Arsenic compounds in the haemolymph of the Dungeness crab, Cancer magister, as determined by using HPLC on-line with inductively coupled plasma mass spectrometry

Abstract

Arsenobetaine, two arsenosugars, dimethylarsinate and several unidentified arsenic species were detected in extracts of the haemolymph of the Dungeness crab, Cancer magister, by using HPLC-ICP-MS. This is the first report of the presence of arsenosugars in the haemolymph/blood of marine animals. Total, extractable and residual arsenic concentrations were determined by ICP-MS. The concentration of total arsenic was in the range of 1.4–3.8 μg ml⁻¹. Nearly all (98%) the arsenic was found to be extractable, and accounted for primarily by arsenobetaine, two arsenosugars and dimethylarsinate. The results demonstrate that arsenic compounds present in the diet of crabs are not fully metabolized in the gut. They are, at least partly, taken up into the haemolymph. The concurrence of arsenobetaine and arsenosugars suggests that the use of repeated haemolymph sampling in crustaceans could facilitate investigations into the kinetics of the biotransformation pathways of arsenic compounds. Finally, the present study clearly demonstrates the unique capabilities of HPLC-ICP-MS for the detection and identification of minor arsenic components amongst the predominant arsenobetaine.

---

Introduction

Arsenic speciation and ecotoxicology in the marine environment has been the topic of several recent reviews, and the presence of organoarsenic compounds in marine algae, molluscs, crustaceans and fish is well documented.^1–4 In seawater the main species present are inorganic arsenate (iAs(V)) and arsenite (iAs(III)), with the organoarsenic compounds methylarsonate (MA) and dimethylarsinate (DMA) as minor constituents. In marine algae the arsenic-containing ribosides (referred to as arsenosugars) dominate, with inorganic arsenic being less important. Following the isolation and identification of arsenobetaine (AB) in the tail muscle of the western rock lobster, Panulirus longipes cygnus, by Edmonds et al.⁵ the ubiquitous presence of AB in marine animals has been well established. Although AB is the predominant form of arsenic in most marine animals, herbivorous and filterfeeding molluscs may contain high levels of tetramethylarsonium ion (TETRA) or arsenosugars. Traces of inorganic arsenic, DMA, trimethylarsine oxide (TMAO), arsenocholine (AC) and a second arsenobetaine ((CH₃)₃As⁺CH₂CH₂COO⁻) have also been reported for marine animals.^3,6,7 Though arsenic clearly bioaccumulates in marine organisms to levels significantly higher than seawater concentrations, i.e. approximately 0.5–2 μg l⁻¹,^8,9 no evidence for biomagnification along food chains exists.^3,4 The structure of arsenic compounds in the marine environment of relevance to the present study are given in Table 1.

Table 1 Arsenic compounds in the marine environment of relevance to the present study

Arsenite	iAs(III)	AsO3^3−
Arsenate	iAs(V)	AsO4^3−
Methylarsonate	MA	CH3AsO3^2−
Dimethylarsinate	DMA	(CH3)2AsO2^−
Trimethylarsine oxide	TMAO	(CH3)3AsO
Tetramethylarsonium ion	TETRA	(CH3)4As^+
Arsenobetaine	AB	(CH3)3As^+CH2COO^−
Arsenocholine	AC	(CH3)3As^+CH2CH2OH
Arsenic-containing ribosides (arsenosugars):
	Arsenosugar	Y
	1	–OH
	2	–OPO3^−CH2CH(OH)CH2OH
	3	–SO3^−
	4	–OSO3^−

---

Despite the substantial effort put into descriptive studies of arsenic speciation in the marine environment, comparatively little is known about the mechanisms or kinetics of arsenic transformations by marine organisms. Furthermore, the source of the ubiquitous AB remains elusive.^3,4 Although pathways for the synthesis or transformation of arsenic compounds by marine microorganisms in sediments,^10,11 algae^2,12,13 and animals³ have been proposed, much work remains to substantiate these pathways and fully elucidate the cycling of arsenic in the marine environment.

