What is the development of pesticide resistance?

It has the menacing sound of an Alfred Hitchcock movie: Millions of rats aren't even getting sick from pesticide doses that once killed them. In one county in England, these "super rats" have built up such resistance to certain toxins that they can consume five times as much poison as rats in other counties before dying. From insect larvae that keep munching on pesticide-laden cotton in the U.S. to head lice that won't wash out of children's hair, pests are slowly developing genetic shields that enable them to survive whatever poisons humans give them.

Rachel Carson predicted such resistance in her groundbreaking book Silent Spring, published soon after the chemical insecticide glory days of the 1950s. And the problem is getting worse. Farmers in the U.S. lost about seven percent of their crops to pests in the 1940s. Since the 1980s, some 13 percent of crops are being lost -- and more pesticides are being used.

It's a huge problem, but the pests are only following the rules of evolution: the best-adapted survive. Every time chemicals are sprayed on a lawn to kill weeds or ants for example, a few naturally resistant members of the targeted population survive and create a new generation of pests that are poison-resistant. That generation breeds another more-resistant generation; eventually, the pesticide may be rendered ineffective or even kill other wildlife or the very grass it was designed to protect.

These days, farmers and backyard gardeners alike are trying to outsmart the pests by using a variety of natural methods. With "integrated pest management," scientists encourage the spread of natural enemies of pests, or they lure the pests with a meal that's even more tasty than the vulnerable crop. Pesticides are only used as a last resort if every other method fails. Integrated pest management has had some successes, but pesticides are still the world's most popular way to kill pests. Something to ponder the next time you can't seem to kill a cockroach.

Page 2

close window

June 19, 2015

This is one of a number of CropWatch articles focusing on resistance management. See more and learn how to reduce the risk of resistance developing on your farm.

Pesticide resistance is a decreased response of a pest population to a pesticide as a result of previous exposure.  It can be thought of as kind of accelerated evolution, where a pest population is exposed to a selection pressure, in this case a pesticide, and over time the "response" of the population may result in resistance.

Pesticide resistance management can be defined in a variety of ways, depending on your perspective:

• A process of delaying the development of resistance in a pest population.
• A set of methods designed to extend the number of generations that a given pest population can be controlled economically with a given technology.
• Managing gene flow and selection.

Factors Affecting Resistance Development of the "Pest" in the Environment

Pesticide resistance occurs in an ecological context where biological and environmental factors are constantly interacting. These factors can be divided into biological, behavioral, and operational.

• Biological factors that can affect resistance development include characteristics of:
  • an organism's life cycle,
  • the organisms rate of its population growth, and
  • the genetics of the organism.
• Behavioral factors that can influence the development of resistance include:
• Operational factors are the things we do to manage pest populations and interact with all the previous factors. Manipulating these operational factors varies selection pressure and is important in resistance management.

To better understand how these factors and their interactions can influence the development of pesticide resistance, visit the online educational module at: http://passel.unl.edu/communities/pesticiderm and click on "Biological/Ecological Factors" to begin the module.

Gary Hein
Director, UNL Doctor of Plant Health Program

As the world anticipates feeding nine billion people as sustainably as possible by 2050, crop protection against pest insects, diseases and weeds has a vital role in maintaining and improving crop yields (Godfray et al., 2010). Whilst awareness is growing of the importance of integrated pest management, pesticides remain a necessary part of the pest‐control toolbox for many crops (Swanton et al., 2011), in combination with other approaches such as disease‐resistant crop varieties (Carolan et al., 2017). However, just as the effectiveness of antibiotics in the control of human disease is under threat due to the evolution of resistant strains of bacteria (World Health Organisation, 2014), the control of agricultural pests and crop diseases is threatened by the evolution of pesticide resistance, affecting insecticides (Bass et al., 2015), herbicides (Powles & Yu, 2010) and fungicides (Lucas, Hawkins & Fraaije, 2015). Modern pesticides aim for specificity to reduce non‐target effects in the environment (Vyas, 1988), but this specificity also means that resistance is more evolutionarily accessible to the intended target pests.

Most studies of pesticide resistance focus on the proximate biochemical mechanism conferring reduced sensitivity to the compound in question, whether by target‐site mutations or over‐expression, or metabolic breakdown or efflux of the pesticide. However, the emergence and spread of resistance are evolutionary processes, and greater understanding of the evolutionary mechanisms involved can inform resistance risk assessment and management strategies (Maclean et al., 2010; Neve et al., 2014). Furthermore, pesticide resistance is a key example of evolution in action, with rapid evolution under novel selective pressures (Palumbi, 2001), and has the potential to contribute to fundamental understanding of general evolutionary processes. Resistance could be considered as an example of evolutionary rescue (Alexander et al., 2014), and could address questions concerning adaptation to changing environments, sources of variation and origins of novel traits.

In this review, we consider what is known about the evolutionary origins of resistance to pesticides, and where further studies are needed. We focus on the relative importance of de novo mutations and standing variation, the role of interspecific gene transfer, and pre‐adaptation through pleiotropic effects of existing adaptations.

In evolutionary terms, a de novo mutation is one which originates once an environmental change has occurred making that mutation selectively advantageous (Messer & Petrov, 2013), and emerges under that selection. In the case of pesticide resistance, this would mean a mutation conferring resistance occurs after the introduction of the pesticide (Fig. ​1A). A mutation may arise once and spread through the population, or there may be multiple, independent de novo origins.

A small number of studies have implicated possible mutagenic effects of β‐lactam antibiotics (Gutierrez et al., 2013) and quinone outside inhibitor (QoI) fungicides (Schnabel & Chen, 2013; Chen et al., 2015), and speculated possible similar effects for herbicides (Gressel, 2011). However, the general assumption for de novo resistance evolution is that mutation continues at the background rate, but strong selection by the pesticide means that mutations conferring resistance will increase in frequency, rather than mostly being lost through genetic drift.

In the case of selection from standing variation, polymorphisms are already present in the population prior to the change in selective pressure (Barrett & Schluter, 2008), whether in a single gene or a polygenic trait (Pritchard & Di Rienzo, 2010). Following the environmental change, a pre‐existing neutral or deleterious allele becomes advantageous and is selected to a higher frequency in the population (Fig. ​1B).

Cryptic variation (Gibson & Dworkin, 2004) could include polymorphisms in genes that are only highly expressed in resistant individuals (Rajon & Masel, 2013), or variation interacting epistatically with resistance mutations, such as compensatory mutations increasing the stability of mutant proteins (Cools et al., 2010). Therefore, even when resistance itself emerges through a de novo mutation, standing variation in genetic background may affect the fitness of resistant individuals.

Standing variation means that resistant alleles are already present, but at low frequencies prior to pesticide selection. However, some species have a degree of pre‐existing resistance already fixed in the population prior to pesticide exposure. In some cases, the level of pre‐existing resistance is high enough that the species is considered intrinsically resistant (Lucas et al., 2015). In other cases, pre‐existing adaptations such as the efflux or metabolism of naturally occurring toxins are insufficient to confer intrinsic high‐level resistance to pesticides, but once under selection for their pleiotropic effects on pesticide resistance, they may evolve increased activity, for example by overexpression (Kretschmer et al., 2009) or gene amplification (Mouches et al., 1986). Such over‐expression or amplification may originate through de novo mutations, including transposons or duplications, or be selected from standing variation. In this scenario, a pre‐existing adaptation is co‐opted as a resistance mechanism, and pesticide resistance, previously a co‐incidental pleiotropic effect, becomes the major selective force on the further evolution of that trait. This pre‐existing resistance may be due to exposure to similar, naturally occurring compounds (Bass et al., 2013), or to pleiotropic effects of non‐xenobiotic‐related adaptations (Song et al., 2011): this is discussed in more detail in Section III.

Where pesticide resistance (intrinsic or acquired) is present in one species, the resistance allele may sometimes move into other, target pest species, through interbreeding or by horizontal gene transfer (HGT) (Fig. ​2A). Alternatively, occurrence of the same resistant allele in different species may be the result of parallel evolution, for genetic changes that are simple enough to have arisen independently in multiple lineages (Fig. ​2B); or, in sister species, shared resistant alleles may be the result of collateral selection, whereby the sensitive and resistant alleles diverged before the species separated, followed by selection of the common resistant allele in both species (Fig. ​2C) (Stern, 2013).

Phylogenetic origins of pesticide‐resistant alleles. The upper part of each panel shows genes superimposed onto a species tree; the lower part shows the resulting gene tree. (A) Allele transfer by horizontal gene transfer or adaptive introgression; (B) parallel evolution by independent de novo mutations; (C) collateral selection from pre‐speciation standing variation. AS: species A, sensitive allele; BS: species B, sensitive allele; AR: species A, resistant allele; BR: species B, resistant allele. Grey lines, sensitive allele; black lines, resistant allele; dashed lines, sensitive allele may have been lost from sister species if resistance has previously reached fixation.

Interbreeding may include adaptive introgression, in which hybridisation between a ‘donor’ species and a target species is followed by back‐crosses with the target species during which positive selection results in the retention of a specific gene from the donor species (Hedrick, 2013). For example, the house mouse Mus musculus domesticus has gained rodenticide resistance alleles through hybridisation with the intrinsically resistant Algerian mouse Mus spretus, followed by introgression under rodenticide selection (Song et al., 2011; Liu et al., 2015).

HGT is most common in bacteria, to the extent that for plasmid‐borne antibiotic resistance the source population for standing variation is a multi‐species ‘resistome’ (Forsberg et al., 2012). However, HGT has also been reported in some eukaryotes, including fungi (Friesen et al., 2006).

Whilst the repeatability of evolution is a long‐standing fundamental question in evolutionary biology (Lässig, Mustonen & Walczak, 2017), when applied to resistance it also has important practical implications. Molecular diagnostics can be developed for the rapid detection of resistant alleles, but the reliability of such diagnostics depends upon the likelihood that different populations have evolved the same or different resistance mechanisms. Furthermore, the applicability of management strategies across different pest populations may depend on whether the same resistance mechanisms are evolving in each case.

Experimental evolution studies suggest that de novo mutations show low repeatability, with replicate populations showing phenotypic but not genotypic convergence (Bedhomme, Lafforgue & Elena, 2013), whereas selection from standing variation is more repeatable (Burke, Liti & Long, 2014) since collateral evolution can take place by selection from the same allele pool (Stern, 2013) without the need for random new mutations within each population.

However, pesticide use exerts an exceptionally specific selective pressure, especially in cases of target‐site resistance where selection acts not just on a single trait but often on a single molecular target and even specific pesticide‐binding residues, with further constraints imposed by the need to maintain the original function of the target protein (Cools, Hawkins & Fraaije, 2013). This can result in parallel de novo mutations, such as G143A in cytochrome b, responsible for most cases of resistance to QoI fungicides (Grasso et al., 2006), and A302S in Rdl, associated with resistance to cyclodiene insecticides in multiple insect orders (Thompson, Steichen & Ffrench‐Constant, 1993). By contrast, the complex constraints and epistatic interactions shown by CYP51 mean that azole resistance has evolved through a wide range of different target‐site mutations (Cools et al., 2013).

Furthermore, experimental evolution studies into selection from standing variation use replicate populations with the same initial genotype mix (Burke et al., 2014), but field populations may have different initial gene pools. Populations that have undergone bottlenecks will have lost some of the initial standing variants, whereas populations that have been isolated for a long time may have independently accumulated subsequent unique mutations into standing variation.

Therefore, the repeatability of selection from common standing variation depends upon the degree of divergence between populations, whereas the repeatability of de novo mutations depends on the functional constraints affecting each target site and other possible resistance mechanisms. In the case of interspecific gene transfer, introgression from a known source such as a herbicide‐resistant crop would be more repeatable than independent evolution of resistance within pest species, whereas introgression from a wider pool of wild relatives, or HGT from more distantly related species, would be less repeatable.

Standing variation comprises alleles already present in the population, whereas for de novo mutations there will be some waiting time for suitable mutations to occur (mutation limitation), depending on population size and mutation rate (Karasov, Messer & Petrov, 2010). Therefore, resistance could be expected to emerge more rapidly if resistant alleles already exist within the standing variation; whereas for de novo mutations, initial emergence as well as subsequent selection of mutations should be considered in resistance management strategies (Hobbelen, Paveley & Van Den Bosch, 2014).

