyze) (Alcock et al., 2022). Using the virulence factor database (VFDB, http://www.mgc.ac.cn/VFs/main.htm) (Liu et al., 2019), we identified genome encoded virulence factor (VF) genes. The database of Clusters of Orthologous Groups of proteins (COGs) is a phylogenetic classification of proteins encoded in completely sequenced genomes and the COGs comprise a framework for functional and evolutionary genome analysis (Tatusov et al., 1997). Furthermore, The average nucleotide identity (ANI Calculator | Ezbiocloud.net; with a threshold of 95–96%) and digital DNA-DNA hybridization (dDDH) (http://ggdc.dsmz.de/ggdc.php; with a threshold of 70%) values of the whole genomes of our four isolates with closely related type strains were calculated using the OrthoANIu algorithm (Yoon et al., 2017b) and Genome-to-Genome Distance Calculator 3.0 (GGDC 3.0) (Meier-Kolthoff et al., 2022), respectively.

To further validate the taxonomic status of the four strains in the genus Ornithinimicrobium, a phylogenomic tree based on core genes was constructed using the FastTree program (Price et al., 2009). The core genes were extracted from 22 whole genomes ( Table S2 ) from the NCBI GenBank database (including genomes of strains HY006^T, HY008, HY1745, HY1793^T, Arthrobacter globiformis NBRC 12137^T serving as outgroup and 17 available genomes within the family of Ornithinimicrobiaceae) as aligned by Mafft and determined by CD-HIT (Fu et al., 2012) software (protein sequence identity threshold, 0.4). The output result was visualized in Dendroscope 3 (version 3.5.10) (Huson and Scornavacca, 2012) and modified by Interactive Tree of Life (https://itol.embl.de/) (Letunic and Bork, 2021). Moreover, the whole-genome comparison and annotation of orthologous clusters across multiple species were conducted with OrthoVenn2 (Xu et al., 2019). The pairwise and multiple whole genome alignments were generated with the progressiveMauve algorithm (version 2015-02-26) (Darling et al., 2010).

The phenotypic and physiological tests were similar to our previously described (Huang et al., 2022a). Bacterial growth test was determined at a range of temperatures (4, 10, 15, 20, 25, 28, 30, 35, 37, 40, and 45°C) in BHI broth for 7 days. The pH levels (4.0–11.0, at intervals of 0.5) and NaCl concentrations (0–12%, w/v, at intervals of 0.5%) for growth were also examined using BHI broth at 28°C. The cells have an incubation period of up to 7 days for observing growth under microaerophilic (5.0% CO₂ incubator) or anaerobic conditions (80.0% N₂, 10.0% CO₂, and 10.0% H₂) by incubating the strains on identical plates. Cell morphology was observed by microscope (Light microscope, Eclipse 80i; Transmission electron microscope, HT7700) after incubating on BHI-1.0% NaCl (w/v) plates at 28°C for 5 days. Oxidase activity was examined by oxidase reagent (bioMérieux) and the catalase activity of strains was tested with bubble production in 3% (v/v) hydrogen peroxide. Gram staining was performed using a Gram-staining kit (Baso) (Austrian, 1960), and motility was tested by observing the spreading growth of cells inoculated by piercing into BHI semisolid (0.3% agar, w/v) medium in test tubes (Xu et al., 2013). Acid production from sucrose, glucose, and other physiological and biochemical tests were determined using API 50CH strips combined with API 50CHB medium and API 20NE, and qualitative enzyme tests were determined with the API ZYM system (bioMérieux). Additionally, Biolog GEN III MicroPlate (catalog No. 1030) with inoculating fluid (catalog No. 72401-IF-A) was also used to obtain the biochemical data following the manufacturers’ instructions (Biolog). Antibiotic sensitivity of strains HY006^T and HY1793^T was tested using the E-test stripes (bioMérieux) as previously described (Brown and Brown, 1991), on BHI supplemented with 5.0% sheep blood agar at 28°C containing the following antibiotics: ceftriaxone, chloramphenicol, clindamycin, erythromycin, levofloxacin, linezolid, meropenem, tetracycline, and vancomycin.

