Comparative genomics of cocci-shaped ... - BMC Genomics

5 downloads 11 Views 3MB Size Report
Park,. CA. Illum ina. 573. 33. 3,328,6. 42. 41.3. 3235. 52. 0. 0. (1). 12. PDYX0. 00000 ..... Region field) as queried against the OpenStreetMap ..... RPS-Blast against the National Center for Biotechnology Information conserved domain database (CDD). ..... Rieu-Lesme F, Dauga C, Morvan B, Bouvet OM, Grimont PA, Doré J.

Oliver et al. BMC Genomics (2018) 19:310 https://doi.org/10.1186/s12864-018-4635-8

RESEARCH ARTICLE

Open Access

Comparative genomics of cocci-shaped Sporosarcina strains with diverse spatial isolation Andrew Oliver1,3, Matthew Kay1 and Kerry K. Cooper2*

Abstract Background: Cocci-shaped Sporosarcina strains are currently one of the few known cocci-shaped spore-forming bacteria, yet we know very little about the genomics. The goal of this study is to utilize comparative genomics to investigate the diversity of cocci-shaped Sporosarcina strains that differ in their geographical isolation and show different nutritional requirements. Results: For this study, we sequenced 28 genomes of cocci-shaped Sporosarcina strains isolated from 13 different locations around the world. We generated the first six complete genomes and methylomes utilizing PacBio sequencing, and an additional 22 draft genomes using Illumina sequencing. Genomic analysis revealed that cocci-shaped Sporosarcina strains contained an average genome of 3.3 Mb comprised of 3222 CDS, 54 tRNAs and 6 rRNAs, while only two strains contained plasmids. The cocci-shaped Sporosarcina genome on average contained 2.3 prophages and 15.6 IS elements, while methylome analysis supported the diversity of these strains as only one of 31 methylation motifs were shared under identical growth conditions. Analysis with a 90% identity cut-off revealed 221 core genes or ~ 7% of the genome, while a 30% identity cut-off generated a pan-genome of 8610 genes. The phylogenetic relationship of the cocci-shaped Sporosarcina strains based on either core genes, accessory genes or spore-related genes consistently resulted in the 29 strains being divided into eight clades. Conclusions: This study begins to unravel the phylogenetic relationship of cocci-shaped Sporosarcina strains, and the comparative genomics of these strains supports identification of several new species. Keywords: Sporosarcina ureae, Cocci, Spore-forming, Comparative genomics, Sporosarcina

Background Sporulation is a crucial survival mechanism for many types of bacteria, which can allow spore-forming bacteria to colonize and/or survive in very diverse environments. Oddly, very few spore-forming cocci have been identified or characterized, and the overall knowledge of coccishaped spore-formers is very limited at best. All six described species are Gram-positive, but three were actually designated as coccobacilli [1, 2] or coccoid [3] and have undergone several reclassifications [4, 5]. In fact, Halobacillus halophilis was originally described as coccoid is now referred to as a bacillus [6]. Three cocci-shaped spore-forming species have been characterized, including * Correspondence: [email protected]; [email protected] 2 School of Animal and Comparative Biomedical Sciences, University of Arizona, Tucson, AZ, USA Full list of author information is available at the end of the article

two anaerobic species (Sarcina ventriculi and Sarcina maxima) [7–9] and one aerobic bacteria (Sporosarcina ureae) [10], but the genomics of the different species have not been investigated, particularly Sporosarcina ureae strains. Currently, the genus Sporosarcina is composed almost entirely of bacilli-shaped species [11], while S. ureae is the only established cocci-shaped Sporosarcina. To date, the only analysis surveying the geographic and physiological diversity of S. ureae or any cocci-shaped Sporosarcina strains comes from work done by Bernadine Pregerson over 40 years ago. During the study, over 50 isolates of cocci-shaped Sporosarcina strains were collected from numerous locations around the world, including four different continents. Pregerson originally identified each isolate as S. ureae based on cell morphology, cell arrangement and spore-forming ability, and examined nutritional

© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Oliver et al. BMC Genomics (2018) 19:310

requirements necessary for growth. The study indicated four major nutritional requirement groups, but failed to reveal any correlation between nutritional requirements and habitat [12]. In 1996, Risen studied the electrophoretic mobilities of 24 metabolic enzymes of these cocci-shaped Sporosarcina strains from Pregerson’s study, and concluded that these strains were non-clonal [13]. We hypothesize that based on the studies of Pregerson and Risen, there are novel species of cocci-shaped Sporosarcina. However, the high resolution of whole genome sequencing (WGS) is required to accurately unravel the diversity of these coccishaped Sporosarcina strains. The goal of this study was to investigate the diversity of these cocci-shaped Sporosarcina strains at the genomic level utilizing next-generation sequencing, and further our understanding of the phylogenetic relationship of the genus Sporosarcina. The study is particularly novel as virtually nothing is known about the overall genomics of the genus especially cocci-shaped Sporosarcina strains. To date the only cocci-shaped Sporosarcina genome is a single draft genome of S. ureae (strain DSM 2281, Genbank Accession: NZ_AUDQ00000000), and there has been no analysis or research utilizing this genome. For this study, we sequenced the genomes of 28 cocci-shaped Sporosarcina strains isolated from 13 different locations around the world, with at least one representative of each of Pregerson’s four original nutritional growth requirement groups [12]. Comparative genomics of cocci-shaped Sporosarcina strains will assist in resolving the diversity of these strains, while also providing a future genetic resource for investigating processes such as sporulation in cocci-shaped bacteria. Examining the genomics of phenotypically different and geographically diverse cocci-shaped Sporosarcina strains; we found an average genome size of 3.3 Mb, which encoded for 3222 CDS, 54 tRNAs and 6 rRNAs. Examination of spore genes found the cocci-shaped Sporosarcina strains were lacking many genes that are found in Bacillus; however, spore genes that are present are well conserved among all the strains. In-depth genomic analysis of the strains demonstrated a highly diverse group, as comparative genomic analysis using a 90% identity cut-off revealed 221 core genes or ~ 7% of the genome, while a 30% identity cut-off generated a pan-genome of 8610 genes. Methylome analysis also supported the diversity, as there were numerous different adenine and cytosine methylation motifs, but only one motif was shared between two of six strains grown under identical conditions. Both core and accessory gene diversity failed to correlate with the nutritional growth requirements or location of isolation for the strains. Overall, this study begins to unravel the genomics and phylogenetic relationship of the genus Sporosarcina, particularly revealing the genomic diversity of cocci-shaped

Page 2 of 17

strains, and indicates there are additional cocci-shaped Sporosarcina species. Furthermore, it provides a genetic resource for investigating the sporulation process in cocci-shaped bacteria.

Methods Strains and accession numbers

All strains sequenced in this study were originally identified as S. ureae by Pregerson, based on cell morphology, cell arrangement and sporulation ability [12]. All 28 genomes generated during this study are publically available on NCBI by their respective accession numbers (Table 1). Additional analysis performed with genome sequences not generated during this study were obtained from the NCBI Genome database under the following accession numbers: NZ_AUDQ00000000 (S. ureae DSM 2281). DNA extraction

DNA was extracted as described by Miller et al. [14] with several modifications. Each cocci-shaped Sporosarcina strain was grown up in triplicate in 5 ml of tryptic soy yeast broth (27.5 g Tryptic Soy Broth, 5 g Yeast Extract; Fisher Scientific, Fairlawn, New Jersey, USA) on a rotator at 30 °C overnight. The replicates were then combined, pelleted (10 min @ 12,000 x g), re-suspended in 1.5 ml Tris-sucrose (10% sucrose; Fisher Scientific; 50 mM Tris, pH 8.0; Research Organics Cleveland, Ohio, USA) and diluted to an optical density of 1.6-1.8. Cells were lysed with 500 μl lysozyme (20 mg/ml in 50 mM Tris, pH 8.0; Fisher Scientific) and 300 μl 10% SDS, and then 600 μl EDTA (100 mM EDTA, pH 8.0; Fisher Scientific) was used to buffer the suspended DNA. Twenty μl RNase (10 mg/ml; Fisher Scientific) was added and incubated at 37 °C for 24 h to ensure total RNA removal. Next, 10 μl proteinase K (20 mg/ml; Fisher BioReagents) was added and incubated at 37 °C for 4 h to remove any remaining proteins, and then 265 μl 3 M sodium acetate (pH 5.5; Fisher Scientific) plus 6 ml absolute ethanol were added to precipitate the DNA. Precipitated DNA was transferred and re-suspended in 400 μl EB buffer (Qiagen) by incubating at 37 °C overnight. Next, 400 μl of phenol:chloroform:isoamyl alcohol (Fisher BioReagents) was added, followed by separation via centrifugation (12,000 x g; 5 min), and the aqueous layer transferred. To remove any traces of phenol from the solution, 400 μl of chloroform (Fisher Scientific) was added and mixed by inverting three times, centrifuged (12,000 x g; 3 min), the top aqueous layer transferred to a new tube, and the DNA precipitated again with absolute ethanol. The precipitated DNA was transferred and again re-suspended in 200 μl EB buffer by incubating at 37 °C overnight. Quality, size and quantity of DNA were confirmed with a Nanodrop spectrophotometer (260/280 = 1.8-2.0), gel electrophoresis

Sporosarcina sp. P29

Sporosarcina ureae str. S204

Sporosarcina sp. P33

Sporosarcina sp. P17a

Sporosarcina sp. P8

Sporosarcina sp. P32a

Sporosarcina sp. P37

CP015027

CP015109

CP015207

CP015348

CP015349

USA: Boston, MA

Japan

USA: Los Angeles, CA

USA: Berkeley, CA

Japan: Tokyo

South Africa: Pretoria

USA: San Diego, CA

USA: San Diego, CA

USA: San Diego, CA

USA: Berkeley, CA

USA: Berkeley, CA

USA: Berkeley, CA

USA: Berkeley, CA

USA: Berkeley, CA

USA: Canoga Park, CA

USA: Berkeley, CA

USA: Berkeley, CA

Japan: Tokyo

Japan: Tokyo

Japan: Yokahama

Japan

Illumina

510

539

800

427

475

529

559

504

866

515

442

572

501

531

573

482

479

506

544

614

443

4457

Pacific Biosciences 158

Illumina

1

1

1

1

1

unknown 36

Pacific Biosciences 150

Pacific Biosciences 169

Pacific Biosciences 707

Pacific Biosciences 183

1

43

42

85

46

44

52

18

34

48

58

50

45

66

160

33

72

70

70

23

75

34

38

3,318,232

3,271,521

3,382,744

3,353,765

3,412,428

3235,441

3,362,333

3,437,370

3,438,859

3,318,423

3,293,369

3,363,873

3,401,750

3,219,274

3,339,001

3,306,654

3,314,917

3,384,350

3,269,093

3,382,782

3,382,122

3,328,642

3,313,921

3,316,673

3,314,266

3,262,144

3,252,669

3,379,590

3,169,294

41.4

44.7

41.4

41.2

41.5

44.5

41.5

41.4

41.4

40.6

41.1

41.1

41.3

41.4

41.4

41.2

41.4

42.2

41.1

41.2

41.4

41.3

41.3

41.3

41.3

41.4

44.5

41.5

41.4

3241 56

3155 68

3183 68

3164 69

3204 68

3050 68

3196 60

3374 47

3372 56

3234 54

3185 56

3281 51

3336 60

3149 51

3216 50

3194 50

3228 56

3292 54

3182 51

3246 50

3325 58

3235 52

3215 58

3218 57

3218 58

3203 44

3247 59

3238 50

0

1

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

1

0

0

0 (2)

