Draft genome sequences of eight bacteria isolated from the indoor environment: Staphylococcus capitis strain H36, S. capitis strain H65, S. cohnii strain H62, S. hominis strain H69, Microbacterium sp. strain H83, Mycobacterium iranicum strain H39, Plantibacter sp. strain H53, and Pseudomonas oryzihabitans strain H72

Classification and features

Two growth media were used for the initial isolation of bacteria: lysogeny broth agar (LB, Difco Laboratories, Detroit, MI) and R2A agar (Difco Laboratories, Detroit, MI). Petri dishes were exposed to residential bathroom air for 1 h; 30 min during which shower water was running to create shower mist and 30 min after the shower was turned off. Petri dishes were mounted on vertical surfaces (door, wall, cabinets) at a height of approximately 1.50 m. Petri dishes were brought back to the laboratory, where LB plates were incubated at 28 °C for 48 h, and R2A plates were incubated at 28 °C for 5 days and at 35 °C for 3 days. Except for Staphylococcus hominis strain H69, which was isolated on LB agar medium at 35 °C, all other strains were isolated on R2A medium (Additional file 1). Research was approved by the University of California Committee for the Protection of Human Subjects Protocol ID 2015-02-7135, and the sampling was conducted in March, 2015.

Taxonomic classification of these isolates was undertaken after genome sequencing, either using the full-length 16S rRNA gene sequences or a concatenated marker gene approach. For Microbacterium , Mycobacterium , and Plantibacter there were insufficient publicly available genome sequences of close relatives for a concatenated marker approach. In these cases, the full length 16S rRNA gene sequence was uploaded to the Ribosomal Database Project [16] and added to alignments containing representatives of all close relatives (as estimated from BLAST [17]). These alignments were downloaded, cleaned with a custom script [18], and an approximately maximum likelihood tree was inferred using the default setting in FastTree [19]. Outgroups for all trees were type strains of another genus or genera within the same family. The sequence alignments supporting the phylogenetic trees of this article are available in the FigShare repository [20].

All strains were given a specific identifier (e.g., H83) based on our internal culture collection. The 16S rRNA gene trees for both Microbacterium and Plantibacter genera were poorly resolved (e.g., low bootstrap values), and these isolates were placed into polyphyletic clades with respect to the names of taxa in the genera (Additional files 2 and 3). In addition, while Microbacterium sp. H83 falls within a clade that contains mostly M. foliorum , this name also occurs outside the clade. Therefore we have not attempted to assign these isolates to a particular species. On the other hand, the rRNA gene for one isolate is found in a monophyletic clade with other M. iranicum isolates (Additional file 4) and thus we have assigned this the name M. iranicum H39, For the Pseudomonas and Staphylococcus isolates, the 16S rRNA gene trees were inadequate for taxonomic classification at the species level, but the genomes of numerous sequenced representatives of close relatives were available for further analysis. All available genome sequences of close relatives (to a max of 20 randomly selected genomes per species) were downloaded from NCBI. The file names and sequences were reformatted for easier visualization. The assemblies were then screened for 37 core maker genes [21] using PhyloSift [22] in search and align mode using “isolate” and “besthit” flags. PhyloSift concatenates and aligns the hits of interest so the sequences are subsequently extracted from the PhyloSift output files and added to a single file for tree-building. An approximately maximum-likelihood tree was then inferred using FastTree.

The concatenated marker genes for one isolate placed it in a well-supported clade of P. oryzihabitans isolates (Additional file 5) and thus we have named this P. oryzihabitans H72. Based on this tree, we believe that one of the (unpublished) strains of P. psychrotolerans has been misclassified and should also be considered P. oryzihabitans . Four of the isolates were Staphylococcus species, for which we created a single concatenated marker tree containing the relevant close relatives of the isolates (Fig. 1). Two of our Staphylococcus isolates placed within a well-supported (i.e., high bootstrap support) monophyletic clade of S. capitis strains and thus we have named these S. capitis H36 and S. capitis H65. One Staphylococcus isolate placed within a well-supported clade of S. cohnii strains and thus we have named it S. cohnii H62. Our fourth Staphylococcus isolate was placed within a well-supported clade containing mostly S. hominis isolates but which also contains a few S. haemolyticus isolates. Because this tree shows a distinct clade containing many S. haemolyticus isolates, we have named this isolate S. hominis H69. It is unclear from this tree alone whether these few S. haemolyticus isolates are misnamed or whether further taxonomic revision of this group is needed.