Previous studies attempting to clarify the metabolic transformations of arsenic compounds in marine animals have focussed on arsenic speciation in various tissues but have largely ignored the blood/haemolymph compartment.^14–16 In these investigations, discrepancies have been found between the arsenic species present in seawater or diet and the species present in organs. Experiments have been designed to quantify the tissue concentrations in freshly caught animals, or at one or more time points, typically separated by days or weeks, following water or dietary exposure in the laboratory. These experiments provide essentially a semi-static picture of internal arsenic distribution; they neither monitor kinetics of arsenic metabolism nor provide detailed information on the site(s) of arsenic transformation.

The circulatory system of marine invertebrates provides the means for transport of absorbed dietary constituents or compounds taken up from the seawater to the various tissues, and the presence and kinetics of certain arsenic compounds in the blood/haemolymph could provide evidence for which organs are responsible for the alterations in arsenic speciation. Decapod crustaceans have easily accessible haemolymph compartments which can be repeatedly sampled at short intervals (minutes or hours), consequently crabs have high potential as model organisms in the study of arsenic transformation pathways. Crustaceans have been reported to contain primarily AB, especially in muscle tissue,^5,17,18 despite the fact that the diet, which for various species may include algae, molluscs, crustaceans or fish, often contains high concentrations of other arsenic compounds. The diet of decapods would appear to be the most likely source of AB, given the high dietary concentrations of various arsenic compounds, including AB and arsenosugars, which have been suggested as precursors of AB.³ Neither AB nor arsenosugars have been detected in seawater, although this may be due to deficiencies in the analytical methods applied in previous studies as discussed by Francesconi and Edmonds.³

The aim of the present investigation was to determine the arsenic speciation in the haemolymph of crabs. The Dungeness crab, Cancer magister, was chosen for this study, because its normal diet includes a wide range of animals, e.g. bivalves, crustaceans and fish,¹⁹ resulting in the ingestion of several of the common arsenic compounds. The large size of the crabs (0.9–1 kg in the present study) furthermore ensures that ample haemolymph samples can be obtained, since approximately 25% of the whole body weight is accounted for by haemolymph.

Experimental

Experimental animals and haemolymph sampling

Three freshly caught, live, male, intermoult Dungeness crabs, Cancer magister, were obtained locally (Vancouver) from a shellfish retailer in late February 2000. The whole body wet weights and carapace widths of crab 1, 2 and 3 were: 1001 g, 19.7 cm; 984 g, 18.5 cm; 905 g, 18.0 cm. The crabs were immediately transported to the laboratory and were kept on ice for a short period until completion of the haemolymph sampling. Haemolymph samples were drawn through the arthrodial membrane of the pereiopods by using disposable syringes. Approximately 50 ml was obtained from each crab. Three 0.5 ml haemolymph subsamples for total arsenic determination, and three 8 ml haemolymph subsamples for arsenic speciation analysis, were taken from each crab.

Extraction procedure

The 8 ml haemolymph samples were freeze-dried in 15 ml centrifuge tubes for extraction, and the dry weight was determined. The dry weight per ml haemolymph in crab 1, 2 and 3 were (mean ± SD, n = 3): 102 ± 1, 63 ± 1 and 74 ± 2 mg ml⁻¹ haemolymph, respectively. One extraction round included the addition of 5 ml methanol ∶ deionized water (1 ∶ 1, v ∶ v), vigorous mixing, sonication for 10 min, centrifugation (bench top) for 10 min and the transfer of the supernatant by Pasteur pipette to a 100 ml round bottom flask. The residue was extracted an additional four times following the same procedure. After the last extraction round the residue was freeze-dried prior to total arsenic determination. The five supernatant fractions were combined in the round bottom flask and the methanol was evaporated. The evaporated sample was made up to an extract volume of 7 ml by the addition of deionized water. Extracts were stored at −20 °C.