For the biomass of chemotaxonomic assays, all the tested strains were harvested from BHI-1% NaCl plates after 5 days of incubation at 28°C. Cellular fatty acids were extracted, methylated, and identified by using the Sherlock Microbial Identification System (MIDI) according to the manufacturer’s standard protocols (Sasser, 1990). The respiratory isoprenoid quinones were extracted, purified, and analyzed by high performance liquid chromatography (HPLC) (Komagata and Suzuki, 1988) with various menaquinones and ubiquinones (Hu et al., 2001) as references. Polar lipids were analyzed using two-dimensional thin-layer chromatography (2D TLC) after hydrolysis with 6 M HCl at 100°C for 18 hours (Staneck and Roberts, 1974; Harper and Davis, 1979). Peptidoglycan amino acid composition was measured with a Hitachi-8900 high speed amino acid analyzer after hydrolyzing the cell wall, and whole-cell sugars were examined according to the method of Hasegawa et al. (Hasegawa et al., 1983).

Bacterial carotenoid extracts were analyzed at 450 nm using an UPLC system with DAD detector (UPLC, U3000; Thermo Scientific). The analytical conditions were as follows, UPLC: column, YMC Carotenoid S-3 μm (150 × 4.6 mm); column temperature, 40°C; flow rate, 1.0 mL/min; injection volume, 2 μL; solvent system (MeOH: MTBE: H₂O=20: 75: 5); gradient program, 100:0 v/v at 0 min, 39:61 v/v at 15 min, 0:100 V/V at 25 min, 100:0 v/v at 25.1min, 100:0 v/v at 30 min. Data were acquired on the U3000 UPLC (Thermo Scientific) and processed using chromeleon 7.2 CDS (Thermo Scientific) (Meléndez-Martínez et al., 2010; Irakli et al., 2011). The carotenoid content in samples was calculated as the formulas: Carotenoid content (µg/100 mL) = Read concentration (µg/mL) × Dilution factor × 100.

Five reference strains were respectively obtained from the Korean Collection for Type Cultures (O. cavernae CFH 30183^T= KCTC 49018^T; O. ciconiae H23M54^T= KCTC 49151^T), Canadian Phytoplankton Culture Collection (O. flavum CPCC 203535^T), the Deutsche Sammlung von Mikroorganismen und Zellkulturen (O. murale 01-Gi-040^T= DSM 22056^T) and the Guangdong Microbial Culture Collection Center (O. pratense W204^T= GDMCC 1.1391^T).

The 16S rRNA gene sequences of strains HY006^T/HY008 (1,488 bp) and HY1745/HY1793^T (1,510 bp) were obtained. According to the comparison results in the EzBioCloud database, strains HY006^T and HY008 were highly similar to those of Ornithinimicrobium pratense W204^T (99.3%) and O. flavum CPCC 203535^T (97.3%), whereas strains HY1745 and HY1793^T were closest to the type strains O. ciconiae H23M54^T (98.7%), O. cavernae CFH 30183^T (98.3%), and O. murale 01-Gi-040^T (98.1%).

The phylogenetic trees constructed using various algorithms (NJ, ML, and MP) ( Figure 1A ) showed that strains HY1745 and HY1793^T formed a separate phylogenetic sub-branch within a sub-clade in the Ornithinimicrobium clade encompassed by three recognized species of the genus (most closely related neighbors mentioned above). In addition, the closest relationship of strains HY006^T/HY008 to O. pratense W204^T was clearly shown in Figure 1A . Furthermore, the phylogenomic tree (based on 536 core genes) indicated that our four isolates belonged to the genus of Ornithinimicrobium, and strains HY006^T/HY008 formed a distinct cluster with species O. pratense W204^T and O. flavum CPCC 203535^T. The strains HY1745 and HY1793^T also formed a clade with O. ciconiae H23M54^T, O. cavernae CFH 30183^T, and O. murale 01-Gi-040^T supported by high bootstrap value ( Figure 1B ), almost identical to the results of 16S rRNA gene phylogenetic trees.

