Insights & Perspectives Life and death with arsenic Think again Arsenic life: An analysis of the recent report ‘‘A bacterium that can grow by using arsenic instead of phosphorus’’ Barry P. Rosen1)Ã, A. Abdul Ajees1) and Timothy R. McDermott2) Arsenic and phosphorus are group 15 elements with similar chemical proper- significant commentary, often as anon- ties. Is it possible that arsenate could replace phosphate in some of the chemi- ymous electronic communications. In this essay, we will examine the data cals that are required for life? Phosphate esters are ubiquitous in biomolecules and the conclusions from that provoca- and are essential for life, from the sugar phosphates of intermediary metabolism tive publication. We emphasize that to ATP to phospholipids to the phosphate backbone of DNA and RNA. Some there are no additional outside data that enzymes that form phosphate esters catalyze the formation of arsenate esters. substantiate or refute the findings in Arsenate esters hydrolyze very rapidly in aqueous solution, which makes it their study. Until their results are improbable that phosphorous could be completely replaced with arsenic to extended in their laboratory, or repli- cated in others, we can only analyze support life. Studies of bacterial growth at high arsenic:phosphorus ratios dem- the available evidence from the onstrate that relatively high arsenic concentrations can be tolerated, and that perspective of the published literature. arsenic can become involved in vital functions in the cell, though likely much less efficiently than phosphorus. Recently Wolfe-Simon et al. [1] reported the isolation of a microorganism that they maintain uses arsenic in place of phos- Chemical considerations phorus for growth. Here, we examine and evaluate their data and conclusions. First we will briefly review some back- Keywords: ground of the chemistry of arsenic and & arsenate; arsenic life; ester hydrolysis; phosphate phosphorus as they relate to biology. Why does life on earth use primarily six elements: hydrogen, carbon, nitro- ‘‘Our children will starve without and coworkers [1] described the iso- gen, oxygen, phosphorus, and sulfur? arsenic’’ wrote Joan Slonczewski, an lation of a microbe, GFAJ-1, from Those six are among the most abundant award-winning science fiction writer Mono Lake, CA, an environment that and the lightest of elements, and they and Yale-educated microbiologist. Her naturally contains high concentrations form strong chemical bonds with one book Brain Plague [2] depicts the of arsenic. They assert that GFAJ-1 uses another. Carbon in particular can form arsenic-contaminated world Prokaryon, arsenic in place of phosphorus to sus- long chain polymers that are the build- where sentient microbes evolved not tain growth, a phenomenon reminiscent ing blocks of life. But these properties do only to use but to require this normally of the microbes described by not preclude the substitution of these toxic element. Recently Wolfe-Simon Slonczewski. This study has generated elements with other less abundant elements in biological molecules, especially those lower in the same group of the periodic table. In general, DOI 10.1002/bies.201100012 elements in the same group have similar chemical properties, but their abun- 1) Department of Cellular Biology and *Corresponding author: dance varies inversely with their atomic Pharmacology, Florida International University Barry P. Rosen Herbert Wertheim College of Medicine, Miami, E-mail: brosen@fiu.edu number. FL, USA A rough analogy of the contrast 2) Department of Land Resources and between phosphorus and arsenic is Environmental Sciences and Thermal Biology Institute, Montana State University, Bozeman, the carbon-silicon dichotomy. Some MT, USA algal diatoms use the group 14 element 350 www.bioessays-journal.com Bioessays 33: 350–357,ß 2011 WILEY Periodicals, Inc. .....Insights & Perspectives B. P. Rosen et al. silicon instead of carbon for building O exoskeletons [3]. So why is carbon, O and not silicon, the backbone of most 1.5 Å 1.7 Å Think again of the molecules of life? Could silicon substitute more widely for carbon? One P As reasonable explanation is the difference O O O in bond lengths between carbon and O silicon. Carbon-carbon single bonds O O are about 1.52 A˚ , while the silicon- ˚ silicon bond length is about 2.34 A. Figure 1. Comparison of phosphate and arsenate anions. The bond angles and distances Shorter bonds are stronger bonds, with are approximated using averages from a number of different phosphate and arsenate bond length inversely related to bond structures. strength and bond dissociation energy. This is why the silicon-silicon bond (230 kJ/mol) is weaker than the carbon- much less toxic arsenate, has chemical structures of a single-stranded 5-mer carbon bond (356 kJ/mol), and why similarities with phosphate. based on the crystal structure of