Studying the Evolution of Adaptation: Doc Insertion, Gene Duplication, and Viral Resistance in Flies
Six years after Aminetzach et al. published their paper documenting the insertion of Doc1420 into the second exon of CHKov1 some ninety-thousand years ago, and the observed insecticide resistance gained from the functional protein generated from this truncated gene , Magwire et al. proposed a likely explanation for how this curious adaptation could have taken place so long before the introduction of insecticide use by humans. Let it be known that unless otherwise noted, all information within this article is sourced from Magwire et al.’s paper entitled “Successive Increases in the Resistance of Drosophila to Viral Infection through a Transposon Insertion Followed by a Duplication” . By analyzing the host-parasite interaction between D.melanogaster and the Sigma virus, which specifically targets D.melanogaster, they have been able to deduce that the insertion of TE Doc1420 into the coding region of CHKov1 provided significant reduction in susceptibility to Sigma virus infection. Since the Sigma virus causes paralysis in infected flies, those with the truncated CHKov1 gene experienced positive selective pressure, resulting in the truncated allele becoming very common worldwide. Magwire et al. go on to describe a much more recent evolutionary event involving a complex rearrangement of CHKov1 and it’s neighboring paralog CHKov2 gene that result in two duplications of each gene, which provides even stronger resistance to Sigma virus infection. It appears that while the Doc1420 truncation of CHKov1 was first documented by Aminetzach et al. as being remarkable for the resistance to insecticides it imbued to those flies containing it, it was originally selected for in the population as a result of adaptation to the host-specific Sigma virus, and this adaptation fortuitously pre-adapted D.melanogaster to better resist organophosphate insecticides.
METHODS AND RESULTS:
In contrast to the abundance of knowledge that was currently available in 2011 with regards to host-parasite relationships in plants, Magwire et al. noted that comparatively little was known about such relationships in animals. It was already known that D.melanogaster was the target of a host-specific virus named Sigma, so they chose to use this host-parasite relationship to better understand the phenomena within invertebrates. This particular relationship was attractive to them because the host-specificity of Sigma to D.melangogaster would have caused strong reciprocal adaptation between the two, while also making it easier to observe how their co-evolution occurred. Sigma is specifically transmitted vertically from parent to offspring via sperm, which offers a further element of simplicity to the understanding of the host-parasite interaction at play.
Magwire et al. set off on their investigation by looking at a resistance gene that was referred to as ref(3)D by Gay P. in 1978, who mapped its’ location on the right arm of the third chromosome between two visible markers . This gene provided its’ carriers with significantly more resistance to the Sigma virus by disrupting the virus’ ability to replicate. Since ref(3)D was originally mapped over thirty years prior, the approximate region attributed to the resistance gene was large due to technological limitation. Magwire et al. set to mapping the exact location of this resistance gene more accurately.
In order to accomplish this, they began with two different lines of D.melanogaster (line OOP, and line 22a), and separately evaluated their reaction to the Sigma virus 8-11 days after injection. They found that less than 5% of the OOP line of flies remained affected by the paralysis of Sigma. Conversely, over 95% of the 22a line of flies remained affected by Sigma paralysis. The researchers knew that the OOP line contained the ref(3)D gene on its’ third chromosome, but also knew that the OOP line contained an allelic variant that reduces how much Sigma virus is transmitted through sperm. Since they wanted to specifically study the resistance gained by the element within ref(3)D without the complication of this other gene affecting their data, they crossed the OOP and 22a lines to generate a new line that had a recombination event between the locations of these two genes. The result was a line which was homozygous for the resistant OOP ref(3)D gene, and the 22a gene that results in high rates of sperm transmission.
By analyzing 191 flies containing recombinant third chromosomes between the resistant and susceptible stocks, Magwire et al. were able to use molecular markers to identify those that underwent a recombination event within the 12 cM interval believed to contain the actual resistance gene. 21 of them were homozygous recombinants within this region, which were used to create 21 lines of flies. Within this defined region there were still multiple different locations that recombination could occur. Where the OOP DNA stopped and the 22a DNA began varied slightly from line to line. All the lines were genotyped using molecular markers around the suspect region, and were all injected with Sigma virus.
They found that the lines of flies were either clearly resistant, or they were clearly susceptible to the Sigma virus, which means that somewhere amongst the slight variations in the way the recombination event occurred within those lines lies the true location of the Sigma resistance gene. By comparing the location of the molecular markers indicating the genotypic characteristic of each line with the proportion of those within each line that were infected, the researchers were able to identify a 182kb region where there was a perfect association between infection rates and genotype. In other words, if a line of flies recombined such that it had OOP DNA within that 182kb region, they were resistant to the Sigma virus; lines of flies with susceptible 22a DNA in this 182kb region after recombination were infected.
