Introduction

Heavy metals, either essential (e.g. Mn, Zn, Cu, Co, Ni and Mo) or toxic (e.g. Hg, Ag and Cd), are generally harmful to microbial cells even at low concentrations, as to other living organisms [1, 2]. However, some microorganisms are able to resist to certain kinds and concentrations of heavy metals through several mechanisms, such as incorporating or precipitating heavy metals into complexes, oxidizing or reducing metals to less toxic valence states, and direct transporting metals out of the cell [3, 4]. These heavy metal resistant microorganisms have been attracting great interests because of their potential biotechnological applications in bio-mining of expensive heavy metals and bioremediation of heavy metal-contaminated environment [2].

Halomonas, the largest genus of the family Halomonadaceae, can be found in most saline environments, including marine environments, salterns, saline lakes and soils, as well as salty foods, etc. [5, 6]. Halomonas zincidurans strain B6T, a moderately halophilic bacterium, was isolated from a deep-sea sediment from the South Atlantic Mid-Ocean Ridge [5]. The strain was able to grow in medium containing high concentrations of heavy metals, especially Zn2+ ion, which is not detected in the reference strains and other moderately halophiles [5, 7]. Therefore, the novel isolate was named as H. zincidurans due to its particular resistance to zinc ion [5]. Here, we present a summary classification and a set of features of H. zincidurans strain B6T, together with the description of the genomic sequencing and annotation.

Organism information

A deep-sea sediment sample, TVG10, was collected from the South Atlantic Mid-Ocean Ridge (Table 1). There were many small hard orange red-colored lumps mixed in the sediment sample, which might be the particles containing ferric oxide and diffusing with hydrothermal plumes [8]. Not surprisingly, the concentrations of heavy metals in sample TVG10 were much higher than those in the samples collected from deep-sea seamount sediment [9], offshore sediment [10] and continental crust [11] (Additional file 1: Table S1), including Fe (98.99 mg/g), Mn (42.48 mg/g), Cu (0.839 mg/g), Ni (0.338 mg/g), Zn (0.285 mg/g), Cr (0.195 mg/g) and Co (0.064 mg/g). With consideration of the heavy metal rich environment, marine broth 2216 medium (MB, BD) containing 20 mM Mn2+ was used to isolate heavy metal resistant strains. Subsequently a strain named B6T was obtained [5].

Table 1 Classification and general features of H. zincidurans B6T according to the MIGS recommendations  [12]
Full size table

H. zincidurans strain B6T is a Gram-stained negative, rod-shaped (Figure 1), moderately halophilic bacterium growing at 0.5-15% (w/v) NaCl (Table 1). Strain B6T exhibited the highest 16S rRNA gene sequence similarity with H. xinjiangensis (96.1%). Phylogenetic analysis based on 16S rRNA gene sequences showed that strain B6T and H. xinjiangensis clustered together in a distinct branch within the genus Halomonas with a high bootstrap value (Figure 2). Strain B6T was able to resist high concentrations of heavy metals in liquid HM medium, including Mn2+ (200 mM), Co2+ (1.0 mM), Cu2+ (2.5 mM) and Zn2+ (14 mM). Its resistance to Zn2+ could be much higher (30 mM) when incubated on marine agar 2216 medium (MA, BD) [5], comparing to only 1 mM Zn2+ resisted by H. xinjiangensis TRM0175T. And the maximum zinc resistance concentration for 250 moderately halophilic bacteria, reported by Nieto et al., was only 2.5 mM [7]. Therefore, H. zincidurans strain B6T is of significant interest due to its prominent resistance to zinc.

Figure 1
figure 1

Micrograph of H. zincidurans strain B6 T obtained by scanning electron microscopy (S260; Cambridge).

Full size image
Figure 2
figure 2

Phylogenetic tree highlighting the position of H. zincidurans strain B6T relative to phylogenetically closely related type strains within the family Halomonadaceae. The sequences were aligned using Clustal W [26], and the neighbor-joining tree [27] was constructed based on kimura 2-parameter distance model [28] by using MEGA5 [29]. Bootstrap values above 60% are shown obtained from 1,000 bootstrap replications. Bar, 0.05 substitutions per nucleotide position. The corresponding GenBank accession numbers are displayed in parentheses.

