The rapid and recurrent evolution of herbicide resistance in agricultural weeds is one of the most striking examples of evolution via natural selection. However, the evolution of herbicide resistance also represents one of the most significant constraints in designing sustainable and effective strategies for weed control. Of particular concern is the recent evolution of resistance in Palmer Amaranth (Amaranthus palmeri) to several herbicides, most notably, glyphosate (Roundup). Prior research has shown that glyphosate resistance is due to an increase in the number of copies of the 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS) gene. Recently, A. palmeri has been found in multiple locations in Ohio where it is not native, and while this is problematic for producers, it also presents a unique opportunity to investigate how adaptive evolution (i.e. evolved glyphosate resistance) and demographic factors influence range expansion and invasive success. In this study we used a combination of dd-RAD sequencing, and q-PCR to 1) identify sources and pathways via which A. palmeri was introduced into Ohio, 2), compare levels of glyphosate resistance and EPSPs copy number variation across the current geographic range of A. palmeri and 3) estimate the frequency that glyphosate resistance has evolved in A. palmeri.
We found that Ohio populations of A. palmeri, were introduced from multiple sources and most likely from Mississippi and Alabama and not from neighboring states. This indicates that as of now, recurrent introductions from out of state are most likely driving the spread of A. palmeri in Ohio. EPSPs copy number varied considerably across the geographic range of A. palmeri, with southern populations having on average the highest number of EPSPs copies (range 2 to 170 copies). Interestingly, Ohio populations had much lower EPSPs copy number than their putative source populations (range 1-10). Similar to prior studies, we found that increased EPSPs copy number was positively correlated with glyphosate resistance. However, we also found that glyphosate resistance could be achieved with relatively few (10-30) EPSPs copies. Our data also indicates that increase in EPSPs copy number occurred de novo in multiple locations – individuals clustered based on geography and not EPSPs copy number or glyphosate resistance. The pattern in variation in copy number is suggestive that there may be a cost to being resistant to glyphosate. In conclusion, our results suggest that demographic factors other than glyphosate resistance are contributing to the spread of A. palmeri in Ohio.