UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The biomethylation of arsenic Li, Hao 1994

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1994-95356x.pdf [ 4.28MB ]
Metadata
JSON: 831-1.0059605.json
JSON-LD: 831-1.0059605-ld.json
RDF/XML (Pretty): 831-1.0059605-rdf.xml
RDF/JSON: 831-1.0059605-rdf.json
Turtle: 831-1.0059605-turtle.txt
N-Triples: 831-1.0059605-rdf-ntriples.txt
Original Record: 831-1.0059605-source.json
Full Text
831-1.0059605-fulltext.txt
Citation
831-1.0059605.ris

Full Text

THE BIOMETHYLATION OF ARSENICbyHao LiB. Sc., University of Science and Technology of China, China, 1985M. Sc., University of British Columbia, Canada, 1989A THESIS SUBM1’rikD IN PARTIAL FULFILLMENT OFTHE REQUIREMENT FOR THE DEGREE OFDOCTOR OF PHILOSOPHYmTHE FACULTY OF GRADUATh STUDIESDEPARTMENT OF CHEMISTRYWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAJUNE, 1994© Hao Li, 1994In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives, It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)____________________Department of___________The University of British ColumbiaVancouver, CanadaDate____ __DE-6 (2/88)ABSTRACTA semi-continuous hydride generation-gas chromatography-atomic absorptionspectrometry (HG-GC-AA) system was developed and optimized for the determinationof arsenite, arsenate, methylarsonate (MMA), dimethylarsinate (DMA), andtrimethylarsine oxide (TMAO). Particularly, this system was used to study the pathwayfor the biomethylation of arsenicals in microorganisms and a marine alga.The HG-GC-AA system was used to separate and identify the extracellulararsenic metabolites produced by the microorganisms Apiotrichum humicola (previouslyknown as Candida humicola) and Scopulariopsis brevicaulis growing in the liquidmedium enriched with arsenicals. Arsenite, MMA, DMA, and TMAO were detectedfollowing incubation with arsenate. With arsenite as a substrate, the metabolites wereMMA, DMA, and TMAO; MMA afforded DMA and TMAO, and DMA affordedTMAO. Trimethylarsine was not detected in these investigations. The production of theanticipated methylated intermediates from the substrates strongly support the metabolicsequence proposed by Challenger (Challenger, F Chem. Rev., 1945, 36:3 15).When L-methionine-methyl-d3was added to the growing culture of Apiotrichumhumicola grown in the presence of either arsenate, arsenite, MMA, or DMA, the CD3label was incorporated intact into the arsenic metabolites (DMA and TMAO) to aconsiderable extent, indicating that S-adenosylmethionine (SAM), or some relatedsuiphonium compound, is involved in the biological methylation. Conclusive evidence ofCD3 incorporation into the arsenicals was provided by using a specially developedhydride generation-gas chromatography-mass spectrometry methodology (HG-GC-MS).When a unicellular marine alga Polyphysa peniculus was grown in artificialseawater enriched with arsenicals, the arsenic metabolites produced in the cells as wellas in the growth medium were identified by using HG-GC-AA methodology. Arsenite11and DMA were detected following incubation with arsenate. When the alga was treatedwith arsenite, DMA was the major metabolite in the cells and in the growth medium;trace amounts of MMA were also detected in the cells. With methylarsonate as asubstrate, the metabolite was dimethylarsinate. Polyphysa peniculus did not metabolizedimethylarsinic acid when it was used as a substrate. Significant amounts of morecomplex arsenic species, such as arsenosugars, were not observed in the cells or mediumbased on the evidence given by flow injection-microwave digestion-hydride generation-atomic absorption spectrometry methodology. Transfer of the exposed cells to freshmedium caused release of most cell associated arsenicals to the surroundingenvironment. The alga seems to follow the biomethylation pathway proposed byChallenger for microbial process, and in the case of P. peniculus, DMA is the endproduct of this biomethylation.When L-methionine-methyl-d3was added to the culture of Polyphysa peniculusenriched with 1 ppm of arsenate, the CD3 label was incorporated intact in the DMAmetabolite to a considerable extent. It thus confirmed that P. peniculus also follows theoxidation-reduction pathway involving carbonium ions originally suggested byChallenger for the alkylation of arsenic by microorganisms.The HG-GC-MS system was also used to identify the antimony hydridesproduced from the trimethylantimony compounds Me3Sb(OH)2 and Me3SbC12.Thepossible causes of the molecular rearrangement of trimethyistibine were investigated.The extracts of plant samples collected from Kam Lake and Keg Lake (Yellowknife)were analyzed by using the HG-GC-MS system. The results provided conclusiveevidence of the presence of methylantimony compounds in these samples.U’TABLE OF CONTENTSABSTRACT.iiLIST OF TABLES viiiLIST OF FIGURES xLIST OF ABBREVIATIONS xviiACKNOWLEDGMENTS xixCHAPTER 1 General introduction 11.1 Biological transformation of arsenic 21.1.1 Redox transformation between arsenate and arsenite 31.1.2 The biomethylation of arsenic 31.1.3 Organoarsenic compounds in the marine environment 101.1.3.1 Arsenic compounds in marine algae 101.1.3.2 Organoarsenic compounds in marine invertebratesandfish 171.2 Scope of work 20CHAPTER 2 Analytical methodology 222.1 Introduction 222.2 Experimental 232.2.1 Chemicals and reagents 232.2.2 Apparatus 242.2.2.1 Graphite furnace atomic absorptionspectrometry (GFAA) 242.2.2.2 Hydride generation-gas chromatography-atomicabsorption spectrometry (HG-GC-AA) 242.2.3 Analytical procedures 262.2.3.1 GFAA 262.2.3.2 HG-GC-AA 282.3 Results and discussion 29iv2.3.1 Graphite furnace atomic absorption spectrometry (GFAA) 292.3.2 Hydride generation-gas chromatography-atomic absorptionspectrometry (HG-GC-AA) 332.3.2.1 The hydride generation system 352.3.2.2 Optimization 382.3.2.3 Interference studies 462.3.2.4 Calibration, limit of detection and precision ofHG-GC-AA analysis 47CHAPTER 3 The identification of extracellular arsenical metabolites in thegrowth medium of microorganisms 503.1 Introduction 503.2 Experimental 523.2.1 Reagents 523.2.2 Microorganism cultures 523.2.3 Hydride generation-gas chromatography-atomic absorptionspectrometry 523.2.4 Experimental procedures 533.3 Results 553.3.1 Transformation of arsenate 553.3.2 Transformation of arsenite 613.3.3 Transformation of methylarsonate 633.3.4 Transformation of dimethylarsinate 673.4 Discussion 67CHAPTER 4 The biomethylation of arsenicals by a microorganism Apiotrichumhumicola in the presence of L-methionine-methyl-d3 754.1 Introduction 754.2 Experimental 774.2.1 Reagents 774.2.2 Microorganism cultures 774.2.3 Hydride generation-gas chromatography-massspectrometry 77V4.2.4 Experimental procedures.804.3 Results and discussion 804.3.1 HG-GC-MS measurements 804.3.2 Characterization of methylated arsenic intermediates in thegrowth medium of A. humicola 864.3.2.1 Transformation of arsenate 864.3.2.2 Transformation of arsenite 924.3.2.3 Transformation of methylarsonic acid 944.3.2.4 Transformation of dimethylarsinic acid 98CHAPTERS The biomethylation and bioaccumulation of arsenicals by a marinealga Polyphysa peniculus 1015.1 Introduction 1015.2 Experimental 1035.2.1 Reagents 1035.2.2 Algal cultures 1045.2.3 Instrumentation 1045.2.3.1 Graphite furnace atomic absorption spectrometry 1045.2.3.2 Hydride generation-gas chromatography-atomicabsorption spectrometry 1065.2.3.3 Flow injection-microwave digestion-hydridegeneration-atomic absorption spectrometry 1065.2.3.4 Hydride generation-gas chromatography-massspectrometry 1085.2.4 Experimental procedures 1085.3 Results 1105.3.1 The accumulation of arsenicals in cells of P. peniculus 1105.3.2 Arsenic speciation analysis in cells of P. peniculus 1115.3.2.1 Water/methanol extracts ill5.3.2.2 Chloroform extracts 1155.3.2.3 Insoluble residues 1155.3.3 Arsenic speciation analysis in the growth media of P.peniculus 1165.3.4 Arsenic efflux studies 1205.3.5 Characterization of dimethylarsenic derivative in growthmedium by using HG-GC-MS 122VI5.4 Discussion 131CHAPTER 6 The identification and characterization of antunony(ffl),antimony(V), and methylantimony species by using HG-GC-MS 1406.1 Introduction 1406.2 Experimental 1416.2.1 Reagents 1416.2.2 Hydride generation-gas chromatography-massspectrometry 1426.2.3 Experimental procedures 1426.3 Results and discussion 143CHAPTER 7 SUMMARY 158BIBLIOGRAPHY 163APPENDIX Determination of hydride-forming and “hidden” arsenicals in theseawater surface microlayer by using microwave digestionfollowed by HG-GC-AA 175A.1 Introduction 175A.2 Experimental 176A.3 Results and discussion 178A.3.1 Determination of hydride-forming arsenicals 178A.3.2 Determination of “hidden” arsenic species 185vuLIST OF TABLESTABLES PAGE1.1 Arsenic concentrations (p.glg) in some marine algae 112.1 Furnace operating parameters for the determination of arsenic 272.2 Typical sampling parameters for Standard Additions Method for GFAAanalysis 282.3 Operating conditions for the HG-GC-AA system 302.4 Reaction conditions for the conversion of inorganic and methylarseniccompounds to volatile arsines 352.5 Percentage deviations of the response of arsenic species in the presence ofinterfering ions (10 ppm ) 483.1 Composition of the growth medium (1 L)32 524.1 HG-GC-MS experimental parameters 794.2 Percentage distribution of dimethylarsenic and trimethylarsenic compoundsdetected in the growth medium 935.1 Composition of Shephard’s medium 1055.2 Total amount of arsenic in cells of P. peniculus determined by using GFAA(.tgIg, dry weight) 1115.3 Arsenic distribution in aqueous extracts of the cells determined by usingHG-GC-AA (p.glg, dry weight) 1135.4 Arsenic distribution in aqueous extracts of the cells harvested from arsenicalenriched media before and after microwave digestion (j.tglg, dry weight) 1145.5 Percentage distribution of dimethylarsenic compounds detected in themedium 131A. 1 Optimum conditions for the determination of hydride-fonning arsenicals in abatch type hydride generator 178yinA.2 Optimum conditions for the decomposition of “hidden” arsenicals. 179A.3 The concentrations of hydride-forming arsenicals (ppb) in seawater surfacemicrolayer samples 184A.4 The concentrations of arsenic (ppb) in seawater surface microlayer sampleswith and without microwave assisted digestion 186IxLIST OF FIGURESFIGURES PAGE1.1 Challenger’s mechanism for the biomethylation of arsenic.30’43Theintermediates in { } are unknown as monomeric species. They can beisolated as (CH3AsO)n and (CH3As)20,respectively, when prepared byconventional methods 71.2 The mechanism for arsenical reduction by U’io46’7 71.3 Structure of S-adenosylmethionine (SAM).49The arrow indicates themethyl group that is used for methylation 81.4 Structure of arsenosugar derivatives 131.5 Proposed mechanism for transformation of arsenic compounds in marinealgae.8°Unidentified compounds are underlined. The A represents Me+from SAM, and B is the adenosyl group from SAM 151.6 Structure of organoarsenicals isolated from marine invertebrates and fish 181.7 Possible pathways for the production of arsenobetaine from arsenosugarderivatives 192.1 Scheme of the major components of a hydride generation system 232.2 Hydride generation-gas chromatography-atomic absorption spectrometryexperimental setup. Dotted lines represent gas lines 252.3 Effect of the concentration of hydrochloric acid on the determination ofhydride-forming arsenicals. The reductant was a 2.0% (wlv) NaBH4aqueous solution. (0) arsenite (•) arsenate (V) MMA (V) DMA(D)TMAO 392.4 Effect of the concentration of acetic acid on the determination of hydride-forming arsenicals. The reductant was a 2.0% (wlv) NaBH4aqueoussolution. (0) arsenite (V) MMA (V) DMA (D) TMAO 412.5 Effect of the concentration of NaBH4on the determination of hydride-forming arsenicals. The acid medium was a 1.0 M hydrochloric acidsolution. (0) arsenite (•) arsenate (V) MMA (I) DMA (D) TMAO 43x2.6 Effect of the concentration of NaBH4on the determination of hydride-forming arsenicals. The acid medium was a 4.0 M acetic acid. (0) arsemte(V) MMA (V) DMA (D) TMAO 442.7 Effect of the carrier gas flow rate (in reaction coil) on the determination ofhydride-forming arsenicals. (0) arsenite (•) arsenate (V) MMA(V) DMA (D) TMAO 453.1 The growth flask fitted with the mercuric chloride trap 543.2 Chromatograms of arsenic compounds in the growth medium of A.humicola (enriched with 1 ppm arsenate) were obtained by using HG-GCAA with 1.0 M HC1 and 2.0% (wlv) NaBH4.(a) the growth mediumcollected on day 0, (b) the growth medium collected on day 2 573.3 Chromatograms of arsenic compounds in the growth medium of A.humicola (enriched with 1 ppm arsenate) were obtained by using HG-GCAA with 4.0 M acetic acid and 2.0% (wlv) NaBH4.(a) the growth mediumcollected on day 0, (b) the growth medium collected on day 2 573.4 Chromatograms of arsenicals in the growth medium of A. humicola(enriched with 1 ppm arsenate) were obtained by using HG-GC-AA with4.0 M acetic acid and 2.0% (wlv) NaBH4.(a) the growth medium collectedon day 0, (b) the growth medium collected on day 5, (c) the growth mediumcollected on day 15, (d) the growth medium collected on 28 583.5 The change in arsenic concentrations in the growth medium of A. humicolaenriched with 1 ppm arsenate. (0) arsenite (•) arsenate (‘V) DMA (D)TMAO 593.6 The effect of adsorption by the cells of A. humicola on the concentration ofTMAO in the growth medium. (a) without agitation before analysis, (b)with agitation before analysis 603.7 Chromatograms of arsenicals in the growth medium ofA. humicola(enriched with 1 ppm arsenite) were obtained by using HG-GC-AA with 4.0M acetic acid and 2.0% (wlv) NaBH4 (a) the growth medium collected onday 0, (b) the growth medium collected on day 5, (c) the growth mediumcollected on day 15, (d) the growth medium collected on day 28 623.8 The change in arsenic concentrations in the growth medium of A. humicolaenriched with 1 ppm arsenite. (0) arsenite (V) DMA (D) TMAO 643.9 Chromatograms of arsenicals in the growth medium of A. humicola(enriched with 1 ppm MMA) were obtained by using HG-GC-AA with 4.0M acetic acid and 2.0% (wlv) NaBH4. (a) the growth medium collected onday 0, (b) the growth medium collected on day 2, (c) the growth mediumcollected on day 15, (d) the growth medium collected on day 28 653.10 The change in arsenic concentrations in the growth medium of A. humicolaenriched with 1 ppm MMA. (V)MMA (V) DMA (D) TMAO 663.11 Proposed biotransformation model of arsenate in A. humicola.a endocellulararsenicals b extracellular arsenicals 734.1 Hydride generation-gas chromatography-mass spectrometry experimentalsetup. Dotted lines represent gas lines 784.2 HG-GC-MS chromatogram of arsine, monomethylarsine, dimethylarsine,and trimethylarsine. Solutions of standard arsenite, MMA, DMA, andTMAO (100 ng of arsenic for each compound) were used 824.3 Mass spectra of arsines derived from standard solutions of arsenite, MMA,DMA, and TMAO. (a) arsine AsH3,(b) monomethylarsineCH3As2,(c)dimethylarsine (CH3)2As , and (d) trimethylarsine (CH3)As 834.4 The predicted fragmentation pattern of (CD3)As.3 854.5 HG-GC-MS chromatogram of a 2 mL culture medium sample that wastaken after 10 days of incubation. The medium was originally enriched with1 ppm arsenate, but not with L-methionine-methyl-d3 884.6 Mass spectrum of the first peak in Figure 4.5, identified as CO2 894.7 Mass spectra of the hydride derivative of the dimethylarsenic speciespresent in the growing culture, enriched with arsenate, after 10 days ofincubation. (a) in the absence of L-methionine-methyl-d3(b) in thepresence L-methionine-methyl-d3 904.8 Mass spectra of the hydride derivative of the trimethylarsenic speciespresent in the growing culture, enriched with arsenate, after 10 days ofincubation. (a) in the absence of L-methionine-methyl-d3(b) in thepresence L-methionine-methyl-d3 91XII4.9 HG-GC-MS chromatogram of a 2 mL culture medium sample that wastaken after 10 days of incubation. The medium was originally enriched with1 ppm MMA, but not with L-methionine-methyl-d3 954.10 Mass spectra of the hydride derivative of the dimethylarsenic speciespresent in the growing culture, enriched with MIvIA, after 10 days ofincubation. (a) in the absence of L-methionine-methyl-d3(b) in thepresence L-methionine-methyl-d3 964.11 Mass spectra of the hydride derivative of the trimethylarsenic speciespresent in the growing culture, enriched with MMA, after 10 days ofincubation. (a) in the absence of L-methionine-methyl-d3(b) in thepresence L-methionine-methyl-d3 974.12 HG-GC-MS chromatogram of a 2 mL culture medium sample that wastaken after 20 days of incubation. The medium was originally enriched with1 ppm DMA, but not with L-methionine-methyl-d3 994.13 Mass spectra of the hydride derivative of the trimethylarsenic speciespresent in the growing culture, enriched with DMA, after 20 days ofincubation. (a) in the absence of L-methionine-methyl-d3(b) in thepresence L-methionine-methyl-d3 1005.1 A schematic diagram of an on-line coupled flow injection-microwave ovendecomposition-hydride generation system. S: Sample flow, A: Acid flow, R:Decomposition reagent flow, B: NaBH4flow, V: Sample injection valve, P:Peristaltic pump, MO: Microwave oven, Ti & T2: T-joints, IB: Ice waterbath, G: Carrier gas (N2), D: To detector (AAS) 1075.2 The change of arsenic species in the growth medium with incubation time.The growth medium was enriched with 10 ppm arsenate before incubation.(0) arsenite (•) arsenate 1175.3 The change of arsenic species in the growth medium with incubation time.The growth medium was enriched with 0.9 ppm arsenate before incubation.(0) arsenite (•) arsenate (V) DMA 1185.4 The change of arsenic species in the growth medium with incubationtime. The growth medium was enriched with 0.9 ppm arsenite beforeincubation. (0) arsenite (•) arsenate (V) DMA 119XII’5.5 The change of arsenic species in media with incubation time after thetransfer of cells that had been previously grown in (a) 10 ppm arsenate (b)0.9 ppm arsenate (c) 10 ppm arsenite (d) 0.9 ppm arsenite (e) 10 ppmMMA (f) 0.9 ppm MMA 1215.6 HG-GC-MS chromatogram and mass spectra of arsine, monomethylarsine,and dimethylarsine. Solutions of standard arsenite, MMA, and DMA (100ng of arsenic for each compound) were used 1235.7 HG-GC-MS chromatogram of a 3 mL culture medium inoculated witharsenate (no methionine present), after 4 days of growth 1245.8 Mass spectrum of AsH3 produced from a 3 mL culture medium inoculatedwith arsenate (no methionine present), after 4 days of growth 1255.9 Mass spectrum of (CH3)2As produced from a 3 mL culture mediuminoculated with arsenate (no methionine present), after 4 days of growth 1265.10 Mass spectrum obtained from the culture medium inoculated with Lmethionine-methyl-d3and arsenate simultaneously, after 4 days of growth 1285.11 Wide scan and Selected Ion Chromatograms (SIC) obtained from theanalysis of culture medium inoculated with both L-methionine-d3andarsenate, a) Wide scan m/z 74-115 b) SIC of m/z 112 c) SIC of m/z 109d)SICofm/z93 1295.12 Mass spectrum obtained from culture medium inoculated with Lmethionine-methyl-d3and arsenate (4 days after the addition ofmethionine). The arsenate was added 3 days prior to the addition of Lmethionine-methyl-d 1305.13 Proposed model for biomethylation of arsenate in marine algae P.peniculus. a endocellular arsenicals b extracellular arsenicals 1386.1 HG-GC-MS chromatogram of stibine SbH3.A standard solution of Sb(Ill)(300 ng of antimony) was used 1456.2 Mass spectrum of stibine SbH3 derived from a standard solution of Sb(Ill) 1456.3 HG-GC-MS chromatogram of trimethylstibine Me3Sb. A standard solutionofMe3Sb(OH)2(200 ng of antimony) was used 1466.4 Mass spectrum of Peak A shown in figure 6.3 146xiv6.5 Mass spectrum of Peak B shown in figure6.3.1476.6 Mass spectrum of trimethyistibine Me3Sb derived from a standard solutionof e3Sb(OH)2 1476.7 HG-GC-MS chromatogram of a blank solution. Solutions of 4.0 M aceticacid and 2.0% (wlv) NaBH4were used for the reaction 1486.8 HG-GC-MS chromatogram of the hydrides derived from a solution ofstandard Me3Sb(OH)2(200 ng of antimony) exhibiting molecularrearrangement. The sample analysis was performed immediately after thehydride generation system was rinsed with distilled water 1506.9 Mass spectrum of the first antimony containing peak in Figure 6.8,identified as SbH3 1506.10 Mass spectrum of the second antimony containing peak in Figure 6.8,identified as MeSbH2 1516.11 Mass spectrum of the third antimony containing peak in Figure 6.8,identified as Me2SbH 1516.12 Mass spectrum of the fourth antimony containing peak in Figure 6.8,identified as Me3Sb 1526.13 HG-GC-MS chromatogram of a 5 mL extract of plant samples collected inKam Lake 1546.14 Mass spectrum of Me3Sb derived from a 5 mL extract of plant samplescollected in Kam Lake 1546.15 HG-GC-MS chromatogram of a 5 mL extract of plant samples collected inKeg Lake 1556.16 Mass spectrum of Peak 1 in Figure 6.15, identified as SbH3 1556.17 Mass spectrum of Peak 2 in Figure 6.15, identified as MeSbH2 1566.18 Mass spectrum of Peak 3 in Figure 6.15, identified as Me2SbH 1566.19 Mass spectrum of Peak4 in Figure 6.15, identified as Me3Sb 157xvA. 1 Schematic diagram of a batch type hydride generation-gas chromatography-atomic absorption spectrometiy 177A.2 Effect of the concentration of hydrochloric acid on the determination ofhydride-forming arsenicais. The NaBH4solution (4.0% w/v) was added tothe reaction system at a rate of 2 mlimin for 3 minutes. (0) arsenite (•)arsenate (V) MMA (V) DMA (0) TMAO 180A.3 Effect of the concentration of acetic acid on the determination of hydride-forming arsenicals. The NaBH4solution (4.0% w/v) was added to thereaction system at a rate of 2 mJJmin for 3 minutes. (0) arsenite (•)arsenate (V) MMA (Y) DMA (0) TMAO 181A.4 Effect of NaBH4concentration on the determination of hydride-formingarsenicals. A 3 mL of 4.0 M HC1 was added to the reaction system toacidify the sample. (0) arsenite (•) arsenate (V) MMA (V) DMA (0)TMAO 182A.5 Effect of NaBH4concentration on the determination of hydride-formingarsenicals. A 2 mL of 4.0 M acetic acid was added to the reaction system toacidify the sample. (0) arsenite (•) arsenate (V) MMA CV) DMA (0)TMAO 183A.6 Effect of microwave digestion time on the decomposition efficiency in thepresence of 2 mL of 2.0% (wlv)K2S08. The sample solution (50 mL)contained 50 ng each of AB (A), AC (A), and Me4As (0) 187A.7 Effect of the quantity of potassium persulfate (2.0% w/v) on decompositionefficiency. The sample solution (50 mL) contained 50 ng each of AB (A),AC (A), and Me4As (0) 188A.8 Effect of microwave digestion time on the decomposition efficiency in thepresence of 5 mL of 6.0% (w/v)K2S08.A 50 mL of seawater sample wasspiked with 50 ng each of AB (A), AC (A), and Me4As (0) 189A.9 Effect of the quantity of potassium persulfate (6.0% w/v) on decompositionefficiency. A 50 mL of seawater sample was spiked with 50 ng each of AB(A), AC (A), and Me4As (0), and irradiated in the microwave oven for5mm 190xv’LIST OF ABBREVIATIONSAA atomic absorption spectrometryAB arsenobetaineAC arsenocholineDMA dimethylarsinate, also dimethylarsinic acidGC gas chromatographyGC-MS gas chromatography-mass spectrometryGFAA graphite furnace atomic absorption spectrometryHGAA hydride generation atomic absorption spectrometryHG-GC-AA hydride generation-gas chromatography-atomic absorptionspectrometryHG-GC-MS hydride generation-gas chromatography-mass spectrometryi.d. inner diameterMMA methylarsonate, also methylarsonic acidMS mass spectrometrym/z mass to charge ratioNMR nuclear magnetic resonanceppb parts per billion, also ng mL4ppm parts per million, also pg mL1rpm revolutions per minuteR.S.D. relative standard deviationSAM S-adenosylmethionineSIM selective ion modeTMAO trimethylarsine oxideTLC thin layer chromatographyxviiUV ultravioletw/v weight per volumexv”ACKNOWLEDGMENTSI wish to express my sincere gratitude to my research supervisor Dr. W. R.Cullen for his guidance, encouragement, patience and fmancial support throughout thecourse of this project.I would also like to thank the members of my guidance committee, Dr. M.Blades, Dr. M. Gerry and Dr. A. Wade for their valuable suggestions.I am very grateful to Dr. M. Gerry, Dr. C. Harrington, Ms. I. Koch, Mr. J.Polson, and Mr. C. Simpson for reading the manuscript of this thesis.My thanks also go to Dr. L. Harrison for kindly providing the marine algaPolyphysa peniculus and for many valuable suggestions and discussions. I am verygrateful to Dr. G. Eigendorf for his support and help during the HG-GC-MSexperiments and also many valuable suggestions and discussions. I would like to thankMr. G. Hewitt for his support during microbial culture experiments and also manyvaluable discussions. I wish to thank Dr. M. Dodd for providing the trimethylantimonycompounds and the extracts of plant samples collected from Yellowknife.I wish to thank members of my research group for helpful discussions. Myspecial thanks go to Dr. S. Pergantis: it has been very rewarding for me to collaboratewith him.Finally, I would like to thank my parents for their encouragement during the pastfew years.XIXCHAPTER 1GENERAL INTRODUCTIONArsenic is an element that is widely distributed in the biosphere. However, it israrely found as the free element in nature, but occurs principally as the sulfides, realgar(As4S),orpiment (As2S3)and arsenopyrite (FeAsS) and is usually found in associationwith lead, zinc, copper, and gold-bearing minerals. The average crustal concentration ofarsenic is estimated to be between 1.5 and 3 ppm.”2Arsenic has also been found in theatmosphere, in the aquatic environment, in soils, and in living organisms.2Anthropogenic input and other factors, such as weathering, volcanic, and biologicalactivity, all contribute to the dispersion of arsenic.3Arsenic compounds are notorious for their toxicity, and arsenic trioxide, knownas “white arsenic”, has been used for criminal purposes more than any other poison.However, arsenic compounds can also have beneficial influences on human and animallife. A paste of realgar As4S was used for the treatment of ulcers in the time ofHippocrates (460-377 B.C.).4 The effectiveness of this tincture is unrecorded. Severalorganoarsenicals such as arsanilic acid, salvarsan, and neosalvarsan, have been used forthe treatment of human syphilis and sleeping sickness.4-7Aromatic arsonic acids, suchas 3-nitro-4-hydroxyphenylarsonic acid and arsanilic acid are still being used as foodsupplements for swine, turkeys and poultry. These compounds are believed to stimulategrowth, improve feed conversion, enhance feathering, and increase egg production andpigmentation,8 although their efficacy has never been scientifically substantiated.Arsenic compounds, such as arsenate, methylarsonate, and dimethylarsinic acid, havealso found use as the active ingredients in pesticides, herbicides, fungicides, harvest1aids, and wood preservatives.”91-Arsenic is a metalloid with a rich chemistry and can form a large number ofinorganic and organic compounds. Its similarity to phosphorus and its ability to formcovalent bonds with sulfur are the two main reasons for its toxicity.12 The arsenictoxicity scale proposed by Penrose’3follows the decreasing order: R3As (R= H, CH3Cl, etc.) > As203 > (RAsO)n > As2O > RnA5O(OH)3.n (n= 1,2) > >Although many arsenic compounds are considered to be non-toxic,2 the lingeringmemories of the criminal use of arsenic trioxide still causes the public mind to equate“arsenic” with the term “poison”. Until recently, the effects of arsenic were discussed interms of “total arsenic”. However, such blanket generalizations cannot be justified sincethis attitude ignores the beneficial uses of some arsenic compounds and the benignnature of others. It also prevents reasonable evaluation of their impact on theenvironment. The fact that the toxic effects of arsenic depend not only on itsconcentration, but also on its speciation has led to detailed investigations of the levelsand chemical structure of arsenic compounds in organisms.2 In recent years, arsenicspeciation in the environment has been extensively studied. As more attention is paid toarsenic speciation our understanding of the interactions of arsenicals with biologicalsystems will deepen.1.1 BIOLOGICAL TRANSFORMATION OF ARSENICThree major modes for the biotransformation of arsenic species have been foundto occur in the environment: (1) redox transformation between arsenate and arsenite; (2)the biomethylation of arsenic; (3) the biosynthesis of more complex organoarsenicals bymarine organisms.21.1.1 Redox transformation between arsenate and arseniteThe reduction of arsenate to the generally more toxic form of arsenite has beenfound to occur in the environment. Aquatic bacteria, activated sewage sludge, wineyeast, mixed bacterial cultures from seawater, cultures of mixed flora from the ratstomach bacteria, fresh water algae, and marine phytoplankton can all carry out thisreduction under either aerobic or anaerobic conditions. 