Arsenic's not so bad!

Research profile of Kevin A. Francesconi

I was born in 1953 in Perth, a very pleasant city on the isolated west coast of Australia. Although my heritage is Italian, my parents attempted the fast track at assimilation—hence the incongruous forename Kevin! They had three sons; I was the third, and my father, a few days after my birth in response to constant questioning about the sex of the new child, flippantly replied “its another Jack”. The name stuck, and it took over 40 years and two new countries to finally loosen the bond. I married young (to Patricia) and we quickly had two daughters (Kristen, now 29, and Danielle 26) who have recently continued the female line each contributing a daughter (Skye and Summer!—my father was not involved in the naming). I currently live in Graz, a delightful small city in the south-east corner of Austria; with Kristen and Skye living in Melbourne and Danielle and Summer on the northwest Australian coast, the family is geographically challenged.

I studied Applied Chemistry at Curtin University of Technology, and towards the end of my Bachelor degree I carried out a small research project as part of a group collaborating with the Western Australian Marine Research Laboratory (WAMRL) to develop a synthetic bait for the lucrative rock lobster fishing industry. Our job was to extract and identify the chemical attractants of natural baits and attempt to synthetically reconstruct these in a suitable slow release matrix. It was my first experience of isolating and identifying compounds—something I was to do a lot of in the following years, and still enjoy doing today.

Metals and metalloids in seafoods

My contact with WAMRL resulted in my being offered a temporary job there as an analyst shortly after graduating in 1975. The position, initially for three months, was extended to nine, and then to 18 months; I stayed for over 20 years. During this time at WAMRL, I also undertook part-time studies in chemistry at the University of Western Australia (UWA) for an MSc (1984) and a PhD (1991).

The marine laboratory was housed in a white concrete building perched on a low cliff overlooking the stunningly blue and beautiful Indian Ocean. I instantly liked the place, not just because of its superb location. My immediate boss was John Edmonds, a recently appointed research chemist who was charged with the responsibility of solving current and foreseeable pollution problems related to metal contamination of the marine environment, and their effects on fisheries products, either directly by harming fish stocks, or indirectly by accumulating in seafood to levels that contravened human health regulations. John was a quiet laconic man with formidable intellect and a matching and infectious passion for science. This was the time when fisheries worldwide were suffering from bad press because of high mercury levels in many species of fish. Furthermore, the newly evolving field of speciation analysis had conclusively shown that most of the fish mercury was present in a methylated form, and methylmercury was known to be extremely toxic. Most countries responded to this threat to human health by imposing a maximum permissible concentration of 0.5 mg Hg kg⁻¹ in fish. Although this restriction did not have a major direct effect on the Australian finfish industries, the repercussions were serious for the valuable shellfish fisheries based on lobsters and prawns because these organisms were known to contain extremely high concentrations of naturally acquired arsenic and cadmium. Our Department head saw the significance of these results and their potential damage to Australian fisheries, and he issued the simple directive to us to determine the chemical form of arsenic in seafood. We were provided with a laboratory and some basic equipment. And, importantly, time—because the problem had remained unsolved for 50 years,¹ we were under no pressure to produce a quick answer. On reflection, I see this as the perfect environment in which to tackle such a problem, and to begin a research career.

The arsenical in seafoods—common but recalcitrant

Our basic equipment was a Varian atomic absorption spectrometer with hydride generation, and we adapted a published technique for determining arsines and applied it to marine animals, mainly seafood samples. It quickly became clear that the native arsenical in lobster and other seafoods did not give volatile arsines without vigorous chemical pre-treatment. Following heating with concentrated NaOH solution, however, a mixture of di- and tri-methylarsines was formed, and the observation² that this mixture of arsines was fairly constant for different types of seafood suggested that the native arsenical was the same in each case. Over time we accumulated useful data about the properties and distribution of the native arsenic compound, but we found it difficult to make significant progress.