2 (1)

0 (4)

0 (3)

0 (2)

1 (1)

0 (1)

0 (1)

0 (1)

1 (2)

0 (3)

0 (1)

1 (5)

1 (1)

0 (3)

0 (2)

0 (1)

1 (0)

0 (3)

0 (2)

0 (2)

0 (1)

0 (2)

0 (2)

0 (2)

1 (1)

2 (1)

0 (2)

0 (6)

2

37

28

12

15

29

46

15

16

16

12

19

11

4

5

13

12

13

13

11

19

12

15

17

15

8

16

11

10

tRNAs Plasmid(s) Prophage(s)a Insertion Sequences

3071 55

Coverage Contigs Base pairs (bp) %GC CDS

Pacific Biosciences 174

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Illumina

Numbers outside parentheses are complete prophages, while number within are total number of prophages

a

NZ_AUDQ00000000 Sporosarcina ureae str. DSM 2281 unknown (Type Strain)

Sporosarcina sp. P10

Sporosarcina sp. P16a

PDYN00000000

CP015108

Sporosarcina sp. P16b

PDYO00000000

PDYK00000000

Sporosarcina sp. P17b

PDYQ00000000

PDYP00000000

Sporosarcina sp. P13

Sporosarcina sp. P18a

PDYR00000000

Sporosarcina sp. P12

Sporosarcina sp. P19

PDYS00000000

PDYL00000000

Sporosarcina sp. P1a

PDYT00000000

PDYM00000000

Sporosarcina sp. P21c

Sporosarcina sp. P20a

PDYU00000000

Sporosarcina sp. P26b

Sporosarcina sp. P2a

PDYY00000000

PDYX00000000

Sporosarcina sp. P25

USA: Canoga Park, CA

Sporosarcina sp. P30

PDYV00000000

Japan

Sporosarcina sp. P31

PDZA00000000

PDYZ00000000

Japan

USA: Waikiki, HI

Sporosarcina sp. P34

Sporosarcina sp. P32b

USA: Honolulu, HI

USA: Reseda, CA

PDZB00000000

PDYW00000000

Sequencing platform

USA: Woodland Hills, CA Illumina

Location Isolated

PDZC00000000

Sporosarcina sp. P3a

Sporosarcina sp. P35

PDZD00000000

Sporosarcina sp. P7

PDZF00000000

PDZE00000000

Strain

Accession ID

Table 1 General genomic characteristics of cocci-shaped Sporosarcina strains

Oliver et al. BMC Genomics (2018) 19:310 Page 3 of 17

Oliver et al. BMC Genomics (2018) 19:310

(high single band, little smearing) and a Picogreen dsDNA assay (Life Technologies’ Quant-iT Picogreen dsDNA kit) per the manufacturer’s instructions, respectively. Pacific Biosciences (PacBio) sequencing

Extracted DNA for six cocci-shaped Sporosarcina strains were sent to the UC Irvine Genomic High Throughput Facility for library preparation and PacBio sequencing. Library preparation involved shearing 15 μg of genomic DNA using Covaris G-Tubes, according to the manufactures instructions, resulting in 20 kb fragments used for generating the PacBio sequencing libraries. Blue Pippen (Sage Science) was used to select DNA fragments of 8 kb-50 kb length. Library and sequencing kits used SMRTbell Template Prep Kit (v1.0), DNA Polymerase binding kit P6, and DNA Sequencing Reagent (v4.0), and a 100pM-125pM concentration was loaded onto the SMRT cell. One SMRT cell/strain allowed for > 150× coverage per strain, ample coverage for the construction of de novo genomes. The sequencing run lasted 4 h for each strain. In total, the six strains resulted in an average of 67,798 reads, an average read length of 13.57 kb, and an average of 166× coverage per genome (Table 1).

Page 4 of 17

filtered using Prinseq [17] using parameters (min_qual_ mean 39, ns_max_n 0) to select for high quality reads. Reads were sub-sampled to roughly 80× coverage (based on a genome size of 3.3 Mb) and assembled with the a5 assembler [18] using default parameters. All genomes were annotated using NCBI’s Prokaryote Genome Automatic Annotation Pipeline (PGAAP). Reverse Position Specific-BLAST (RPS-BLAST) was used to find the Cluster of Orthologous Groups (COG) data for each of the genomes. Briefly, each set of query proteins were BLASTed against NCBI’s Conserved Domain Database (CDD) using RPS-BLAST [19]. CDD contains well-annotated multiple sequence alignment models for ancient domains and full-length proteins, allowing for fast identification of conserved domains in the query proteins. After matching the query proteins to the CDD proteins with RPS-BLAST, a perl script [20] was used to pair the correct COG information to each matched query protein. NCBI provides the COG data for each of the genomes contained in the CDD, and this was cross-referenced with the BLAST results to obtain COG information for the query proteins. 16S rRNA analysis

Illumina sequencing

Fifteen μg of genomic DNA from each of the 22 coccishaped Sporosarcina strains were sheared into 400 bp fragments using a Covaris ME220 Focused Acoustic Shearer per the manufacturer’s recommended protocol. Barcoded Illumina sequencing libraries were prepared from the sheared fragments using the NEBNext Multiplex Oligos for Illumina (96 Index Primers; New England BioLabs, Ipswich, MA, USA) and NEBNext Ultra II DNA Library Preparation Kit for Illumina (New England BioLabs) following the manufacturer’s instructions. Barcoded libraries were quality checked using an Experion Automated Electrophoresis System (Bio-Rad, Hercules, CA, USA), quantified using a Picogreen dsDNA assay, and then pooled in equal molar ratios for sequencing. The pooled sequencing libraries were then sent to GeneWiz (South Plainfield, NJ, USA) for paired-end Illumina sequencing (2 × 150 bp) on an Illumina HiSeq X machine. In total, sequencing the 22 strains resulted in a median of 11.7 million reads and 522× coverage per genome (Table 1). Assembly and annotation

PacBio genomes were assembled using SMRTanalysis software (v2.3.0.1), and any genomes that needed further assembly were done in silico by using Geneious software (Biomatters, v9.0.0) to map the corrected reads to the contigs and subsequently linking the contigs into a complete genome sequence [15]. The raw Illumina sequence reads for each of the strains were examined using Fastqc [16] and quality

After annotation, each PacBio complete genome was BLASTed, using the BLAST plugin in Geneious, with a copy of its own 16S rRNA gene to verify that the genomes did not contain unidentified rRNA loci or genes. Due to small known variation between copies of the gene, even within the same genome, a consensus sequence of all copies of the gene within the genome helps capture the most accurate single sequence to use in a comparative analysis [21]. Therefore, for each of the six strains that were sequenced with PacBio, we created a strain specific consensus sequence for the 16S rRNA gene from the different copies of the gene throughout the genome. The 16S gene in the Illumina-sequenced draft genomes was predicted using Barrnap (http://www.vicbioinformatics. com/software.barrnap.shtml), and 198 genus Sporosarcina 16S rRNA gene sequences were downloaded from the Ribosomal Database Project (RDP) [22]. The sequences were aligned using SILVA Incremental Aligner (SINA, www.arb-silva.de) an alignment tool that takes into account ribosomal secondary structure when aligning sequences [23]. FastTree2 was used to build a phylogenetic tree using GTR + gamma20 parameters and 1000 bootstrap replicates [24]. Tree visualization was done using the web-based program interactive Tree of Life (iTOL) [25]. Geographic distribution analysis

To investigate the geographic distribution of the genus Sporosarcina, location data was gathered from the Earth Microbiome Project (EMP) [26] and RDP [22]. Data was

Oliver et al. BMC Genomics (2018) 19:310

extracted from the EMP using the Redbiom tool (https://github.com/biocore/redbiom), which queried the database for the genus Sporosarcina across all available contexts. Sample metadata from matching features returned were parsed for latitude and longitude data. Data from the RDP was downloaded in Genbank format, and the location information (generally City/ Region field) as queried against the OpenStreetMap Nomatim (nominatim.openstreetmap.org) database to obtain approximate latitude and longitude. The data from both sources were then categorized based on general sample type into one of four groups; environment, animal, plant, or human. A map was created using the Matplotlib and Basemap packages in Python, with rendering using GEOS (Geometry Engine - Open Source).

Pan/Core genome analysis

There are currently no standard parameters to elucidate the core genome of related species, therefore we used the following core genome parameters (percent amino acid sequence identity (PI), percent query coverage (PC), and E-value), and set those cutoffs to the strict values of > 90% PI, > 90% PC and > 1e− 4 E-value [27]. To determine the pan genome, we used established sequence parameters (30% PI) [28] to identify orthologous gene clusters. Any cluster generated in this step that was unique to a strain was identified as a strain-specific gene. To generate the core genome sequence comparison data, we created a protein BLAST database of all protein sequences from the 28 sequenced genomes and the downloaded S. ureae DSM 2281 genome. Next, the protein sequences were individually compared for each genome using the BLASTp command from the BLAST+ software [29] against the created protein BLAST database. The output showed if the gene in the query genome were present in all the other database genomes, and how related they were to each other. This resulted in a raw data file for each genome that would contribute to the core genome. A large dataset is generated in the previous step and a python program called Geneparser (https://github. com/mmmckay/geneparser) was written to parse the files and identify core/pan/strain-specific genes present in each genome. Geneparser uses the organism specific amino acid sequence files and generates concatenated gene sequences of all the shared genes, where all shared genes are placed in the same order for each genome. The concatenated sequences were aligned with MAFFT using the default settings [30]. Using the resulting alignment file, a phylogenetic tree was constructed with FastTree [31] using JTT + CAT parameters and 1000 bootstrap replicates.

Page 5 of 17

Methylome analysis

Previously described SMRTanalysis software was used to identify any base modifications by identifying locations of methylation associated with different motifs between all six PacBio sequenced genomes. Additionally, each motif was run through the REBASE database (http://rebase.neb.com/rebase/rebase.html, New England Biolabs) to check if the motifs were associated with any known restriction enzymes and their associated organisms. Only methylation sites that have a Phred-like Quality Value (QV) score of 50 or greater were presented in this study [32]. To visualize the methylomes, all modifications were plotted against each genome using Circos (v.0.69) [33]. Synteny analysis

Contigs for each of the draft genomes (Illumina sequenced strains and S. ureae str. DSM 2281) were reordered using the program Mauve (v2.4.0) [34], with the closest related strain with a complete genome utilized as the reference genome. Next, Artemis Comparison Tool (ACT) comparison files were generated between two targeted genomes for comparison by using the blastall command from BLAST+ software, and this was repeated for all 29 cocci-shaped Sporosarcina genomes. Finally, the alignments between the different genomes were generated and visualized using the program ACT (v13.0.0) [35]. Mobile genetic element analysis

Potential prophage sequences in the genome were identified and categorized (intact, incomplete or questionable) using the PHASTER website (http://phaster.ca/) [36]. Insertion sequences located within the genome were identified using the ISfinder website (https://wwwis.biotoul.fr/) [36] using a cutoff of 75% identity across 75% of the insertion sequence. Additionally, each of the genomes was manually reviewed for the presence of transposase sequences. In order to determine the presence of plasmids in the draft genomes, a database of all the < 200 kb size contigs from all the draft genomes was generated. Then the sequence from the plasmid pSporoP37 identified in PacBio sequenced strain P37 was used to identify potential plasmid sequences among the contigs by using the BLASTn command from the BLAST+ software against the database. Finally, each of the < 200 kb size contigs were further examined using BLASTn against the nonredundant (nr) database (https://blast.ncbi.nlm.nih.gov/ Blast.cgi), and determining any plasmid sequence hits. Whole genome comparison analysis

To compare the average nucleotide identity between each of the 29 cocci-shaped Sporosarcina genomes, a BLAST Atlas comparing all the genomes to each strain

Oliver et al. BMC Genomics (2018) 19:310

as the reference genome were generated using the montage project command in the CGView Comparison Tool [37]. To visualize the results, strains with complete genomes were utilized as the reference genomes. The visualized reference strains included S204 that is closely related to the type strain DSM 2281, and P33 that is distantly related to DSM 2281. The average amino acid identity (AAI) matrix was generated using the Genome-based distance matrix calculator website (http://enve-omics.ce.gatech.edu/gmatrix/), with the default parameters, and the species cutoff value was set at 95% as suggested in Konstantinidis and Tiedje [38].

