Home » Evolution » To Stop Evolution: New Way Of Fighting Antibiotic Resistance Demonstrated By Scripps Scientists

To Stop Evolution: New Way Of Fighting Antibiotic Resistance Demonstrated By Scripps Scientists

To Stop Evolution: New Way Of Fighting Antibiotic Resistance Demonstrated By Scripps Scientists

http://www.sciencedaily.com/print.php?url=/releases/2005/05/050518175350.htm

A team of scientists at The Scripps Research Institute and the University of Wisconsin have demonstrated a new way of fighting antibiotic resistance: by stopping evolution.

In the June issue of the open-access journal PloS Biology, the team describes how a protein called LexA in the bacterium Escherichia coli promotes mutations and helps the pathogen evolve resistance to antibiotics. The scientists also show that E. coli evolution could be halted in its tracks by subjecting the bacteria to compounds that block LexA. Interfering with this protein renders the bacteria unable to evolve resistance to the common antibiotics ciprofloxacin and rifampicin.

“If you inhibit this pathway, the bacteria cannot evolve,” says Scripps Research Assistant Professor Floyd Romesberg, Ph.D., who led the study.

Since the evolution of resistance is under the control of LexA, compounds that block the protein might prolong the potency of existing antibiotics.

Evolution — Its not Like Death and Taxes Anymore

This research raises fundamental questions about evolution. Biologists have often thought about evolution in the same way many think about death and taxes — something inevitable. But Romesberg is a chemist, and he found himself asking not only how, but why evolution happens.

For the last few years, Romesberg has led an effort to understand the genes that drive evolution, an innovative way of thinking because scientists have more often understood evolution as the force that drives genes. What underlies evolution is mutation — changes in the DNA. “Mutations are the fuel for evolution,” says Romesberg.

Because of the potential harm of mutations, humans and other mammals have evolved to make as few as possible. The machinery inside our cells has the ability to replicate our genomes extremely well, and the “polymerase” enzymes that replicate our DNA rarely make mistakes. Even when they do, we have multiple, redundant repair and proofreading mechanisms that would make even the most six-sigma-compliant NASA engineer jealous.

Nevertheless, all organisms are prone to some level of damage because no replication machinery is perfect, and with large genomes of billions of bases of DNA to be copied many billions of times over the lifetime of an organism, a certain level of spontaneous mutations will occur — mistakes that escape repair and become part of the DNA of the cell in which they occur. Scientists have generally thought that slowly over time, these mutations accumulate and species diverge.

However, Romesberg and his colleagues believe that cells are not just the passive victims of random mutations, but have ways of initiating mutations in their own DNA. Evidence for this includes the fact that the rates of mutation in some cells does not seem consistent with the mutation rates associated with DNA replication.

Some cells, like bacteria subjected to antibiotics, seem to acquire mutations at a much higher rate. Romesberg reasons that these rapid mutations may sometimes be part of a mechanism organisms have to rapidly evolve when they need to.

Stress and Mutations

A few years ago, when he was first starting to think about this, Romesberg encountered a paper in a scientific journal that discussed certain genes that “make mutations,” as he put it. When these genes are deleted from cells, the cells lose their ability to mutate, even when subjected to massive amounts of ultraviolet light.

This brought Romesberg to the conclusion that mutation is a programmed stress response — a survival mechanism. If the cell senses damage, and if the damage persists beyond its ability to repair it, the cell will turn on its mutation machinery and open the floodgates for evolution.

Take the bacterium Escherichia coli for instance. When E. coli cells are subjected to damage, they upregulate repair enzymes, which then go to work trying to fix the problem. If the damage persists, the cell upregulates recombination enzymes, which are tasked with recombining the DNA — another way to repair it. And, says Romesberg, if the damage still persists, the cells upregulate enzymes whose sole task is to make mutations.

Presumably, inducing mutations is an effective evolutionary strategy for dealing with environmental changes that maximize the chances that a progeny cell will be better adapted. In order to evolve, organisms have to mutate, so they turn on the mutation process when they are threatened with extinction.

Romesberg reasoned that since mutations can be turned on full-force, perhaps they could be shut off as well. Doing so, he says, would put a halt to evolution — an interesting prospect because the mutations responsible for evolution are the underlying causes of cancer and aging as well.

