The complete genome sequences of sulfur-oxidizing Gammaproteobacteria Sulfurifustis variabilis skN76^T and Sulfuricaulis limicola HA5^T

Abstract

Sulfurifustis variabilis and Sulfuricaulis limicola are autotrophic sulfur-oxidizing bacteria belonging to the family Acidiferrobacteraceae in the order Acidiferrobacterales. The type strains of these species, strain skN76^T and strain HA5^T, were isolated from lakes in Japan. Here we describe the complete genome sequences of Sulfurifustis variabilis skN76^T and Sulfuricaulis limicola HA5^T. The genome of Sulfurifustis variabilis skN76^T consists of one circular chromosome with size of 4.0 Mbp including 3864 protein-coding sequences. The genome of Sulfuricaulis limicola HA5^T is 2.9 Mbp chromosome with 2763 protein-coding sequences. In both genomes, 46 transfer RNA-coding genes and one ribosomal RNA operon were identified. In the genomes, redundancies of the genes involved in sulfur oxidation and inorganic carbon fixation pathways were observed. This is the first report to show the complete genome sequences of bacteria belonging to the order Acidiferrobacterales in the class Gammaproteobacteria.

Introduction

Sulfurifustis variabilis skN76^T and Sulfuricaulis limicola HA5^T are gammaproteobacterial sulfur-oxidizing bacteria isolated from sediments of Lake Mizugaki and Lake Harutori, respectively [1, 2]. They both belong to the family Acidiferrobacteraceae in the order Acidiferrobacterales . In this order, only three species have been isolated in pure culture. They are all chemolithoautotrophs and can grow by oxidation of inorganic sulfur compounds. Sulfurifustis variabilis and Sulfuricaulis limicola are neutrophilic, whereas the other species, Acidiferrobacter thiooxydans , is acidophilic [3]. Taxonomy of Acidiferrobacter thiooxydans has been revised several times, and the family Acidiferrobacteraceae and order Acidiferrobacterales were recently established to accommodate the species [1, 3–5]. The members of the family Acidiferrobacteraceae have been frequently detected in various environments as gene sequences [2, 3, 6].

Here we show the complete genome sequences of Sulfurifustis variabilis skN76^T and Sulfuricaulis limicola HA5^T as the first genomes of the order Acidiferrobacterales .

Organism information

Classification and features

The cells of Sulfurifustis variabilis skN76^T are rod-shaped or filamentous form with varying length, and 0.3–0.5 μm in width (Fig. 1a, Table 1). The cells of Sulfuricaulis limicola HA5^T are rod-shaped, 1.2–6.0 μm in length and 0.3–0.5 μm in width (Fig. 1b, Table 1). They are both Gram-stain-negative. Sulfurifustis variabilis and Sulfuricaulis limicola belong to the family Acidiferrobacteraceae within the class Gammaproteobacteria (Fig. 2). They both utilized thiosulfate, tetrathionate and elemental sulfur as electron donors for chemolithoautotrophic growth under aerobic conditions [1, 2].

Fig. 1

Phase-contrast micrographs of Sulfurifustis variabilis skN76^T (a) and Sulfuricaulis limicola HA5^T (b), grown with thiosulfate at 45 and 28 °C, respectively. Bars, 5 μm

Table 1 Classification and general features of Sulfurifustis variabilis skN76^T and Sulfuricaulis limicola HA5^T according to MIGS recommendations

Fig. 2

Phylogenetic tree showing the relationships of Sulfurifustis variabilis skN76^T and Sulfuricaulis limicola HA5^T with other members of the class Gammaproteobacteria based on 16S rRNA gene sequences aligned by using CLUSTAL W. Desulfatitalea tepidiphila S28bF^T was used as an outgroup. This tree was reconstructed using 1412 sites with the neighbor-joining method by using MEGA6 [27]. Percentage values of 1000 bootstrap resamplings are shown at nodes; values below 50 % were not shown

Genome sequencing information

Genome project history

Sulfurifustis variabillis skN76^T and Sulfuricaulis limicola HA5^T were selected for sequencing as representatives of sulfur-oxidizing bacteria belonging to the order Acidiferrobacterales , to reveal characteristics of their genomes. A summary of the project information is shown in Table 2.