14-22 In oxygenated seawatersarsenate is the major arsenic species, with the predominant dissolved form beingHAsO42.3’4However, arsenite has been found to comprise significant amounts ofthe total arsenic in marine waters, and are in higher quantities than thermodynamicequilibrium predicts from the redox pair arseniteIarsenate.226 The presence ofarsenite in seawater can result from the reduction of arsenate by marine phytoplankton,bacteria, and zooplankton.16’20227Blasco et al.18”9 also reported that Chiorellasp., a fresh water alga, can reduce arsenate to arsenite. The reduction of arsenate toarsenite has also been described in higher plants, and animals, including man.5The oxidation of arsenite to arsenate is considered to be one of the mechanismsby which an organism protects itself against undesirable species. This process has beendescribed for several bacterial strains isolated from cattle dipping fluids, sewage andsoil.21.1.2 The biomethyladon of arsenicThe biological methylation of arsenic is a common phenomenon in nature. Eventhough a variety of microorganisms and higher organisms including some plants, mice,monkeys and man are known to methylate inorganic arsenic compounds, the mechanismof methylation is still not fully understood.23Cases of arsenic poisoning were reported to have been caused by the evolution oftoxic gas from wallpaper containing such arsenic compounds as Scheele’s green (cupricarsenite) and Schweinfurt green (copper acetoarsenite, Paris green). The first systematicstudy of this toxic gas was carried out in 1891 by Gosio who identified a number ofmolds that could grow on wallpaper colored with arsenite-containing pigments.28 Thepure cultures of these molds such as Scopulariopsis brevicaulis and Monilia sitophila,produced a volatile, toxic, garlic smelling arsenic species from added arsenate whengrown on a potato medium.28 Biginelli incorrectly identified this gas as (C2H5)AsHbased on an analysis of the precipitate obtained from passage of this gas through anacidified (HC1) mercuric chloride solution.29 The correct characterization of thisvolatile gas as trimethylarsine (CH3)As was not achieved until 1933 by Challenger andhis coworkers.3°They grew the mold Scopulariopsis brevicaulis on sterile breadcrumbs treated separately with As203,methylarsonic acid (MMA)CH3AsO(OH)2,anddimethylarsinic acid (DMA) (CHAsO(OH). The volatile arsine product wascharacterized as the mercuric chloride adduct ((CH3As.2HgCl2),which precipitatedout when the arsine gas stream was passed through a solution of HgCl2 in hydrochloricacid (Biginelli’s solution29). The methylarsines were also detected ashydroxytrialkylarsonium nitrates or picrates, or as benzyltrialkylarsonium picrates whenthey were passed through nitric acid, or alcoholic benzyl chloride.30 Since then, severalfungi isolated from soil and sewage have been found to be active in producingtrimethylarsine in the presence of arsenate, arsenite, MMA, and DMA.335 Thesefungi show selective methylation depending on the fungal species, the pH of the growthmedium, the species of arsenic substrates added to the growth medium, and theconcentration of phosphate in the growth medium. Bird et al.34 reported that (CH3)Aswas produced from MMA, and DMA but not from arsenite by the fungi Aerobacter4niger, Penicillium notatum, and Penicillium chrysoganum. Cox and Alexander32madethe first “modern” identification of (CH3)As produced by three fungi Apiotrichumhumicola (originally known as Candida humicola), Gliocladium roseum, and a speciesof Penicillium when arsenate, arsenite, MMA, and DMA were added to the growingculture of these microorganisms. They identified the volatile arsine by using gaschromatography-mass spectrometry. Cullen et aL33 studied Apiotrichum humicolaextensively, and used gas chromatographic techniques to monitor trimethylarsineproduction. They also employed a liquid oxygen trap to cryofocus the trace amounts ofvolatile arsines prior to mass spectrometric identification.33 Other alkylarsonic acidsRAsO(OH)2(R = CH32-,CH32-,CH2=CHCH-) and diallcylarsinic acidsR’R”AsO(OH) (R’ = CH32-,R” = CH32-)can also be metabolized by fungito produce RAs(CH3)2 and R’R”As(CH).31335 Using a simple method ofchemofocusing and mass spectrometry, Cullen et al.36 have found that Apiotrichumhumicola methylates or reduces PhAsO(OH)2,Ph(CH)AsO(O ) and Ph(CH3)2AsO toPh(CH3)2As. The reduction of trimethylarsine oxide (TMAO) (CH3AsO to(CH3)As by A. humicola has also been studied in detail by using gas chromatographytechniques.37 This reduction is rapid and requires biologically intact cells. The pH inthe growth medium, the growth temperature, electron transport inhibitors, anduncouplers of oxidative phosphorylation such as azide and oligomycin, can all affect thereduction rate of TMAO.37The first well-documented report of an arsine being produced by bacteria wasmade in 1971.38 A Methanobacterium strain, MoH, growing in anaerobic ecosystems,such as sewage sludge, or freshwater sediment, was found to produce a volatile amine(probably (CH3)2As ) from arsenate. Since then, bacterial methylation of inorganicarsenic has been studied extensively. A number of nonmethanogenic bacteria are known5to synthesize CH3As2,(CH3)2As , and (CH3)As from arsenate.39 A study byCheng and Focht39 showed that AsH3,CH3As2,and (CH3)2A were formed bysoils treated separately with arsenate, arsenite, MMA, and DMA. They could not fmdtrimethylarsine in the headspace gas. The soil bacteria Pseudomonas and Alcaligeneswere shown to be responsible for the reduction. When TMAO is added to fresh water,sewage sludge, and sea sediments, (CH3)As can be detected and is believed to arisefrom bacterial action.42The production of volatile arsines by algae in the presence of arsenate, arsenite,methylarsonate, and dimethylarsinate has not been reported.Challenger30’43 proposed a pathway for the biomethylation of arsenicals bymicroorganisms, involving alternating oxidation and reduction steps (Figure 1.1).Support for this sequence comes from the observation that arsenate, arsenite, MMA,DMA, and TMAO are all substrates for the production of trimethylarsine by fungi.303537 In studies connected with Challenger’s mechanism, it should be noted that thespecies {CH3As(O )2}and {(CH3)2As(OH)} probably do not exist as such; thesearsenic(III) compounds are better represented as (CH3AsO) and ((CH3)2As)0in anoxic environment.44 Cullen et al.45 incubated the organoarsenic(ffl) derivatives(CH3AsO) and (CH3ASS)n with active cultures of Apiotrichum humicola,Scopulariopsis brevicaulis, and other organisms such as Veillonella alcalescens. Bothcompounds are metabolized by A. humicola and S. brevicaulis to (CH3)As and(CH3)2As . Moreover, (CH3)2AsO(OH) was found in the culture medium incubatedwith (CH3AsO).45 This was the first time that a methylated arsenic metabolite hadbeen isolated from the culture medium, a result which could be predicted from aconsideration of Figure 1.1.6H3As’O4 2e As111(OH)3 Me MeAsVO(OH)zMeAs’O(OH)2 2e {MeAsm(OH)2) Me Me2As”O(Ou)Me2AsVO(OH) 2e {Me2AsW(OH)) Me4 Me3As’OMe3As’O 2e Me3AsFigure 1.1 Challenger’s mechanism for the biomethylation of arsemc.30’43 Theintemiediates in {) are unknown as monomeric species. They can be isolated as(CH3AsO) and (CH3As)20,respectively, when prepared by conventional methods.In Challenger’s mechanism, each methylation step is preceded by a reductionstep. This two-electron reduction can be carried out in vitro by using lipoic acid and arange of thiols and dithiols, including cysteine, glutathione, mercaptoethanol, anddithiothreitol.46’7It is possible these are the reducing agents for the biological process.A possible mechanism for this reaction is shown in Figure 1.2.MeAsO(OH).x ÷ 2RSH MeAs(SR)2(OH)3.X + H20Me1As(SR)z(OH)a.x - MeAs(OH)3.X + RSSRMexAs(OH)3..x + (3-x)RSH MeAs(SR)3.Figure 1.2 The mechanism for arsenical reduction by thiols.46’77The methyl groups added to the arsenic atom during metabolism come from amethyl donor. In his 1945 review,30 Challenger favored the hypothesis that themethylation of arsenic involved the transfer of a methyl group from some methylcontaining compounds such as betaine, methionine, or a choline derivative. Toinvestigate the possible methyl donor, Challenger et aL48 added 14C labeled methionine(14CH3SCH2(N)H OOH) to the culture medium and was able to detect 14(labeled trimethylarsine. Based on these results it was proposed that “active methionine”,later characterized as S-adenosylmethionine (SAM),49 is possibly involved in thetransfer of the methionine methyl group to arsenic during mycological methylation.48The structure of SAM is given in Figure 1.3.NCH3HOOC-CH-CH2-CS- HI +NH2Figure 1.3 Structure of S-adenosylmethionine (SAM).49 The arrow indicates themethyl group that is used for methylation.cHcH8Cullen et al.33 grew cultures of Apiotrichum humicola and Scopulariopsisbrevicaulis aerobically in the presence of L-methionine-methyl-d3 and found that CD3was transferred intact to arsenite, arsenate, MMA, and DMA. The presence of the labelwas established by using mass spectrometry, and a high incorporation of the CD3 group,80-90%, was found in the isolated trimethylarsine. For example, the arsine producedfrom As203 consisted of 83% (CD3)As, 13% (CD)2AsCH,3% CDAs(CH)2and1% (CH3)As. These results reinforce the suggestion that S-adenosylmethionine orrelated suiphonium compounds are the source of the [CH3]+ shown in Figure 1.1. Inorder to develop these ideas further, broken-cell homogenates of A. humicola wereincubated with arsenate, arsenite, MMA, DMA, SAM, and NADPH.51 The methylatedarsenic metabolites in the supernatant were confirmed by using a combination ofelectrophoresis analysis and column chromatography. Yet it is still unclear if SAM is theonly in vivo alkylating source for fungi or whether other methylated sulfur compoundswill cause the same reaction to occur.4For bacteria, the source of the methyl donor in the arsenic biomethylationprocess is still not well understood.2Cell-free extracts of the Methanobacterium strainMoH produced dimethylarsine when incubated anaerobically with arsenate, CH3-B12(methylcobalamin, a derivative of vitamin B 12) H2, and ATP.38 CH3-B12 was addedbecause at that time CH3-B12,usually a donor of the carbanion CH3, was believed tobe involved in methane production as well as in the methylation of mercury52.The truemethane precursor was later identified as HSCH2CHSO3-(coenzyme-M, HSC0M).38’534 Although CH3-B12 served as a precursor in the production ofdimethylarsine, the oxidative addition of CH3 to arsenic(ffl) is unlikely.33 Cullen etal.2 suggested three possibilities for the role of CH3-B12 in the arsenic biomethylationprocess: (a) CH3S-CoM is produced from CH3-B12,and this becomes involved with the9methylation of arsenic; (b) CH3-B12 reacts with the arsenicals in a purely chemical orbiological process; (c) CH3-B12 provides the methyl group to a component in the cellextract, e.g., methionine, which is the ultimate methyl donor to the arsenic. So far, theavailable data do not allow any distinctions to be made among these possibilities, andwithout any strong evidence, there seems little need to invoke a different mechanism forarsenic methylation by bacteria from that discussed above for fungi.2 The only majordifference between fungi and bacteria seems to be that reduction of methylarsenic(V)species to arsines, (CH3)nAsH3n (n=O-2), is a more common response by bacteria.2The possible source of the methyl donor for arsenic methylation by algae has notbeen investigated previously, although it has been speculated that SAM is the primarymethyl donor in the production of the MMA and DMA detected in marine algae.91.1.3 Organoarsemc compounds in the marine environment1.1.3.1 Arsenic compounds in marine algaeAlgae, located at the bottom of the aquatic food web, have often been the subjectof arsenic metabolism studies because of their ecological and nutritional importance.Through bioaccumulation, algae exhibit concentrations of arsenic that are much higherthan those of the surrounding water.5557 Arsenic concentrations in a range of marinealgae are shown in Table 1.1.5864 Studies of the interaction of marine algae witharsenicals are relevant because arsenic compounds produced by algae are generallybelieved to be the source of the arsenic compounds found in marine animals, although itis not well established how and when these transformations take place.10Table 1.1 Arsenic concentrations (jiglg) in some marine algaeArsenic concentration (igfg)Type of algae (dry weight)(No. of species) Location Range Mean ReferencesBrown (8) India 8-68 30 58Red(5) 0-5.0 1.5Green (5) 0.1-6.3 2.2Brown (7) Norway 15-109 44 59Red(2) 10-13 12Brown (3) UK 26-47 39 60Red(2) 11-39 25Brown (15) Japan <1-230 46 61Red (18) <1-12 4.4Green (3) <1-8 3.8Brown (24) USA 1.06-31.6 10.30 62Red (15) 0.43-3.16 1.43Green (16) 0.17-23.3 1.54Brown (14) Canada 40.8-92.4 57.0 63Brown (14) Australia 21.3-179 62.0 64Red (10) 12.5-3 1.3 19.2Green (9) 6.3-16.3 10.711In seawater, arsenate is the predominant arsenic species and is present atapproximately 1.0-2.0 ppb.’20325 However, significant amounts of arsenite,monomethylarsenicals, probably as MMA, and dimethylarsenicals, probably as DMA,have also been observed and are believed to be a consequence of the biological activityof marine algae. It is believed that arsenate, being chemically similar to phosphate, isreadily taken up by algae from the water via phosphate transport systems located in thealgal cell membranes.65-7 In highly productive environments, oceanic phosphatelevels, reduced at the surface by phytoplankton and bacterial consumption, approachthose of arsenate, and are sometimes even less than those of arsenate.24’68This createsa favorable environment for algae to take up arsenate. Once inside the algal cell,arsenate can remain unaltered or it can be reduced and transformed to a variety oforganic arsenic compounds. The algal reduction of arsenate, as well as its methylationleading to various water- and lipid-soluble compounds, is considered as a detoxificationprocess.698°Most of the arsenic in marine macroalgae exists in complex forms including avariety of arsenosugar derivatives which have been isolated and characterized. The firstsuccessful isolation of arsenosugars was achieved from the brown kelp Ecklonia radiataby Edmonds and Francesconi.8L2The arsenicals were extracted by using methanol,subsequently isolated by using column purification and preparative TLC, and fmallycharacterized by using microanalysis and spectroscopic techniques such as 1H- and 13C-NMR spectrometry. These methodologies provided convincing evidence for thesuggested structures of two arsenic containing ribofuranosides, 4a and 4b (Figure 1.4).These two compounds accounted for 81% of the total arsenic present in this brownkelp.81’2 Since then a variety of arsenosugars and their derivatives has been reportedto exist in macroalgae.8389 Methanol extracts of the edible brown seaweed Laminaria12japonica showed the existence of three arsenic sugar derivatives, namely 4b, 4c, 4d.83Another edible seaweed, Hizikia fusiforme, which belongs to the fucales, containsessentially half of its total arsenic as sugar derivatives, as 4a, 4c, 44 and 4e, and half asarsenate.84Arsenosugars were also isolated from the kidney of the giant clam, Tridacnanuzima, in western Australia;9°these sugars were proposed to be the metabolicproducts of symbiotic, unicellular green algae living in the clam.OjS_C!270N /0 CH2— CH—CH2—R’CH3/%H__C\cHHR4a -OH -OH4b -OH -SO3H4c -OH -OSO3H4d NH2 -SO3H04e -OH- O--O-CH2rOH CHOHCH2OFigure 1.4 Structure of arsenosugar derivatives.13Edmonds and Francesconi80have proposed a pathway for the biotransformationof arsenate by marine algae (Figure 1.5). This pathway initially follows the mechanismoutlined by Challenger (CH3donation indicated by A in figure 1.5).30,43 However, inthe fmal steps the adenosyl group (indicated by B in Figure 1.5) of the methylatingagent SAM is transferred to the arsenic atom of dimethylarsinate, ultimately to formarsenosugars and arsenolipids. Enzymatic hydrolysis of the resulting intermediate (7)would lead to (8) which could form the arsenosugars (4a-e) by reaction with availablealgal metabolites.8°Recently, the proposed key intermediate (7), an arsenic containingnucleoside, has been isolated from the kidney of giant clam, Tridacna maxima byEdmonds and Francesconi.91Although arsenosugars have been isolated from seaweeds, there are only alimited number of reports that describe the biotransformation of arsenicals bymacroalgae grown in culture media. The marine macroalgae Fucus spiralis (L) andAscophyllum nodosum (L) assimilate arsenate to produce both water-soluble and lipid-soluble organoarsenicals,92’3although these compounds were not positively identifiedas arsenosugars. Sanders and Windom2°used arsenate, arsenite, and DMA as substratesfor cultures of a marine macroalga Valonia macrophysa. Digestion of the cells in dilutenitric acid led to an increase in methylated arsenicals, suggesting that more complexarsenic compounds were produced.Studies on arsenic biotransformation in marine phytoplankton in cell cultureexperiments have demonstrated the ability of some marine phytoplankton to producealkylated arsenic compounds and more complex water-soluble and lipid-solubleorganoarsenicals in the presence of inorganic arsenic.20’6979 However, there has notbeen any strong evidence for the production of arsenosugars in marine phytoplankton.Investigations of arsenate uptake by marine phytoplankton have established that the14H3As”04 2 e As1(OH)3 A MeAs’O(OH)2 2 e(1) (2) (3)(MeAsT11OH)2) A Me2As’O(OH) 2e (Me2As”kOH)) B(5) (6)NHMe2s_ç01(7) (8) (9)Me2AS— AArsenolipidsOH OH OH OH(10) (11)IflAi /HaNyO%N’ S-adenosylmethionineCW‘OHOH BI.Figur 1.5 Proposed mechanism for transformation of arsenic compounds in marinealgae.0 Unidentified compounds are underlined. The A represents Me+ from SAM, andB is the adenosyl group from SAM.15arsenic is distributed between the CH3OHICHC1 extractable fraction and insolublecomponents of the cells, although no individual arsenic compound has been positivelyidentified apart from arsenite, MMA, and DMA. Lunde75 reported that the marineunicellular algae Chiorella ovalis, Phaedaciylum tn cornutum, and Skeletonemacostatum incorporated[74As]arsenate and[74As]arsenite from seawater into variouswater-soluble and lipid-type fractions. Arsenate was the preferred substrate for thebiosynthesis of organoarsenic compounds by these algae. Irgolic et al.74 grew 12species of marine unicellular algae at arsenate levels ranging from 500 to 50,000 ppb (inarsenic), but were not able to characterize the lipid-type arsenicals that were produced.Andreae et al.7° showed that four classes of marine phytoplankton: the diatoms,coccolithophorids, dinoflagellates, and green algae (Prasinophyceae) can transformarsenate to arsenite and subsequently to MMA and DMA. A portion of the arsenate wasretained within the cell. An increase in methylated arsenicals was observed after basedigestion of the aqueous extracts, suggesting the presence of more complex organiccompounds. Cooney et al.69 incubated the marine diatom Chaetoceros concavilorniswith[74Asjarsenate, and found that a large amount of complex water-soluble arseniccompounds (62% of the total arsenic) and lipid-soluble arsenicals (33% of the totalarsenic) could be extracted with water or CHC13. This water-soluble compound wasinitially characterized as trimethylarsoniolactate, but the author retracted theidentification later in favor of an arsenosugar.78 The complexity of arsenicbiotransformation in marine phytoplankton has been revealed in studies of arsenateuptake by the unicellular alga Dunaliella tertiolecta.72’9Wrench and Addison72foundthat three arsenolipids, which are not related to arsenosugars, were produced when D.tertiolecta was treated with 0.2 MBq of[74As]arsenate for 45 mm. They suggested thatone of these is a complex between arsenite and phosphatidyl inositol, the second a16neutral or zwitterionic complex between arsenite and a glycolipid, and the third anunidentified phospholipid-like arsenical. However, other workers79 have reported thatabout 47% of the arsenic in the same alga is present as a phospholipid (0-phosphatidyltrimethylarsoniumlactate, later recharacterized as an arsenosugar derivative80)and as anunknown lipid (48% of the total arsenic in cells) following exposure for 48 hours to[74As]arsenate.1.1.3.2 Organoarsemc compounds in marine invertebrates and fishIt has been known for many years that the concentration of arsenic species inmarine and freshwater animals is considerably higher than the backgroundconcentrations in the surrounding water.2 The “fish arsenic” found in marine animals ischemically and physiologically different from arsenate and arsenite (Figure l.6).9O0Evidence indicates that a large proportion of this arsenic’ is present as organoarseniccompounds which are non-toxic.18’209700It was not until 1977 that arsenobetaine(6a) was isolated from the rock lobster Palinurus cygnus.10U02 Since thenarsenobetaine has been shown to be the most abundant arsenical in most marine animalsso far investigated, including lobster, fish, clam, crab, mussels and scallops. 103108 Inmost of the studies, organoarsenicals in marine animals were isolated and identifiedfollowing solvent extraction, ion-exchange chromatography, HPLC, and TLC. Otherorganoarsenicals have also been found in marine animals. The tetramethylarsonium ion(6b) has been found in the clam Meretrix lusoria along with arsenobetaine and smallamounts of two unidentified arsenicals.’°9 Similar results are observed for otherbivalves, clams, mussels and scaiiopsJO8J-0Arsenocholine (6c) has also beenclaimed to be found in shrimps,112-114 but this result could not be confirmed by17others.106’15 TMAO (6d) has been found as a minor component in a number of fishspecies and in clams.108 It has been suggested that (CH3)AsO is the breakdownproduct of some unidentified arsenicals present in the fish since its concentration infrozen perch is much higher than in fresh perch.116(CH3)AsCH2COO (CH3)4As6a 6b(CH3)AsCH2CO (CH3)AsO6c 6dFigure 1.6 Structure of organoarsenicals isolated from marine invertebrates and fish.It is important to establish at what stage in the food chain the interconversion ofthe organoarsenicals takes place. This can be achieved only by closely controlledexperiments. Limited experimentation has indicated that marine animals acquire theirarsenic burdens through the food chain rather than directly from ambient water.80’91The organoarsenicals found in the marine organisms are believed to result from theaccumulation of compounds that have been synthesized from arsenate at low trophiclevels.2Edmonds eta!.80’91117 suggested that arsenosugars are likely to be convertedto arsenobetaine within the food chain. The facile transformation of arsenosugarspresent in the brown algae Ecklonia radiata into dimethyloxarsylethanol under18anaerobic conditions supports this view.’17 Two routes are possible to producearsenobetaine via dimethyloxarsylethanol as shown in Figure 1.7. Such processes aremost likely to occur (probably microbially mediated) in marine sediments.8°However,it is not obvious where, or if, any of these transformations would occur in a naturalecosystem.Recently Shibata and Morita118 isolated a trimethylarsenoriboside (compoundii in Figure 1.5) from the marine alga Sargassum thunbergii. Anaerobic degradation ofthis compound could directly produce arsenocholine which would be readily convertedinto arsenobetaine in fish or elsewhere. Cullen and Nelson119 reported the probableexistence of arsenobetaine in seawater.0Me2As o OR Anaerobic0decomposition (CH3)2AsCHHO °‘°° (CH3)2AsCHCO0OH OHMethylation MeEhylation(CH3)AsCH2HO °,uthi (CH3)AsCH2COO6c 6aFigure 1.7 Possible pathways for the production of arsenobetaine from arsenosugarderivatives.191.2 SCOPE OF WORKThis thesis is concerned with the interaction of arsenic with the microorganismsApiotrichum humicola and Scopulariopsis brevicaulis as well as a marine unicellularalga Polyphysa peniculus. The effect of several arsenicals on cell cultures as well as themechanism of biomethylation of arsenicals has been examined.A continuous hydride generation-gas chromatography-atomic absorptionanalytical system is described in Chapter 2. This system was optimized to determinearsenate, arsenite, MMA, DMA, and TMAO in the cell extracts and in the growthmedium of the microorganisms Apiotrichum humicola and Scopulariopsis brevicaulisand the marine alga Polyphysa peniculus. Graphite furnace atomic absorptionspectrometry was also optimized for the determination of total arsenic in the samples.The biotransformation of arsenate, arsenite, methylarsonate, anddimethylarsinate by the microorganisms Apiotrichum humicola and Scopzdariopsisbrevicaulis is described in Chapter 3. The separation, identification, and quantificationof extracellular arsenical metabolites in the growth medium was carried out by using thehydride generation-gas chromatography-atomic absorption spectrometry systemdescribed in Chapter 2.In Chapter 4, a hydride generation-gas chromatography-mass spectrometrytechnique is described. This system was used to characterize the extracellular arsenicaLsin the media of microorganisms grown in the presence of arsenate, arsenite,methylarsonate, dimethylarsinic acid, and L-methionine-methyl-d3.The nature of themethyl donor in the biomethylation of arsenic compounds was investigated.Chapter 5 reports on the effect of adding arsenate, arsenite, methylarsonate anddimethylarsinic acid to an unicellular marine alga Polyphysa peniculus. The arsenicaccumulation, methylation and excretion by the alga was investigated. In the presence of20L-methionine-methyl-d3and arsenate, the arsenic metabolites excreted by the alga in thegrowth medium were characterized by using hydride generation-gas chromatography-mass spectrometiy.In Chapter 6, a continuous type hydride generation system coupled with GC-MSis described. This system was used to identify the antimony hydrides produced from thetrimethylantimony compounds Me3Sb(OH)2and Me3SbC12.The possible causes of themolecular rearrangement of the trimethylstibine were investigated. The HG-GC-MSsystem was also used to analyze the extracts of plant samples collected fromYellowknife.21CHAPTER 2ANALYTICAL METHODOLOGY2.1 INTRODUCTIONSince the introduction of atomic absorption spectrometry, applications of thisanalytical technique have become routine in laboratories through the world.120-’22Grapjiite furnace atomic absorption spectrometry (GFAA) has become one of the mostwidely used analytical methods because of its high sensitivity and applicability forsmaller sample volumes. It has been used for total arsenic determination in a variety ofsamples such as drinking water, seawater and brines, biological fluids, soils, industrialproducts, and wastes. 120-126 Hydride generators coupled to atomic absorptionspectrometers have also found wide application in trace analysis because of the 10-1000fold increase in sensitivity compared to other liquid sample nebulizationprocedures.127431 Hydride generators coupled to a range of detectors have been usedto determine the total amounts of arsenic compounds that can be transformed intohydride-forming species through acid digestion102’108,132 136 or basedigestion.’13,137-139 The speciation analysis of arsenate, arsenite, MMA, DMA andTMAO that are precursors to the volatile arsines AsH3,CH3As2,(CH3)2As and(CH3)As, can be achieved by using a hydride generator coupled to a gaschromatograph, with an atomic absorption spectrometer as the detector.1391 Themajor components of such a hydride generation system are shown in Figure 2.1.In the present work, optimum conditions for determination of arsenic compoundsby using GFAA were established. In addition, continuous hydride generation-gaschromatography-atomic absorption spectrometry methodology was developed andoptimized for applications to the work outhned in the introduction (Chapter 1).22SAMPLE PREPARATION_________________HYDRIDEFiLTRATIONDIGESTION GENERATION__________________________1DETECTOR HYDRIDE SEPARATOR HYDRIDE COLLECTOR4AA / MS OC COOLED TRAPFigure 2.1 Scheme of the major components of a hydride generation system.2.2 EXPERIMENTAL2.2.1 Chemicals and reagentsAll chemicals used were of reagent grade. Deionized water was used for dilution.Glass and plasticware were cleaned by soaking overnight in 2.0% Extran solution,followed by a water rinse, a soak in dilute hydrochloric acid, and finally a deionizedwater rinse.Arsenic standards were freshly prepared each day by serial dilution from stocksolutions (1000 ppm of elemental arsenic) of the following compounds: sodiumarsenate, Na2HAsO4.7H (Baker); sodium arsenite, NaAsO2 (Baker); disodiummethylarsonate, CH3AsONa.60(Alfa); dimethylarsinic acid, (CH32AsO(OH)(Alfa); trimethylarsine oxide, (CH3)AsO, which was synthesized according to the23literature.’45 Solutions of 2.0% (wlv) NaBH4 prepared in 0.1% (w/v) NaOH, 1.0 MHC1, and 4.0 M acetic acid were freshly made daily.2.2.2 Apparatus2.2.2.1 Graphite furnace atomic absorption spectrometry (GFAA)A Varian Techtron Model AA 1275 Atomic Absorption Spectrometer was usedfor the arsenic determination. It was equipped with a Varian Spectra AA hollow cathodelamp set at 8 mA, a deuterium background corrector, and was connected to a Hewlett-Packard 82905A printer. The monochromator was set at 193.7 nm, and the slit width at1 nm. A Varian Techtron GTA-95 atomizer accessory fitted with a pyrolytically coatedgraphite tube was used to achieve graphite furnace atomization. Argon was used as thepurge gas.2.2.2.2 Hydride generation-gas chromatography-atomic absorption spectrometry(HG-GC-AA)The hydride generation system used in most of the work is shown in Figure 2.2.A Gilson Minipuls 2 four-channel peristaltic pump was used to deliver and mix sample,acid, and sodium borohydride solution. The sample solution first met with a continuousflow of acid and then borohydride. Hydride generation took place in an 18-turn reactioncoil, producing both arsines and hydrogen. The gases and spent solutions were separatedin a gas-liquid separator.24PeristakicPumpReactionCoilHeliumSampleAcidNaBH4Gas/LiquidSeparator [fiLlDry Ice/AcetoneWalerTrapII1. ArsineTrapLiquidNitrogen2. ReleasingArsinesWater bathVentHeliumInFigure2.2Hydridegeneration-gaschromatography-atomicabsorptionspectrometryexperimentalsetup.Dottedlinesrepresentgas lines.I’NaAtomicAbsorptionSpecLromerGas ChromatographThe effluent gases were passed through two U-tube traps connected in series.The first U-tube was a moisture trap made of Teflon (30 cm x 0.8 cm i.d.) which wascooled by a dry ice-acetone slurry (-78°C). The second trap (Teflon, 30 cm x 0.4 cmi.d.) was immersed in liquid nitrogen (-196°C) and was used to collect the volatilearsines. A six-way valve was interfaced between the arsine trap and a gaschromatograph. The valve could be switched from the stripping gas stream of thegenerating apparatus, to the carrier stream of the gas chromatograph. The interior partsof the six-way valve were made of stainless steel.A 5830A Hewlett Packard Gas Chromatograph was used to separate theproduced volatile arsines. The GC column consisted of a Teflon tube (50 cm x 0.4 cmi.d.) hand packed with Porapak-PS (80-100 mesh, Chromatographic Specialties).Atomic absorption measurements were performed by using a Jarrell-Ash Model810 atomic absorption spectrometer equipped with a Varian Spectra arsenic hollow-cathode lamp set at 10 mA. A hydrogen-air flame in a quartz cuvette mounted on thespectrometer was used to atomize the volatile arsines. Light from the arsenic hollowcathode lamp was aligned to pass through the hydrogen/air flame. The monochromatorwas set at 193.7 nm, and the slit width at 1 nm. A Hewlett-Packard 3390A integratorwas connected to the AA spectrometer to record the absorbance values as peak areas.2.2.3 Analytical procedures2.2.3.1 GFAAOperating parameters such as temperature, time, and gas flow need to beoptimized in order to achieve good analytical sensitivity in GFAA analysis. These26parameters have been optimized by other workers in this research group for the purposeof determination of arsenic in standard solutions and in the extracts of plant cells.Therefore, only minor changes in the furnace operating parameters were made when theacid digestates of the algal cells were analyzed.The GTA-95 accessory was used to program the furnace operating parameterssuch as temperature and time for drying, asbing, and atomization stages. In theoptimization process, the arsenic standard solution (volume 20 liL) was injected into thegraphite furnace by using the automatic delivery system of the GTA 95 accessory.Palladium solution (20 iL), prepared as palladium nitrate (100 ppm) in citric acid (2.0%w/v), was used as a modifier. By using the single parameter variation method,temperature and heating time were optimized to give the maximum absorbance signal(peak area) for arsenic in a standard solution. These optimized parameters are shown inTable 2.1.Table 2.1 Furnace operating parameters for the determination of arsenicStep Temperature Time Gas Flow Comment# (°C) (sec) (L mm4)1 70 5 3.0 dry2 120 30 3.0 dry3 1200 20 3.0 ash4 1200 1.0 0 ash5 2300 1.0 0 atomize6 2300 1.0 0 atomize7 2300 2.0 3.0 clean27The standard additions technique159 was used to compensate for matrixinterferences when the cell digestates were analyzed. The sample (volume 10 ilL) andthe standard solution (volume 3-15 liL) were mixed with palladium modifier by usingthe automatic delivery system of the GTA 95 accessory. The mixture was then injectedinto the graphite tube before the furnace was heated. The parameters established for thestandard additions are given in Table 2.2.Table 2.2 Typical sampling parameters for Standard Additions Method for GFAAanalysisSolution Standard Sample Blank ModifierVolume Volume Volume Volume(IlL) (ilL) (jiL) (jiL)Blank --- --- 25 20Addition 1 3 10 12 20Addition 2 6 10 9 20Addition 3 9 10 6 20Addition 4 12 10 3 20Addition 5 15 10 0 20Sample --- 10 15 202.2.3.2 HG-GC-AAWhen HG-GC-AA analysis was performed, the sample solution, 1.0 Mhydrochloric acid or 4.0 M acetic acid, and 2.0% (wlv) sodium borohydride solutionwere pumped from their respective containers by using a peristaltic pump and mixedcontinuously in an 18-turn reaction coil (— 2.5 m) with the help of a carrier gas (helium)28and the evolved hydrogen. The solution introduction rate was 3 mL!min. After thesample solution (typically between 1-5 mL) was taken up, the probe in the sample cupwas transferred into a wash solution of deionized water for another 2 minutes to cleanout the apparatus. The mixed reagents were carried into the gasfliquid separationapparatus where the waste solution constantly drained out.After trapping the volatile arsines in the U-tube immersed in liquid nitrogen, theliquid nitrogen bath was removed and the U-tube was immersed in a 700C water bath.The gas-sampling valve was switched, and the arsines were carried by a stream ofhelium gas to the GC column, where they were separated by means of carefultemperature programming (Table 2.3). The effluents from the GC were then carried tothe atomic absorption spectrometer, where they were combusted in a hydrogen/air flamein a quartz cuvette, and quantified. The experimental conditions are summarized inTable 2.3.2.3 RESULTS AND DISCUSSION2.3.1 Graphite furnace atomic absorption spectrometry (GFAA)The GFAA technique permits the direct analysis of liquid samples. A measuredvolume of the sample solution (5-70 pL) is placed in the furnace and electrically heatedin a series of stages so that the analyte element is freed from as many of the impuritiesas possible before it is atomized. 