The draft genome sequence of strain HY008 consisted of 35 contigs and yielded a genome of 4,147,990 bp (contained 3,817 genes) after assembly. The complete genome of type strain HY006^T was 4,169,359 bp (N50: 4,169,359 bp; genome coverage: 240.0×) in length and contained 3,805 genes, with one plasmid (length: 46,843 bp). The plasmid contains 59 coding sequences (CDS) and the G + C content is 66.8 mol%. Most of the protein are hypothetical protein (88.1%), some are enzymes involved in DNA replication (6.8%), two uncharacterized protein (3.4%) and one structural protein (1.7%), no proteins with specific functions were found ( Table S3 ). In addition, the DNA G+C content of strains HY006^T and HY008 were 71.4 and 71.5 mol%, respectively. Strains HY1745 and HY1793^T contained 4,421 and 4,262 genes, respectively. The number of tRNA genes in our four strains was 46, and the G + C content was between 68.7 and 71.5% ( Table 1 ). Moreover, stains HY006^T, HY008, HY1745 and HY1793^T contained ten, nine, five, and ten predicted virulence factors, respectively ( Table S4 ). Specifically, two virulence factor-like genes, namely sodA (superoxide dismutase, VFDB ID: VFG001421) and HSI-3 (type VI secretion system ATPase TssH, VFDB ID: VFG041043) were discovered in most our isolates except for strain HY1745. In addition, five virulence factor-like genes, including those encoding isocitrate lyase icl (VFG001381), nitrate reductase subunit alpha narG (VFG001391), nitrate reductase subunit beta narH (VFG001814), AlgW protein algW (VFG014984), and UDP-glucose 6-dehydrogenase ugd (VFG048797), were also dominantly found in the type strains HY006^T and HY1793^T, but were absent in strains HY008 and HY1745.

The values of dDDH between the four isolates and their closely related species or other available genome sequences in the family Ornithinimicrobiaceae were all below the 70% threshold (19.6–33.6%), by contrast to the finding that strains HY006^T and HY008 had a dDDH value of 99.9% and strains HY1745 and HY1793^T shared 78.9%, indicating that they belong to the same species ( Table S2 ). Similarly, the ANI values of strain HY006^T with strains O. flavum CPCC 203535^T and O. pratense GDMCC 1.1391^T were 79.3% and 87.4%, respectively, which were below the threshold value (95%) for species delineation; strain HY1793^T with its three closely related species (O. ciconiae H23M54^T, O. cavernae KCTC 49018^T and O. murale DSM22056^T) were between 80.5–82.9 and ranged from 71.2 to 75.0% with other species in the genus Ornithinimicrobium. However, the ANI relatedness within each strain pair is 99.9% (HY006^T and HY008) and 97.5% (HY1745 and HY1793^T). Moreover, the dDDH and ANI values between strains HY006^T and HY1793^T were 20.4% and 74.4% respectively, suggesting that they belong to different species. In short, these results support the notion that these four strains represent two different novel species in the genus Ornithinimicrobium.

Moreover, the Clusters of Orthologous Groups (COG) database was used for the classification of the genes in the sequenced genomes ( Figure 2A ). The results revealed that the highest numbers of genes contained in these genomes were associated with transcription (COG-K), translation (COG-J), and DNA replication and repair (COG-L) for information storage and processing. For cellular processes and signaling, the genes involved in cell wall/membrane/envelope biogenesis (COG-M) and signal transduction mechanisms (COG-T) were commonly abundant in sequenced genomes. Based on our analysis of the genes associated with metabolism, we found that the species of the genus Ornithinimicrobium focused on amino acids transport and metabolism (COG-E) and carbohydrate transport and metabolism (COG-G), additionally, the inorganic ion transport and metabolism (COG-P), energy production and conversion (COG-C) and coenzyme transport and metabolism (COG-H) were also abundant in the bacterial genomes. However, the proportion of secondary metabolites biosynthesis, transport and catabolism (COG-Q) and nucleotide transport and metabolism (COG-F) were relatively small. Notably, a large quantity of the genes in these bacterial genomes were poorly characterized, and their functions remain to be identified (COG-S) or general function prediction only (COG-R) ( Figure 2A ).

