Article pubs.acs.org/est
Role of Morphological Growth State and Gene Expression in Desulfovibrio af ricanus Strain Walvis Bay Mercury Methylation James G. Moberly,† Carrie L. Miller,‡ Steven D. Brown,† Abir Biswas,‡,§ Craig C. Brandt,† Anthony V. Palumbo,† and Dwayne A. Elias†,* †
Biosciences Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States § The Evergreen State College, Olympia, Washington 98505, United States ‡
S Supporting Information *
ABSTRACT: The biogeochemical transformations of mercury are a complex process, with the production of methylmercury, a potent human neurotoxin, repeatedly demonstrated in sulfate- and Fe(III)-reducing as well as methanogenic bacteria. However, little is known regarding the morphology, genes, or proteins involved in methylmercury generation. Desulfovibrio af ricanus strain Walvis Bay is a Hg-methylating δproteobacterium with a sequenced genome and has unusual pleomorphic forms. In this study, a relationship between the pleomorphism and Hg methylation was investigated. Proportional increases in the sigmoidal (regular) cell form corresponded with increased net MeHg production but decreased when the pinched cocci (persister) form became the major morphotype. D. af ricanus microarrays indicated that the ferrous iron transport genes (feoAB), as well as ribosomal genes and several genes whose products are predicted to have metal binding domains (CxxC), were up-regulated during exposure to Hg in the exponential phase. Whereas no specific methylation pathways were identified, the finding that Hg may interfere with iron transport and the correlation of growth-phase-dependent morphology with MeHg production are notable. The identification of these relationships between differential gene expression, morphology, and the growth-phase dependence of Hg transformations suggests that actively growing cells are primarily responsible for methylation, and so areas with ample carbon and electron-acceptor concentrations may also generate a higher proportion of methylmercury than more oligotrophic environments. The observation of increased iron transporter expression also suggests that Hg methylation may interfere with iron biogeochemical cycles.
■
INTRODUCTION Elemental mercury (Hg0) is a naturally occurring, toxic heavy metal with no known biological function.1 Inorganic mercury (Hg2+) occurs naturally, primarily from geological cinnabar deposits with human activity adding Hg to the biosphere primarily through fossil fuel combustion, waste incineration, and industrial/commercial uses.2−5 Atmospheric Hg0 cycles in the biosphere6 where gaseous Hg0 is photochemically oxidized and deposited by rainfall and deposition.7 Most Hg2+ deposited to terrestrial systems is sequestered by ligands in the soil and vegetation, with a small fraction being reduced to Hg0 and returned to the atmosphere. The various Hg complexes then control its subsurface mobility and bioavailability into lower redox zones where methylating microorganisms may convert it to monomethylmercury (MeHg).7−10 MeHg is readily transferred through the food web in aquatic ecosystems with biomagnification at each trophic level11 with fish tissue concentrations that are 106−107 fold higher than in water,12−15 thus posing toxicological risks to fish consumers even in pristine environments.14,15 Microbial MeHg generation is highly sensitive to environmental factors16−20 and so the © 2012 American Chemical Society
ecology, physiology, and genetics of microorganisms within subsurface redox transition zones are critical components in understanding Hg2+ methylation. The transformation of Hg2+ to MeHg is believed to be driven by anaerobic microbial processes although the exact mechanism is unknown.21 The primary culprits for methylation have been shown to be the sulfate-reducing (SRB) and Fe(III)-reducing (IRB) bacteria12,19,22−25 within the δ-proteobacteria but this capability is not phylogenetically consistent as some members of a given genus can methylate Hg2+, whereas others do not.26,27 Orders within the δ-proteobacteria that can methylate Hg2+ to MeHg include the Desulfovibrionales,27−29 Desulfobacterales,28,30 and Desulf uromonadales.29,31,32 Of these orders, Desulf uromonadales and Desulfovibrionales and particularly the genera Desulfovibrio and Geobacter have received the greatest attention. The incomplete oxidizer Desulfovibrio desulf uricans strain ND132 Received: Revised: Accepted: Published: 4926
January 12, 2012 April 10, 2012 April 13, 2012 April 13, 2012 dx.doi.org/10.1021/es3000933 | Environ. Sci. Technol. 2012, 46, 4926−4932
Environmental Science & Technology
Article