diamond, crystalline carbon with a Arsenate is able to form esters that B-DNA (PDB ID 1BNA). The resulting Mohs hardness of 10, can scratch crys- are similar to phosphate esters. However, two models of ‘‘arsenic DNAs’’ provide talline silicon, with a Mohs hardness these compounds are less stable – or to a first approximation of what DNA would of 7. To put it another way, longer bonds put it another way, more brittle – than look like if all of the phosphate was make for more brittle structures: picture the corresponding phosphate esters. This substituted with arsenate. Comparison the carbon-carbon polymer diamond, is at least in part a consequence of bond of the normal phosphate 5-mer with and the silicon dioxide polymer window lengths: the P–O bond length in phos- either arsenate structural models shows glass. This does not mean that there phates is about 1.5 A˚ [5], while the As–O small but clear differences in the orien- could not be organisms where silicon bond of arsenates is about 1.6–2 A˚ (the tation and location of the bases. It is replaces carbon, but those organisms bond lengths vary in different com- difficult to predict from this modeling would be more fragile. pounds, but arsenic generally makes lon- how the changes would affect Watson- A closer analogy is the sulfur- ger bonds than phosphorus) (Fig. 1). Crick DNA base pairing or base stacking, selenium similarity. Although the group Again, longer bonds are more easily bro- but even small differences could 16 element selenium is not universally ken, so arsenate esters hydrolyze more represent a challenge for the enzymes used in biology in place of sulfur, it readily than phosphate esters. Another of replication or transcription. In the readily substitutes for sulfur in amino consideration is bond angles. The P–O eons of time since the origin of life on acids such as selenocysteine and sele- bond angle is approximately 1178,while earth, enzymes might have adapted to nomethionine in spite of its longer bond the As–O bond angle is about 1008 (again these constraints, but would they be able length and inherent lower stability. In these can vary in different compounds, to switch back and forth between phos- fact, selenium is an essential trace but the arsenic angles are generally less phate DNA and arsenate DNA? Even if element because some enzymes use obtuse than the phosphorus angles). the overall structure of DNA or RNA with selenocysteine in their active site [4]. Although no force field parameters have arsenate substituting for phosphate were So, would it be reasonable to suppose been determined for arsenic, we compatible with biological function, the that the group 15 element arsenic could attempted to model ‘‘arsenic DNA’’. In instability of the arsenate esters that form replace phosphorus in some of the the Cambridge Structural Database, As– the backbone of the nucleic acids would chemicals required for life? In biology, O distances and O–As–O angles vary most likely present an insurmountable phosphorus is found primarily in the from about 1.7 to 2.0 A˚ and 1008 to obstacle to formation of an arsenate stable þ5 oxidation state as phosphate 1098, respectively. In the Molecular backbone in DNA or RNA. and phosphate esters. The esters of pen- Modeling Database (MMDB), which is The lability of ATP, the currency of tavalent phosphorus are ubiquitous in part of the Entrez system, there are a bioenergetics, with one, two, or three biomolecules, from the sugar phos- number of arsenate structures. For arsenates in place of phosphate is also phates of intermediary metabolism to example, two protein structures with a serious issue of concern. The short phospholipids to the phosphate back- bound arsenate ligands are PDB IDs half-life of ADP-arsenate, which spon- bone of DNA and RNA. Furthermore, (protein data bank identifiers) 3ENZ taneously hydrolyzes about 100,000- the high energy phosphate ester bond and 1TA4, which have As–O bond fold faster than ATP, is the primary of ATP is the foundation of cellular lengths of 1.7 and 1.93 A˚ , respectively, source of the uncoupling action of energy cycling. Arsenic, in contrast, and which both have average bond arsenate in oxidative phosphoryla- has two biologically relevant oxidation angles of 1098.Becausethebondangles tion [6]. For AMP-diarsenate or adeno- states, þ3 and þ5, and it cycles between and distances vary in different arsenate sine triarsenate this issue is likely to be these states. Arsenite, its þ3 oxidation salts and esters, we chose two sets of even more dire – to our knowledge these state, is quite reactive and toxic, form- initial parameters, with As–O bond dis- compounds have never been synthe- ing strong bonds of metallic character tances and O–As–O angles of 1.7 A˚ and sized, probably because they are too with thiols in proteins and small mol- 1098 (Fig. 2A) and 1.957 A˚ and 100.18 labile to exist long enough to identify.