Seeing double: Could arsenic replace phosphorus in bacteria?

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Seeing double: Could arsenic replace phosphorus in bacteria?

Phosphorus is a key pentavalent element in biological molecules, playing a crucial role in the formation, function and control of nucleic acids, proteins, and phospholipids, as well as energy molecules like adenosine triphosphate and many co-enzymes; chiefly present in the form of the phosphate molecule, PO43-, with its subsequent esters and anhydrides (Westheimer 1987) . This essay aims to validate the2011 claim by Wolfe-Simon et al. that a bacterium could sustain its growth using arsenic to replace of phosphorus.

As Westheimer (1987) explains, phosphate has several key chemical properties that allow it to perform the variety of functions it does in different biomolecules. Firstly, the ionisation constants of phosphate ensure it is always ionised under physiological conditions, preventing the loss of phosphate-biomolecules from the cell. Pentavalent phosphate also easily forms removable covalent bonds, and can be used to link two moities, while retaining ionisation. Finally, phosphate has a lower hydrolysis rate than electrically neutral carboxylic acids, with the negative charge reducing nucleophilic attack by electronegative nucleophiles.

Westheimer (1987) also explores the possibility of other molecules being capable of substituting phosphates. Tribasic arsenate, an arsenic acid, is identified as a major possibility, since it is also forms pentavalent esters, but has a higher hydrolysis rate. Wolfe-Simon, Davies & Anbarthemselves identify in 2009 these chemical similarities over the range of biological pH and redox conditions, as well as the higher hydrolysis rate of arsenate relative to phosphate, but argue that higher relative arsenate concentrations, as well as the formation of arsenic-sulfide minerals that would allow arsenate-based biomolecules to generate spontaneously in conditions where life was believed to have originated, would nullify the effect of quick arsenate hydrolysis.

In order to look for evidence of their above argument, Wolfe-Simon et al. isolated a bacteria from the hypersaline, alkaline and high in dissolved arseninic Mono Lake, California, with the bacteria in question, GFAJ-1 of the Holmonadaceae family, being capable of growing in media containing 40mM AsO43-, but only trace (3.1±0.3μM) PO43- (2011). Bacterial culturing showed that it exhibited slower growth, was larger in size and possessed vacuole-like structures under As+/P- conditions compared to growth under As-/P+ (1.5mM PO43-) conditions, and that it did not grow under As-/P- conditions. NanoSIMS (Second Ion Mass Spectrometry) data in this paper from isolated DNA show higher cellular levels of radiolabelled As to C in the As+/P- conditions compared to the As-/P+ conditions, as well as the inverse. Inducibly coupled plasma mass spectrometry (ICP-MS) data also show similar results by measuring dry weight intracellular element percentages. X-ray data of the As+/P--growing bacteria seemed in line, showing evidence for As in the pentavalent state, bound to O and what is comparable to distal C, within reasonable covalent bond lengths. Thus, the authers claim to have identified arsenate-containing biomolecules within GFAJ-1 cells growing on media containing AsO43- without PO43-.

The release of the paper was, however, met with widespread scepticism and critisism, and many groups attempted to replicate, disprove or find an alternate hypothesis.

Firstly, a paper by Fekry, Tipton & Gates (2011) used model compounds to estimated the half-life of asenate-containing nucleotides at 0.006s, compared to the estimate of phosphate-containing nucleotides at 30 000 000 years, demonstrating the vastly different kinetic properties of arsenate diesters compared to phosphate diesters.

Next, Rosen, Ajees & McDermott (2011), modelled slight differences in “arsenic-DNA” compared to “phosphate-DNA”, which could impact base-pairing, transcription and translation, as well as the 100 000-fold increase in ATP hydrolysis rate when one or more arsenate moities replace the phosphates. They also argued for the formation of arseno-lipids that reduce membrane stabilitity and cause the swelling observed under light microscopy, as well as account for X-ray data of arsenic in the lipid fraction(Rosen, Ajees & McDermott 2011). As for the vacuole-like structures, Rosen, Ajees & McDermott suggest that they are poly-β-hydroxybutyrate granules that store carbon under nutrient starvation conditions (2011).

Rosen, Ajees & McDermott also suggest caution with the NanoSIMS, ICP-MS and X-ray data, since each of the methods could be plagued by contaminating AsO43-, or the inability to normalise the data with regards to the ICP-MS data, leading to inaccurate results (2011).