Digestion procedure

Samples for total arsenic analysis, i.e. 0.5 ml whole haemolymph, 0.5 ml extract and 100 mg dry weight of the freeze-dried residue, were transferred to 20 ml glass test tubes. Two ml of concentrated nitric acid and a few Teflon boiling chips were added to each tube. The tubes were transferred to a block heater and the temperature increased in 10 °C steps every hour, starting at 70 °C and reaching 150 °C. At 150 °C the samples were evaporated to dryness over the next 1 to 2 d. Whole haemolymph samples were then redissolved in 4 ml 1% nitric acid containing 5 ng Rh ml⁻¹, while extract and residue samples were redissolved in 2 ml and 4 ml, respectively, in the same rhodium-nitric acid solution. The rhodium served as an internal standard during ICP-MS analysis. The redissolved digests were filtered (0.45 μm), and stored at −20 °C.

Total arsenic analysis by inductively coupled plasma mass spectrometry (ICP-MS)

The redissolved digests were diluted appropriately with the rhodium-nitric acid solution and analysed for total arsenic by using ICP-MS. A VG Plasma Quad 2 Turbo Plus ICP-MS (VG Elemental, Fisons Instrument), equipped with a SX 300 quadropole mass analyser, a standard ICP torch, and a de Galan V-groove nebulizer, was used. Samples were analyzed in the “peak jump” mode. Standard interference corrections for ⁴⁰Ar³⁵Cl⁺ at m/z 75, i.e. at the same mass as ⁷⁵As, were made by monitoring m/z 75, 77 (⁴⁰Ar³⁷Cl⁺ and ⁷⁷Se⁺) and 82 (⁸²Se⁺). Signals were corrected according to the internal rhodium standard. Recoveries were checked using certified reference materials. TORT-2 Lobster Hepatopancreas Reference Material, NRC of Canada, was found to contain 21.8 ± 2.7 μg As g⁻¹ (mean ± SD, n = 5), compared to the certified value of 21.6 ± 1.8 μg As g⁻¹; DORM-2 Dogfish Muscle Reference Material, NRC of Canada, to contain 18.2 ± 1.9 μg As g⁻¹ (mean ± SD, n = 13), certified value of 18.0 ± 1.1 μg As g⁻¹; Oyster Tissue Standard Reference Material 1566a, NIST, USA, to contain 13.7 ± 1.2 μg As g⁻¹ (mean ± SD, n = 15), certified value of 14.0 ± 1.2 μg As g⁻¹ dry weight.

Arsenic speciation analysis by HPLC-ICP-MS

Anion pairing (AP) reversed phase (RP) HPLC and cation pairing (CP) RP HPLC were used to separate a series of 12 anionic, cationic and neutral arsenic species of known structure (arsenic standards). A Discovery C₁₈ column (2.1 mm i.d. × 5 cm length, 5 μm packing material, Supelco) with a mobile phase flowing at 0.7 ml min⁻¹, containing 5 mM tetrabutylammonium hydroxide (TBAH) adjusted to pH 6.28 with malonic acid, was used for the AP-RP HPLC separations. A Luna C₁₈ column (1 mm i.d. × 5 cm length, 5 μm packing material, Phenomenex), with a mobile phase flowing at 0.2 ml min⁻¹ containing 10 mM heptanesulfonate (pH 2.6), was used for CP-RP HPLC separations. The HPLC outlet was connected directly to the Micromist nebulizer (Glass Expansion, Australia) of the ICP-MS (VG PlasmaQuad 3; TJA Solutions, Winsford, Cheshire, UK) via a 0.13 mm i.d. × 90 cm PEEK tubing.

Chemicals

All chemicals were of analytical grade unless otherwise stated: sodium arsenate heptahydrate (Na₂HAsO₄·7H₂O, Sigma), arsenic(III) oxide (As₂O₃, Alfa products), methylarsonate (CH₃AsO₃²⁻, Pfalz and Bauer, Stamford), dimethylarsinate ((CH₃)₂AsO₂⁻, Aldrich), methanol (HPLC grade, Fisher), tetrabutylammonium hydroxide (TBAH: 20 wt.%, Aldrich), malonic acid (BDH), nitric acid (69%, sub-boiling distilled, Seastar Chemicals), and rhodium solution (ICP standard, 1000 μg ml⁻¹ in 20% HCl, Specpure, Alfa). Additional arsenic standards included dimethylarsinoylriboside derivatives (arsenosugars 1, 2, 3, 4), isolated from biological sources, purified, and characterised as previously described.^20,21 These were kindly donated by Dr K. A. Francesconi (Institute of Chemistry, Karl-Franzens-University, Graz, Austria). AB,⁵ AC,²² TMAO,²³ and TETRA iodide²⁴ were synthesized using literature methods.

