The Influence of Multiple Herbicide Resistance on Growth And

Iowa State University Capstones, Theses and Graduate Theses and Dissertations Dissertations

2018 The influence of multiple herbicide resistance on growth and development of Amaranthus tuberculatus and the efficacy of the very long chain fatty acid-inhibiting herbicides on multiple herbicide-resistant Amaranthus tuberculatus; Investigating a putative 4-hydroxylphenylpyruvate dioxygenase-inhibiti Eric Alexander Lyle Jones Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/etd Part of the Agricultural Science Commons, Agriculture Commons, and the Agronomy and Crop Sciences Commons

Recommended Citation Jones, Eric Alexander Lyle, "The influence of multiple herbicide resistance on growth and development of Amaranthus tuberculatus and the efficacy of the very long chain fatty acid-inhibiting herbicides on multiple herbicide-resistant Amaranthus tuberculatus; Investigating a putative 4-hydroxylphenylpyruvate dioxygenase-inhibiti" (2018). Graduate Theses and Dissertations. 16386. https://lib.dr.iastate.edu/etd/16386

This Thesis is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected].

The influence of multiple herbicide resistance on growth and development of Amaranthus tuberculatus and the efficacy of the very long chain fatty acid-inhibiting herbicides on multiple herbicide-resistant Amaranthus tuberculatus; Investigating a putative 4-hydroxylphenylpyruvate dioxygenase-inhibiting herbicide–resistant Ambrosia trifida population

Eric Alexander Lyle Jones

A thesis submitted to the graduate faculty in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE

Major: Agronomy (Weed Science)

Program of Study Committee: Micheal D.K. Owen, Co-Major Professor Robert G Hartzler, Co-Major Professor Allen D. Knapp The student author, whose presentation of the scholarship herein was approved by the program of study committee, is solely responsible for the content of this thesis. The Graduate College will ensure this thesis is globally accessible and will not permit alterations after a degree is conferred. Iowa State University Ames, Iowa 2018

DEDICATION

I dedicate my thesis to Gary Park, Robert Park, and Turner Jones. I love and greatly miss you three. While I may not have adopted the “outlaw” lifestyle as most Park and Jones men are destined to, I hope you all are looking down upon me with smiling faces. You all shaped me into the person I am today. The many fond memories I have with you three will continue to bring me joy. You all left too early and I am eager until we meet again.

iii

TABLE OF CONTENTS Page ACKNOWLEDGEMENTS iv ABSTRACT vi CHAPTER 1 GENERAL INTRODUCTION AND LITERAURE REVIEW 1 CHAPTER 2 INFLUENCE OF MULTIPLE HERBICIDE RESISTANCE ON GROWTH AND DEVELOPMENT IN AMARANTHUS TUBERCULATUS 12 CHAPTER 3 INVESTIGATING EFFICACY OF SELECTED VERY LONG CHAIN FATTY ACID-INHIBITING HERBICIDES ON AMARANTHUS TUBERCULATUS POPULATIONS WITH EVOLVED MULTIPLE HERBICIDE RESISTANCES 36 CHAPTER 4 INVESTIGATING A PUTATIVE HPPD-RESISTANT AMBROSIA TRIFIDA POPULATION 56 CHAPTER 5 GENERAL CONCLUSIONS 72 BIBLIOGRAPHY 75

ACKNOWLEDGEMENTS

Foremost, I would like to address my Mother, Leah Jones, for providing me a life that only the finest people deserve. From cleaning my wrestling clothes in years past to helping me figuring out the “real world”, my Mother has been my number one consultant. Secondly, I would like to address my father, Terry Jones. My father taught me how to have a strong work ethic and to never half-heartedly finish a job and be satisfied with the outcome. I am forever grateful for my parent’s guidance and love during my presently short journey of life.

Next, I would like to thank my committee consisting of Drs. Micheal Owen, Robert Hartzler,

Allen Knapp, and John Hinz. I will be forever grateful to be the last “gradual” student of Dr.

Owen’s prestigious lab group. I was granted the freedom to study my specific interests in weed science. I will always appreciate the guidance you offered and the challenge of your hands-off approach. It has also been a pleasure to share a similar lineage, interests, and hobbies with my major professor.

I would like to thank Dr. Hartzler for challenging my understanding of weed science topics and allowing me to assist in teaching his famed weed identification class (Agronomy 217).

While I may not have been the greatest teaching assistant, I would like to say it was a great opportunity to teach future agronomists a valuable skill. It should be mentioned our weekly chats (discussion, civil arguments, etc.) were never taken in vain; you are a great teacher, mentor, and “nozzle head” (despite your many notions that you are indeed not!).

Dr. Knapp provided me with my basic understanding of plant physiology (Agronomy 316), which did not seem basic at the time. Prior knowledge of the two photosystems, carotenoids, and plastoquinones has made learning herbicide physiology easier than expected. Your helpful comments and insights were invaluable to writing this thesis and journal manuscripts.

I would like to thank Dr. John Hinz for exposing me to the addictive subject of “weed science”. What may have been an internship to another Iowa State University student became my lifelong passion. I am glad you decided to give this “city slicker” a chance to prove himself! I would like to extend my thanks to Angela Rieck-Hinz as well for contributing to improve my interest in agronomy and weed science.

Finally, I would like to acknowledge Damian (Damos) Franzenburg, James (Jimmi) Lee,

Iththiphonh (Alex) Macvilay, James (Lux-um-burger) Lux, and Daniel (Dandy Danielson)

Kohlhase. I am certain without the help and guidance from all of you I would not have made it to this point. It is doubtful I will ever be able to work with another group that will be able to rival the fun and memories I established with you all.