In attempting to replicate the results, Reaves et al. (2012) observed a lack of arsenate detectable in GFAJ-1 DNA using High Performance Liquid Chromatography after washing steps not performed by Wolfe-Simon et al. (2011), indicating free arsenate contamination in the results, as cautioned by Rosen, Ajees & McDermott (2011).

Adding to the growing argument, Erb et al. (2012) could not replicate the growth of GFAJ-1 on media containing arsenate with <1.7μM phosphate, and used high resolution mass spectrometry to demonstrate that most metabolites, especially nucleotides, were detected in the phosphorlyated, and not the arsenylated, form. Even the arsenylated metabolites were also detected in control media, indicating spontaneous formation (Erb et al. 2012).

A highly likely explanation for the growth of GFAJ-1 in As+/P- conditions came from the results of Basturea, Harris & Deutscher (2012), who showed, using a radiolabelled [3H]-uridine assay, that arsenate actually accellerates the ribosomal degradation observed in E. coli under nutrient starvation conditions, and that E. coli grown in arsenate containing media leads to an arsenate-resistant population of bacteria. This ribosomal degradation was even alluded to in Wolfe-simon et al.'s results (2011), where 2 large bands that Basturea, Harris & Deutscher (2012), as well as Rosen, Ajees & McDermott (2011), believe to be rRNA, are present in DNA from As-/P+ conditions, but not in As+/P- conditions.

Basturea, Harris & Deutscher in 2012 argue conclusively that a number of arsenate-resistant emerge after an extensive lag period, which then grow on the phosphate released from ribosomal degradation of other bacteria.

Based on the results and arguments put forward, the misleading results observed by Wolfe-Simon et al. (2011) are largely due to poor washing steps to remove arsenic contamination, and conclusions were reached without first attempting to explain all observed results. A bacteria that can utilise arsenic instead of phosphorus remains, until proven otherwise, fiction.


  1. Basturea, GN, Harris, TK & Deutscher, MP 2012, 'Growth of a Bacterium That Apparently Uses Arsenic Instead of Phosphorus Is a Consequence of Massive Ribosome Breakdown', Journal of Biological Chemistry, vol. 287, no. 34, pp. 28816-28819. Available from: Journal of Biological Chemistry. [6 March 2014].
  2. Erb, TJ, Kiefer, P, Hattendorf, B, Günther, D & Vorholt, JA 2012, 'GFAJ-1 Is an Arsenate-Resistant, Phosphate-Dependent Organism', Science, vol. 337, no. 6093, pp. 467-470. Available from: Science Online. [9 March 2013].
  3. Fekry, MI, Tipton, PA & Gates, KS 2011, 'Kinetic Consequences of Replacing the Internucleotide Phosphorus Atoms in DNA with Arsenic', ACS Chemical Biology, vol. 6, no. 2, pp. 127-130. Available from: ACS Online. [9 March 2014].
  4. Reaves, ML, Sinha, S, Rabinowitz, JD, Kruglyak, L & Redfield, RJ 2012, 'Absence of Detectable Arsenate in DNA from Arsenate-Grown GFAJ-1 Cells', Science, vol. 337, no. 6093, pp. 470-473. Available from: Science Online. [6 March 2014].
  5. Rosen, BP, Ajees, AA & McDermott, TR 2011, 'Life and death with arsenic', BioEssays, vol. 33, no. 5, pp. 350-357. Available from: Wiley Online Library. [12 March 2014].
  6. Westheimer, F 1987, 'Why nature chose phosphates', Science, vol. 235, no. 4793, pp. 1173-1178. Available from: Science Online. [6 March 2014].
  7. Wolfe-Simon, F, Blum, JS, Kulp, TR, Gordon, GW, Hoeft, SE, Pett-Ridge, J, Stolz, JF, Webb, SM, Weber, PK, et al. 2011, 'A Bacterium That Can Grow by Using Arsenic Instead of Phosphorus', Science, vol. 332, no. 6034, pp. 1163-1166. Available from: Science Online. [6 March 2014].
  8. Wolfe-Simon, F, Davies, PCW & Anbar, AD 2009, 'Did nature also choose arsenic?', International Journal of Astrobiology, vol. 8, no. 02, pp. 69-74. Available from: Cambridge Journals. [11 March 2014].