Furthermore, a de novo mutation starts as a single copy, whereas alleles present as standing variation will already be present in multiple individuals (Hermisson & Pennings, 2005). Since the probability of loss or fixation of a resistance allele depends on both the initial frequency and the selection coefficient, a single‐copy de novo mutation would only be likely to emerge if it had a high positive selection coefficient, whereas a standing variant already present at a higher frequency may increase in frequency even with a weaker selective benefit. For pesticide resistance, the selection coefficient is higher for mechanisms conferring greater levels of resistance, but reduced by any associated fitness costs. Therefore, resistance mechanisms conferring high resistance factors can emerge from de novo mutations, whereas mechanisms conferring only partial resistance, or carrying significant fitness penalties, are more likely to emerge from standing variation. The emergence risk of partial rather than complete resistance has implications for the optimal dose rates for resistance management (Van Den Bosch et al., 2011).

Similarly, in diploid and polyploid species, a recessive allele is less likely to arise de novo or by hybridisation, since a new gene would initially be heterozygous, whereas a standing variant may already have reached sufficient frequency for homozygotes to be present in a recombining population so selection is less dependent on genetic dominance (Orr & Betancourt, 2001). Introgressed genes would also be heterozygous in the F1 generation, with further crosses, or backcrosses with other hybrid progeny, required before a recessive allele could confer a fitness effect, so emergence probability would depend on dominance as well as hybridisation frequency.

The risk of resistance emerging through pleiotropic co‐option depends on the presence of suitable efflux pumps or metabolic pathways, and the evolutionary distance between their initial state and acquired resistance to the pesticide. There may be a greater risk of a suitable efflux pump or metabolic pathway being available in species dealing with a wider range of natural toxins, but structure–substrate relationships are not sufficiently well understood to predict which specific efflux or metabolic genes may contribute to resistance risk for a given pesticide.

Whilst de novo adaptations to sudden environmental changes must emerge rapidly, standing variation may have accumulated over a longer time scale, allowing rarer genetic changes to occur and more complex traits to evolve. It has also been argued that complex genetic changes are more likely to originate as cryptic variation, since less‐fit intermediate states would not be exposed to negative selection (Rajon & Masel, 2013).

In the case of interspecific transfer from an intrinsically resistant species, the longer timescales over which resistance has evolved in the source species also allow rare or multiple genetic changes to occur, but wider genomic co‐adaptation may be lost with crossing or with the horizontal transfer of a small genetic region.

A de novo mutation with a single origin may spread through the movement of insects, seeds and pollen, or spores, whereas an allele present in standing variation may already be present throughout the range of a pest if it arose prior to the spread of the pest species to different regions. Therefore, the origin of resistance can determine the most effective containment strategy. Where a de novo resistance mutation is present in a limited area, quarantine measures could limit its spread, whereas if resistance is already present within standing variation across a pest's range, resistance management to prevent parallel selection within each region is more important. For interspecific origins, a resistant strain originating from a rare HGT, or a hybrid with a relative with a restricted range, should be treated more like a de novo mutation, with quarantine measures considered, whereas crosses with a widespread relative or the crop itself are more likely to recur independently in other regions and efforts should be focussed on resistance‐management practices that reduce local selection.

A major current debate concerning the origin of resistance is centred on azole resistance in the soil‐dwelling saprophyte and opportunistic clinical pathogen Aspergillus fumigatus. Resistance to clinical azole anti‐fungal drugs is a growing problem, and research is ongoing to determine the origin of resistance in previously untreated patients. The occurrence of specific combinations of coding mutations and promoter inserts points to selection from standing variation in inoculum rather than de novo mutations within each patient (Verweij et al., 2013), but the origins of those resistant genotypes in the environmental population, and whether they have been selected by agricultural or other uses of azole fungicides, are not yet clear (Gisi, 2014).

In the case of genetically modified or mutagenised herbicide‐resistant crops, resistant weeds may evolve due to gene flow from the crop, or from independent parallel evolution within the weed species (whether from de novo mutations or standing variation within the weed population). This has implications for whether genetic modification itself is culpable for the evolution of ‘superweeds’, or whether it is purely the result of associated agronomic practices involving extensive repeated use of a single herbicide mode of action. Risk assessments for the release of new herbicide‐resistant crops consider the presence of related weed species in areas where the crop will be grown, but growing evidence that most ‘superweeds’ result from parallel selection rather than hybridisation (see Section V.2) indicates that agronomic risk factors and resistance‐management guidelines are more important.

Next‐generation high‐throughput sequencing methods are becoming increasingly important tools in resistance studies, especially for non‐target‐site mechanisms (e.g. Van Leeuwen et al., 2012; Gaines et al., 2014; Omrane et al., 2015). Population genomic analyses can identify genes under selection. New methods are being developed to detect selective sweeps that are currently underway before the selected gene has reached a high frequency in the population, but the success of these methods depends on whether the selective sweep underway is a hard sweep from a single de novo origin or a softer sweep from multiple or standing mutations.

Resistance studies more commonly look for genomic or transcriptional differences correlated with resistant phenotypes (e.g. Omrane et al., 2015). The evolutionary origin of resistance mutations is also relevant to the design of these studies: where a resistance mechanism originates from a single de novo mutation with a recent hard selective sweep, hitchhiking genes may also appear to be correlated with resistance, so it would be preferable to carry out crosses if possible and sequence the progeny rather than field strains, through a bulk segregant approach as used in previous random amplification of polymorphic DNA (RAPD) marker studies (Fabritius, Shattock & Judelson, 1997), or a quantitative trait locus (QTL) approach (Lendenmann, Croll & Mcdonald, 2015). Where a resistance mechanism has been selected from standing variation in an outcrossing species, field isolates will already have undergone extensive recombination, so additional resistance‐correlated genes may point to intragenomic co‐adaptation, with those genes conferring a selective advantage in the presence of the primary resistance allele.

Evolutionary researchers have considered the relative roles of de novo mutations and standing variation in experimental evolution (Burke et al., 2014). However, for pesticide resistance, selection experiments also have practical significance in attempting to predict the future evolution of resistance to new pesticides.

Selection experiments from a single parental isolate investigate the potential for de novo mutations, since there is no standing variation, and may use ultraviolet (UV) or chemical mutagenesis to increase the mutation rate. Scalliet et al. (2012), Fraaije et al. (2012) and Gutiérrez‐Alonso et al. (2017) used in vitro selection of individual fungal isolates with UV mutagenesis to explore possible mutations conferring resistance to newly developed succinate dehydrogenase inhibitor (SDHI) fungicides. This approach was appropriate given the apparent prevalence of de novo point mutations in target‐site resistance to fungicides, and the sdhC‐H152R substitution generated by Scalliet et al. (2012) has subsequently been reported in the field (Dooley et al., 2016). However, UV mutagenesis did not predict the emergence of field strains with slightly reduced SDHI sensitivity due to a non‐target‐site mechanism comprising overexpression of the major facilitator efflux pump MgMfs1 (Omrane et al., 2015).

Lagator, Colegrave & Neve (2014) and Lagator et al. (2013) used the alga Chlamydomonas reinhardtii as a laboratory model organism for the experimental evolution of herbicide resistance, investigating the effects of alternating or mixing different modes of action. The parental strain was isolated from a single‐cell colony, and C. reinhardtii is asexual in laboratory conditions; therefore, all differences in herbicide sensitivity must have arisen de novo during the experiment. This may affect the likelihood of strains emerging with simultaneous resistance to multiple herbicides.

Selection experiments with chemical mutagenesis in the model insect species Drosophila melanogaster have been used to identify or confirm target sites of neonicotinoids, spinosyns and methoprene, but their accuracy in predicting which mutations will arise in field populations of pest insects has been variable (Shemshedini & Wilson, 1990; Perry et al., 2008; Watson et al., 2010). However, mutagenesis and selection of laboratory populations of Australian sheep blowfly, Lucilia cuprina, have successfully reproduced insecticide‐resistance alleles previously detected in the field, including target‐site resistance to dieldrin and major‐gene metabolic resistance to organophosphates (McKenzie & Batterham, 1998). In this case, the metabolic resistance was conferred by gain‐of‐function single‐nucleotide polymorphisms (SNPs), which could readily arise de novo.

In contrast to such in vitro mutagenesis studies, resistance risk assessment through the screening of field populations to detect outliers from the baseline sensitivity distribution (Russell, 2004) investigates the presence of reduced sensitivity in standing variation (Espeby, Fogelfors & Milberg, 2011; Ulber, Nordmeyer & Zwerger, 2013).

The evolutionary origins of insecticide‐resistance alleles may include standing genetic variation, de novo mutation or, as has recently been demonstrated, adaptive introgression (Table ​1).

Summary of cases of pesticide resistance where evolutionary origins have been inferred

Many early studies of insecticide resistance assumed de novo mutation as the source of resistance, particularly in the case of target‐site resistance where mutations were identified in functionally constrained receptors in the insect nervous system. More recent work has provided evidence supporting these initial assumptions with two examples: target‐site resistance to pyrethroids, and to organophosphates (OPs) and carbamates (CMs), in mosquitoes. Pyrethroid resistance in Anopheles gambiae is commonly associated with knockdown resistance (kdr) mutations (L1014F and L1014S) in the target site, the voltage‐gated sodium channel. Lynd et al. (2010) employed haplotype diversity analyses to investigate genetic variation around the kdr locus in An. gambiae s.s. A loss of genetic diversity was associated with the kdr mutations, with the L1014F kdr mutation exhibiting an especially pronounced footprint of a hard selective sweep, strongly suggestive of evolution by de novo mutation.

The G119S substitution in the gene encoding acetylcholinesterase 1 (Ace‐1) is widespread in mosquitoes and confers resistance to OPs and CMs. Weetman et al. (2015) identified a single Ace‐1 resistant haplotype, with reduced nucleotide diversity and high linkage disequilibrium compared to Ace‐1 wild‐type haplotypes, again providing strong evidence of a recent de novo origin rather than selection from standing genetic variation.

In contrast to the classical hard‐sweep scenarios described above, investigation of OP/CM resistance in Drosophila melanogaster has provided an example of adaptation from multiple de novo mutations producing a soft selective sweep (Karasov et al., 2010). Four different point mutations, I161V, G265A, F330Y and G368A, in the Ace gene lead to OP and CM resistance in D. melanogaster, and all resistant alleles are strongly deleterious in the absence of insecticide. By sequencing the Ace‐1 gene in a large number of resistant and susceptible fly strains, Karasov et al. (2010) provided evidence of multi‐step adaptation, with resistant haplotypes comprising three SNPs arising from de novo mutations just a few years after the introduction of OPs and CMs. Significantly, the authors found that Ace‐1 mutations arose repeatedly, even within the same continent, demonstrating a far higher rate of adaptive mutation in fruit flies than previously envisaged, and indicating that in some cases, adaptive potential from de novo mutations may be higher than that predicted from standing variation levels.

In other cases, the rapid emergence of insecticide resistance is thought to reflect evolution from standing genetic variation. Resistance to the new diamide insecticides evolved in the diamondback moth, Plutella xylostella, within just two years of use. Resistance was found to be associated with an amino acid substitution, G4946E, in the target site, the ryanodine receptor (RyR) (Troczka et al., 2012). Interestingly, the nucleotide changes that result in the G4946E substitution in two geographically separated resistant strains were different and susceptible strains also show synonymous polymorphism at this codon. The P. xylostella genome is known to be extremely polymorphic (You et al., 2013) and the sequence diversity observed in the RyR of resistant and susceptible strains, combined with the extremely rapid emergence of resistance, is suggestive of selection from standing variation.

Population‐genetic studies examining the evolution of insecticide resistance in D. melanogaster have also suggested that adaptation from standing genetic variation can occur in the case of metabolic resistance (Catania et al., 2004). Resistance to dichlorodiphenyltrichloroethane (DDT) and several other insecticide classes in D. melanogaster is conferred by enhanced expression of the P450 cyp6g1 resulting from the insertion of an Accord transposable element in the promotor of the gene encoding this enzyme. Investigation of the distribution of the Accord element in 673 D. melanogaster lines from 34 world‐wide populations revealed a much narrower selective sweep around the insertion than would be expected for a de novo mutation under strong selection, suggesting that this insertion may pre‐date the use of DDT (Catania et al., 2004).