“Evolution is not an unstoppable force,” says Romesberg. “There is a biochemistry underlying it and it is subject to intervention.”

The Scourge of Antibiotic Resistance

One thing that scientists would like to change about bacteria and cancer cells is their changeability.

At the dawn of the 20th century, bacterial infections accounted for several of the leading causes of death in the United States. But then came the antibiotic revolution. Antibiotic “wonder drugs” toppled tuberculosis (TB) and typhoid fever, controlled cholera and gonorrhea, reduced staphylococcal dysentery, and lowered the incidence of many other pandemic bacterial infections. These antibiotics are basically natural chemicals (or derivatives of natural chemicals) produced by other bacteria or fungi in the environment to kill off the competition. Scientists in the last century have discovered a number of these natural “antibiotic” products and have used them as the basis for treating bacterial infections.

By the middle of the century, the threat posed by many types of bacteria seemed to be waning. Bacterial infections that once topped the list as leading causes of death in the United States were no longer among the top ten. The average life expectancy in the United States soared from 47.3 years in 1900 to almost 80 years today, and antibiotics are partly to thank for this.

But in the last few decades, mutant strains of several types of bacteria with the ability to resist antibiotics have emerged, including those that cause TB, pneumonia, cholera, typhoid, salmonella, and staphylococcal dysentery. These diseases are coming back resistant to the antibiotics that have been used to treat them, and people infected with these resistant strains have to be treated with alternative antibiotics. For years scientists have been combating drug resistance by coming up with new types of antibiotics, and for years bacteria have been evolving ways around these antibiotics.

Antibiotic resistance is a major health problem in the United States. According to the U.S. Food and Drug Administration (FDA), about 70 percent of bacteria that cause infections in hospitals are resistant to at least one antibiotic, and some are so resistant that no antibiotics are effective, and they must be treated with experimental and potentially toxic drugs.

Now some super bugs — multiple drug-resistant bacteria — are emerging as an even greater threat. Multiple drug-resistant TB is no longer susceptible to broad categories of antibiotics, such as rifampicin, isoniazid, and streptomycin. Some strains of the common hospital infection-causing bacteria Staphylococcus aureus are resistant to all antibiotics except vancomycin, which is a drug of last resort, and some strains of Streptococcus pneumoniae are even resistant to vancomycin. Certain strains of Shigella dysenteriae, the cause of epidemic dysentery, have even become resistant to all but a single drug — the quinolone ciprofloxacin — and may soon become completely untreatable. This is a major concern for public health because, according to the World Health Organization (WHO), large-scale epidemics of dysentery driven by this pathogen have been known to cause tens of thousands of deaths in Central America, South Asia, and central and southern Africa.

Treating multiple drug-resistant bacterial infections can be a hundred times more expensive than treating normal infections, and the WHO estimates the total cost of treating all hospital-borne antibiotic resistant bacterial infections is around $10 billion a year. Worse, with modern rapid transit and world travel, multiple drug-resistant bacteria could potentially spread beyond the isolated confines of a hospital and into the general population.

If one could design drugs to halt the enzymes that make mutations in bacteria, this could be a way of combating the evolution of antibiotic resistance.

The Dramatic Effect of LexA Inhibition

Halting evolution is exactly what Romesberg and his colleagues demonstrated in their latest PloS paper. They showed that when E. coli cells were treated with the antibiotics ciprofloxacin and rifampicin, they turned on their mutation pathways and rapidly evolved resistance to the antibiotics. Wanting to find the proteins that turn on the mutations, Romesberg and his colleagues identified a master regulatory switch that, if inhibited, blocks the ability of the cells to mutate.

This was the protein LexA, which belongs to the class of signaling enzymes known as serine proteases and works by cutting the amino acid chain of other proteins. Romesberg and his colleagues showed that LexA’s action was necessary for the evolution of resistance to the antibiotics in vitro.

The effect in vivo was dramatic. Blocking LexA in rodent models of E. coli infections halted the growth of antibiotic resistance. Three days after being subjected to antibiotics, the rodents showed no level of resistance to rifampicin or ciprofloxin, when they harbored a variant of LexA that was catalytically inactive. In control experiments, the rodents were given rifampicin or ciprofloxin, and the scientists found that nearly all the wild-type E. coli cells had grown fully resistant to the antibiotics after three days.