146 The first stage is the drying step where the solventis evaporated from the sample solution by holding the temperature of the furnaceslightly above the boiling point of the solvent. Arsenic is not likely to be lost during thiscycle unless arsenic trichloride is present. 147 The formation of arsenic trichioride is29Table 2.3 Operating conditions for the HG-GC-AA SystemAA spectrometer 810 Jarrell-AshWavelength 193.7 rimSlit Width 1 rimLamp Current 10 mAFlame Air/HydrogenAir Flow Rate 120 mL min1Hydrogen Flow Rate 75 mL mink’Carrier Gas HeliumCarrier Gas Flow Rate 30 mL min1(reaction coil)Carrier Gas Flow Rate (GC) 20 mL min1Size of Quartz Burner 10 cm x 0.8 cm i.d.Trapping Time 3 mmSample Size 1-5mLPump Rate 3 mL mm4GC Temperature Program Temperature 1 50°CTimel 0Temp. Rate 30°C min1Temperature 2 150°CTime2 2min30favored by a high-chloride medium. This occurs particularly when a hydrochloric acidsolution is present. In the second ashing stage, the temperature is raised to a value that issufficient to remove all organic materials by thermal decomposition as gaseoussubstances, but not sufficient to vaporize the arsenicals. This cycle was carried out attemperatures ranging from 120°C to 12000C. The fmal step is the atomization stage, inwhich the temperature is rapidly increased to a point at which gaseous atoms of thearsenicals are formed. The absorption of the analyte is measured during this stage. Astream of argon gas is usually used to protect the furnace and the analyte fromoxidation, and to transport impurities vaporized during thermal pretreatment steps out ofthe furnace.Samples were analyzed for arsenic by using methods involving either simplecalibration curves or standard additions. Typically, calibration curves were obtaineddaily by plotting absorbance (peak area) against concentration for 20 iL injections of aseries of standard arsenic solutions. A linear relationship was obtained for arsenicconcentrations up to 200 ng/mL. The relative standard deviation for twenty injections of20 EiL of 20 nglmL standard arsenic solution was calculated to be 6.9%. The limit ofdetection is defined as the analyte concentration that gives a signal equal to the blankplus three standard deviations of the blank, and was determined to be 5 nglmL.When samples containing different arsenic compounds are analyzed, anadditional complication arises: arsenic compounds do not all have the same volatility ata particular temperature, and some volatile arsenic compounds may be lost during theashing stage. To overcome these difficulties in the direct determination of total arsenicby GFAA, matrix modifiers were added.’52”3Chemical modification of the sample isessential for the analysis of many biological samples. The aim of chemical modificationin GFAA is to prevent the loss of elemental analyte at the high pyrolysis temperatures31required to remove the bulk of the matrix components during thermal pre-treatment ofthe sample.148’54 Among the chemical modifiers, nickel nitrate is commonly used inthe analysis of arsenic, where the added nickel forms a stable arsenide that atomizes at ahigh temperature.’48”9525Lately, palladium has been suggested as a veryeffective chemical modifier for arsenic and many other elements.156’7 Typically, 20jiL of palladium modifier, prepared as palladium nitrate (100 ppm) in citric acid (2.0%wlv), was added to the sample inside the graphite tube. Palladium nitrate needs to bereduced to Pd(0) metal by citric acid in order to act as an effective modifier. Palladiumwas found to form a stable intermetallic compound with arsenic in the graphitefurnace.’58 Under these conditions, the ashing temperature can be increased up to1450°C without any loss of arsenic by volatilization. A temperature of 1200°C was usedas the ashing temperature in our studies.Other matrix interferences can be minimized by using the standard additionstechnique which consists of taking a number of replicate aliquot portions of the samplesolution, and adding to them known increasing quantities of the arsenic standardsolution.159 Since all calibration solutions have the same composition with theexception of their analyte contents, the influence of impurities will be the same. Thestandard additions technique was used in the work to determine the total amount ofarsenic accumulated in cells of P. peniculus described in Chapter 5.The presence of background interferences due to the incomplete isolation of theradiation absorbed by the analyte element from other radiation, or radiation absorptiondue to the interferent, can be overcome by using a background correction device such asa deuterium lamp. In this system, the radiation from the line source (arsenic hollowcathode lamp) and a continuous source (deuterium lamp) is passed alternately throughthe graphite furnace. The measured absorbance with the line source is due to both the32analyte and background interference. When the continuous beam of radiation passesthrough the cell, the absorbance is almost entirely the result of background absorption.After the subtraction of the background absorption from the total absorbance, theabsorbance due to the analyte can be obtained.2.3.2 Hydride generation-gas chromatography-atomic absorption spectrometry(HG-GC-AA)The hydride generation procedure involves the reduction of the analyte to avolatile covalent hydride that is subsequently swept into a detector where quantitativemeasurements can be made. Holak, in 1969, was the first to use hydride generation forthe determination of arsenic using atomic absorption spectrometry.16°He generatedarsine by adding zinc to the sample solution acidified with hydrochloric acid andcollected the arsine in a trap cooled in liquid nitrogen. Finally the arsine was atomizedin an argon-hydrogen flame for quantification. This procedure proved to be extremelyuseful because the analyte is separated from the sample matrix, and therefore, matrixinterferences are substantially reduced, and even eliminated.There are two main reactions used for the generation of arsine. The first involvesa metal-acid reducing system such as zinc-hydrochloric acid.’6°The reaction can bewritten as below:4Zn + 8H+ + As(OH)3 = 4Zn2+ AsH3 + 3H20 + H2The major disadvantages of this method are that the time for complete reactioncan be as long as 10 minutes and it is possible to introduce contamination from the33metal. As a consequence, this process has not gained general acceptance for analyticalpurposes.The second and more effective method for the production of arsine involves theuse of sodium borohydride (sodium tetrahydroborate, NaBH4)and acid. The reaction iswritten as follows:6BH4- + 31120 + 3H + 5As(OH)3 = 5AsEL3 + B033-÷5H3B03 + 9H2The NaBH4-acid system can afford better yield, involves a shorter reaction time,and less contamination is present in the analysis. Initial applications involved the use ofNaBH4 as pellets dropped into a reaction flask containing an acid solution ofarsenic. 164-166 It is now customary to use a solution of NaBH4: the recommendedconcentration ranges from 0.5 to 10%. 117-139 Since 1969, the method has beenextended to the determination of other elements such as Hg, Ge, Sn, Pb, Sb, Bi, Se, andTe.’27439”61 63Braman et al.’62 noted that the reduction of arsenic compounds with sodiumborohydride was pH dependent and also related to the pKa of the individual arsenicacids (Table 2.4). This allowed for a selective reduction of arsenate or arsenite. Toseparate arsenate (or arsenite) from MMA and DMA, Braman et aL’62 and Andreae14°established cold-trapping procedures that effectively pre-concentrate several arsinesderived from the hydride-forming arsenicals present in 10-50 mL sample volumes (<1ppb). As a result of the volatility differences of the arsines, slow warming of the trapfacilitates their sequential determination with a suitable detector. Alternatively, GC34separation has been successfully used to separate AsH3,CH3As2,(CH3)2A5 and(CH3)As.139-144Table 2.4 Reaction conditions for the conversion of inorganic and methylarseniccompounds to volatile arsinesArsenic pKj Reduction Product b.p.Compounds pHarsenite 9.23 <7 AsH3 -55arsenate 2.20 > 4.0 no reaction1 AsH3 -55MMA 4.10 > 5.0 little reaction1-3 CH3A5H2 2DMA 6.19 1-4 (CH3)2As 36TMAO --- 1 (CH3)As 522.3.2.1 The hydride generation systemTwo hydride generators, batch type and continuous type, have been widely usedfor hydride production. In the batch type hydride generator, the acidified sample is firstpurged by a helium gas stream in a reaction vessel, and then reacted with the reducingagent (NaBH4)which is injected by means of a syringe. The hydrides produced arecarried by the helium gas, either directly into a detector or into a collection trap beforebeing introduced to a detector. In general, the batch type hydride generator is suitablefor samples containing a low arsenic level (usually < I ppb). The disadvantages of thisbatch-type hydride generator are: (1) a large volume of sample (>10 mL) is needed toobtain enough hydride for analysis; (2) the experimental results may critically depend on35the experience and skill of the operator; (3) the procedure can be time consuming, onesample analysis taking as long as 20-30 minutes.The continuous type hydride generator coupled with an arsine trap and GC-AAdetector is simple to operate and easy to automate. 167,168 Better precision can beachieved since a peristaltic pump is used for introducing reactants, and operatinginconsistency can therefore be eliminated. However, the arsenic concentrations requiredfor successful analysis are relatively high, usually above 5 ppb in arsenic. In spite of thislimitation, we have chosen to use a continuous type hydride generator for most of oursample analysis because of its fast sample introduction rate and good reproducibility.The time for one sample analysis can be reduced to less than 7 minutes. Only a smallvolume of sample (<5 mL) is needed for an analysis. This system has proved to beefficient and gives reproducible results in dealing with a large number of samples ofrelatively high arsenic concentration (5 ppb- 10 ppm).Volatile arsines produced in the continuous mode hydride generator (Figure 2.2)are carried by an inert gas and the produced hydrogen into a cooled trap, where they arecollected. This step concentrates the arsenic species before detection. Arsines are thenvolatilized by warming the trap, separated by using a gas chromatograph, and thendetected by using an atomic absorption spectrometer. Carbon dioxide and water thataccompany the arsine production can cause clogging of hydride trap (Teflon U-tube, 30cm x 0.4 cm i.d.), and affect the column efficiency. Sodium hydroxide, calciumchloride, potassium carbonate, silica gel, and magnesium perchloride have been used forremoving water and carbon dioxide.’40”62However, these reagents also absorb someof the arsines. To achieve the maximum trapping of water and the minimum loss ofarsines, we used a Teflon tube immersed in dry ice-acetone slurry as the water trap. Itwas found that the diameter of the tube is very critical. When the diameter is 0.4 cm, the36water trap is easily clogged by moisture produced during hydride generation. If theevolved gases are passed through a Teflon tube with a diameter of 2.0 cm, the water trapcould not remove moisture efficiently: clogging of the hydride trap is soon observed. ATeflon U-tube, 30 cm x 0.8 cm i.d., was chosen as the water trap because it is effectivein removing moisture carried out of the reaction system by the helium gas and theevolved hydrogen. The efficiency of this water trap was judged by the absence of wateraccumulated in the hydride trap immersed in liquid nitrogen. Although the water trapwas efficient in trapping moisture and easy to assemble, it should be noted that the U-tube needed to be cleaned after 8 to 10 runs because of clogging from the accumulatedwater. The possible loss of arsines in the water trap was assayed by comparing the arsineresponse signals with and without the presence of moisture trap. No significantdifference was observed. Other water traps such as the U-tube packed with NaOH orglass wool were also investigated: both water moisture and arsines are absorbed by thesetraps.Different cold traps have been used for arsine collection following hydridegeneration. Odanaka et al.169 trapped the arsines in n-heptane cooled with dry iceacetone. Aliquots of the heptane solutions were then injected into the GC-MS for arsenicanalysis. The liquid nitrogen-cooled trap, filled with different packing materials such asglass beads,140”44’62 and glass wool (silanized),’4°are by far, the most frequentlyused trapping techniques. Reimer17°used a Teflon U-tube (30 cm x 0.4 cm i.d.)immersed in liquid nitrogen (-196°C) as a hydride trap. In the present work, this trapwas found to collect the arsines efficiently and to give reproducible results, and wastherefore adopted for general use.37After all of the arsines are collected, the liquid nitrogen Dewar is removed andthe trap is warmed in a water bath (70°C) to volatilize the arsines. The arsines can befed directly into the detector where they arrive in the order of increasing boiling pointand molecular mass.147 In order to separate the arsines, Braman et al.’62 packed thehydride U-trap with glass beads and wrapped it with Nichrome wire connected to avariable transformer which was used to elevate the tube’s temperature. Quartz wool anddifferent chromatographic packing have also been used in the U-trap to increase theretention time.14°If a better and more reliable separation of the arsines is desired, a gaschromatograph can be placed between the liquid nitrogen trap and thedetector.139J41’70”72Chromosorb 101172 with 16.5% silicone oil DC-550,17 orChromosorb W (AW DMCS)13944°have served as stationary phases. Reimer’70reported that a column packed with Porapak-PS (80-100 mesh) gave a good separationof arsines. In the present work, several packing materials such as chromosorb W,silanized quartz wool, and Porapak-PS were investigated. Only Porapak-PS provides agood separation of the arsines and this packing was adopted for general use.2.3.2.2 OptimizationThe acid concentration, sodium borohydride concentration, and carrier flow (inreaction coil) were optimized by using the single parameter variation method. The peakarea of the signals was chosen as the response to be optimized.The effect of hydrochloric acid concentration on the determination of 60 ng eachof arsenite, arsenate, MMA, DMA, and TMAO is shown in Figure 2.3. The resultsindicate that the concentration of hydrochloric acid is more critical in the determinationof DMA and TMAO than it is for arsenate, arsenite, and MMA. The responses of38C,,CDU00001301201101009D80706050403020100HCI Concentration, MFigure 2.3 Effect of the concentration of hydrochloric acid on the determination ofhydride-forming arsenicals. The reductant was a 2.0% (wlv) NaBH4aqueous solution.(0) arsenite (•) arsenate (V) MMA (V) DMA (0) TMAO0 1 2 3 4 5 639arsenite, arsenate, and MMA increase rapidly with increasing acid concentration,reaching a constant value at concentrations above 0.5 M. The response for DMA isshown to reach a maximum at 0.75-1.0 M HCI, but falls thereafter, with the absorbanceapproaching 30% of its maximum absorbance in 4.0 M hydrochloric acid. Theproduction of trimethylarsine is not favored by the higher HC1 concentrations: thehighest response for TMAO is achieved in the presence of 0.1 M HC1. Similar resultswere reported by other workers using automated systems for the determination of “total”arsenic. 143,144,173 The reasons for the difference in responses are not known, althoughkinetic factors or mixing dynamics may be responsible. 143 Optimum ranges ofhydrochloric acid concentration for maximum sensitivity are 0.5-5 M for arsenate,arsenite, MMA; 0.75-1.0 M for DMA; and 0.1 M for TMAO. A solution of 1.0 M ofhydrochloric acid concentration was adopted for the work, and was primarily used forthe determination of total inorganic arsenic (arsenate and/or arsenite), MMA, and DMA.The selective reduction of arsenic species has been used for arsenic speciationanalysis by a number of researchersJ40’14214462873Most of the studies werecarried out by using a batch type hydride generator. Acetate, KHP, citrate, Tris-Trismaleate, and Tris-HC1 were used to give a sample pH value of 5 before injection of thereductant into the sample. Under these conditions, arsenite, but not arsenate, wasreduced to arsine by sodium borohydride. We have studied the effects of Tris-HC1 (1.0M, pH 6.2), acetate (2.0 M, pH 5), citric acid/citrate (1.0 M, pH 5.5), and acetic acid(4.0 M, pH 4.0) on the production of arsines in the continuous type hydride generator.They all gave satisfactory results for arsenate and arsenite separation. We chose aceticacid for routine use because it is economical and easy to obtain. The effects of aceticacid on arsine production are shown in Figure 2.4, where it can be clearly seen that theresponses from arsenite, DMA, and TMAO are at constant levels over almost the entire40U,CD0C13012011010090_80706050403020100Acetic Acid Concentration, MFigure 2.4 Effect of the concentration of acetic acid on the determination of hydride-forming arsenicals. The reductant was a 2.0% (w/v) NaBH4aqueous solution.(0) arsenite (V) MMA (V) DMA (D) TMAO0 1 2 3 4 5 6 741concentration range studied (1.0-6.0 M). The response for arsenate is almost negligibleand so not shown on the graph (Figure 2.4). The signal from MMA is largely suppressedin almost the entire range studied. Since acetic acid provides a wider optimum range forthe production of diniethylarsine and trimethylarsine, it was used not only to separatearsenite from arsenate but also to quantify arsenite, DMA and TMAO. A solution of 4.0M acetic acid was chosen for the work.The effects of the sodium borohydride concentration on the absorbance ofarsenate, arsenite, MMA, DMA, and TMAO in the presence of hydrochloric acid andacetic acid are shown in Figure 2.5 and Figure 2.6, respectively. When 1.0 Mhydrochloric acid is used, the results indicate that the concentration of sodiumborohydride is less critical in the determination of all the arsenicals studied exceptTtvIAO. The optimum concentration of sodium borohydride falls in the range of 1.0%-4.0% (wlv). The small decrease in the responses of all the arsenicals at higherconcentration (> 4.0%) of sodium borohydride (Figure 2.5) is probably due to poordecomposition of the borohydride; a large amount of hydrogen gas is produced from thereaction of the sodium borohydride with the hydrochloric acid. If 4.0 M acetic acid isused, the optimum range of sodium borohydride concentration for maximum sensitivityis 1.0-6.0% for arsenite, DMA, and TMAO (Figure 2.6). Therefore, a solution of 2.0%of sodium borohydride was adopted in the determination of arsenicals when either acidwas used.The effect of the flow rate of the carrier gas (used for carrying reactant intoliquid/gas separator) on the peak area of signals was also investigated (Figure 2.7).When flow rates are between 10 mL/min and 60 mL/min the arsenic atomic absorptionresponses are not changed significantly. When the flow rate is above 80 mL/min, theresponses of the arsines declines. This is probably due to a lowering of the time the42110 -100 -Sodium Borohydride Concentration, % (w/v)Figure 2.5 Effect of the conceniration of NaBH4 on the determination of hydride-forming arsenicals. The acid medium was a 1.0 M hydrochloric acid solution. (0)arsenite (•) arsenate (V) MMA (V) DMA (D) TMAO43130 -120 -110 -100 -90-8070-ci)< 60--0, 50-40 -::: /2 3 4 5 6 7Sodium Borohydride Concentration, % (w/v)Figure 2.6 Effect of the concentration of NaBH4 on the determination of hydride-forming arsenicals. The acid medium was a 4.0 M acetic acid. (0) arsenite (V) MMA(Y)DMA (D)TMAO44(1)CCVC‘a)0Carrier Gas Flow Rate, mL/minFigure 2.7 Effect of the carrier gas flow rate (in reaction coil) on the determination ofhydride-forming arsemcals. (0) arsenite (•) arsenate (V) MMA (Y) DMA (C)TMAO45110 -100 -90 -80 -70 -60 -50 -40 -30 -20 -10 -0-0 20 40 60 80 100 120reactants spend in the reaction coil. The high flow rate can also carry more water vaporto the moisture trap and the hydride trap, causing clogging of both traps. The slow flowrate does not affect the response of the arsines, but increases the operation time. Theoptimum flow rate of the carrier gas that does not alter arsenic response, and provides areasonable operation time without clogging the moisture and hydride traps is 30mL/min. The flow rate of the second carrier gas, used as the GC carrier gas, could affectthe separation of AsH3,CH3As2,(CH3)2As , and (CH3)As. If the flow rate isabove 70 mLlmin, the elutants are not separated completely. When the flow rate isreduced to 10 mL/min, the elution time can be as long as 6 miii. A good separation ofarsunes as well as a reasonable retention time can be achieved when the carrier gas flowrate is 20 mLfmin, and so this flow rate is adopted in the work.2.3.2.3 Interference studiesInterferences with the arsenic determination may be encountered during theformation of arsine in the hydride generation vessel, during transport to the detector, orin the detector. It has been reported that the presence of certain anions’73475 andcations143’4J736 82 can influence the production of arsine. Some transitionmetal ions can suppress arsine evolution during the reduction stage of hydridegeneration. 143 Part of these interferences can be avoided by adding complexung agentsto the arsenic-containing analyte. Complexing agents that have been used includethiosemicarbazide, 1,10-phenanthroline, EDTA, tartaric acid, thiourea and potassiumiodide. 144,176-182Possible interference effects were assessed by comparing the result of standardarsenic solutions in the absence or presence of Fe3 (FeCl3),Zn2 (Zn(N03)2), Mii246(MnC12), Cu2 (Cu(N03)2,Co2(CoC12), and Ni2 (NiC12). In the present work, theconcentrations of each arsenic species and the interfering metal ions used were 20 ppb,and 10 ppm, respectively. A 3 mL of sample was used for most of the studies. Table 2.5shows the effect of various metal ions on the response of arsenic species in two acidicsystems. The effects are expressed as percentage deviations from the interference-freeresponse. Differences of less than 5.0% may not be significant. Deviations greater than10% are considered to be the result of metal ion interference.In the hydrochloric acid reaction medium, the metal ions interfere with thereduction of all the arsenic species. Nickel has the greatest effect: it suppresses thesignal responses of arsine, CH3As2and (CH3)2As by almost 40%. When acetic acidis used for arsenic speciation analysis, the interference is quite significant in thepresence of Fe3+, Zn2+, Cu2+, and Ni2+. These results are consistent with those foundpreviously. 