Evolution of insecticide resistance from standing genetic variation and by de novo mutation are not necessarily mutually exclusive. Work on the sheep blowfly, Lucilia cuprina, provides an example of both origins in resistance to OPs (Hartley et al., 2006; Rose et al., 2011). Resistance to the OP insecticides diazinon and malathion in this species is conferred by two independent mutations in esterase E3 encoded by the Lcα7 gene. The G137N substitution confers resistance to diazinon while alterations at a second site, W251L/S/T, confers resistance to malathion. Analysis of Lcα7 in Australasian strains of L. cuprina, including pinned specimens collected prior to the introduction of OP insecticides, has suggested that two incomplete soft sweeps occurred at this locus. Several resistance mutations at position 251 were observed in pinned specimens, providing unambiguous evidence of genetic variation existing before the first use of OPs. Conversely, the G137N mutation was only identified in post‐OP samples, suggesting that diazinon resistance may have evolved later by de novo mutation. The different origins for two resistance mutations in the same gene is intriguing and is likely related to their relative fitness cost, with mutation of position 137 carrying a much greater fitness penalty than substitutions at 251 (Hartley et al., 2006). Indeed, the G137N substitution may only have risen to high frequency because it confers much greater resistance to the more widely used diazinon insecticide and because of the proliferation of a compensatory mutation at a separate Modifier locus (Hartley et al., 2006).

Beyond de novo mutation and standing genetic variation, recent work has highlighted a third potential evolutionary origin of insecticide resistance alleles, through adaptive introgression (Norris et al., 2015). The malaria mosquitoes An. gambiae and An. coluzzii are sympatric across much of sub‐Saharan Africa with hybridisation occurring at varying frequencies across the range of the two species, although hybrids typically suffer a fitness disadvantage and the species generally show assortive mating. This results in a degree of reproductive isolation, and the two species differ in their level of resistance to insecticides, with several insecticide‐resistance alleles in An. gambiae located within a genomic island of divergence on Chromosome 2. Norris et al. (2015) carried out a longitudinal population‐genetic study of the two mosquito species in Selinkenyi, Mali, and demonstrated that during a breakdown in assortive mating in 2006, An. coluzzi inherited the genomic island from An. gambiae, gaining several insecticide‐resistance alleles including the kdr mutation L1014F that was previously absent from An. coluzzii. Hybrid individuals carrying the resistant alleles were strongly selected for, and backcrossed with parental populations. The introgression event was coincident with the start of a major vector control initiative in Mali using insecticide‐treated bed nets, suggesting that a change in the fitness landscape allowed the genomic island of divergence to cross the reproductive barrier between the two species by favouring the survival of the normally less fit hybrids (Norris et al., 2015).

The molecular and biochemical mechanisms conferring resistance to herbicides have been well established, particularly for acetyl‐CoA carboxylase (ACCase), acetolactate synthase (ALS) and 5‐enolpyruvylshikimate‐3‐phosphate (EPSPS) inhibiting herbicides (Powles & Yu, 2010). For all three modes of action, target‐site and non‐target‐site mechanisms of resistance have been documented, with prevailing evidence suggesting that non‐target‐site metabolic resistance is the predominant mechanism. Target‐site and non‐target‐site mechanisms are not mutually exclusive, as both can be found within the same population, and even occur within the same individual (Yu & Powles, 2014). There have been relatively few studies that have employed population‐genetic and genomic techniques to address questions related to the evolutionary origins of the mutations underpinning these adaptations (Table ​1). This is, at least in part, due to the lack of development of genomic resources for major weed species, and, in the case of non‐target‐site resistance, to scant knowledge of the genetic basis of resistance, although this situation is now being addressed (Cummins et al., 2013; Gaines et al., 2014). The available evidence largely supports the phenomenon of multiple evolutionary origins of resistance (Délye et al., 2013), implying that the signature of soft selective sweeps would be detected for herbicide resistance.

For ACCase‐ (Kaundun, 2014) and ALS‐inhibiting (Yu & Powles, 2014) herbicides, multiple SNPs have been documented, resulting in a number of different amino acid substitutions that confer resistance to these herbicides, across and within species and individuals. The existence and global distribution of these mutations in a number of species also suggests multiple independent origins of resistance at a broad scale. Délye et al. (2004) used phylogenetic analysis of the ACCase sequence in nine populations of the grass weed, A. myosuroides to demonstrate that there had been four and six independent origins of the L1781I and N2041I substitutions, respectively, across nine populations of the species. Further ACCase sequencing over a larger set of A. myosuroides populations in France identified L1781, N2041 and A2096 haplotypes differing by up to eight, 11 and 10 mutations, respectively (Menchari et al., 2006), further supporting multiple independent origins. These studies did not explore signatures of selection around resistance‐conferring loci to infer whether mutations arose from standing genetic variation or by de novo mutation. However, Délye et al. (2013) reported the existence of the L1781 mutation in a herbarium sample of A. myosuroides collected in 1881, almost 100 years prior to the introduction of these herbicides. In Avena fatua, Mengistu et al. (2003) showed that ACCase‐resistant individuals were present in populations collected prior to herbicide use, and subsequent investigation of genetic variation in a series of resistant populations indicated multiple origins of resistance (Mengistu et al., 2005).

For non‐target‐site resistance to ACCase herbicides, evolution of resistance from standing genetic variation has been strongly implicated by selection experiments in the greenhouse (Neve & Powles, 2005a, b ) and in the field (Manalil et al., 2011). Neve & Powles (2005a) found mean phenotypic frequencies of resistance (survival at field recommended rates) of 0.4% across 31 populations of Lolium rigidum with no previous history of herbicide exposure. Combining these results with in vitro assays, they demonstrated that high‐frequency resistance within standing genetic variation in these populations was due to non‐target‐site resistance mechanisms.

For glyphosate resistance, a number of studies have coupled population‐genetics and phenotypic data sets from population samples collected prior to the introduction of glyphosate, suggesting that multiple independent evolutionary origins are due to standing genetic variation. Using simple‐sequence repeat (SSR) markers, Fernández et al. (2013) showed that glyphosate‐resistant Sorghum halepense populations have independently evolved glyphosate resistance. Okada et al. (2013) demonstrated that glyphosate resistance in Conyza canadensis populations across California has evolved via multiple independent evolutionary origins, with subsequent dispersal after resistance has evolved. Baucom & Mauricio (2010) grew Ipomoea purpurea from preserved seed collected in the 1980s, prior to the widespread introduction of glyphosate in the mid‐1990s, and compared levels of glyphosate sensitivity to a modern population. They found variation in glyphosate resistance in the historical as well as the modern population, indicating that glyphosate resistance was present in standing variation prior to herbicide selection. Kuester et al. (2015) analysed neutral genetic markers in I. purpurea and demonstrated that populations experienced gene flow pre‐glyphosate usage in the USA but glyphosate resistance then evolved independently in multiple populations, and concluded that this independent evolution of resistance was through selection from the standing variation demonstrated by Baucom & Mauricio (2010).

Herbicide‐resistance genes have also been reported to originate through processes of hybridisation and introgression between crops and related weed species, although there is little evidence that this is a major source of field‐evolved herbicide resistance. There has been much debate about the potential for transfer of resistance genes between genetically modified crops and closely related wild, weedy relatives through processes of hybridisation and subsequent introgression (see review by Kwit et al., 2011). Warwick et al. (2008) demonstrated that herbicide‐resistance genes from transgenic Brassica napus crops were introduced into weedy B. rapa populations and then maintained for six years in these weedy populations in the absence of herbicide selection and despite the fitness cost of hybridisation. Watrud et al. (2004) reported gene flow between genetically modified (GM) Agrostis stolonifera and weedy populations of the same species and Reichman et al. (2006) followed this report by demonstrating that individuals with the transgene had established in natural habitats. However, in most cases, herbicide‐resistance has evolved independently within weed species, due to the strong selection imposed by repeated use of the single herbicide to which the GM crop is resistant (Powles, 2008).

Although much work has been focussed on the hybridisation of genetically modified organisms (GMOs) and their wild relatives, allele transfer through hybridisation may also occur between conventionally bred herbicide‐resistant crops and wild relatives (Mallory‐Smith & Olguin, 2011), or between related weed species [see Gaines et al., 2012 and citations therein]. Martins et al. (2015) detected the imazimox‐resistant Imi1 allele from imazimox‐resistant wheat in wheat × jointed goatgrass (Aegilops cylindrica) hybrids and backcrossed plants. Gaines et al. (2012) demonstrated through field and greenhouse crosses that inter‐specific hybridisation can transfer glyphosate resistance, conferred by EPSPS gene amplification, from Amaranthus palmeri to other weedy Amaranthus species.

Most studies so far that have explicitly considered the evolutionary origins of fungicide‐resistant alleles have found evidence for de novo mutations as the main source of resistance (Table ​1).

Resistance to QoI fungicide is commonly conferred by the G143A substitution in mitochondrially encoded target site cytochrome b. Chen et al. (2007) analysed the cytochrome b haplotypes in the grapevine powdery mildew Plasmopara viticola, finding that G143A is present in two different haplotypic backgrounds, with the reconstructed haplotype network indicating parallel de novo mutations. Torriani et al. (2009) analysed cytochrome b haplotypes in the wheat pathogen Zymoseptoria tritici. The G143A substitution was present in 24 different haplotypes, with phylogenetic analysis indicating four independent de novo origins. Estep et al. (2015) subsequently investigated cytochrome b haplotypes in Z. tritici populations from North America, finding G143A present in several haplotypic backgrounds, with phylogenetic analysis supporting multiple parallel origins. However, they also report evidence of reduced mitochondrial genetic diversity among isolates with G143A, so the overall picture is of a selective sweep intermediate between a soft and hard sweep, following a small number of parallel de novo mutations.

Resistance to azole fungicides in Z. tritici is due to multiple mutations in the target‐site‐encoding gene CYP51. Earlier isolates had only one or few mutations, conferring low levels of resistance, but more recent isolates have accumulated more mutations, building up higher levels of resistance (Cools et al., 2006; Cools & Fraaije, 2013). The sequential accumulation of CYP51 mutations indicates de novo origins, with more recent mutations appearing in genetic backgrounds already including previously selected mutations. For example, the recently emerged mutation S524T, conferring reduced sensitivity to the recently introduced fungicide prothioconazole, is not found as a single mutation in a wild‐type background, only in combination with other mutations such as substitutions or deletions at codons 459–461. Site‐directed mutagenesis studies have shown that Z. tritici CYP51 with only the S524T substitution in a wild‐type background would be adaptive under prothioconazole selection, conferring reduced sensitivity to prothioconazole whilst retaining enzyme function (Cools et al., 2011); therefore its occurrence only in combination with other mutations and not in a wild‐type background does not reflect functional constraints, but is a result of its de novo origin after other mutations had been selected to higher frequencies in the population. However, Z. tritici populations continue to contain a mixture of many different CYP51 haplotypes (Curvers et al., 2014; Wieczorek et al., 2015), with no single allele being selected to fixation, and substitutions such as I381V, V136A and S524T appear to have arisen multiple times through parallel or convergent evolution in different backgrounds. Thus Z. tritici CYP51 is likely only to have undergone soft selective sweeps despite the de novo mutations.

In the grapevine powdery mildew Uncinula necator, azole resistance is conferred by CYP51 target site mutations, or CYP51 overexpression due to a promoter insert. All CYP51 overexpressing isolates sequenced so far carry a linked synonymous mutation in the coding region (Frenkel et al., 2015), indicating a hard selective sweep following a de novo mutation in a single genetic background. By contrast, in the oilseed rape pathogen Pyrenopeziza brassicae, CYP51 over‐expression is due to one of several promoter inserts of different lengths (Carter et al., 2014), suggesting that CYP51 over‐expression has multiple origins, which would result in a softer selective sweep whether from standing variation or multiple de novo origins. In the opportunistic clinical pathogen Aspergillus fumigatus, a 34‐base pair tandem repeat in the promoter region is found in combination with the L98H mutation (Mellado et al., 2007), whereas a 46‐base pair tandem repeat is found in combination with Y121F/T298A, and isolates with TR34/L98H also have reduced genetic diversity overall as measured by microsatellites and cell surface protein types (Camps et al., 2012), indicating a single de novo origin for each.

Fernández‐Ortuño et al. (2014) investigated multi‐fungicide‐resistant strains of the soft fruit pathogen Botrytis cinerea. They sequenced five target‐site‐encoding genes and an efflux transporter transcription factor for isolates resistant to seven different classes of fungicides. No sequenced resistant isolates had identical genotypes across the six genes, leading the authors to conclude multiple independent origins of multi‐fungicide‐resistance. However, it is not clear whether the multiple resistance originates from independent parallel mutations, or independent recombination events between single‐fungicide‐resistant strains, and therefore it is not possible to infer whether those mutations occurred de novo, or whether at least the single‐resistance mutations were selected from standing variation.