At this time we met Jack Cannon, a classical natural products chemist at UWA; I recall a second meeting held in Jack's office. Jack, a chain smoker, was seated behind his desk puffing on his Camel non-filters while John related some of the fundamental data we had accumulated about the unknown arsenic compound. Jack became very interested; he began scribbling and puffing with equal and increasing intensity, and soon could hardly be seen through a cloud of smoke. The scribbling stopped, Jack let out a chuckle, and as the smoke cleared he was once again visible tapping his pencil at the bottom of a page full of tiny arrows and scratchy calculations. “It might be possible” he mused, “if we start with 5 kg of lobster tail meat and do this-that-this, and assuming that-this-that, we could end up with about 20 mg of pure compound to crystallise for X ray analysis”. The way Jack described it sounded simple enough—why hadn't we thought of that!

Arsenobetaine—the form of arsenic in lobster

Soon after, Jack made the short trip to the lobster processing plant at Fremantle harbour. Here, in the early evening, the fishing fleet unloads its daily catch: the still-live lobsters are immediately beheaded and the severed lobster tails stored on ice awaiting shipment abroad. The next morning, Jack, safely back at UWA, proudly displayed his catch of lobster tail meat—he described the previous night's scene as Dante's Inferno on ice! So we now had Jack's scribbled calculations and flow sheet, and sufficient lobster meat. Jack then divulged two working maxims that were to guide us in the coming months: (a) one at a time is good fishing (Jack was quickly in tune with his new marine collaborators), and (b) if you want to solve a problem, you have to hit it hard!

We began immediately and the three of us worked essentially every day for the next 3–4 months. The isolation was carried out at UWA; at the end of each stage we would take portions of each fraction back to WAMRL to perform total arsenic analyses while Jack would obtain dry mass data on all the fractions. The combined data allowed us to precisely evaluate the effectiveness of each stage of the isolation procedure. It started with solvent extraction (MeOH), then solvent partitioning (phenol/water) which generated a modest number of fractions, and then moved on to ion-exchange chromatography on huge preparative columns which generated many hundreds of samples. At each stage, our extract was becoming increasingly concentrated in arsenic; when we reached the level of about 10% As by mass (from an initial 0.002%), preparative TLC was used to further purify the arsenical. In accord with Jack's original plan, the pure natural product finally crystallised from acetone/methanol and its structure was determined as arsenobetaine [(CH₃)₃As⁺CH₂COO⁻] by single crystal X ray structure analysis.³ Jack wasn't content, however, until he had chemically synthesised arsenobetaine and obtained a full data set to confirm the structure of the natural product, and demonstrated that it wasn't an artefact of the isolation procedure. This was a valuable lesson for me and I sometimes wonder what Jack would think of the modern way, in the analytical and environmental sciences at least, of reporting novel chemical structures on preliminary and often dangerously tenuous data.

The availability of chemically synthesised arsenobetaine allowed a full toxicological assessment which showed that it was innocuous, and the long held view that seafood arsenic is not harmful was finally given firm scientific support. The results from this work were quickly translated into food safety legislation in some countries, but, surprisingly, European Union legislation regarding arsenic in foods still fails to incorporate information on arsenic speciation.

Arsenosugars—arsenic species in algae

The origin of arsenobetaine then became a relevant and interesting topic, and John and I began investigating the arsenic compounds in the brown alga Ecklonia radiata which was an important component of the food web supporting the lobsters. By now, we were true disciples of Jack and the prospect of extracting 10 kg of fresh alga did not deter us. But the algal arsenicals proved considerably more difficult to isolate than arsenobetaine, and it was almost 2 years of laborious and frustrating work before we obtained pure compound. Two compounds in fact—5 mg of one and 20 mg of another—neither of which would crystallise. With the skilled help of Lindsay Byrne at UWA we obtained good NMR data of the compounds, and with helpful advice from Don Cameron at Melbourne University we were able to propose structures based on the NMR data and some degradative chemical work.⁴ The two compounds were the novel and unusual arsenic-containing carbohydrates: they were the first of 15 or so arsenosugars to be isolated from marine organisms in the following years, many of which today serve as standards for arsenic speciation analysis.