143,177The chelating agent EDTA was added in an attempt to remove the metal ioninterferences in the two acid systems. It was found that mixing 2 mL of 10% (w/v)EDTA (disodium salt) with the samples containing one of the five metals (10 ppm) 3mm before the reduction, effectively prevents interference from the metal ions. Aquantitative recovery (90-100%) of arsenic was achieved under these conditions.2.3.2.4 Calibration, limit of detection and precision of HG-GC-AA analysisQuantification of arsenicals can be easily achieved through calibration curvesderived from different standard arsenicals. Typically, calibration curves are obtaineddaily for the arsenic compounds by plotting absorbance (peak area) against the amountof arsenicals in the standard arsenic solutions. The responses of arsenate and arsenite are47Table 2.5 Percentage deviations of the response of arsenic species in the presence ofinterfering ions (10 ppm)Interfering arsenite arsenate MMA DMA TMAOionsin the presence of HC1Fe3 +7 +12 -10 -13 -21Zn2 -12 -5 -7 -12 -10Mn2 +7 +5 -10 +5 -21Cu2 -17 -22 -12 -15 -35Co2 -15 -12 -17 -11 -23Ni2 -38 -32 -41 -29 -45in the presence of acetic acidFe3 -15 -10 -17 -12 -15Zn2 -23 -10 -15 -23 -33Mn2 +5 +7 -8 -14 -12Cu2 -28 -33 -35 -50 -45Co2 +5 +10 -3 +15 -15Ni2 -70 -55 -62 -71 -8048similar when 1.0 M hydrochloric acid is used (Figure 2.3). Therefore, the calibrationcurves for arsenate and arsenite can be used interchangeably to calculate the totalamount of inorganic arsenic (arsenate and arsenite). The quantification of MMA can beachieved by using the calibration curve obtained in 1.0 M hydrochloric acid. When 4.0M acetic acid is used, the hydride production from arsenate is completely suppressed,and under these conditions arsenite, DMA and TMAO can be quantified. The calibrationcurves are linear to 200 ng for arsenate, arsenite and MMA, and 250 ng for DMA andTMAO when 1.0 M HC1 is used. Under 4.0 M acetic acid conditions, the calibrationcurves are linear to 200 ng for arsenite, DMA and TMAO, and 400 ng for MMA.On the basis of 8 replicate runs of standards containing 60 ng of arsenic, theanalytical precision for arsenate and arsenite are 7% relative standard deviation (%RSD), 5% RSD for MMA, 9% RSD for DMA, and 9% RSD for TMAO.The limits of detection, defmed as the amount of analyte which gives a signalequal to the blank plus three standard deviations of the blank, are 0.20 ng (of arsenic)for arsenate and arsenite, 0.15 ng for MMA, and 0.30 ng for DMA and TMAO in thepresence of 1.0 M hydrochloric acid. When 4.0 M acetic acid is used, the detectionlimits are 0.20 ng for arsenite, DMA and TMAO, and 0.40 ng for MMA.49CHAPTER 3THE IDENTIFICATION OF EXTRACELLULAR ARSENICAL METABOLITESiN ThE GROWTh MEDIUM OF MICROORGANISMS3.1 INTRODUCTIONThe first investigation of the biological production of an arsine was published in1891 and was concerned with the formation of “Gosio gas” (Section 1.1.2).28 It wasonly after Challenger started to work on the problem that the metabolic product wascorrectly identified as trimethylarsine.3°A metabolic pathway was proposed for thebiomethylation of arsenicals to trimethylarsine (Figure 1.l).3O43 This pathway is madeup from two basic steps: (1) reduction of the arsenic(V) species to arsenic(lll) species,possibly oxides; and (2) subsequent oxidative methylation to methylarsenic(V) moieties.Evidence supporting Challenger’s pathway has come from studies onScopulariopsis brevicaulis and Apiotrichum humicola (previously known as Candidahumicola).3035’7Experiments demonstrated that arsenate, arsenite, MMA, DMA, andTMAO served as precursors for trimethylarsine synthesis. These arsenicals are likelyintermediates in the biomethylation reactions and are known to be present in theenvironment and the biosphere. However, the presence of arsenite, MMA, DMA andTMAO as metabolic intermediates in the biomethylation of, for instance, arsenate, hasrarely been reported.2 Challenger stated that arsenic intermediates from the proposedmetabolic pathway (Figure 1.1) were not found in the culture medium of S. brevicaulis,although no details were given regarding the methodology used to support thisconclusion.43Baker et at. 183 incubated arsenite and arsenate, respectively, in a nutrientmedium containing sediment collected from a small acidic, oligotrophic lake. By using50HG-GC-AA techniques, they reported that 0-0.7% of the total arsenic was transformedto MMA and DMA and a variety of microorganisms are believed to have contributed tothese biological methylation processes. Cullen et al.5’ incubated labeled arsenicals withbroken-cell homogenates of Apiotrichum humicola in order to look for metabolicintermediates in the growth medium. They employed a combination of molecularsieving, anion exchange, and elect.rophoresis to separate the arsenicals from each other.Arsenite, MMA and DMA were found to be metabolites of[74As]arsenate, and TMAOwas a metabolite of[14Cjmethylarsonate and[14C]dimethylarsinate. In addition, ademethylation product,14C]methylarsonate, was observed when[14C]dimethylarsinatewas used. Replacement of the cell preparation by buffer failed to bring about anytransformations. Hence, the various compounds identified represent probableintermediates in the biosynthesis of trimethylarsine. When S. brevicaulis and A.humicola were treated with the model arsenic(ffl) intermediate (CH3AsO), theproduction of DMA, together with trace amounts of MMA and trimethylarsine wasobserved.45This was the first time that non-volatile methylated intermediates had beenidentified in the growth medium of a pure culture.In the present work we report on the effect of adding low levels of four arseniccompounds, arsenate, arsenite, MMA and DMA to cultures of the microorganisms A.humicola and S. brevicaulis growing aerobically in a liquid medium. Hydridegeneration-gas chromatography-atomic absorption spectrometry (HG-GC-AA) was usedto identify the extracellular arsenic metabolites present in the culture medium.513.2 EXPERIMENTAL3.2.1 ReagentsAll chemicals and reagents are as previously described in Section 2.2.1.3.2.2 Microorganism culturesApiotrichum humicola was obtained from the American Type Culture Collection(ATCC 26699) and Scopulariopsis brevicaulis was provided by the “Fungus CultureCollection” of the Chemistry Department at U.B.C. The cultures were grown aerobicallyin a synthetic inorganic liquid medium at pH 5 as described by Cox and Alexander(Table 3.1).32Table 3.1 Composition of the growth medium (1 L)32(NH4)2S0 2.0 gKH2PO4 0.1 gMgSO.7H0 0.05 gFeSO4.7H2 0.0018 gthiamine hydrochloride 0.01 gglucose 10 gsuccimic acid buffer (pH 5) 0.05 M3.2.3 Hydride generation-gas chromatography-atomic absorption spectrometry(HG-GC-AA)The HG-GC-AA system described in Section 2.3.2 was used for arsenic analysis.A peristaltic pump was used to mix the sample solution (1-3 mL) with an acid (1.0 M52HC1 or 4.0 M acetic acid) and 2.0% (w/v) NaBH4 solutions, and the arsines producedwere collected in a U-tube immersed in liquid nitrogen. After warming the hydride trapin a water bath (70°C), these arsines were separated in a Porapak-PS GC column byusing a Hewlett Packard Model 5830A gas chromatograph. An 810 Jarrell-Ash atomicabsorption spectrometer was used as the arsenic detector at 193.7 nm. The arsines wereatomized in an air/hydrogen flame in a quartz cuvette. The signal was recorded on aHewlett Packard 3390A integrator.3.2.4 Experimental proceduresAqueous solutions of the appropriate arsenical were filter sterilized (0.2 .tmmembrane) separately and added to the autoclaved culture medium in the flask.Typically, a 10 mL aliquot of an actively growing culture of A. humicola or a 2 mLaliquot of an actively growing culture of S. brevicaulis was added to 250 mL of themedium which contained 1 ppm of arsenic. During the growth period the cultures weremaintained at 2 1-22°C and were agitated by using a rotary shaker running at 130 rpm.Once each day for 2 weeks, a 6 mL culture aliquot was removed and stored frozen priorto HG-GC-AA analysis. After the first 2 weeks of incubation, a glass microfiber paperpre-soaked in 5.0% mercuric chloride was suspended in the head-space of cultures totrap the volatile arsines (chemofocusing) as shown in Figure 3.1.36 The medium (6 mL)was then collected once each week for another 2 to 3 weeks. The experiment wasterminated after 4 weeks incubation of A. humicola and 5 weeks incubation of S.brevicaulis. Terminal cultures were centrifuged, and the cells were washed and freeze-dried for future analysis.53plugglass filter paper5% HgC1Figure 3.1 The growth flask fitted with the mercuric chloride trap.To determine the arsenic species in the cells, the freeze-dried samples wereweighed and transferred into an Erlenmeyer flask (250 mL) containing 30 mL of mixedsolvent CHC13/MeOHIH2O(1/1/1). The mixture was sonicated for 2 hours and thenagitated on a mechanical shaker for 24 hours. The extracts were centrifuged to separatethe aqueous fraction from the organic fraction. The aqueous layer was kept at .4°C priorto analysis. The organic extract and the residue were air dried, digested in 4 mL of 2.0M NaOH in a water bath at 95°C for 3 hours, then neutralized with 6.0 M hydrochloricacid prior to HG-GC-AA analysis.The hydride generation system described in Section 2.3.2 was used to identifyhydride-forming arsenicals in growth media as well as in the cells. To determine thearsenic species in the growth medium, the sample was diluted to an appropriate volumebefore it was subjected to HG-GC-AA analysis. Usually, 0.2 mL of the sample in thegrowth medium was diluted to 3 mL and mixed with 1.0 M HC1 or 4.0 M acetic acidand 2.0% (w/v) NaBH4. To monitor the production of arsenite from arsenate byplastic strip54microorganisms, the hydrochloric acid solution was replaced by 4.0 M acetic acid.Under this condition, arsenate is not reduced to arsine by sodium borohydride solution.During culture incubations, the presence of triinethylarsine was assessed byusing two methods: one based on odor and the other on chemofocusing. An intense anddistinctive garlic-like odor has been used as qualitative evidence of arsineproduction.184-8 The odor threshold for (CH3)As now appears to be 2 pg. g1 indilute aqueous solution.’89 This allows qualitative evaluation of arsine production bycautious sniffing of the gas in the culture head-space. The chemofocusing method hasalso proved to be an effective means to trap the volatile arsines.36 If trimethylarsine isproduced, crystals of the HgCl2 adduct are formed on a glass fiber filter soaked in 5.0%mercuric chloride solution that is suspended in the head-space of cultures (Figure 3.1).Subsequent heating of the filter decomposes the mercuric chloride adduct and thevolatile arsines are released for mass spectrometric analysis.363.3 RESULTSA. humicola and S. brevicaulis grown in the presence of each of the foursubstrates arsenate, arsenite, MMA, and DMA produced a number of compounds;therefore each substrate will be discussed separately.3.3.1 Transformation of arsenateThe growth medium was analyzed by using HG-GC-AA. Selective reduction ofarsenicals can be accomplished by varying the pH, and this was used to distinguisharsenate from arsenite in the samples. Thus when 1:0 M HC1 is used, both arsenate and55arsenite in these samples are reduced to AsH3 by 2.0% (wlv) sodium borohydride. Inthe presence of 4.0 M acetic acid, arsenite forms arsine, but arsenate does not.A. humicola exposed to I ppm arsenate reduced more than 90% of the substrateto arsenite (detected as AsH3)within 2 days of incubation as shown in Figures 3.2 and3.3. The chromatograms of AsH3 in Figure 3.2 were obtained when 1.0 M HC1 wasused to assist the hydride production. The AsH3 was derived from arsenate in thesample on day 0, and from a mixture of arsenate and a.rsenite in the sample on day 2.Figure 3.3 shows that only traces of arsenite, detected as AsH3 following reduction inthe presence of acetic acid, were in the sample on day 0. However, a strong AsH3 signalwas observed on day 2 under the same conditions, indicating the production of arsenitefrom arsenate. Oxidation of the arsenite to arsenate was not observed during the rest ofthe growth period although the concentration of arsenite in the growth medium furtherdecreased with time, and only background levels were detected after 4 weeks ofincubation (Figure 3.4). In addition to arsenite, small amounts of DMA and TMAOwere detected as (CH3)2As and (CH3)As, respectively, in the culture mediumcollected on day 5. The concentration of DMA increased from 0 to 0.02 ppm by day 7,but a further increase was not observed. The TMAO concentration in the growthmedium increased rapidly with time, reaching 0.75 ppm at the end of the fourth week.The formation of the CH3As2 from MMA is largely suppressed when the hydridegeneration is carried out in acetic acid as described in Chapter 2, and trace amounts ofMMA (if present in the growth medium) would not be detected. However, when HC1was used instead the acetic acid, trace amounts of MMA were detected as CH3As2inthe culture medium after 2 weeks of incubation, but the concentration remained constantfor a further 2 weeks. The change in the concentrations of the arsenic species as afunction of incubation time is shown in Figure 3.5. It should be noted that adsorption of560 1.0 2.0 3.0 4.0Time (mm) Time (miii)(a) (b)Figure 3.2 Chromatograms of arsenic compounds in the growth medium of A. humicola(enriched with 1 ppm arsenate) were obtained by using HG-GC-AA with 1.0 M HC1 and2.0% (w/v) NaBH4.(a) the growth medium collected on day 0, (b) the growth mediumcollected on day 2Figure 3.3 Chromatograms of arsenic compounds in the growth medium of A. humicola(enriched with 1 ppm arsenate) were obtained by using HG-GC-AA with 4.0 M aceticacid and 2.0% (w/v) NaBH4.(a) the growth medium collected on day 0, (b) the growthmedium collected on day 2AsR3I0 1.0 2.0 3.0 4.0I’ I0 1.0 2.0 3.0 4.0Time (mm)(a)0 1.0 2.0 3.0 4.0Time (miii)(b)57_LMH3________ITI’-.I • • ______Io 1.0 2.0 3.0 4.00 1.0 2.0 3.0 4.0Time (mm)Tmm (mm)(a) (b)((CHASI • —0 1.0 2.0 3.0 4.00 1.0 2.0 3.0 4.0Tmm (mm)1m (mm)(c) (d)Figure 3.4 Chromatograms of arsenic compounds in the growth medium of A. humicola(enriched with 1 ppm arsenate) were obtained by using HG-GC-AA with 4.0 M aceticacid and 2.0% (w/v) NaBH4.(a) the growth medium collected on day 0, (b) the growthmedium collected on day 5, (c) the growth medium collected on day 15, (d) the growthmedium collected on day 2858Incubation time (days)Figure 3.5 The change in arsenic concentrations in the growth medium of A. humicolaenriched with 1 ppm arsenate. (0) arsenite (•) arsenate (V) DMA (D) TMAO**The values are averages of four replicate determinations by HG-GC-AA of separate aliquots (0.2 mL)from the 6 anL sample taken at each time interval. The error bars represent 90% confidence intervals,and are smaller than the size of the symbols when they are not evident on the graph.1 .00.80.60.40.20.00 5 10 15 20 25 30594TMAO on the surface of the cells can also take place; the concentration of TMAO inthe growth medium decreases dramatically if the sample is not shaken before analysis(Figure 3.6).0 1.0 2.0 3.0Time (mm)The change of arsenic speciation in the growth medium of S. brevicaulis is lessdramatic. The total reduction of arsenate to arsenite was achieved 4 days afterinoculation of the stock culture. The quantity of arsenite (- 1 ppm) in the medium didI (CH3)A(cH3)As(cH3)1As4.0 0 1.0 2.0 3.0Time (n3in)4.0(a) (b)Figure 3.6 The effect of adsorption by the cells of A. humicola on the concentration ofTMAO in the growth medium. (a) without agitation before analysis, (b) with agitationbefore analysis.60not change significantly over a 5 week incubation period. TMAO (0.01 ppm) and traceamounts of DMA were detected on day 5. The amount of TMAO had increased slowlyto approximately 0.04 ppm at the end of the experiment. During this period, the quantityof DMA did not change significantly. No MMA was detected in these samples.Neither A. humicola nor S. brevicaulis produced trimethylarsine as judged by theodor test or by the use of the chemofocusing trap. In the latter case no arsine wasdetected by mass spectrometry when strips of the filter were analyzed.No significant amounts of arsenic were found in cells of A. humicola and S.brevicaulis at the end of the incubation time by using HG-GC-AA.3.3.2 Transformation of arseniteActively growing cultures of A. humicola (10 mL) and S. brevicaulis (2 mL)were inoculated with 250 mL of liquid media containing arsenite (1 ppm) and grownaerobically at 21-22°C for 4 or 5 weeks, respectively.The concentration of arsenite in the culture medium of A. humicola decreasedrapidly from 1 ppm to 0.29 ppm in 2 weeks, and reached a background level after 4weeks of incubation (Figure 3.7). The rate of arsenite disappearance was similar to thatdiscussed in Section 3.3.1. The oxidation of arsenite to arsenate in the growth mediumwas not observed. Both TMAO and DMA were detected as (CH3)As and (CH3)2As ,respectively, in the growth medium on day 5. The DMA concentration increased from 0to 0.02 ppm in 2 weeks of incubation, and no further change was found thereafter. TheTMAO concentration increased from 0.05 ppm on day 5 to 0.69 ppm at the end of thefourth week. Trace amounts of MMA were also detected after 2 weeks of incubation,but the quantity of MMA did not increase significantly over a further 2 weeks61AsH3 AsH3ii- (CH3>ASIII.III ‘1p I p I0 1.0 2.0 3.0 4.0 0 1.0 2.0 3.0 4.0Time (mm) Tune (miii)(a) (b)(cH3)Asring (mm) Time (miii)(c) (d)Figure 3.7 Chromatograms of arsenic compounds in the growth medium of A. humicola(enriched with 1 ppm arsenite) were obtained by using HG-GC-AA with 4.0 M aceticacid and 2.0% (wlv) NaBH4.(a) the growth medium.collected on day 0, (b) the growthmedium collected on day 5, (c) the growth medium collected on day 15, (d) the growthmedium collected on day 2862incubation. The change in concentrations of arsenite, DMA and TMAO in the growthmedium is shown in Figure 3.8. The adsorption of TMAO on the surface of the cells ofA. humicola was also observed.In the growing culture of S. brevicaulis, the substrate arsenite was not oxidizedto arsenate, and the change in its concentration was minimal. TMAO and trace amountsof DMA were detected after 8 days of incubation. TMAO in the medium increasedslowly with time, reaching about 0.04 ppm at the end of 5 weeks of incubation. Therewas no further increase in the concentration of DMA over the same period. No MMAwas detected in the culture medium.No significant accumulation of arsenic in cells of A. humicola and S. brevicauliswas found by using HG-GC-AA.No trimethylarsine was produced by the two microorganisms in the experiments.3.3.3 Transformation of methylarsonateFollowing the addition of MMA, DMA (0.02 ppm in the medium) and TMAO(0.01 ppm in the medium) were detected in the A. humicola culture on the second day ofincubation (Figure 3.9). The concentration of DMA and TMAO increased to 0.15 ppmand 0.33 ppm, respectively, at the end of the experiment. The MMA concentration in themedium decreased from 1 ppm to 0.5 ppm at the end of the incubation. The change inarsenic concentration is shown in Figure 3.10. TMAO adsorption on the surface of cellsof A. humicola was observed as judged by a significant increase in TMAO concentrationin the sample after it was agitated a few times.For S. brevicaulis, TMAO and trace amounts of DMA were detected in thegrowth medium after 2 weeks of incubation. TMAO concentration had increased from 0631 .00.830Incubation time (days)Figure 3.8 The change in arsenic concentrations in the growth medium of A. humicolaenriched with 1 ppm arsenite. (0) arsenite (Y) DMA (0) TMAO**The values are averages of four replicate determinations by HG-GC-AA of separate aliquots (0.2 mL)from the 6 niL sample taken at each time interval. The error bars represent 90% confidence intervals,and are smaller than the size of the symbols when they are not evident on the graph.640 5 10 15 20 251O.O4OTi (=) Time (mm)(a) (b)_2J10 1.0 2,0 3.0 4.0 0 1.0 2.0 3.0 4.0Time (mm) Ti (mm)(c) (d)Figure 3.9 Chromatograms of arsenic compounds in the growth medium of A. humicola(enriched with 1 ppm MMA) were obtained by using HG-GC-AA with 4.0 M aceticacid and 2.0% (wlv) NaBH4.(a) the growth medium collected on day 0, (b) the growthmedium collected on day 2, (c) the growth medium collected on day 15, (d) the growthmedium collected on day 28651.00.8 -JN060.4 -T0.20.00510 15 20 2’5 30Incubation time (days)Figure 3.10 The change in arsenic concentrations in the growth medium of A. humicolaenriched with 1 ppm MMA. (V)MMA (V) DMA (D) TMAO**The values are averages of four replicate determinations by HG-GC-AA of separate aliquots (0.2 rnL)from the 6 mL sample taken at each time interval. The error bars represent 90% confidence intervals,and are smaller than the size of the symbols when they are not evident on the graph.66to 0.02 ppm when the experiment was ended. The MMA concentration in the mediumdid not change significantly in the time period of the experiment.No trimethylarsine was detected when MMA (1 ppm) was incubated with A.humicola or S. brevicaulis. The possible demethylation products, arsenate and/orarsenite were not observed.Accumulation of arsenicals in the cells of both microorganisms was notobserved.3.3.4 Transformation of dimethylarsinateFollowing the addition of DMA, trace amounts of TMAO were detected in thegrowing culture of A. humicola after 15 days of incubation, and the decrease in theDMA concentration was insignificant.TMAO was detected in the culture medium of S. brevicaulis but not until day 10.The concentration of TMAO increased with time and was about 0.07 ppm in the growthmedium after 5 weeks of incubation.Incubation of A. humicola and S. brevicaulis with DMA (1 ppm) did not producedetectable amounts of trimethylarsine or demethylation products, such as MMA.The accumulation of arsenicals in cells of A. humicola and S. brevicaulis was notdetected by using HG-GC-AA.3.4 DISCUSSIONA combination of molecular sieving, anion exchange, and electrophoresis hasproved to be a reliable method to isolate and identify the arsenicals present in biological67material.45’51 These procedures, however, are lengthy and time consuming especiallywhen dealing with large numbers of samples; there may also be problems with detectionwhen the concentration of analytes is low. The HG-GC-AA system used in this studycan minimize the time for analysis (less than 10 mm per sample) and can be used todetect low levels of arsenate, arsenite, MMA, DMA and TMAO in samples providedthat it is assumed that these are the precursors to the arsines that are ultimately formedand quantified. On this basis, the technique is capable of determining the concentrationof these extracellular arsenic metabolites at concentrations which are too low to bedetected by other convenient techniques. However, a possible disadvantage of themethod is that the separation and identification of the arsenic species is based on theproperties of derivatives of the arsenicals in the medium rather than on the properties ofthe arsenicals themselves.Exposure of A. humicola and S. brevicaulis to arsenate yields arsenite, DMA andTMAO in the growth medium, and MMA is found in the growth medium of A.humicola but not S. brevicaulis. The substrate arsenite is metabolized by A. humicola toproduce MMA, DMA and TMAO and by S. brevicaulis to produce DMA and TMAO.Both microorganisms transform MMA to DMA and TMAO, and DMA to TMAO.The identification of these arsenicals is noteworthy as it is the first time that anyof the non-volatile arsenic intermediates (shown in Figure 1.1) have been identified inthe growth medium of a pure culture spiked with either arsenate, arsenite, MMA orDMA. As was mentioned in the introduction, methylated intermediates have only beenisolated from broken-cell extracts of A. humicola5’and from the growth medium ofcultures of A. humicola, S. brevicaulis, V. alcalescens and L. brevis when methylarsineoxide [(CH3AsO)n] was used as a substrate.4568The reduction of arsenate to arsenite is the first step towards methylation inChallenger’s mechanism.3043Our results support this assertion, as most of the arsenateis reduced to arsenite prior to the detection of any methylated arsenicals in the medium.The production of the anticipated methylated intermediates from the substrates, theabsence of oxidation of arsenite to arsenate, and the lack of demethylation productsstrongly support the metabolic sequence of Figure 1.1 proposed by Challenger.30’43Cells of A. humicola take up and metabolize arsenate to arsenite, and thenexcrete the arsenite into the medium. The arsenate is believed to be reduced inside thecells by thiols or dithiols as a detoxification process.2 The uptake of arsenate by thecells probably involves an active transport system because when the phosphateconcentration is equimolar with arsenate, the rate of arsenate uptake by Apiotrichumhumicola is reduced to 18% of that observed in the absence of phosphateJ90 Theuptake of MMA and DMA probably occurs by means of slow passive diffusion.’9°Cullen et aL51 reported that only trace amounts of MMA are produced fromarsenate by broken cell homogenates of A. humicola and that MMA is the leasttransformed arsenical by the broken-cell homogenates. They speculated that MMAwould not be found as a free intermediate in Challenger’s arsenate-to-trimethylarsinepathway. The present results show that MMA is a metabolite of arsenate and arsenite inwhole cell cultures of A. humicola, but it is produced in a much lower concentrationthan either DMA or TMAO. We also fmd that the production of DMA in the growthmedium of A. humicola is a rapid process when MMA is used as a substrate since DMAand TMAO are detected within 2 days of incubation. This result indicates that MMA, ifit exists as an intermediate in the arsenate/arsenite-trimethylarsine methylation process,would be metabolized rapidly inside the cells. Therefore, little MMA would be excretedand detected as an extracellular metabolite of the cells. Cullen and Nelson’993 have69studied the biomobility of MMA and DMA by measuring the rate at which thesearsenicals diffuse through the walls of liposomes, regarding these as models forbiological membranes. It was found that the membranes are less permeable to MMAthan DMA. The diffusion coefficient of MMA is 10 times lower than DMA. Because ofthe low diffusion coefficient, it is possible that the cells metabolize the endocellularMMA to DMA and TMAO faster than MMA can diffuse into the growth medium. TheDMA thus produced has a higher permeability coefficient, and can diffuse into thegrowth medium. This premise seems to be consistent with what we have observed.In our work, the transformation of DMA to other arsenic metabolites is a slowprocess in cultures of A. humicola. This is a surprising result as DMA has been found tobe a better precursor to trimethylarsine than is arsenate or MMA.32 We suggest thatthese latter experiments were carried out with high concentrations of arsenicals (> 100ppm) and that this may be necessary to trigger the methylation process from DMA to(CH3)As, as is discussed next. Other factors may also affect the low production ofTMAO, and they are discussed in the proposed methylation model.Apart from the substrate DMA, which is hardly metabolized, the concentrationsof all other arsenic substrates decrease with time in the growing culture of A. humicola,corresponding to an increase in the concentrations of arsenic metabolites. Only traceamounts of inorganic arsenic are detected in the growth medium at the end ofincubations when arsenate or arsenite is used as a substrate: the major transformationproduct is TMAO. The total concentration of arsenicals in the two growing cultures isrelatively constant during the growth period (approximately 0.8-0.9 ppm), indicatingthat the uptake, the methylation, and the excretion of arsenic proceed simultaneously.Since neither the production of the volatile trimethylarsine is observed during thegrowth period nor significant amounts of arsenic are found in the cells harvested at the70end of the mcubations, the small decrease in the total arsenic concentration of the twogrowth media may be due to the adsorption of TMAO on cell surface of A. humicola.When MMA is used as a substrate, the arsenic speciation in the growth medium ischanged dramatically while the total concentration of arsenic is kept relatively constant(0.9-1.0 ppm). This also indicates that the accumulation of arsenic and the production ofvolatile trimethylarsine by the cells are limited.The detection of TMAO in the growing culture medium of microorganisms is avery significant observation. Not only is this the first time that it has been found as anextracellular metabolite of fungi, but also it is the major arsenic metabolite of themicroorganism. It is well known that both A. humicola and S. brevicaulis methylatearsenate, arsenite, MMA and DMA to trimethylarsine.3035 However, this volatilearsine was not detected during the present experiments. It is unlikely that this volatilemetabolite is lost during sampling since the chemofocusing trap is an effective arsinecollection device.36 Pickett et al.37 have demonstrated that TMAO can be reduced totrimethylarsine rapidly by whole cells of A. humicola. The rate of arsine productionfrom TMAO increases linearly with the TMAO concentration, and is considerably fasterthan from arsenate or DMA. Because of this rapid reduction of TMAO, it was suggestedthat TMAO would not be detectable as an intermediate in cultures of A. humicola andthat it would be unlikely to be found in the environment.37 However, in the presentexperiments TMAO is found to be the major methylation product and seems to be theend product of the methylation process. It now seems likely that low concentrations ofTMAO (< 1 ppm) do not greatly affect the living fungal system. Therefore, furtherdetoxification by reducing TMAO to trimethylarsine is not necessary. Previously, muchhigher concentrations (> 100 ppm) of the arsenicals including arsenate, arsenite, MIVIA,DMA and TMAO, were added to the growing culture of A. humicola to produce71thmethylarsine.3035’7These high concentrations were used because of analyticalexpedience and not for any scientific reason. The observation of TMAO is not withoutprecedent. Kaise et al.’94 incubated arsenobetaine (100 ppm) in an inorganic mediumcontaining bottom sediment collected from coastal waters. After 100 hours ofincubation, arsenobetaine was degraded into TMAO completely. This degradation isbelieved to be caused by indiginous microorganisms. Trimethylarsine was not detectedin these experiments.194The limited metabolism of the arsenic substrates by S. brevicaulis in the growthmedium is not surprising. This liquid medium was optimized by Cox and Alexander forthe production of trimethylarsine by A. humicola: it has since been shown to support thegrowth of S. brevicaulis but no comparative growth studies were made.32 Breadcultures of S. brevicaulis may be more productive than cultures in liquid media, eventhough the reported rates of arsine production are not great.2Although our fmdings lend general support to Challenger’s proposed pathway,they indicate that it is an over-simplification of the many processes involved. Therefore,an extended model is proposed in Figure 3.10, for the methylation of arsenate bygrowing cells of Apiotrichum humicola. At first, the cells take up arsenate from themedium through a phosphate transport system, reduce the arsenate to arsenite inside thecells probably by using thiols and/or dithiols, and excrete most of the arsenite into thegrowth medium, again probably by using an active transport system. This result can beachieved within 2 days. Any arsenite in the cells can be methylated to MMA by Sadenosylmethionine, SAM, but because of its low passive diffusion coefficient, most ofthe endocellular MMA does not diffuse into the growth medium. Rather, theendocellular MMA is more likely to be reduced and methylated to DMA and thenTMAO. It is also possible that two methyl groups from SAM can be transferred to the72phosphate Arsenatea thiols and/or Arsenitea’ active transport bdithiols II system - Arsenitetransport systemIIcell membranemedium cells mediumassive birseniteb Y!P.a.YL llArsemtea 2e MMAa 2e DMAa:iffusion____‘DMA>%\5 Ittransport system CH3 CH3++________TMAOa passive b“TMAOI diffusioncell membranemedium cells mediumFigure 3.11 Proposed biotransformation model of arsenate in A. humicola.a endocellular arsenicals b extracellular arsenicalsintermediate MMA sequentially to form TMAO, but without the formation of DMA as afree intermediate. This is indicated by the low and constant level of extracellular DMAin the growth medium and by the slow methylation process observed when DMA is usedas a substrate. Both DMA and TMAO are able to diffuse into the growth medium. Theexcreted or added arsenite enters the cells of A. humicola by means of an active73transport system and/or by passive diffusion, and is metabolized in the same sequence toform MMA, DMA and TMAO. When MMA and DMA are used as substrates, theiruptake is achieved by means of passive diffusion and the methylation process is similarto that presented above.Although this model can explain some of the results of our work, thebiotransformation process is undoubtedly more complex in reality.74CHAPTER 4THE BIOMETHYLATION OF ARSENICALS BY A MICROORGANISMAPIOTRICHUM HUMICOLA IN THE PRESENCE OF L-METHIONINEMETHYL-D34.1 INTRODUCTIONSince the first systematic investigation of the biotransformation of arsenicconducted by Gosio in 189 1,28 many studies have shown that the biological methylationof arsenic is a ubiquitous phenomenon in nature;2however, the process is still not fullyunderstood.2Challenger showed that Scopulariopsis brevicaulis produced trimethylarsinefrom arsenate and proposed a biosynthetic pathway as outlined in Figure i.i.30,43Evidence supporting the Challenger pathway has come from studies ofmicroorganisms.30-5’7This work demonstrated that arsenate, arsenite, MMA, DMA,and TMAO are precursors to trimethylarsine.Challenger suggested that some methyl containing compounds such as betaine,methionine, or a choline derivative could be the possible methyl donor to arsenic.3°Tofurther investigate this hypothesis, Challenger et at.48 added these compounds, labeledwith 14C, to cultures of S. brevicaulis growing on bread crumbs enriched with arsenite.These experiments showed that only 14C labeled methionine14CH3SCH2(N)H OOH was able to transfer its label to arsenite to asignificant extent. The maximum methylation percentage observed was 28% after 5 daysof incubation. Cullen et al.33 demonstrated