Glass- and plasticware were cleaned by soaking in 2% Extran solution overnight, rinsing with water, then deionized water. This was followed by soaking in 0.1 M HNO₃ solution overnight, rinsing with deionized water and air-drying.

Results and discussion

Total arsenic concentration in the haemolymph of the Dungeness crab

The concentrations of total, extractable and residual arsenic in the haemolymph of the three crabs are presented in Table 2. The concentration of total arsenic was in the range 1.4–3.8 μg As ml⁻¹ and nearly all the arsenic was found to be extractable (98 ± 1%, mean ± SD, n = 3, calculated as extractable As/(extractable As + residual As) × 100%). The total arsenic concentration in the haemolymph of the Dungeness crab in the present study compares well with values for the shore crab, Carcinus maenas.²⁵ In the 3 crabs used in the present investigation an approximately 3-fold difference between the lowest and the highest haemolymph concentration was observed. Variability in trace element concentrations in the tissues of crustaceans, and marine invertebrates in general, are well known, and can often be attributed to a combination of factors, e.g. bioavailability of the element affected by physiochemical conditions of the surrounding medium,²⁶ physiological condition,^27,28 reproductive status or the crustacean moult cycle.²⁹

Table 2 Concentrations of total, extractable and residual arsenic in the haemolymph of three Dungeness crab. Mean ± SD/μg As ml⁻¹, n = 3

	Total	Extractable	Residual
Crab 1	2.12 ± 0.02	2.11 ± 0.02	0.060 ± 0.003
Crab 2	1.35 ± 0.06	1.11 ± 0.04	0.023 ± 0.002
Crab 3	3.77 ± 0.47	3.68 ± 0.10	0.059 ± 0.007

---

Arsenic speciation in the haemolymph of the Dungeness crab

Arsenic speciation in the haemolymph extracts was carried out by comparing the retention times of the arsenic containing peaks (obtained under AP-RP and CP-RP HPLC conditions), with the retention times obtained for 12 arsenic standards. Initially the haemolymph extract was analysed using AP-RP HPLC-ICP-MS. Under these chromatographic conditions 5 arsenic-containing peaks were observed (Fig. 1). The major arsenic peak (containing 97.9% of the total As in the extract) had a retention time of 35 s, and the remaining four minor peaks eluted at 51, 63, 132 and 163 s. The haemolymph extract was also analysed using CP-RP HPLC-ICP-MS. Under these conditions 7 arsenic-containing peaks were observed (Fig. 2). The major arsenic containing peak (accounting for 94.5% of the total As in the extract) occurred at 70 s, while the remaining six minor peaks appeared at 17, 34, 102, 159, 232 and 311 s. The retention times of the 12 arsenic standards were also determined under identical conditions (both AP- and CP-RP HPLC). The pair of retention times obtained for each of the 12 arsenic standards are represented by single points (solid circles) in Fig. 3. The 12 arsenic standards used in this study account for all the major arsenic species known to exist in marine organisms.¹ Also included in Fig. 3 are the retention times obtained for the arsenic species detected in the haemolymph extracts. These retention times are listed on the plot axes (Fig. 3); 7 retention times on the CP-RP axis and 5 retention times on the AP-RP axis. Dotted lines drawn from these retention times allow for a direct comparison with retention times obtained for each of the 12 arsenic standards. Positive identification of an arsenic species is achieved when its retention times coincide with the retention times of an arsenic standard, i.e. a point at which dotted lines intercept as well as coincide with a solid circle. All the arsenic present in the extract was accounted for in the chromatograms obtained.