One case of fungicide resistance selected from standing variation was reported by Hawkins et al. (2014b). In the barley pathogen Rhynchosporium commune, isolates possessing CYP51A, an additional paralogue of the target‐site‐encoding gene CYP51, show reduced azole sensitivity and have increased in frequency under azole selection since the 1980s, but phylogenetic analysis shows that CYP51A diverged from the other CYP51 paralogue, CYP51B, around 400 million years ago. In other species with CYP51A, its presence appears to be fixed, conferring intrinsic partial resistance to azoles (Fan et al., 2013), and possibly conferring some pre‐disposition for the evolution of further resistance, as mutations and overexpression occur predominantly in that paralogue where present (e.g. Mellado et al., 2007), perhaps because the second paralogue is less affected by functional constraints than the constitutively expressed ‘workhorse’ CYP51B (Hawkins et al., 2014a). CYP51A expression is induced by azole exposure (Fan et al., 2013; Hawkins et al., 2014b), so paralogue presence or mutations could be seen as cryptic variation, with phenocopy induction by fungicides, although its as yet unidentified pre‐azole function is likely to involve induced expression under certain other conditions too.

There are currently no known cases of fungicide resistance originating from adaptive introgression or horizontal transfer. However, evidence of introgression has been found in fungal genomes (Neafsey et al., 2010), and adaptive introgression has played a role in fungal adaptation to anthropogenic environments such as the domestication of wine yeasts (Marsit et al., 2015). Evidence of HGT has also been found in fungal genomes (Cheeseman et al., 2014), and it has played a role in the evolution of a plant pathogenic fungus, Pyrenophora tritici‐repentis, which emerged in the mid‐20th century after acquiring the ToxA toxin‐encoding gene from Stagonospora nodorum (Friesen et al., 2006), with further studies inding that the P. tritici‐repentis genome also contains genes transferred from bacteria (Sun et al., 2013). Therefore, adaptive introgression and HGT are potential sources of fungal adaptation.

From studies reported to date, the emerging picture is that herbicide resistance evolves largely by selection from standing genetic variation, especially in the case of metabolic resistance with fewer cases of target‐site resistance selected from standing variation. Insecticide resistance has evolved variously through selection from standing variation and de novo mutations, with both metabolic and target‐site resistance mechanisms commonly occurring. Fungicide resistance has evolved predominantly through de novo mutations, with mostly target‐site resistance.

Various ad hoc reasons may be suggested to explain these differences: perhaps plants have greater standing variation in order to metabolise other toxins; perhaps pathogens have higher de novo mutation rates to engage in host–pathogen arms races. However, individual population‐genetic parameters such as mutation rate or genetic diversity should be considered in the context of a broader evolutionary model of when resistance will evolve from standing variation, de novo mutations, a combination of both, or not at all.

Our proposed model schematic is shown in Fig. ​3. Where resistance is present in standing variation, those variants will be selected, since they are already present, usually in more than one individual, and can therefore emerge more rapidly than de novo mutations (see Section II.2). If the initial population lacks resistant alleles, then de novo mutations will be selected when and if they arise. However, if standing variation confers only partial resistance relative to field doses, those variants will be selected initially, but subsequent de novo mutations that confer higher resistance will then be selected. A similar scenario would be possible if the initially selected standing variants carried higher fitness penalties than potential de novo mutations, although this would make them less likely to persist prior to pesticide selection unless the fitness cost was recessive (and unlikely to be homozygous at pre‐selection frequencies) or conditional (such as an inducibly expressed detoxification pathway). The overall risk of resistance evolving reflects the combined risk from standing variation and de novo mutations.

Therefore, the outcome will depend on levels of standing variation, rates of de novo mutation, and the relative resistance and fitness conferred by each at field doses of the pesticide. The effect of these three key factors is summarised in Fig. ​3, and the pathogen‐ and pesticide‐related parameters affecting the outcomes in different pesticide classes are discussed below.

Observed patterns of evolution imply that levels of standing variation are highest in weeds, then insect pests, then fungal pathogens.

A model for evolution from standing variation proposed by Hermisson & Pennings (2005) comprises terms relating to effective population size, mutation rate, the fitness coefficients before and after the change in selection, and the dominance of those fitness effects. Mutation rate, initial fitness and effective population size determine variation levels at mutation–selection–drift balance. Mutation rate refers to relevant mutations, and will therefore be affected by the likelihood and complexity of the resistance mechanism in question as well as overall mutation rate.

Experimental evolutionary studies have shown that selection from standing variation can drive adaptations in sexually outcrossing species, where recombination contributes to genetic variation, forming new combinations of existing alleles (Burke et al., 2014). Therefore, higher standing variation might be expected in outcrossing weed species, compared to many polycyclic fungal pathogens and some insects such as aphids, which undergo repeated clonal reproduction during a crop growing season.

Alternatively, weeds may maintain higher levels of standing variation due to lower rates of loss of diversity. Standing variation reflects mutation–selection–drift balance; reduced mutation rates would be expected to reduce de novo mutation, so a species with less standing variation but high rates of de novo mutation may be experiencing stronger purifying selection. This could reflect more stringent selection overall, or higher fitness costs of resistance mutations, but there is little evidence of either at present.

Loss of diversity can also result from demographic factors. Plant pathogens and phytophagous insects have undergone past population bottlenecks during host jumps (Groman & Pellmyr, 2000; Messina, Mendenhall & Jones, 2009); whereas weeds, which act as competitors rather than parasites to crop plants (with very few exceptions), are not host dependent or host specialised to the same extent. However, the high and stable host populations provided by agricultural crops mean that once a pest or pathogen has colonised a crop species, they will experience fewer bottlenecks than relatives on wild plants (Stukenbrock et al., 2007), and there is more evidence of bottlenecks from geographic range shifts (Munkacsi, Stoxen & May, 2008; Linde et al., 2009), which would be expected to affect all pest groups.

Population‐genetic factors affect the overall level of standing genetic variation in a population, but other factors may specifically affect variation conferring resistance. Plants may have more relevant standing variation, especially for metabolic resistance, due to a wider pre‐existing metabolic range. This is often explained on the grounds that plants are non‐motile, and therefore must rely on detoxification rather than avoidance. However, fungal pathogens, with the lowest propensity to evolve pesticide resistance from standing variation in metabolic detoxification, are also sessile; and whilst insects are capable of chemotaxis and host choice, they have evolved metabolic adaptations to defence compounds produced by their own host plant. Fungal pathogens cannot move to find a suitable host, instead relying on the production of large numbers of spores so that some will land on the right plant. Therefore, the contrast between plant weeds and fungal pathogens is due not to differences in mobility, but perhaps to the types of eco‐evolutionary differences sometimes simplified to r/K selection (Reznick, Bryant & Bashey, 2002) or competitor–stress tolerator–ruderal (CSR) strategies (Grime, 1988). Whilst weed and pest species tend to be r‐strategists compared to other species within their taxa, comparisons among taxa would suggest that micro‐organisms are more extreme r‐strategists than macro‐flora. Therefore, plants may have evolved a higher degree of phenotypic plasticity so that an individual can cope in a range of surroundings, while pathogens rely more on the extreme over‐production of propagules for dispersal so that some will fall in favourable conditions – a strategy which may also favour higher rates of de novo mutation. This may be because pathogens must repeatedly colonise new host plants and new growth of host plants, such as new leaf layers, whereas weed seed banks may accumulate leading to a more competitive situation.

The main parameters in models of evolution by de novo mutation are mutation rate, and selection coefficient. Whilst the generation of standing variation also depends on the rate of relevant mutations, this is as a component of longer‐term mutation–selection–drift balance such that a lower relevant mutational supply could be balanced out by lower rates of variant loss, whereas for rapid evolution by de novo mutations, a high enough rate of relevant mutation is crucial.

Fungal pathogens generally have shorter generation times than weeds, with most species going through multiple generations per growing season. Therefore, even with a similar mutation rate per generation, they would have a higher mutation rate per unit time. Furthermore, pathogens tend to have larger populations given their smaller size, so mutation rates per individual would be multiplied by a greater population size to give a higher total mutational supply (Messer & Petrov, 2013).

The emergence of mutations may be faster in haploid organisms, including most plant pathogens, since a single mutation is subject to selection immediately; in diploid or polyploid organisms, including weeds, dominant alleles are subject to selection immediately, but recessive alleles must first reach homozygosity without being lost through genetic drift (Orr & Betancourt, 2001). Rust fungi are diploid, and have been considered to be at low risk of evolving fungicide resistance, although their low resistance risk is partly due to mutational constraints such as intron splice sites (Oliver, 2014), and some cases of fungicide resistance due to target‐site mutations have now been reported in rust fungi (Schmitz et al., 2014).

The influence of ploidy on resistance evolution also needs to be considered for several economically important arthropods such as whiteflies, thrips and spider‐mites. These species are haplodiploid and this is predicted to accelerate resistance development as a consequence of novel mutations being exposed directly to selection, irrespective of dominance, in hemizygous males. However, this prediction is based on the assumption that the resistance of R males is equivalent to RR females, which is valid for some species and insecticides but not all (Carrière, 2003).

The potential for a de novo mutation from a single origin to spread through a population also depends on migration rates (McDonald & Linde, 2002). Fungal ascospores can carry resistance alleles long distances (Fraaije et al., 2005; Torriani et al., 2009). Insecticide resistance can also spread through insect migration (Raymond et al., 1991), although in other cases resistance has been selected independently in geographically isolated populations (Troczka et al., 2012). A study of gene flow and herbicide resistance in the weed Ipomoae purpura concluded that historical gene flow had led to shared standing variation, but lower gene flow within the timeframe of glyphosate use meant that resistance had evolved independently in local populations (Kuester et al., 2015).

The selection of resistance depends not just on the presence or absence of standing variation or de novo mutations, but on the levels of resistance conferred by each variant, relative to the selecting dose of the pesticide. When the dose is sufficiently low for standing variants to confer resistance, those variants will be selected; when the dose is beyond the sensitivity range of the standing variation, then more highly resistant de novo mutations will be selected. If doses are lower than the resistance possible from small‐effect alleles present as standing variation, those alleles will be selected rapidly before any de novo mutations emerge, and there will then be no further selection for new mutations unless the dose rate is subsequently increased beyond the range of polygenic resistance. If dose rates are higher, small‐effect alleles offer little selective advantage as the pesticide will still be lethal to individuals carrying such alleles, but once highly resistant de novo mutations occur, those will be selected.

The importance of the nature of selection, as well as the population genetics and overall mutation rate of the target organism, is demonstrated by cases of evolution from standing variation and de novo mutations occurring within the same species. In the clinical bacterial pathogen Streptococcus pneumoniae, drug resistance evolved from de novo mutations, but standing variation in serotype led to vaccine escape (Croucher et al., 2014). In Drosophila melanogaster, target‐site resistance to OPs resulted from multiple de novo mutations (Karasov et al., 2010), whereas metabolic resistance to DDT was selected from standing variation (Catania et al., 2004).

As discussed in Section II.2, de novo mutations, starting from a lower frequency than standing variants, need to be under stronger positive selection to be likely to emerge: this means highly resistant alleles are more likely than partial‐resistance alleles to emerge and persist following de novo mutations. Conversely, metabolic resistance may be more likely than target‐site resistance to be present as standing variation: pesticides target essential genes, which may be more conserved due to functional constraints than secondary metabolism and detoxification pathways which may be more variable to allow adaptation to various naturally occurring toxins.

This explanation would be testable by analysing baseline sensitivity distributions for herbicides, fungicides and insecticides, with the prediction that whilst standing variation may be present across all groups, the highest levels of fungicide resistance would all lie well below field dose rates, whereas the most herbicide‐resistant plants would be at levels closer to field resistance. However, this would require genuinely unexposed populations of a range of species. Alternatively, experimental evolution approaches could be used, using both a mixed starting population and mutagenesis if required, and comparing the results of selection with high and low doses of a pesticide.