that the CD3 group in L-methionine-methyld3 was incorporated into the trimethylarsine that was evolved from cultures of75Apiotrichum humicola and S. brevicaulis grown in the presence of arsenite, arsenate,MMA, and DMA. In these experiments, the arsine was collected by cryofocusing inliquid oxygen, and characterized by using mass spectrometry.33These results stronglyindicate that “active methionine”, identified as S-adenosylmethionine (SAM),49 isinvolved in the transfer of the methiomne methyl group to arsenic during mycologicalmethylation.48In Chapter 3, we reported that extracellular arsenic metabolites, mainlydimethylarsenic species (probably DMA) and trimethylarsenic species (probablyTMAO), were found in the growth medium of pure cultures of Apiotrichum humicolaand Scopulariopsis brevicaulis grown in the presence of arsenate, arsenite, MMA orDMA. These results reinforce the validity of the mechanism proposed by Challenger;however, it has not been verified that SAM is the methyl donor for the extracellulararsenic metabolites produced by A. humicola.In this chapter we report on the effect of adding L-methionine-methyl-d3andeach of the four arsenic substrates arsenate, arsenite, MMA, and DMA, separately, to thegrowing culture of Apiotrichum humicola. The incorporation of the deuterated methylgroups from methionine in the extracellular arsenic metabolites was monitored by usinghydride generation-gas chromatography-mass spectrometry (HG-GC-MS). The massspectra obtained by using this technique provided conclusive evidence of CD3incorporation in the arsenic metabolites.764.2 EXPERIMENTAL4.2.1 ReagentsAll chemical reagents were prepared as previously described in Chapters 2 and 3.A solution of L-methionine-methyl-d3,CD3SCH2H(NH)COO (Aldrich) wasprepared by dissolving the compound in deionized water.4.2.2 Microorganism culturesThe source of the cultures and the growth medium is as described in Section 3.2.4.2.3 Hydride generation-gas chromatography-mass spectrometryA continuous type hydride generation apparatus connected to a gaschromatograph-mass spectrometer system was used to generate, separate, andcharacterize arsines (Figure 4.1). The hydride generator was thoroughly described inChapter 2 and 3. A peristaltic pump was used to introduce and mix the sample solution(1-3 mL) with 4.0 M acetic acid and 2.0% (wlv) sodium borohydride solutions. In thepresence of acetic acid, arsenite, MMA, DMA, and TMAO can all be reduced to theircorresponding arsines, but arsenate does not react. These conditions were optimized forefficient hydride production of (CH3)2As and (CH3)As. The arsines produced wereseparated on a Porapak-PS column (80-100 mesh) by using a gas chromatograph(Varian, Vista 6000 GC) with a pre-set temperature program (Table 4.1), and werecharacterized by using a quadrupole mass spectrometer (Nermag Rb-bC). Dataacquisition and processing were performed by using a PC based data system (Teknivent,77PeristalticPumpReactioncoil,Sample))____AcidI___NaBH4HeliumVent___________HeliumI_Gas/LiquidDry-Ice/AcetoneAGasChromatograph-MassSpec.SeparatorWaterTrap1. f31)metrapLiquidNitrogen2.ReIeasingArsinesWaterbathFigure4.1Hydridegeneration-gaschromatography-massspectrometryexperimentalsetup.Dottedlinesrepresentgaslines.Vector 2) interfaced to the mass spectrometer. The HG-GC-MS experimental conditionsused are listed in Table 4.1. These conditions were arrived at after optimizing thehydride generation efficiency, the chromatographic resolution, and the massspectrometric sensitivity for the arsines of interest.Table 4.1 HG-GC-MS experimental parametersHydride Generation2.0% NaBH4solution, 4.0 M acetic acid, 3 mL sampleGas ChromatographInitial temperature: 50°CRamp rate: 400C. mm4Final temperature: 150°C for 2 mmHelium flow rate: 35 mIJminMass SpectrometerInterface temperature: 140°CIon source temperature: 160°CScan range: m1z74-130Scan time: Scan every 0.1 s for 4 miii794.2.4 Experimental proceduresAqueous solutions of the appropriate arsenical and L-methionine-methyl-d3,filter sterilized (0.2 p.m membrane) separately, were added to the autoclaved growthmedium (see Table 3.1). The initial concentration of each of the arsenicals was 1 ppm(in arsenic) and L-methionine was 1.3 mM. A 10 mL aliquot of an actively growingculture of A. humicola was added to the growth medium, and the day of addition isreferred to as day 0. A control experiment (without the addition of methionine) was alsocarried out. The incubation conditions, and the sampling of cultures were handled in thesame manner as those previously described in Chapter 3. The transfer of the CD3 groupfrom L-methionine-methyl-d3 to the arsenic substrates arsenate, arsenite, MMA, andDMA was investigated by examining the mass spectra of the arsines obtained by usingHG-GC-MS.4.3 RESULTS AND DISCUSSION4.3.1 HG-GC-MS measurementsHydride generation has proved to be an effective tool for detecting trace levels ofarsenate, arsenite, MMA, DMA and TMAO in samples.-391 These arsenicals areprecursors to the arsines that are ultimately detected by using atomic absorptionspectrometry (AA),1391’and mass spectrometry (MS).’6’1396 The reduction ofthese hydride-forming arsenicals to their corresponding arsines by sodium borohydrideis a pH dependent process. Under pH < 1 conditions, all hydride forming arsenicals canbe reduced to the arsines. At pH >4 arsenite, but not arsenate, forms arsine.The majority of previous studies reporting the use of a mass spectrometer as a80detector for arsines have used the selected-ion-monitoring (SIM) mode forcharacterizing arsine, methylarsine, dimethylarsine or trimethylarsine. In many cases themost abundant ions of each arsine were used for detection in the SIM mode, e.g. m/z 78[AsH3]and 76 [AsH1 for arsine; m/z 92 [CH3As],90 [CH3As] and 76 [AsH]for methylarsine; m/z 106 [(CH3)2AsH1 and 90 [CH3As] for dimethylarsine; m/z 120[(CH3)As], 105 [(CH3)2As] and 103 [(CH2)As] for trimethylarsine. Odanaka etal.’69 were able to quantify arsines with detection limits between 10-20 ng of arsenicby using a SIM mode GC-MS system and an off-line hydride generation heptane coldtrap procedure. Later, in a study involving the analysis of arsenic in NaOH digestedtissue extracts of shellfish, fish, crustaceans, and seaweeds, Kaise et al.’39 collected thearsines, generated by NaBH4 reduction, in a liquid-nitrogen trap coupled to a GC-MSsystem. The SIM mode gave detection limits of 0.3 ng arsenic. Norm et al.11-6 reportedthe existence of TMAO in fish tissue based on analysis carried out on a HG-GC-MSsystem equipped with a liquid nitrogen trap for arsine collection.In the present work, the mass spectrometer was set up to scan from mlz 74 to130 at 1 scan per 0.1 s. Scans above m/z 130 were not made since the highest m/z forany anticipated volatile arsine is 129, corresponding to (CD3)As. The gaschromatogram resulting from a mixture of arsines which were generated from a solutioncontaining synthetic standards of arsenite, MMA, DMA, and TMAO, is shown in Figure4.2. The mass spectra of each of the arsines AsH3,CH3As2,(CH3)2As , and(CH3)As is shown in Figure 4.3. The wide-scan monitoring mode was used instead ofthe SIM mode because the former would allow for the observation of any unsuspectedfragment ions resulting from the deuterium labeled arsines. This precaution was takenbecause the fragmentation patterns of deuterium labeled arsines apart from (CD3)As,were unknown. Cullen et al.33 demonstrated that the mass spectrum of (CD3)As has81550000 --___________________________500000 AsH3(CH3)2As450000400000350000 CO2 CH3As2ci) 300000 (CH3)As2500003 200000150000100000500000-0.00 0.50 1.00 1.60Retention Time (miii.)Figure 4.2 HG-GC-MS chromatogram of arsine, monomethylarsine, dimethylarsine, andtnmethylarsine. Solutions of standard arsenite, MMA, DMA, and TMAO (100 ng ofarsenic for each compound) were used.8276 (a)90%180%-.• 70%4’ 60% [AsH3]t:. 50%• 40%-.30%20%10%0%-——-—-70 80 90 100 110 120 130M/Z819 (b)90%80%] [CH3AsH2]t.—çj 70%j60%-j5%1—• 40% 7630%20%10%70 80 90 100 110 120 130MIZFIgure 4.3 Mass spectra of arsines derived from standard solutions of arsenite, MMA,DMA, and TMAO. (a) arsme AsH3, (b) monomethylarsine CH3As2,(c) dimethylarsine(CH3)2As , and (d) trimethylarsine (CH3As.8390%80%70%60%— 50%. 40%? 30%20%10%90%80%70%-i•; 40%30%20%’10%140 110M/Z9) (c)i[(CH3)2AsH]t169675Q-70 80 90 100 110 120 130M/Z103758910 (d)[(CH3)As]tI-70 80 90 120 13084an identical fragmentation pattern to that of (CH3)As except that H is replaced by D.The main fragmentation path is as shown in Figure 4.4. Cullen et aL33 also prepared(CD3)As(CH2for comparison purposes, and essentially the same fragmentationpattern is obtained. Fragment ions formed from the loss of both CD3 and CH3 from theparent ion are observed affording [(CH3) DAs]+ and [(CH3)2As] in high relativeabundance.33 Although the mass spectra of standards CD3AsH2and (CD3)2AsH arenot available, it is believed that their fragmentation patterns are similar to those ofCH3As2,and (CH3)2As . The other reason for acquiring spectra in the wide-scanmode was to establish the identity of all hydride generated volatile compounds resultingfrom the cell culture samples because any interfering ions might lead to erroneousresults in the SIM mode.By using the HG-GC-MS methodology described above it was possible to detectdown to 25 ng of arsenic. In the SIM mode the detection limit could be easily improvedby at least an order of magnitude.CD2 Xs —D2 CD3As 4-C2D5 (CD3)As.m/z91 ni/z95 [M]mJz 129—CD3CD-D2 + —D +2 ACD22 (CD3)2AsmfzlO3 m/z107 mIzlllFigure 4.4 The predicted fragmentation pattern of (CD3)As.3854.3.2 Characterization of methylated arsenic intermediates in the growth mediumof A. humicolaIn order to verify that S-adenosylmethionine is the methyl donor of theextracellular arsenic metabolites produced by A. humicola, 1 ppm each of arsenate,arsenite, MMA, and DMA were added separately to the culture of A. humicola growingin the presence and absence of L-methionine-methyl-d3.The arsenic metabolites in thegrowth medium were analyzed by using HG-GC-MS in the wide-scan mode.4.3.2.1 Transformation of arsenateWhen arsenate (1 ppm) was added to the growth medium in the absence of Lmethionine-methyl-d3,three arsenic compounds were found to be present in the growthmedium after 5 days of incubation. Positive identification of the volatile arseniccompounds generated by borohydride treatment was made based on a comparison oftheir retention times and mass spectra with those exhibited by standards (Figures 4.2 and4.3). Figure 4.5 shows the HG-GC-MS chromatogram of the sample collected on day 10.The first peak was identified as C02, which is produced probably during hydridegeneration. Its mass spectrum is shown in Figure 4.6. The second peak was identified asarsine AsH3,derived from arsenite that was produced by A. humicola from the arsenatesubstrate as described in Chapter 3. The third peak was (CH3)2As (Figure 4.7a)derived from DMA, and the fourth peak was (CH3)As (Figure 4.8a) derived fromTMAO. The mass spectra of these arsines are identical with those of the arsines derivedfrom standard arsenicals (Figure 4.3).In the presence of L-methionine-methyl-d,dimethylarsenic and trimethylarsenicspecies were also detected in the growth medium. No apparent difference in the quantity86of these methylated arsenic intermediates was observed when L-methionine-methyl-d3was either present or absent from the medium. ft has been proposed that the addedmethionine could increase the cell’s internal concentration of SAM and thus mightenhance the methylation process.45 This assumption was based on some results thatshowed high methyl incorporation in arsenicals in the presence of methionine.33However, Cullen et al.45 also reported that preculturing whole cell cultures of A.humicola or their cell-free extracts with methionine did not effect the amount of themethylated arsenic species produced from methylarsine oxide (CH3AsO).The mass spectra of the dimethylarsenic (Figure 4.7b) and trimethylarsenic species(Figure 4.8b) which were produced by HG-GC-MS from cultures grown in the presenceof L-methionine-methyl-d3,exhibit additional ions, strongly indicating that a considerableamount of CD3 was incorporated onto arsenic. The mass spectrum of the dimethylarsenicderivative contains additional ions at m/z 112 [(CD3)AsH]t m/z 109[(CH3) DAsH1, and m/z 93 [CDAs]. The mass spectrum of the trimethylarsenicspecies contains additional ions at m/z 129 [(CD3)As], m/z 126 [(CH3) DAs],m/z 123 [(CH)CDAs], m/z 111 [(CD3)2As], m/z 108 [(CH3) DAs], and mlz93 [CDAsJ. These ions are absent from the HG-GC-MS traces resulting from themedia containing no L-methionine-methyl-d3(Figure 4.7a, 4.8a).The identification of deuterated arsenic metabolites in the growing cultures of A.humicola is noteworthy, as this is the first time that methionine has been demonstrated tobe involved in the production of the non-volatile methylarsenic intermediates which havebeen proposed in Challenger’s pathway (Figure 1.1). As described above, the labeledmethyl group from methionine had been shown to be transferred to arsenate, arsenite,MMA, and DMA to form labeled volatile trimethylarsine.33’48 Our work stronglyreinforces the suggestion that S-adenosylmethionine or some related sulphonium87300000025000002000000c)15000001000000500000Figure 4.5 HG-GC-MS chromatogram of a 2 mL culture medium sample that was takenafter 10 days of incubation. The medium was originally enriched with 1 ppm arsenate, butnot with L-methionine-methyl-d3Retention Time (mm.)8890%80%70%. 60%50%40%30%20%10%760%- I70 80 90 100 110 120 130MJZFigure 4.6 Mass spectrum of the first peak in Figure 4.5, identified as CO2.89100%-9090% (a)80% F[(CH3)2ASH]t_ 7Q% r19660%— 50%40%—— 30%20%1(1110%75_________-. I_____________70 80 90 100 110 120 130M/Z— I91090% 1 (b)80%70% [(CH3)2AsH]t60% 1650%— [(CH3) DAs ]40%[CD3As]/30% 120% 93I JJ 19110% 4270 80 90 100 110 120 130M/ZFigure 4.7 Mass spectra of the hydride derivative of the dimethylarsenic species presentin the growing culture, enriched with arsenate, after 10 days of incubation. (a) in theabsence of L-methionine-methyl-d3, (b) in the presence L-methionine-methyl-d390100%- I___________________________13 10 (a)90%80% [(CH3JM].+70%.—4 60%—..89— 50%40%.——.30%20%10%75T70 80 90 100 110 120 130MJZ100%- I________-(b)90% [(CH3)As)+ •80% \ .70%+10.—60% C.) +.89— 50%40%+ C.).18—’ —‘ ——30%20% 126ç)I%./10% 75 I0%- JI __j jJ70 80 90 100 110 120 130MIZFigure 4.8 Mass spectra of the hydride derivative of the trimethylarsenic species presentin the growing culture, enriched with arsenate, after 10 days of incubation: a) in theabsence of L-methionine-methyl-d3, b) in the presence L-methionine-methyl-d391compounds are the source of the [CH3]+shown in Challenger’s pathway.The percentage of the CD3 incorporation was determined by comparing therelative intensities of the parent ions, mlz 106 [(CH3)2AsHI, 109 [(CH3) DAsHj,and 112 [(CD3)AsH1 for the dimethylarsenic species, and m/z 120 [(CH3)As], m/z123 [(CHCDAs], m/z 126 [(CH3) DAs], and m/z 129 [(CD3)As] for thetrimethylarsenic species. The distribution change of the CD3 label in the dimethyl andtrimethylarsenic metabolites with incubation time is presented in Table 4.2. The relativestandard deviations of the peak intensity of m/z 106, and m/z 120 for 4 determinations ofstandard dimethylarsinic acid and trimethylarsine oxide is 6%. For these calculations, weassumed that the responses of the deuterated arsines are identical to the responses of theundeuterated arsines; however, this assumption needs to be verified by using theappropriate deuterated methylarsenic standards. Consequently, the results presented inTable 4.2 should only be viewed as an indication of low or high incorporation.4.3.2.2 Transformation of arseniteFollowing the addition of arsenite to the growing culture of A. humicola,dimethylarsenic and trimethylarsenic species were detected in the growth medium after 5days of incubation. The concentration of the arsenic metabolites was found to beindependent of the presence or absence of L-methionine-methyl-d3. In the absence of Lmethionine-methyl-d3,the HG-GC-MS traces of the dimethylarsenic and thetrimethylarsenic species found in the medium were similar to those of the standard(CH3)2As and (CH3)As. While in the presence of L-methionine-methyl-d3, the massspectra of the two arsenic species showed the incorporation of the CD3 group asindicated by the ions at m/z 112, 109, and 93 for the dimethylarsenic species, and mlz92Table 4.2 Percentage distributiondetected in the growth mediumof dimethylarsenic and trimethylarsenic compoundsIncubation Time (days)Arsenic metabolites day 5 day 10 day 15 day 20L.O.D.: limit of detection.Arsenate (1 ppm) substrate303733552811647 5628 2625 1861 6328 278 73 3Arsenite (1 ppm) substrate(CH3)2As(CH)(CDAs(CD32AsH(CH3)As(CH2D)As(CH3)(CDAs(CD3)As(CH3)2As(CH)(CDAs(CD32AsH(CH3)As(CH2D)M(CH3)(CDAs(CD3)As(CH3)2As(CH)(CDAs(CH3)As(CH2D)As(CH3)(CDAs(CH3)As(CH’)9D)As51 45 51 7726 30 28 1423 25 21 945 50 52 5336 34 33 3319 12 12 12<L.O.D. 4 3 2MMA (1 ppm) substrate33 66 62 6467 34 38 3642 53 58 6031 29 28 2827 18 14 12DMA (1 ppm) substrate594166346931712993129, 126, 123, 111, 108, and 93 for the trimethylarsenic species. The distribution of theCD3 label in the dimethyl and trimethylarsenic metabolites with incubation time ispresented in Table 4.2.4.3.2.3 Transformation of methylarsomc acidApiotrichum humicola transformed MMA into dimethylarsenic andtrimethylarsenic species both in the presence and absence of L-methionine-methyl-d3inthe growth medium. Again, the presence of methionine did not significantly affect theproduction of arsenic metabolites. Figure 4.9 shows that a large amount of DMA andTMAO is present in the growing culture collected on day 10 (in the absence of Lmethionine-methyl-d3).The mass spectra of their hydride derivatives are presented inFigure 4. lOa and 4.1 la, respectively.When deuterated methionine was added to the growth medium, the CD3 groupfrom methionine was incorporated into MMA to form deuterated dimethyl andtrimethylarsenic species. The mass spectra clearly show the presence of ions at m/z 109[(CH3) DAsHj, and m/z 93 {CDAsJ for the dimethylarsenic species (Figure4.lOb), and m/z 126 [(CH3) DAsJ, m/z 123 [(CH)CDAs], m/z 111[(CD3)2As], mlz 108 [(CH) DAs], and m/z 93 [CDAs] for the trimethylarsenicspecies (Figure 4.1 ib). The absence of ions at m/z 112 [(CD3)AsH] fordimethylarsenic species and mIz 129 [(CD3)As] for trimethylarsenic species isexpected since only one and two deuterated methyl groups can be incorporated intoMMA to form dimethylarsenic and trimethylarsenic species, respectively. The massspectrum of monomethylarsenic species does not show the presence of ion at m/z 95[CD3AsH2].The absence of ions at m/z 129, 112, and 95 indicates that the cleavage of943000000 I ICH3As2250000020000001500000(CH3)2As1000000 (CH3)As50000:,.,.S’I,0.00 0.50 1.00 1.50Retention Time (mm.)Figure 4.9 HG-GC-MS chromatogram of a 2 mL culture medium sample that was takenafter 10 days of incubation. The medium was originally enriched with 1 ppm MMA, butnot with L-methionine-methyl-d3.95100% -I (a)90%80%. 70% [(CH3)2AsHJt1660%— 50%.— 40%— 30%20%7510%_________________I____________________9670 80 90 100 110 120 130M/Z9)90% V(b)80% -.— 70%60%[(CH3)2AsH]t— 50%•[(CH3) DAS It.— 40%— [CD3As] V3Q% / 1920%1110% I___________0%- V I70 80 90 100 110 120 130M/ZFigure 4.10 Mass spectra of the hydride derivative of the dimethylarsenic species presentin the growing culture, enriched with MMA, after 10 days of incubation. (a) in theabsence of L-methionine-methyl-d3, (b) in the presence L-methionine-methyl-d3961 flfl.88 13 (a)80% [(CH3)As]t70% rI—60% 76-.50%-a?‘ 40%30%a?20% I10%91O%J 1 -______I70 80 90 100 110 120 130M/ZlOvi- I_____10590% [(CH3)As]t(b)80%I —70%+ 120<-II < —60%a?—.. x50%— c)a?ii. 40% +—“ 1830% 88126a?20%10% I ill 1110%---_—_70 80 90 100 110 120 130M/ZFigure 4.11 Mass spectra of the hydride derivative of the iiimethylarsenic species presentin the growing culture, enriched with MMA, after 10 days of incubation. (a) in theabsence of L-methionine-methyl-d3, (b) in the presence L-methiornne-methyl-d397the H3C-As bond is insignificant. The distribution of the CD3 label in the arsenicmetabolites is shown in Table 4.2.4.3.2.4 Transformation of climethylarsinic acidThe biotransformation of DMA to trimethylarsenic species in the absence of Lmethionine-methyl-d3by the microorganism A. humicola is a slow process compared tothe formation of trimethylarsenic species from arsenate, arsenite, and MMA. Most of theDMA substrate remained unchanged in the growing culture after 20 days of incubation(Figure 4.12). However, a small amount of a trimethylarsenic species was produced, andits HG-GC-MS spectrum is presented in Figure 4.13a. In the presence of L-methioninemethyl-d3,the production of the trimethylarsenic species did not seem to be enhanced.The mass spectrum of this arsenic species (Figure 4.13b) exhibits ions at mlz 123[(CH3)2CDAs], m/z 108 [(CH3) DAs], and m/z 93 [CDAs], indicating thatthe CD3 group is incorporated into DMA. Not unexpectedly, the mass spectrum of thetrimethylarsenic species does not show ions present at m/z 129 [(CD3)As], m/z 126[(CH3) D2As], and 111 [(CD3)2As], suggesting that H3C-As bond cleavage isnot significant. The distribution of the CD3 label in the trimethylarsenic species is shownin Table 4.2.In conclusion, the results obtained from this study strongly suggest thatmethionine, or SAM, is the source of the methyl groups in the biological methylation ofarsenic in the microorganism A. humicola, and thus conform with the oxidation-reductionpathway involving carbomum ions suggested by Challenger.30’43983500000 ---‘- .I--i(CH3)2As30000002500000hiC 20000000150000001000000500000 (CH3)As0.00 0.50 1.00 1.60Retention Time (mm.)Figure 4.12 HG-GC-MS chromatogram of a 2 mL culture medium sample that was takenafter 20 days of incubation. The medium was originally enriched with 1 ppm DMA, butnot with L-methionine-methyl-d3.99100%-13 190%o (a)80% [(CH3)As]t. 70%60%— 7589— 50%40%30%20%10%I, —-70 80 90 100 110 120 130M/ZI_________-1O0% (b)90% [(CH3)As]t E10 L.80%•_ 70%60% ÷C 50% 89c)• 40%— 13— 30%20%7510% II I —_ ___ __________ __—___ ____—70 80 90 100 110 120 130M/ZFigure 4.13 Mass spectra of the hydride derivative of the trimethylarsenic species presentin the growing culture, enriched with DMA, after 20 days of incubation. (a) in theabsence of L-methionine-methyl-d3, (b) in the presence L-methionine-methyl-d3100CHAPTER 5THE BIOMETHYLATION AND BIOACCUMULATION OF ARSENICALS BYA MARINE ALGA POLYPHYSA PENICULUS5.1 iNTRODUCTIONArsenic is probably one of the better known elements because of the toxicproperties of some of its compounds. Against this background it is not surprising thatreports, at the beginning of this century, of high levels of arsenic in marine organismsattracted much interest. Since then, concerns regarding the forms of arsenic in marineorganisms, their toxicity, how they were accumulated, and what role they played in thebiochemical functions of the organisms have been investigated. But even now, ourknowledge of marine arsenic chemistry is incomplete.Marine algae have been the subject of many arsenic metabolic studies because oftheir ecological and nutritional importance. Such studies of the interaction of marinealgae with arsenicals are relevant because arsenic compounds produced by algae aregenerally believed to be the source of the arsenic compounds found in marine animals,although it is not well established how and when these transformations take place.A number of investigators have reported that the total arsenic concentrations in arange of marine algae have much higher levels than seawater.58M Arsenate, thepredominant form of arsenic in seawater, is readily taken up by algae, possibly becauseof its similarity to the essential phosphate.65-7Once inside the algal cell, the arsenic isbelieved to be transformed by algae into a range of arsenic compounds, includingMMA, DMA and other more complex aqueous-soluble or lipid-type arsenicals.69-80Several workers have been successfully isolated and characterized arsenosugar101derivatives from different seaweeds.81-89 Many people now believe that the presence ofarsenosugar derivatives in marine algae is ubiquitous.Edmonds and Francesconi8°have proposed a pathway for the biotransformationof arsenate by marine algae as shown in Figure 1.5. In order to provide support for thisproposed mechanism a number of questions need to be addressed, such as: (1) whetherthis mechanism applies to all species of marine micro- and macroalgae; (2) whether themethylation of arsenic follows Challenger’s proposed mechanism and involve theutilization of SAM as a donor of both methyl and adenosyl groups. We believe thatexperiments with controlled cultures may provide some of this information.As mentioned in Chapter 1, there are only a limited number of reports thatdiscuss the biotransformation of arsenicals by macroalgae in culture medium.20’923Complex arsenic compounds are reported to be produced by these algae, but they havenot been positively identified as arsenosugars.Studies on arsenic biotransformation by marine phytoplankton, mainlyunicellular marine phytoplankton, only confirm the production of arsenite, MMA, andDMA from arsenate.69’7024The presence of complex arsenic compounds is alsoreported although no positive identification of these compounds can be made.We have chosen to study Polyphysa peniculus, a unicellular marine alga whichhas been cultivated in the laboratory of Dr. L. G. Harrison in artificial seawater understerile conditions for more than two decades. Its cells are unusually large (4-5 cm inlength, 0.4 mm in diameter), but, like many phytoplankton, it is a unicellular alga(Chiorophyta). In this chapter we report on the effect of adding arsenate, arsenite,methylarsonate and dimethylarsinic acid to artificial seawater containing P. peniculus.Arsenic accumulation, methylation and excretion by the alga were examined by usinggraphite furnace atomic absorption spectrometry (GFAA) and hydride generation-gas102chromatography-atomic absorption spectrometry (HG-GC-AA) methodology. Theinability of P. peniculus to synthesize significant amounts of complex water and lipid-soluble arsenic compounds was also established by using flow injection-microwavedigestion-hydride generation-atomic absorption spectrometry methodology.We also report on the effect of adding L-methionine-methyl-d3 and arsenate (1ppm) to artificial seawater containing P. peniculus. The arsenic metabolites excreted bythe alga in the growth medium, principally DMA, were identified by using hydridegeneration-gas chromatography-mass spectrometry (HG-GC-MS). These resultsprovided conclusive evidence of CD3 incorporation from L-methionine-d3 in thedimethylarsenic species produced by P. peniculus, and supports the hypothesis that Sadenosylmethionine is the biological methyl donor.5.2 EXPERIMENTAL5.2.1 ReagentsStandard solutions of arsenic compounds were freshly prepared by serial dilutionfrom stock solutions (1000 ppm of elemental arsenic) as described in Section 2.2.1.Solutions of 1.5% (wlv) potassium persulphate (BDH) in 0.1% (wlv) NaOH, 1.0 MHC1, 4.0 M acetic acid, and 2.0% (wlv) NaBH4in 0.1% (wlv) NaOH were freshly madedaily. A solution of L-methionine-methyl-d3CD3SCH2H(NH)COO (Aldrich)was prepared by dissolving the compound in deionized water.1035.2.2 Algal culturesThe alga used throughout the experiments was Polyphysa peniculus(Dasycladales, Chlorophyta), a marine alga closely related to the better known genusAcetabularia, and sometimes known as A. cl4ftonii and several other synonyms.195 TheP. peniculus culture used in this work has been maintained in sterile artificial seawater(Shephard’s medium,’96pH 7.9-8.3) for more than two decades, first in the laboratoryof Dr. B. R. Green (Botany, this university) and then in the laboratory of Dr. L. G.Harrison. The composition of Shephard’s medium is shown in Table 5.1. The culturehad not been rendered anexic, but had been maintained and handled by using steriletechniques, and treated with antibiotics (a mixture of amphotericin, penicillin G, andkanamycin) if any bacterial infection arose.5.2.3 Instrumentation5.2.3.1 Graphite furnace atomic absorption spectrometryTotal amount of arsenic accumulated in cells of P. peniculus were measured byusing graphite furnace atomic absorption spectrometry methodology as previouslydescribed in Chapter 2. The standard additions technique was used in these studies. Thefurnace operating parameters and the conditions for the standard additions method arefound in Chapter 2 (Tables 2.1 and 2.2).104Table 5.1 Composition of Shephard’s mediumNutrients and Trace Metals in 1 L MediumNaC1 24 g NaNO3 0.04 gMgSO4.7H20 12 g Na2HPO4 0.001 gTris(base)(HOCH)3CN 1 g Micronutrients 10 mLCaC12.2H0 1 g NaHCO3 0.1 gKC1 0.75 g Vitamins 0.2 gMicronutrient Salts Stock Solution (1 L)Na2EDTA 1.2 g MnC12.4H0 0.02 gZnSO4 0.2 g CoC12.6H0 0.2 mgNa2MoO.2H0 0.1 g CuSO4.5H20 0.2 mgFeC13.6H20 0.05 g H3B0 0.1 gVitamin Stock Solution (100 mL)Thiamine HC1 0.3 g p-Aminobenzoate 20 mgVitamin B 12 4 mg Calcium Pantothenate 10 mg1055.2.3.2 Hydride generation-gas chromatography-atomic absorption spectrometryA hydride generation system was used for arsine production and collection aspreviously described in Chapter 2. After the sample introduction was completed, thearsines trapped in liquid nitrogen were volatilized when the hydride trap was warmed ina water bath (700C). By using a Hewlett Packard Model 5830A gas chromatograph witha pre-set temperature program the arsines were then separated on a Porapak-PS column(80-100 mesh), atomized by using a hydrogen-air flame in a quartz cuvette, and detectedby using a Jarrell-Ash Model 810 atomic absorption spectrometer equipped with aVarian Spectra arsenic hollow-cathode lamp set at 10 mA. The monochromator was setat 193.7 nm, and the slit width at 1 nm. Absorbance was recorded as peak area by usinga Hewlett-Packard 3390A integrator.5.2.3.3 Flow injection-microwave digestion-hydride generation-atomic absorptionspectrometryThe flow injection-microwave digestion-HGAA system (Figure 5.1) describedby Le et al.,197498 was used to determine non-hydride-forming, hiddenu arseniccompounds. The evolved arsines were carried into an open-ended T-shape quartz tube(11.5 cm long x 0.8 cm i.d.) which was mounted in the flame of a Varian Model AA1275 atomic absorption spectrometer equipped with a standard Varian air-acetyleneflame atomizer. The signals were recorded on a Hewlett-Packard 3390A integrator.106PSRABFigure 5.1 