Results

Page 6 of 17

sides of the country (Hawaii, California, and Massachusetts) and three different continents (North America, Africa, and Asia), we examined the most common environments and spatial distribution for the entire genus Sporosarcina. Genbank and the EMP revealed that the genus Sporosarcina has been found on all seven continents of the world, confirming that not only do cocci-shaped Sporosarcina strains have a diverse global distribution, but the entire genus does as a whole (Fig. 1). Additionally, the cocci-shaped Sporosarcina strains were all isolated from soil environments, but the analysis of the genus did also find species associated with animals, plants, and humans as well as other environments. However, compiling all the different environments finds that Sporosarcina is most commonly associated with terrestrial environments.

Biogeographical analysis of the genus Sporosarcina

As the cocci-shaped Sporosarcina strains used in this study were isolated from soil samples from vastly different geographical locations, including three U.S. states on opposite

Phylogenetic analysis of the genus Sporosarcina

Utilizing public data and the sequences generated during this study (226 16S rRNA gene sequences), we examined

Fig. 1 Geographic distribution of the genus Sporosarcina, using location data from the Earth Microbiome Project (circles), and Genbank (triangles). Colors indicate the general source of isolation, based on sequence metadata, with the exception of orange that indicates cocci-shaped Sporosarcina strains. When exact GPS coordinates were not available, coordinates were approximated based on location data provided. The map was created using the Matplotlib and Basemap packages in Python, with rendering using GEOS (Geometry Engine - Open Source)

Oliver et al. BMC Genomics (2018) 19:310

Page 7 of 17

'

rRNA'

_rRNA

s - 16S_

tigs - 16S

ly.contig

bly.con 8_assem

'p31_S5

S204

'p19_S

A'

rk

en

sis

26

in

a sp

a

nsi

.-

in rc

ke

or

.-

a sp

61_ass embly 'p10_S .conti 66_ass gs - 16S embly _rRNA 'p12_ .conti ' S62_ gs - 16S assem P32a _rRNA bly.co ntigs ' 'p20 - 16S_ a_S4 rRNA 3_as ' 'Spo sem rosa bly.c rcin ontig 'Spo a ur s rosa - 16 eae S_rR rcin -K 'Spo F026 a ur NA rosa ' 347' eae rcin 'Spo -K a ur F026 rosa eae 341' rcin 'Spo -K a ur rosa F026 eae 'Spo rcin 346' -K rosa a aq F0 'Sp 2634 uim rcin oros arin 5' a so 'Sp arci ali or EU na HM osar 3081 'Sp sp 75 .D cin or 62 20' AB os 85 a sp 'Sp ' ar _M .D cin oro 0R AB 'S a sp sa 50 po rc _A - JF .P ro ina TA 'S A3 sa 72 po 92 sp rc 89 -F .G ro - JF in 57 'u N39 sa ag ' IC nc 72 rc 76 lob 9ult 89 'u in 40 80 AY isp a sp nc ure ' ' ult 43 ora 'u .T 92 nc ure d Sp 0 61 52 JQ ult oro d ' -11 38 ur Sp sa 87 ed oro -K rc 28 in Sp J9 sa ' a sp 50 rc or 07 in osa .a 1 G ' rc sp Q in .91 a FJ 10 sp 67 19 .' 81 FJ 81 19 ' 80 85 '

A'

RNA'

16S_rRN ontigs -

_assemb

'p30_S49

s - 16S_rRNA'

- 16S_rRNA'

ntigs - 16S_r

embly.contigs

assembly.co

assembly.c 'p29_S57_

ly.contigs - 16S_rRN

ew

osa or

91

05 Q sa 'un .W ro 11 -H 3' rc sa cu 10 ina B5 rc Q ltu 64 in 11 sp 'un -H red ' a sp .H oros cu Q11 106 Sp W ltu .2' arci or 10 red 10 KP os C2 na 63 ar Sp 01 sp. ' cin -H or 66 osar a sp mix Q6 77 ' ed 03 .cin cu 00 EF a sp ltu 2' 07 .re 37 EU J1 16 Bha -9F 22 ' 39 rgav -K 59 R02 aea_ ' 91 cece 68 ' mbe nsis

'Spor osarci na ko reens 'Sporo is - KM sarcin a ko 2651 reensi 21' s - KF xed cul 7307 ture 66' J2-3F 'Sporo - KR sarcin 029179 a lute ' ola sp. mix KM117 ed cult 164' ure J3-1 4 - KR 'Sporosa 029202 rcina lute ' ola - KC7 71233' 'Sporosa rcina luteo la - LN7 74391' 'Sporosarc ina luteol a - KC89 4026' 'Sporosarcina sp. JSM 221512 2 - KJ685859' 'Sporosarcina sp. IDA3546 - AJ54477 6' 'Sporosarcina sp. LKS-72 - JF502800'

sp. mi

F7

-K

5 79

ure

ult

'Sporosarcina sp. AG25 - KR045826'

'uncultured Sporosarcina sp. - EF075111'

'Sporosarcina sp. 309-63 - AY444830'

'Sporosarcina koreensis - KR028013'

'Sporosarcina psychrophila - KF555615'

'Sporosarcina aquimarina - AJ581991'

'Sporosarcina

ra

o isp

nc

'u

'Sporo

' 11 23

83

JF

0' 92 12 3' ' 66 76 S J 811 ' d lob F6 -F 20 -4 ag -K 63 ' in .5 ra 22 rc F02 po 63 a sp sa -K 02 in bis ro a o lo rc KF or g p ' sa 'S 27 aisp ro ina 63 hil ob rc po p ' 02 gl sa 'S ro 21 a KF 63 oro cin sych aar F02 'Sp ap os por -K n is or ci ' hila 'Sp glob osar 8896 na or hrop ci p 92 yc F ar 'S ' -K a ps oros 6813 12 ' rcin .P 'Sp Q40 sa sp G 6088 ro a I4 M03 rcin 'Spo .L -K rosa a sp ora 18' rcin 'Spo 0474 obisp rosa GU a gl 4' .8 'Spo rcin 8182 . PE rosa KR1 a sp 9'Spo 60' rcin 33 sa O 9644 . ro - KF a sp 'Spo 90' ora rcin 2277 bisp rosa a glo 'Spo 8 - DQ rcin C-C2 83' rosa . PI 9340 po sp 'S - EU cina 300' pora osar globis KM656 'Spor a a sp. ' sarcin sarcin 656304 'Sporo Sporo - KM ed sp. tur ina ' 'uncul rosarc 656291 ed Spo - KM tur sp. 'uncul arcina 94' Sporos - EF0744 tured sp. cul a 'un rosarcin 74492' red Spo sp. - EF0 'uncultu cina osar red Spor 9' 'uncultu - KC49319 AT28 ina sp. 'Sporosarc 8 - DQ227775' na sp. PIC-C 'Sporosarci 827' sp. AG26 - KR045 'Sporosarcina 8' AT27 - KC49319 'Sporosarcina sp.

ro po

.-

a sp

in

rc sa

sp

sa

ro

po

'S

-H

4

RL

.M

a

in

rc

p

'S

'Sporo sarcina

'Sporosarcina ginsengisoli

'

03

43

80

M

-F

M

sp

in rc

P1

'Sp

sarcin a

'p51_S50_ass

'p16a_S54_assembly.contig

an

sa ro

in

wy

po

rc

ne

'S

sa

a

rc

oro

in rc

sa

Sp

'

32

83

29

N

8

20

.b

a

-K

-H

'p32b_S42_

'p21c_S60_assemb

in

d

oro

sa ro

Sp

po

ure

d

yo

9

B6

The 28 cocci-shaped Sporosarcina strains sequenced in the study and the previously sequenced S. ureae str. DSM 2281, revealed some general genomic characteristics of the cocci-shaped Sporosarcina. The average cocci-shaped Sporosarcina genome is 3.33 Mb in size with a GC content range of 41.7–44.0%, encoding for an average 3222 CDS, 54 tRNAs and six ribosomal loci (Table 1), in comparison to the closely related bacillus-shaped S. newyorkensis that has a predicted slightly larger average genome of 3.61 Mb encoding 3673 CDS or over 450 more genes. To establish a general functional role of the 3222 CDS present in the average genome, the clusters of orthologous groups of proteins (COGs) for each of the genomes were determined. On average, 89.2% CDS (2874 out of 3222) could be assigned to a COG category for the 29 cocci-shaped Sporosarcina genomes

rc

ew

00

.W

B1

an

LJ

sp

.W

sa

.W

a

a sp

a sp

P8

ro

'S

ult

re

nc

ltu

'u

ncu

po

sp

in

in

in

dS

a

rc

rc

sa

rc

re

oro

sa

sa

in

'u

rc

ltu

'Sp

rc in

ro

ro

ro

General genomic characteristics of cocci-shaped Sporosarcina strains

cin ar

po

sa

ro sa

po

po

po

that geographic isolation location was a poor predictor of relatedness of the genus Sporosarcina (Additional file 1: Figure S1).

os or

'S

ro

po

ncu

'p25_S45_assembly.contigs - 16S_rRNA'

- 16S_rRNA' 'p1_S65_assembly.contigs A' bly.contigs - 16S_rRN 'p16b_S53_assem - 16S_rRNA' mbly.contigs 'p7_S37_asse RNA' ntigs - 16S_r assembly.co A' 'p18a_S55_ 16S_rRN ontigs assembly.c NA' 'p2_S59_ - 16S_rR gs bly.conti NA' 40_assem - 16S_rR 'p26b_S ontigs ' mbly.c _rRNA 56_asse - 16S 'p17b_S .contigs P17a embly 47_ass ' 'p13_S rRNA - 16S_ tigs P33 ly.con ssemb 51_a P37 'p3_S ' NA S_rR - 16 18' tigs 7745 y.con LN 58' mbl is 6895 _asse kens _S44 - FN wyor 11' sp. 'p35 a ne 9715 rcin cina ' - JX rosa osar 5298 sis 'Spo Spor ' rken M24 red -H wyo 4079 ultu sis a ne ' U99 'unc 99 rken rcin -G 52 wyo sis rosa ' M24 97 a ne 'Spo rken -H 52 rcin 24 ' wyo sis M 87 rosa en a ne rk 40 -H 'Spo ' rcin sis wyo U99 85 en rosa 40 -G a ne rk ' yo 'Spo sis U99 83 rcin en 40 new -G 9 ' rk rosa 9 na yo sis 81 'Spo GU en ew 40 arci ' rk 99 san 56 si oros U yo cin 95 en ' ew -G 68 rk 74 osar an 81 yo or FN 95 ' 68 26 03 FN 7' 91 06 AM 111 sQ -H . I1 sp