Fig. 1 Anion pairing RP HPLC-ICP-MS chromatogram of the haemolymph extract originating from the Dungeness crab. Signal obtained while monitoring m/z 75. Numbers on peak tops represent retention times in seconds.

Fig. 2 Cation pairing RP HPLC-ICP-MS chromatogram of the haemolymph extract originating from the Dungeness crab. Signal obtained while monitoring m/z 75. Numbers on peak tops represent retention times in seconds.

Fig. 3 Retention times for 12 arsenic standards (solid circles) obtained using AP- and CP-RP HPLC-ICP-MS. Numbers on the plot axes (left and bottom) signify retention times observed for the arsenic species present in the haemolymph extracts. A logarithmic scale is used for the CP-RP retention times. Numbers on the top and right axes indicate the % of total As present in the extract which gives rise to each chromatographic peak.

The data presented in Fig. 3 suggests AB, arsenosugar 1 and 2, and DMA to be present in the haemolymph extract analysed. Under AP conditions the major arsenic species in the extract occurs at 35 s (abbreviated as: AP-35), and under CP conditions it occurs at 70 s (abbreviated as: CP-70). These retention times intercept at the point where the retention times of standard AB occur. This strongly suggests that AB is the major arsenic species present in the haemolymph extract. Arsenic compound(s) with retention times at AP-35 and CP-311 matched the retention times exhibited by the arsenosugar 1 standard. In addition a minor arsenic species occurring at AP-51 and CP-34 matched the retention times exhibited by the DMA standard. Arsenosugar 2 was identified to be present in the haemolymph extract as its retention times coincided with those of the extract peaks occurring at AP-63 and CP-17. Results obtained from spiking experiments gave further support to the above identifications. TMAO was investigated as the possible species occurring at AP-35 and CP-232, however, closer examination of retention time data did not show a satisfactory match. TMAO spiked into the extracts revealed separate chromatographic peaks. Also hydride generation ICP-MS analysis of the extract revealed only 0.3% of the total arsenic in the extract readily formed volatile hydrides. If species AP-35/CP-232 was TMAO then the arsenic forming volatile hydrides would be considerably higher than 0.3% of the total arsenic. Finally the chromatographic behaviour of the arsenic species occurring at AP-132, AP-163, CP-102, CP-159 and CP-232 did not match that of any of the 12 arsenic standards investigated in this study and therefore no suggestions as to their identity could be put forward.

Biotransformation of arsenic compounds has been investigated previously in crustaceans. Hunter et al.¹⁴ provided data for the shrimp Crangon crangon supporting that the major source of arsenic compounds in crustaceans is their food. AB ingested by the shrimp was accumulated unchanged, as was the case for iAs(V), although a reduction of some of the iAs(V) to iAs(III) could not be excluded. No methylation of iAs(V) was observed. Ingested TMAO was metabolized predominantly to dimethylarsinate.¹⁴ The uptake and transformation of AB and arsenosugars fed to Crangon crangon was studied by Francesconi et al.¹⁵ While AB was efficiently retained (57%) as unchanged compound, only small amounts (0.9%) of a dimethylated arsenosugar was accumulated, mainly unchanged, by the shrimp, and no transformation into AB was observed. A trimethylated arsenosugar was more readily accumulated (4.2%), and subsequently approximately half was biotransformed into AB.¹⁵

The present study demonstrates that arsenic compounds in the diet of crabs are not fully metabolized in the gut and are at least partly taken up. AB and small amounts of two arsenosugars and DMA along with several unidentified arsenic species are detectable in the haemolymph. The presence of arsenosugars in the haemolymph/blood of marine animals has not been reported previously. It is possible that arsenosugars may have been overlooked during speciation analysis due to the large difference in concentration compared to AB and the present study clearly demonstrates the unique capabilities of HPLC-ICP-MS for the detection and identification of minor arsenic components amongst the predominant AB. It should, however, be stressed that identification of arsenic species by ICP-MS is only possible when matching standard compounds are available.