The different resistance levels likely to persist as standing variation or to emerge de novo have practical implications for the optimal pesticide dose rates for resistance management. The long‐standing advice to use full dose rates applies in cases where standing variation includes individuals that would be selected by lower doses but controlled at a full dose. However, where the greater risk is highly resistant de novo mutations, higher doses will exert stronger selective pressure for the emergence of such mutations and therefore lower doses would slow the evolution of resistance (Van Den Bosch et al., 2011). Where both types of resistance evolution are possible, a ‘revolving dose strategy’ alternating low and high doses may be optimal (Gardner, Gressel & Mangel, 1998).

In some cases, the model predicts a series of resistance mechanisms evolving over time. When a population is exposed to a pesticide, initially selection from standing variation would result in rapid but smaller shifts in sensitivity, followed by the selection of more resistant de novo mutations as they arise. A classic example of multiple adaptive steps occurring in rapid succession is the evolution of insecticide resistance in D. melanogaster mediated by the overproduction of the P450 Cyp6g1. This example of allelic succession at the Cyp6G1 locus has occurred over a 70‐year period with each step resulting in increasing resistance (Schmidt et al., 2010). However, sequential increases in resistance could also result from recombination of multiple quantitative resistance alleles; or from the accumulation of multiple de novo mutations, as seen with CYP51 mutations conferring azole resistance in Zymoseptoria tritici (Cools & Fraaije, 2013). Furthermore, where de novo mutation rates are very high, the waiting time may be negligible, with selection operating on standing and de novo mutations simultaneously, selecting whichever has the optimal balance of resistance to field doses and low fitness penalties. More research is needed to test the model proposed here: quantifying the relevant parameters (diversity levels, mutation rates, baseline sensitivity distributions, temporal processes in resistance evolution) in field populations, defining mutant‐selective windows for pest species in the laboratory, and testing various scenarios through experimental evolution. More explicit studies of the evolutionary origins of resistance are also needed across the three groups to increase the sample size for each and see whether the pattern of standing variation in weeds, mixed origins in insects, and de novo mutation in fungi actually holds across a greater number of species and modes of action. Further analysis of differences within, as well as between, the three groups may also be informative; for example, is polygenic resistance selected from standing variation more common in outcrossing insects, and de novo emergence of major metabolic or target‐site resistance more common in insects with clonal reproduction?