Furthermore, the Venn diagram displays the distribution of shared orthologous clusters among our four isolates and the other two strains (O. kibberense DSM 17687^T and O. pekingense CGMCC 1.5362^T, which should be deserved more attention because they were possessing cytotoxicity activity against to some mammary adenocarcinoma cell lines or causing ocular infection, respectively. In addition, MK-8(H₄) was detected as the major predominant menaquinone, which was consistent with strains HY006^T and HY1793^T but different from the other strains isolated from Rousettus leschenaultia or Taphozous perforates. Furthermore, both of them were color in light yellow, which was also similar to species O. sufpigmenti). The six strains form clusters number between 2,698 to 3,836 ( Figure 2B ), and a total of 1805 clusters are shared with the above strains. In addition, there are 3, 20, 17, 8, and 20 single-copy gene clusters in strains HY008, HY1745, HY1793^T, O. kibberense DSM 17687^T, and O. pekingense CGMCC 1.5362^T, respectively ( Figure 2C ). Additionally, our four strains and the above two strains have large segment sequence rearrangement. There are at least five locally collinear blocks, and the direction and arrange sequence are consistent, suggesting that genome evolution may experience a similar path ( Figure S1 ).

Collectively, cells of strains HY006^T, HY008, HY1745, and HY1793^T were convex, opaque, and circular (0.5–1.0 mm in diameter) with smooth surfaces. Cells were Gram-stain-positive, catalase-positive, oxidase-negative, circle-shaped (HY006^T 1.0 × 1.2 µm; HY1793^T 1.0 × 1.2 µm) ( Figure S2 ), non-motile. Interestingly, the color was totally different between the two type strains. The strain HY006^T was yellow whereas strain HY1793^T was cream, which may be associated to the production of carotenoids. Extraction of the HY006^T cells with a solvent system showed that they accumulated two different carotenoids at significant concentrations ( Figure S3 ). The contents of β-cryptoxanthin (C₄₀H₅₆O) and lycopene (C₄₀H₅₆) in strain HY006^T were 26.0 µg/mL and 2.3 µg/mL, respectively. In the contrast, any carotenoid components were not found in the strain of HY1793^T, which was consistent with its color (cream) but different from strain HY006^T (yellow).

Optimal growth conditions for the strains HY006^T and HY008 were 28–30°C (range, 15–40°C) for 5 days on BHI with 1.0% (w/v) NaCl [range, 0–7.0 (w/v)] at pH 8.0–8.5 (range, 7.0–11.0). Additionally, the optimal growth conditions of strains HY1745 and HY1793^T were at 28°C (range, 15–37°C), pH 8.0 (range, pH 6.5–10.0), and with 1.0–1.5% (w/v) NaCl [up to 9.0% (w/v)] in aerobic. Notably, our strains were tolerant to 7.0% NaCl, while no growth of O. flavum CPCC 203535^T was observed under this condition. However, O. cavernae KCTC 49018^T, O. ciconiae JCM 33221^T, O. murale DSM 22056^T, and O. pratense GDMCC 1.1391^T can grow well with 7.0% NaCl. The four isolates (HY006^T, HY008, HY1745 and HY1793^T) and their closely related strains (O. cavernae KCTC 49018^T, O. ciconiae JCM 33221^T, O. flavum CPCC 203535^T, O. murale DSM 22056^T and O. pratense GDMCC 1.1391^T) were also not observed growth on pH value of lower than 6.0, which were different from strains O. pekingense DSM 21552^T and O. kibberense DSM 17687^T but consistent with most the species of Ornithinimicrobium (Huo et al., 2019; Fang et al., 2020; Cao et al., 2022). Also, strains HY006^T and HY008 were positive for α-chymotrypsin, which were different from strains HY1745, HY1793^T, O. cavernae KCTC 49018^T, O. ciconiae JCM 33221^T, O. flavum CPCC 203535^T, O. pratense GDMCC 1.1391^T but consistent with O. murale DSM 22056^T. Conversely, alkaline phosphatase and cystine arylamidase were utilized by strains HY1745 and HY1793^T whereas strains HY006^T and HY008 were negative for them. The more differential biochemical characteristics of all the compared strains were shown in Table S5 . As detailed in Figure S4 , all strains positively assimilated aztreonam, d-serine and lithium chloride while tested with the Biolog GEN III MicroPlate. Of note is the strains HY006^T and HY008 were positive for acetoacetic acid, d-arabitol, d-glucuronic acid, l-galactonic acid lactone, tetrazolium violet and α-hydroxy-butyric acid, distinctly different from the other two potential novel strains (HY1745 and HY1793^T). Most notably, lincomycin was found to be utilized by strains HY1745 and HY1793^T, which was consistent with strains O. cavernae CFH 30183^T, O. murale DSM 22056^T, O. kibberense K22-20^T and O. panacihumi DCY118^T (Mayilraj et al., 2006; Huo et al., 2019; Zhang et al., 2019). Conversely, strains HY006^T, HY008 and the other four closely related strains (O. ciconiae JCM 33221^T, O. flavum CPCC 203535^T, and O. pratense GDMCC 1.1391^T) were sensitive to lincomycin according to the results of Biolog GEN III MicroPlate ( Figure S4 ). Furthermore, strain HY006^T was resistant [minimum inhibitory concentration (MIC) ≥8.0 µg/mL] to chloramphenicol (MIC, >20.0 µg/mL), linezolid (MIC, 12 µg/mL), and intermediately resistant (MIC, 1.0–6.0 µg/mL) to erythromycin (MIC, 3.0 µg/mL) and tetracycline (MIC, 6.0 µg/mL), but susceptible to ceftriaxone, clindamycin levofloxacin, meropenem and vancomycin (MIC, ≤0.5 µg/mL). Additionally, strain HY1793^T was resistant to erythromycin (MIC, 12.0 µg/mL), intermediately resistant to clindamycin (MIC, 2.0–3.0 µg/mL) and levofloxacin (MIC, 1.0 µg/mL) about the antibiotic sensitivity whereas susceptible to ceftriaxone, chloramphenicol, linezolid, meropenem, tetracycline and vancomycin (MIC, ≤0.5 µg/mL) ( Figure S5 ). Of note, both strains contain the corresponding drug resistance genes in their genome but the value of identities is low (<60.0%), indicating that the resistant phenotype may be regulated by other different genes and will require more research to be identified.