'Sp

'Sp

'Sp

po

'S

'S

'S

'S 'S

'u

'p34_S64_assembly.contigs - 16S_rRNA'

the phylogenetic relatedness of the entire genus to begin to understand its diversity. This permitted us to determine the phylogenetic relationship the cocci-shaped Sporosarcina strains have to the other species within the genus Sporosarcina, and demonstrates exactly where these cocci-shaped strains fit in a genus of mostly bacillus-shaped bacteria. The analysis revealed that the closest bacillus-shaped Sporosarcina species to the cocci-shaped strains is S. newyorkensis, which at the 16S rRNA gene level share 98.1% pairwise identity with P37, 97.2% pairwise identity with P13, and 96. 9% pairwise identity with S204 (Fig. 2). The 28 sequenced cocci-shaped Sporosarcina strains clustered together with the few strains of S. ureae from the public data; however, there was still quite a bit of diversity within the group as the strains were separated into 11 different clades. For example, P33, P35, and P37 were grouped into clade 1 with 100% pairwise identity with each other, and 99.3% pairwise identity with the next closest relatives, P3 and P17a. However, they only have 97.6% pairwise identity to P13, which is just slightly higher than P13 to S. newyorkensis. Plotting pairwise 16S rRNA gene identity against distance between the approximate isolation site failed to show a correlation (R2 = .0002), and overall the 16S rRNA gene analysis found

'Sporosarcina sp. ArzA-13 - JQ929013'

- JX517274'

'Sporosarcina sp. TMB3-18-1

5' ginsengisoli - JX51727

- JX949779'

'Sporosarcina

globispora - AM2374 00' 'Sporosarcin a globispora - JF970589' 'Sporosarc ina psych rophila JX429018' 'Sporosa rcina psyc hrophila 'Sporosa - KM87814 rcina psyc 7' hrop 'Sporo hila - KM sarcina 878177 psychr ' 'Sporo ophila sarcin - KM 878206 a sp. 6-2 'Sporo ' FJ7956 sarcin 66' a sp. 'Sporo GIC1 sarcin 7 - AY 439227 a sp. 'Spor DAB_ ' osarci AT A116 na ps 'Spo - JF ychr rosa 7289 ophil rcin 49' 'Spo a - KF a ps rosa ychr 0263 ophi rcin 33' 'Spo a ps la rosa JX42 ychr rcin ophi 9016 'Spo a gl ' la rosa obisp KJ7 rcin 'Spo 1332 ora a ps 8' rosa -K ychr F026 rcin 'Spo ophi 330' a sp rosa la .C 'Spo rcin KF5 L3. a sp 5562 rosa 46 - FM .D 'Spo 0' rcin AB 1737 rosa a gl _A 'Sp 13' TA ob rcin ispo 9oros a sp JF ra 'Sp arci 7289 .D -K or AB na F02 13' osar 'Sp glob _M 6325 cin oro OR is ' a gl por 'S sa 51 po rc ob a- JF ina ro isp KF 'S 72 sa or po glo 89 02 rc a72 ro 63 bis in 'S ' sa 52 JX ag po po rc ' 42 lob ra ro in 'S 90 -K sa isp ag po 17 rc J7 ora ' lob ro 'S in 13 sa po ag isp -K 32 rc ro 'S ora lo 6' M in sa bis po 87 a -K rc ro lu po 81 in te M 48 sa ra a ola 87 ' rc -K lu 81 in te -K M 78 a o la 87 sp F2 ' 82 -J . ID 54 07 N6 73 A ' 6' 44 34 55 41 9' -A J5 44 78 2'

AB700599' sp. eSP04 - AB686545' sp. KW019 ' - AJ544773 IDA0953 cina sp. 9' 'Sporosar - EU18521 rcina sp. rosa 81' red Spo FN4237 'uncultu isoli ginseng 381' sarcina - AB245 'Sporo gisoli raceae a ginsen cin spo sar omono ' 'Sporo Therm 519413 s - KC 34' reensi 3422 a ko - KP sarcin eola 70' 'Sporo 0128 na lut HM osarci la 01' 'Spor luteo 0741 cina - EF sp. osar 299' a or 'Sp R029 arcin -K 33' oros -14F d Sp 2970 re X1 re - EU ultu 277' sp. 'unc cultu F075 ' cina ixed E ar m os 3997 sp. sp. na ina Spor Q68 09' sarc red arci -H ro tu os a ul in or 7946 'Spo ' 'unc 05 mar d Sp - JQ 99 ture aqui 5-3 ' ul P a M34 K 49 'unc rcin -K 81 sp. 87 rosa -02 7' ina M po rc 02 32 'S N -K 13 rosa ' .M rii J7 25 'Spo teu a sp -K 29 i as in p 71 ri 3' rc na 71 KF teu rosa 72 ' iarci pas ri F6 68 a os 'Spo or 95 teu cin -K ' as 68 'Sp 26 rii osar ap FN eu 91 n or st ci 4' .00 'Sp ar pa 77 os a sp - KM 38 5' ina or in p i rc P9 41 rc 'S ri ' sa sa 11 -K eu 67 ro oro ii N4 ast 23 po ur p 'Sp S -J e 66 a d ii in ast re EF ur rc ltu ap .sa ste in cu a sp ro n p rc a 'u a po sa in 'S in ro rc rc o p sa osa 'S ro or po Sp 'S ed ur ult nc 'u 'Sporosarcina

'S

'Sporosarcina aquimarina - JX292285'

'S

8' 47 ' 22 59 J7 38 -K 24 4' sis 86 AB en 43 s8' m B2 83 ro ' nsi A 2 e 0 sa 80 m a 15 sis Q2 ro in 37 en sa -H rc m a EU sa sis ro in .ro en rc ' ' m a sa po sa a sp 17 43 in ro 'S in ro 66 74 rc rc po 01 39 a sa sa sa 'S in ro KP ro GU rc .po po sa 3' 'S ro dS a sp .B 27 re po 37 cin ' a sp 'S ltu ar 47 76 in os 89 ncu sarc - JX or 36 'u -5 Sp KT 06' oro .W red 3'Sp 3432 ltu a sp HQ JF cu 5' cin .K a'un 1332 osar a sp arin or KJ7 rcin 03' uim 'Sp a8783 rosa a aq arin KM rcin uim 'Spo 74' arosa a aq 2670 arin rcin 'Spo AM uim 6' arosa a aq 1794 arin 'Spo rcin uim KP7 rosa aa aq 73' arin 'Spo rcin 5172 uim rosa a - JX a aq 'Spo arin rcin 39' uim rosa 3976 a aq 'Spo - FN rcin PA7 43' rosa sp. 3976 'Spo ina rc ' 1 - FN rosa 6326 . PN 'Spo KF02 na sp na osarci 323' uimari 'Spor a aq KF026 na sarcin ' imari 'Sporo a aqu 026324 sarcin a - KF imarin 'Sporo 785' a aqu FR873 sarcin rina 'Sporo 00' aquima - FN6466 sarcina -PON4 'Sporo SS-2009 rcina sp. 1' 'Sporosa AM90077 PA21 rcina sp. 'Sporosa rangiaceae Streptospo 607' 9-PA7 - FN646 a sp. SS-200 'Sporosarcin HG421018' aquimarina -

'Sporosarcina aquimarina - KP745566'

- HQ284937' 'Sporosarcina sp. 11AL4 FN646611' 'Sporosarcina sp. SS-2009-PA3 -

'Sporosarcina

po ro 'S sa po rc 'u ro in nc sa a ult gin rc ure in se a ng d gin po Sp iso se ro oro li ng sa -H sa rc iso rc in Q li in a sp 'S 33 -K a sp po 15 .L F7 ro 32 .KS sa 71 'Sp ' EF rc 40 -4 in oro 507 8' a so 'Sp 37 sa 'Sp JF rc li 18 or 5 or ina -K 02 os ' osar 'un ar 78 M so cin cu cin 3' 00 li lt a 9 a u KF 'Spo 13 so red gin li 2' 37 rosa sen Sp 86 KM oros giso rcin 4 6' 00 arci li 'Spo a sp 91 EU na 34 .B rosa sp. ' 30 GU rcin 81 -E MS1 'Spo 21 a sp F07 ' 01 rosa .T 43 -K HG rcin 98 C99 'Spo -b65 ' a sp 10 rosa .C -K 36 rcin L2. F99 ' 'Spo 96 a sp 9707 rosa - FM .C ' rcin L7 17 .2 a sp 3614 5'Spo . TB FM ' rosa 1741 21-8 rcin 295' a sp 'Spo KJ5 . LL rosa 4213 rcin 512 'uncu 7' a glo -K ltured J542 bisp 140' ora Spor - EF osar 'Sporo 0108 cina 50' sarcin sp. EF66 a glo 'uncul bis 5963 tured pora ' Sporo - KF 9566 sarcin 'uncul 59' a sp. tured - KJ Sporos 473567 arcina ' 'Sporo sp. GQ911 sarcin a kor 020' eensis 'Sporosa - KM 009125 rcina aqu ' imarina - KP7 'Sporosa 45564' rcina aqui marina - KP74557 'Sporosarc 0' ina aquim arina - KP74 5591' 'Sporosarcin a sp. A3M1 S16 - KP83 6259' 'Sporosarcina sp. 13x - KR902 559' 'Sporosarcina sp. NB90 - GU479626 ' 'Sporosarcina aquimarina - KM036087'

ina 'Sporosarc

KEY

Cocci Sporosarcina species sequenced by this study Cocci Sporosarcina species Bootstrap values (0.5 1.0, scaled)

Fig. 2 Phylogenetic tree based on all the 16S rRNA gene sequences of the genus Sporosarcina deposited on the Ribosomal Database Project. Sequences were filtered to be greater or equal to 1200 bases long, good sequence quality, and both cultured and uncultured organisms. The sequences were aligned using SILVA, and refined using MUSCLE. The tree was built using FastTree and visualized in iTOL

Oliver et al. BMC Genomics (2018) 19:310

(Fig. 3). Excluding categories R (General function prediction only) and S (Function unknown) the top three COG categories were E (Amino acid transport and metabolism), K (Transcription), and P (Inorganic ion transport and metabolism), while the lowest three with at least two genes were D (Cell cycle control, cell division, chromosome partitioning), V (Defense mechanisms), and U (Intracellular trafficking, secretion, and vesicular transport). Presence of mobile genetic elements was variable between the genomes depending on the type of element. Only strains P35 and P37 were found to contain a plasmid. Whereas, the cocci-shaped Sporosarcina genomes ranged from 2 to 46 insertion sequence (IS) elements, with an average of 15.6 per genome. The amount of IS elements present in a genome were widely variable among the different strains, S204 had 46 IS elements, while the closely related type strain DSM 2281 only had 2 IS elements. Moreover, the genomes of the closely related strains, P33, P35 and P37, had 29, 16 and 37 IS elements, respectively. Although only having draft genomes might be having a slight effect on the analysis, those with complete genomes still ranged from 12 to 46 IS elements. Additionally, the cocci-shaped Sporosarcina

Page 8 of 17

genomes contain a range of 1 to 6 prophages with an average of 2.3 per genome, however the vast majority were incomplete prophages. The cocci-shaped Sporosarcina genomes demonstrate a significant amount of synteny, although seven strains (P10, P12, P19, DSM 2281, S204, and P18a) contain a single ~ 791 kb inversion located between 1,498,920 and 2,290,700 (S204 genome coordinates). The inversion appears to be due to the result of mobile genetic elements, as it contains an ISSpglI element (Sporosarcina globispora) on both ends (Additional file 2: Figure S2). Analysis of methylome