Full size image

Genome sequencing information

Genome project history

The next-generation shotgun-sequencing and quality assurance was performed at the Beijing Genome Institute (BGI, Shenzhen). The gap closure and annotation processes were performed by the authors. The Whole Genome Shotgun project of H. zincidurans strain B6T has been deposited at DDBJ/EMBL/GenBank under the accession JNCK00000000. The version described in this paper is version JNCK01000000. Table 2 presents the project information and its association with MIGS version 2.0 compliance [12].

Table 2 Project information
Full size table

Growth conditions and DNA isolation

H. zincidurans strain B6T was aerobically cultivated in MB medium at 30°C. Total genomic DNA was extracted using the method described by Marmur [32]. The quality and quantity of the genomic DNA was determined by 0.6% agarose gel electrophoresis with λ-Hind III digest DNA marker (TaKaRa, Dalian, China) and by a Qubit® fluorometer (Invitrogen, CA, USA) with Qubit dsDNA BR Assay kit (Invitrogen, CA, USA). About 350 μg DNA with a concentration of 450 ng/μl was obtained.

Genome sequencing and assembly

Whole-genome shotgun DNA sequencing of H. zincidurans strain B6T was performed using Solexa paired-end sequencing technology (HiSeq2000 system, Illumina, USA) [33]. Two libraries with insert size 494 bp and 2,586 bp were constructed and a total of 519 Mb and 416 Mb raw data were produced before filtering. After removing the adapter, duplicated reads and short inserts from the data of large library, there remained 433 Mb (~120-folds genome coverage) and 328 Mb (~90-folds genome coverage) clean data from the small and large libraries for assembling, respectively. Then these sequences were assembled into 15 contigs using the SOAPdenovo v.1.05 [30], the contig N50 length of which was 1,864,365 bp. PCR primers for gap closure were designed by Primer Premier v.5. PCR reactions were performed with PrimeSTAR HS Polymerase (TaKaRa, Dalian, China) and the amplicons were sequenced using Sanger and primer walking technologies. The sequenced fragments were subsequently assembled with the contigs using SeqMan of the Lasergene package (DNAstar, Madison, WI) into 2 contigs.

Genome annotation

The whole genomic tRNAs were identified using tRNAscan-SE v.1.21 [34] with bacterial model, and rRNAs were found by RNAmmer v.1.2 Server [35]. ORFs were predicted using Glimmer v.3.0 [31]. The predicted ORFs were translated and analyzed using the NCBI nonredundant, Swiss-Prot [36] and COG [37] databases, as well as RAST server online [38] for genome annotation. KAAS [39] was used to assign the predict proteins into KEGG pathway [40] with BBH method. Genes with signal peptides and transmembrane helices were predicted using TMHMM server v.2.0 [41] and SignalP server v.4.1 [42], respectively. The G+C content, G+C content at the third-codon position and RSCU were calculated by CodonW v.1.4.4.

Genome properties

The genome was assembled into 2 contigs, one with a size of 3,546,937 bp and the other with 7,823 bp (Table 3). The G+C content determined based on the total 3,554,760 bp sequences was 66.41%. A total of 3,392 genes were predicted, including 3,325 protein-coding genes, 61 tRNA genes and two copies of 16S-23S-5S rRNA gene operons (Table 4 and Figure 2). Among the protein coding genes, 2,848 were assigned to putative functions, and the remaining was annotated as hypothetical proteins. In total, 1,938 and 442 protein coding genes were assigned to KEGG and subsystems, respectively. The detailed properties and the statistics of the genome as well as the distribution of genes into COG functional categories are summarized in Tables 3, 4 and 5, Figure 3 and Additional file 2: Table S2.

Table 3 Summary of genome: two contigs
Full size table
Table 4 Nucleotide content and gene count levels of the genome
Full size table
Table 5 Number of genes associated with the 25 general COG functional categories
Full size table
Figure 3
figure 3

Circular map of the chromosome of H. zincidurans strain B6T . Labeling from the outside to the inside circle: ORFs on the forward strand (colored by COG categories), ORFs on the reverse strand (colored by COG categories), RNA genes (tRNAs red, rRNAs blue), G+C content (peaks out/inside the circle indicate values higher or lower than the average G+C content, respectively), GC skew (calculated as (G-C)/(G+C), green/purple peaks out/inside the circle indicates values higher or lower than 1, respectively).

