Salt-tolerance Function in Halotolerant Bacteria from Coastal Soil and Sediment

Understanding the salt-tolerance function in microbial communities may help illustrate the influences of salt concentration and other environmental factors on the soil biodiversity. Here, we combined high-throughput sequencing of 16S rDNA and ITS to investigate the distribution and salt-tolerance function of microbial communities in coastal soils and sediments present near the Yinggehai saltern field of Hainan Island, China. The microbial communities in the soil and sediments of the land zone (YGHLS), the intertidal zone (YGHIS), and the inshore zone (YGHWS) were compared. PCoA of weighted and unweighted UniFrac distances showed the difference in soil microbial communities across the samples. ANOSIM analysis revealed that the three samples were separated from each other. Halomonas , Halobacillus , and Wallemia presented in the samples and accounted for 0.0335 ± 0.0586%, 0.0241 ± 0.0304%, and 0.0308 ± 0.0445% of the total microbial community. Relative abundance of Na + :H + antiporter, Trk system potassium uptake protein, and Kdp operon response regulator in the samples exceed 0.02%, 0.09%, and 0.06% respectively. It revealed that the halotolerant bacteria of Yinggehai coastal soil and sediment mainly take the Trk system for salt-tolerance function.


Introduction
Island ecosystem is a unique and fragile ecosystem on earth. They provide special habitat for a wide variety of terrestrial and aquatic organisms and play important ecological and economic roles. Due to the high salt environment and other special environment factors, it owns abundant microbial genetic resources (Somboonna et al. 2012, Ng and Tan 2018, Sottorff et al. 2019. Characteristics of microbial community that exists in the coast near the saltern of an island is an interesting theme in ecology. To assess the effects of hypersaline environment on microbial composition, many microbial biogeography studies examined coastal soils and sediments present in saltern fields. The phyla Proteobacteria, Firmicutes, and Actinobacteria were isolated from coastal solar salterns and revealed they have the ability to synthesize bioactive molecules (Boyadzhieva et al. 2018). Tang et al. (2011) reported 47 bacterial isolates either in the phylum Firmicutes or in the class Gammaproteobacteria isolated from a hypersaline pond in Sichuan, China.
Characterization of heterotrophic prokaryote subgroups in the Sfax coastal solar salterns showed that the Alphaproteobacteria class in the non-crystallizer pond samples and the Bacteroidetes phylum in the crystallizer pond samples (Trigui et al. 2011). The different microbial genera present in the saline sediments and soils belonged mostly to halophilic bacteria, suggesting that the coast near saltern may be a valuable system to investigate saline influence the microbial community.
Yinggehai saltern is located in the southwest of Hainan Island, China. It was built in 1958, a high-salinity environment in the coastal waters. The saltern area is warm and hot all year and has very distinct wet and dry season. Evaporation is much greater than rainfall, and this superior climate condition has become a suitable place for salt drying. This study aims to explore the roles of the microbial community in the coastal soil and sediment near Yinggehai saltern by using 16S and ITS ribosomal DNA sequencing. In addition, we focus on salt-tolerance function in the microbial community of the soil and sediment. It will unravel the geographical factors affect the halotolerant bacteria community.

Study Site and Sample Collection
Yinggehai saltern is located in Hainan Island, China (Fig. 1). Three successional areas were selected for analysis, including the land zone (YGHLS, surrounded by plants, 18.4948N 108.7158E), the intertidal zone (YGHIS, ebb and flow of the tide, 18.4931N 108.7156E), and the inshore zone (YGHWS, immersed by 4 seawater, 18.4898N 108.7153E). The samples were collected in July 2018. Quadruplicate 1 × 1 m 2 plots were separated by 10 m in each area. For each plot, 5 soil cores (5-10 cm diameter × 5-10 cm depth) were taken to make a composite sample.

Determination of Physicochemical Characteristics
Soil and sediment samples were air-dried and sieved through a 100-mesh sieve for determination of organic matter (OM). Electrical conductivity (EC), pH, total nitrogen, sodium (Na), sulfate (S-SO 4 2− ), nitrate (N-NO 3 − ) and ammonium (N-NH 4 + ) were sieved through a 10-mesh sieve. The physicochemical characteristics were tested according to the method described by Chen et al. (2020).

Bioinformatics and Statistical Analysis
Chao1 richness estimator, Shannon diversity index, Simpson index, and ACE metric (abundance-based coverage estimator) were estimated using the OTU table in QIIME. Beta diversity analysis was performed using UniFrac distance metrics (Lozupone and Knight, 2005;Lozupone et al. 2006) and visualized via principal coordinate analysis (PCoA). Significant differences in microbial structures were evaluated based on analysis of similarity (ANOSIM) performed by the R package "vegan" (v3.2.0) (Clarke 1993). Venn diagrams were generated using the R package "VennDiagram" (Zaura et al. 2009). The linear regression analysis was conducted with Origin v 9.1. PICRUSt2 analysis was performed to gain insight for the salt-tolerance function (Langille et al. 2013). The gene family counts for each sample were derived from the KEGG ortholog (KO).

Sequence Accession Numbers
The raw sequence data are available in the NCBI Sequence Read Archive (SRA) database under the accession numbers PRJNA699381 and PRJNA699383.