The major cellular fatty acids (>20.0%) of our four isolates (HY006^T, HY008, HY1745 and HY1793^T) were iso-C_15:0 and iso-C_16:0, which were consistent with strain O. humiphilum HKI 0124^T (Groth et al., 2001) and O. murale 01-Gi-040^T (Kampfer et al., 2013). In addition, Summed Feature 9 (C_16:010-methyl and/or iso-C_17:1 ω9 _C ) was also accounted for a high proportion (>10.0%) in strains HY1745, HY1793^T and other closely related strains (O. cavernae KCTC 49018^T, O. ciconiae JCM 33221^T, O. flavum CPCC 203535^T, O. murale DSM 22056^T, and O. pratense GDMCC 1.1391^T) but lower in strains HY006^T and HY008 (<6.5%). The detailed fatty acid profiles of the strains are presented in Table S6 . MK-8(H₄) was detected as the predominant menaquinone in strains HY006^T and HY1793^T, which was accounted for 59.0% and 93.3%, respectively. Furthermore, a significant amount of MK-8 (33.9%) was present in strain HY006^T rather than MK-6 (6.7%) was shown in strain HY1793^T, different from other species that had either simpler or more complicated compositions. For example, O. pekingense LW6^T has been reported to contain partially saturated menaquinone MK-8(H₂) (Liu et al., 2008); O. cerasi CPCC 203383^T and O. flavum CPCC 203535^T had MK-8(H₄), MK-8(H₂), and MK-8 as the predominant quinone (Fang et al., 2017; Fang et al., 2020). Both type strains (HY006^T and HY1793^T) had diphosphatidylglycerol (DPG), phosphatidylglycerol (PG), phosphatidyl inositol mannoside (PIM), two unidentified lipids (L1-2) and two unknown phosphoglycolipids (PGL1-2) in their polar lipid profiles ( Figure S6 ), additionally, HY006^T had two phospholipids (PL1-2) and two unidentified glycolipids (GL1-2) whereas HY1793^T had five phospholipids (PL1-5) and five unidentified glycolipids (GL1-5). The whole-cell peptidoglycan of both type strains contained ornithine, alanine, glycine, and glutamic acid ( Figure S7A ), significantly, ornithine is the diagnostic diamino acid component within the cell wall of the genus Ornithinimicrobium. They also contained glucose in the whole cell sugar, however, ribose was only detected in strain HY1793^T ( Figure S7B ).