In order to investigate with more accuracy where the Sigma resistance gene was located within this 182kb region, they repeated their procedure using a larger sample of flies. Rather than using 191, 1920 flies were screened for informative recombination events within this tightened region and 32 new lines were created. Just like with the first experiment, it was clearly observable which lines were resistant and which were susceptible. Thus, by again comparing where the genotypic OOP DNA stopped and the 22a DNA started with the proportion of flies that were infected in each line, the researchers were able to reduce the region of 182kb where they had previously identified a perfect association between genotype and infection, to a smaller 60kb region within it. So long as one of the given lines of flies contained OOP rather than 22a DNA within this 60kb region, they were not susceptible to the Sigma virus.
The researchers wanted to be able to select for recombinants within this smaller region so that they could even more accurately map the exact location of the Sigma virus resistance gene. In order to accomplish this within the identified 60kb suspect region, phenotypic markers were utilized in lieu of molecular markers. This was done with the use of P-elements, which are transposons that are present specifically in D.melanogaster and are widely used for mutagenesis. The researchers took two P-elements that each carried eye-color markers, and combined them into one to produce a susceptible mapping stock. This mapping stock was then crossed to a resistant fly line, and they selected recombinants that carried only one of the two genes, which ensured that recombination had occurred within the target region in question. Using this method the researchers were able to create 10 lines of flies that were homozygous for the recombinant chromosome, and by again comparing the genotype with the proportion of flies that were infected in each of those 10 lines across this 60kb region they were able to narrow the suspect region that could contain the Sigma resistance gene to a 36kb span within it.
In order to map the gene within this even smaller 36kb region, Magwire et al. chose to use site-specific recombination in male flies with P-elements. Recombination does not usually occur in the male Drosophila germ line, but in crosses involving wild-type strains containing P-elements male recombination can occur. In this fourth experiment the researchers crossed P-element lines that were susceptible to Sigma virus with a resistant line, and were able to induce recombination at the site of the element. This procedure yielded them four recombinants that were viable homozygotes within the target region. Other flies that did not undergo a recombination event with this procedure were also used to generate lines that served as controls for the effects of genetic background within the flies. Thus, the flies of study in this fourth experiment either had the P-element-containing susceptible chromosome, or the chromosome containing the resistant gene. Using molecular markers around the location of the P-element, the researchers were able to confirm that recombination had indeed occurred. After injecting the recombinant lines and control lines with Sigma virus, they found that two of the recombinants were located only 3089bp apart, yet had dramatically different levels of Sigma virus resistance; one of them was highly resistant while the other was highly susceptible. One of the other experimental recombinant lines showed significant resistance to the virus as well, but not as much resistance as the highly resistant line just mentioned. This region corresponds with a gene called CHKov2 in the published genome of D.melanogaster.
With a better idea of where exactly to look for the location of the individual polymorphism causing Sigma resistance, Magwire et al. sequenced the region in and around CHKov2 in each of these lines. They discovered that the highly resistant line of flies just mentioned had previously underwent a complicated rearrangement within this region, whereas the line that was not as robustly resistant had the same gene order as is published in the Drosophila genome. They also found that both of these resistant lines contained a naturally occurring Doc1420 (TE) insertion within an exon of CHKov1, a neighboring paralog of CHKov2. The major difference between these two lines is a complex rearrangement involving two duplications of CHKov1 and CHKov2 in the former, and no such duplication event in the latter.
Magwire et al. believe it is probable that this gene duplication event is responsible for the increased amount of resistance observed between the highly resistant line and the moderately resistant line because the region they mapped using male recombination contains only one SNP that differs between the two of them besides the rearrangement itself. But, was the added resistance due to the fact that CHKov1 had be duplicated twice, or because CHKov2 had been duplicated twice? To find out, the researchers used PCR to see if the expression of either gene was different in the resistant and highly resistant flies after injection. They injected the lines with Sigma, and six days later they found that CHKov2 expression was 5.6-fold greater, and 9.6-fold greater twelve days after injection. Even though CHKov1 had been duplicated twice, its’ expression did not change as a result of injection by Sigma virus like CHKov2 did after being duplicated twice.
At this point Magwire et al. had located the likely gene region responsible for D.melanogaster resistance to Sigma virus, but was the resistance gene actually reducing viral titres, or was it just altering the flies’ sensitivity to the virally produced CO2 that was causing them to be paralyzed? To investigate this question, the researchers used PCR to gauge the approximate number of copies of the viral genome within both resistant and highly resistant flies. For this PCR they used the same samples they previously utilized to evaluate gene expression, and found a 79-fold and 138-fold decrease in in Sigma virus load in the highly resistant line six and twelve days later, respectively. Thus, it appeared that the resistance gene was indeed reducing viral titres rather than just affecting CO2 sensitivity.