This research is also significant for cancer because cancer cells mutate readily and often acquire resistance to common chemotherapies. Romesberg hopes that if he and his colleagues can identify the molecules that human cancer cells use to drive evolution, they can find ways of intervening and preventing the evolution of chemotherapy resistance in these cancer cells. Last year Romesberg was awarded a Department of Defense Breast Cancer Research Program Idea Award of $563,100 to identify the genes in Saccharomyces cerevisiae yeast and human cells that induce mutations that can transform normal cells into cancer cells.

###

The article, “Inhibition of Mutation and Combating the Evolution of Antibiotic Resistance” by Ryan T. Cirz, Jodie K. Chin, David R. Andes, Valöée de Crï’-Lagard, William A. Craig, and Floyd E. Romesberg appears in the June, 2005 issue of the journal PloS Biology. See: http://dx.doi.org/10.1371/journal.pbio.0030176

This work was supported by the Office of Naval Research.

  • Delicious
  • Facebook
  • Reddit
  • StumbleUpon
  • Twitter
  • RSS Feed

10 Responses to To Stop Evolution: New Way Of Fighting Antibiotic Resistance Demonstrated By Scripps Scientists

  1. Hmm. This seems to suggest that mutations aren’t as random as we thought. Could mutations be the result of a mechanism? And if so, how could that mechanism have evolved in the first place?

  2. Not all microevolutionary processes are “undirected”. There is evidence that some processes are the result of mechanisms instead of pure “chance-mutations”, natural selection and some other processes. And I think that this could become a future line of evidence for ID. I have outlined this concept together with a friend under the name “Potentielle Komplexität als ID-Forschungsprogramm. Ursprünge der Variabilität” (“Potential Complexity as ID-research program: Origins of variability”). This paper is not online but the basic idea is (I hope my english is good enough to describe it): Organisms are equipped with more as it is necessary to their immediate survival, for example alternative phenotypes activated in response to special conditions(look for example at the phenotypic realization of C3/C4-photosynth. Systems with Eleocharis vivipara). If we have “preprogrammed” Mechanisms of variation we have evidence for ID. As we know evolution is a blind process. As we know a intelligence is capable to look forward. Organisms who have some features in respect not to present but to potential challenges in the future show a form of telelogy. There are some basic objections (selection conditions in the past…) to that concept but I hope they are refutable with a close look at the evidence.

  3. Markus, are you familiar with the work of Richard Goldschmidt? I’m reading his “Material Basis of Evolution”. He talks there of “directed mechanisms.” Any good resources/references out there on this sort of stuff?

  4. http://biology.plosjournals.or.....io.0030176

    For those interested in reading and interpreting the primary literature for themselves, here’s the abstract to the actual article in question:

    The emergence of drug-resistant bacteria poses a serious threat to human health. In the case of several antibiotics, including those of the quinolone and rifamycin classes, bacteria rapidly acquire resistance through mutation of chromosomal genes during therapy. In this work, we show that preventing induction of the SOS response by interfering with the activity of the protease LexA renders pathogenic Escherichia coli unable to evolve resistance in vivo to ciprofloxacin or rifampicin, important quinolone and rifamycin antibiotics. We show in vitro that LexA cleavage is induced during RecBC-mediated repair of ciprofloxacin-mediated DNA damage and that this results in the derepression of the SOS-regulated polymerases Pol II, Pol IV and Pol V, which collaborate to induce resistance-conferring mutations. Our findings indicate that the inhibition of mutation could serve as a novel therapeutic strategy to combat the evolution of antibiotic resistance.

  5. Hi PaV! Not really familiar(Goldschmidt, Schindewolf, hopeful monsters – i know not much more). I know one german author – Ferdinand Schmidt and his concept of “cybernetic evolution” – who has proposed a really directed evolution: Organisms should be capable with help of ominous “Feedback-Loops” (?>”Regelkreise”) to change her genes and blueprints purposefully. It was Schmidt’s answer to the classical evolution-problems concerning the orgin of the eye and other systems of complex (syn)organization. And a real solution of this “problems” needs purpose, teleology (here was Schmidt right…) – but in Schmidts case was this a sort of evolution-directing-”computer” in organisms originated with the origin of life – through undirected mechanisms, eventually in a darwinian manner as he says;-).
    Okay, I’m not a advocate of “directed evolution” but I think this concept (in the sense of preprogrammed variation-potential and mechanisms) could apply to some of the observable evolutionary phenomena. Look for example at the Paper *Reznick, David N. et al. 2002: “Independent Origins and Rapid Evolution of the Placenta in the Fish Genus Poeciliopsis” Science 298: 1018-1020* and the summary under http://www.sciencedaily.com/re.....071202.htm Here is the independent origin of a complex system within a genus (!) described. Should we believe that this is the outcome of undirected mechanisms? Or should we look for a sort of “program” or control-mechanism?