Site Characteristics
The selected sampling sites were from the land zone to the inshore zone, which is nearby Yinggehai saltern. 6 The variations of a wide range of physicochemical factors were observed. This analysis could reflect the ecosystem nearby the saltern zone (Table S1). The NH 4 + -N of the samples from the three sites did not vary much. The YGHWS sample had significant (Duncan test, P < 0.05) higher sodium (2338.47±78.58 mg/kg) and sulfate (187.35±8.50 mg/kg) contents. Detailed descriptions of the data are provided in Supplementary   Table S1.

Microbial Community Diversity and Structure
The microbial communities could be adequately represented by the observed OTUs based on the relatively stable rarefaction curves (Fig. S1). Statistical results demonstrated that the most diversity indices were significantly different among the three sampling points at bacterial and fungal communities (Table 1).
Although Shannon, evenness, and simpson index of bacterial community were not significantly different among the three sampling points, the other two diversity indices were significantly related to each other. Venn diagrams showed that the unique bacterial OUTs in the YGHLS, YGHIS, and YGHWS were 2118, 1587 and 2416, the unique fungal OUTs in the YGHLS, YGHIS, and YGHWS were 775, 716 and 1351 (Fig. S2). The difference in soil microbial communities across the samples was showed by PCoA (Fig. 2). ANOSIM analysis (Bray-Curtis) revealed that the three samples were separated from each other (Table S2,

Microbial Community Composition
The taxonomy of the bacterial community at the phylum and genus levels is presented in Fig. 3A and Fig. 3B.
Proteobacteria, Actinobacteria, Acidobacteria, Chloroflexi, and Gemmatimonadetes were the five most abundant phyla in all samples in decreasing order and accounted for 84.41 ± 3.23% of the total microbial community. Halomonas and Halobacillus presented in the samples and accounted for 0.0335 ± 0.0586% and 0.0241 ± 0.0304% of the total microbial community, respectively. The identified phyla and genera of samples from the three sites were shown in Fig. 3A and Fig. 3B. Furthermore, the samples from three sites clustered independently in the bacterial genus distribution (Fig. S 3A).
Furthermore, we analyzed the relative abundance of fungi in the three samples at the phylum and genus 7 levels. The fungal phyla Ascomycota, Basidiomycota, Ciliophora, Zygomycota, Cercozoa, and Chytridiomycota were more abundant (Fig. 3C). Schizophyllum, Bullera, Cryptococcus, Talaromyces, Aspergillus, Tomentella, Spiromastix, and Auricularia were detected in all the samples (Fig. 3D). A rare genus Wallemia presented in the samples and accounted for 0.0308 ± 0.0445% of the total microbial community.
The patterns of fungal genus distribution are showed in Fig. S 3B.

Bacteria and Fungi
Sodium and sulphate were selected from our initial physicochemical indicators in the linear model, showing that they did not have linear relationships with Halobacillus and Wallemia (Fig. 4). However, a bacterial genus Halomonas showed positive relationship with sodium (R 2 = 0.501, P < 0.05) and sulphate (R 2 = 0.494, P < 0.05) content. It indicated that the sodium and sulphate content significantly affected the Halomonas relative abundance, whereas they did not relate to Halobacillus and Wallemia.

Relationship between Microbial Indicators and Halophilic Bacteria Abundance
The structure (weighted PCoA1) and richness were selected as microbial indicators in the linear model, as they could explain the relationship with halophilic bacteria abundance. As a result, bacterial weighted PCoA1 (R 2 = 0.065, P > 0.05) and richness (R 2 = 0.118, P > 0.05) did not show linear relationships with Halobacillus and Halomonas abundance respectively. However, we found that bacterial weighted PCoA1 (R 2 = 0.393, P < 0.05) had a significant positive relationship with Halomonas abundance. Similarly, we also found that bacterial richness (R 2 = 0.348, P < 0.05) had a significant positive relationship with Halobacillus abundance (Fig. 5).

Discussion
Proteobacteria, Bacteroidetes, Chloroflexi, and Firmicutes dominated the bacterial communities of sediment samples present near the Mediterranean Sea coast of Camargue, while members belonging to Proteobacteria, Bacteroidetes, Gemmatimonadetes, Actinobacteria, Firmicutes, and Acidobacteria dominated the bacterial communities of the soil samples (Osman et al. 2018). In this study, Proteobacteria, Actinobacteria, Acidobacteria, Chloroflexi and Gemmatimonadetes were the five most abundant phyla in the samples.
Pomorie salterns and Burgas salterns located at Burgas Bay, Black Sea coast, Bulgaria. And, bacterial genera Halomonas was isolated from the two salterns (Boyadzhieva et al. 2018 Interestingly, the bacterial weighted PCoA1 had a significant positive relationship with Halomonas abundance. Thus, Halomonas abundance was directly influenced by physicochemical characteristics or bacterial weighted PCoA1? It still needs more studies to ascertain. The Na + /H + antiporter is ubiquitous from bacteria to mammals and plays an important role in extrusion of Na + or Li + in exchange for H + to keep the cytoplasm isosmotic with the environment and to avoid intoxication of living cells (Yang et al. 2006). At a high external salt concentration, bacteria take up K + from the surrounding medium and accumulate K + as an osmoregulatory solute to achieve an osmotic equilibrium. The major transport systems for K + accumulation in these organisms is the transporter Kdp, the Ktr system, and the Trk transporter (Kraegeloh et al. 2005). Relative abundance of Trk transporter was higher than Na + :H + antiporter and Kdp transporter in the samples. It indicated that the halotolerant bacteria of Yinggehai coastal soil and sediment mainly take the Trk system in response to Na + osmotic stress.