Still pursuing the origin of arsenobetaine, we then investigated the possible role of arsenosugars. Although we could show production of likely intermediates in the biotransformation of arsenosugars into arsenobetaine, we could not demonstrate conversion at significant levels, and the intriguing question of where arsenobetaine comes from still remains unanswered today.

Mercury and cadmium: the art of the possible

During the arsenic studies, I was also working on related problems with mercury in fish and cadmium in prawns (shrimps). Although the form of mercury in fish was known to be mainly methylmercury, and presumed to be protein bound, it was not known if the toxicology of this protein bound methylmercury was different from that of free methylmercury (as MeHgCl, for example). In other words, would a fuller speciation analysis of fish mercury demonstrate that it was less toxic than people feared, based on comparison with free methylmercury, and thus provide data for a more lenient permissible concentration for mercury in fish. A fairly simple investigation involving specific methylmercuric chloride analysis by gas chromatography, however, indicated that in the acidic conditions of the human gut, the fish mercury would be readily converted to methylmercuric chloride. Hence, in toxicological terms, investigation of the protein moiety would not result in much additional information, and we abandoned this area of research. Probably, this work is worth repeating with today's improved techniques of speciation analysis to put it on a firmer scientific basis, particularly in view of renewed human health concerns regarding mercury in fish.

The work with cadmium produced some surprises and highlighted the role of biological variability in monitoring studies. An initial sampling of prawns, collected from major fishing grounds in Shark Bay Western Australia, showed that the individuals fell into two distinct groups of high and low cadmium concentrations. On the second sample collection, the sex of the individual prawns was recorded (not always an easy task), and the resultant analyses clearly demonstrated that the females contained about 100-fold as much cadmium as the males! Examination of the proteins responsible for binding the cadmium revealed an interesting metallothionein variant which may be related to the prawns' high natural cadmium exposure in Shark Bay.⁵ Although the binding of cadmium in specific proteins was of fundamental interest, the relevance in human toxicological terms was less clear because cadmium ions would likely be readily released from the metallothionein in the human gut. In the absence of scientific data supporting diminished toxicity of natural cadmium in seafoods, the problem was addressed by setting the maximum permissible concentration at a level just above that found in most edible marine organisms! Truly the art of the possible.

North to Denmark

Changed personal circumstances and a desire to work for a period abroad led to a move to Denmark in 1996 where I joined the ecotoxicology group of Poul Bjerregaard at the University of Southern Denmark in Odense. It was my first appointment in a university, and, after 20 years in a research institute, some adjustments were necessary. It was also my first extended period away from Australia. I had luck from the start, however, because my first research student, Lone Andreason, was married to an Australian (actually raised less than 1 km from my home in Perth), and spoke English with a superb Australian accent. Lone tackled her research project with intelligence and enthusiasm and I quickly realised that supervising students could be very enjoyable. Of particular interest to me was witnessing the undergraduate students' first steps into research and the accompanying transition from text-book learning to reading scientific journal articles. I was surprised to see that the students accepted everything they read in a journal article as being proven and correct, and I realised that they needed considerable guidance at the beginning to critically read and learn to question the primary scientific literature. My research in Denmark was initially quite biological, but later I collaborated with Søren Pedersen developing methods of arsenic speciation analysis using liquid chromatography coupled to electrospray mass spectrometry. The method proved extremely useful for our biological studies on the biotransformation of arsenic in algae and in humans, in addition to giving us the prized possibility of determining structures for novel compounds.⁶