A schematic diagram of an on-line coupled flow injection-microwave ovendecomposition-hydride generation system.S -- Sample flow R — Decomposition reagent flow A -- Acid flowB — NaBH4flow V — Sample injection valve P— Peristaltic pumpMO -- Microwave oven Ti & T2 -- T-joints TB -- Ice water bathG — Carrier gas (N2) D -- To detector (AAS)MO lB DG1075.2.3.4 Hydride generation-gas chromatography-mass spectrometryThe hydride generation-gas chromatography-mass spectrometry system (HGGC-MS) was used for characterization of the methylated arsenic species in the growthmedium as previously described in Chapter 4. The experimental conditions arepresented in Table 4.1. In this study, the mass spectrometer was scanned from m/z 74 to115 at 1 scan per 0.1 second.5.2.4 Experimental proceduresThe alga (approximately 0.6 to 1 g dry weight) was added to sterile Erlenmeyerflasks (2 L) each containing 1 L of sterile Shephard’s medium. Arsenicals were filtersterilized (0.2 pm membrane), and added to the medium separately. Two concentrations,10 ppm and 0.9 ppm, were employed in our studies. During the growth period thecultures were maintained at 20°C. Fluorescent lamps that gave 3,200 lux intensityaround the flasks were used as the light source, and the light:dark cycle was 16:8 hours.Once each day the culture was agitated and 10 mL aliquots of the medium wereremoved and frozen prior to analysis. The alga was harvested on day 7 and thoroughlyrinsed with sterile Shephard’s medium. A half portion of the alga was freeze-dried andstored in a freezer for future analysis. The rest of the alga was transferred to a 1 L sterileErlenmeyer flask containing 500 mL fresh arsenic-free sterile Shephard’s medium. Theincubation conditions were not changed and the day of transfer is referred to as day 0 ofthe second cycle. The culture was handled in the same manner as in the first cycle, themedium was sampled each day, and the alga was again harvested after 7 days ofincubation, rinsed, freeze-dried, weighed, and stored in a freezer. The growth of thecells in a 7 days period was limited as indicated by the length of the cells.108The possibility of non-metabolic interactions between arsenic and cells of P.peniculus was studied by exposure of heat-treated or 4.0% formalin-treated cells (4.0%formalin is an effective reagent for killing cells but causes only minimum damage totissue integrity92’3)to arsenate emiched medium. The culture medium was collectedonce each day for 7 days, and frozen prior to analysis.The total amount of arsenic in the cells was determined following acid digestion.Freeze-dried cells (50-100 mg) were dissolved in 1 mL of concentrated HNO3 (69.0-71.0%), left overnight, then boiled with 1 mL of H20 for 5-10 minutes prior toanalysis. The resultant pale-yellow transparent solution was neutralized with NaOHsolution, diluted to an appropriate volume, and subjected to GFAA analysis. The amountof arsenic was quantified by using the standard additions technique. Palladium nitrate(100 ppm), prepared in citric acid (2.0% wlv), was used as a modifier.For the arsenic speciation analysis, freeze-dried algal cells (0.1-0.2 g) wereweighed and transferred into an Erlenmeyer flask (250 mL) containing 30 mL of mixedsolvent, CHC13JMeOHJH2O(1:1:1). The mixture was sonicated for 2 hours and thenstoppered with a solid rubber plug and left on a mechanical shaker for 24 hours. It wasthen centrifuged and the residue was re-extracted with 10 mL of the mixed solvent foranother 24 hours. The extracts were combined and centrifuged to separate the aqueousfraction with the organic fraction. The colorless aqueous layer was kept at 40C prior toanalysis. The organic extract and the residue were air dried, digested in 4 mL of 2.0 MNaOH in a water bath at 95°C for 3 hours, and fmally neutralized with 6.0 Mhydrochloric acid prior to analysis.The HG-GC-AA system was used to analyze hydride-forming arsenicals in eachof the three fractions of the cell extracts as well as in the growth medium. The flowinjection-microwave digestion-HGAA technique was used to detect the existence of109“hidden” arsenic in the cells.To investigate the utilization of L-methionine-methyl-d3 in the biomethylationprocess of P. peniculus, arsenate and L-methionine-methyl-d3, filter sterilizedseparately, were added to the growth medium. The initial concentration of arsenate andL-methionine-methyl-d3was 13 jiM (1 ppm) and 130 jiM, respectively. The day ofarsenate addition is referred to as day 0. Four separate experiments were carried out. Forthe first experiment, only arsenate was added to the growth medium. In the secondexperiment L-methionine-methyl-d3and arsenate were inoculated in the medium at thesame time. The third experiment was carried out by adding L-methionine-methyl-d3three days prior to the addition of arsenate and for the fourth experiment L-methioninemethyl-d3was not added until three days after the arsenate. Once each day, until day 11,the culture was agitated and a 10 mL of culture aliquot was removed and stored frozenprior to the HG-GC-MS analysis.5.3 RESULTS5.3.1 The accumulation of arsenicals in cells ofPolyphysapeniculusTwo arsenic concentrations, 10 ppm and 0.9 ppm, were employed in this study.The total amount of arsenic accumulated in P. peniculus was determined by usingGFAA. The results are presented in Table 5.2. With the exception of arsenate, algaexposed to 10 ppm arsenicals accumulated more arsenic in their cells than those exposedto 0.9 ppm arsenicals. In the presence of MMA, the arsenic accumulation in the cellswas very low.110Table 5.2 Total amount of arsenic in cells of P. peniculus determined by using GFAA(igIg, dry weight)Arsenic Cells treated with 10 ppm Cells treated with 0.9 ppminoculated arsenicals arsenicalsA Bb BbArsenate 36.0±2.5C 11.5±0.7 43.3±2.4 7.7±0.5Arsenite 52.7±4.2 5.6±0.5 17.4±1.2 3.5±0.3MMA 18.3±1.1 4.4±0.3 3.1+0.2 traceDMA 34.8±1.9 5.3±0.4 25.1±1.8 3.1±0.3a 7 days after the cells were exposed to arsenicalsb 7 days after the cells were transferred to fresh mediac large fraction of dead cells was present within 2 days of incubationArsenate at high concentration (10 ppm) was more toxic to P. peniculus asindicated by the large percentage of dead cells found within 2 days of initiating theincubation. Certainly, high concentrations of arsenate affect the arsenic accumulatingability of the alga as shown in Table 5. 2.Only small amounts of arsenic compounds were retained in the cells after thealga was transferred to an arsenic-free medium for 7 days (Table 5.2).5.3.2 Arsenic speciation analysis in cells of Polyphysa pernculus5.3.2.1 Water/methanol extractsMost of the arsenic accumulated in cells of P. peniculus is present in thewater/methanol fractions (> 90%) after the solvent extraction. The hydride-forming111arsenicals, arsenate, arsenite, MMA, and DMA in the aqueous extracts were detected asAsH3,CH3As2and (CH3)2As , respectively, by using HG-GC-AA. TMAO, whichwould have been detected as trimethylarsine, was not found in any of the samples. Thesearsenic speciation results are presented in Table 5.3. As expected the total amount ofhydride-forming arsenicals in the alga is proportional to the external arsenicconcentrations applied.A large amount of inorganic arsenic, mainly arsenite (25.1 ppm), was detected incells incubated with 10 ppm arsenate. Small amounts of DMA (1.9 ppm) were alsofound in the same sample. When P. peniculus was treated with 0.9 ppm arsenate, theamount of DMA found in the aqueous extract of the alga was as high as 21.7 ppm, andis about 45% of total hydride-forming arsenicals found in the aqueous extract.Substantial amounts of arsenite were also detected in both extracts. Neither MMA norarsenate were found in cells incubated with 0.9 ppm arsenate. When the alga wastransferred to an arsenic-free medium and incubated for 7 days, only small amounts ofarsenate were retained.Four arsenicals, principally arsenite, but also arsenate, MMA and DMA werefound in the aqueous extracts after the alga was treated for 7 days with 10 ppm and 0.9ppm arsenite. The percentage of MMA in both aqueous extracts is about 7%. Thepercentage of DMA in cells exposed to 0.9 ppm arsenite is 37% of total aqueoushydride-forming arsenicals. Little DMA was detected after the alga was treated with 10ppm arsenite. The arsenical content of the alga was greatly reduced after the alga wastransferred to an arsenic-free medium.Similar amounts of MMA and DMA were found in cells incubated with 10 ppmand 0.9 ppm MMA. However, after the alga was transferred to an arsenic-free medium,small amounts of arsenicals remained in cells originally treated with 10 ppm MMA, but112Table 5.3 Arsenic distribution in aqueous extracts of the cells determined by usingHG-GC-AA (.tgIg, dry weight)Arsenic Arsenic species Cells treated with 10 ppm Cells treated with 0.9 ppminoculated found in cells arsenicais arsenicalsBb Aa BbArsenate 12.8±0.8 9.7±0.7 0 5.1±0.3Arsenite 25.1±1.3 0 26.3±1.6 0Arsenate MMA 0 0 0 0DMA 1.9±0.2 0 21.7±2.0 0Total 39.8 9.7 48.0 5.1Arsenate 8.0±0.5 0 1.8±0.1 0Arsenite 37.7±2.5 2.3±0.2 6.5±0.4 1.2±0.1Arsenite MMA 3.1±0.2 0.4±0.03 1.0±0.1 0DMA 0.8±0.1 0.8±0.08 5.4±0.5 0.8±0.1Total 49.6 3.5 14.7 2.0Arsenate 0 0.3±0.03 0 0Arsenite 0 0.9±0.07 0 0MMA MMA 5.3±0.3 0.8±0.06 1.4±0.1 0DMA 6.5±0.6 2.7±0.3 1.6±0.2 0Total 11.8 4.7 3.0 0Arsenate 1.6±0.1 0 0 0Arsenite 0 0 0 0DMA MMA 0.6±0.04 0. 1±0.01 0.7±0.04 0DMA 28.1±2.5 3.6±0.3 20.4±1.8 2.5±0.3Total 30.3 3.7 21.1 2.5a 7 days after the cells were exposed to arsenicalsb 7 days after the cells were transferred to fresh medianone were found in the cells exposed to 0.9 ppm MMA.Arsenic speciation analysis of cells exposed to DMA shows that the accumulatedarsenic exists mainly as DMA. Trace amounts of MMA were also present in the aqueousextracts. Most of the accumulated arsenic was discharged from the cells after the algawas transferred to an arsenic-free medium.113In order to determine if any “hidden” arsenic, possibly arsenosugars, existed inthe aqueous extracts of the algal cells, a flow injection-microwave digestion-HGAAtechnique was applied. In this methodology, the “hidden” arsenic species aredecomposed and oxidized by potassium persulphate to arsenate with the aid ofmicrowave radiation. 197,198 The product, arsenate, can easily be reduced to arsine.Thus, by comparison of the arsine absorbance before and after microwave assisteddigestion, the amount of total “hidden” arsenic in a sample can be calculated. The resultsare shown in Table 5.4: no significant differences are apparent in the amounts of arsenicdetected before and after microwave assisted digestion. This suggests that very smallamounts, if any, of “hidden” arsenic species such as arsenosugars, were produced andaccumulated in the cells during the growth cycle.Table 5.4 Arsenic disthbution in aqueous extracts of the cells harvested from arsenicalenriched media before and after microwave digestion (.tg/g, dry weight)Arsenic Cells treated with 10 ppm Cells treated with 0.9 ppminoculated arsenicals arsenicalsbefore digestion after digestion before digestion after digestionArsenate 37.3±2.6 38.7±2.7 45.4±3.2 47. 1±3.8Arsenite 48.7±2.9 50.6±4.1 15.9±0.8 14.7±0.9MMA 12.7±0.8 11.5±1.1 3.8±0.3 3.4±0.3DMA 32.2±2.1 33.1±2.2 20.4±1.2 20.0±1.81145.3.2.2 Chloroform extractsChloroform fractions from the original CHC13IMeOH/H20cell extracts were airdried, digested with 2.0 M NaOH, neutralized with concentrated HC1, and then analyzedby using HG-GC-AA. Arsenosugars, if present, would be decomposed to DMA underthese conditions and would be detected as dimethylarsine by using HG-GC-AA. Noarsenicals were detected in these CHC13 extracts of cells exposed to arsenate andarsenite. Only trace amounts of dimethylarsenic compounds were detected in cellsexposed to MMA and DMA. Arsenic compounds were not found in the CHC13 extractsfollowing transfer of the cells to arsenic-free media.The flow injection-microwave digestion-HGAA technique was used to detect ifany “hidden” arsenic compounds, which may not have been hydrolyzed by NaOH, werepresent in the organic extracts. There was no significant difference in the amount ofhydride-forming arsenicals present before and after microwave assisted digestion.5.3.2.3 Insoluble residuesThe residues were digested by using 2.0 M NaOH, and were subsequentlyanalyzed by using HG-GC-AA. Trace amounts of monomethylarsenic anddimethylarsenic compounds were detected in cells exposed to arsenite, MMA and DMA.The flow injection-microwave digestion-HGAA technique did not show the presence of“hidden” arsenic species in these digested samples.1155.3.3 Arsenic speciation analysis in the growth media of Polyphysa peniculusArsenic speciation analysis of the growth media was carried out by using HGGC-AA. The change in the chemical form of arsenic was very dramatic in the mediaenriched with arsenate. When the alga was exposed to 10 ppm arsenate, more than 70%of the substrate was reduced to arsenite after one day of incubation. The concentrationof arsenate and arsenite remained relatively constant until the alga was harvested (Figure5.2). Reduction of arsenate to arsenite was also observed in the growth medium of P.peniculus spiked with 0.9 ppm arsenate, but the reaction was slower (Figure 5.3). Theconcentration of arsenite reached a steady state after 2 days of incubation. In addition toarsenate and arsenite, DMA was detected in this culture medium on day 3. Theconcentration of DMA had increased slowly with time from 0 to about 0.15 ppm at thetime when the experiment was terminated. No MMA was found in either of theexperiments mentioned above. When the heat-treated or 4.0% formalin-treated cellswere exposed to 0.9 ppm of arsenate, neither arsenite nor DMA was observed in thegrowth medium.No MMA was detected in the growth medium when 10 ppm or 0.9 ppm arsenitewas used as a substrate. When the alga was treated with 10 ppm arsenite, about 90% ofthe arsenite in the medium remained unchanged throughout the incubation period.Arsenate was detected in this medium after two days of incubation, but at a low level.After incubation of P. peniculus with 0.9 ppm arsenite for 1 day, small amounts ofDMA were detected in the growth medium as well as arsenate (Figure 5.4). Theconcentration of DMA increased from 0 to 0.08 ppm after another five days ofincubation.When MMA or DMA was used as a substrate, the changes of arsenic species andarsenic concentration were minimal in the growth medium.11611109-.- 8E0cL 7C00U)320Incubation Time (days)Figure 5.2 The change of arsenic species in the growth medium with incubation time.The growth medium was enriched with 10 ppm arsenate before incubation. (0) arsenite(•) arsenate**The values are averages of four replicate detenninations by HG-GC-AA of separate aliquots (0.02inL) from the 6 mL sample taken at each time intervaL The error bars represent 90% confidenceintervals.1170 1 2 3 4 5 6 7 81 .00.40.30.20.1Incubation Time (days)Figure 5.3 The change of arsenic species in the growth medium with incubation time.The growth medium was enriched with 0.9 ppm arsenate before incubation. (0) arsenite(•) arsenate (V) DMA**The values are averages of four replicate determinations by HG-GC-AA of separate aliquots (0.2 mL)from the 6 mL sample taken at each time interval. The error bars represent 90% confidence intervals,and are smaller than the size of the symbols when they are not evident on the graph.0.90.80.70.60.50.00 1 2 3 4 5 6 7 81181 .00.90.80.7E0.30.20.10.0Incubation Time (days)Figure 5.4 The change of arsenic species in the growth medium with incubation time.The growth medium was enriched with 0.9 ppm arsenite before incubation. (0) arsenite(•) arsenate (Y) DMA**The values are averages of four replicate determinations by HG-GC-AA of separate aliquots (0.2 mL)from the 6 mL sample taken at each time interval. The error bars represent 90% confidence intervals,and are smaller than the size of the symbols when they are not evident on the graph.0 1 2 3 4 5 6 7 81195.3.4 Arsenic efflux studiesAfter the cells were exposed to arsenicals for 7 days, they were washed andtransferred to fresh arsenic-free media, and left for a further 7 days. Arsenic speciationanalysis in this “fresh” media was carried out by using HG-GC-AA. Figure 5.5 showsthat the accumulated arsenicals were rapidly excreted to the media by the cells, usuallywithin 1-2 days. The difference in the amounts of hydride-forming arsenicals in the cellsafter the first 7 days and after 14 days was compared with the amounts of hydride-forming arsenicals released in the medium during the second 7 days period. The resultsare, in general, in agreement, indicating that “hidden” arsenic species were not producedduring this period. This conclusion is also reinforced by the results obtained by using theflow injection-microwave digestion-HGAA technique.The amount of DMA found in the medium after 14 days from cells treated witheither 10 ppm arsenate or arsenite was higher than that detected in the cells before thetransfer. Consequently, the amount of inorganic arsenicals, mainly arsenite, was greatlydecreased. In contrast, in the 0.9 ppm arsenate or arsenite experiments, a decrease in theamount of DMA and an increase in the amount of inorganic arsenicals was observed,indicating that a demethylation process took place. After the cells which had beenexposed to MMA were transferred to an arsenic-free medium, the endocellular MMAand DMA were excreted to the medium. An increase in the concentration of DMA and adecrease in the concentration of MMA was also observed after 1-2 days.12025 250C00I.C0)C.)C0C)of arsenic species in media with incubation timebeen previously grown in (a) 10 ppm arsenate (b)arsenate (c) 10 ppm arsenite (d) 0.9 ppm arsenite (e) 10 ppm MMA (f)MMA. (0) arsenite (•) arsenate (V) MMA (Y) DMA**The values are averages of four replicate determinations by HG-GC-AA from the sample taken ateach time interval. The error bars represent 90% confidence intervals, and are smaller than the size ofthe symbols when they are not evident on the graph.20 201510502015151050201510S020151010502015510050012345678 012345678Figure 5.5 The changetransfer of cells that hadIncubation time (days)after the0.9 ppm0.9 ppm1215.3.5 Characterization of dimethylarsenic derivative in growth medium by usingHG-GC-MSThe HG-GC-MS methodology can be used for isolation and characterization ofarsines derived from hydride-forming arsemcals. In this study, the mass spectrometerwas scanned from m/z 74 to 115 at 1 scan per 0.1 s by using a wide scan mode. Scansabove m/z 115 were not made since no trimethylarsenicals were produced as mentionedpreviously. The present study mainly concerned the detection of dimethylarsine. Thechromatogram and mass spectra of arsine, methylarsine and dimethylarsine which wereproduced from standard arsenate, MMA, and DMA, obtained by using wide-scanmonitoring mode are shown in Figure 5.6; the spectra are similar to those shown inChapter 4.In order to investigate the arsenic methylation pathway in P. peniculus, we addedL-methionine-methyl-d3and arsenate (1 ppm) to the growing culture of the alga andanalyzed the aqueous arsenic metabolites in the medium by using HG-GC-MS. In thefirst experiment, only arsenate was added to the growth medium. After four days ofincubation, a dimethylarsenic derivative was detected in the medium, as can be seenfrom the gas chromatogram obtained by using HG-GC-MS (Figure 5.7). This result issimilar to those obtained by using HG-GC-AA. The first peak in the gas chromatogramwas identified as arsine [AsH3] (Figure 5.8), and the second peak as dimethylarsine[(CH3)2AsH] (Figure 5.9). This assignment was made based on a comparison of theirretention times and mass spectra with those produced by standards (Figure 5.6).In the second experiment arsenate and L-methionine-methyl-d3were added tothe growing culture at the same time. The HG-GC-MS chromatogram obtained fromsamples collected on day 4 (the day of the addition of arsenate is referred to as day 0) byusing HG-GC-MS is similar to that shown in Figure 5.7. The mass spectrum for the122Figure 5.6 HG-GC-MS chromatogram and mass spectra, of arsine, monomethylarsine,and dimethylarsine. Solutions of standani arsenite, MMA, and DMA (100 ng of arsenicfor each compound) were used.AsH3NI1000000I—500000IIIII—[M-2JAsH371 1 SI IS II IS 100308 US 315MIZ200% I‘01(CH3)2ASSi’‘I’40%S...4%50l [M]T0.00 •.)S 2.’IS 1.50Retention Time (miii.)100180%CH3AsH280%70%Iii [M]+.4%III50%20%10105 170 78 SI U Si *0 300 101UIMIZUs 70 75 4% U SiISMIZiie 105 3101231.8e+007 IAsH31.5e+007‘I1.Oe+00705e+006 (CH3)2AsHOe+000-0.00 0.50 1.00 1.25Retention Time (mm.)Figure 5.7 HG-GC-MS chromatogram of a 3 mL culture medium inoculated witharsenate (no methionine present), after 4 days of growth.124100%-90%80%70%60%50%40%30% [AsH3]20%10%0%-70 75 80 85 90 95 100 105 110115MIZFigure 5.8 Mass spectrum of AsH3 produced from a 3 mL culture medium inoculatedwith arsenate (no methiomne present), after 4 days of growth.125[CH3AS] + (CH3)2AsH90%80%70%60%C50%-40%30% [(CH3)2AsH]+[As]+ 1(620%19110% 10370 75 80 85 90 95 100 105 110 115MIZFigure 5.9 Mass spectrum of (CH3)2As produced from a 3 mL culture mediuminoculated with arsenate (no methiomne present), after 4 days of growth.126second peak indicates that the CD3 incorporation in the dimethylarsenic species issignificant as indicated by ions at mlz 112 [(CD3)AsHJ, mlz 109[(CH3) DAsH], m/z 94 [CD3AsH1 and mlz 93 [CD3As] (Figure 5.10). Theseions are absent from the mass spectra resulting from the medium containing no Lmethionine-methyl-d3 (Figure 5.9). Further evidence of the incorporation of the CD3 inthe dimethylarsenic species is provided by the single ion chromatograms of m/z 93, 109,and 112 shown in Figure 5.11. It can be seen that their retention times are identical tothe retention time of dimethylarsine shown in Figure 5.8. These peaks do not exist in theabsence of L-methionine-methyl-d3.The amount of DMA detected in the growthmedium had increased slightly with time during the 11 days incubation. This result issimilar to those obtained by using HG-GC-AA.In the third experiment, L-methionine-methyl-d3was inoculated into themedium three days prior to the addition of arsenate. Two peaks are shown in the gaschromatogram from samples collected on day 4 (the day of the addition of arsenate isreferred to as day 0). The mass spectrum of the second peak also confirms theincorporation of the deuterated CD3 group in the dimethylarsenic species as shown bythe presence of ion mlz 112, 109, and 93. The extent of CD3 incorporation is similar tothe second experiment.When L-methionine-methyl-d3was added to the culture medium three days afterthe incubation of arsenate with the alga, the mass spectrum of the dimethylarseniccompound (Figure 5.12) shows that the incorporation of CD3 is less than that found inthe experiments described above.The distribution of the CD3 label in the dimethylarsenic metabolites withincubation time is presented in Table 5.5. The percentage of CD3 incorporation wasdetermined by comparing the relative peak intensities of the parent ions, m/z 10612790%80% [(CH3)2AsJ4J +70%[(CD3)(CH3)AsH] +.—60%50% \D3)2As+40%[CD3A5]+30%7520%191 19610%J 112H 119_____________________________________I —70 75 80 85 90 95 100 105 110 115MIZFigure 5.10 Mass spectrum obtained from the culture medium inoculated with Lmethionine-methyl-d3 and arsenate simultaneously, after 4 days of growth.128q,QQ7ia...0073.4.001OI000r0.0 0.Z 0.4 0.8 0.8 1.0 1.Zb200000cE__d0.2 .4 .6 .8 1 1.2Retention Time (mm.)Figure 5.11 Wide scan and Selected Ion Chromatograms (SIC) obtained from theanalysis of culture medium inoculated with both L-methionine-d3 and arsenate. a) Widescan mlz 74-115 b) SIC of m/z 112 c) SIC of m/z 109 d) SIC of mlz 93129100%- II90%80%70%60%50%40%30%75 1620% 1110% 193I I i L ].Q970 75 80 85 90 95 100 105 110 115MIZFigure 5.12 Mass spectrum obtained from culture medium inoculated with Lmethiornne-methyl-d3and arsenate (4 days after the addition of methionine). Thearsenate was added 3 days prior to the addition of methionine.130[(CH3)2AsHj, 109 [(CH3) DAsH1, and 112 [(CD3)AsHi. The relative standarddeviation of the peak intensity of mlz 106 for 4 determinations of standarddimethylarsinic acid was 6%. For lack of a standard, the standard deviation of the peakintensity of mlz 112 from deuterated dimethylarsine was not determined.Table 5.5 Percentage distribution of dimethylarsenic compounds detected in themediumExperiment 1 2 3 4Number:Experimental Arsenatea Arsenate and Arsenate Arsenate addedcondition: added methionine added added 3 days 3 days prior toonly on the same day after methioninemethionine inoculationinoculationDayb 5 4 7 11 4 7 4 7 11Species detected:(CH3)2As 100 73 62 67 60 69 92 91 86(CD)( H’)AsH 0 20 29 20 27 21 4 4 9(CD’)2AsH 0 7 9 13 13 10 4 5 5a The medium in all experiments contained 1 ppm arsenate.b The day of arsenate addition is referred to as Day 0.5.4 DISCUSSIONAs mentioned in the introduction, Edmonds and Francesconi8°have proposedthe biotransformation pathway (Figure 1.5) to explain the production of arsenosugarsfound in marine macroalgae. In the present work, we demonstrate that when arsenate,arsenite and MMA are used as substrates for the unicellular alga P. peniculus theprincipal methylation product is DMA. No trimethylarsenical species are found in thecells or in the growth medium. Exposure of P. peniculus to arsenate yields arsenite (in131cells, and in the media) and DMA (in cells and in the medium spiked with 0.9 ppmarsenate). When the alga is treated with arsenite, MMA and DMA are detected in thecells; the metabolite DMA can also be found in the growth medium spiked with 0.9 ppmarsenite. The substrate MMA is biotransformed by P. peniculus to produce DMA in thecells. When DMA is used as a substrate, trace amounts of the demethylation productMMA are detected in the cells. When P. peniculus is transferred from arsenic enrichedmedia to arsenic free media, the accumulated arsenicals in the algal cells are excretedinto the “fresh media. Biotransformation of arsenic, including methylation anddemethylation, also takes place in this media.The most significant result from these studies is that no complex arseniccompounds, such as arsenosugars, are produced by P. peniculus. When theconcentration of arsenicals varies from as high as 10 ppm to as low as 20 ppb (in thesecond cycle), there are no “hidden” arsenicals in either the cells or the media as judgedby the flow injection-microwave digestion-HGAA methodology which has been shownto be very effective in decomposing and converting complex arsenicals to hydride-forming species. 