Arsenic and high affinity phosphate uptake gene distribution in shallow submarine hydrothermal sediments

Abstract

The toxicity of arsenic (As) towards life on Earth is apparent in the dense distribution of genes associated with As detoxification across the tree of life. The ability to defend against As is particularly vital for survival in As-rich shallow submarine hydrothermal ecosystems along the Hellenic Volcanic Arc (HVA), where life is exposed to hydrothermal fluids containing up to 3000 times more As than present in seawater. We propose that the removal of dissolved As and phosphorus (P) by sulfide and Fe(III)(oxyhydr)oxide minerals during sediment-seawater interaction, produces nutrient-deficient porewaters containing < 2.0 ppb P. The porewater arsenite-As(III) to arsenate-As(V) ratios, combined with sulfide concentration in the sediment and/or porewater, suggest a hydrothermally-induced seafloor redox gradient. This gradient overlaps with changing high affinity phosphate uptake gene abundance. High affinity phosphate uptake and As cycling genes are depleted in the sulfide-rich settings, relative to the more oxidizing habitats where mainly Fe(III)(oxyhydr)oxides are precipitated. In addition, a habitat-wide low As-respiring and As-oxidizing gene content relative to As resistance gene richness, suggests that As detoxification is prioritized over metabolic As cycling in the sediments. Collectively, the data point to redox control on Fe and S mineralization as a decisive factor in the regulation of high affinity phosphate uptake and As cycling gene content in shallow submarine hydrothermal ecosystems along the HVA.

Keywords: Arsenic biogeochemistry; Arsenic speciation; Hydrothermal activity; Phosphate biogeochemistry.

Fig. 1

Location of field site in the Aegean Sea. a Location of Milos along the Hellenic Volcanic Arc (HVA). b Sampling site. Dots are areas where shallow submarine activity has been located on the seafloor (Dando et al. 1995)

Fig. 2

Sample collection. a Seafloor diffused hydrothermal activity characterized by gas bubbles and the seafloor white and brown mats. b Close up view showing the sharp boundary between the white and brown seafloor deposit. c Close up view of white deposit on the seafloor. d Push core sample for the white-capped sediment. e Push core sample for the brown-capped sediment

Fig. 3

Total sediment sulfide obtained by the Cr distillation method and average porewater ICP-OES concentrations of S, Mn, Fe and As in sediment core down to 20 cm. a Total porewater sulfur versus depth. b Average porewater S, Mn, Fe and As. c Magnification of average porewater Mn, Fe and As content from (b). Bars are standard deviation from the mean. d Total sediment sulfide and Fe versus depth and habitat. e Total sediment As and P versus depth and habitat. f Total reactive Fe (Fe(III)(oxyhydr)oxides) to total reactive sediment sulfide content plotted against habitat and sediment depth

Fig. 4

Raman mineralogical analysis. a Raman spectra for marcasite intergrowth in pyrite in the white-capped sediment. b Raman spectra for Fe oxides phases. c Raman spectrum for quartz

Fig. 5

Arsenate, arsenite and phosphate distribution in sediment porewater. a Arsenite. b Arsenate. c Phosphate. d Arsenite/Arsenate ratio. e Arsenate/phosphate ratio. f Mean concentration, arsenite/arsenate and arsenate/phosphate ratios. Bars standard deviation from the mean. Concentrations are per gram of sediment

Fig. 6

Average arsenic and high affinity phosphate gene abundance in sediment and seawater. a Average gene abundance for the periplasmic aoxB, arsB and acr3-1 and acr3-2 and arrA genes. b pstB gene abundance specific for the Geobacteraceae

Fig. 7

Geobacteraceae 16S rRNA gene abundance against As and high affinity phosphate uptake genes. Linear fit with R² values of 0.98403, 0.96795, 0.91712, 0.84513, 0.84194 and 0.73033 for aoxB, arsB, acr3-2, acr3-1, pstB and arrA, respectively

Fig. 8

Principal component analysis (PCA). Plots are for selected parameters, As respiration, As resistance and high affinity phosphate uptake genes, total sediment P and As and reactive Fe to S ratios. a Screen plot showing a maximum of three-factor analysis possible for the dataset. b Three-factor PCA analysis of the correlation matrix plotted against habitat (see Table S3 for calculated variables). c Sediment biplot. d Complete linkage-correlation distance similarity dendrogram

Fig. 9

Conceptual model for As and P cycling along the sampled transect. The size of the red and black arrows are proportional to the quantity of As sulfides and Fe(III)(oxyhydr)oxides precipitated and stored in sediments. Open circles represent hydrothermal gas bubbles