Complete genome sequence of the Sulfodiicoccus acidiphilus strain HS-1^T, the first crenarchaeon that lacks polB3, isolated from an acidic hot spring in Ohwaku-dani, Hakone, Japan

Abstract

Objective

Sulfodiicoccus acidiphilus HS-1^T is the type species of the genus Sulfodiicoccus, a thermoacidophilic archaeon belonging to the order Sulfolobales (class Thermoprotei; phylum Crenarchaeota). While S. acidiphilus HS-1^T shares many common physiological and phenotypic features with other Sulfolobales species, the similarities in their 16S rRNA gene sequences are less than 89%. In order to know the genomic features of S. acidiphilus HS-1^T in the order Sulfolobales, we determined and characterized the genome of this strain.

Results

The circular genome of S. acidiphilus HS-1^T is comprised of 2353,189 bp with a G+C content of 51.15 mol%. A total of 2459 genes were predicted, including 2411 protein coding and 48 RNA genes. The notable genomic features of S. acidiphilus HS-1^T in Sulfolobales species are the absence of genes for polB3 and the autotrophic carbon fixation pathway, and the distribution pattern of essential genes and sequences related to genomic replication initiation. These insights contribute to an understanding of archaeal genomic diversity and evolution.

Introduction

Sulfodiicoccus acidiphilus HS-1^T, represented a novel genus, was recently isolated in our laboratory and validly described [1]. The genus belongs to the order Sulfolobales, a well-known taxon of the phylum Crenarchaeota, widely inhabits hot acidic environments all over the world [2,3,4,5]. The 16S rRNA gene sequence similarities between S. acidiphilus and other species in the order Sulfolobales were less than 89%. Given the low 16S rRNA gene similarities with other Sulfolobales species, we hypothesized that the strain also harbored distinct genomic features in Sulfolobales species. Therefore, we determined the complete genome of S. acidiphilus HS-1^T and compared it with other genomes in Sulfolobales species. Genomic analysis revealed that S. acidiphilus HS-1^T has several distinguishing genomic features.

Main text

Methods

Organism information

The isolation and characterization of S. acidophilus HS-1^T representing a novel genus Sulfodiicoccus was reported previously [1, 6]. The phylogenetic position of S. acidiphilus based on 16S rRNA gene sequences is shown in Additional file 1: Figure S1. The general features of S. acidiphilus HS-1^T are shown in Additional file 1: Table S1. HS-1^T is an irregular cocci, non-motile, thermoacidophilic archaeon. Optimal growth occurs at 65–70 °C and pH 3–3.5. The strain is obligately aerobic and can utilize the following organics as a sole carbon source: yeast extract, beef extract, casamino acids, peptone, tryptone, xylose, galactose, glucose, maltose, sucrose, raffinose, lactose, aspartic acid and glutamic acid. Chemolithotrophic growth does not occur when S⁰, FeS₂, K₂S₄O₆, Na₂S₂O₃ or FeSO₄ acts as an electron donor under aerobic conditions (O₂ as an electron acceptor). The cells are regular to irregular cocci with a diameter of 0.8–1.5 μm (Additional file 1: Figure S2).

Genomic DNA preparation, genome sequencing and assembly

HS-1^T was cultivated in a 5 L glass bottle using ~ 4 L modified Brock’s basal salt (MBS) medium [7], supplemented with yeast extract (1 g/L) and glucose (1 g/L) under aerobic conditions (65 °C, pH 3). Approximately 1 L of the culture in early exponential phase was centrifuged (OD₆₀₀ = ~ 0.1, 8000×g, 4 °C, 10 min), and the supernatant was removed. DNA was extracted from the resulting cell pellets using the Genomic DNA Buffer Set (QIAGEN) and the Genomic-Tip 500/G (QIAGEN), according to the manufacturer’s protocols. The quantity and purity of the extracted DNA was checked spectrophotometrically and through agarose gel electrophoresis. The genome sequencing of the S. acidiphilus strain HS-1^T was performed by Macrogen Inc. (South Korea) using a PacBio RS II sequencer (Pacific Biosciences, Menlo Park, CA, US). De novo assembly was conducted using the Hierarchical Genome Assembly Process v.3.0 (https://github.com/PacificBiosciences/Bioinformatics-Training/wiki/HGAP).