• Abou Ammar, G. , Tryono, R. , Döll, K. , Karlovsky, P. , Deising, H. B. & Wirsel, S. G. R. (2013). Identification of ABC transporter genes of Fusarium graminearum with roles in azole tolerance and/or virulence. PLoS One 8, e79042. [PMC free article] [PubMed] [Google Scholar]
• Alexander, H. K. , Martin, G. , Martin, O. Y. & Bonhoeffer, S. (2014). Evolutionary rescue: linking theory for conservation and medicine. Evolutionary Applications 7, 1161–1179. [PMC free article] [PubMed] [Google Scholar]
• Barrett, R. D. H. & Schluter, D. (2008). Adaptation from standing genetic variation. Trends in Ecology & Evolution 23, 38–44. [PubMed] [Google Scholar]
• Bass, C. , Denholm, I. , Williamson, M. S. & Nauen, R. (2015). The global status of insect resistance to neonicotinoid insecticides. Pesticide Biochemistry and Physiology 121, 78–87. [PubMed] [Google Scholar]
• Bass, C. , Zimmer, C. T. , Riveron, J. M. , Wilding, C. S. , Wondji, C. S. , Kaussmann, M. , Field, L. M. , Williamson, M. S. & Nauen, R. (2013). Gene amplification and microsatellite polymorphism underlie a recent insect host shift. Proceedings of the National Academy of Sciences of the United States of America 110, 19460–19465. [PMC free article] [PubMed] [Google Scholar]
• Baucom, R. S. & Mauricio, R. (2010). Defense to the herbicide roundup predates its wide spread use. Evolutionary Ecology Research 12, 131–141. [Google Scholar]
• Bearchell, S. J. , Fraaije, B. A. , Shaw, M. W. & Fitt, B. D. L. (2005). Wheat archive links long‐term fungal pathogen population dynamics to air pollution. Proceedings of the National Academy of Sciences of the United States of America 102, 5438–5442. [PMC free article] [PubMed] [Google Scholar]
• Bedhomme, S. , Lafforgue, G. & Elena, S. (2013). Genotypic but not phenotypic historical contingency revealed by viral experimental evolution. BMC Evolutionary Biology 13, 46. [PMC free article] [PubMed] [Google Scholar]
• Berg, J. J. & Coop, G. (2014). A population genetic signal of polygenic adaptation. PLoS Genetics 10, e1004412. [PMC free article] [PubMed] [Google Scholar]
• Beseli, A. , Amnuaykanjanasin, A. , Herrero, S. , Thomas, E. & Daub, M. (2015). Membrane transporters in self resistance of Cercospora nicotianae to the photoactivated toxin cercosporin. Current Genetics 61, 601–620. [PubMed] [Google Scholar]
• Burke, M. K. , Liti, G. & Long, A. D. (2014). Standing genetic variation drives repeatable experimental evolution in outcrossing populations of Saccharomyces cerevisiae . Molecular Biology and Evolution 31, 3228–3239. [PMC free article] [PubMed] [Google Scholar]
• Camps, S. M. T. , Rijs, A. J. M. M. , Klaassen, C. H. W. , Meis, J. F. , O'Gorman, C. M. , Dyer, P. S. , Melchers, W. J. G. & Verweij, P. E. (2012). Molecular epidemiology of Aspergillus fumigatus isolates harboring the TR34/L98H azole resistance mechanism. Journal of Clinical Microbiology 50, 2674–2680. [PMC free article] [PubMed] [Google Scholar]
• Carolan, K. , Helps, J. , Van Den Berg, F. , Bain, R. , Paveley, N. & van den Bosch, F. (2017). Extending the durability of cultivar resistance by limiting epidemic growth rates. Proceedings of the Royal Society of London B: Biological Sciences 284, 20170828. [PMC free article] [PubMed] [Google Scholar]
• Carrière, Y. (2003). Haplodiploidy, sex, and the evolution of pesticide resistance. Journal of Economic Entomology 96, 1626–1640. [PubMed] [Google Scholar]
• Carter, H. E. , Fraaije, B. A. , West, J. S. , Kelly, S. L. , Mehl, A. , Shaw, M. W. & Cools, H. J. (2014). Alterations in the predicted regulatory and coding regions of the sterol 14 alpha‐demethylase gene (CYP51) confer decreased azole sensitivity in the oilseed rape pathogen Pyrenopeziza brassicae . Molecular Plant Pathology 15, 513–522. [PMC free article] [PubMed] [Google Scholar]
• Catania, F. , Kauer, M. O. , Daborn, P. J. , Yen, J. L. , Ffrench‐Constant, R. H. & Schlotterer, C. (2004). World‐wide survey of an accord insertion and its association with DDT resistance in Drosophila melanogaster . Molecular Ecology 13, 2491–2504. [PubMed] [Google Scholar]
• Charlesworth, D. , Barton, N. H. & Charlesworth, B. (2017). The sources of adaptive variation. Proceedings of the Royal Society of London B: Biological Sciences 284, 20162864. [Google Scholar]
• Cheeseman, K. , Ropars, J. , Renault, P. , Dupont, J. , Gouzy, J. , Branca, A. , Abraham, A.‐L. , Ceppi, M. , Conseiller, E. , Debuchy, R. , Malagnac, F. , Goarin, A. , Silar, P. , Lacoste, S. , Sallet, E. , et al. (2014). Multiple recent horizontal transfers of a large genomic region in cheese making fungi. Nature Communications 5, 2876. [PMC free article] [PubMed] [Google Scholar]
• Chen, F. , Everhart, S. E. , Bryson, P. K. , Luo, C. , Song, X. , Liu, X. & Schnabel, G. (2015). Fungicide‐induced transposon movement in Monilinia fructicola . Fungal Genetics and Biology 85, 38–44. [PubMed] [Google Scholar]
• Chen, W. J. , Delmotte, F. , Richard‐Cervera, S. , Douence, L. , Greif, C. & Corio‐Costet, M. F. (2007). At least two origins of fungicide resistance in grapevine downy mildew populations. Applied and Environmental Microbiology 73, 5162–5172. [PMC free article] [PubMed] [Google Scholar]
• Colosimo, P. F. , Hosemann, K. E. , Balabhadra, S. , Villarreal, G. , Dickson, M. , Grimwood, J. , Schmutz, J. , Myers, R. M. , Schluter, D. & Kingsley, D. M. (2005). Widespread parallel evolution in sticklebacks by repeated fixation of Ectodysplasin alleles. Science 307, 1928–1933. [PubMed] [Google Scholar]
• Cools, H. J. & Fraaije, B. A. (2013). Update on mechanisms of azole resistance in Mycosphaerella graminicola and implications for future control. Pest Management Science 69, 150–155. [PubMed] [Google Scholar]
• Cools, H. J. , Fraaije, B. A. , Kim, S. H. & Lucas, J. A. (2006). Impact of changes in the target P450 CYP51 enzyme associated with altered triazole‐sensitivity in fungal pathogens of cereal crops. Biochemical Society Transactions 34, 1219–1222. [PubMed] [Google Scholar]
• Cools, H. J. , Hawkins, N. J. & Fraaije, B. A. (2013). Constraints on the evolution of azole resistance in plant pathogenic fungi. Plant Pathology 62, 36–42. [Google Scholar]
• Cools, H. J. , Mullins, J. G. L. , Fraaije, B. A. , Parker, J. E. , Kelly, D. E. , Lucas, J. A. & Kelly, S. L. (2011). Impact of recently emerged sterol 14α‐demethylase (CYP51) variants of Mycosphaerella graminicola on azole fungicide sensitivity. Applied and Environmental Microbiology 77, 3830–3837. [PMC free article] [PubMed] [Google Scholar]
• Cools, H. J. , Parker, J. E. , Kelly, D. E. , Lucas, J. A. , Fraaije, B. A. & Kelly, S. L. (2010). Heterologous expression of mutated eburicol 14 alpha‐demethylase (CYP51) proteins of Mycosphaerella graminicola to assess effects on azole fungicide sensitivity and intrinsic protein function. Applied and Environmental Microbiology 76, 2866–2872. [PMC free article] [PubMed] [Google Scholar]
• Croucher, N. J. , Chewapreecha, C. , Hanage, W. P. , Harris, S. R. , McGee, L. , Van Der Linden, M. , Song, J.‐H. , Ko, K. S. , De Lencastre, H. , Turner, C. , Yang, F. , Sá‐Leão, R. , Beall, B. , Klugman, K. P. , Parkhill, J. , et al. (2014). Evidence for soft selective sweeps in the evolution of Pneumococcal multidrug resistance and vaccine escape. Genome Biology and Evolution 6, 1589–1602. [PMC free article] [PubMed] [Google Scholar]
• Cummins, I. , Wortley, D. J. , Sabbadin, F. , He, Z. , Coxon, C. R. , Straker, H. E. , Sellars, J. D. , Knight, K. , Edwards, L. , Hughes, D. , Kaundun, S. S. , Hutchings, S.‐J. , Steel, P. G. & Edwards, R. (2013). Key role for a glutathione transferase in multiple‐herbicide resistance in grass weeds. Proceedings of the National Academy of Sciences of the United States of America 110, 5812–5817. [PMC free article] [PubMed] [Google Scholar]
• Curvers, K. , Pycke, B. , Kyndt, T. , Haesaert, G. & Gheysen, G. (2014). Sensitivity towards DMI fungicides and haplotypic diversity of their CYP51 target in the Mycosphaerella graminicola population of Flanders. Journal of Plant Diseases and Protection 121, 156–163. [Google Scholar]
• D'Costa, V. M. , King, C. E. , Kalan, L. , Morar, M. , Sung, W. W. L. , Schwarz, C. , Froese, D. , Zazula, G. , Calmels, F. , Debruyne, R. , Golding, G. B. , Poinar, H. N. & Wright, G. D. (2011). Antibiotic resistance is ancient. Nature 477, 457–461. [PubMed] [Google Scholar]
• Délye, C. , Deulvot, C. & Chauvel, B. (2013). DNA analysis of herbarium specimens of the grass weed alopecurus myosuroides reveals herbicide resistance pre‐dated herbicides. PLoS One 8, e75117. [PMC free article] [PubMed] [Google Scholar]
• Délye, C. , Straub, C. , Matjicek, A. & Michel, S. (2004). Multiple origins for black‐grass (Alopecurus myosuroides Huds) target‐site‐based resistance to herbicides inhibiting acetyl‐CoA carboxylase. Pest Management Science 60, 35–41. [PubMed] [Google Scholar]
• Di Rago, J. P. , Coppee, J. Y. & Colson, A. M. (1989). Molecular basis for resistance to myxothiazol, mucidin (strobilurin A), and stigmatellin. Journal of Biological Chemistry 264, 14543–14548. [PubMed] [Google Scholar]
• Dooley, H. , Shaw, M. W. , Mehenni‐Ciz, J. , Spink, J. & Kildea, S. (2016). Detection of Zymoseptoria tritici SDHI‐insensitive field isolates carrying the SdhC‐H152R and SdhD‐R47W substitutions. Pest Management Science 72, 2203–2207. [PubMed] [Google Scholar]
• Espeby, L. A. , Fogelfors, H. & Milberg, P. (2011). Susceptibility variation to new and established herbicides: examples of inter‐population sensitivity of grass weeds. Crop Protection 30, 429–435. [Google Scholar]
• Estep, L. K. , Torriani, S. F. F. , Zala, M. , Anderson, N. P. , Flowers, M. D. , Mcdonald, B. A. , Mundt, C. C. & Brunner, P. C. (2015). Emergence and early evolution of fungicide resistance in North American populations of Zymoseptoria tritici . Plant Pathology 64, 961–971. [Google Scholar]
• Fabritius, A. L. , Shattock, R. C. & Judelson, H. S. (1997). Genetic analysis of metalaxyl insensitivity loci in Phytophthora infestans using linked DNA markers. Phytopathology 87, 1034–1040. [PubMed] [Google Scholar]
• Fan, J. , Urban, M. , Parker, J. E. , Brewer, H. C. , Kelly, S. L. , Hammond‐Kosack, K. E. , Fraaije, B. A. , Liu, X. & Cools, H. J. (2013). Characterization of the sterol 14α‐demethylases of Fusarium graminearum identifies a novel genus‐specific CYP51 function. New Phytologist 198, 821–835. [PubMed] [Google Scholar]
• Fernández, L. , De Haro, L. A. , Distefano, A. J. , Carolina Martínez, M. , Lía, V. , Papa, J. C. , Olea, I. , Tosto, D. & Esteban Hopp, H. (2013). Population genetics structure of glyphosate‐resistant Johnsongrass (Sorghum halepense L. Pers) does not support a single origin of the resistance. Ecology and Evolution 3, 3388–3400. [PMC free article] [PubMed] [Google Scholar]
• Fernández‐Ortuño, D. , Grabke, A. , Li, X. & Schnabel, G. (2014). Independent emergence of resistance to seven chemical classes of fungicides in Botrytis cinerea . Phytopathology 105, 424–432. [PubMed] [Google Scholar]
• Ferrer‐Admetlla, A. , Liang, M. , Korneliussen, T. & Nielsen, R. (2014). On detecting incomplete soft or hard selective sweeps using haplotype structure. Molecular Biology and Evolution 31, 1275–1291. [PMC free article] [PubMed] [Google Scholar]
• Forsberg, K. J. , Reyes, A. , Wang, B. , Selleck, E. M. , Sommer, M. O. A. & Dantas, G. (2012). The shared antibiotic resistome of soil bacteria and human pathogens. Science 337, 1107–1111. [PMC free article] [PubMed] [Google Scholar]
• Fraaije, B. A. , Bayon, C. , Atkins, S. , Cools, H. J. , Lucas, J. A. & Fraaije, M. W. (2012). Risk assessment studies on succinate dehydrogenase inhibitors, the new weapons in the battle to control Septoria leaf blotch in wheat. Molecular Plant Pathology 13, 263–275. [PMC free article] [PubMed] [Google Scholar]
• Fraaije, B. A. , Cools, H. J. , Fountaine, J. , Lovell, D. J. , Motteram, J. , West, J. S. & Lucas, J. A. (2005). Role of ascospores in further spread of QoI‐resistant cytochrome b alleles (G143A) in field populations of Mycosphaerella graminicola . Phytopathology 95, 933–941. [PubMed] [Google Scholar]
• Frenkel, O. , Cadle‐Davidson, L. , Wilcox, W. F. & Milgroom, M. G. (2015). Mechanisms of resistance to an azole fungicide in the grapevine powdery mildew fungus, Erysiphe necator . Phytopathology 105, 370–377. [PubMed] [Google Scholar]
• Friesen, T. L. , Stukenbrock, E. H. , Liu, Z. H. , Meinhardt, S. , Ling, H. , Faris, J. D. , Rasmussen, J. B. , Solomon, P. S. , Mcdonald, B. A. & Oliver, R. P. (2006). Emergence of a new disease as a result of interspecific virulence gene transfer. Nature Genetics 38, 953–956. [PubMed] [Google Scholar]
• Gaines, T. A. , Lorentz, L. , Figge, A. , Herrmann, J. , Maiwald, F. , Ott, M.‐C. , Han, H. , Busi, R. , Yu, Q. , Powles, S. B. & Beffa, R. (2014). RNA‐seq transcriptome analysis to identify genes involved in metabolism‐based diclofop resistance in Lolium rigidum . The Plant Journal 78, 865–876. [PubMed] [Google Scholar]
• Gaines, T. A. , Ward, S. M. , Bukun, B. , Preston, C. , Leach, J. E. & Westra, P. (2012). Interspecific hybridization transfers a previously unknown glyphosate resistance mechanism in Amaranthus species. Evolutionary Applications 5, 29–38. [PMC free article] [PubMed] [Google Scholar]
• Garcia‐Leon, G. , Hernandez, A. , Hernando‐Amado, S. , Alavi, P. , Berg, G. & Martineza, J. L. (2014). A function of SmeDEF, the major quinolone resistance determinant of Stenotrophomonas maltophilia, is the colonization of plant roots. Applied and Environmental Microbiology 80, 4559–4565. [PMC free article] [PubMed] [Google Scholar]
