November 2, 2010 § 2 Comments
What is electricity? It’s moving electrons. Every living thing moves electrons around, not just in nerves (for those of us that have them) but also in metabolism (oxidize one thing, reduce another). Is it possible to use this metabolic electricity to communicate with man-made devices? If you could, you might be able to make very sensitive biosensors, or even use bacteria to charge your batteries. The first question you would need to address is whether you could get the electrons generated by metabolism out to the surface of the cell where they could be captured by a metal electrode.
Several species of bacteria do this naturally. One of the best-studied of these is Shewanella oneidensis, and the reason it needs to move electrons to the surface of the cell is so that it can use metal oxides as electron acceptors when there’s no oxygen around: in effect, these bacteria “breathe” metal. Lots of applications have been suggested based on this unusual property, including uses in bioremediation. The possibility that these bacteria might be interesting for applications involving electricity was recently strengthened by a paper showing that these bacteria can produce currents of up to 204 fA/cell; what’s more, they extrude wire-like protrusions — “bacterial nanowires” — that can conduct electricity at quite respectable rates (as do other bacteria). But S. oneidensis is not exactly the best-characterized organism for engineering purposes; to make really good use of the ability to detect what’s going on, electrically, within the cell — or to maximize the production of electrons to be captured in batteries — you might very well want a species that’s easier to manipulate.
A collaborative team led by Caroline Ajo-Franklin [a Silver lab alumnus] therefore set out to move the key electron transfer chain from S. oneidensis into the canonical engineerable organism, E. coli (Jensen et al. 2010. Engineering of a synthetic electron conduit in living cells. Proc. Natl Acad. Sci. doi: 10.1073/pnas.1009645107). In S. oneidensis the pathway appears to consist of four heme-containing cytochromes, arranged in a chain from the inside of the cell to the outside: one is in the inner membrane, a second is in the periplasm, and the third and fourth are in the outer membrane. There’s an additional protein that helps with folding of the outer membrane cytochromes, and may also be involved in communication between those proteins and the periplasmic cytochrome. Since each of the cytochromes involves multiple (4-10) heme groups, the authors expected that correct folding and localization was quite likely to be a challenge. They therefore went after a minimal version of the natural S. oneidensis chain: they missed out the inner membrane cytochrome, hoping [in grant language, anticipating] that an E. coli cytochrome that has 52% similarity to the S. oneidensis one (NapC) would substitute, and they used only one of the two extracellular cytochromes. This left them with two cytochromes and the protein that helps with folding to express, characterize for redox activity, and test in various combinations to see whether they reproduced electron transport from inside to outside.
So they did all that. The first test they used for metal reduction was reduction of soluble Fe(III). Interestingly, all you need to insert into E. coli to get soluble Fe(III) reduction is the periplasmic cytochrome — you don’t need the extracellular cytochrome or the folding protein at all. This seemed to be good evidence that the E. coli inner membrane cytochrome, NapC, is indeed passing electrons to the introduced periplasmic cytochrome, as the authors hoped. Being good scientists, however, the authors went back and checked what happened when they deleted the E. coli NapC gene. In the absence of NapC, bacteria expressing the S. oneidensis periplasmic cytochrome do reduce Fe(III) more slowly than when NapC was present, but only by about 1/3. So it seems that another E. coli source as well as NapC can pass electrons to the S. oneidensis periplasmic cytochrome.
But what about passing electrons to solid metal? For this, you need both the periplasmic cytochrome and the outer membrane one. Jensen et al. test both solid Fe(2)O(3) and crystalline nanoparticles of the same metal oxide: electron transfer from the engineered E. coli to the metal ion was significant in both cases, though stronger for the nanoparticles, indicating that surface area might be limiting in the solid case.
These engineered bacteria are much better at transferring electrons to metal than are unmodified E. coli, but they’re considerably worse than S. oneidensis itself. This is not terribly surprising: the transfer of electrons from NapC to the S. oneidensis cytochromes would not be expected to be optimal, and indeed the authors present evidence that it isn’t. In part this may be because there simply isn’t enough NapC; the natural expression of this protein is not very strong. The reengineered system also lacks one of the two extracellular membrane cytochromes. So more work will undoubtedly be needed to increase the efficiency of electron transfer. But overall, this seems like a solid first step in the direction of being able to make biosensors out of engineered bacteria, and read their output in a convenient electrical form.
Jensen HM, Albers AE, Malley KR, Londer YY, Cohen BE, Helms BA, Weigele P, Groves JT, & Ajo-Franklin CM (2010). Engineering of a synthetic electron conduit in living cells. Proceedings of the National Academy of Sciences of the United States of America PMID: 20956333