  6. Markus:

    Not having time at the moment to look at the article you cite, let me quickly tell you that Goldschmidt’s view was that macroevolution came about by a shifting in the patterns of the chromosomal material. If this reshifting was on a large scale, then it would be (1) much more difficult for the organism to be viable, but (2)if it did become viable, then we would see organismal change at least at the genus level, if not at the Class level (or higher).

    Goldschmidt also implied through this chromosomal view that the “genotype” of most organisms is contained in, using his terms (from the 1940′s), in the germ plasm of living cells. He doesn’t elaborate it much in his famous book of 1940, but this is the impression he gives.

    I must say that it has crossed my mind–thinking now in terms of ID–that something along this line may be the actual case of evolution, since one is forced to answer the question as to “where” this added “information” comes from. It simplifies things if, indeed, the “germ plasm” of all living cells contain the information for every type of combination possible, i.e., every life form present. But that does represent an extravagant claim, and so I tend to dismiss it. However, Goldschmidt, being a preeminent geneticist and embryologist, makes me want to reconsider this possibility. If indeed Goldschmidt is correct about germ plasm, then we shouldn’t be surprised by the sudden origin of a placenta in a fish line–the information is there already in the chromosomes, and if they can rearrange themselves into a new–and VIABLE–configuration, then, voilà, a fish with a placenta. (Not having read the article, but along the lines of Goldschmidt, I rather suspect that the “placenta” was induced by some kind of chemical method. [That is, chemical induction of some kind of chromosomal change] I’ll have to go look and see.)

    PaV

  7. Markus, here are two quotes I picked out of the paper:
    “Reznick and his colleagues studied guppy-like fish in the genus Poeciliopsis. They report that placentas have evolved independently three times in closely related Poeciliopsis species. Other species in the genus lack placentas, and some have partial maternal provisioning via tissues that may be precursors of placentas.”
    AND:
    “In the Science paper, the researchers show that: 1) Fish in the genus Poeciliopsis have placentas in various stages of evolution, and 2) There are clusters of closely related species that either have highly evolved placentas, placentas in intermediate stages of evolution, or no placentas at all. These provide ideal material for studying how such complexity evolves.”

    Based on these two quotes, I would say the evidence fits almost perfectly with what Goldschmidt proposes. His interpretation would likely be something along the lines of this: Over the life history of this particular form a ‘systemic mutation’ has occurred, that is, a ‘repatterning’ of the chromosomal material. The net affect of this change is to disturb the rates of reaction within the developmental pattern of poecilioppus. When the inducer for the placenta rises above its threshold value while the embryo is still developing, then the full placenta appears. If it reaches its threshold value later on, then the placenta will be partially formed, and, if very late, then it will not appear at all.” In other words, he would see this as being hormonally based (or through other potent chemicals) and directly related to chemicals manufactured by germ plasm material (Yes, even back in 1940, Goldschmidt already realized the chemical producing power of germ plasm (DNA))

  8. Thank you for your thoughts! I’m reading at the moment Wallace Arthur’s “Biased Embryos and Evolution” and as far as I can see is the “EvoDevo”-approach kindly in the direction of Goldschmidt. It seems to me as the right time to read Goldschmidt in original.

  9. Give hard sci-fi novel “Darwin’s Radio” by Greg Bear a read for an intriguing and entertaining view of saltation driven by computational capability in the DNA molecule. I really liked the first book (Radio). The second book in the series, “Darwin’s Children”, was a little slow for me and had far less hard science driving the plot.

    Here’s a review of it by biologist Michael Goldman that appeared in NATURE

    http://www.gregbear.com/A55885.....ges/300040

  10. [...] et al. E. Coli Paper As I noted before, you can read Lenski’s et al. paper on evolution (PDF) in an experimental population of bacteria here. This paper has been touted as the first [...]

Leave a Reply