Limits...
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.

Kuester A, Chang SM, Baucom RS - Evol Appl (2015)

Bottom Line: We uncovered a mosaic pattern of resistance across the landscape, with some populations exhibiting high-survival postherbicide and other populations showing high death.SSR genotyping revealed little evidence of isolation by distance and very little neutral genetic structure associated with geography.An approximate Bayesian computation (ABC) analysis uncovered evidence for migration and admixture among populations before the widespread use of glyphosate rather than the very recent contemporary gene flow.

View Article: PubMed Central - PubMed

Affiliation: Department of Ecology and Evolutionary Biology, 830 North University, University of Michigan Ann Arbor, MI, USA.

ABSTRACT
Strong human-mediated selection via herbicide application in agroecosystems has repeatedly led to the evolution of resistance in weedy plants. Although resistance can occur among separate populations of a species across the landscape, the spatial scale of resistance in many weeds is often left unexamined. We assessed the potential that resistance to the herbicide glyphosate in the agricultural weed Ipomoea purpurea has evolved independently multiple times across its North American range. We examined both adaptive and neutral genetic variations in 44 populations of I. purpurea by pairing a replicated dose-response greenhouse experiment with SSR genotyping of experimental individuals. We uncovered a mosaic pattern of resistance across the landscape, with some populations exhibiting high-survival postherbicide and other populations showing high death. SSR genotyping revealed little evidence of isolation by distance and very little neutral genetic structure associated with geography. An approximate Bayesian computation (ABC) analysis uncovered evidence for migration and admixture among populations before the widespread use of glyphosate rather than the very recent contemporary gene flow. The pattern of adaptive and neutral genetic variations indicates that resistance in this mixed-mating weed species appears to have evolved in independent hotspots rather than through transmission of resistance alleles across the landscape.

No MeSH data available.


Related in: MedlinePlus

Three simple scenarios used in the ABC analyses of Ipomoea purpurea populations sampled from NC and TN. Four populations each trial were used to model (A) no gene flow among populations, (B) admixture (denoted r) that occurred before the widespread use of the herbicide, and (C) admixture after widespread glyphosate use. The origin of the populations occurred at t1, or the time at which the species was identified as an agricultural weed in the United States, taken from (Defelice 2001). Population divergence between TN and NC populations is indicated by t2. Widespread glyphosate use is indicated in t3 and denoted with a dotted line, corresponding to the year (1974) at which glyphosate was released for commercial use in US agriculture. Admixture events are indicated either with t4 (gene flow prior to glyphosate use) or t5 (gene flow after glyphosate use). Also included in the model were the effective population size of each population at t2 (Ne) and subsequent effective size postglyphosate bottleneck (N1). Parameter estimates are shown in Table S4.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4561571&req=5

fig02: Three simple scenarios used in the ABC analyses of Ipomoea purpurea populations sampled from NC and TN. Four populations each trial were used to model (A) no gene flow among populations, (B) admixture (denoted r) that occurred before the widespread use of the herbicide, and (C) admixture after widespread glyphosate use. The origin of the populations occurred at t1, or the time at which the species was identified as an agricultural weed in the United States, taken from (Defelice 2001). Population divergence between TN and NC populations is indicated by t2. Widespread glyphosate use is indicated in t3 and denoted with a dotted line, corresponding to the year (1974) at which glyphosate was released for commercial use in US agriculture. Admixture events are indicated either with t4 (gene flow prior to glyphosate use) or t5 (gene flow after glyphosate use). Also included in the model were the effective population size of each population at t2 (Ne) and subsequent effective size postglyphosate bottleneck (N1). Parameter estimates are shown in Table S4.

Mentions: We next used a Bayesian coalescent approach (approximate Bayesian computation; Beaumont 2010) to examine the likelihood that migration occurred recently between resistant populations in the southeastern USA (see Results) compared to a scenario of gene flow between populations prior to the widespread use of glyphosate across the landscape. If the former scenario were more likely, we would reason that contemporary gene flow (such as through the movement of contaminating morning glory seed between farms in seed lots) is likely to be responsible for resistance across populations. If, however, the latter scenario of migration before the widespread use of glyphosate were the more likely one, we would infer that resistance has independently evolved in separate populations. Although ABC analysis can be used to make inferences about complex population histories, estimate population parameters such as effective population size (Tallmon et al. 2008), and has recently been used to model many different scenarios of herbicide resistance evolution (Okada et al. 2013), we elected to model the relatively simple alternative scenarios of migration between I. purpurea populations pre- or postglyphosate use. We employed the software DIYABC v 2 (Cornuet et al. 2008) to test three scenarios (Fig. 2) using the microsatellite data from North Carolina and Tennessee populations—two areas of the landscape where we observed the highest resistance (see Results). The first scenario assumed no admixture across populations. The second scenario assumed admixture before the use of glyphosate (at time t4) and a third scenario assumed admixture after the use of glyphosate (time t5) within NC and TN regions.