A subset of six cocci-shaped Sporosarcina strains that were sequenced exclusively with PacBio sequencing technology allowed for methylome analysis. This analysis revealed that at the epigenetic level there are significant differences among these six genomes, as only one shared DNA methylation motif (TTCGGA), between P33 and P37, was identified between the genomes under the growth conditions used for the study. Interestingly, S204 is the only strain to contain a Dam methylation motif, which also happened to be the only methylated motif throughout the genome. Strains P17a, P32a, and P33 each contain multiple methylated motifs including both

Fig. 3 Average number of genes for 29 cocci Sporosarcina strains in the different categories of the clusters of orthologous groups (COG) using RPS-Blast against the National Center for Biotechnology Information conserved domain database (CDD). Error bars represent standard deviation

Oliver et al. BMC Genomics (2018) 19:310

m6A and m4C methylation, while P37 lacks any apparent cytosine methylation. Additionally, P8 has no currently identified adenine or cytosine methylation motifs, but did have several base modifications of unknown types (Table 2; Additional file 3: Figure S3). The different phylogenetic analyses used throughout the study places these six strains in five different clades, suggesting some genomic diversity among these strains, which is further supported by high level of variation at the epigenetic level. However, even those that belong in the same clade (P33 and P37) and contain the only shared methylation motif, variation is still seen in number of motifs: P37 had three motifs methylated, while P33 had nine. Comparative genomic analysis of cocci-shaped Sporosarcina strains

Based on the initial diversity observed between the coccishaped Sporosarcina strains at the 16S rRNA gene level, we investigated how that diversity held up at the genomic level by utilizing whole genome sequence (WGS) analysis to reveal higher levels of resolution between stains. The pan-genome analysis of the 28 sequenced cocci-shaped Sporosarcina strains plus the previously sequenced S. ureae str. DSM 2281 using 30% identity cutoff contains a total of 8610 genes. Interestingly, core genome analysis of all 29 cocci-shaped Sporosarcina strains at a 90% identity cutoff established only 221 core genes or only about 7% of the total genome (Fig. 4). Overall, the presumably more reliable and higher resolution of a phylogenetic analysis based on the identified core genome placed the 29 cocci-shaped Sporosarcina strains into eight clades, which was down from the 11 clades from the 16S rRNA gene level analysis (Fig. 5). The cocci-shaped Sporosarcina genomes averaged 57 strainspecific genes, but the amount varied widely. Strains that lacked a very close phylogenetic neighbor tended to have a larger pool of strain-specific genes, for example strain P13, which forms its own clade, has 213 strain-specific genes or 6.6% of the genome. Phylogenetic analysis based on the accessory genes generated a phylogenetic relationship that was nearly identical to the core genome as it also separated the strains into eight different clades. All strains were also placed in the exact eight clades, but there were minor modifications as to the relationship within the clade (Additional file 4: Figure S4). Furthermore, the core and accessory gene phylogenetic relationships failed to cluster strains based on isolation location or Pregerson’s original nutritional requirement grouping phenotypes. Sporulation is a key characteristic of the genus Sporosarcina, therefore, to further understand the diversity between the cocci-shaped Sporosarcina strains we examined spore-related genes. Examining coccishaped Sporosarcina genomes for 66 spore-related genes present in Bacillus subtilis, revealed that many

Page 9 of 17

of these genes are actually missing. Nevertheless, the overall presence or absence of these sporulation genes in cocci-shaped Sporosarcina strains was fairly well conserved across all the strains (Fig. 5). However, the amino acid identity between the strains did have some variation, as phylogenetic analysis of the 29 cocci-shaped Sporosarcina strains based on the sporerelated genes generated a tree nearly identical to the core genome tree. In fact, it was identical to the phylogenetic tree generated by the accessory genes, and placed the strains into the exact same eight clades as both the core genome and accessory. The sporulation gene tree also generated the same shifts in the relationships between strains within the same clades as the accessory gene tree (Data not shown). Potential novel cocci-shaped Sporosarcina species

Since the phylogenetic relationship based on core genes, accessory genes and spore-related genes all indicate that these 29 cocci-shaped Sporosarcina strains should be separated into eight different clades, we examined if these were potentially novel species of cocci-shaped Sporosarcina based on the average amino acid identity (AAI) between strains (Fig. 6). Strains within clade 1 (P35, P33 and P37) share 99.3% AAI, but only 82% AAI with the clade 2 (P13) and 86% AAI with clade 8 (P1a, P8, P21c, P16a, and P25). Whereas, strains within clade 7 (P7, P2, P16b, P18a, and P34) share 97.1% AAI, but only 89% AAI with clade 3 (P26b, P32a, and P17b). Furthermore, clade 6 (P10, P12, P19, S204 and DSM 2281) that includes the S. ureae type strain (DSM 2281) share 97.8% AAI between the strains, but just 93.7% AAI with the nearest neighbor clade 5 (P17a and P3a). Overall, all the strains within a clade share the 95% AAI minimum for identical species, but none of the clades share the 95% minimum between them (Fig. 6). Furthermore, to investigate the average nucleotide identity (ANI) variation between the different clades, and also further the analysis at the DNA level, BLAST atlases using each strain as a reference were produced using the BLASTn command to compare each of the genomes against the reference genome (Fig. 7). For visualization only those strains with a complete genome were utilized as a reference genome, therefore setting S204 as the reference demonstrates that only the strains present in clade 6 (P10, P12, P19, S204 and DSM 2281) share ≥94% ANI across the vast majority of the genome. Whereas, all the remaining 24 strains share ≤92% ANI with S204 and the other strains present in clade 6. On the other hand, applying P33, a member of clade 1 (P33, P35, and P37), as the reference genome found it shared ≥96% ANI with the other strains in the clade. However, all 26 other cocci-shaped Sporosarcina strains

Oliver et al. BMC Genomics (2018) 19:310

Page 10 of 17

Table 2 Methylation profiles of six strains of cocci-shaped Sporosarcina Recognition Sequence

Type/ subtype

Unique

% Detected

Coverage

BCGCCGANRD

II

yes

50.6

90.6

CCGYAG

II

yes

100

90.5

CGCCGTTNNNB

II

yes

21.4

89.2

CGCCGVNY

II

yes

59.3

93.2

CGGCGNYD

II

yes

42.5

91.5

CGSCGNBV

II

yes

18.3

84.9

GCGGTAVYR

II

yes

21

92.2

TGAAATT

II

yes

99.9

82.5

Potential Methylases

Methylation Type

SurP17aORF5150P

6 mA

SurP17aORF5155P

6 mA

Sporosarcina sp. P17a

Sporosarcina sp. P32a CCAG

II

no

30.9

74.5

CAAYNNNNNGTAA

I gamma

yes

100

80.6

M.SurP32aI

6 mA

ACRGAG

II G,S,gamma

yes

100

82.4

SurP32aII

6 mA

II

no

29.2

84.2

CGTCGANA

II

yes

73.9

351.8

CGTCGTNGD

II

yes

21.2

363.7

Sporosarcina sp. S204 GATC

6 mA

Sporosarcina sp. P8

CGTCGTNYR

II

yes

76.8

350.6

CGTCGTTNY

II

yes

44.8

356.2

CGWCGVNB

II

yes

69.2

355.3

DNGCCACNCA

II

yes

23.5

365.9

GGGGCATNNNNNNNH

II

yes

16.9

341.2

GACGAG

II

no

99.6

72.8

SspP37ORF15190P

6 mA

GCCATC

II

no

100

73.2

M.SspP37ORF13670P

6 mA

TTCGAA

II

no

100

71.7

M.SspP37I

6 mA

ACGNNNNNNTAYNG

I

yes

100

84.6

ANCDGGGAC

II

yes

28.4

82.7

DNCGCGGTANY

II

yes

26.5

86.3

GGHANNNNNNTTTA

I

yes

99.8

84.7

GTCCCBVNY

II

yes

52.2

87.6

GTCCCGBANNNNNNH

II

yes

29.4

85.9

SGTCCCNY

II

yes

23.2

85.6

TTCGAA

II

no

100

81.7

M.SspP33I

6 mA

GGGAC

II

no

100

84.8

SspP33II

6 mA

Sporosarcina sp. P37

Sporosarcina sp. P33

share ≤86% ANI with P33, which also demonstrates at the DNA level the diversity between the eight different clades. Overall, the AAI and ANI variability between the strains present in the eight different clades suggests these clades may represent novel species of cocci-shaped Sporosarcina strains.

Discussion Members from the genus Sporosarcina have been isolated from very diverse environments such as soil [39], food production facility [40], or clinical samples [41] just to name a few. In the 1970s, Pregerson isolated over 50 cocci-shaped Sporosarcina strains from three different continents, and

Number of genes

Oliver et al. BMC Genomics (2018) 19:310

Number of genomes

Fig. 4 The core and pan-genomes of cocci-shaped strains of Sporosarcina. Using BLASTp, and cutoff values of 90% amino acid identity across 90% of the gene, 221 conserved core genes were identified among the strains. The pan genome has 8610 genes using 30% identity across 70% of the gene as cutoff parameters

found they were most commonly isolated from soils exposed to human or animal urine [12]. However, no study has collective examined the general global distribution of the genus Sporosarcina, or the most common environment associated with members. Investigating the geographic distribution of the genus Sporosarcina showed, similar to the cocci-shaped Sporosarcina strains, the other species have a global dissemination. Furthermore, the genus could be found in terrestrial, human, animal, and plant environments, but was most commonly associated with terrestrial colonization, again similar to the soil associated coccishaped Sporosarcina strains. It may be that other environments such as plants or animals get colonized through soil contamination, but additional surveillance studies are needed to determine for sure. Ultimately, the global distribution of cocci-shaped Sporosarcina strains appears to be similar to the genus Sporosarcina as a whole. Phylogenetic relatedness based on the 16S rRNA gene indicates that these cocci-shaped Sporosarcina strains including S. ureae belong in the genus with the other bacilli-shaped Sporosarcina species. The 16S rRNA gene analysis predicts the closest neighbor is the bacillishaped S. newyorkensis, but it also confirmed the diversity of the cocci-shaped Sporosarcina strains predicted from previous studies, as it divided the 28 sequenced strains, DSM 2281, and four additional S. ureae 16S rRNA sequences into 11 clades. However, analysis failed to find a direct correlation between distance isolated and 16S rRNA gene similarity. Additionally, using the single 16S rRNA gene did not provide the resolution needed to decipher the phylogenetic relationships of the coccishaped Sporosarcina strains. For example, organisms