197,198 Results obtained by using flow injection-microwave digestionHGAA are also in agreement with those obtained by using HG-GC-AA for speciation. Itseems that the alga P. peniculus follows the microbial biomethylation pathway proposedby Challenger for microbial processes (Figure 1.1), and in the case of P. peniculus,DMA is the probable end product of this biomethylation.The reduction of arsenate to arsenite is proposed to be the first step towardsmethylation,30’43 and our results are in agreement: most arsenate in the medium isreduced to arsenite by P. peniculus prior to the detection of DMA in the medium.Arsenate reduction to arsenite proceeds rapidly and most of the arsenate is reduced toarsenite within 1-2 days. It is possible that arsenate, being chemically similar to132phosphate, is readily taken up by a1gae6567 and then reduced by thiols or dithiols as adetoxification process.2 This reduction could be enzymatic or could be a chemicalreaction resulting from the interaction of arsenate with an enzymatically producedreducing agent.36 Regardless of which mechanism is correct, the results show that it isnecessary to have a biologically intact organism capable of generating the appropriatereducing conditions, because nonmetabolizing, enzymatically inactive cells do notreduce arsenate to arsenite in the growth medium.Compared to cells exposed to 0.9 ppm arsenate, the accumulation of arsenic incells is much lower when the cells are exposed to the same concentration of arsenite,MMA, or DMA; entry of MMA and DMA into the cell probably occurs by means ofpassive diffusion.2In particular, the uptake of MMA seems to be the least efficient. Thisresult agrees with that reported by Cullen and Nelson’91-193 that MMA has a muchlower diffusion coefficient than DMA. The diffusion coefficient was determined whenliposomes were employed as model membranes and showed that the diffusioncoefficient of MMA was shown to be 10 times lower than DMA.Arsenate at high concentration (10 ppm) is more toxic to P. peniculus, andaffects the arsenic accumulating ability of the alga as shown in Table 5.2. It is possiblethat high concentrations of arsenate inactivate the phosphate transport system andinterfere with oxidative phosphorylation.199201A high concentration (10 ppm) of arsenate or arsenite also seems to inhibit thebiomethylating ability of P. peniculus. As shown in Table 5.2, only a small amount ofmethylated arsenicals is found in the cells after 7 days growth. In a lower arsenate orarsenite concentration (0.9 ppm), P. peniculus can efficiently methylate inorganicarsenic to DMA which can either be excreted into the medium or kept in the cells. Whenthe alga is transferred from a hostile environment, such as medium containing 10 ppm133arsenate or arsenite, to a fresh arsenic-free environment, the biomethylating ability of P.peniculus seems to be restored. This was indicated by an increase in the amount ofDMA in the new medium compared to the amount of DMA in the cells harvested at theend of the first cycle.It is not surprising to see that little or no MMA is detected when the alga istreated with arsenate and arsenite. Cullen et al.5 reported that only traces of MMAwere produced from arsenate by broken cell homogenates of Apiotrichum humicola andthat MMA is the least transformed arsenical substrate. They tentatively concluded thatMMA is not a free intermediate in Challenger’s arsenate-to-trimethylarsine pathway.The results obtained from whole cell cultures of A. humicola as previously described inChapter 3 also show that only limited amounts of MMA are found as an extracellulararsenic metabolite in the growth medium. As mentioned previously, MMA has a verylow diffusion coefficient, and the cells may prefer to metabolize the endocellular MMAto DMA than to wait for MMA to diffuse into the growth medium. The DMA thusproduced has a higher permeability coefficient, and can be excreted by the cells to thegrowth medium. This surmise seems to be consistent with what we have observed. Workon arsenic speciation in seawater shows that arsenite and DMA are the main arsenicalproducts of natural phytoplankton blooms; MMA has not been detected in highconcentrations.202-4It has been shown that a variety of marine phytoplankton take up arsenate fromseawater and produce arsenite, MMA and DMA,20’67249and release them intothe surrounding media.20’79The efflux studies of P. peniculus demonstrate that theexcretion of arsenicals is a rapid process. No “hidden” arsenic species are detected in thecells and in the medium, indicating that P. peniculus does not produce arsenosugarswhen it is exposed to low concentrations of arsenic species. We suggest that the fast134excretion of the reduced and br methylated arsenic compounds to the media reduces therequirement for further detoxification process. It seems that the cells interact not onlywith endocellular arsenicals but also with the excreted arsenicals as indicated by thedifferences in the amount of individual arsenical species before and after the transfer.The biotransformation, the excretion, and the re-uptake of the excreted arsenicals maytake place simultaneously. These interactions seem to reach a steady states after 2-3 daysin the “fresh” media.Both HG-GC-AA and HG-GC-MS technique have proved to be very useful toolsfor the determination of hydride-forming arsenic compounds.116439440162J6’7°The HG-GC-AA system is more widely used because it is cheaper and easier to accessthan the latter. However, HG-GC-MS system is a very useful tool for characterizing anunknown volatile compound because both the retention time and its mass spectrum arethus available. The results from the culture samples obtained by using the massspectrometer in the wide scan monitoring mode clearly shows that only adimethylarsenic species is produced, and excreted to the growth medium by the alga.The most significant result is that when L-methionine-methyl-d3is added to thegrowth medium, the CD3 group is incorporated to a considerable extent by the cells toform deuterated dimethylarsenic species. This is revealed by the ions at mlz 112, 109,and 93. The mass spectrum of the volatile dimethylarsenic compound indicates that amixture of deuterated and non-deuterated DMA is present in the growth medium of algaP. peniculus.The confirmation of CD3 incorporation in the dimethylarsenic compound isimportant, as it is the first time that methionine, or S-adenosylmethionine, has beenshown to be the source of the methyl groups in the biological alkylation of arsenic inmarine algae. This result confirms that P. peniculus also follows the oxidation-reduction135pathway involving carbonium ions originally suggested by Challenger for the alkylationof arsenic by microorganisms.30’43As previously mentioned, DMA is probably the endproduct of this methylation process in P. peniculus.In addition to systematic and random errors, the errors such as the assumptionthat the ionization efficiencies, and thus the responses, of the deuterated arsines areidentical to the responses of the undeuterated arsines, may affect the calculatedincorporation percentages of CD3. A previous study by Cullen et al.33 demonstratedthat the deuterated trimethylarsine shows slightly lower sensitivity than the nondeuterated trimethylarsine under mass spectrometry conditions. Thus, the resultspresented in Table 5.5 should only be viewed as an indication of low or highincorporation without too much reliance being placed on the actual values. Nevertheless,it can be clearly seen that a high incorporation of the CD3 label in the dimethylarsenicspecies does occur.The added methionine is expected to be incorporated into the cell’s pooi of SAM,and could enhance any methylation process. This expectation is based on previous resultobtained by Cullen and coworkers.33 In their experiment different concentrations ofmethionine were applied to the growing culture of microorganisms for production oftrimethylarsine, and it was found that the methyl incorporation in arsenicals isproportional to the concentration of methionine. Later, Cullen et al.45 reported that preincubation of whole cell cultures of A. humicola or of their cell-free extracts withmethionine did not effect the amount of the methylated arsenic species produced frommethylarsine oxide. In our experiments, no apparent increase or decrease in the quantityof the dimethylarsenic metabolite was observed when L-methionine-methyl-d3waseither present or absent in the medium. The data of Table 5.5 show that the quantity ofCD3 label incorporattion in the dimethylarsenic species in experiments 2 and 3 is quite136similar, about 27-40%. Pre-incubation of methionine with the cells neither affects theamount of DMA produced nor increases the percentage of deuterated DMA in theculture medium. In the fourth experiment L-methionine-methyl-d3 was not added until 3days after the incubation of arsenate with P. peniculus. The deuterated dimethylarsenicderivative was observed one day after the addition of methionine but in lower amountsthan those observed in experiments 2 and 3. In addition, no significant increase isobserved in the total amount of dimethylarsenic species and in the amount of deuteratedarsenic species present after 11 days of incubation. This suggests that most of thearsenate had been incorporated and metabolized by the cells prior to the addition of Lmethionine-methyl-d3. After the methylated dimethylarsenic species is excreted into thegrowth medium, the methylation process slows down and reaches a steady state.Based on these results we now propose a model for the methylation of arsenateby cells of P. peniculus. The basic steps are outlined in Figure 5.13. First, algal cellstake up arsenate from the medium via the phosphate transport system, reduce thearsenate to arsenite inside the cells by using thiols and/or dithiols, and excrete most ofthe arsenite into the growth medium by means of an active transport system. Thisprocess is reasonably fast. Second, arsenite in the cells is methylated to MMA by usingSAM; however, due to the low passive diffusion coefficient, the endocellular MMA isnot excreted to the growth medium. As a consequence, the MMA remains in the cellwhere it is more likely to be reduced and further methylated to DMA. This arsenicalwhich has a greater diffusion coefficient can diffuse into the growth medium by meansof passive diffusion. The methylation of MMA could be a fast process. In Chapter 3, wereported that MMA is metabolized to DMA by whole cell cultures of A. humicola and isthen excreted rapidly to the growth medium. After the excretion of DMA by the algalcells, the methylation process slows down as indicated by the absence of any change in137cell membranephosphate I Arsenatea thiols and/crtransport system dithiolsmedium cells—cellmembrane.b active transport II irsenitea 2e 2e DMAaArsemtesystem II Me Memedium cellsactive transport Arsenite1systemmediumpassive DMAbdiffusionmediumFigure 5.13 Proposed model for biomethylation of arsenate in marine algae P.peniculus. a endocellular arsenicals b extracellular arsenicalsthe concentrations of DMA in the growth medium and by the results obtained inExperiment 4 where the addition of the deuterated methionine 3 days prior to theaddition of arsenate does not increase the extent of CD3 incorporation in the DMA. Thismodel explains many of the results obtained in the present investigations, but in realitythe biotransformation process is probably much more complex. For example, anuptake/excretion equilibrium between endocellular and extracellular arsenicals, and thecleavage of As-C bonds may also be involved in the metabolic process.138In conclusion, the results obtained from this study strongly suggest thatmethionine via S-adenosylmethionine, is the source of the methyl groups in thebiological alkylation of arsenic in marine algae. The fmal arsenical biotransformationproduct of P. peniculus is a dimethylarsenic compound, probably DMA.139CHAPTER 6THE IDENTIFICATION AND CHARACTERIZATION OF ANTIMONY(ffl),ANTIMONY(V), AND METHYLANTIMONY SPECIES BY USING HG..GC-MS6.1 INTRODUCTIONAntimony is present in the aquatic environment as a result of the weathering ofrocks, from soil runoff, and through effluents from mining and smelting.205 It is nowrecognized that the toxicity and physiological behavior of antimony depend on itsoxidation state.206 Typical concentrations of antimony in unpolluted waters are lessthan 1 ppb.207’8Hydride generation-atomic absorption spectrometry (HGAA) has been by far themost frequently used technique for determination of antimony in aqueoussolutions.26’0509212 The hydride forming antimony compounds are reduced tostibines by sodium borohydride solution (NaBH4),and detected by an atomic absorptionspectrometer.Using a batch type hydride generator, Andreae etal.26’051 identified themethylantimony compounds in natural water samples by using HG-GC-AA where thehydrides cochromatographed with those produced from supposedly authentic samples.On this basis the organoantimony species found in the samples were said to bemethyistibonic acid MeSbO(OH)2and dimethylstibinic acidMe2SbO(OH).However, Dodd et al.212 reported that molecular rearrangement of stibineproduced from pure methylantimony compounds Me3Sb(OH)2,Me3SbC12, andMe2SbC1(OH) occurred when a batch type hydride generator was used, giving amixture of antimony hydrides, possibly SbH3,MeSbH2,Me2SbH, and Me3Sb from all140three antimony(V) compounds. These results cast some doubt on the ability of thehydride generation technique to clearly distinguish different alkylated compounds ofantimony. As a consequence, the results of earlier studies of antimony speciation in theaquatic environment26’0511need to be regarded with caution.In the present work, we use a continuous type hydride generation system coupledwith GC-MS to identify the antimony hydrides produced from the trimethylantimonycompounds Me3Sb(OH)2 and Me3SbCl2. The possible cause of the molecularrearrangement noted above was investigated. The HG-GC-MS system was also used toanalyze the extracts of plant samples collected in Kam Lake and Keg Lake(Yellowknife). The results provided conclusive evidence of the presence ofmethylantimony compounds in these samples.6.2 EXPERIMENTAL6.2.1 ReagentsAll chemicals used were of reagent grade. Distilled water was used for alldilution. Glass and plasticware were cleaned by soaking overnight in 2.0% Extransolution, followed by a water rinse, a soak in dilute hydrochloric acid, and finally awater rinse.A solution of antimony potassium tartrate containing 1000 ppm of Sb was usedto prepare the standards for the determination of antimony(ffl). The antimony(V)standard (1000 ppm of Sb) was prepared from potassium antimonate. The standards forMe3Sb(OH)2and Me3SbC12were supplied by Dr. M. Dodd (Royal Roads MilitaryCollege, Victoria, B.C., Canada). They were dissolved in distilled water, and diluted to141make a stock solution containing 100 ppm of Sb. The standard solutions were diluted tothe appropriate concentrations daily for analytical use. Solutions of 2.0% (wlv) NaBH4prepared in 0.1% (wlv) NaOH, and 4.0 M acetic acid were freshly made daily.6.2.2 Hydride generation-gas chromatography-mass spectrometryThe hydride generation-gas chromatography-mass spectrometry system (HGGC-MS) used was previously described in Chapter 4. The experimental conditions arepresented in Table 4.1. In this study, the mass spectrometer was scanned from m/z 120to 170 at 1 scan per 0.3 second. The highest m/z value for any anticipated volatilestibine was 168, corresponding to (CR3)‘23Sb.6.2.3 Experimental proceduresA Gilson Minipuls 2 four-channel peristaltic pump was used to deliver and mixsample, acid and sodium borohydride solution. The standard solution of eitherantimony(III), antimony(V), or the trimethylantimony standards was first mixed with4.0 M acetic acid, and then 2.0% (w/v) NaBH4.The hydrides were produced in an 18-turn glass reaction coil, carried by a helium flow to a glass liquid/gas separator wherethe spent solution constantly drained out, and then collected in a Teflon U-tube trapimmersed in liquid nitrogen. The liquid nitrogen bath was then removed and the U-tubetrap immersed in a 80°C water bath. The released stibines were separated by using GCon a silanized Porapak-PS column, and characterized by using the mass spectrometer.When the plant extracts were analyzed, 2 mL of 10% (w/v) EDTA was added to thesample (5 mL) prior to analysis.1426.3 RESULTS AND DISCUSSIONThe HG-GC-MS technique has proved to be an effective tool for arsenicdetermination.’16’39”6 However, there has been no published report regarding theuse of a HG-GC-MS system for the determination of antimony compounds. Most of thereported work concerning the analysis of Sb(ffl), Sb(V), and methylantimony speciesused a batch-type hydride generator coupled with a GC-AA system. The hydridesproduced were usually concentrated in a cold trap, separated by using temperatureprogrammed GC, and detected by using an AA2620521 1,212 Since the methylatedantimony species can be separated from inorganic antimony on the basis of theirchromatographic behavior, only Sb(ffl) and Sb(V) need to be differentiate by using thehydride generation step. The efficiency of the hydride generation process for Sb(ffl) andSb(V) is related to their pKa values, and depends strongly on the pH of the reactionmedium.205 At pH >4, Sb(llI) (pKal 11.0) is reduced to SbH3 by NaBH4;and at pH =1, both Sb(Ill) and Sb(V) (pKa1 2.7) form SbH3.The continuous type HG-GC-MS system described in Chapters 4 and 5 was usedsuccessfully to characterize the arsenic metabolites produced by a microorganism and bya marine alga. We showed that 4.0 M acetic acid could be used successfully for theanalysis of arsenite, MMA, DMA, and TMAO. This approach was also applied to theanalysis of Sb(ffl) and methylantimony species. Andreae et at.205 suggested thatslightly higher values of pKal are to be expected for the methylantimony species,compared to methylarsenic species, due to the larger ionic radius of antimony. Thereforesamples acidified with 4.0 M acetic acid could be analyzed successfully formethylantimony species. Indeed tnmethylstibine is easily produced from(CH3)Sb(OH)2and (CH3)SbCl2under these conditions as will be shown later. Theprinciple aim of the present work was to identify the methylated antimony compounds143present in the samples, so further optimization of the hydride generation process was notattempted.When a solution of Sb(ffl) was mixed with 4.0 M acetic acid and 2.0% (wlv)NaBH4,the production of stibine SbH3 was confirmed by the mass spectrometer. TheHG-GC-MS chromatogram and the mass spectrum of SbH3 are shown in Figure 6.1 andFigure 6.2, respectively. Because antimony has two major isotopes, ‘21Sb (57%) and123Sb(43%), and arsenic has only one major isotope 75As, the mass spectrum of stibineSbH3 is more complex than that of arsine AsH3.The ions at m/z 126 and 125 are from[123SbH3]and[123SbH],and the ions at m/z 122 and 121 are from[‘21SbHi and[121 SbJ, respectively.When a trimethylantimony compound Me3Sb(OH)2 was examined by usingHG-GC-MS, the chromatogram (Figure 6.3) shows the presence of three peaks. Themass spectra of the first two peaks (Peak A and Peak B) are shown in Figures 6.4 and6.5, and can not be assigned to either SbH3,MeSbH2,or Me2SbH and do not appear tobe Sb containing compounds as judged by the isotope pattern in the mass spectra. Themass spectra of the third peak in Figure 6.3 clearly shows the presence of (CH3)Sb asindicated by the ions at m/z 168 [(CH3)123SbJ, m/z 166 [(CH3)121Sb], m/z 153[(CH3)2123SbJ, mlz 138 [CH123SbJ, and rnlz 121[121SbJ (Figure 6.6). The HGGC-MS chromatogram (Figure 6.7) obtained from a blank solution shows the presenceof two peaks corresponding to Peak A and Peak B, indicating that these compounds areprobably not from the original standard solution. The HG-GC-MS chromatogram andmass spectrum of the hydride produced from Me3SbC12 are similar to that fromMe3Sb(OH)2.As mentioned above Dodd et al.212 reported that the trimethylantimonycompoundsMe3Sb(OH)2andMe3SbC12formed four hydrides SbH3,MeSbH2,14455000005000000 SbH345000004000000350000030000002500000200000015000001000000500000 . -0-0.0 0.5 1.0 1.5 2.0Retention Time (mm.)Figure 6.1 HG-GC-MS chromatogram of stibine SbH3. A standard solution of Sb(ffl)(300 ng of antimony) was used.100%- —i1190%+-C80% ‘70% —60%+40%30%20%10%0%- —‘ -117 130 140 150 160 170MIZFigure 6.2 Mass spectrum of stibine SbH3 derived from a standard solution of Sb(ffl).14511000000-900000 Me3Sb800000700000600000500000400000300000 A200000100000 Bv-‘--I I0.0 0.5 1.0 1.5 2.0Retention Time (mm.)Figure 6.3 HG-GC-MS chromatogram of trimethyistibine Me3Sb. A standard solutionof Me3Sb(OH)2(200 ng of antimony) was used.90%80%70%50%40%30%20%10%—I I 147115 130 140 150 160170M/ZFigure 6.4 Mass spectrum of Peak A shown in figure 6.3.146100% 1790%80%70%, 60%50%40%30%20%10% 1?21390% Ii i f? •14115 130 140 150 160 170MIZFigure 6.5 Mass spectrum of Peak B shown in figure 6.3.100% I_____________________-11126 +90%+80%7O%60% —C—50%40% 16Z30%20%10%___—______________115 130 140 150 160 170MIZFigure 6.6 Mass spectrum of trimethyistibine Me3Sb derived from a standard solutionof Me3Sb(OH)2147.12.0Retention Time (mEn.)Figure 6.7 HG-GC-MS chromatogram of a blank solution. Solutions of 4.0 M aceticacid and 2.0% (w/v) NaBH4were used for the reaction.Me2SbH, and Me3Sb when a batch type hydride generator was used for hydridegeneration, indicating that the molecular rearrangement had occurred during the hydridegeneration. When the same trimethylantimony compounds (from the same source) wereanalyzed by using the semi-continuous mode hydride generation system described inChapter 2, the only stibine produced is Me3Sb as can be seen from Figure 6.3. However,the molecular rearrangement of the trimethyistibine can take place if certain precautionsare not taken. This usually happens when the hydride generation system including theprobes (for the introduction of acid, NaBH4 and sample), the reaction coil, and theliquid/gas separator are rinsed with distilled water and then used immediately for sampleanalysis. The HG-GC-MS chromatogram (Figure 6.8) obtained from a sample of(CH3)Sb(OH)2 (with the molecular rearrangement) shows the presence of four0.0 0.5 1.0 1.5148stibines. The first antimony containing peak was identified as SbH3 whose massspectrum (Figure 6.9) is similar to that seen in Figure 6.2. The second antimonycontaining peak was characterized as MeSbH2 (Figure 6.10) on the basis of ions at mlz140 [CH3123Sb],m/z 139 [CH123SbH], and m/z 121 [121Sb]. The massspectrum of the third antimony containing peak is shown in Figure 6.11. The ions at m/z154 [(CH3)2123SbH], m/z 152 [(CH3)121SbH , mlz 139 [CH23SbHi’, m/z 138[CH123Sb], and m/z 121[121Sb] confirm the presence of Me2SbH. The fourthantimony containing peak is due to Me3Sb on the basis of the mass spectrum shown inFigure 6.12 (cf. Figure 6.6).The above results are noteworthy as they confirm the identity of the molecularrearrangement products. These spectra can also be used as qualitative standards for theanalysis of environmental samples.The possible causes of the rearrangement were investigated. At first it wassuspected that the rearrangement could be due to active sites on the inner glass surfaceof the hydride generation system. However, the use of a Teflon reaction coil andliquid/gas separator did not eliminate the rearrangement. Only when acetic acid,NaBH4, and distilled water were introduced by using a peristaltic pump andcontinuously mixed in the reaction coil with the aid of a stream of helium for a fewminutes (usually more than 3 minutes) prior to sample analysis was the rearrangementprevented (Figure 6.3). At the present time the cause of the rearrangement remainsunknown. After the system was conditioned by analyzing the blank solution atbeginning of a routine analysis session, no more conditioning was needed.In the light of these results it is not surprising that Dodd et al.212 encounteredthe rearrangement of trimethyistibine when using the batch type hydride generator.Their apparatus was cleaned with deionized water after each analysis, and was never149800000•700000 MCSbH2 I600000500000 Me3SbL.AZe7cIN\S00.0 0.5 10 1.5 .2.0Retention Time (mm.)Figure 6.8 HG-GC-MS chromatogram of the hydrides derived from a solution ofstandard Me3Sb(OH)2(200 ng of antimony) exhibiting molecular rearrangement. Thesample analysis was performed immediately after the hydride generation system wasrinsed with distilled water.100%- -1190% +80%70%I:40%30%20%10%0%- — I -115 130 140 150 160 170M/ZFigure 6.9 Mass spectrum of the first antimony containing peak in Figure 6.8, identifiedas SbH3.150100%-1190% +Cl)80%70%60%50% l’6 +=• 40%-30%20%10%0%- - I115 130 140 150 160 170MJZFigure 6.10 Mass spectrum of the second antimony containing peak in Figure 6.8,identified as MeSbH2.100%-11 1690%++80%N Cl),60% +50%.40% +30%.1220% C.)10%0%--I I I115 130 140 150 160170M/ZFigure 6.11 Mass spectrum of the third antimony containing peak in Figure 6.8,identified as Me2SbH.1511II.170MIZFigure 6.12 Mass spectrum of the fourth antimony containing peak in Figure 6.8,identified as Me3Sb.conditioned with the mixture described in the above paragraph. Andreae etal.262O52H also used a batch type hydride generator for the analysis of Sb(ffl),Sb(V), and methylantimony species. They reported that the methylantimony standardshad been stored for several years, and thus the solutions prepared from them containedsome demethylation products: methyistibonic acid contained inorganic Sb; and thedimethyistibinic acid contained inorganic Sb and methyistibonic acid as judged by usingHG-GC-AA. In light of the present work, it is possible that these demethylatedcompounds were produced in the hydride generation process rather than in the storageperiod. The results reported by Andreae etal.26’0511 that Sb(ffl), methyistibonicacid, and dimethylstibinic acid were found in natural water samples must be interpretedwith caution because they used a batch type HG-GC-AA system.115 130 140 150 160152In the present work, the HG-GC-MS system was used to analyze the extracts ofplant samples collected in Yellowknife. Identification of the antimony species producedfrom these samples was achieved based on the comparison of their retention times andmass spectra with those shown in Figures 6.8, 6.9, 6.10, 6.11, and 6.12. The HG-GCMS chromatogram (Figure 6.13) of the extract of plant samples collected in Kam Lakeshows three major peaks. The first two peaks are from the same unknown compounds(Peak A and Peak B) described previously. The mass spectrum of the third peakindicates that this is a trimethylantimony compound (Figure 6.14). In the extract of plantsamples collected in Keg Lake, four antimony containing peaks were found (Figure6.15). Peak 1 was identified as SbH3 (Figure 6.16) produced from Sb(ffl), Peak 2 asmonomethyistibine MeSbH2 (Figure 6.17), Peak 3 as dimethyistibine Me2SbH (Figure6.18), and Peak 4 as Me3Sb (Figure 6.19).These results are significant as it is the first time that convincing evidence isprovided for the existence of methylantimony compounds in the environment. Thesecompounds are likely to be biological metabolites since the anthropogenic sources ofmethylantimony compounds are not known.Sb(V) must be determined under hydrochloric acid conditions since Sb(V) cannot form stibine when 4.0 M acetic acid is used. In the presence of 1.0 M hydrochloricacid, both Sb(V) and Sb(ffl) are reduced to SbH3 by sodium borohydnde. Because ofthe limited time available and the lack of standard compounds of methyistibonic acidand dimethylstibinic acid, we did not quantify any of these antimony compounds in thetwo samples.153230000A Me3Sb200000-.150000c.B— 100000500000_J0.0 0.5 1.0 1.5 1.8Retention Time (mm.)Figure 6.13 HG-GC-MS chromatogram of a 5 mL extract of plant samples collected inKam Lake.100%-1____________________11+90%80%[121SbJ 1 5—70%+ +, 60%q 50%1540%30%1620%10% 127 1560%_hI l12 iIi_ J’’JI I Iii___________ __ _____ __ __ ____117 130 140 150 160170M/ZFigure 6.14 Mass spectrum of Me3Sb derived from a 5 mL extract of plant samplescollected in Kam Lake.1543A350000 Me2SbH3000002500002200000 MeSbH2150000 1 4SbH3 B Me3Sb100000500000-0.0 0.5 1.0 1.5 1.8Retention Time (mm.)Figure 6.15 HG-GC-MS chromatogram of a 5 mL extract of plant samples collected inKeg Lake.100%-______I+90%80%70%, 60%— +50%z.0. 40%30% —20%14310%0%-96 I.117 130 140 150 160 170M/ZFigure 6.16 Mass spectrum of Peak 1 in Figure 6.15, identified as SbH3.155—- --12190% +1680% I, 70% —I_+-.—x4? 50%40%—4?30%20%10% 128. 157II’ 1 f9 I117 130 140 150 160 170M/ZFigure 6.17 Mass spectrum of Peak 2 in Figure 6.15, identified as MeSbH2..LVV 16• 1180%p70%, 60%Z—• 50%40%—30%20%10% I0%-—— II142 H117 130 140 150 160 170MIZFigure 6.18 Mass spectrum of Peak 3 in Figure 6.15, identified as Me2SbH.1561 -- - 111690% +80% +[121Sb]70%1760%50%- 1 640%30%20%16L___116 , 1110%0%-117 130 140 150 160170MIZFigure 6.19 Mass spectrum of Peak 4 in Figure 6.15, identified as Me3Sb.157CHAPTER 7SUMMARYThe hydride generation technique is a very useful tool for the determination ofhydride-forming arsenic species such as arsenite, arsenate, MMA, DMA and TMAO.The semi-continuous flow HG-GC-AA system developed in the course of this project iseasy to construct and simple to operate. The operational procedures are simplifiedcompared to those of the batch type hydride generation systems, and the time for onesample analysis can be reduced to less than 7 minutes. The system has proved to beefficient and gives reproducible results in analyzing a large number of samples (>40samples per day).The biotransformation processes for low levels of arsenic (1 ppm) in themicroorganisms A. humicola and S. brevicaulis were investigated.Exposure of A. humicola to arsenate yields arsenite prior to the detection of anymethylated arsenicals MMA, DMA and TMAO in the growth medium. Most of thearsenate substrate is transformed into TMAO, and the production of MMA and DMA islimited. The substrate arsenite is metabolized by A. humicola to produce MMA (in traceamounts), DMA (in small amounts), and TMAO (the major metabolite). The oxidationof arsenite to arsenate by the microorganism is insignificant. When MMA is used as asubstrate, DMA and TMAO are the major metabolites, and the demethylation of MMAto arsenate and/or arsenite by A. humicola is insignificant. The transformation of thesubstrate DMA to the metabolite TMAO is a slow process, and only small amounts ofTMAO are present in the growth medium at the end of the experiment.The microorganism S. brevicaulis also transforms the arsenic substrates into avariety of metabolites including arsenite from arsenate, DMA and TMAO from arsenate,158arsenite and MMA, and TMAO from DMA. However, the yields of the methylatedarsenic metabolites are low due probably to the growth medium used in the experimentssince this medium was optimized for the production of trimethylarsine by A. humicola.Trimethylarsine is not produced by either of the microorganisms under theexperimental conditions used in the work. It now seems likely that low concentrations ofTMAO (< 1 ppm) do not greatly affect the living fungal system. Therefore, furtherdetoxification by reducing TMAO to trimethylarsine is not necessary.The identification of these arsenicals is noteworthy as it is the first time that anyof the non-volatile arsenic intermediates have been identified in the growth medium of apure culture of microorganisms spiked with either arsenate, arsenite, Mlv.IA or DMA.The production of the anticipated methylated intermediates from the substrates, theabsence of oxidation of arsenite to arsenate, and the lack of demethylation productsstrongly support the metabolic sequence proposed by Challenger.On the basis of the results obtained from the experiments, an extended model isproposed to supplement Challenger’s proposed pathway, which explains ourexperimental observations in terms of the uptake, the biotransformation, and theexcretion of the arsenicals by the cells of microorganisms.The effect of adding L-methionine-methyl-d3and the arsenic substrates (1 ppm)to the growing culture of A. humicola was studied by using a specially developed HGGC-MS methodology. When either arsenate or arsenite is added to the growth mediumof A. humicola in the presence of L-methionine-methyl-d3, the mass spectra of thehydride derivative of the dimethylarsenic metabolite exhibit ions at m/z 112[(CD3)2AsHjt and m/z 109 [(CH3) DAs ] indicating incorporation of the CD3moiety. Similarly, the mass spectra of the produced trimethylarsenic species containsions at m/z 129 [(CD3)As], mlz 126 [(CH3) DAs], and m/z 123159[(CH3)2(CDAs]. When MMA is used as the substrate in the presence of Lmethionine-methyl-d3, the mass spectra of the produced diinethylarsenic andtrimethylarsenic species show the ions at m/z 109 [(CH3) DAsH], mlz 126[(CH3) D2As]and m/z 123 [(CH)(CDAs]. The absence of ions at mlz 112[(CDAsH] and m/z 129 [(CD3)As] indicates that the cleavage of the H3C-Asbond is insignificant. When both DMA and L-methionine-methyl-d3 are added to thegrowing culture of A. humicola, the mass spectrum of the produced trimethylarsenicspecies exhibits the ions at mlz 123 [(CH)(CDAs].The addition of L-methionine-methyl-d3 does not alter the production of thearsenic metabolites by A. humicola.This is the first time that methionine has been demonstrated to be involved in theproduction of the non-volatile methylarsenic intermediates which have been proposed inChallenger’s pathway. Thus this work strongly reinforces the suggestion that methioninevia SAM is the source of the [CH3]+ shown in Challenger’s pathway.The arsenic accumulation in cells of a unicellular marine alga P. peniculus wasfound to be proportional to the concentration of the arsenic substrates with the exceptionof arsenate. Arsenate at high concentrations (10 ppm) is more toxic to P. peniculus, andaffects the bioaccumulating and biomethylating ability of the alga. When arsenate,arsenite and MMA are used as substrates for P. peniculus the principal methylationproduct is DMA. No trimethylarsenic species are found in the cells or in the growthmedium. Exposure of P. peniculus to arsenate yields arsenite (in cells and in the media)and DMA (in cells and in the medium spiked with 0.9 ppm arsenate). The reduction ofarsenate to arsenite by the alga is a rapid process. When the alga is treated with arsenite,MMA and DMA are detected in the cells; the metabolite DMA can also be found in thegrowth medium spiked with 0.9 ppm arsenite. The substrate MMA