Table 2 Project information

Growth conditions and genomic DNA preparation

Sulfurifustis variabilis skN76^T and Sulfuricaulis limicola HA5^T were grown with 20 mM thiosulfate as an energy source in a bicarbonate-buffered medium previously described [1], at 45 and 28 °C, respectively. Genomic DNA samples were prepared by using Wizard® genomic DNA purification kit (Promega, Madison, WI, USA) from approximately 0.2 ml (skN76) or 0.1 ml (HA5) of cell pellets. Amounts of the obtained DNA assessed by spectrophotometry were ca. 270 μg (skN76) and 90 μg (HA5) respectively, and the UV absorption ratio of 260/280 nm was greater than 1.8 in both samples.

Genome sequencing and assembly

The genomic DNA was sheared into approximately 20 kb using g-TUBE (Covaris, Inc., Woburn, MA, USA). The SMRTbell^TM templates were prepared from the fragments using SMRTbell^TM Template Prep Kit 1.0 (Pacific Biosciences, Menlo Park, CA, USA). The size-selected libraries for sequencing were prepared by using BluePippin (Sage Science, Baverly, MA, USA). The libraries were sequenced on a PacBio RS II instrument (Pacific Biosciences) with P6-C4 chemistry (for Sulfurifustis variabillis skN76^T) or P5-C3 chemistry (for Sulfuricaulis limicola HA5^T). De novo assembly was performed by using RS_HGAP Assembly.3 (for Sulfurifustis variabillis skN76^T) or RS_HGAP Assembly.2 (for Sulfuricaulis limicola HA5^T), implemented within the SMRT Analysis v2.3 (Pacific Biosciences) software environment. By assembling 79,017 subreads (837,333,548 bp) of Sulfurifustis variabillis skN76^T, two contigs with the lengths of ca. 4.0 Mbp and ca. 5.4 kbp were obtained. The shorter one was identical to a partial sequence of the larger one, and a circular chromosome was manually constructed from the larger contig by finding self-overlapping regions using the in silico Molecular Cloning (R) Genomic Edition (In Silico Biology, Inc., Yokohama, Japan) application. As for Sulfuricaulis limicola HA5^T, a single contig (ca. 2.9 Mbp) was obtained by assembling 61,565 subreads (409,124,339 bp), and circular chromosome was manually constructed in the same manner.

Genome annotation

The genomes were annotated automatically using the Microbial Genome Annotation Pipeline [7]. Further manual annotation of the predicted protein-coding sequences was performed on the basis of BLASTP searches against the NCBI nonredundant database. CDSs were annotated as hypothetical protein-coding genes when they met any of the following four criteria in the top hit of the BLASTP analysis: (1) E-value >1e-8, (2) length coverage <60 % against query sequence (3) sequence identity <30 % or (4) function of the hit was unidentified. The WebMGA server was used to assign the genes to Clusters of Ortholog Groups and Protein family domains [8–11]. The Phobius server was used to predict signal peptides and transmembrane helices [12]. Clustered Regularly Interspaced Short Palindromic Repeat loci were detected using CRISPRfinder [13].

Genome properties

The basic statistics of the genomes are shown in Table 3. Both genomes contained 46 tRNA genes and one rRNA operon. The genome size of Sulfurifustis variabillis skN76^T was approximately 1.4 times larger than that of Sulfuricaulis limicola HA5^T. CRISPR loci were found only in the genome of Sulfurifustis variabillis skN76^T (Table 3). The distribution of genes into COGs functional categories is presented in Table 4.

Table 3 Genome statistics of Sulfurifustis variabilis skN76^T and Sulfuricaulis limicola HA5^T

Table 4 Number of genes associated with general COG functional categories

Insights from the genome sequences

In both the genomes of Sulfurifustis variabillis skN76^T and Sulfuricaulis limicola HA5^T, genes involved in the sulfur oxidation pathway were identified. The genomes of both strains contain genes of the DSR system related to the oxidation of elemental sulfur to sulfite [14, 15]. They contain a dsr gene cluster of identical composition, dsrABEFHCMKLJOPNR (SVA_1954-1967, SCL_1274-1261). There are some dsr genes outside of the gene cluster, dsrAB (SVA_0258-0259, SCL_0256-0257), dsrS (SVA_2921, SCL_0781) and dsrC (SVA_0281, SVA_0284, SVA_0358, SVA_0917, SVA_0969, SVA_1205, SVA_1793, SVA_1949, SVA_2832, SVA_3655; SCL_0275, SCL_0524, SCL_0785, SCL_1279, SCL_1423, SCL_2646).