Page 11 of 17

that share at least 97% pairwise identity at the 16S rRNA gene level, along with other phenotypic markers, are considered the same species [42]. Nonetheless, S. newyorkensis shares 98.1% pairwise identity with P33 and P37, and 97.2% pairwise identity with P13, although they are lacking the phenotypic markers, at the 16S rRNA gene level it suggests they are the same species. In fact, it is not until the distant clades including S. ureae DSM 2281 16S rRNA gene identity drops below the 97% level (96.9% pairwise identity) with S. newyorkensis, but current research has suggested moving the cutoff to 98. 65% particularly when combined with other genomic metrics such as ANI, might clarify the process of distinguishing novel species [43]. In fact, using 98.65% would resolve the issue between the cocci-shaped Sporosarcina strains and S. newyorkensis. Moreover, many of the cocci-shaped Sporosarcina strains would also not be considered the same species as S. ureae, which supports the comparative genomic analysis data too. Notwithstanding, the analysis does suggest that the cocci-shaped Sporosarcina strains P33, P37, and P35 are closely related to the bacilli-shaped S. newyorkensis, thus future comparative genomic studies could help resolve how the cocci-shaped strains fit in the genus, as well as a genetic resource for investigating cell morphology and sporulation in cocci-shaped bacteria. The diversity predicted by the 16S rRNA gene analysis, nutritional growth requirement analysis, and enzymological analysis were also supported by methylome analysis under the growth conditions utilized in this study. In fact, no motif was common to all six strains and only one motif (TTCGAA) of 31 determined by PacBio sequencing, is shared between P33 and P37. Interestingly, both of these strains, which share 99% AAI across the entire genome still contain substantial variation in their methylomes. Moreover, P33 and P37 have different phenotypes as they were grouped into different nutritional groups by Pregerson, which suggests that methylome difference could result in gene expression differences in the strains [44]. We hypothesize that these variations in the methylome allow the closely related cocci-shaped Sporosarcina strains to adapt to different environments or slightly different ecological niches, as these two strains were isolated on opposite sides of the world. Future work, such as epigenomic and transcriptomic studies of the various cocci-shaped strains, is needed to completely address the role DNA methylation has in variation of phenotype. The study found that on average a cocci-shaped Sporosarcina strain contains a 3.3 Mb genome that encodes for 3222 CDS, approximately 11.1% of those CDS have only a general function predicted, 8.9% CDS an unknown function, and 8.7% CDS for amino acid transport and metabolism. In the soil, nitrogen

Oliver et al. BMC Genomics (2018) 19:310

cwlJ gerAA gerAB gerAC gpr rapA safA sigF spo0A spo0B spo0E spo0F spo0M spoIIAA spoIIAB spoIIB spoIID spoIIE spoIIGA spoIIIAA spoIIIAB spoIIIAC spoIIIAD spoIIIAE spoIIIAF spoIIIAG spoIIIAH spoIIID spoIIIE spoIIIJ spoIIM spoIIP spoIIQ spoIIR spoIISA spoIISB spoIVA spoIVB spoIVCA spoIVFA spoIVFB spoVAA spoVAB spoVAC spoVAD spoVAEA spoVAEB spoVAF spoVB spoVD spoVE spoVFA spoVFB spoVG spoVID spoVIF spoVK spoVM spoVR spoVS spoVT sspA yabP yabQ yqfC yqfD

Page 12 of 17

Sporosarcina sp. P35 Sporosarcina sp. P37 Sporosarcina sp. P33 Sporosarcina sp. P13 Sporosarcina sp. P26b Sporosarcina sp. P32a Sporosarcina sp. P17b Sporosarcina sp. P20a Sporosarcina sp. P29 Sporosarcina sp. P31 Sporosarcina sp. P30 Sporosarcina sp. P32b Sporosarcina sp. P17a Sporosarcina sp. P3a Sporosarcina sp. P10 Sporosarcina sp. P12 Sporosarcina sp. P19 Sporosarcina sp. DSM2281 Sporosarcina sp. S204

Minimal acetate media without added growth factors

Sporosarcina sp. P34 Sporosarcina sp. P18a Sporosarcina sp. P16b

Definable growth factor requirements. Require biotin Complex, undetermined growth factor requirements

Sporosarcina sp. P8

Nutritionally fastidious

Sporosarcina sp. P21c

Sporosarcina sp. P2a Sporosarcina sp. P7 Sporosarcina sp. P1a

Sporosarcina sp. P16a

Bootstrap values (0.9 -1.0, scaled)

Sporosarcina sp. P25

USA: Honolulu, HI USA: Boston, MA Japan: Tokyo USA: San Diego, CA Japan: Tokyo Japan USA: Berkeley, CA USA: Berkeley, CA Japan: Yokahama Japan Japan Japan USA: Berkeley, CA USA: Reseda, CA USA: San Diego, CA USA: San Diego, CA USA: Berkeley, CA Type Strain South Africa: Pretoria USA: Waikiki, HI USA: Berkeley, CA USA: Berkeley, CA USA: Canoga Park, CA USA: Woodland Hills, CA USA: Canoga Park, CA USA: Los Angeles, CA USA: Berkeley, CA USA: Berkeley, CA Japan: Tokyo

Fig. 5 Phylogenetic tree of the cocci-shaped Sporosarcina strains based on the core genome, and rooted based on the 16S rRNA gene tree. Tree was built using core genes shared at 90% amino acid sequence identity and 90% sequence coverage by all strains. Those sequences were concatenated in the same order for each genome, aligned using MAFFT, and the tree was built using FastTree. The genome names are colored to reflect the phenotypic class they were assigned during their initial isolation by Pregerson (1973). Spore genes found in B. subtilis and other spore-forming bacteria were protein-blasted to determine whether similar sequences exist in cocci-shaped strains of Sporosarcina

is very limited for plants, bacteria and other microbes, therefore there is a high level of competition for available nitrogen. Ammonium is a preferred form of nitrogen for many soil bacteria and fungi [45], however amino acids, such as glutamine and glutamate, are another critical source of nitrogen for bacteria [46]. Cocci-shaped Sporosarcina strains contain urease that can break urea down into ammonia when it is available, and probably represents a major method of nitrogen acquisition for these strains based on Pregerson isolating from soils with frequent urine exposure. Yet it is possible that the high level of amino acid transport and metabolism genes are present as a backup system to acquire critical nitrogen from the soil environment in the absence of urea. Again, future work examining the role urease and amino acid acquisition has among the coccishaped Sporosarcina strains survival in different types of soils is needed to directly answer these questions. The exact role mobile genetic elements have in the diversity of cocci-shaped Sporosarcina strains is unclear, particularly since the presence or absence of certain types of mobile genetic elements were quite variable. For example, out of the 29 genomes analyzed during this study only the closely related strains P37 and P35

contained a plasmid. Additionally, on average there were 2.3 prophages per genome, but those were almost always incomplete or prophage scars, as only 27.6% (8 out of 29) of the genomes contained an intact prophage. There was definitely no evidence of large fluctuations in the genome size due to the presence or absence of prophage sequences, particularly like Escherichia coli where the prophages are a critical evolutionary driver and cause massive changes to the genome size [47]. However, there was a lot of variability with the amount of IS elements present in the various genomes of cocci-shaped Sporosarcina strains, and IS elements are also a known driver of E. coli evolution particularly O157:H7 [48]. In fact, the study found a role in the cocci-shaped Sporosarcina evolution, as there is fairly strong synteny among most of the strains, except for seven strains that contain an approximately 791 kb (~ 24% of the genome) inversion that is due to the presence of ISSpglI elements (Sporosarcina globispora) on each end of the inversion. Yet, the exact role mobile genetic elements particularly IS elements play in the evolution and diversity of cocci-shaped Sporosarcina strains will need particular in-depth analysis beyond the scope of this current study. Sporulation is a key characteristic of the genus Sporosarcina including the cocci-shaped strains, but analysis

Oliver et al. BMC Genomics (2018) 19:310

Page 13 of 17

Fig. 6 Average amino acid identity matrix between the 29 cocci-shaped Sporosarcina strains sequenced during this study. Thick black boxes indicate species 95-96% cutoffs as proposed by Konstantinidis and Tiedje (2005), and were generated with the Genome-based distance matrix calculator website

with 66 spore-related genes from Bacillus subtilis found that at ≥20% AAI cocci-shaped Sporosarcina strains only contained 38% (25 out of 66) of those genes. Yet, those 25 spore-related genes are well conserved among all 29 strains of cocci-shaped Sporosarcina, nonetheless, the other 41 spore-related genes have AAI variation among the cocci-shaped Sporosarcina. These 41 sporerelated genes may be critical drivers of the diversity of these strains, as phylogenetic relatedness analysis based on all 66 of the spore-related genes generated a phylogenetic tree nearly identical to that of the core genome. It is possible that there are novel spore-related genes not identified in this study, but it does provide a framework for future work investigating sporulation in coccishaped bacteria. Inferring bacterial relationships based on whole genome DNA sequences is a difficult endeavor due to the vast amount of sequence shared during horizontal gene transfer (HGT). To counter this, studies indicate that using a smaller subset of “core” genes would minimize the effect of HGT skewing phylogenetic analysis [49].

While there are multiple methodologies to generate a core genome, a consensus as to what defines a core genome does not exist. Furthermore, using WGS data to define novel species also still under debate, but Konstantinidis and Tiedje determined that the 70% DDH species cutoff corresponds with an average amino acid identity of 95-96% [38, 50]. Thus, using a series of highly conserved genes, such as those found in a core genome, we can resolve the phylogenetic relationships of very closely related strains and identify novel species among those strains. Utilizing a 90% AAI cut-off generated a core genome of 221 genes or 7% of the genome among the cocci-shaped Sporosarcina strains. Using a lower 75% AAI cut-off expands the core genome to 881 genes or 27.3% of the genome. However, these are both lower than other described core genomes, for example Rasko et al. used an 80% AAI cut-off and found E. coli had a core genome of 2344/5020 or 46.7% of the genome [51]. While, using a 90% cut-off, Leekitcharoenphon et al. found Salmonella enterica had a core genome of 2882 genes or approximately 64% of the genome [52]. Again, this low level of core genes further supports the large amount of genomic diversity

Oliver et al. BMC Genomics (2018) 19:310

a

Page 14 of 17

b

Fig. 7 BLAST Atlases comparing the 29 cocci-shaped Sporosarcina genomes against one of two complete reference genomes (S204 and P33), circular plots were generated with CGView Comparison Tool using BLASTn. Genomes are arranged with the genetically closest to reference genome on the outer ring, and most distantly related on the inner most ring. a S204 reference genome; ordered from outer ring to inner ring: 1) Forward CDS, tRNA and rRNA; 2) Reverse CDS, tRNA and rRNA; 3) S204; 4) DSM 2281; 5) P19; 6) P12; 7) P10; 8) P29; 9) P31; 10) P30; 11) P32b; 12) P17a; 13) P20a; 14) P3; 15) P1; 16) P21c; 17) P16a; 18) P25; 19) P18a; 20) P2; 21) P8; 22) P16b; 23) P7; 24) P34; 25) P26b; 26) P32a; 27) P17b; 28) P37; 29) P35; 30) P33; 31) P13. b P33 reference genome; ordered from outer ring to inner ring: 1) Forward CDS, tRNA and rRNA; 2) Reverse CDS, tRNA and rRNA; 3) P33; 4) P37; 5) P35; 6) P12; 7) P10; 8) P17a; 9) P31; 10) P30; 11) P29; 12) P32b; 13) P19; 14) S204; 15) P3; 16) P17b; 17) P26b; 18) P20a; 19) P32a; 20) DSM 2281; 21) P34; 22) P21c; 23) P8; 24) P16a; 25) P25; 26) P1; 27) P16b; 28) P18a; 29) P7; 30) P2; 31) P13

among these cocci-shaped Sporosarcina strains. Ultimately, the higher resolution provided by WGS and comparative genomics refined the 29 cocci-shaped Sporosarcina strains done from the 11 clades predicted by 16S rRNA gene analysis to just eight clades. In fact, phylogenetic relatedness predicted by core gene, accessory gene or spore-related gene analysis all place the strains into the exact same eight clades. In fact, using the Konstantinidis and Tiedje suggested 95% AAI cut-off for species, only those strains clustered within a common clade would be the same species. Additionally, clade 1 (P33, P35, and P37) only has 85.9% AAI and clade 2 (P13) only 81.5% AAI to the other cocci-shaped Sporosarcina strains, supporting that both clades comprise novel coccishaped Sporosarcina species. Again, all these results support Pregerson’s result from the original 1973 phenotype study of these strains, as she predicted that the cocci-shaped Sporosarcina strains isolated from around the world were quite diverse. As more genomes are becoming sequenced, the definition of what constitutes a prokaryotic species is being challenged. Until recently, a prokaryotic species was defined as a strain (including the type strain) characterized by certain phenotypic consistency, 70% DNA-DNA hybridization (DDH) and over 97% identity of the 16S rRNA gene [38]. With the advent of affordable whole genome DNA sequencing (WGS), the ability to study organisms at the individual nucleotide level allows for refining phylogenetic relationships that were originally

based on the classic polyphasic approach. There currently exists a push to include parameters derived from whole genome sequencing, such as average nucleotide identity or average amino acid identity to delineate species [53–55]. One such study used ANI and alignment fraction to calculate the probability that two genomes belong to the same species, showing these metrics are often far more accurate than DDH and 16S rRNA identity. Moreover the same study shows these metrics will help reclassify organisms that currently have the same taxonomic classification, but cluster separately based on genomic metrics [56]. In this study, we show that 29 strains of cocci-shaped Sporosarcina are much less related to each other than the polyphasic metrics would suggest. Although it has been suggested for a long time, Hug et al. demonstrated that using more genes resolved phylogenetic relationships particularly those that were more ambiguous when using just one gene [57].