is transformed by P.160peniculus to produce DMA in the cells. When DMA is used as a substrate, traceamounts of the demethylation product MMA are detected in the cells. When P.peniculus is transferred from arsenic enriched media to arsenic free media, theaccumulated arsenicals in the algal cells are excreted into the “fresh” media.Biotransformation of arsenic, including methylation and demethylation, also takes placein this media. No complex arsenic compounds, such as arsenosugars, are produced by P.peniculus.The alga transforms arsenate (1 ppm) to a deuterated dimethylarsenic species inthe presence of L-methionine-methyl-d3. The mass spectrum of the dimethylarsenicspecies obtained by using the HG-GC-MS system shows the presence of ions at m/z 112[(CD3)2AsH], and m/z 109 [(CH3) DAsHj. The CD3 label is incorporated intactby the cells to form deuterated DMA. This is the first time that methionine, or SAM, hasbeen shown to be the source of the methyl groups in the biological alkylation of arsenicin marine algae.These results suggest that the alga follows the biomethylation pathway proposedby Challenger for microbial processes involving alternating oxidation and reductionsteps and the use of carbonium ions for the alkylation of arsenic. In the case of P.peniculus, DMA is the end product of this methylation. A model for the methylation ofarsenate by the cells of P. peniculus is proposed to explain the results obtained.The semi-continuous HG-GC-MS system has proved to be very useful for theidentification of antimony hydrides produced from the trimethylantimony compoundsMe3Sb(OH)2 and Me3SbCl2. The identification of the molecular rearrangementproducts of trimethylstibine is achieved by using this system. The possible causes of themolecular rearrangement of trimethylstibine were investigated. The appropriateoperational procedures have been established to effectively eliminate the causes of the161molecular rearrangement of trimethyistibine during the hydride generation processes.The system has been successfully used to analyze the extracts of plant samples collectedfrom Kam Lake and Keg Lake (Yellowknife). These samples were found to containmethylantimony compounds. This is first time that convincing evidence is provided forthe existence of methylantimony compounds in the environment.The HG-GC-AA and HG-GC-MS developed dunng the course of this projecthave proved to be useful in the determination and identification of the arsenicmetabolites produced by the microorganisms and a marine alga. The biotransformationprocesses of antimony in organisms have not been fully investigated. Therefore, it isalso possible to use the two systems to study the interactions between antimony andbiota including microorganisms, algae, and plants.162BIBLIOGRAPHY1. National Research Council, Arsenic; National Academy of Science: Washington,DC, 1977, p162. Cullen, W R and Reimer K 3 Chem.. Rev., 1989, 89:7 133. Ferguson, 3 F and Gavis 3 Water Res., 1972,6:12594. Squibb, K S and Fowler, B A In Biological and Environmental Effects ofArsenic; Fowler, B A, Ed.; Elsevier: Amsterdam, 1983,p2335. National Research Council of Canada, The Effects ofArsenic in CanadianEnvironment, Ottawa, 19786. Thomas, H and Brein, L AMem. Liverpool School Trop. Med., 1905, No.167. Uhlenhuth, H Dtsch. Med. Wochenschr, 1907, 33:12378. Arsenic: Industrial, Biological, Environmental Perspectives (Proc. ArsenicSymp., 1981), Lederer, W H and Fensterheim, R 3, Eds., Van Nostrand: NewYork, 1983, p899. Arsenic: Industrial, Biological, Environmental Perspectives (Proc. ArsenicSymp., 1981), Lederer, W H and Fensterheim, R J, Eds., Van Nostrand: NewYork, 1983, p5710. Arsenauk, R D Proc. American Wood-Preserver Assoc., 1975, 71:12611. IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals toHumans, 1980, 23:3912. Tamaki, S and Frankenberger Jr. W T Rev. Environ. Contam. ToxicoL, 1992,124:7913. Penrose, W R CRC Crit. Rev. Environ. Control, 1974,46514. Myers, D J, Heimbrook, M E, Osteryoung, 3 and Morrison, S M Environ. Lett.,1973, 5:5316315. Crecelius, E A Bull. Environ. Contain. Toxicol., 1977, 18:22716. Johnson, D L Nature (London), 1972, 240:4417. Rowland, I R and Davies, M J J. App!. Toxicol., 1977, 18:22718. Blasco, F, Gaudm, C and Jeanjean R C.R. Hebd. Seances Acad. Sci. Ser. D,1971, 273:81219. Blasco, F, Gaudin, C and Jeanjean R CR. Hebd. Seances Acad. Sci. Ser. D,1972, 275:122320. Sanders, J G, Windom H L Estuar. Coast. Mar. Sci., 1980, 10:55521. Blake, N J and Johnson, D L Deep Sea Res., 1976, 23:77322. Johnson, D L and Braman, R S Deep Sea Res., 1975, 22:50323. Turner, D R, Whitfield, M and Dickson, A G Geochim. Cosmochim. Acta., 1981,45:85524. Andreae, M 0 Limnol. Oceanogr., 1979, 24:44025. Andreae, M 0 Deep Sea Res., 1978, 25:39 126. Andreae, M 0 and Froelich, P N Jr. Tellus, 1984, 36B: 10127. Pilson, M E Q Liinnol. Oceanogr., 1974, 19:33928. Gosio, B Ber. Duet. Chem. Ges., 1897, 30: 102429. Biginelli, D Gazz. Chim. Ital., 1901, 3 1:5830. Challenger, F Chem. Rev., 1945, 36:3 1531. Challenger, F, Higginbottom, C and Effis, L J. Chem. Soc., 1933, 9532. Cox, D P and Alexander, M Bull. Environ. Contain. Toxicol., 1973,9:8433. Cullen, W R, Froese, C L, Lui, A, McBride, B C, Patmore, D, J and Reimer, MJ. Oganomet. Chem., 1977, 139:6134. Bird, M C, Challenger, F, Chariton, PT and Smith, JO Biochem. J., 1948,43:7835. Challenger, F and Ellis, L J. Chem.. Soc., 1935, 39616436. Pickett, A W, McBride, B C, Cullen, W R and Manji, H Can. J. Microbiol.,1981, 27:77337. Cullen, W R, Erdman, A E, McBride, B C and Pickett, A W J. ofMicrobiol.Methods, 1983, 1:29738. McBride, B C and Wolfe R S Biochemistry, 1971, 10:43 1239. Cheng C M and Focht, D D Appl. Environ. Microbiol., 1979, 38:49440. Shariatpanahi, M, Anderson, A C and Abdeighani, A A J. Environ. Sci. Health,1981, B 16:3541. Shariatpanahi, M and Anderson, A C Trace Subst. Environ. Health, 1982,16:17042. Pickett, A W, McBride, B C and Cullen, W RAppi. Organomet. Chem., 1988,2:47943. Challenger, FAdv. Enzymol., 1951, 12:42944. Cullen, W R Adv. Organomet. Chem., 1966, 4:14545. Cullen, W R, McBride, B C, Hasseini, M, Pickett, A W and Reglinski, J Appl.Organomet. Chem., 1989, 3:7146. Cullen, W R, McBride, B C and Reglinski, J J. Inorg. Biochem., 1984, 21:4547. Cullen, W R, McBride, B C and Reglinski, J J. Inorg. Biochem., 1984, 21:17948. Challenger, F, Lisle, D B and Dransfield, P B J. Chem. Soc., 1954, 176049. Cantoni, G L J. Biol. Chem., 1953, 204:40350. Simmonds, S, Keller, E B, Chandler, 3 P and du Vigneaud, V J. Biol. Chem.,1950, 183:19151. Cullen, W R, McBride, B C and Pickett, A W Can. J. Microbiol., 1979,25:120152. Wood, J M Naurwissenschaften, 1975, 62:35716553. Taylor, C D and Wolfe, R S J. Biol. Chem., 1974, 249:487954. McBride B C and Wolfe, R S Biochemistry, 1971, 10:23 1755. Andreae, M 0 IN: Organometallic Compounds in the Environment, Craig, P J(ed) Wiley, NewYork, 198656. Woolson, E A Amer. Chem. Soc. Symp. Ser. No. 7, Washington, D C, 197557. Penrose, W R, Conacker, H B S, Black, R, Meranger, 3 C, Miles, W,Cunningham, H M and Squires, W R Environ. Health Persp., 1977, 19:5358. Dhandhukia, MM and Seshadri, K Phykos, 1969, 8:10859. Lunde, G J. Sci. FoodAgric., 1970, 21:41660. Leatherland, T M and Burton, J D J. Mar. Bio. Assoc. U.K., 1974, 54:45761. Tagawa, S and Kojima, Y J. Shimonoseki Univ. Fish., 1976, 25:6762. Sanders, J G Chemosphere, 1979, 3:13563. Whyte, J N C and Englar, JR Botanica Marina, 1983, 26:15964. Maher, W A and Clarke, S M Mar. Pollut. Bull., 1984, 15:11165. Benson, A A Science, 1981, 211:48266. Blasco, F Physiol. Veg., 1975, 13:18567. Planar, D and Healey, F P .1. Phycol., 1978, 14:33768. Brunskill, G L, Graham, B W and Rudd, J W M Can. J. Fish. Aquat. Sci., 1980,37:41569. Cooney, R V, Mumma, R 0 and Benson, A A Proc. Nati. Acad. Sci., USA 1978,75:426270. Andreae, M 0 and Klumpp, D Environ. Sci. Technol., 1979, 13:73871. Sanders, J G J. Phycol., 1979, 15:42472. Wrench, J J and Addison, R F Can. J. Fish. Aquat. Sci., 1981, 38:5 1873. Klumpp, D W Mar. Biol., 1980, 58:26516674. Irgolic, K J, Woolson, E A, Stockton, R A, Newman, R D, Bottino, N R,Zingaro, R A, Kearney, P C, Pyles, R A, Maeda, S. Mcshane, W J and Cox, E REnviron. Health Perspect., 1977, 19:6175. Lunde, G Acta Chim.. Scand., 1973, 27:158676. Benson, A A Royal Society of Chemistry Symposium, London, 198777. Bottino, N R, Newman, R D, Cox, E R, Stockton, R, Hoban, M, Zingaro, R Hand Irgolic, K J J. Exp. Mar. Biol. Ecol., 1978, 33:15378. Knowles, F C and Benson, A A Trends Biochem. Sci., 1983, 8:17879. Cooney, R V and Benson, A A Chemosphere, 1980, 9:33580. Edmonds, J S and Francesconi, K A Experientia, 1987,43:55381. Edmonds, J S and Francesconi, K A Nature (London), 1981, 289:60282. Edmonds, J S and Francesconi, K A J. Chem. Soc., Perkin Trans.1, 1983, 237583. Shibata, Y, Morita, M and Edmonds 3 S Agric. Biol. Chem., 1987, 51:39184. Edmonds, 3 S, Morita, M and Shibata, Y .1. Chem. Soc., Perkin Trans.1, 1987,57785. 3m, K, Hayashi, T, Shibata, Y and Morita, M Appl. Organomet. Chem., 1988,2:36586. 3m, K, Shibata, Y and Morita, MAgric. Bio. Chem., 1988, 52:196587. Morita, M and Shibata, Y Chemosphere, 1988, 17:114788. Francesconi, K, Edmonds, 3 S, Stick, R V, Skelton, B W and White, A H J.Chem. Soc., Perkin Trans.1, 1991, 270789. Shibata, Y and Morita, M Agric. Biol. Chem., 1988, 52: 108790. Edmonds, J S and Francesconi, K A 3. Chem. Soc., Perkin Trans.1, 1982, 298991. Francesconi, K A, Stick, R V and Edmonds, 3 S J. Chem. Soc. Chem. Commun.,1991, 92816792. Klumpp, D W Mar. Biol., 1980, 58:25793. Klumpp, D Wand Peterson, D 3 Mar. Biol., 1981, 62:29794. Chapman, A C Analyst 1926, 5 1:54895. Coulson, £ J, Remington, R B and Lynch, K M J. Nutr., 1935, 10:25596. Lunde, G Nature (London), 1969, 224:18697. Lunde, G J. Sci. FoodAgric., 1973, 24: 102198. Lunde, G Acta. Chem. Scand., 1973, 27:158699. Lunde, G Fiskeridir. Skr. Ser. Teknol. Unders., 1972, 5:1100. Lunde, G Environ. Health Perspect., 1977, 19:47101. Cannon, J R, Edmonds, J S, Francesconi, K A, Raston, C L, Sanders, 3 B,Skelton, B W and White, A HAust. J. Chem., 1981, 34:787102. Edmonds, J 5, Francesconi, K A, Cannon, J R, Raston, C L, Skelton, B W andWhite, A H Tetrahedron Lett., 1977, 1543103. Edmonds, 3 S and Francesconi, K A Chemosphere, 1981, 10:1041104. Edmonds, 3 S and Francesconi, K A Mar. Polit. Bull., 1981, 12:92105. Shiomi, K, Shinaga, A, Yamanaka, H and Kikuchi, T Bull. Jpn. Soc. Sci. Fish.,1983, 49:79106. Luten, J B, Riekwel-Booy, G, Van der Greeg, 3 and Ten Noever de Brauw, M CChemosphere, 1983, 12:131107. Maher, W and Butler, EAppi. Organomet. Chem., 1988, 2:191108. Cullen, W R and Dodd, M Appl. Organomet. Chem., 1989, 3:79109. Shiomi, K, Kakehashi, Y, Yamanaka, H and Kikuchi, TAppl. Organomet.Chem., 1987, 1:177110. Shiomi, K, Orii, M, Yamanaka, H and Kikuchi, TNippon Suisan Sukkaishi,1987, 53:103168111. Francesconi, K A, Edmonds, J S and Hatcher, B G Comp. Biochem. Physiol.,1988, 90C:313112. Lawrence J F, Michalik, P, Tam, G, Conacher, H B S J. Agric. Food Chem.,1986, 34:3 15113. Norm, H, Christakopoulos, A, Chemosphere, 1982, 11:287114. Norm, H, Ryhage R, Christakopoulos, A and Sandstrorn, M Chemosphere, 1983,12:299115. Shiomi, K, Shinagawa, A, Igarashi, T, Yamanaka, H and Kikuchi, TExperientia, 1984, 40:1247116. Norm, H, Christakopoulos, A and Sandstrom, M Chemosphere, 1985, 14:313117. Edmonds, J S and Francesconi, K A Experientia, 1982, 38:643118. Shibata, Y and Morita, MAgric. Biol. Chem., 1988, 52: 1087119. CuJien, W R and Nelson J C Appl. Organomet. Chem., 1993,120. Kirkbright, G F IN: Elemental Analysis ofBiological Material, IEAE, Vienna,1980121. Cantle, J E TN: Atomic Absorption Spectrometiy (Techniques andInstrumentation in Analytical Chemistry), Elsevier, 1982122. Welz, B IN Atomic Absorption Spectrometry, VCH, Weinhein, Deerfield Beach,FL, Basel, 1985123. Suzuki, M and Ohta K, Prog. Anal. At. Spectrosc., 1983, 6:49124. Subramanian, K S, Prog. Anal. At. Spectrosc., 1986,9:237125. Greenfield, S, Hieftje, G M, Omenetto, N, Scheeline, H and Slavin, A Anal.Chim. Acta, 1986, 180:69126. Slavin, W, Manning, D C and Carnrick, G R J. Anal. At. Spectrom., 1987, 2:201127. Godden, W B and Thomerson, D R Analyst, 1980, 105:1137169128. Nakahara, T Prog. Anal. At. Spectrosc., 1983, 6:163129. Rabbins, W B and Caruso, J A Anal. Chem., 1979, 51:889A130. Welz, B Chem. Brit., 1986, Febuary, p130131. Dedina, 3 Frog. Analyst. Spectrosc., 1988, 11:25 1132. Van der Veen, NO, Keukens, H J and Vos GAnal. Chim. Acta, 1985, 171:285133. Francesconi, K A, Micks, P, Stockton, R A and Irgolic, K J, Chemosphere, 1985,14:1443134. Raptis, S E,Wegscheider, W and Knapp, G MicrochimActa (Wien), 1981, 1:93135. Van Loon, 3 C IN: Selected Methods of Trace Analysis (Biological andEnvironmental Samples), Wiely, NewYork, 1985136. Maher, W A Mar. Biol. Lett., 1984, 5:47137. Edmonds, 3 S and Francesconi, K A Nature (London), 1977, 265:436138. Yamauchi, H and Yamamura, Y Toxicology, 1985, 34:113139. Kaise, T, Yamauch, H, Hirayama, T and Fukui, S Appl. Organomet. Chem.,1988, 2:339140. Andreae, M 0 Anal. Chem., 1977,49:820141. Ebdon, L, Walton, A P, Miliward, G E and Whitfield, M Appl. Organomet.Chem., 1987, 1:427142. Hinners, TA Analyst, 1980, 105:751143. Anderson, R K, Thompson, M and Culbard, EAnalyst, 1986, 111:1143144. Howard, AG and Arbab-Zavar, M H Analyst, 1981, 106:213145. Merijanian, A and Zingaro, R A Inorg. Chem., 1966, 5:187146. Rothery, E IN: Analytical Methods for Graphite Tube Atomizers, VarianTechtron, Victoria, Australia, 1982170147. Irgolic, K J IN: Hazardous Methods in the Environment Stoeppler M (ed),elsevier, Amsterdam, 1992148. Chakraborti, D, Irgolic, K 3 and Adams, F Intern. J. Environ. Anal. Chem., 1984,17:241149. Koreckova, J, Frech, W, Lundberg, E, Perssoon, J A and Cedergren, A Anal.Chim. Acta., 1981, 130:267150. Walsh, PR, Fasching, 3 L and Duce, R A Anal. Chem., 1976, 48:1014151. Riley, KWAtom. Spectrosc., 1982, 3:120152. Ediger, R E At. Absorpt. Newslett., 1975, 14:127153. Callio, S Atom. Spectrosc., 1980, 1:80154. Matousek, 3 P Prog. Anal. Atom. Spectros., 1981,4:247155. Punemans, F and Massart, D C MikrochimActa (Wien), 1984, 11:261156. Shan, X Q, Ni, Z M and Zhang, LAnai. Chim. Acta., 1983, 151:179157. Schiemmer, G and Weilz, B Spectrochim Acta, 1986, 41B: 1157158. Styris, D L, Prell, L J and Redfield, D A Anal. Chem., 1991, 63:503159. Welz, B Fresenius Anal. Chem., 1986, 325:95160. Holak, W Anal. Chem., 1969,41:1712161. Robbins, W B and Caruso, 3 A Anal. Chem., 1979, 51:894162. Braman, R S, Johnson, D L Foreback, C C, Ammons, 3 M and Bricker, 3 LAnai.Chem., 1977,49:621163. Braman, R S and Tompkins, M A Anal. Chem., 1978, 50:1088164. Yamamoto, Y and Kumamaru, T FreseniusAnal. Chem., 1976, 28 1:353165. Schmidt, F J and Royer, 3 C Anal. Lett., 1973, 6:17166. Mesman, B B and Thomas, T C Anal. Lett., 1975, 8:449167. Vijan, P N, Wood, G R At. Absorpt. Newslett., 1974, 13:33171168. Thompson, M, Palavanpour, B, Walton, S J, Kirkbright, G F Analyst, 1978,103:568169. Odanaka, Y, Tsuchiya, N, Matano, 0 and Goto, S Anal. Chem., 1983, 55:929170. Reimer, K JAppl. Organomet. Chem., 1989, 3:475171. Talmi, Y and Bostick, D TAnal. Chem., 1975, 47:2145172. Narasaki, H and Ikeda, MAnal. Chem., 1984, 56:2059173. Aggett,JandAspell,ACAnalyst, 1976, 101:341174. Pierce, F D and Brown, H R Anal. Chem., 1976,49:693175. Sandhu, S S and Nelson, P Anal. Chem., 1978, 50:322176. Thompson, M, Palavanpour, B, Walton, S J, Kirkbright, G F Analyst, 1978,103:705177. Welz, B and Meicher, M Analyst, 1984, 109:573178. Pruszkowska, E, Barrett, P, Ediger, R and Wallace, G At. Spectrosc., 1983,4:494179. Thompson and M, Palavanpour, B Anal. Chim. Acta, 1979, 109:25 1180. Peacock, C 3 and Singh, S C Analyst, 1981, 106:931181. Kirkbright, G F and Taddia, M Anal. Chim. Acta, 1978, 100:145182. Inui, T Terada, S and Tamura, H Freseniusz Anal. Chem., 1981, 305:189183. Baker, M D, Wong, P T 5, Chau, Y K, Mayfield, C I and Inniss, W E184. Epps, E A and Sturgis, M B Soil Sci. Soc. Am. Proc., 1939,4:215185. Woolson, E A and Keamey, P Environ. Sci. Technol., 1973, 7:47186. Merrill, W and French, D W Proc. Minn. Acad. Sci., 1964, 31:105187. Woolson, E A J. Agric. Food Chem., 1975, 23:677188. Thom, C and Raper, K B Science, 1932, 76:548189. Parris, GE and Brinckman, F EEnviron. Sci. Technol., 1976, 10:1128172190. Cullen, W R, McBride, B C and Pickett, A W App!. Organomet. Chem., 1990,4:119191. Cullen, W R and Nelson, JCApp1. Organomet. Chem., 1992, 6:179192. Herring, F G, Cullen, W R and Nelson, J C Bull. Magnetic Resonance, 1994,14:289193. Cullen, W R, Herring, F G and Nelson, 3 C Bull. Environ. Contam.. Toxicol.,1994, 52(2):171194. Kaise, T, Hanaoka, K and Tagawa S Chemosphere 1985, 16:255 1195. Berger, S and Kaever, M J Dasycladales, Georg Theme Verlag, Stuttgart K,New York, 1992,p163196. Shephard, D S Methods of Cell Physiology, Prescott, D ed., vol. IV, AcdemicPress, New York, 1970, p49197. Le, X C, Cullen, W R and Reimer K JAppl. Organomet. Chem., 1992, 6:161198. Le, X C, Cullen, W R and Reimer, K J Talanta, 1993,40:185199. Rothstein, A J. Gen. Physiol., 1963,46:1075200. Scarborough, G A Biochem. Biophys., 1975, 166:245201. DaCosta, E W B App!. Microbiol., 1972, 23:46202. Howard, A G, Arbab-Zavar, M H and Apte, S C Mar. Chem., 1982, 11:493203. Howard, A G, Arbab-Zavar, M H and Apte, S C Estuar. Coastal Mar. Sci.,1984, 19:493204. Howard, A G and Comber, S D W App!. Organomet. Chem., 1989, 3:509205. Andreae, MO, Asmode, J-F, Foster, P and Van’t dack L Anal. Chem., 1981,53:1766206. Berman, E Toxic Metals and Their Analysis, Heyden, London, 1980173207. Onishi, H Handbook of Geochemistry, Wedepohi, K H ed., Springer:Berlin,Heidelberg, New York, 1969; antinwny208. Schutz, D F and Turekian, K K Geochim. Cosmochim. Acta 1965, 29:259209. Braman, R S, Justen, L L, and Foreback, C C Anal. Chem., 1972,44:2195210. Fermanderz, F 3 J. At. Absorpt. Newsi., 1973, 12:93211. Andreae, M 0 Trace Metals in Seawaters (eds. Wong C S, Boule, E, Bruland, KW, Burton, 3 L, and Goldberg, E D). Plenum Press, New York and London212. Dodd, M, Grundy, S L, Reimer, K 3 and Cullen, W RAppl. Organomet. Chem.,1992, 6:207174APPENDIXDETERMINATION OF HYDRIDE-FORMING AND “HIDDEN” ARSENICALSIN THE SEAWATER SURFACE MICROLAYER BY USING MICROWAVEDIGESTION FOLLOWED BY HG-GC-AAA.1 INTRODUCTIONArsenobetaine (AB), arsenocholine (AC), tetramethylarsonium salts (Me4Asj,and arsenosugars, the so called “hidden” arsenicals, are known to be present in marineorganisms in significant concentrations.9440°The presence of “hidden” arsenicals inseawater has not been well documented, partly because of the low concentration of thesecompounds in the water column and the limitation of analytical techniques. It is possiblethat arsenobetaine, arsenosugars, and their breakdown products may in fact be present inthe water column.The seawater surface microlayer (about 50 tm in depth) represents an interfacebetween the atmosphere and ocean systems. Large microlayer phytoplanktonenrichments have been found. They frequently occur at densities 10-100 times greaterthan the underlying phytoplankton, and play an important role in the productivity ofmany waters. As a result, arsenic speciation in the microlayer may be markedly differentfrom that in the bulk seawater.A convenient batch-type hydride generation system which was developed todetermine trace amounts of hydride-forming arsenic compounds in seawater surfacemicrolayer samples collected from coastal sites of British Columbia (Canada). Whenused in conjunction with batch-type microwave digestion, this system can detect“hidden” arsenic species in these samples.175A.2 EXPERIMENTALA.2.1 Sample Collection and PretreatmentSeawater surface microlayer samples were collected from coastal sites of BritishColumbia (Canada) in April of 1989. A glass plate of convenient size (20 cm square byabout 4 mm thick) was introduced vertically through the surface. The glass plate wastaken out of the water, held by hand and the surface film and water layer adhering to theplate were removed from both sides with a Teflon wiper blade and collected in a Teflonbottle. Samples were preserved in Teflon bottles by rapid freezing on dry ice, and thenstored at -20°C prior to analysis.A.2.2 Analytical ProceduresThe apparatus for arsine production, collection, and determination is shown inFigure A. 1. The sample (30-50 mL) and the acid (3 mL of 4.0 M hydrochloric acid or 2mL of 4.0 M acetic acid) were introduced into the reaction vessel (250 mL Erlenmeyerflask) by using a pipette. The NaBH4 solution (4.0% w/v) was then introduced into thereaction vessel by using the peristaltic pump. The volatile arsines were then collected,separated, and quantified as previously described in Chapter 2. The optimum conditionsfor the determination of hydride forming arsenicals are shown in Table A. 1.A Sharp Model microwave oven operating at maximum power (500 W) wasused for sample digestion. The sample solution (30-50 mL) and 2-5 mL of potassiumpersuiphate solution (2.0-6.0%) were introduced into a 250 mL Erlenmeyer flask. Theflask was loosely covered with a Teflon cap, placed in the microwave oven, andirradiated for 2-5 min at full power. After digestion, 2-5 mL of ascorbic acid (10% wlv),176j:PenstalticPumpHeliumNaBH4‘I_____11r.iiDry IcelAcetoneWaterTrap__________AtomicAbsorptionISpectrometerGasChromatographI____ISampleMagneticStirrerFigureA.1Schematicdiagramofabatchtypehydridegeneration-gaschromatography-atomicabsorptionspectrometry.Table A.1 Optimum conditions for the determination of hydride-forming arsenicals in abatch type hydride generator___________________________In hydrochloric acid In acetic acidSample volume 50 mL 50 mLArsenic concentration 1 ppb 1 ppbAcid concentration 1 M 4 MAcid quantity 3 mL 2 mLNaBH4concentration (wlv) 4% 4%Peristaltic pump rate 2 mLfmin 2 mLfminReaction time 2 mm 2 mmArsenicals to be determined arsenate, arsenite, MMA arsenite, DMA, TMAO3 mL of 4.0 M HC1, and an appropriate amount of deionized water were added to theflask to bring the total volume to 50 mL. The samples were then cooled to roomtemperature prior to HG-GC-AA analysis by using the system described above. Theoptimum conditions for the decomposition of “hidden” arsenicals are displayed in TableA.2.A.3 RESULTS AND DISCUSSIONA.3.1 Determination of hydride-forming arsenicalsThe batch type hydride generator described here (Figure A. 1) uses a peristalticpump for the introduction of the NaBH4solution and a magnetic stirring bar for mixingof the reactants, thus improving the analysis efficiency, reducing the potentialoperational inconsistency, and significantly reducing the analysis time. A complete178analysis takes about 7 mm.Table A.2 Optimum conditions for the decomposition of “hidden” arsenicalsin a standard solution In a seawater sampleSample volume 50 mL 30 mLK2S08 concentration (wlv) 2% 6%K2S08 volume 2 mL 5 mLMicrowave digestion time 2 miii 5 mmAscorbic acid concentration (wlv) 10% 10%Ascorbic acid volume 2 mL 2 mLThe quantity of hydrochloric acid, acetic acid, and NaBH4 used for hydridegeneration was optimized by using the single parameter variation method. The peak areaof the signals from arsine, methylarsine, dimethylarsine and trimethylarsine was chosenas the response to be optimized. A 50 mL of 1 ppb each of arsenate (or arsenite), MMA,DMA, and TMAO aqueous solution was used for the optimization.The effect of hydrochloric acid (1.0 M), acetic acid (4 M), and NaBH4 on thedetermination of arsenate, arsenite, MMA, DMA and TMAO is shown in Figures A.2,A.3, A.4, A.5. The optimum conditions for the determination of these arsenicals areshown in Table A.1.The calibration curve for each arsenic compound is linear from 0.1-5 ppb of Asin a 50 mL of aqueous solution. In the presence of hydrochloric acid, the detectionlimits were determined as 0.3 ng (of arsenic) for arsenate, arsenite, and MMA; and 0.4ng for DMA and TMAO. In the presence of acetic acid, the detection limits were 0.3 ng1791101009080IVolume of 4 N HC1 (mL)Figure A.2 Effect of the concentration of hydrochloric acid on the determination ofhydride-forming arsenicals. The NaBH4solution (4.0% w/v) was added to the reactionsystem at a rate of 2 mL/min for 3 minutes. (0) arsenite (•) arsenate (V) MMA (V)DMA (D)TMAO0 1 2 3 4 5 6 71801 401 20100U,D 800-600G)0402006Volume of Acetic Acid (mL)Figure A.3 Effect of the concernration of acetic acid on the determination of hydrideforming arsenicals. The NaBH4solution (4.0% w/v) was added to the reaction system ata rate of 2 mL/min for 3 minutes. (0) arsenite (•) arsenate (V) MMA (V) DMA (D)TMAO0 1 2 3 4 5181110 -100 -90 -80- /70 - /7v—vI E7Sodium Borohydride Addition Time (mm)Figure A.4 Effect of NaBH4 concentration on the determination of hydride-formingarsenicals. A 3 mL of 4.0 M HC1 was added to the reaction system to acidify the sample.(0) arsenite (•) arsenate (V) MMA (V) DMA (D) TMAO182140 -_________________________________6OflSodium Borohydride Addition Time (mm)Figure A.5 Effect of NaBH4 concentration on the determination of hydride-formingarsenicals. A 2 mL of 4.0 M acetic acid was added to the reaction system to acidify thesample. (0) arsenite (•) arsenate (V) MMA (Y) DMA (C) TMAO183for arsenite, DMA, and TMAO; and 0.5 ng for MMA.The reproducibility was 6% for arsenate and arsenite, 5% for MMA, and 10%for DMA and TMAO in the presence of either hydrochloric acid or acetic acid.The interference from the seawater matrix in the determination of arsenate,arsenite, MMA, DMA, and TMAO was eliminated by adding 3 mL of 10% (w/v)EDTA. Calibration curves obtained from standard solutions of arsenicals were used forquantitative analysis of seawater samples.Determination of the total amount of inorganic arsenic (arsenate and arsenite)and the amount of MMA was achieved when the seawater surface microlayer sampleswere acidified by using hydrochloric acid. The amount of arsenite and DMA in theseawater surface microlayer samples was quantified when acetic acid was used. Theresults are shown in Table A.3.Tabie A.3 The concentrations of hydride-forming arsenicals (ppb) in seawater surfacemicrolayer samplesStations Arsenite Arsenate MMA DMA Total*RB1 0.30 0.12 <L.O.D. d 0.42*R19 0.35 0.35 < L.O.D. d 0.70*PR1 0.22 0.39 < L.O.D. d 0.61*MSA1 0.26 0.37 < L.O.D. d 0.63*MSft. 0.27 0.12 <L.O.D. d 0.39*RS1 0.25 0.36 < L.O.D. d 0.61*RS3 0.21 0.34 < L.O.D. d 0.55d: detectableL.O.D.: limit of detectionRB1: Desolation Sound; Rl9: Rupert Inlet; MSA1: Moira Sound, Alaska;MSA2: Moira Sound, Alaska; RS1: Rennell Sound; RS3: Rennell Sound;184A3.2 Determination of “hidden” arsenic speciesInitially the aqueous solutions (1 ppb, 50 mL) of arsenobetaine, arsenocholine,tetramethylarsonium ion were used to establish the optimum digestion conditions. Thesingle parameter variation method was used to study the factors affecting thedecomposition efficiency, in particular the digestion time and the amount of K2S08.The decomposition efficiency is presented as the arsenate peak area of the signals fromthe “hidden” arsenicals relative to that from the same amount of arsenate, thus thearsenate peak area produced from 50 ng of AB was compared with the peak area from50 ng of arsenate.The effect of microwave irradiation time and the amount of K2S08 ondigestion efficiency is shown in Figures A.6 and A.7.Any excess K2S08 and NaOH remaining in the flask after the microwaveassisted digestion could affect the hydride generation reaction since they can react withthe NaBH4 and HC1 added to the reaction vessel. The addition of ascorbic acid (2 mL,10% w/v) which reacts with both K2S08and NaOH, could solve this problem. Theoptimum conditions for the decomposition of “hidden” arsenicals are shown in TableA.2.The calibration curve of arsenate was used for quantitative analysis of the“hidden” arsenicals.Most environmental samples contain a complex matrix, which could affect thedecomposition efficiency of “hidden” arsenicals in the microwave assisted digestion.Initial attempts to achieve a complete decomposition of organoarsenicals in a seawatersample matrix were made by optimizing quantity of persulfate, and the microwavedigestion time. A known amount (50 ng) of AB, AC, or Me4As+ were spiked into aseawater matrix (50 mL) and recoveries were evaluated. The effect of the amount of185potassium persulfate and the digestion time on the decomposition of AB, AC, Me4As+is shown in Figures A.8 and A.9. The maximum recoveries of the “hidden” arsenicals in50 mL of seawater is 65-75%. Complete recovery is achieved when the sample volumeis reduced to 30 mL. The optimum conditions for the complete decomposition of AB,AC, and Me4As are shown in Table A.2.Arsenate prepared in distilled water was chosen as the standard for quantitativeanalysis in the work.By using microwave digestion-HG-GC-AA, it was found that some seawatersurface microlayer samples contain “hidden” arsenic as shown in Table A.4. In thesestudies, the samples which were filtered through a 0.45 p.m pore-size glass filter show asimilar concentration of arsenic to those without filtration. This indicates that the“hidden” arsenicals are dissolved in seawater, and the amount of arsenicals present inphytoplankton is insignificant.Table A.4 The concentrations of arsenic (ppb) in seawater surface microlayer sampleswith and without microwave assisted digestionStations RB1 R19 PR1 MSA1 MSA2 RS1 RS3Without microwavedigestion 0.42 0.70 0.61 0.63 0.39 0.61 0.55With microwavedigestion 0.45 0.66 0.79 0.65 0.59 0.60 0.57186>‘C)C)wC0.U)00E0C)0)10090807060_50403020100Digestion Time (mm)Figure A.6 Effect of microwave digestion time on the decomposition efficiency in thepresence of 2 mL of 2.0% (w/v)K2S08.The sample solution (50 mL) contained 50 ngeach of AB (A), AC (A), and Me4As (a).1870 1 2 3 4 5 6 7ioo5o1:7Volume of Potassium Persulfate (mL)Figure A.7 Effect of the quantity of potassium persulfate (2.0% wlv) on decompositionefficiency. The sample solution (50 mL) contained 50 ng each of AB (ba), AC (A), andMe4As (0).1881009080>70ci)C)60w.2 50(1)00- 40E0C-)ci, 3020100Digestion Time (mm)Figure A.8 Effect of microwave digestion time on the decomposition efficiency in thepresence of 5 mL of 6.0% (wlv)K2S08.A 50 mL of seawater sample was spiked with50 ng each of AB (ti), AC (A), and Me4As ().0 1 2 3 4 5 6 7189100 -90 -60-50- /40-E0C)G) 30-20 -10 -0 1L I I I0 1 2 3 4 5 6 7 8Volume of potassium persulfate (mL)Figure A.9 Effect of the quantity of potassium persulfate (6.0% w/v) on decompositionefficiency. A 50 mL of seawater sample was spiked with 50 ng each of AB (z), AC (A),and Me4As+ (c), and irradiated in the microwave oven for 5 mm.190

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059605/manifest

Comment

Related Items