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.

Kuester A, Chang SM, Baucom RS - Evol Appl (2015)

Three simple scenarios used in the ABC analyses of Ipomoea purpurea populations sampled from NC and TN. Four populations each trial were used to model (A) no gene flow among populations, (B) admixture (denoted r) that occurred before the widespread use of the herbicide, and (C) admixture after widespread glyphosate use. The origin of the populations occurred at t1, or the time at which the species was identified as an agricultural weed in the United States, taken from (Defelice 2001). Population divergence between TN and NC populations is indicated by t2. Widespread glyphosate use is indicated in t3 and denoted with a dotted line, corresponding to the year (1974) at which glyphosate was released for commercial use in US agriculture. Admixture events are indicated either with t4 (gene flow prior to glyphosate use) or t5 (gene flow after glyphosate use). Also included in the model were the effective population size of each population at t2 (Ne) and subsequent effective size postglyphosate bottleneck (N1). Parameter estimates are shown in Table S4.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4561571&req=5

fig02: Three simple scenarios used in the ABC analyses of Ipomoea purpurea populations sampled from NC and TN. Four populations each trial were used to model (A) no gene flow among populations, (B) admixture (denoted r) that occurred before the widespread use of the herbicide, and (C) admixture after widespread glyphosate use. The origin of the populations occurred at t1, or the time at which the species was identified as an agricultural weed in the United States, taken from (Defelice 2001). Population divergence between TN and NC populations is indicated by t2. Widespread glyphosate use is indicated in t3 and denoted with a dotted line, corresponding to the year (1974) at which glyphosate was released for commercial use in US agriculture. Admixture events are indicated either with t4 (gene flow prior to glyphosate use) or t5 (gene flow after glyphosate use). Also included in the model were the effective population size of each population at t2 (Ne) and subsequent effective size postglyphosate bottleneck (N1). Parameter estimates are shown in Table S4.
Mentions: We next used a Bayesian coalescent approach (approximate Bayesian computation; Beaumont 2010) to examine the likelihood that migration occurred recently between resistant populations in the southeastern USA (see Results) compared to a scenario of gene flow between populations prior to the widespread use of glyphosate across the landscape. If the former scenario were more likely, we would reason that contemporary gene flow (such as through the movement of contaminating morning glory seed between farms in seed lots) is likely to be responsible for resistance across populations. If, however, the latter scenario of migration before the widespread use of glyphosate were the more likely one, we would infer that resistance has independently evolved in separate populations. Although ABC analysis can be used to make inferences about complex population histories, estimate population parameters such as effective population size (Tallmon et al. 2008), and has recently been used to model many different scenarios of herbicide resistance evolution (Okada et al. 2013), we elected to model the relatively simple alternative scenarios of migration between I. purpurea populations pre- or postglyphosate use. We employed the software DIYABC v 2 (Cornuet et al. 2008) to test three scenarios (Fig. 2) using the microsatellite data from North Carolina and Tennessee populations—two areas of the landscape where we observed the highest resistance (see Results). The first scenario assumed no admixture across populations. The second scenario assumed admixture before the use of glyphosate (at time t4) and a third scenario assumed admixture after the use of glyphosate (time t5) within NC and TN regions.

Bottom Line: We uncovered a mosaic pattern of resistance across the landscape, with some populations exhibiting high-survival postherbicide and other populations showing high death.SSR genotyping revealed little evidence of isolation by distance and very little neutral genetic structure associated with geography.An approximate Bayesian computation (ABC) analysis uncovered evidence for migration and admixture among populations before the widespread use of glyphosate rather than the very recent contemporary gene flow.

View Article: PubMed Central - PubMed

Affiliation: Department of Ecology and Evolutionary Biology, 830 North University, University of Michigan Ann Arbor, MI, USA.

ABSTRACT
Strong human-mediated selection via herbicide application in agroecosystems has repeatedly led to the evolution of resistance in weedy plants. Although resistance can occur among separate populations of a species across the landscape, the spatial scale of resistance in many weeds is often left unexamined. We assessed the potential that resistance to the herbicide glyphosate in the agricultural weed Ipomoea purpurea has evolved independently multiple times across its North American range. We examined both adaptive and neutral genetic variations in 44 populations of I. purpurea by pairing a replicated dose-response greenhouse experiment with SSR genotyping of experimental individuals. We uncovered a mosaic pattern of resistance across the landscape, with some populations exhibiting high-survival postherbicide and other populations showing high death. SSR genotyping revealed little evidence of isolation by distance and very little neutral genetic structure associated with geography. An approximate Bayesian computation (ABC) analysis uncovered evidence for migration and admixture among populations before the widespread use of glyphosate rather than the very recent contemporary gene flow. The pattern of adaptive and neutral genetic variations indicates that resistance in this mixed-mating weed species appears to have evolved in independent hotspots rather than through transmission of resistance alleles across the landscape.

No MeSH data available.


Related in: MedlinePlus