the enriched isotope was added just prior to the reduction of the Hg with hydroxylamine and stannous chloride. An enriched MeHg, synthesized from Hg2+ using methylcobalamin,43 was added to the sample prior to distillation. Enriched Hg isotopes (purchased from Oak Ridge National Laboratory), were all greater than 95% pure and the small abundance of other isotopes was taken into account during data processing.44 Details of the Hg and MeHg analyses can be found in the EPA methods. DNA Microarray Analysis. Three D. af ricanus cultures with and without exposure to 5 μg/L Hg (as HgNO3) were sampled at 24 and 48 h postinoculation corresponding to midexponential and early stationary phase, respectively. Total RNA was isolated and quantified as described previously.40 Commercial DNA microarray probes were designed from a draft version of the D. af ricanus strain Walvis Bay (ATCC 19997) genome (Roche NimbleGen, IN) and annotations updated once the genome was finished.36 The cDNA was labeled, hybridized, washed, and scanned following the NimbleGen protocols. Statistical analyses were conducted with JMP Genomics 4.1 software (SAS Institute, NC), essentially as described previously.45 The data were normalized using the standard normalization algorithm within JMP Genomics. An analysis of variance was performed to determine differential expression levels between conditions using the false discovery rate (FDR) testing method (p < 0.05) that resulted in a corresponding significance cut off value of −log(p) = 5.6. Additionally, only genes which were significantly differentially expressed and showed a ≥2-fold change in expression were included in the analysis.46 The entire data set was submitted to the National Center for Biotechnology Information Gene Expression Omnibus under accession number GSE32580. Geochemical Speciation. Speciation calculations of the medium constituents were performed using the geochemical modeling program PHREEQC Interactive (v 2.17.4799) 47 with a modified version of the database that included mercurysulfide aqueous complexes and solid phases.48 No term was employed for sorption of Hg onto cell surfaces. Additionally, speciation terms were not used for the medium buffer TRIS and vitamin constituents. Redox potential calculations were based on the sulfate/sulfide couple with an initial value for pE at −4. The pH of the medium (pH 7) was modeled as constant throughout the calculations.
is a strong producer of MeHg that has had its genome sequenced 33and is considered a model organism for studying mercury methylation.27 Desulfovibrio af ricanus has also been studied for its moderate ability to methylate Hg independent of cobalt or vitamin B12 as opposed to the complete oxidizer Desulfococcus multivorans that produced three times less MeHg during cobalt limitation.28 D. af ricanus also displays an unusual pleomorphism34 that varies with growth phase. Recent work35 showed that growth phase may be important in MeHg production in strain ND132 but the impact of D. af ricanus pleomorphic growth on Hg2+ methylation is unknown. With the genome of D. af ricanus strain Walvis Bay now sequenced,36 the present study assessed the possible links between MeHg generation and growth-phasedependent cell morphology. Transcriptomic profiles were generated for D. af ricanus exponential and stationary phase cultures in the presence and absence of Hg2+ in an attempt to identify genes or pathways involved in Hg2+ transport and MeHg production.
■
METHODS D. af ricanus Growth. D. af ricanus strain Walvis Bay (ATCC 19997) was grown in MOYLS4 medium37 except that modified vitamin and mineral solutions 38 were used instead of Thauers vitamin and trace metals, and iron(III)-ethylenediaminetetraacetic acid was not included in the medium. Sodium lactate (60 mM) and sodium sulfate (30 mM) were added as the sole electron donor and acceptor, respectively. One liter D. af ricanus cell cultures were grown in two liter anaerobic bottles at 30 °C with 10% inocula (v/v) from late exponential phase cultures and corresponded to A660 = 0.05. Samples were taken every 6 h for organic acid quantification, cell counts and morphology characterization utilizing a Petroff-Hauser cell counting chamber on a Zeiss Axioskop2 plus microscope with 40× objective (Zeiss Plan-NEOFLUAR), colorimetric sulfide determination (Hach method 8131),39 and Hg2+ methylation assays. Organic acids were measured using HPLC (Waters model 2414 binary HPLC) with a refractive index detector (Waters model 2414 refractive index detector). The method for HPLC analysis utilized isocratic flow (5 mM sulfuric acid, 0.6 mL/min, 50 °C) on an Aminex HPX-87H column as has been described previously.40 Mercury Methylation Assays. Mercury methylation profiles were ascertained using enriched stable isotopes.41,42 At 6 h intervals, aliquots of cell culture (15 mL) were injected into N2-purged, anaerobic, sterile Balch tubes, and inorganic 201 Hg2+ (was added (5 μg/L). The conversion of 201Hg2+ to 201 MeHg (1 h) was used to qualify the Hg2+ methylation potential. Cultures were then preserved within the Balch tube (9 M H2SO4, 500 μL) and stored at 4 °C until analysis. Sample pH was verified to be