Next the researchers wanted to find out just how prevalent the complex rearrangement of CHKov1 and CHKov2 the resulted in being highly resistant to Sigma virus actually was in the wild. They acquired a set of highly inbred North American fly lines whose genomes have been sequenced from the Drosophila Genetic Reference Panel (DGRP), then used PCR to genotype all the lines for the Doc1420 insertion into CHKov1 and the complex rearrangement found in the highly resistant lines previously identified. Whereas the Doc1420 insertion was common within the lines, the rearrangement was found in none of them. Thus the resistance this population had to the Sigma virus could not be attributed to the complex rearrangement and duplication of the CHKov1 and CHKov2 genes.
Seeing as how both the moderately resistant and highly resistant lines from their fourth experiment contained the Doc1420 insertion into the coding region of CHKov1, they decided to investigate how much, if any, resistance is imbued on the carrier by the Doc1420 insertion alone. This was accomplished by injecting 11870 flies from 186 different DGRP lines with Sigma virus and evaluating proportions of infection 13 days later in comparison to susceptible lines. They found that the Doc1420 insertion alone was responsible for a 52% drop in infection rates.
A pecking order of sorts had been observed by Magwire et al. with regards to allelic resistance to Sigma infection. The ancestral gene associated with D.simulans contained neither the Doc1420 insertion nor any form of complex rearrangement of the CHKov1 and CHKov2 genes, and was the most susceptible to infection by Sigma virus. The DGRP lines did not contain any gene rearrangement, but did contain the Doc1420 insertion that provided moderate virus resistance. The lines of flies they had which contained both the Doc1420 insertion and the complex rearrangement were all highly resistant to the Sigma virus, but were relatively rare in the wild. This means that the susceptible ancestral form eventually acquired a TE that provided resistance, then recently underwent a duplication event that provided even more resistance in some lines that has not as of yet proliferated to the extend of the previous Doc1420 polymorphism.
Magwire et al. were still not quite convinced that there were no other polymorphisms responsible for resistance to the Sigma virus. They used 150 lines from the inbred DGRP flies, and sequenced their genomes within the 60kb region they identified in their second experiment. The researchers found that out of the 468 different polymorphisms contained within this region of the DGRP lines, 32 that were located around the CHKov1 gene showed significant association with resistance to the Sigma virus. But, were any of these 32 SNPs really responsible? The researchers found that there was a large degree of linkage disequilibrium between the Doc1420 insertion and the sites surrounding its’ location within the CHKov1 gene, which means that those 32 possibly-responsible SNPs could very well only be registering association with resistance as a result of one, singular polymorphism. So, they ran the test again but instead decided to compare both the presence and absence of the Doc1420 insertion in their model. After including Doc1420 as an explanatory factor in the analysis, none of the previously observed associations of the 32 SNPs with resistance remained. Since the Doc1420 insertion involves a dramatic alteration in the coding sequence of CHKov1, this singular polymorphism is therefore the most likely one responsible for the resistance gene. To further confirm that there weren’t any other polymorphisms that could be responsible, the OOP and 22s lines originally used at the beginning of the study were also sequenced within this same 60kB region. As with the inbred DGRP line, none of the other polymorphisms identified could be significantly associated with Sigma virus resistance.
Looking more closely at the Doc1420 insertion itself, Magwire et al. wanted to analyze the extent of the linkage disequilibrium between it and its’ surrounding sites. The disequilibrium was found to extend over a >25kB region around the gene, and within this region it was noted that the genomes of the susceptible chromosomes varied much more than those with the resistant chromosome. This implies that while the Doc1420 inserted into the genome long ago, it just recently began to proliferate in frequency.
