Hacker News Reader

Comment by DanBC | Hacker News Reader

> I am astonished by the letter from Bernadette Dunham, the director of the Food and Drug Administration’s Center for Veterinary Medicine, especially since she told a House committee on April 9 that she was unsure how antibiotic resistance develops.

I've posted this before, but some people may have missed it.

"Defeating the superbugs" (http://www.bbc.co.uk/programmes/b01ms5c6) has a segment showing bacteria developing resistance to antibiotics.

(http://v6.tinypic.com/player.swf?file=24goih4&s=6) (Sorry about the lousy host; YouTube's content sniffing detects this as BBC property and blocks it.)

They have a slab of jelly. The jelly has sections of differing strength of antibiotic. There's a section with no antibiotic, then 10x, then 100x then 1000x. (They cannot dissolve any more antibiotic into the jelly at that point.)

A time lapse camera shows the bacteria growing, and developing resistance to each section.

It's an excellent bit of video.

View on HN

Excellent video. What would be even more powerful is if they repeated the experiment, but with the 1000x agar right next to the antibiotic-free medium to illustrate what happens when antibiotics are used properly.

I would like to see that as well, but I would want them to let it sit for a long time - months, maybe. Because eventually, you may still get a mutation that can withstand the jump from 0 to 1000x.

Quite possible, especially because there's a safe haven of antibiotic-free medium just across the border, and the likelihood of a beneficial mutation is correlated with the number of individuals present.

In a properly-applied therapeutic situation, there is no such safe haven, and one hopes that the subject's immune system would be able to exterminate the much smaller load once the antibiotic's done its job, instead of leaving any stragglers behind to develop resistance.

Without iterative improvement, the chance of a complex adaption occurring by random chance are much much smaller. I don't know how complex an adaption to become resistant to this substance is, so it's possible it could occur by random chance. But even if so, it would likely still take much longer.

I can't speak about what the adaptation is, so I don't know if the adaptations at the difference levels of concentration are different in kind or degree. If degree, it is more feasible. Certainly the environment is only different in degree. Keep in mind the adaptation to go from 0 to 10x did not happen iteratively - and I may be remembering wrong, but I think that the bacteria stopped at the 10x boundary longer than the others. That may indicate that the subsequent mutations were of degree, not kind. Anyway, I think it would be worth doing this experiment.

I wonder if this can be used to develop useful traits in bacteria.

For example, develop bacteria that can digest some pollutants.

Depends. In many cases of antibiotic resistance, the actual "resistance" is a matter of degrees. For example, Penicillin interrupts the cell walls of bacteria, so in Gram negative bacteria it is only active in the periplasm. Some forms of Penicillin resistance work by metabolizing the Penicillin before it can do its thing. Other forms work by simply pumping the Penicillin out of the periplasm. The former mechanism requires a specific enzymatic activity. The later usually just requires tuning the pre-existing pumps that maintain the environment of the periplasm to be more specific to Penicillin or pump faster over all. In fact, most "multidrug resistance" genes are simply pumps that can exclude drugs from the insides of bacteria in general.

But really, this is just the shortest of short answers. If you want to know more about how these things can evolve, you should read about Richard Lenski's long-term experiments: http://en.wikipedia.org/wiki/E._coli_long-term_evolution_exp...

Something similar to this is done with bacteria and other life forms for all kinds of reasons. It's called directed evolution, where you take a strain of bacteria with a certain property (say aerobic nitrogen fixation at 65 Centigrade) and try to force that property to evolve (say force the aforementioned aerobic fixation to work at STP).

It's a very understood method but it's really difficult to develop proper methodology since almost every experiment will be different. The success percentage is also relatively low and is like playing the lottery (literally, you never know if the evolution is even physically possible or it might be so statistically rare that you'd need millions of parallel cultures).

It would depend on what sort of pollutant you are talking about. Many antibiotics act in subtle and complex ways because they need to be able to damage bacteria without damaging the human. Effective, evolving resistance is just a small "fix" for the bacteria.

However, some kinds of things like strong detergents, high temperatures, and so on qould require too much change to protect against so the bacteria can't evolve resistance to it. (sure, there are bacteria out there that might better resist these extreme conditions but you won't easily be able to evolve any bacteria you want to do the same)

Sure it could probably evolve resistance, it would just take much more time, and in an environment that constantly maintains a reasonable amount of selection pressure, increasing it as they evolve. Humans could assist the process by "island hopping". Like trying to select for ones that produce a certain chemical, then another chemical that requires that, step by step. As opposed to hoping they get a mutation that lets them produce a complex chemical by random chance.

It would still take a while, but bacteria have very short generations and can support huge populations, so it's far, far easier for them to evolve than anything else.

What you are suggesting seems awefully hard to do though. Its pobably easier to just get a ready-made gene from some other bacteria and insert it into the bacteria you care about.

That only works if such a gene exists already though.

This is definitely being studied. Here's one such grant by the EPA:

http://cfpub.epa.gov/ncer_abstracts/index.cfm/fuseaction/dis...

That's the principle behind selective breeding.

It and genetic engineering are pretty much exactly how such traits are developed.

That was easily one of the best bits of video I've seen in a long time. That explains a lot of science really well. Thank you!

Reply on news.ycombinator.com