Conclusions In conclusion, this is the first study to investigate the genomics of not just cocci-shaped Sporosarcina, but any species of the genus Sporosarcina, in fact, the study more than tripled the amount of WGS sequence data available for the genus Sporosarcina. During this study, genomes of 28 cocci-shaped strains of the genus Sporosarcina were sequenced and characterized, and comparative genomics of these cocci-shaped strains

Oliver et al. BMC Genomics (2018) 19:310

isolated from around the world revealed a high level of diversity. In fact, we have shown that, although they share morphological, biochemical, and 16S rDNA similarity, they are remarkably variable in their gene content, genome sequence identity, and methylomes. Based on the phylogenetic relationship generated from either core genes, accessory genes or spore-related genes these cocci-shaped Sporosarcina strains are always divided into eight different clades, thus suggesting there may be up to seven novel cocci-shaped Sporosarcina species in addition to S. ureae. Although it requires additional phenotypic analysis to confirm these different clades, based on the strong AAI and ANI variation among the strains, we conclude that clade 1 (P33, P35, and P37) and clade 2 (P13) represent new species.

Additional files Additional file 1: Figure S1. Correlation of the pairwise distance isolated compared to 16S rRNA gene similarity. All sequence information was retrieved from the Earth Microbiome Project or Genbank. Red line indicates an R value of 0.0147. (PPTX 178 kb) Additional file 2: Figure S2. ACT (Artemis Comparison Tool) alignment plot of strains of cocci-shaped Sporosarcina. Bands indicated shared genes. Red bands are genes shared in the same direction and blue bands are genes share in reverse directions (sequence inversions). (PPTX 6951 kb) Additional file 3: Figure S3. Circos plot showing type and location of DNA methylation modifications of six strains of Sporosarcina that were sequenced with Pacific Biosciences technology. Color of lines indicate type of modification: adenine (blue), cytosine (red), and unknown (yellow). The lower table is a key for each ring present on the circos plot. (PPTX 23969 kb) Additional file 4: Figure S4. The phylogenetic tree is based on the core genome, and the matrix displays the presence or absence of genes in blue and white respectively, for each strain, in the pan-genome. At 30% amino acid sequence idenity, across 70% of the gene, there are 8610 unique gene clusters that make up the pan-genome of the 29 Sporosarcina strains. (PPTX 15892 kb)

Abbreviations AAI: Average amino acid identity; ACT: Artemis Comparison Tool; ANI: Average nucleotide identity; BLAST: Basic Local Alignment Search Tool; CDD: Conserved Domain Database; CDS: Coding DNA sequence; COG: Cluster of Orthologous Groups; DDH: DNA-DNA hybridization; EMP: Earth Microbiome Project; GEOS: Geometry Engine – Open Source; HGT: Horizontal gene transfer; IS: Insertion sequence; ISSpglI: Insertion sequence of Sporosarcina globispora; iTOL: Interactive Tree of Life; NCBI: National Center for Biotechnology Information; PC: Percent query coverage; PGAAP: Prokaryote Genome Automatic Annotation Pipeline; PI: Percent amino acid sequence identity; QV: Quality Value; RDP: Ribosomal Database Project; RPS-BLAST: Reverse Position Specific Basic Local Alignment Search Tool; SMRT: Single Molecule Real Time Sequencing; WGS: Whole genome sequencing Acknowledgements We thank Bernardine Pregerson for her hard and meticulous work laying the foundation of this research. Thanks to Larry Baresi for insightful conversation and helpful guidance throughout the study. Additional thanks to Aaron Alexander, Tabitha Bayangos, Cristina Alcaraz, and Courtney Sams for assistance with DNA extractions and Illumina library preparations. We also appreciate the advice and insights provided by Gilberto Flores and Sean Murray. Finally, to Melanie Oakes at the University of California Irvine

Page 15 of 17

Genomics High Throughput Facility for assistance with protocols and sequencing. Funding This work was made possible, in part, through access to the Genomics High Throughput Facility Shared Resource of the Cancer Center Support Grant (CA-62203) at the University of California, Irvine and NIH shared instrumentation grants 1S10RR025496-01 and 1S10OD010794-01. The work was funded through laboratory start-up funds provided to Dr. Kerry Cooper. The funding sources of this study did not have any role in the study design, data collection, data analysis, interpretation of the data or writing of the manuscript. Availability of data and materials The datasets generated and/or analyzed during the current study are available in the Genbank repository, under their respective accession numbers: PDZF00000000 (Sporosarcina sp. P7), PDZE00000000 (Sporosarcina sp. P3a), PDZD00000000 (Sporosarcina sp. P35), PDZC00000000 (Sporosarcina sp. P34), PDZB00000000 (Sporosarcina sp. P32b), PDZA00000000 (Sporosarcina sp. P31), PDYZ00000000 (Sporosarcina sp. P30), PDYY00000000 (Sporosarcina sp. P2a), PDYX00000000 (Sporosarcina sp. P29), PDYW00000000 (Sporosarcina sp. P26b), PDYV00000000 (Sporosarcina sp. P25), PDYU00000000 (Sporosarcina sp. P21c), PDYT00000000 (Sporosarcina sp. P20a), PDYS00000000 (Sporosarcina sp. P1a), PDYR00000000 (Sporosarcina sp. P19), PDYQ00000000 (Sporosarcina sp. P18a), PDYP00000000 (Sporosarcina sp. P17b), PDYO00000000 (Sporosarcina sp. P16b), PDYN00000000 (Sporosarcina sp. P16a), PDYM00000000 (Sporosarcina sp. P13), PDYL00000000 (Sporosarcina sp. P12), PDYK00000000 (Sporosarcina sp. P10), CP015108 (S. ureae str. S204), CP015027 (Sporosarcina sp. P33), CP015349 (Sporosarcina sp. P37), CP015109 (Sporosarcina sp. P17a), CP015348 (Sporosarcina sp. P32a), CP015207 (Sporosarcina sp. P8). The Geneparser program is freely available at https://github.com/mmmckay/geneparser. Additional analysis performed with genome sequences not generated during this study were obtain from the NCBI Genome database under the following accession numbers: NZ_AUDQ00000000 (S. ureae DSM 2281). Authors’ contributions AO contributed to experimental design, data generation and analysis, and data interpretation and manuscript writing. KKC supervised the study and contributed to experimental design, data analysis, and interpretation and manuscript writing. MK contributed to data analysis. All authors have read and approved the final version of this manuscript. Ethics approval and consent to participate All strains of cocci-shaped Sporosarcina sequenced for this study were originally isolated from soil samples around the world by Bernadine Pregerson over 40 years ago. All strains are available from Dr. Kerry Cooper upon request. Competing interests The authors declare that they have no competing interests.

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Author details 1 Department of Biology, California State University Northridge, Northridge, CA, USA. 2School of Animal and Comparative Biomedical Sciences, University of Arizona, Tucson, AZ, USA. 3Present Address: Molecular Biology and Biochemistry, University of California Irvine, Irvine, CA, USA. Received: 18 August 2017 Accepted: 28 March 2018

References 1. Kaneuchi C, Benno Y, Mitsuoka T. Clostridium coccoides, a new species from the feces of mice. Int J Syst Bacteriol. 1976;26:482–6. https://doi.org/10.1099/00207713-26-4-482. 2. Rieu-Lesme F, Dauga C, Morvan B, Bouvet OM, Grimont PA, Doré J. Acetogenic coccoid spore-forming bacteria isolated from the rumen. Res Microbiol. 1996;147:753–64. https://doi.org/10.1016/S0923-2508(97)85122-4.

Oliver et al. BMC Genomics (2018) 19:310

3.

4.

5.

6.

7. 8.

9.

10.

11.

12.

13.

14.

15.

16. 17.

18.

19.

20.

21.

22.

23.