General description of the isolates are summarized in Table 1, and micrographs appear in Fig. 2.

MIGS ID	Property	Microbacterium sp. H83	Evidence codea	Mycobacterium iranicum H39	Evidence code	Plantibacter sp. H53	Evidence code	Pseudomonas oryzihabitans H72	Evidence code	Staphylococcus capitis H36	Evidence code	S. capitis H65	Evidence code	S.cohnii H62	Evidence code	S. hominis H69	Evidence code
	Classification
	Domain	Bacteria	TAS [61]	Bacteria	TAS [61]	Bacteria	TAS [61]	Bacteria	TAS [61]	Bacteria	TAS [61]	Bacteria	TAS [61]	Bacteria	TAS [61]	Bacteria	TAS [61]
	Phylum	Actinobacteria	TAS [62]	Actinobacteria	TAS [62]	Actinobacteria	TAS [62]	Proteobacteria	TAS [63]	Firmicutes	TAS [64]	Firmicutes	TAS [64]	Firmicutes	TAS [64]	Firmicutes	TAS [64]
	Class	Actinobacteria	TAS [65]	Actinobacteria	TAS [65]	Actinobacteria	TAS [65]	Gammaproteobacteria	TAS [66, 67]	Bacilli	TAS [68, 69]	Bacilli	TAS [68, 69]	Bacilli	TAS [68, 69]	Bacilli	TAS [68, 69]
	Order	Micrococcales	TAS [70, 71]	Corynebacteriales	TAS [72, 73]	Micrococcales	TAS [70, 71]	Pseudomonadales	TAS [71, 74]	Bacillales	TAS [70, 71]	Bacillales	TAS [70, 71]	Bacillales	TAS [70, 71]	Bacillales	TAS [70, 71]
	Family	Microbacteriaceae	TAS [75, 76]	Mycobacteriaceae	TAS [71, 77]	Microbacteriaceae	TAS [75, 76]	Pseudomonadaceae	TAS [71, 78]	Staphylococcaceae	TAS [79]	Staphylococcaceae	TAS [79]	Staphylococcaceae	TAS [79]	Staphylococcaceae	TAS [79]
	Genus	Microbacterium	TAS [71, 80]	Mycobacterium	TAS [71, 81]	Plantibacter	TAS [15]	Pseudomonas	TAS [71, 82]	Staphylococcus	TAS [71, 83]	Staphylococcus	TAS [71, 83]	Staphylococcus	TAS [71, 83]	Staphylococcus	TAS [71, 83]
	Species	Microbacterium sp.	NAS	M. iranicum	TAS [11]	Plantibacter sp.	NAS	P. oryzihabitans	TAS [84]	S. capitis	TAS [71, 85]	S. capitis	TAS [71, 85]	S. cohnii	TAS [71, 86]	S. hominis	TAS [71, 85]
	Strain	H83	IDA	H39	IDA	H53	IDA	H72	IDA	H36	IDA	H65	IDA	H62	IDA	H69	IDA
	Gram stain	Positive	TAS [23]	Positive	TAS [11]	Positive	TAS [87]	Negative	TAS [30]	Positive	TAS [88]	Positive	TAS [88]	Positive	TAS [88]	Positive	TAS [88]
	Cell shape	Rod	TAS [23]	Rod	TAS [11]	Rod	TAS [87]	Rod	TAS [30]	Coccus/grape-like clusters	TAS [88]	Coccus/grape-like clusters	TAS [88]	Coccus/grape-like clusters	TAS [88]]	Coccus/grape-like clusters	TAS [88]
	Motility	nd		Non-motile	TAS [11]	Non-motile	TAS [87]	Motile	TAS [30]	Non-motile	TAS [88]	Non-motile	TAS [88]	Non-motile	TAS [88]	Non-motile	TAS [88]
	Sporulation	Non-spore forming	TAS [23]	Non-spore forming	TAS [11]	Non-spore forming	TAS [87]	Non-spore forming	NAS	Non-spore forming	TAS [88]	Non-spore forming	TAS [88]	Non-spore forming	TAS [88]]	Non-spore forming	TAS [88]
	Temperature range	Mesophile	TAS [23]	25-40°	TAS [11]	Mesophile	IDA	Mesophile	TAS [30]	18-45 °C	TAS [88]	18-45 °C	TAS [88]	15-45 °C	TAS [88]	20-45 °C	TAS [88]
	Optimum temperature	nd		37°	TAS [11]	30 °C	TAS [87]	nd		nd	IDA	nd		nd		30-40 °C	TAS [88]
	pH range; Optimum	5-11; nd	TAS [23]	nd		nd		nd		nd		nd		nd		nd
	Carbon source	Yeast extract, Peptone, Dextrose, Starch, Casamino acids	IDA	Yeast extract, Peptone, Dextrose, Starch, Casamino acids	IDA	Yeast extract, Peptone, Dextrose, Starch, Casamino acids	IDA	Yeast extract, Peptone, Dextrose, Starch, Casamino acids	IDA	Yeast extract, Peptone, Dextrose, Starch, Casamino acids	IDA	Yeast extract, Peptone, Dextrose, Starch, Casamino acids	IDA	Yeast extract, Peptone, Dextrose, Starch, Casamino acids	IDA	Yeast extract, Tryptone	IDA
GS-6	Habitat	Indoor air	NAS	Indoor air	NAS	Indoor air	NAS	Indoor air	NAS	Indoor air	NAS	Indoor air	NAS	Indoor air	NAS	Indoor air	NAS
6.3	Salinity	Normal	IDA	5% NaCl (w/v)	TAS [11]	Normal	IDA	6.5% NaCl (w/v)	TAS [84]	10% NaCl (w/v)	TAS [88]	10% NaCl (w/v)	TAS [88]	10% NaCl (w/v)	TAS [88]	10% NaCl (w/v)	TAS [88]
22	Oxygen requirement	Aerobic	TAS [23]	Aerobic	TAS [89]	Aerobic	TAS [87]	Aerobic	TAS [30]	Facultative anaerobes	TAS [88]	Facultative anaerobes	TAS [88]	Facultative anaerobes	TAS [88]	Facultative anaerobes	TAS [88]
15	Biotic relationship	Free living	NAS	Symbiont	TAS [11]	Free living	TAS [87]	Free living; symbiont	TAS [84]	Free living	NAS	Free living	NAS	Free living	NAS	Free living	NAS
14	Pathogenicity	nd		nd		nd		nd		nd		nd		nd		nd
4	Geographic location	USA: California: Piedmont	NAS	USA: California: Oakland	NAS	USA: California: Walnut Creek	NAS	USA: California: Oakland	NAS	USA: California: Oakland	NAS	USA: California: Milpitas	NAS	USA: California: Milpitas	NAS	USA: California: Milpitas	NAS
5	Sample collection	2015-03-16	NAS	2015-03-18	NAS	2015-03-17	NAS	2015-03-18	NAS	2015-03-18	NAS	2015-03-31	NAS	2015-03-31	NAS	2015-03-31	NAS
4.1	Latitude	37°49'25.6"	NAS	122°16'21.9"	NAS	122°03'50.1"	NAS	122°16'21.9"	NAS	122°16'21.9"	NAS	121°53'59.0"	NAS	121°53'59.0"	NAS	121°53'59.0"	NAS
4.2	Longitude	122°13'53.9"	NAS	37°48'41.1"	NAS	37°54'49.4"	NAS	37°48'41.1"	NAS	37°48'41.1"	NAS	37°25'57.7"	NAS	37°25'57.7"	NAS	37°25'57.7"	NAS
4.4	Altitude	100 m	NAS	13 m	NAS	40 m	NAS	13 m	NAS	13 m	NAS	6 m	NAS	6 m	NAS	6 m	NAS

^aEvidence codes – IDA inferred from direct assay, TAS traceable author statement, NAS non-traceable author statement, nd not determined. These evidence codes are from the Gene Ontology project [90]

Staphylococcus are non-spore-forming, non-motile round-shaped cells (Fig. 2 e-h). They demonstrate habitat preference in the human body with S. capitis mainly being found on the adult head and S. cohnii on the feet [10]. S. hominis is the main colonizer of head, axillae, arms, and legs, and is frequently encountered in nosocomial infections.

Organisms within the genus Microbacterium spp. are yellow-pigmented, aerobic, rod-shaped, Gram-positive bacteria [23] (Fig. 2a). They have been isolated from numerous and variable environments, including soil and water [24], the phyllosphere [25], human patients [26], and a residential toilet [27], and they have been associated with endophthalmitis [28] and catheter infections [29].

Pseudomonas oryzihabitans (Fig. 2d) is a Gram-negative, non-fermenting, yellow-pigmented bacterium [30]. Despite its environmental origin, P. oryzihabitans has been recognized as a potential pathogen in recent years, especially in immunocompromised hosts, both in nosocomial or community-level settings. It can form biofilms in aquifers in association with suspended particulate matter, which can be subsequently entrained into the drinking water distribution systems, posing a potential risk for human health given their resistance to chlorine compared to their planktonic counterparts [13]. This species has been associated with catheter [31] and bloodstream infections, endophthalmitis [32], necrotic enteritis [33], and peritonitis ([34] and references therein). There are two instances in which the source of human infection has been well documented, and the source has been found to be a synthetic sponge, one used by an immunocompromised individual [31] and another in the milk kitchen of a neonatal intensive care unit [33].

Mycobacterium iranicum (Fig. 2b) is a newly described, rapidly growing, orange-pigmented scotochromogenic, non-tuberculous mycobacterial species. Its clinical significance is still under study but it has been associated with patients with pulmonary infections, such as pneumonia, chronic obstructive airway disease, and bronchiectasis [11, 35].

Lastly, Plantibacter (Fig. 2c) are pleomorphic, rod-shaped, yellow-pigmented, aerobic, Gram-positive bacteria that belong to the class of Actinobacteria .

Genome project history

Growth conditions and genomic DNA preparation

Strains were initially collected through environmental sampling (see Classification and features section) and were subsequently deposited into the DMSZ. Glycerol stocks of all isolates were initially grown at 28 °C on LB plates. A single colony was then inoculated in LB and incubated at 28 °C for 18 h (except for M. iranicum strain H39, grown at 37 °C for 5 days). DNA was subsequently extracted from the cultures using the DNeasy Blood and Tissue kit (Qiagen), and the quality was assessed using a NanoDropTM spectrophotometer.

Genome sequencing and assembly

Barcoded Illumina paired-end libraries were generated from all samples using the Nextera XT kit (Illumina). After pooling, the libraries were size-selected for a range of 600–900 bp on a Pippin Prep (Sage Science) and then sequenced on an Illumina MiSeq (Paired End 300 bp). After demultiplexing with a custom script, the reads from each sample were assembled using the A5-miseq pipeline, which automates the process of adapter removal, quality trimming, error-correction, and contig generation [37, 38]. The completeness and contamination of the assemblies was estimated using PhyloSift [22] and CheckM [39]. Across all strains, genome completeness was determined to be a minimum of 98.9%, and the maximum contamination was 0.99% (Additional file 1).

Genome annotation

Phylogenetic comparisons

The genomes of the sequenced isolates were compared to publicly available closely related genomes to determine the ANI values [43]. For those six isolates in which a species epithet was given based on gene trees, ANI values were greater than 90% (Additional file 6), and were greater than 96% for the Staphylococcus isolates. The genomes of those isolates that were assigned to genera based on gene trees were compared to closely related publicly available genomes. For Microbacterium sp. H83, the ANI value with M. hydrocarbonoxydans was 84.1% and for Plantibacter sp. H53 was 87.8% with another Plantibacter sp. (Additional file 6).