• Gardner, S. N. , Gressel, J. & Mangel, M. (1998). A revolving dose strategy to delay the evolution of both quantitative vs major monogene resistances to pesticides and drugs. International Journal of Pest Management 44, 161–180. [Google Scholar]
• Gibson, G. & Dworkin, I. (2004). Uncovering cryptic genetic variation. Nature Reviews. Genetics 5, 681–690. [PubMed] [Google Scholar]
• Gisi, U. (2014). Assessment of selection and resistance risk for demethylation inhibitor fungicides in Aspergillus fumigatus in agriculture and medicine: a critical review. Pest Management Science 70, 352–364. [PubMed] [Google Scholar]
• Godfray, H. C. J. , Beddington, J. R. , Crute, I. R. , Haddad, L. , Lawrence, D. , Muir, J. F. , Pretty, J. , Robinson, S. , Thomas, S. M. & Toulmin, C. (2010). Food security: the challenge of feeding 9 billion people. Science 327, 812–818. [PubMed] [Google Scholar]
• Grasso, V. , Palermo, S. , Sierotzki, H. , Garibaldi, A. & Gisi, U. (2006). Cytochrome b gene structure and consequences tor resistance to Qo inhibitor fungicides in plant pathogens. Pest Management Science 62, 465–472. [PubMed] [Google Scholar]
• Gressel, J. (2011). Low pesticide rates may hasten the evolution of resistance by increasing mutation frequencies. Pest Management Science 67, 253–257. [PubMed] [Google Scholar]
• Grime, J. P. (1988). The C‐S‐R model of primary plant strategies‐ origins implications and tests In Plant Evolutionary Biology (eds Gottlieb L. D. and Jain S. K.), pp. 371–394. Routledge, Chapman and Hall, New York. [Google Scholar]
• Groman, J. D. & Pellmyr, O. (2000). Rapid evolution and specialization following host colonization in a Yucca moth. Journal of Evolutionary Biology 13, 223–236. [Google Scholar]
• Gutierrez, A. , Laureti, L. , Crussard, S. , Abida, H. , Rodriguez‐Rojas, A. , Blazquez, J. , Baharoglu, Z. , Mazel, D. , Darfeuille, F. , Vogel, J. & Matic, I. (2013). β‐lactam antibiotics promote bacterial mutagenesis via an RpoS‐mediated reduction in replication fidelity. Nature Communications 4, 1610. [PMC free article] [PubMed] [Google Scholar]
• Gutiérrez‐Alonso, O. , Hawkins, N. J. , Cools, H. J. , Shaw, M. W. & Fraaije, B. A. (2017). Dose‐dependent selection drives lineage replacement during the experimental evolution of SDHI fungicide resistance in Zymoseptoria tritici. Evolutionary Applications 10, 1055–1066. [PMC free article] [PubMed] [Google Scholar]
• Hartley, C. J. , Newcomb, R. D. , Russell, R. J. , Yong, C. G. , Stevens, J. R. , Yeates, D. K. , La Salle, J. & Oakeshott, J. G. (2006). Amplification of DNA from preserved specimens shows blowflies were preadapted for the rapid evolution of insecticide resistance. Proceedings of the National Academy of Sciences of the United States of America 103, 8757–8762. [PMC free article] [PubMed] [Google Scholar]
• Hawkins, N. J. , Cools, H. , Fan, J. & Fraaije, B. (2014a). The role of multiple CYP51 paralogues in intrinsic and acquired azole sensitivity differences In Modern Fungicides and Antifungal Compounds VII: Proceedings of the 17th International Reinhardsbrunn Symposium (eds Dehne H. W., Deising H. B., Fraaije B., Gisi U., Hermann D., Mehl A., Oerke E. C., Russell P. E., Stammler G., Kuck K. H. and Lyr H.), pp. 97–102. Deutsche Phytomedizinische Gesellschaft, Braunschweig. [Google Scholar]
• Hawkins, N. J. , Cools, H. J. , Sierotzki, H. , Shaw, M. W. , Knogge, W. , Kelly, S. L. , Kelly, D. E. & Fraaije, B. A. (2014b). Paralog re‐emergence: a novel, historically contingent mechanism in the evolution of antimicrobial resistance. Molecular Biology and Evolution 31, 1793–1802. [PMC free article] [PubMed] [Google Scholar]
• Hedrick, P. W. (2013). Adaptive introgression in animals: examples and comparison to new mutation and standing variation as sources of adaptive variation. Molecular Ecology 22, 4606–4618. [PubMed] [Google Scholar]
• Hermisson, J. & Pennings, P. S. (2005). Soft sweeps: molecular population genetics of adaptation from standing genetic variation. Genetics 169, 2335–2352. [PMC free article] [PubMed] [Google Scholar]
• Hobbelen, P. H. F. , Paveley, N. D. & Van Den Bosch, F. (2014). The emergence of resistance to fungicides. PLoS One 9, e91910. [PMC free article] [PubMed] [Google Scholar]
• I5K Consortium (2013). The I5K Initiative: advancing arthropod genomics for knowledge, human health, agriculture, and the environment. Journal of Heredity 104, 595–600. [PMC free article] [PubMed] [Google Scholar]
• Karasov, T. , Messer, P. W. & Petrov, D. A. (2010). Evidence that adaptation in Drosophila is not limited by mutation at single sites. PLoS Genetics 6, 10. [PMC free article] [PubMed] [Google Scholar]
• Kaundun, S. S. (2014). Resistance to acetyl‐CoA carboxylase‐inhibiting herbicides. Pest Management Science 70, 1405–1417. [PubMed] [Google Scholar]
• Kraiczy, P. , Haase, U. , Gencic, S. , Flindt, S. , Anke, T. , Brandt, U. & Vonjagow, G. (1996). The molecular basis for the natural resistance of the cytochrome bc(1) complex from strobilurin‐producing basidiomycetes to center Q(p) inhibitors. European Journal of Biochemistry 235, 54–63. [PubMed] [Google Scholar]
• Kretschmer, M. , Leroch, M. , Mosbach, A. , Walker, A. S. , Fillinger, S. , Mernke, D. , Schoonbeek, H. J. , Pradier, J. M. , Leroux, P. , De Waard, M. A. & Hahn, M. (2009). Fungicide‐driven evolution and molecular basis of multidrug resistance in field populations of the grey mould fungus Botrytis cinerea . PLoS Pathogens 5, 13. [PMC free article] [PubMed] [Google Scholar]
• Kuester, A. , Chang, S.‐M. & Baucom, R. S. (2015). The geographic mosaic of herbicide resistance evolution in the common morning glory, Ipomoea purpurea: evidence for resistance hotspots and low genetic differentiation across the landscape. Evolutionary Applications 8, 821–833. [PMC free article] [PubMed] [Google Scholar]
• Kwit, C. , Moon, H. S. , Warwick, S. I. & Stewart, C. N. Jr. (2011). Transgene introgression in crop relatives: molecular evidence and mitigation strategies. Trends in Biotechnology 29, 284–293. [PubMed] [Google Scholar]
• Lagator, M. , Colegrave, N. & Neve, P. (2014). Selection history and epistatic interactions impact dynamics of adaptation to novel environmental stresses. Proceedings of the Royal Society of London B: Biological Sciences 281, 20141679. [PMC free article] [PubMed] [Google Scholar]
• Lagator, M. , Vogwill, T. , Colegrave, N. & Neve, P. (2013). Herbicide cycling has diverse effects on evolution of resistance in Chlamydomonas reinhardtii . Evolutionary Applications 6, 197–206. [PMC free article] [PubMed] [Google Scholar]
• Lässig, M. , Mustonen, V. & Walczak, A. M. (2017). Predicting evolution. Nature Ecology & Evolution 1, 0077. [PubMed] [Google Scholar]
• Lendenmann, M. H. , Croll, D. & Mcdonald, B. A. (2015). QTL mapping of fungicide sensitivity reveals novel genes and pleiotropy with melanization in the pathogen Zymoseptoria tritici . Fungal Genetics and Biology 80, 53–67. [PubMed] [Google Scholar]
• Linde, C. C. , Zala, M. & Mcdonald, B. A. (2009). Molecular evidence for recent founder populations and human‐mediated migration in the barley scald pathogen Rhynchosporium secalis . Molecular Phylogenetics and Evolution 51, 454–464. [PubMed] [Google Scholar]
• Liu, K. J. , Steinberg, E. , Yozzo, A. , Song, Y. , Kohn, M. H. & Nakhleh, L. (2015). Interspecific introgressive origin of genomic diversity in the house mouse. Proceedings of the National Academy of Sciences of the United States of America 112, 196–201. [PMC free article] [PubMed] [Google Scholar]
• Lucas, J. A. , Hawkins, N. J. & Fraaije, B. A. (2015). The evolution of fungicide resistance. Advances in Applied Microbiology 90, 29–92. [PubMed] [Google Scholar]
• Lynd, A. , Weetman, D. , Barbosa, S. , Egyir Yawson, A. , Mitchell, S. , Pinto, J. , Hastings, I. & Donnelly, M. J. (2010). Field, genetic, and modelling approaches show strong positive selection acting upon an insecticide resistance mutation in Anopheles gambiae s.s . Molecular Biology and Evolution 27, 1117–1125. [PMC free article] [PubMed] [Google Scholar]
• Mackie, J. A. , Levinton, J. S. , Przeslawski, R. , Delambert, D. & Wallace, W. (2010). Loss of evolutionary resistance by the oligochaete Limnodrilus hoffmeisteri to a toxic substance‐cost or gene flow? Evolution 64, 152–165. [PubMed] [Google Scholar]
• Maclean, R. C. , Hall, A. R. , Perron, G. G. & Buckling, A. (2010). The population genetics of antibiotic resistance: integrating molecular mechanisms and treatment contexts. Nature Reviews. Genetics 11, 405–414. [PubMed] [Google Scholar]
• Mallory‐Smith, C. A. & Olguin, E. S. (2011). Gene flow from herbicide‐resistant crops: it's not just for transgenes. Journal of Agricultural and Food Chemistry 59, 5813–5818. [PubMed] [Google Scholar]
• Manalil, S. , Busi, R. , Renton, M. & Powles, S. B. (2011). Rapid evolution of herbicide resistance by low herbicide dosages. Weed Science 59, 210–217. [Google Scholar]
• Marsit, S. , Mena, A. , Bigey, F. , Sauvage, F.‐X. , Couloux, A. , Guy, J. , Legras, J.‐L. , Barrio, E. , Dequin, S. & Galeote, V. (2015). Evolutionary advantage conferred by an eukaryote‐to‐eukaryote gene transfer event in wine yeasts. Molecular Biology and Evolution 32, 1695–1707. [PMC free article] [PubMed] [Google Scholar]
• Martins, B. A. B. , Sun, L. N. & Mallory‐Smith, C. (2015). Resistance allele movement between imazamox‐resistant wheat and jointed goatgrass (Aegilops cylindrica) in Eastern Oregon wheat fields. Weed Science 63, 855–863. [Google Scholar]
• McDonald, B. A. & Linde, C. C. (2002). Pathogen population genetics, evolutionary potential, and durable resistance. Annual Review of Phytopathology 40, 349–379. [PubMed] [Google Scholar]
• McKenzie, J. A. & Batterham, P. (1998). Predicting insecticide resistance: mutagenesis, selection and response. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences 353, 1729–1734. [PMC free article] [PubMed] [Google Scholar]
• Mellado, E. , Garcia‐Effron, G. , Alcazar‐Fuoli, L. , Melchers, W. J. G. , Verweij, P. E. , Cuenca‐Estrella, A. & Rodriguez‐Tudela, J. L. (2007). A new Aspergillus fumigatus resistance mechanism conferring in vitro cross‐resistance to azole antifungals involves a combination of CYP51A alterations. Antimicrobial Agents and Chemotherapy 51, 1897–1904. [PMC free article] [PubMed] [Google Scholar]
• Menchari, Y. , Camilleri, C. , Michel, S. , Brunel, D. , Dessaint, F. , Le Corre, V. & Délye, C. (2006). Weed response to herbicides: regional‐scale distribution of herbicide resistance alleles in the grass weed Alopecurus myosuroides . New Phytologist 171, 861–873. [PubMed] [Google Scholar]
• Mengistu, L. W. , Messersmith, C. G. & Christoffers, M. J. (2003). Diversity of herbicide resistance among wild oat sampled 36 years apart. Weed Science 51, 764–773. [Google Scholar]
• Mengistu, L. W. , Messersmith, C. G. & Christoffers, M. J. (2005). Genetic diversity of herbicide‐resistant and ‐susceptible Avena fatua populations in North Dakota and Minnesota. Weed Research 45, 413–423. [Google Scholar]
• Messer, P. W. & Petrov, D. A. (2013). Population genomics of rapid adaptation by soft selective sweeps. Trends in Ecology & Evolution 28, 659–669. [PMC free article] [PubMed] [Google Scholar]
• Messina, F. J. , Mendenhall, M. & Jones, J. C. (2009). An experimentally induced host shift in a seed beetle. Entomologia Experimentalis et Applicata 132, 39–49. [Google Scholar]
• Mouches, C. , Pasteur, N. , Berge, J. , Hyrien, O. , Raymond, M. , De Saint Vincent, B. , de Silvestri, M. & Georghiou, G. (1986). Amplification of an esterase gene is responsible for insecticide resistance in a California Culex mosquito. Science 233, 778–780. [PubMed] [Google Scholar]
• Munkacsi, A. B. , Stoxen, S. & May, G. (2008). Ustilago maydis populations tracked maize through domestication and cultivation in the Americas. Proceedings of the Royal Society of London B: Biological Sciences 275, 1037–1046. [PMC free article] [PubMed] [Google Scholar]
• Neafsey, D. E. , Barker, B. M. , Sharpton, T. J. , Stajich, J. E. , Park, D. J. , Whiston, E. , Hung, C.‐Y. , Mcmahan, C. , White, J. , Sykes, S. , Heiman, D. , Young, S. , Zeng, Q. , Abouelleil, A. , Aftuck, L. , et al. (2010). Population genomic sequencing of Coccidioides fungi reveals recent hybridization and transposon control. Genome Research 20, 938–946. [PMC free article] [PubMed] [Google Scholar]
• Neve, P. , Busi, R. , Renton, M. & Vila‐Aiub, M. M. (2014). Expanding the eco‐evolutionary context of herbicide resistance research. Pest Management Science 70, 1385–1393. [PubMed] [Google Scholar]
• Neve, P. & Powles, S. B. (2005a). High survival frequencies at low herbicide use rates in populations of Lolium rigidum result in rapid evolution of herbicide resistance. Heredity 95, 485–492. [PubMed] [Google Scholar]
• Neve, P. & Powles, S. B. (2005b). Recurrent selection with reduced herbicide rates results in the rapid evolution of herbicide resistance in Lolium rigidum . Theoretical and Applied Genetics 110, 1154–1166. [PubMed] [Google Scholar]
• Norris, L. C. , Main, B. J. , Lee, Y. , Collier, T. C. , Fofana, A. , Cornel, A. J. & Lanzaro, G. C. (2015). Adaptive introgression in an African malaria mosquito coincident with the increased usage of insecticide‐treated bed nets. Proceedings of the National Academy of Sciences of the United States of America 112, 815–820. [PMC free article] [PubMed] [Google Scholar]