Thus, Magwire et al. propose that there has been two separate polymorphisms that have bestowed D.melanogaster with an increasing level of resistance to the Sigma virus, which strongly illustrates the selective pressure to adapt within specific host-parasite relationships. The ancestral state of the gene was truncated by the insertion of the Doc1420 element into one of its’ exons. Typically speaking, such an event would be disastrous, because to alter the actual coding region of a gene usually means disrupting the function of the gene itself. This time, however, the insertion of the TE produced a new protein which provides significant resistance to a specific pathogen; this is very rare. This event took place around 90,000 years ago, but just recently (within the last 25-240 years) began to experience selective pressure to proliferate in the population to the point that now the majority of the flies in the world today have it . The second event involved a complex rearrangement of the CHKov1 and CHKov2 region which consisted of two duplications of these genes. This second polymorphism is much more rare in the wild currently, but is likely to proliferate as it imparts the highest degree of Sigma virus resistance and therefore maximizes the fitness of those flies which have it. The idea was that the duplication events increase the expression of the the CHKov1 and CHKov2 genes, but it was found that only CHKov2 expressed an elevated level of expression after injection with the Sigma virus within those that underwent the rearrangement, so it is more likely that the duplications are benefiting the flies by amplifying the dosage of CHKov2 while the duplications of CHKov1 occurred in tandem due to their close proximity to CHKov2. D.melanogaster gained a moderate amount of resistance to the Sigma virus due to a TE insertion, then gained a high degree of resistance later on through gene duplication events. This shows that changes to the genome over time can cause significant alterations in the level of resistance a host has for a specific parasite, be it through the insertion of transposable elements or through the rearrangement and duplication of genes.
Magwire et al. set out in their study with the intention of better understanding host-parasite relationships in animals. By choosing to study the Sigma/D.melanogaster relationship they were able to simplify their ability to study such phenomena, since D.melanogaster is an optimal genetic test subject and Sigma expresses specificity only to D.melanogaser. This specific relationship was anticipated to have produced powerful selective pressure on the flies to adapt resistance to the Sigma virus because the virus paralyzes those infected and therefore significantly reduces fitness, and indeed that is what was observed in their observations. The researchers arrived at data that contradicted a prevailing view of the time: that the development of pathogenic resistance comes at a high price . Rather than coming at a cost, it appears that the gain of a TE insertion into the coding region of a gene resulted in increased pathogenic resistance, and pleiotropically pre-adapted the flies to better sustain exposure to insecticides in a rare, fortuitous fashion.
It should be noted that at the time of this paper’s publication the molecular mechanisms for how, exactly, the truncation of CHKov1 by Doc1420 and the subsequent complex rearrangement that the region underwent provides resistance to the Sigma virus and resistance to insecticides was unknown. Magwire et al.’s conjecture is that since CHKov1 contains a choline kinase domain, the truncation might affect acetylcholine esterase. Since Sigma is a Rhabdovirus, and Rhabdoviruses use acetylcholine receptors to enter the cell, this alteration might be inhibiting the virus’ ability to enter the host’s cells. Subsequently, this affect on acetylcholine esterase could also increase resistance to insecticides because organophosphates target acetylcholine esterase, providing a lucky pre-established means for resisting modern pesticide use on crops.
Back in 2005 Aminetzach et al. discovered that the insertion of Doc1420 into the second exon of CHKov1 produced a functional protein, presenting evidence against the common view that the truncation of coding regions results in decreased fitness . This new protein coded for by the altered CHKov1 gene proved to imbue its’ carriers with increased resistance to insecticides, but was also found to have been inserted 90,000 years ago. This was well before human development of agriculture, let alone the practice of spraying crops with organophosphates. Thus, this selective pressure can only be used to account for the recent proliferation of the resistant allele within the last 25-240 years, not to account for how it became set in a proportion of the D.melanogaster population in the first place. This question stumped Aminetzach et al., who could only guess that the insertion was maintained due to some other benefit to the fitness of the flies . In this paper, Magwire et al. have presented a likely solution to this question by discovering that the insertion of Doc1420 into the gene originally provided carriers with significantly more resistance to a host-specific virus, and this adaptation conveniently provided its carriers with unexpected resistance to common modern agricultural insecticides. This data indicates that specific host-parasite relationships in animals can cause rapid adaptation to occur in order to maintain a status-quo of fitness, and that not all adaptations to pathogenic interactions come at a cost.
1.) Yaul T. Aminetzach et al. (2005). “Pesticide Resistance via Transposon-Mediated Adaptive Gene Truncation in Drosophila”. Science 309, 764. DOI:10.1126/science.1112699.
2.) Magwire MM, Bayer F, Webster CL, Cau C, Jiggins FM (2011). “Successive Increases in the Resistance of Drosophila to Viral Infection through a Transposon Insertion Followed by a Duplication”. PLoS Genet 7(10):e1002337.doi:10.1371/journal.pgen.1002337.
3.) Gay P (1978). “Drosophila Genes Which Intervene in Multiplication of Sigma Virus”. Molecular & General Genetics 159:269-283.
4.) Kraaijeveld AR, Godfray HCJ (1997) “Trade-off between parasitoid resistance and larval competitive ability in D.melanogaster”. Nature 389:278-280.
5.) Mckean K, Yourth C, Lazzare B, Clark A (2008) “The evolutionary costs of immunological maintenance and deployment”. BMC Evolutionary Biology 8:76.