What does "In most cases" mean to you?
>
> "Despite the general trend toward increased susceptibility, we saw diverse outcomes with different antibiotics. For streptomycin, which was ***the only drug to which the ancestral strain was highly resistant***, none of the evolved lines showed any increased susceptibility."
>
> Many antibiotic resistance mutations do impose a fitness cost, because they are mutations in the essential genes targetted by the antibiotic. But over time compensatory mutations can emerge which reduce that fitness cost and allow the resistance gene to persist even in the absence of drug pressure.
That's not what they found in the Lenski study:
"In summary, we examined changes in antibiotic-resistance profiles for
12 lineages of E. coli that evolved for 50,000 generations in antibiotic-free
medium. We observed a clear trend toward increased susceptibility, even though
the ancestor was already highly susceptible to all but one of the 15
antibiotics in this study. Most of the increases in susceptibility arose during
the first 2,000 generations, a period when adaptation to the experimental
environment was most rapid, and lineages that evolved hypermutability showed no
greater tendency to become more susceptible. These patterns suggest that the
increases in susceptibility were side effects of beneficial mutations in the
experimental environment, rather than resulting from the decay of unused genes
and functions. Against these overall trends, however, there were some exceptions, including small increases in resistance of some clones to a few
antibiotics and the retention of high-level resistance to streptomycin."
Most of the resistance alleles impose an energy burden on those variants giving them a fitness disadvantage in the low glucose (energy) environment. Only 1 of the 15 antibiotics tested showed "the retention of high-level resistance" and that was the case of streptomycin. And that is no guarantee that this allele will not cause fitness disadvantage in a different environment such as thermal stress or in the presence of other antibiotics. If these alleles had fitness advantage or were at least neutral in all environments, they would exist at all times in the wild-type, but they don't. If they did, antibiotics would never work but in the vast majority of cases they do work.
They only ran their experiment for 300 generations. And if rifampicin resistance is retained, why do I still see Pseudomonas aeruginosa infections in my medical practice that are still drug-sensitive.
>
> But more important than the examples of compensatory mutations in individual cases is the evidence for maintenance and spread of resistance genes in environmental bacteria, where antibiotic treatment is not a factor and the bacteria are exposed, at most, only to low levels of antibiotic contaminating the environment from wastewater. You can find lots of papers about this
>
> Continental-scale pollution of estuaries with antibiotic resistance genes
>
>
https://www.researchgate.net/profile/Michael_Gillings/publication/313124948_Continental-scale_pollution_of_estuaries_with_antibiotic_resistance_genes/links/58bdcfb8a6fdcc2d14eb4f11/Continental-scale-pollution-of-estuaries-with-antibiotic-resistance-genes.pdf
>
> "The widespread dissemination of ARGs (antibiotic resistance genes) into aquatic environmentsis alarming and has implications for controlling the spread of resistance. The movement of the MCR1 polymixin resistance gene onto a conjugative plasmid41allows it to move from cell to cell for the first time. The release of such plasmids into environmental compartments rapidly increases the diversity of species that can acquire such an element and increases the risk of ARG transfer, especially if those waters are also polluted with relevant antibiotics. In turn, this vastly increases the potential for rapid global dissemination and the emergence of resistant opportunistic pathogens."
You clearly don't recognize the problem with these studies. I attended a conference last year where they presented a similar example. Two problems with these studies, first you need to demonstrate that the resistant bacteria didn't enter these waters as fecal bacteria from patients treated with antibiotics and second, you need to demonstrate that resistance can evolve in the water polluted with these antibiotics. The presenter at that conference demonstrated that resistance can evolve to the drug in growth media but not in the river water. River water presents a totally different environment to the bacteria than does a test tube of growth media in an incubator. I'll let you try to imagine the differences in the environments under these two conditions and explain to us how resistance can evolve in river water. Maybe you should do the experiment and demonstrate that it happens because that's what I suggested to the presenter at the conference. The presenter didn't look too happy with that suggestion.
>
> Here's a review...
>
> Dissemination of Antimicrobial Resistance in Microbial Ecosystems through Horizontal Gene Transfer
>
>
https://www.frontiersin.org/articles/10.3389/fmicb.2016.00173/full
>
> There's lots in this review, but, for example...
>
> "The mobilization or transfer of ARGs by bacteriophages has been documented for various bacterial species: the transduction of erythromycin (Hyder and Streitfeld, 1978), tetracycline or multiple resistances between strains of S. pyogenes (Ubukata et al., 1975); the transfer of tetracycline and gentamicin resistance between enterococci (Mazaheri Nezhad Fard et al., 2011); the carriage of β-lactamase genes by bacteriophages in Escherichia coli (Billard-Pomares et al., 2014) and Salmonella (Schmieger and Schicklmaier, 1999); or the transfer of antibiotic resistance plasmids in MRSA (Varga et al., 2012)."
You still miss the point. Until the resistance allele evolves, there is nothing to laterally transfer. And if you use antibiotics correctly, you will drive the infective agent to extinction. If you use the antibiotics incorrectly, too low dosage, incomplete treatment, single drug therapy where resistant variants already exist in the population,..., you will introduce resistant variants into the environment and that environment starts with the patient. If you use the antibiotics correctly, there will be no resistance alleles to transfer laterally. If you think that the antibiotics eliminated from the patient and ending up in sewage water and rivers, demonstrate that resistant variants can evolve under the conditions provided by that environment.
>
> You are correct that antibiotic resistance genes often impose a fitness cost. But you are incorrect where you ignore the compensatory mutations that reduce those fitness costs.
So, are you claiming the drug sensitive Pseudomonas doesn't exist anymore? And where were the compensatory mutations in the Lenski study?
>
> You are correct when you note that, in theory, if resistance genes have fitness costs, that avoiding use of the drug, thereby removing the selection pressure, would lead to loss of the resistance gene and the need for it to evolve de novo once the selection pressure was applied again. But you are wrong to ignore the evidence that, in practice, this strategy usually does not work, because of the compensatory mutations that allow resistance genes to persist in the absence of drug pressure.
I think you mean loss of the resistance allele, not loss of the gene. From the Lenski paper:
"These patterns suggest that the increases in susceptibility were side effects of beneficial mutations in the experimental environment, rather than resulting
from the decay of unused genes and functions."
And the Lenski study is done with only a single selection condition, energy limitation. If your claim was correct, there would be no drug-sensitive bacteria at all. But that's not the case. Drug-resistance is a much bigger problem in the hospital environment where the patients tend to be more debilitated and immune compromised. I work in the community environment where patients tend to be healthier and generally have good functioning immune systems. Do you suggest that I deny them antibiotics when they come in with respiratory tract infections? If so, that would explain why pneumonia and sepsis are the most common medical reason for admission to the hospital. Too many medical providers are taking your bad advice.
>
> And you are just incorrect when you worry about "how the bacteria know which genes to transfer" in lateral gene transfer. They don't know anything. Selection just favors those cases in which they transfer the right genes (and of course many of those genes are found together on plasmids of mobile genetic elements anyway, and are therefore more likely to be transfered than chromosomal genes.
The "right genes" have to exist first. And your simple minded understanding of this problem and selection needs to be tested. What if you are treating someone with two drugs. What is the probability that the correct two alleles will be laterally transferred to some member of the population. Let "N" be the total population size, nA be the number of members of the population with resistance allele A to one drug and nB be the number of members of the population with resistant allele B to the second drug.
>
> But don't worry, it's definitely true that to find the joint probability of independent events you multiply their individual probabilities. And yes, you ca still multiply probabilities if the events are not independent, as long as you use the correct conditional probabilities. Nobody can take that away from you.
That's good that you understand this principle, now compute the joint probability of resistance allele A and resistance allele B be laterally transferred to some member of a population size "N". Don't forget to multiply these probabilities. We'll get you to understand the correct mathematics of evolution if we have to drag you kicking and screaming.