Full size image

Insights into the genome

The genome of H. zincidurans strain B6T contains 31 genes related to heavy metal resistance, especially to zinc resistance (Table 6). Zinc is an essential but also toxic metal for living being [2, 43]. The concentration of zinc inside bacterial cells is maintained by importing limitation, efflux, accumulation and sequestration [44, 45]. H. zincidurans strain B6T possesses four heavy metal translocating P-type ATPases (HALZIN_733, HALZIN_1240, HALZIN_2196 and HALZIN_2262), which may participate in the transport of Zn2+, Mn2+, Cu2+, Cd2+, Pb2+, Ag + and Hg2+ against the concentration gradient to the periplasm [2, 44]. Especially the two ZntA P-type ATPases (HALZIN_733 and HALZIN_2196) may mediate resistance to Zn2+, Cd2+ and Pb2+ [46, 47]. Zn2+, Co2+, Cu2+, Cd2+ and Ni2+ are able to be transported by RND family efflux transporter protein (HALZIN_54, HALZIN_1411, HALZIN_2047, HALZIN_2208 and HALZIN_2209) from both the cytoplasm and the periplasm to outside [2, 44]. Usually the P-type ATPases are regulated by MerR family regulators responding to the intracellular heavy metal concentration [44, 48, 49]. Six analogues of MerR family regulators (HALZIN_399, HALZIN_922, HALZIN_2261, HALZIN_2264, HALZIN_2469 and HALZIN_2675) were found in the genome of H. zincidurans strain B6T. Additionally, a zinc uptake regulation protein ZUR (HALZIN_1413), which is a repressor regulator during zinc uptake, is also detected [44, 50]. The presence of these genes is accordance with zinc resistance phenotype of H. zincidurans strain B6T.

Table 6 Description of the genes related to heavy metal resistance
Full size table

Among the 31 ORFs related to heavy metal resistance, it is noteworthy of two mer-operons. One mer-operon encodes a mercuric transport protein (MerE, HALZIN_916) for organic mercury uptake [51], a transcriptional regulator (MerD, HALZIN_917), three alkylmercury lyases (MerB, HALZIN_918-920) catalyzing organomercurials yielding Hg2+ [52] and a transcriptional regulator (MerR, HALZIN_922). The other one encodes a transcriptional regulator (MerR, HALZIN_2469), two mercuric transport proteins (MerT and MerP, HALZIN_2470-2471) for inorganic mercury uptake [51] and a mercuric reductase (MerA, HALZIN_2472) catalyzing Hg2+ to Hg0 [53]. According to the genomic data, H. zincidurans strain B6T is able to survive in both inorganic and organic mercury environments. Interestingly, the four ORFs of the inorganic mer-operon showed the highest sequence identities to those of Halomonas lutea. Nevertheless, all the six ORFs of the organic mer-operon did not show the highest sequence identities to those of the genus Halomonas, but to the genera Burkholderia, Pseudomonas, Gladiecola and Stenotrophomonas, which indicates that the organic mer-operon might be acquired by HGT. Of special interest are the three alkylmercury lyases (MerB, HALZIN_918-920), which had obvious differences between the G+C content (56.6%; 57.1, 56.6 and 56.0% for these three gene sequences, respectively) as well as the G+C content at the third-codon positions (60.3%; 60.4, 61.0 and 59.4% for these three gene sequences, respectively) and those of the total protein-coding genes (65.4 and 82.8%, respectively). Besides, the RSCUs of nearly half of the 59 codons used by the three genes (23, 27 and 26 codons for HALZIN_918-920, respectively) change more than 2 folds, compared with those used by total protein-coding genes. 13 of the 31 ORFs (41.9%) were not related to Halomonadaceae genes according to the gene sequence similarity analysis, 9 of the 13 ORFs had RSCU change larger than 2 folds in more than 25% codons. These results indicated the existence of HGT events among the heavy metal resistance-related genes. Thus, HGT events might be an important way for H. zincidurans strain B6T to acquire heavy metal resistant ability and to adapt to the heavy metal rich environment.

Conclusion

The draft genome sequence of the heavy metal resistant bacteria H. zincidurans strain B6T isolated from the South Atlantic Mid-Ocean Ridge provide an insight into the genomic basis of its heavy metal resistance ability. And HGT may play an important role in its adaption to the heavy metal rich environment. On the basis of analysis and characterization of genome, H. zincidurans strain B6T might be resistant more kinds of heavy metal than we tested, such as Hg2+, Cd2+, Pb2+, Ni2+ and Ag+, etc. And it may have the potential for the bioremediation of multi-metal-contaminated environments. In addition, further analysis will be performed to confirm its resistant ability to other heavy metals and determine the mechanism of heavy metal resistance that we don’t know yet.