Genome annotation

Annotation of the protein coding genes and the COG (cluster of orthologous groups) assignments were performed using the on-line annotation server DFAST [8]. The tRNA and rRNA genes were identified using tRNAscan-SE [9, 10] and DFAST, respectively. Pseudogenes were identified using LAST [11] implemented in DFAST. Protein coding genes with Pfam domains [12] were searched using a CD-search program [13] with an e-value threshold of less than 1e−2 (database: Pfam v.30.0). Signal peptides were predicted using PRED-SIGNAL [14]. Transmembrane helices were predicted using TMHMM [15]. CRISPR repeats were detected using the CRISPR recognition tool CRT [16]. Genes in internal clusters were predicted using the CD-HIT Suite (sequence identity cut-off: 0.3, minimal alignment coverage for longer and shorter sequences: 0.7, other parameters: default) [17].

Search of replication origin

The replication origin (oriC) and origin recognition box (ORB) in the chromosome were predicted by Ori-finder 2 [18]. Another replication origin in the chromosome was manually searched. Repeat sequences in the predicted oriC region were searched by the REPuter program [19].

Reexamination of chemolithotrophic growth on hydrogen

The autotrophic growth of HS-1^T on hydrogen was reexamined. MBS medium (10 mL, pH 3) was added to a glass test tube and the headspace was filled with H₂/CO₂/air (80:20:10, 120 kPa). A 50 μL of active culture was inoculated into the test tube and incubation occurred at 65 °C for 2 weeks.

Genome project history

The complete genome sequence of HS-1^T (= JCM 31740^T = InaCC Ar79^T) was deposited in GenBank under accession number AP018553. The raw data (PacBio reads) used for the assembly was deposited in the DNA Data Bank of Japan under accession number DRA008516. The Bioproject accession number is PRJDB6753. A summary of the genome project is provided in Additional file 1: Table S2.

Results and discussion

Genome properties

A complete circular genome sequence (2,353,189 bp) was successfully obtained from a total of 225,345 subreads (a total of 1,538,043,255 bp), and a plasmid was absent. The G+C content is 51.15 mol%, which is identical to the reported value of 52.0 mol% estimated by HPLC method [1]. The genome was predicted to contain a total of 2459 genes, of which 2411 code for proteins and 48 code for RNAs (rRNA: 3, tRNA: 45). The genome harbors each one copy of 5S, 16S and 23S rRNA genes. Genes of 16S and 23S rRNA are encoded in a gene cluster with a 191 bp spacer region, while 5S rRNA gene is found in different location. Among the 2411 protein coding genes, 1219 were assigned putative functions. A total of 837 and 244 genes could be assigned COG functional categories and pseudogenes, respectively (Additional file 1: Tables S3, S4, Figure S3). Three CRISPR repeat regions were detected (positions: 1664810–1679609, 1679646–1680862, and 1688612–1701897). No plasmids were detected. Other genomic statistics such as predicted Pfam domains, signal peptides, and transmembrane helices are summarized in Additional file 1: Table S3.

Insights from the genome sequence

Replication initiation genes and oriC

All previously identified species of the order Sulfolobales (whose complete genomic sequences are available) have three copies of the cdc6 gene (a cell division control gene, Additional file 1: Table S5). In contrast, only one copy of the cdc6 gene (HS1genome0091) was found in the HS-1^T genome (Fig. 1a). One replication origin (named oriC-1, position: 87,046–87,335) located directly upstream of the cdc6 gene was predicted by the Ori-finder 2. The oriC-1 had a relatively high content of adenine and thymine residues (A/T rich) (56.55%) and contained three ORBs as typical archaeal genomes with the following sequence: ACCCCTCTGTTTCCACTGGA [18, 20,21,22] (Fig. 1b).A previously reported uncharacterized motif (UCM) that exists in the oriC regions of several Crenarchaea [23] has also been located in HS-1^T oriC-1. These facts indicate that oriC-1 is a DNA replication origin. Another replication initiation-related gene, whiP (HS1genome1070) [23], was found on the opposite side of the cdc6 (Fig. 1a), however, neither oriC nor ORB sequences were found around whiP by Ori-finder 2. We then manually searched for putative oriC regions, and an A/T rich intergenic region (64.91%, position: 1,016,172–1,016,607, named as oriC-2) was found directly upstream of whiP. Further attempts to search for repeat sequences in oriC-2 using the REPuter program [19] confirmed the presence of 12 direct repeats (8–11 bp, Fig. 1b). The role of the direct repeats is unclear, although they may be involved in the recognition of the replication origin for the DNA initiation protein WhiP. Perhaps the direct repeats alternatively function as an ORB. Further experiments such as DNaseI footprint analyses [23, 24] are required to know the function.

Fig. 1

Predicted oriCs in the Sulfodiicoccus acidiphilus HS-1^T genome. a Positions of the predicted oriCs, b ORB, UCM and repeat sequences found upstream of cdc6 and whiP

Pathways involved in autotrophic growth

Two CO₂ fixation pathways have been reported in the thermophilic autotroph Crenarchaeota, namely the 3-hydroxypropionate/4-hydroxybutyrate (HP/HB) cycle and the dicarboxylate/4-hydroxybutyrate (DC/HB) cycle [25]. Most species in the order Sulfolobales are thought to possess the HP/HB cycle but not the DC/HB cycle [25,26,27]. Both cycles are absent in HS-1^T (Fig. 2). The other known autotrophic pathways (Calvin–Bassham–Benson cycle, reductive citric acid cycle (Amon–Buchanan cycle), reductive acetyl-CoA pathway (Wood–Ljungdahl pathway) and 3-hydroxypropionate cycle) [28] have not been identified in HS-1^T. These observations are consistent with the incapability of chemolithoautotrophic growth of HS-1^T [1]. In the previous paper, we mentioned the capability of chemolithoautotrophic growth of HS-1^T using hydrogen as an electron donor. Based on the genomic information, we carefully reexamined the capability of chemolithoautotrophic growth with serial inoculation using the chemolithoautotrophic medium, and no growth occurred after 2nd inoculation. Thus, we revise our previous description regarding the autotrophic growth of HS-1^T on hydrogen, to “HS-1^T does not grow on hydrogen autotrophically.”

Fig. 2

Incomplete DC/HB and HP/HB cycles of Sulfodiicoccus acidiphilus HS-1^T. The annotated ORF number is shown in parentheses. Gray arrows indicate pathways that are missing in HS-1^T. HP/HB: 3-hydroxypropionate/4-hydroxybutyrate cycle; DC/HB: dicarboxylate/4-hydroxybutyrate cycle

Lack of polB3 in the genome of HS-1^T

Three groups of family B DNA polymerases (PolB1, PolB2, and PolB3) have been associated with archaea: PolB1 is only distributed in the superphylum that includes Thaumarchaeota, Aigarchaeota, Crenarchaeota and Korarchaeota (also known as TACK superphylum); PolB2 is patchily distributed in most of the archaeal lineages; and PolB3 is distributed in all archaea except for Thaumarchaeota [29]. Surprisingly, the HS-1^T genome lacks polB3, although all the crenarchaeal genomes reported before harbor the gene. In the order Sulfolobales, polB3 is located on the downstream region of dnaG, with a high genomic synteny around the dnaG sequence among Sulfolobales species, while the genome structure of the downstream region of dnaG of HS-1^T is different from those in the genomes of other Sulfolobales species (Fig. 3). Since all the other species in the order Sulfolobales have a polB3 downstream of a dnaG, HS-1^T may have lost polB3 and the downstream region of dnaG during the course of its evolution. The roles of PolB1, PolB2 and PolB3 in Saccharolobus solfataricus (synonym: Sulfolobus solfataricus), a model organism of the order Sulfolobales, were previously investigated in vivo by Choi et al. [30]. The authors showed that PolB1 was catalytically much more efficient and processive than PolB2 and PolB3, suggesting that PolB1 plays a catalytic role as the main replicative DNA polymerase. They also suggested that PolB2 and PolB3 have limited catalytic roles in translesion DNA synthesis and may not be involved in chromosomal DNA replication [30]. The lack of a polB3 gene in S. acidiphilus HS-1^T indicates that PolB3 is not essential for either DNA replication or translesion DNA synthesis in Sulfolobales or the phylum Crenarchaeota.

Fig. 3

Organization of genes around dnaG and polB3 in the genomes of the Sulfolobales species. The downstream dnaG region, containing polB3 and several other genes, is not present in the HS-1^T genome. Each color represents similar genes except for white, which indicates genes detected only in this region of HS-1^T

Limitations

This study focused on noteworthy genomic features of S. acidiphilus HS-1^T that are distinct from other Sulfolobales species. Although our genomic analyses revealed some exceptions of genomic features in Sulfolobales, further molecular biology assessment and biochemical analyses are needed to resolve the issues raised in this manuscript.