As genes encoding proteins involved in oxidation of sulfite to sulfate in the cytoplasm, both genomes contain two copies of the aprAB genes encoding an adenosine-5’-phosphosulphate reductase (SVA_2607-2608, SVA_3565-3564; SCL_0600-0601, SCL_2474-2473), along with the sat gene encoding a sulfate adenylyltransferase (SVA_3563, SCL_2472) and the aprM gene (SVA_2609, SCL_0602). In addition, the genome of Sulfuricaulis limicola HA5^T contains the hdrAACB genes encoding a Hdr (SCL_2523-2520), but that of Sulfurifustis valiabilis skN76^T does not. The AprM and Hdr complex are thought to have similar function that interacts with the adenosine-5’-phosphosulphate reductase [16–18]. The genomes also contain the soeABC genes (SVA_2734, SVA_2736-2737; SCL_0523-0521), encoding a membrane-bound polysulfide reductase-like iron-sulfur molybdoprotein, which is suspected to be involved in sulfite oxidation in the cytoplasm [19]. Further, the genome of Sulfurifustis valiabilis skN76^T contains the sorAB genes (SVA_1391-1390) related to the direct oxidation of sulfite to sulfate in the periplasm [20].

For thiosulfate oxidation, both genomes contain the soxXYZAB gene cluster (SVA_2999-3003, SCL_2229-2233). Although sulfide oxidation by these bacteria has not been demonstrated, genes related to sulfide oxidation were identified; the fccAB (soxEF) genes encoding a flavocytochrome c/sulfide dehydrogenase (SVA_0067-0066, SVA_3594-3595; SCL_0078-0077) and the sqr gene encoding a sulfide:quinone oxidoreductase (SVA_1781, SVA_2675, SVA3205).

Sulfurifustis variabillis skN76^T and Sulfuricaulis limicola HA5^T are autotrophic bacteria. They both have two copies of the rbcL and rbcS genes, encoding large and small subunits of ribulose bisphosphate carboxylase/oxygenase (SVA_3460-3459, SVA_3471-3470; SCL_2417-2416, SCL_2425-2424), which is the key enzyme in the Calvin-Benson-Bassham cycle to catalyze inorganic carbon fixation. The two copies of RuBisCO in each genome are phylogenetically distinct, and belong to lineages referred to as green-like form IA and red-like form IC (Fig. 3) [21]. In the form IC RuBisCO coded by rbcL gene (SVA_3460, SCL_2417), Sulfurifustis variabillis skN76^T and Sulfuricaulis limicola HA5^T have six-amino-acid inserts at the same position where a similar insert was reported from Nitrosospira sp. 40KI [22]. There are two other RuBisCO sequences which have six-amino-acid inserts at the same position, and these sequences with inserts formed a monophyletic cluster in the tree of RuBisCO (Fig. 3). In general, RuBisCO of form IA and IC have different properties which are thought to be advantageous to fix inorganic carbon under different concentrations of carbon dioxide and/or oxygen [21]. Possession of the genes for these two distinct RuBisCO forms may be beneficial to cope with changing environmental conditions, or to thrive in various types of ecosystems.

Fig. 3

Neighbor-joining tree showing the phylogenetic positions of RuBisCO amino acid sequences coded in the genomes of Sulfurifustis variabilis skN76^T and Sulfuricaulis limicola HA5^T. The sequences aligned by using CLUSTAL W. This tree was reconstructed using 421 sites with MEGA6 [27]. Percentage values of 1000 bootstrap resamplings are shown at nodes; values below 50 % were not shown. The sequences shown in box have six-amino-acid inserts at the same position

Conclusion

This is the first report on complete genome sequences of bacteria belonging to the order Acidiferrobacterales . The genome analysis of Sulfurifustis variabillis skN76^T and Sulfuricaulis limicola HA5^T revealed that they have similar sets of genes involved in sulfur oxidation pathways. In the both genomes, redundancies of the genes for sulfur oxidation and inorganic carbon fixation were observed, as represented by multiple copies of dsrAB, aprAB and rbcLS. Such redundancies may provide physiological flexibility to the chemolithotrophic sulfur oxidizers which are fully depending on these functions to obtain energy and carbon source for growth.