Claus D, Fahmy F, Rolf HJ, Tosunoglu N. Sporosarcina halophila sp. nov., an obligate, slightly halophilic bacterium from salt marsh soils. Syst Appl Microbiol. 1983;4:496–506. https://doi.org/10.1016/S0723-2020(83)80007-1. Liu C, Finegold SM, Song Y, Lawson PA. Reclassification of Clostridium coccoides, Ruminococcus hansenii, Ruminococcus hydrogenotrophicus, Ruminococcus luti, Ruminococcus productus and Ruminococcus schinkii as Blautia coccoides gen. nov., comb. nov., Blautia hansenii comb. nov., Blautia hydroge. Int J Syst Evol Microbiol. 2008;58:1896–902. https://doi.org/10.1099/ijs.0.65208-0. Spring S, Ludwig W, Marquez MC, Ventosa A, Schleifer K-H. Halobacillus gen. nov., with descriptions of Halobacillus litoralis sp. nov. and Halobacillus trueperi sp. nov., and transfer of Sporosarcina halophila to Halobacillus halophilus comb. nov. Int J Syst Bacteriol. 1996;46:492–6. https://doi.org/10.1099/00207713-46-2-492. Kim KH, Jia B, Jeon CO. Identification of Trans-4-Hydroxy-L-Proline as a compatible solute and its biosynthesis and molecular characterization in Halobacillus halophilus. Front Microbiol. 2017;8:2054. https://doi.org/10.3389/fmicb.2017.02054. Knoll H, Horschak R. Zur sporulation der garungssarcinen. Monatsber Dtsch Akad Wiss Berl. 1971;13:222–4. Claus D, Wilmanns H. Enrichment and selective isolation of Sarcina maxima lindner. Arch Microbiol. 1974;96:201–4. https://doi.org/10.1007/ BF00590176. Lowe SE, Pankratz HS, Zeikus JG. Influence of pH extremes on sporulation and ultrastructure of Sarcina ventriculi. J Bacteriol. 1989;171:3775–81. https://doi.org/10.1128/jb.171.7.3775-3781.1989. Beijerinck MW. Anhäufungsversuche mit ureumbakterien. Ureumspaltung durch urease und durch katabolismus. Zentralbl Bakteriol Parasitenkd Infekt Hyg II Abt. 1901;7:33–61. Vos P, Garrity G, Jones D, Krieg NR, Ludwig W, Rainey FA, et al. Systematic bacteriology. New York: Springer New York; 2009. https://doi.org/10.1007/978-0-387-68489-5. Pregerson B. The distribution and physiology of Sporosarcina ureae : California State University Northridge; 1973. http://scholarworks.csun.edu/ bitstream/handle/10211.2/4517/PregersonBernardine1973.pdf;sequence=1 Accessed 10 Feb 2018 Risen LP. Multilocus genetic structure in populations of Sporosarcina ureae and the assessment of hexose utilization: California State University Northridge; 1996. http://scholarworks.csun.edu/handle/10211. 3/180855 Miller WG, Pearson BM, Wells JM, Parker CT, Kapitonov VV, Mandrell RE. Diversity within the Campylobacter jejuni type I restriction-modification loci. Microbiology. 2005;151:337–51. https://doi.org/10.1099/mic.0.27327-0. Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, et al. Geneious basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics. 2012;28: 1647–9. https://doi.org/10.1093/bioinformatics/bts199. Andrews S. FastQC: a quality control tool for high throughput sequence data. 2010. http://www.bioinformatics.babraham.ac.uk/projects/. Schmieder R, Edwards R. Quality control and preprocessing of metagenomic datasets. Bioinformatics. 2011;27:863–4. https://doi.org/10. 1093/bioinformatics/btr026. Tritt A, Eisen JA, Facciotti MT, Darling AE. An integrated pipeline for de novo assembly of microbial genomes. PLoS One. 2012;7:e42304. https://doi.org/10.1371/journal.pone.0042304. Marchler-Bauer A, Lu S, Anderson JB, Chitsaz F, Derbyshire MK, DeWeese-Scott C, et al. CDD: a conserved domain database for the functional annotation of proteins. Nucleic Acids Res. 2011;39 Database: D225–9. https://doi.org/10.1093/nar/gkq1189. Leimbach A. bac-genomics-scripts: bovine E. coli mastitis comparative genomics edition (https://zenodo.org/record/215824#.Wlr8B1Q-dTY). Zenodo. 2016. https://doi.org/10.5281/zenodo.215824. Acinas SG, Marcelino LA, Klepac-Ceraj V, Polz MF. Divergence and redundancy of 16S rRNA sequences in genomes with multiple rrn operons. J Bacteriol. 2004;186:2629–35. https://doi.org/10.1128/JB.186.9.2629-2635.2004. Maidak BL, Cole JR, Lilburn TG, Parker Jr CT, Saxman PR, Farris RJ, et al. The RDP-II (ribosomal database project). Nucleic Acids Res. 2001;29:173–4. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC29785/ Pruesse E, Peplies J, Glöckner FO. SINA: accurate high-throughput multiple sequence alignment of ribosomal RNA genes. Bioinformatics. 2012;28:1823–9. https://doi.org/10.1093/bioinformatics/bts252.

Page 16 of 17

24. Price MN, Dehal PS, Arkin AP. FastTree 2 – approximately maximumlikelihood trees for large alignments. PLoS One. 2010;5:e9490. https://doi.org/10.1371/journal.pone.0009490. 25. Letunic I, Bork P. Interactive Tree Of Life (iTOL): an online tool for phylogenetic tree display and annotation. Bioinformatics. 2007;23:127–8. https://doi.org/10.1093/bioinformatics/btl529. 26. Thompson LR, Sanders JG, McDonald D, Amir A, Ladau J, Locey KJ, et al. A communal catalogue reveals Earth’s multiscale microbial diversity. Nature. 2017;551:457–63. https://doi.org/10.1038/nature24621. 27. Alicea BJ, Carvallo-Pinto MA, Rodrigues JLM. Towards a core genome: pairwise similarity searches on interspecific genomic data. arXiv:0807.3353 [q-bio.GN]. 2008. http://arxiv.org/abs/0807.3353. 28. Rasko DA, Myers GS, Ravel J. Visualization of comparative genomic analyses by BLAST score ratio. BMC Bioinformatics. 2005;6:2. https://doi.org/10.1186/ 1471-2105-6-2. 29. Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, et al. BLAST+: architecture and applications. BMC Bioinformatics. 2009;10:421. https://doi.org/10.1186/1471-2105-10-421. 30. Katoh K, Misawa K, Kuma K, Miyata T. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 2002;30:3059–66. https://doi.org/10.1093/nar/gkf436. 31. Wu M, Eisen JA. A simple, fast, and accurate method of phylogenomic inference. Genome Biol. 2008;9:R151. https://doi.org/10.1186/gb-2008-9-10-r151. 32. Pirone-Davies C, Hoffmann M, Roberts RJ, Muruvanda T, Timme RE, Strain E, et al. Genome-wide methylation patterns in Salmonella enterica subsp. enterica serovars. PLoS One. 2015;10:e0123639. https://doi.org/10.1371/journal.pone.0123639. 33. Krzywinski M, Schein J, Birol I, Connors J, Gascoyne R, Horsman D, et al. Circos: an information aesthetic for comparative genomics. Genome Res. 2009;19:1639–45. https://doi.org/10.1101/gr.092759.109. 34. Darling ACE, Mau B, Blattner FR, Perna NT. Mauve: multiple alignment of conserved genomic sequence with rearrangements. Genome Res. 2004;14: 1394–403. https://doi.org/10.1101/gr.2289704. 35. Carver TJ, Rutherford KM, Berriman M, Rajandream M-A, Barrell BG, Parkhill J. ACT: the Artemis comparison tool. Bioinformatics. 2005;21:3422–3. https://doi.org/10.1093/bioinformatics/bti553. 36. Arndt D, Grant JR, Marcu A, Sajed T, Pon A, Liang Y, et al. PHASTER: a better, faster version of the PHAST phage search tool. Nucleic Acids Res. 2016;44: W16–21. https://doi.org/10.1093/nar/gkw387. 37. Grant JR, Arantes AS, Stothard P. Comparing thousands of circular genomes using the CGView comparison tool. BMC Genomics. 2012;13:202. https://doi. org/10.1186/1471-2164-13-202. 38. Konstantinidis KT, Tiedje JM. Towards a genome-based taxonomy for prokaryotes. J Bacteriol. 2005;187:6258–64. https://doi.org/10.1128/JB.187. 18.6258-6264.2005. 39. Kwon S-W, Kim B-Y, Song J, Weon H-Y, Schumann P, Tindall BJ, et al. Sporosarcina koreensis sp. nov. and Sporosarcina soli sp. nov., isolated from soil in Korea. Int J Syst Evol Microbiol. 2007;57(8):1694. https://doi.org/10.1099/ijs.0.64352-0. 40. Tominaga T, An S-Y, Oyaizu H, Yokota A. Oceanobacillus soja sp. nov. isolated from soy sauce production equipment in Japan. J Gen Appl Microbiol. 2009;55:225–32. https://doi.org/10.2323/jgam.55.225. 41. Wolfgang WJ, Coorevits A, Cole JA, de Vos P, Dickinson MC, Hannett GE, et al. Sporosarcina newyorkensis sp. nov. from clinical specimens and raw cow’s milk. Int J Syst Evol Microbiol. 2012;62:322–9. https://doi.org/10.1099/ijs.0.030080-0. 42. Stackebrandt E, Goebel BM. Taxonomic note: a place for DNA-DNA reassociation and 16S rRNA sequence analysis in the present species definition in bacteriology. Int J Syst Evol Microbiol. 1994;44:846–9. https://doi.org/10.1099/00207713-44-4-846. 43. Kim M, Oh H-S, Park S-C, Chun J. Towards a taxonomic coherence between average nucleotide identity and 16S rRNA gene sequence similarity for species demarcation of prokaryotes. Int J Syst Evol Microbiol. 2014;64 Pt 2:346–51. https://doi.org/10.1099/ijs.0.059774-0. 44. Casadesús J, Low DA. Programmed heterogeneity: epigenetic mechanisms in bacteria. J Biol Chem. 2013;288:13929–35. https://doi.org/10.1074/jbc.R113.472274. 45. Merrick MJ, Edwards RA. Nitrogen control in bacteria. Microbiol Rev. 1995; 59:604–22. http://www.ncbi.nlm.nih.gov/pubmed/8531888 46. Geisseler D, Horwath WR, Joergensen RG, Ludwig B. Pathways of nitrogen utilization by soil microorganisms – a review. Soil Biol Biochem. 2010;42: 2058–67. https://doi.org/10.1016/j.soilbio.2010.08.021.

Oliver et al. BMC Genomics (2018) 19:310

Page 17 of 17

47. Perna NT, Plunkett G, Burland V, Mau B, Glasner JD, Rose DJ, et al. Genome sequence of enterohaemorrhagic Escherichia coli O157:H7. Nature. 2001;409: 529–33. https://doi.org/10.1038/35054089. 48. Rump LV, Fischer M, Gonzalez-Escalona N. Prevalence, distribution and evolutionary significance of the IS629 insertion element in the stepwise emergence of Escherichia coli O157:H7. BMC Microbiol. 2011;11:133. https://doi.org/10.1186/1471-2180-11-133. 49. Uchiyama I. Multiple genome alignment for identifying the core structure among moderately related microbial genomes. BMC Genomics. 2008;9:515. https://doi.org/10.1186/1471-2164-9-515. 50. Wayne LG, Brenner DJ, Colwell RR, Grimont PAD, Kandler O, Krichevsky MI, et al. Report of the ad hoc committee on reconciliation of approaches to bacterial systematics. Int J Syst Evol Microbiol. 1987; 37(4):463. https://doi.org/10.1099/00207713-37-4-463. 51. Rasko DA, Rosovitz MJ, Myers GSA, Mongodin EF, Fricke WF, Gajer P, et al. The Pangenome structure of Escherichia coli: comparative genomic analysis of E. coli commensal and pathogenic isolates. J Bacteriol. 2008;190:6881–93. https://doi.org/10.1128/JB.00619-08. 52. Leekitcharoenphon P, Lukjancenko O, Friis C, Aarestrup FM, Ussery DW. Genomic variation in Salmonella enterica core genes for epidemiological typing. BMC Genomics. 2012;13:88. https://doi.org/10.1186/1471-2164-13-88. 53. Richter M, Rossello-Mora R. Shifting the genomic gold standard for the prokaryotic species definition. Proc Natl Acad Sci. 2009;106:19126–31. https://doi.org/10.1073/pnas.0906412106. 54. Qin Q-L, Xie B-B, Zhang X-Y, Chen X-L, Zhou B-C, Zhou J, et al. A proposed genus boundary for the prokaryotes based on genomic insights. J Bacteriol. 2014;196:2210–5. https://doi.org/10.1128/JB.01688-14. 55. Meier-Kolthoff JP, Auch AF, Klenk H-P, Göker M. Genome sequence-based species delimitation with confidence intervals and improved distance functions. BMC Bioinformatics. 2013;14:60. https://doi.org/10.1186/1471-2105-14-60. 56. Varghese NJ, Mukherjee S, Ivanova N, Konstantinidis KT, Mavrommatis K, Kyrpides NC, et al. Microbial species delineation using whole genome sequences. Nucleic Acids Res. 2015;43:6761–71. https://doi.org/10.1093/nar/gkv657. 57. Hug LA, Baker BJ, Anantharaman K, Brown CT, Probst AJ, Castelle CJ, et al. A new view of the tree of life. Nat Microbiol. 2016;1:16048. https:// doi.org/10.1038/nmicrobiol.2016.48.

Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit