Genetic variations that alter gene activity or protein function can introduce different traits in an organism. If a trait is advantageous and helps the individual survive and reproduce, the genetic variation is more likely to be passed to the next generation a process known as natural selection. Over time, as generations of individuals with the trait continue to reproduce, the advantageous trait becomes increasingly common in a population, making the population different than an ancestral one.
Sometimes the population becomes so different that it is considered a new species. Not all variants influence evolution. Only hereditary variants , which occur in egg or sperm cells, can be passed to future generations and potentially contribute to evolution.
Also, many genetic changes have no impact on the function of a gene or protein and are not helpful or harmful. In addition, the environment in which a population of organisms lives is integral to the selection of traits. Some differences introduced by variants may help an organism survive in one setting but not in another—for example, resistance to a certain bacteria is only advantageous if that bacteria is found in a particular location and harms those who live there.
So why do some harmful traits, like genetic diseases, persist in populations instead of being removed by natural selection? There are several possible explanations, but in many cases, the answer is not clear.
For some conditions, such as the neurological condition Huntington disease , signs and symptoms occur later in life, typically after a person has children, so the gene variant can be passed on despite being harmful. For other harmful traits, a phenomenon called reduced penetrance , in which some individuals with a disease-associated variant do not show signs and symptoms of the condition, can also allow harmful genetic variations to be passed to future generations.
See S12 Fig for data on additional timepoints, and nitrofurantoin and pH conditions. To this end, we used fluctuation assays for mutations that cause rifampicin resistance, and estimated the genomic mutation rate U using Drake's approach [ 74 ]. Having estimated the mutation rates for the ancestor and evolved populations, we also wanted to examine whether prominent theoretical models that predict declines in mean population fitness at high mutation rates apply to our populations S2 Text.
While some of the models we studied e. Each evolved strain's mean change in mutation rate is shown as the percentage of its ancestor's mutation rate. Because mutation rates changed between the beginning and the end of the experiment, we wondered whether the final mutation rates were correlated with our measured phenotypes. We found significant correlations between a replicate's mutation rate and its effective population size, standing genetic diversity, and number of high frequency derived alleles, but no correlations between a replicate's mutation rate and its final relative fitness, or normalized cell density after 24 hours of growth in acidic medium or medium containing nitrofurantoin Spearman's rank correlation, S13 Fig.
Interpretation of these results requires caution for two reasons. First, for any one population, we do not know exactly when during the generations of evolution the mutation rate changed from its ancestral value. Second, we compared the mutation rate of a single randomly-selected clone from populations which can have considerable genetic diversity, and thus potentially also show diversity in mutation rates.
Despite these caveats, we found that the correlations between a representative clone's mutation rate and our other metrics are consistent with our previous analyses and figures Fig 4 , S2 Fig , S12 Fig , which simply considered the effects of ancestral mutation rate strain identity.
We call the set of genes potentially involved in modulating the mutation rate the "mutation rate genome". We wondered whether this part of the whole genome was a preferential target of mutation or selection in our experiments.
To find out, we first identified a set of 96 genes potentially involved in modulating the mutation rate S2 Table from the literature and EcoCyc [ 49 , 75 — 77 ]. If mutations or selection did not preferentially affect the mutation rate genome, the amount of genetic change we observe in it would be proportional to its length relative to the rest of the genome.
This is indeed the case: We counted the number of synonymous mutations occurring at any frequency in any replicate population at generation , and observed no statistically significant increase in the incidence of such genetic change in the mutation rate genome for any of our evolving strains S14A Fig.
We also found no difference in mean diversity between synonymous sites in the mutation rate genome relative to the rest of the genome S14B Fig. Although the mutation rate genome is not a preferential target of genetic change, its genes still accumulated many non-synonymous and nonsense changes, which are the kinds of changes that are especially likely to affect protein function S15 Fig.
Mutations in ten genes met these criteria rpoS , umuC , dinB , dinG , dps , glyS , glyW , mutL , phr , and vsr , and two were found in multiple replicate populations rpoS : 7 of 8; umuC : 2 of 8. Populations with rpoS mutations can hold a fitness advantage in nutrient-limiting environments [ 79 ], but at a cost to fitness in a variety of stressful environments [ 28 , 80 ]. We found the same rpoS ND mutation in 2. This mutation reached Thus, the mutation was likely distributed to the eight replicate populations from the ancestor, and either increased in frequency due to its direct fitness effects, or because it was hitchhiking with a beneficial mutation.
Each of the remaining genes with high frequency mutant alleles in a single replicate population were involved in DNA repair and replication dinB , dinG , glyS , glyW , mutL , phr , vsr or protection of DNA in stationary phase dps in a single replicate population and could have also affected the evolved mutation rate.
Here, we studied the effects of mutational pressure on evolutionary adaptation and the evolution of the mutation rate itself. To this end, we engineered four isogenic E. At the opposite extreme was our strain with the highest ancestral mutation rate MR XL. We originally expected this strain to have a mutation rate approximately fold higher than wildtype [ 35 ], consistent with the large effects that mutations in the dnaQ and mutL genes have on the mutation rate [ 75 ].
The discrepancy could in principle be due to the acquisition of an anti-mutator allele during the transfer of the strain between laboratory locations.
Alternatively, our mutation rate could be an underestimate for technical reasons discussed in the Methods. The mutation rate for our MR XL strain was also somewhat lower than that of a hypermutable clone which spontaneously evolved from an E.
The mutation rate of our hypermutable MR XL strain is low enough that we expected its populations to be viable [ 21 ]. We first characterized the general patterns of adaptation in our four strains, and found that their fitness increased significantly by generation for all replicate populations.
Previous experimental evolution studies in constant environments have observed fitness gains that are initially large but decrease over time [ 17 , 18 , 83 , 84 ], which is consistent with diminishing returns epistasis, in which the size of the fitness gain in an evolving population depends on its current fitness, such that populations with lower fitness can improve their fitness to a greater extent [ 85 , 86 ].
However, our fitness trajectories differ from those predicted by diminishing returns epistasis in two ways. First, they do not show a decreasing fitness gain over time [ 18 ].
Second, the mean fitness of replicate populations with small or modestly high mutation rates MR S , MR M did not immediately improve, but unexpectedly remained largely unchanged for the first generations compared to [ 87 ].
While delayed adaptive response is consistent with a lower overall beneficial mutation supply rate, it may not be sufficient to explain our observations. We expected to wait just 44 generations for a new beneficial mutation to establish in our slowest-evolving replicate population S3 Text. An instance of such contingent evolution has been documented in E. We next characterized the effect of mutational pressure on adaptation. We found that strains with higher ancestral mutation rates increased in fitness more than those with lower mutation rates, except for MR XL populations, which we will discuss below.
These observations are in agreement with theory [ 15 , 91 ] and previous experimental studies which found that large asexual populations of E. We do not actually observe the loss of fitness on average across the MR XL replicate populations, but rather a prolonged period in which fitness remains unchanged as a whole. Interestingly, however, the fitness of several MR XL replicate populations decreases from its maximum and arrives at a value that is approximately equal to that of the ancestral population.
This is reminiscent of models of extreme mutational pressure developed over the past forty years that predict reduced adaptation and eventual extinction [ 19 , 20 , 22 , 92 — 94 ].
However, these models predict a loss of fitness only at higher mutation rates than we observed, and require unrealistic assumptions S2 Text , together emphasizing the importance of additional theoretical work.
Another possibility is Hill-Robertson interference [ 7 ], which can reduce the rate of adaptive evolution by background selection—negative selection against deleterious alleles that removes the most deleterious lineages from a population—and can reduce genetic diversity [ 8 , 12 ].
Empirical evidence supports the action of this mechanism in natural populations of several eukaryotic species reviewed in [ 13 , 14 ]. However, because background selection removes deleterious mutations from a population, it cannot alone reduce the fitness of a population and it can therefore not explain the loss of fitness we observed in the three MR XL replicates.
While a lowering of the mutation rate has been previously observed [ 46 — 48 , 50 ] and predicted to be favored in some conditions [ 38 , 40 , 42 , 43 , 95 ], its extent and consistency across multiple of our evolving populations is remarkable. The mutation rate decrease probably did not occur very early during evolution, because the MR XL populations show greater genetic diversity than all other populations throughout the experiment Fig 3.
The decreasing mutation rate, together with the observation that the MR XL populations failed to adapt after more than generations, suggests that the maladaptive effects of hypermutation begin at even lower mutation rates than those in our initial MR XL strain.
While we cannot predict whether our hypermutable populations would eventually go extinct, the observation that their mutation rate can decrease adaptively makes this less likely.
Indeed, recent mutation accumulation experiments with small bacterial populations suggested that populations with higher mutation rates tend to go extinct more often and have reduced fitness than populations with lower mutation rates [ 47 ]. However, we cannot definitively identify the proximal mechanisms driving the drop in mutation rates using bioinformatics alone.
Future experimental studies to evaluate the effect of each "mutate rate genome" mutant allele on the mutation rate and fitness would be necessary. We emphasize that all our experiments use asexual populations, and that the evolutionary dynamics of mutation rates and adaptation may be different in sexual, recombining populations.
For example, in our non-recombining populations, any mutator allele remains completely linked to the mostly deleterious mutations it helps bring forth, resulting in indirect negative selection on the mutator allele. However, such an allele and its associated mutations can become unlinked in recombining populations, which reduces the strength of indirect selection on the mutator allele see [ 33 , 39 ] for reviews.
Additionally, beneficial and deleterious alleles can become unlinked in recombining populations, which can lead to increased levels of adaptation and diversity see [ 13 , 14 ] for reviews. We also characterized the effect of mutational pressure on the ability of an evolving population to grow better or worse than its ancestor in a variety of chemically novel environments, which contain chemical agents that include heavy metal stressors, antibiotics, or acids.
Importantly, our populations were never exposed to any of these conditions during the evolution experiment. A priori , we reasoned that two outcomes were possible. First, populations with high mutation rates may grow better in novel environments, because they can accumulate more beneficial mutations while evolving in their original environment, and these mutations may also be beneficial in novel environments through pleiotropy.
High mutation rate populations can also generate more genotypic diversity, which in turn increases the chances that a population harbors a clone with a latent beneficial mutation that allows it to grow better in a novel environment.
Such latent beneficial mutations can indeed occur, as demonstrated by the classic fluctuation test, which relies on such mutations to estimate mutation rates towards resistance to lethal selection [ 96 , 97 ]. Second, populations with high mutation rates may grow worse in novel environments, because they may accumulate more mutations that are either beneficial or neutral in the current environment, but deleterious in a novel environment. Such latent deleterious mutations do indeed exist [ 36 , 70 , 98 ].
In sum, strains with high mutational pressure may harbor more latent beneficial alleles, but also more latent deleterious alleles, and it is not clear a priori which dominates in their effect on fitness.
We conducted two tests on how mutational pressure can affect growth in novel conditions. However, it yielded a very clear pattern for our MR XL populations: They were not able to grow in any one of these environments, which illustrates that at the highest mutation rates we consider, latent deleterious mutations outweigh beneficial ones in both the ancestor and evolved populations. One possible explanation is that the MR XL strain is inherently more sensitive to novel environments, including the assay environment.
Because the MR XL ancestor population could not grow at all, we were unable to further quantify the effect of the highest mutation rate in these 96 novel environments. In the second test, we periodically measured growth of all 32 replicate populations relative to their ancestors in two stressful conditions: the antibiotic nitrofurantoin a specific narrow stressor and an acidic medium a broader stressor.
For both, we found that strains with higher ancestral mutation rates could grow better than those with lower mutation rates, except for MR XL replicate populations, which grew worst of all populations. This experiment shows that latent beneficial alleles may predominate at low and intermediate mutational pressure, but no longer at high mutational pressure. Our observations are consistent with a previous study showing that multidrug resistance in E.
In sum, a modest increase in mutation rates can provide an evolutionary advantage in both the constant environment of our long-term laboratory evolution experiment and in novel environments. These mutation rates are below those commonly considered to limit adaptation, and highlight the need for additional theoretical work. Our observations show that biological systems may be more sensitive to mutational pressure than simple theoretical models suggest, at least when the effects of mutations are allowed to accumulate over many generations.
This observation may improve the prospects of using elevated mutagenesis to drive pathogen or tumor populations to extinction [ 20 , — ], if high mutation rates can be sustained for a sufficiently long amount of time. We utilized four isogenic E. Strain genotypes are summarized in Table 1. This gene is involved in the methyl-directed mismatch repair system.
We constructed the mutator strain MR L by replacing the mutL region in MR M with the mutL region from ES4 with a kanamycin resistance gene inserted upstream of the region, using the method of Datsenko and Wanner [ ]. We then excised the kanamycin resistance gene using pCP20 [ ], which left a small scar immediately upstream of the mutL gene.
We confirmed the mutation rates of these ancestral strains using fluctuation tests [ ] see "Mutation rate measurements and calculations" for details , and found that the MR M , MR L , and MR XL strains had , , and fold higher mutation rates to rifampicin resistance than MR S.
In additional to these strains, we used the strain E. See Fig 1 for an overview. Each plate held 24 populations arranged in a checkerboard pattern, such that each well was surrounded only by wells with blank medium, and the populations were assigned to the 24 wells at random by a custom R script.
We diluted each culture ,fold every 24 hours into fresh DM medium, which allows almost 17 generations of growth per day.
We delayed the start of the MR S replicates by 63 days for technical reasons. We controlled for contamination in several ways. Second, we periodically checked each evolving culture for contamination by confirming its resistance profile and approximate mutation rate using spot tests. MR S and MR XL replicates can grow on tetracycline, and the replicates with higher mutation rates display more colonies on rifampicin. Third, we examined the genome sequence data for cross-contamination, but detected no evidence for cross-contamination in it.
For populations that do not have a constant number of cells, the effective population size is given by the harmonic mean of population sizes over the course of the dilution and growth cycles of the experiment. Previous studies have estimated the effective population size only from the size of the bottleneck measured during one dilution [ , ].
In contrast, because we recorded the census size of the population at carrying capacity N max every 7 days, we were able to estimate the effective population size as the harmonic mean of the population sizes both at the beginning and at the end of a cycle of growth and dilution. To obtain N max , d at any one day d , we counted the number of cells in each evolving replicate population in stationary phase just before transferring the population into fresh media.
We discarded plates with fewer than 20 or more than colonies for the purpose of this analysis. Thus, during each generation g of each growth cycle, a population assumed population sizes. We then determined the nominal effective population size N e of a replicate population during its entire lab evolution as which is the harmonic mean of all the population sizes. We calculated it for all 25 days on which we collected population size data.
The number 18 corresponds to the total number of generations g for which we computed population sizes during any one of these 25 days. We also estimated the effect of linkage on reducing the effective population size due to background selection or interference selection [ 53 — 55 , ]. We periodically obtained a proxy for the fitness of the evolving strains by measuring growth curves of the archived populations.
For each time point, we restarted all evolving populations as well as three replicates from each ancestral population and three replicates of wild type E.
During this time, we read the absorbance at nm every 10 minutes. We fit the classic logistic equation describing population growth to the data [ ], using the Growthcurver R package [ ], and defined the relative fitness of each population as r evo —r anc.
Here, r evo is the growth rate of the evolved population and r anc is the mean growth rate of the three replicates of the ancestor grown in the same plate reported in units of cell divisions per hour. We measured each growth curve three times. We used the R package lme4 v1. In this analysis, we chose the mutation rate classes as fixed effects, and the replicate population as well as the well plate as random effects.
For all linear mixed effects analyses conducted in this paper, we observed no deviations from homoscedasticity according to Levene's test for homogeneity of variance [ ] implemented in the R package car v2. Also, all residuals were normally distributed unless otherwise specified.
We obtained significance values using a likelihood ratio test of the full model against a null model that did not contain the fixed effects.
Using the data from the above growth curve experiments, we also compared the fitness of the ancestor populations against each other by obtaining the relative fitness of the ancestors as r anc —r K12 , where r anc is the growth rate of the ancestor population and r K12 is the mean growth rate of the three replicates of E.
We performed a linear mixed effects analysis of the relationship between the ancestral fitness relative to E. In this analysis, we chose the mutation rate classes as fixed effects, and the identity of the original glycerol stock and of the well plate as random effects.
We used the R package multcomp v1. We sequenced samples from the four ancestral populations day 0, generation 7 and from each of the 32 evolving replicate populations at days 63, , and generations , , and For simplicity, we hereafter designate these time points as generations 0, , , and No , with modifications as previously described [ ]. Importantly, we used no PCR steps in preparing the libraries.
We used breseq v0. We developed scripts in R to identify the alterations that occurred in the evolved populations, but were not fixed in their ancestors. Because DNA is double-stranded, the remaining possible point mutations are covered by their reverse complements, e.
We computed the relative frequencies of each mutational class for each replicate population, and used these to perform a principal component analysis PCA in R with prcomp, which uses singular value decomposition for the PCA.
We defined the center of the mutational cloud of genomes as the location in genotype space defined by the majority allele at each site. We define the population spread metric C as the average of C n over all sites in the genome. A related quantity is the approximate sequence distance D that an evolving population has moved from its ancestral genotype, i. In other words, D corresponds to the total number of sites at which the majority allele is different from the ancestral allele.
We also computed each population's average genome-scale nucleotide site diversity [ , ] using the pairwise alignment position nucleotide counting approach [ , ].
We estimated the proportion of pairwise nucleotide differences at each site n as where m p is the number of reads corresponding to the majority allele and m is the total number of reads at site n.
We estimated the average nucleotide diversity for the L positions in our genome having non-zero coverage as. In this analysis, we chose the mutation rate classes as fixed effects, and the time points and each of the 32 evolving replicates as random effects. We identified putatively beneficial mutations as mutations that occurred in a genomic region more often than one would expect by chance alone. To identify such mutations, we used a numerical approach that focuses on a given gene g among a larger set of genes or genomic regions G e.
If all sites in the genomes of all samples were equally likely to experience a mutation, and if different genes were likely to experience mutations only in proportion to their length, then the probability p g that any one gene g receives such a mutation in any given replicate would depend only on the length of the gene l g ,. We computed the probability of observing the n g mutations in any given set of replicates as the probability that gene g was mutated in each member of the set of replicates times the probability that it was not mutated in any of the other replicates.
This quantity P g is our null expectation that two replicates acquire mutations in gene g , if each replicate population's mutations were randomly distributed across its genome. We were interested in genes containing mutations in improbably many replicate populations, which we identified as those genes having less than a 0. Similarly, the probability of observing a mutation in exactly 3 replicate populations is given by.
We estimated the mutation rate of a single clone isolated from each ancestor and from each evolved replicate population through fluctuation assays that screened for mutants resistant to rifampicin [ ], which can be caused by mutations in the rpoB gene. Specifically, we performed the following procedure for each replicate population. We obtained the genomic mutation rate U using Drake's approach [ 74 ] by first determining the "correction factor" C , which counts the number of single nucleotide mutations in rpoB that show rifampicin resistance.
This mutation rate may be an underestimate because we neglected other types of mutations e. Our phenotype screening revolved around the density of cells after growth in various chemicals. We resuspended the colonies from the final third round plates in IF-0 solution Biolog, Inc. To determine the minimum threshold for detection of growth in a given compound C , we computed the absolute difference between the readings in a given well across all pairs of samples i , j after 10 minutes before cells had started to grow and divide , i.
The values of A i-j , C , 10m quantify the expected experimental noise of wells with no growth. Each compound in the Biolog Phenotype MicroArrays we used occurs in four wells at increasing concentrations.
For further analysis, we used data only from the concentration the well that showed the highest variation in the difference between matched evolved and ancestor strains across all samples.
We considered a sample to have evolved tolerance to a compound C if it improved its phenotype after generations of evolution more than expected based on experimental noise, i. Likewise, we considered that a sample had lost tolerance if its phenotype had degenerated after generations of evolution, i. We note that both cellular growth and respiration contribute to the Biolog phenotype B S , C , because respiration can occur independently of cellular growth [ 70 , ].
We were also interested in observing the evolutionary dynamics of phenotypes over time. The phenotypes we selected for this analysis are the cell density after 24 hours of growth of the evolved populations relative to their ancestors in two conditions: a narrow antibiotic nitrofurantoin stress, and a broader environmental low pH stress.
Specifically, we chose DM medium with 1. Nitrofurantoin is one of the phenotypes where evolved populations gained tolerance in the Biolog analyses, and acid stress has been well-studied in E. To control for changes in cell density at stationary phase, we also performed a control measurement in the standard medium, DM Specifically, we measured the growth of evolved replicate populations at days 28, 63, 91, , , and generations , , , , , and , hereafter designated as generations , , , , , and We incubated the resulting well plates for 24 hours, and then measured the absorbance at nm.
In order to quantify the relationship between the normalized fold-change in cell density G and ancestral mutation rate, we performed a linear mixed effects analysis using the R package lme4 v1. For the nitrofurantoin and pH stressors, we used data from the experimental condition with the most variability between replicates 2.
We tested for homoscedasticity using the R package car v 2. We counted the number of cells in stationary phase just before our daily transfer at regular intervals. Each point is the average cell density of an evolving replicate population at a given generation.
One standard deviation above and below the mean is depicted with a shaded line. We counted the number of cells at regular intervals, and used these counts to estimate A the nominal effective population size N e for each replicate population. Because our populations are asexual, the effects of selection on polymorphisms linked to neutral sites will make drift at neutral sites appear much stronger than indicated by these estimates.
To account for such effects, we also made rough estimates of the effect of linkage on the effective population size using two published methods further described in Methods , which compute the "Gordo" N e B , and the "Good" N e C. Together, panels B and C suggest that the effective population size may be much smaller than the nominal population size. Each circle shows the N e estimate of a replicate population, the center line of the box plot is the median value, and the top and bottom edges of the box correspond to the first and third quartiles.
A Fitness differences between ancestral replicate populations and E. Each circle shows the growth rate of a replicate population for a given strain horizontal axis minus the growth rate of E.
Overall, 54 experimental estimates were made for each strain. B Fitness differences between each evolving replicate population and a common reference strain E. Shaded areas indicate one s. A C The fitness difference between each evolving replicate population and its ancestor and its change over time is depicted in separate panels for each strain and replicate.
Panels corresponding to the replicates randomly chosen for further characterization in Biolog plates are outlined with a heavy black border. B Variance in relative fitness for the replicate populations of each strain. Strains with higher ancestral mutation rates have more variability in the relative fitness of their evolving populations than those with lower mutation rates.
Each line in a given panel shows the frequency of one SNP in one replicate population vertical axis at generations 0, , , and horizontal axis. The color of the line indicates the type of SNP. Types of SNPs with likely functional consequences are emphasized in brown nonsense mutations and green nonsynonymous mutations. The frequency of newly-arising SNPs after one day of growth in the ancestral populations. Several of the observed SNPs, particularly those occurring at higher frequencies, may have been transferred to the eight replicates.
A Nucleotide changes are depicted along the horizontal axis. For each type of mutation, we computed how often it occurred at any time point during the evolution experiment relative to all other types Methods. B The mutational spectra from replicate populations evolved from ancestors with different mutation rates do not clearly separate when projected onto the first two principal components PC1 and PC2 in a principal component analysis Methods.
Because we evolved eight replicate populations for each strain, each vertical stack of dots can harbor at most eight dots. For many genes, all MR XL replicates share the same nucleotide change, which likely already occurred in the shared ancestor. A Genes with different mutations in the same gene in different replicates, and B genes where all the MR XL replicates share the same nucleotide change the nucleotide changes found in the MR L replicate populations for betI and torA are not the same as found in the MR XL populations.
Each circle corresponds to one evolving replicate population. The size of a circle is proportional to the frequency at which a mutation is found in a population, and can change over time horizontal axes. All replicates for all strains circles inside each panel are depicted for each gene labeled in the top, left of each panel. Importantly, all tested ancestor and evolved MR XL strains failed to grow in every one of the 96 environments. Each circle represents the ancestor's density horizontal axes and the evolved replicate population's density vertical axes in a particular environment.
Points above the diagonal line correspond to conditions in which an evolved replicate population outperformed its ancestor; points below the line correspond to conditions in which an evolved replicate population underperformed its ancestor.
We consider a population to have evolved tolerance to a condition when its density is larger than the ancestral density in the same condition, excluding differences attributable to experimental noise.
Conversely, we consider a population as having experienced decay if its density after evolution is smaller than that of its ancestor see Methods. Both gains and decays are indicated by solid circles. Open circles indicate that no gain or decay was detected for that condition, or that the difference between the evolved and ancestral cell density could be due to experimental noise. When the environment under which the bacteria are grown is changed, however, in a way that is somehow detrimental to the population, it will often adapt itself rapidly and effectively to the new conditions.
A good example of the way in which a bacterial culture may adapt to an unfavorable environment is the reaction of Escherichia coli to streptomycin. Most strains of this bacterium are sensitive to streptomycin, and are unable to multiply in the presence of even very small amounts of the antibiotic.
Sensitivity to streptomycin is an inherited trait and is transmitted, unchanged, through countless generations. If a high concentration of streptomycin is added to the culture tube in which a sensitive strain is growing, the outcome depends upon the size of the population at the time.
If the number of bacteria in the tube when the antibiotic is added is relatively small a hundred or a thousand , multiplication will stop at once, and no further growth will take place in the tube, no matter how long it is incubated. If the population is large a hundred million bacteria or more , the addition of streptomycin will arrest multiplication sharply, but incubation of the tube for a few days will almost always result in the ultimate appearance of a fully grown culture containing tens of billions of bacteria.
When the bacteria in this culture are tested, they prove to be completely resistant to streptomycin, and are able to multiply vigorously in its presence. Further, we find that resistance to streptomycin is a stable, hereditary characteristic, transmitted indefinitely to the descendants of these bacteria. Thus, by exposing a large population of streptomycin-sensitive bacteria to a high concentration of the antibiotic, the emergence of a genetically resistant strain can be brought about.
This, indeed, is a strikingly adaptive change, and at first sight it may seem to substantiate the old idea that the environment can cause useful modifications that are then inherited.
The careful study of the events leading to the appearance of a streptomycin-resistant strain proves without doubt that this is not so.
It can be readily demonstrated, first of all, that the adaptation to streptomycin does not come about by the mass conversion of the entire sensitive population, but rather is the result of the selective overgrowth of the culture by a few individuals that are able to multiply in its presence, while the division of the rest of the population is inhibited.
It is for this reason that adaptation occurs only when the exposed population is large enough to contain at least one such individual. The critical question is this: how did these rare individuals acquire the properties that enabled them and their descendants to multiply in the presence of streptomycin? This question has deep roots in biological controversy. It recalls, in a new form, the arguments over Lamarck's idea that modifications of the individual caused by environment can be inherited by descendants.
Although Lamarckism has long since been disproved to the satisfaction of most biologists by repeated demonstrations that such inheritance just doesn't happen, the idea has persisted in bacteriology until very recently that microorganisms are somehow quite different from other plants and animals, and that permanent hereditary changes of an adaptive kind can be produced in bacteria directly as a result of the action of the conditions of life.
Two alternative hypotheses can be considered in planning experiments to determine the true origin of streptomycin-resistant variants. The first is that a small number of initially sensitive bacteria were modified as a direct result of the action of streptomycin, thereby acquiring permanent resistance.
This would be an example of an adaptive hereditary change caused by the environment, as Darwin envisaged the origin of most hereditary variations. The second possibility is that the resistant individuals had already acquired the properties necessary for resistance before coming into contact with streptomycin, as a result of a mutation during the normal division of the sensitive population.
In this case, the role of the antibiotic would be entirely passive, providing conditions that favor selectively the multiplication of those rare individuals present in the population that are already equipped, by virtue of the previous occurrence of a chance rearrangement of a particular gene, to withstand its inhibitory action.
During the past fifteen years, a great many experiments have been designed and conducted in a number of laboratories for the purpose of determining which of these hypotheses is correct. They have established beyond doubt that the second one is right, and that streptomycin-resistant variants originate by mutation, at a very low rate, during the growth of sensitive strains that have never been exposed to streptomycin.
The proof depends upon the demonstration that the very first generation of resistant individuals in a culture, to which streptomycin has just been added already consists of related family groups, or clones, in just the way that would be predicted if their resistance were the consequence of a hereditary change that had taken place some generations back.
The development of resistance to streptomycin illustrates the way in which mutations provide the basis for adaptive changes in bacterial populations. Actually, any culture of Escherichia coli, apparently quite homogeneous when hundreds or even thousands of bacteria are compared, contains within it rare variants that differ from the predominant type in one or more of countless ways. When a suitable selective environment is provided, it can be shown that a culture contains mutants resistant to many antibiotics, to the action of radiation, to all sorts of chemicals that inhibit particular steps in metabolism — mutants that differ from the standard type in the sugars they can ferment, in their rate of growth, in the complexity of their nutritional requirements, in their antigenic properties, and in almost any characteristic for which a method of detection can be found.
In every case that has been carefully studied, these differences are found to originate without any contact with the conditions under which they happen to be advantageous, and their rates of occurrence are ordinarily not increased by such contact. This is true not only in bacterial cultures, where mutations can be demonstrated rapidly and dramatically.
Natural populations of other plants and animals, including man, are known to contain mutations of many kinds that occur with no apparent causal relation to the conditions of growth. Thus, in a way that Darwin could not have surmised, chance, through mutation, plays a most important part in evolution.
It would be difficult indeed to imagine how a species could long survive, or progress in evolution, if it were dependent for its flexibility upon variations directly caused by the conditions of life.
Quite aside from the fact that modifications produced in this way are not inherited, except in very special cases, it would require the intervention of some purposive and prescient agent to guarantee that previously unencountered conditions could typically provoke in the organism just those responses that are required to enhance adjustment.
Of course, the occurrence of a diversity of mutations in populations of bacteria and other organisms does not necessarily equip them to meet successfully every environmental challenge. Some strains of bacteria, for instance; are unable to adapt to streptomycin, since their spectrum of mutations does not include the particular modification of metabolism that is required for streptomycin resistance. Furthermore, since there are limits to the range of conditions that can support life, any sufficiently drastic changes, such as those that would take place in the center of a hydrogen bomb explosion, are not likely to prove conducive to the survival of any living thing.
Even within the range of more tolerable conditions, the suddenness of change is sometimes more decisive than its magnitude. For example, the bacterium Escherichia coli can be made resistant to streptomycin, penicillin, and chloromycetin, if the mutants resistant to each of these antibiotics are selected sequentially, but such a triply resistant strain cannot be obtained if the sensitive strain is exposed simultaneously to all three agents.
This is explained by the negligible probability that any one individual in a finite population will have undergone mutation in three particular genes, each of which mutates very infrequently and independently of the others.
Observations of this kind, incidentally, although originally made in laboratories of genetics,, have found important applications in medical practice. Many people who have used antibiotics to combat infection have had the experience of dramatic relief of symptoms, only to be followed within a few days by a recurrence, this time failing to respond to the same antibiotic.
Sometimes this can be explained by selection of a variant, present in the infecting population of bacteria, that is resistant to the antibiotic and that has its chance to multiply once the sensitive population is eliminated by the first round of treatment.
In some cases, a physician will recommend the use of a combination of two or more unrelated antibiotics simultaneously, knowing that mutants resistant to more than one such drug are much less likely to be present.
While the use of combinations of antibiotics is not always feasible for medical reasons, under certain conditions it has effectively prevented the occurrence of relapses caused by selection of resistant variants. There is, of course, much more involved in the complicated saga of evolution than the simple picture of mutation and selection that accounts for bacterial adaptation to streptomycin.
0コメント