Virulence and biofilm production

CoNS are opportunistic pathogens and they do not encode for virulence factors (e.g., exotoxins) commonly found in pathogenic species such as S. aureus . However, they do encode genes related to biofilm formation, persistence and immune invasion [44]. The attachment to a surface is the first step to successful colonization and a precursor for the establishment of infection. In the IMG annotation, we found genes with predicted functions to be associated with cell wall-associated FBP, such as fbe, and several other surface-associated proteins such as a bifunctional autolysin and putative adhesins. However, the gene fbe was not found in S. capitis H36, and another gene known to be important for surface adhesion in Staphylococcus, ebh [44], was not observed in any isolate. Both Ebh and FBP act as adhesins but FBP also acts as an invasin, facilitating binding and internalization in host cells [45]. Additionally, we found genes with predicted functions to be associated with Microbial Surface Components Recognizing Adhesive Matrix Molecules, such as the sdrG gene. Further biofilm accumulation is mediated by exopolysaccharides such as PNAG and PGA. Genes related only to PGA (cap operon), which have been shown to provide resistance to phagocytosis and to a host’s antimicrobial peptides in S. epidermidis [46], were identified. Genes encoding predicted pro-inflammatory molecules with cytolytic and antimicrobial properties such as β-type phenol soluble modulins (PSM) [44] were found in all four staphylococci strains, along with genes encoding their accessory regulator B (Agr) [47]. Other systems important for the regulation of virulence in staphylococci that were found in our strains included the staphylococcal accessory regulator Sar, one of the two components of each of the regulatory systems, SaeRS and ArlRS, and an infection-related protease, ClpC [44].

Antibiotic resistance

We used the Resistance Gene Identifier of CARD [48] to explore possible genes related to antimicrobial resistance. Microbial genome sequencing has the potential to be used as a prediction tool of antibiotic resistance in clinical settings [49, 50], and in fact has been shown to be a promising approach in S. aureus [51, 52] as well as other bacteria [53]. However, at the moment, clinical testing of antibiotic resistance is restricted to PCR-based targeting of specific genes [54, 55], and many of the genes in antibiotic databases have not been verified in clinical settings and are subject to errors in annotation (e.g., [56]). Nevertheless, we surveyed genes predicted to confer antibiotic resistance in order to explore commonalities across the different isolates. Additional file 7 details the Gene ID and other information stemming from the IMG annotation of putative antibiotic resistance genes identified in CARD. Limiting the results to “perfect” and “strict” hits, many of these genes included efflux pumps predicted to confer resistance to more than one class of antimicrobials (e.g., fluoroquinolones, tetracyclines, polymyxins) as well as genes predicted to be associated with resistance to specific antimicrobials (e.g., beta-lactams, aminocoumarins, chloramphenicol, aminoglycosides, and fosfomycin). Some antimicrobial genes were common to many strains; others were limited to specific taxonomic groups. For example, all eight strains were found to contain genes predicted to confer resistance to mupirocin and fosfomycin, while genes for fusidic acid resistance were only observed in S. capitis H65 (Additional file 7).

In addition to general targeting of antibiotic resistance genes, we also looked specifically for genes related to triclosan resistance. TCS is a synthetic antimicrobial agent that is commonly used in home and personal care products such as hand soaps, toothpastes, deodorants, body washes, hand creams, body lotions, and cosmetics. It has been directly associated with the development of multidrug antibiotic resistance in a variety of primarily pathogenic bacteria via in vitro assays [57]. TCS induces resistance through mutations in the gene (fabI) that encodes TCS’s target enzyme (enoyl-acyl carrier protein reductase FabI) through overexpression, or through efflux pumps, with the latter only to be associated with multi-antibiotic resistance [57]. The fabI gene was identified only in one out of four staphylococci isolates, S. capitis H65, as well as in the M. iranicum H39 and P. oryzihabitans H72 genomes. We found several genes related to non-specific multidrug efflux pumps, such as mex genes (mexJKL) in their genomes. The MexJK efflux pump can efflux triclosan, but also requires the outer membrane protein channel composed of the OprM in order to efflux other antibiotics in Pseudomonas aeruginosa [58]. MexJK-OprM was found through CARD in all our genomes, except for Plantibacter sp. H53 that did not carry OprM. The triclosan efflux transporter TriABC–OpmH [59] was only partially present in P. oryzihabitans H72 (TriB was absent). Additionally, P. oryzihabitans H72 was the only isolate to contain an efflux pump predicted to offer triclosan resistance (Additional file 7). Susceptibility to TCS or other antibiotics has not been experimentally tested for the strains described here.