Current position in Graz: research and teaching

I very much enjoyed my time in Odense but after five Danish winters of varying bleakness I took the opportunity to change, and in July 2001 I moved to Karl-Franzens University in Graz—initially as a guest Professor and, since February 2002, as Professor of Analytical Chemistry. I am very pleased with the move to Graz because my new colleagues are very active in the field of speciation analysis and have many years experience with arsenic compounds. Our techniques are based on liquid chromatographic separations with mass spectrometric detection, and our current research projects include arsenic-containing lipids in foods, human metabolism of arsenic and selenium compounds present in food and water, and element-selective analysis of metal binding proteins. Our long-term goal in the arsenic area is to provide definitive data regarding its possible essential role in organisms. We currently have an enthusiastic core of students with a real talent and appetite for research.

My teaching includes a course called Critical Evaluation of Scientific Publications which developed from my earlier observations in Denmark, and is designed for students beginning their research studies, normally at the PhD level. We cover the philosophy, planning and protocol of scientific research, the preparation of a scientific manuscript, and the machinations and ramifications of the publication process. We then examine in detail some recent research papers discussing their strengths and weaknesses, both in terms of the underlying science and the way that this science is presented and discussed. The students seem to find this a useful process, and quite quickly come to recognise deficiencies in some scientific papers, and to gain a feel for what constitutes good research and good scientific reporting.

Research philosophy and future directions

Research is the attempt to discover something new. It doesn't always succeed, but then it doesn't always have to. In fact, research should contain some failures (otherwise the bar has been set too low), some dead-ends, in addition to those few moments of success. A research institute that accepts this fact creates an environment that encourages scientists to tackle more challenging and difficult problems; chances of success may be reduced but the results are likely to have lasting impact. I try to infuse this philosophy into our research students, and encourage them to take risks and pursue a particular idea that they may have. I also stress that their chosen topic is totally theirs, and that they need to take full possession of it. At the end of their PhD studies, they should know more about their research topic than anyone else, worldwide. Logical really—if they have been pursuing new ideas and implementing new techniques for a specific problem over a three year period, no-one else could possibly know their topic better.

I have a very positive view of the future of my general research area, bioanalytical chemistry, and of the role of speciation analysis in the field of metals in biology, in particular metals and human health. Previous work with total metal concentrations has provided a wealth of valuable data linking certain diseases or conditions with metal imbalance, either deficiency or excess. Analytical techniques are now being developed to distinguish and quantify various forms (protein or other) of a metal in biological tissues. Ultimately this may enable relationships between different metal species and disease to be studied on a fine scale, and perhaps valuable correlations will be uncovered that have, up to now, been shrouded in the total metal data.

References

1. A. C. Chapman, On the presence of compounds of arsenic in marine crustaceans and shell fish, Analyst, 1926, 51, 548–563.
2. J. S. Edmonds and K. A. Francesconi, Methylated arsenic from marine fauna, Nature, 1977, 265, 436.
3. J. S. Edmonds, K. A. Francesconi, J. R. Cannon, C. L. Raston, B. W. Skelton and A. H. White, Isolation, crystal structure and synthesis of arsenobetaine, the arsenical constituent of the western rock lobster Panulirus longipes cygnus George, Tetrahedron Lett., 1977, 18, 1543–1546.
4. J. S. Edmonds and K. A. Francesconi, Arseno-sugars from brown kelp (Ecklonia radiata) as intermediates in cycling of arsenic in a marine ecosystem, Nature, 1981, 289, 602–604.
5. K. A. Francesconi, K. L. Pedersen and P. Højrup, Sex-specific accumulation of Cd-metallothionein in the abdominal muscle of the coral prawn Metapenopsis crassissima from a natural population, Mar. Environ. Res., 1998, 46, 541–544.
6. K. A. Francesconi, S. Khokiattiwong, W. Goessler, S. N. Pedersen and M. Pavkov, A new arsenobetaine from marine organisms identified by liquid chromatography-mass spectrometry, Chem. Commun., 2000, 12, 1083–1084.

---