The concurrence of AB, arsenosugars and DMA in the haemolymph of Dungeness crabs suggests a possible avenue of future research into the biotransformation pathways of arsenic compounds. The use of repeated haemolymph sampling in crustaceans would facilitate investigations of the kinetics of the metabolism of arsenic species following, for example, feeding on arsenic-containing food or injection of arsenic compounds.

Acknowledgements

The authors wish to thank Mr Bert Mueller, Department of Oceanography, University of British Columbia, for his technical expertise in relation to the ICP-MS analysis. We are grateful to NSERC Canada for financial support.

References

1. W. R. Cullen and K. J. Reimer, Chem. Rev., 1989, 89, 713.
2. D. J. H. Phillips, Aquat. Toxicol., 1990, 16, 151.
3. K. A. Francesconi and J. S. Edmonds, Adv. Inorg. Chem., 1997, 44, 147.
4. J. M. Neff, Environ. Toxicol. Chem., 1997, 16, 917.
5. J. S. Edmonds, K. A. Francesconi, J. R. Cannon, C. L. Raston, B. W. Skelton and A. H. White, Tetrahedron Lett., 1977, 18, 1543.
6. E. H. Larsen, G. Pritzl and S. H. Hansen, J. Anal. At. Spectrom., 1993, 8, 1075.
7. K. A. Francesconi, S. Khokiattiwong, W. Goessler, S. N. Pedersen and M. Pavkov, Chem. Commun., 2000, 1083.
8. M. O. Andreae, Deep-Sea Res., 1978, 25, 391.
9. S. J. Santosa, S. Wada and S. Tanaka, Appl. Organomet. Chem., 1994, 8, 273.
10. J. S. Edmonds, K. A. Francesconi and J. A. Hansen, Experientia, 1982, 38, 643.
11. K. A. Francesconi, J. S. Edmonds and R. V. Stick, Appl. Organomet. Chem., 1992, 6, 247.
12. J. S. Edmonds and K. A. Francesconi, Experientia, 1987, 43, 553.
13. J. S. Edmonds, Bioorg. Med. Chem. Lett., 2000, 10, 1105.
14. D. A. Hunter, W. Goessler and K. A. Francesconi, Mar. Biol., 1998, 131, 543.
15. K. A. Francesconi, D. A. Hunter, B. Bachmann, G. Raber and W. Goessler, Appl. Organomet. Chem., 1999, 13, 669.
16. J. Kirby and W. Maher, Appl. Organomet. Chem., 2002, 16, 108.
17. K. A. Francesconi and J. S. Edmonds, Comp. Biochem. Physiol., 1987, 87C, 345.
18. E. H. Larsen, C. R. Quétel, R. Munoz, A. Fiala-Medioni and O. F. X. Donard, Mar. Chem., 1997, 57, 341.
19. B. G. Stevens, D. A. Armstrong and R. Cusimano, Mar. Biol., 1982, 72, 135.
20. K. A. Francesconi, J. S. Edmonds, R. V. Stick, B. W. Skelton and A. H. White, J. Chem. Soc., Perkin Trans. 1, 1991, 2707.
21. K. A. Francesconi, J. S. Edmonds and R. V. Stick, J. Chem. Soc., Perkin Trans. 1, 1992, 1349.
22. K. J. Irgolic, T. Junk, C. Kos, W. S. McShane and G. C. Pappalardo, Appl. Organomet. Chem., 1987, 1, 403.
23. A. Merijanian and R. A. Zingaro, Inorg. Chem., 1966, 5, 187.
24. W. R. Cullen and M. Dodd, Appl. Organomet. Chem., 1989, 3, 401.
25. J. L. Andersen and M. H. Depledge, Mar. Biol., 1994, 118, 285.
26. P. S. Rainbow, J. Mar. Biol. Assoc. UK, 1997, 77, 195.
27. P. Bjerregaard, Mar. Biol., 1990, 106, 199.
28. P. Bjerregaard, Comp. Biochem. Physiol., 1991, 99A, 75.
29. M. Bondgaard, U. Nørum and P. Bjerregaard, Mar. Biol., 2000, 137, 995.

---