• Okada, M. , Hanson, B. D. , Hembree, K. J. , Peng, Y. , Shrestha, A. , Stewart, C. N. Jr. , Wright, S. D. & Jasieniuk, M. (2013). Evolution and spread of glyphosate resistance in Conyza canadensis in California. Evolutionary Applications 6, 761–777. [PMC free article] [PubMed] [Google Scholar]
• Oliver, R. P. (2014). A reassessment of the risk of rust fungi developing resistance to fungicides. Pest Management Science 70, 1641–1645. [PubMed] [Google Scholar]
• Omrane, S. , Sghyer, H. , Audéon, C. , Lanen, C. , Duplaix, C. , Walker, A.‐S. & Fillinger, S. (2015). Fungicide efflux and the MgMFS1 transporter contribute to the MDR phenotype in Zymoseptoria tritici field isolates. Environmental Microbiology 17, 2805–2823. [PubMed] [Google Scholar]
• Orr, H. A. & Betancourt, A. J. (2001). Haldane's sieve and adaptation from the standing genetic variation. Genetics 157, 875–884. [PMC free article] [PubMed] [Google Scholar]
• Palumbi, S. R. (2001). Humans as the world's greatest evolutionary force. Science 293, 1786–1790. [PubMed] [Google Scholar]
• Pedro, H. , Maheswari, U. , Urban, M. , Irvine, A. G. , Cuzick, A. , McDowall, M. D. , Staines, D. M. , Kulesha, E. , Hammond‐Kosack, K. E. & Kersey, P. J. (2016). Phytopath: an integrative resource for plant pathogen genomics. Nucleic Acids Research 44, D688–D693. [PMC free article] [PubMed] [Google Scholar]
• Peng, Y. , Lai, Z. , Lane, T. , Nageswara‐Rao, M. , Okada, M. , Jasieniuk, M. , O'Geen, H. , Kim, R. W. , Sammons, R. D. , Rieseberg, L. H. & Stewart, C. N. (2014). De novo genome assembly of the economically important weed Horseweed using integrated data from multiple sequencing platforms. Plant Physiology 166, 1241–1254. [PMC free article] [PubMed] [Google Scholar]
• Pennings, P. S. , Kryazhimskiy, S. & Wakeley, J. (2014). Loss and recovery of genetic diversity in adapting populations of HIV. PLoS Genetics 10, e1004000. [PMC free article] [PubMed] [Google Scholar]
• Perry, J. & Wright, G. (2013). The antibiotic resistance ‘mobilome’: searching for the link between environment and clinic. Frontiers in Microbiology 4, 138. [PMC free article] [PubMed] [Google Scholar]
• Perry, T. , Heckel, D. G. , McKenzie, J. A. & Batterham, P. (2008). Mutations in Dα1 or Dβ2 nicotinic acetylcholine receptor subunits can confer resistance to neonicotinoids in Drosophila melanogaster . Insect Biochemistry and Molecular Biology 38, 520–528. [PubMed] [Google Scholar]
• Powles, S. B. (2008). Evolved glyphosate‐resistant weeds around the world: lessons to be learnt. Pest Management Science 64, 360–365. [PubMed] [Google Scholar]
• Powles, S. B. & Yu, Q. (2010). Evolution in action: plants resistant to herbicides. Annual Review of Plant Biology 61, 317–347. [PubMed] [Google Scholar]
• Pritchard, J. K. & Di Rienzo, A. (2010). Adaptation ‐ not by sweeps alone. Nature Reviews Genetics 11, 665–667. [PMC free article] [PubMed] [Google Scholar]
• Rajon, E. & Masel, J. (2013). Compensatory evolution and the origins of innovations. Genetics 193, 1209–1220. [PMC free article] [PubMed] [Google Scholar]
• Raymond, M. , Callaghan, A. , Fort, P. & Pasteur, N. (1991). Worldwide migration of amplified insecticide resistance genes in mosquitoes. Nature 350, 151–153. [PubMed] [Google Scholar]
• Reichman, J. R. , Watrud, L. S. , Lee, E. H. , Burdick, C. A. , Bollman, M. A. , Storm, M. J. , King, G. A. & Mallory‐Smith, C. A. (2006). Establishment of transgenic herbicide‐resistant creeping bentgrass (Agrostis stolonifera l.) in nonagronomic habitats. Molecular Ecology 15, 4243–4255. [PubMed] [Google Scholar]
• Reznick, D. , Bryant, M. J. & Bashey, F. (2002). r‐ and K‐selection revisited: the role of population regulation in life‐history evolution. Ecology 83, 1509–1520. [Google Scholar]
• Roesti, M. , Gavrilets, S. , Hendry, A. P. , Salzburger, W. & Berner, D. (2014). The genomic signature of parallel adaptation from shared genetic variation. Molecular Ecology 23, 3944–3956. [PMC free article] [PubMed] [Google Scholar]
• Rose, C. J. , Chapman, J. R. , Marshall, S. D. , Lee, S. F. , Batterham, P. , Ross, H. A. & Newcomb, R. D. (2011). Selective sweeps at the organophosphorus insecticide resistance locus, Rop‐1, have affected variation across and beyond the α‐esterase gene cluster in the Australian sheep blowfly, Lucilia cuprina . Molecular Biology and Evolution 28, 1835–1846. [PubMed] [Google Scholar]
• Russell, P. E. (2004). Sensitivity Baselines in Fungicide Resistance Research and Management. FRAC Monograph Number 3. CropLife International, Brussels. [Google Scholar]
• Scalliet, G. , Bowler, J. , Luksch, T. , Kirchhofer‐Allan, L. , Steinhauer, D. , Ward, K. , Niklaus, M. , Verras, A. , Csukai, M. , Daina, A. & Fonné‐Pfister, R. (2012). Mutagenesis and functional studies with succinate dehydrogenase inhibitors in the wheat pathogen Mycosphaerella graminicola . PLoS One 7, e35429. [PMC free article] [PubMed] [Google Scholar]
• Schmidt, J. M. , Good, R. T. , Appleton, B. , Sherrard, J. , Raymant, G. C. , Bogwitz, M. R. , Martin, J. , Daborn, P. J. , Goddard, M. E. , Batterham, P. & Robin, C. (2010). Copy number variation and transposable elements feature in recent, ongoing adaptation at the Cyp6g1 locus. PLoS Genetics 6, e1000998. [PMC free article] [PubMed] [Google Scholar]
• Schmitz, H. K. , Medeiros, C.‐A. , Craig, I. R. & Stammler, G. (2014). Sensitivity of Phakopsora pachyrhizi towards quinone‐outside‐inhibitors and demethylation‐inhibitors, and corresponding resistance mechanisms. Pest Management Science 70, 378–388. [PubMed] [Google Scholar]
• Schnabel, G. & Chen, F. (2013). Fungicide‐induced mutagenesis in Monilinia fructicola and implications for resistance management. Phytopathology 103, 128–128. [Google Scholar]
• Shemshedini, L. & Wilson, T. G. (1990). Resistance to juvenile hormone and an insect growth regulator in Drosophila is associated with an altered cytosolic juvenile hormone‐binding protein. Proceedings of the National Academy of Sciences of the United States of America 87, 2072–2076. [PMC free article] [PubMed] [Google Scholar]
• Song, Y. , Endepols, S. , Klemann, N. , Richter, D. , Matuschka, F.‐R. , Shih, C.‐H. , Nachman, M. W. & Kohn, M. H. (2011). Adaptive introgression of anticoagulant rodent poison resistance by hybridization between old world mice. Current Biology 21, 1296–1301. [PMC free article] [PubMed] [Google Scholar]
• Stern, D. L. (2013). The genetic causes of convergent evolution. Nature Reviews Genetics 14, 751–764. [PubMed] [Google Scholar]
• Stukenbrock, E. H. , Banke, S. , Javan‐Nikkhah, M. & Mcdonald, B. A. (2007). Origin and domestication of the fungal wheat pathogen Mycosphaerella graminicola via sympatric speciation. Molecular Biology and Evolution 24, 398–411. [PubMed] [Google Scholar]
• Sun, B.‐F. , Xiao, J.‐H. , He, S. , Liu, L. , Murphy, R. W. & Huang, D.‐W. (2013). Multiple interkingdom horizontal gene transfers in Pyrenophora and closely related species and their contributions to phytopathogenic lifestyles. PLoS One 8, e60029. [PMC free article] [PubMed] [Google Scholar]
• Swanton, C. J. , Mashhadi, H. R. , Solomon, K. R. , Afifi, M. M. & Duke, S. O. (2011). Similarities between the discovery and regulation of pharmaceuticals and pesticides: in support of a better understanding of the risks and benefits of each. Pest Management Science 67, 790–797. [PubMed] [Google Scholar]
• Thompson, M. , Steichen, J. C. & Ffrench‐Constant, R. H. (1993). Conservation of cyclodiene insecticide resistance‐associated mutations in insects. Insect Molecular Biology 2, 149–154. [PubMed] [Google Scholar]
• Torriani, S. F. F. , Brunner, P. C. , Mcdonald, B. A. & Sierotzki, H. (2009). QoI resistance emerged independently at least 4 times in European populations of Mycosphaerella graminicola . Pest Management Science 65, 155–162. [PubMed] [Google Scholar]
• Troczka, B. , Zimmer, C. T. , Elias, J. , Schorn, C. , Bass, C. , Davies, T. G. , Field, L. M. , Williamson, M. S. , Slater, R. & Nauen, R. (2012). Resistance to diamide insecticides in diamondback moth, Plutella xylostella (Lepidoptera: Plutellidae) is associated with a mutation in the membrane‐spanning domain of the ryanodine receptor. Insect Biochemistry and Molecular Biology 42, 873–880. [PubMed] [Google Scholar]
• Ulber, L. , Nordmeyer, H. & Zwerger, P. (2013). Resistance risk assessment within herbicide authorisation ‐ a call for sensitivity data. Pest Management Science 69, 160–164. [PubMed] [Google Scholar]
• Unckless, R. L. , Clark, A. G. & Messer, P. W. (2017). Evolution of resistance against CRISPR/Cas9 gene drive. Genetics 205, 827–841. [PMC free article] [PubMed] [Google Scholar]
• Van Den Bosch, F. , Paveley, N. , Shaw, M. , Hobbelen, P. & Oliver, R. (2011). The dose rate debate: does the risk of fungicide resistance increase or decrease with dose? Plant Pathology 60, 597–606. [Google Scholar]
• Van Leeuwen, T. , Demaeght, P. , Osborne, E. J. , Dermauw, W. , Gohlke, S. , Nauen, R. , Grbić, M. , Tirry, L. , Merzendorfer, H. & Clark, R. M. (2012). Population bulk segregant mapping uncovers resistance mutations and the mode of action of a chitin synthesis inhibitor in arthropods. Proceedings of the National Academy of Sciences of the United States of America 109, 4407–4412. [PMC free article] [PubMed] [Google Scholar]
• Vatsiou, A. I. , Bazin, E. & Gaggiotti, O. E. (2016). Detection of selective sweeps in structured populations: a comparison of recent methods. Molecular Ecology 25, 89–103. [PubMed] [Google Scholar]
• Verweij, P. E. , Kema, G. H. J. , Zwaan, B. & Melchers, W. J. G. (2013). Triazole fungicides and the selection of resistance to medical triazoles in the opportunistic mould Aspergillus fumigatus . Pest Management Science 69, 165–170. [PubMed] [Google Scholar]
• Vyas, S. C. (1988). Nontarget Effects of Agricultural Fungicides. CRC Press, Boca Raton, Florida, USA. [Google Scholar]
• Warwick, S. I. , Légère, A. , Simard, M.‐J. & James, T. (2008). Do escaped transgenes persist in nature? The case of an herbicide resistance transgene in a weedy Brassica rapa population. Molecular Ecology 17, 1387–1395. [PubMed] [Google Scholar]
• Watrud, L. S. , Lee, E. H. , Fairbrother, A. , Burdick, C. A. , Reichman, J. R. , Bollman, M. A. , Storm, M. J. , King, G. A. & Van De Water, P. K. (2004). Evidence for landscape‐level, pollen‐mediated gene flow from genetically modified creeping bentgrass with CP4 EPSPS as a marker. Proceedings of the National Academy of Sciences of the United States of America 101, 14533–14538. [PMC free article] [PubMed] [Google Scholar]
• Watson, G. B. , Chouinard, S. W. , Cook, K. R. , Geng, C. , Gifford, J. M. , Gustafson, G. D. , Hasler, J. M. , Larrinua, I. M. , Letherer, T. J. , Mitchell, J. C. , Pak, W. L. , Salgado, V. L. , Sparks, T. C. & Stilwell, G. E. (2010). A spinosyn‐sensitive Drosophila melanogaster nicotinic acetylcholine receptor identified through chemically induced target site resistance, resistance gene identification, and heterologous expression. Insect Biochemistry and Molecular Biology 40, 376–384. [PubMed] [Google Scholar]
• Weetman, D. , Mitchell, S. N. , Wilding, C. S. , Birks, D. P. , Yawson, A. E. , Essandoh, J. , Mawejje, H. D. , Djogbenou, L. S. , Steen, K. , Rippon, E. J. , Clarkson, C. S. , Field, S. G. , Rigden, D. J. & Donnelly, M. J. (2015). Contemporary evolution of resistance at the major insecticide target site gene Ace‐1 by mutation and copy number variation in the malaria mosquito Anopheles gambiae . Molecular Ecology 24, 2656–2672. [PMC free article] [PubMed] [Google Scholar]
• Wieczorek, T. M. , Berg, G. , Semaskiene, R. , Mehl, A. , Sierotzki, H. , Stammler, G. , Justesen, A. F. & Jorgensen, L. N. (2015). Impact of DMI and SDHI fungicides on disease control and CYP51 mutations in populations of Zymoseptoria tritici from Northern Europe. European Journal of Plant Pathology 143, 861–871. [Google Scholar]
• World Health Organisation (2014). Antimicrobial Resistance: Global Report on Surveillance. WHO, Geneva. [Google Scholar]
• Yan, X. , Ma, W.‐B. , Li, Y. , Wang, H. , Que, Y.‐W. , Ma, Z.‐H. , Talbot, N. J. & Wang, Z.‐Y. (2011). A sterol 14α‐demethylase is required for conidiation, virulence and for mediating sensitivity to sterol demethylation inhibitors by the rice blast fungus Magnaporthe oryzae . Fungal Genetics and Biology 48, 144–153. [PubMed] [Google Scholar]
• Yoshida, K. , Sasaki, E. & Kamoun, S. (2015). Computational analyses of ancient pathogen DNA from herbarium samples: challenges and prospects. Frontiers in Plant Science 6, 771. [PMC free article] [PubMed] [Google Scholar]
• You, M. , Yue, Z. , He, W. , Yang, X. , Yang, G. , Xie, M. , Zhan, D. , Baxter, S. W. , Vasseur, L. , Gurr, G. M. , Douglas, C. J. , Bai, J. , Wang, P. , Cui, K. , Huang, S. , et al. (2013). A heterozygous moth genome provides insights into herbivory and detoxification. Nature Genetics 45, 220–225. [PubMed] [Google Scholar]
• Yu, Q. & Powles, S. B. (2014). Resistance to AHAS inhibitor herbicides: current understanding. Pest Management Science 70, 1340–1350. [PubMed] [Google Scholar]
• Zhang, X.‐S. & Hill, W. G. (2005). Genetic variability under mutation selection balance. Trends in Ecology & Evolution 20, 468–470. [PubMed] [Google Scholar]
• Zheng, D. S. , Olaya, G. & Koller, W. (2000). Characterization of laboratory mutants of Venturia inaequalis resistant to the strobilurin‐related fungicide kresoxim‐methyl. Current Genetics 38, 148–155. [PubMed] [Google Scholar]

Page 2

Evolution of pesticide resistance: (A) by de novo mutation; (B) by selection from standing genetic variation.

Click on the image to see a larger version.