ABSTRACT Iowa farmers rely mainly on herbicides for weed management. Inconsistent and unsatisfactory weed control is being realized as the spread of herbicide-resistant weed populations has increased. Populations of waterhemp (Amaranthus tuberculatus (Moq.) J.D.

Sauer) and giant ragweed (Ambrosia trifida L.) have evolved resistance to glyphosate and other herbicides commonly used in crop fields. This research focuses on the growth and development of multiple herbicide-resistant waterhemp and assessing the evolution of new herbicide-resistant waterhemp and giant ragweed populations.

Significant differences in growth, flowering, accumulated biomass, and seed production were detected when the multiple herbicide-resistant waterhemp populations were compared to a herbicide-susceptible waterhemp population. While statistically significant differences were detected, the small differences are not likely to select herbicide-susceptible waterhemp populations over MHR waterhemp populations. Thus, it can be concluded that plants with multiple herbicide resistances are not likely incurring a fitness penalty and may remain in the agroecosystem.

Currently, very long chain fatty acid- and 4-hydroxyphenylpyruvate dioxygenase (HPPD, EC

1.13.11.27)-inhibiting herbicides are still efficacious on many waterhemp and giant ragweed populations, respectively. Since herbicides impart such a large selection pressure on weed populations, the recurrent use of specific herbicides may decrease efficacy longevity as only herbicide-resistant individuals will remain.

CHAPTER 1. GENERAL INTRODUCTION

Introduction Weeds are plants with the ability to colonize areas disturbed by man, a consequence of plant domestication (Harlan 1992, Dekker 2011). Every control tactic that has been employed to eradicate weeds has failed (Gressel et al. 2016). Herbicides, once considered the end-all-be- all weed control tools, are becoming less efficacious as the evolution of herbicide-resistant weeds increases. In addition to selecting for herbicide-resistant weeds, herbicides have caused shifts in weed population demographics to more resilient weed species (Keeley 1987).

If farmers want to continue to manage weeds successfully, basic weed biology must be learned (Swanton and Weise 1991). An integrated weed management (IWM) approach will consist of “many little hammers” rather than “one big hammer”, meaning many less efficacious tactics used together will provide high efficacy with relatively low selection pressure (Liebman and Gallandt 1997). Little hammer tactics, including tillage, row spacing, cover crops, and utilization of grainivores, all rely on knowledge of weed biology in order to kill weeds efficiently (Liebman and Gallandt 1997, Harrison et al. 2003).

Herbicide Resistance Besides tillage, herbicides provide the highest selection pressure for evolution in agricultural weeds (Gressel 2009, Zelaya and Owen 2005). Select individuals within the population possess natural mutations that allow the plant to survive a lethal herbicide dose. When the herbicide is applied, herbicide-susceptible individuals will die and herbicide-resistant individuals will produce seeds and add to the soil seed bank. Recurrent application of the same herbicide(s) will likely lead to the evolution of a herbicide-resistant weed population. 2

Weeds with multiple herbicide resistances are now considered common instead of rare (Heap

2018). Weeds such as Lolium rigidum Gaud. (rigid ryegrass), Echinochloa crus-galli (L.) P.

Beauv. (barnyardgrass), Amaranthus tuberculatus (Moq.) J.D. Sauer (waterhemp), and

Ambrosia artemisiifolia L. (common ragweed) have evolved resistance to 11, 9, 6, and 5 unrelated herbicide groups, respectively (Heap 2018). Farmers are running out of viable chemical means to control weeds.

Herbicide resistance is facilitated by two main mechanisms, target-site (TS) and non-target- site (NTS). Target-site resistance is an alteration of the enzyme such that the herbicide molecule can no longer bind or inhibit the enzyme or protein function (Délye et al. 2013).

Another mechanism of TS resistance is when an overabundance of the target enzyme is produced which ultimately dilutes the effect of the herbicide (Powles and Yu 2010). Non- target-site resistance inhibits any phytotoxicity imposed by the herbicide on the plant before it reaches the intended target site. Mechanisms of NTS resistance include reduced herbicide uptake or translocation, and increased herbicide metabolism (Yuan et al. 2006). Herbicide metabolism is of great interest as cross-resistance to unrelated herbicides may result from this

NTS resistance mechanism. Xenobiotic metabolism enzymes, specifically cytochrome P450 mixed function oxidases and glutathione-S transferases, are a plentiful and diverse family of enzymes within plants (Yuan et al. 2006). Thus, it is very possible that such an enzyme could detoxify many xenobiotics non-selectively. Ultimately, this could render all herbicides useless.

Resistance to xenobiotics should come at a cost to the organism, or all organisms, in theory, would be xenobiotic-resistant (Coustau et al. 2000). Mutations in receptors and enzymes might alter protein function or divert valuable resources for essential plant functions 3

(Coustau et al. 2000, Vilva-Auib et al. 2009). If the selection agent such as an herbicide is not present, the resistant individual is not at an advantage compared to the susceptible individual thus making the cost of resistance visible. Research provides evidence that some herbicide resistance alleles cause significant measureable fitness costs, while others do not

(Vila-Auib et al. 2009, Jordan 1996, Wu et al. 2017). However, compensatory evolution may mask putative fitness penalties caused by herbicide resistance mutation(s) by manipulating background genetics (Darmency et al. 2014).

Knowledge of multiple herbicide resistance and incurred fitness penalties will help understand the longevity of multiple herbicide-resistant (MHR) waterhemp populations in the field. TS triazine-resistant waterhemp was reported to be less fit to compete with the triazine- susceptible waterhemp (Soltani 2008). Rigid ryegrass and Bromus tectorum L. (downy brome) populations that exhibit multiple herbicide resistance were reported to be less fit to compete with the herbicide-susceptible populations of their respective weed species (Vila-

Auib et al. 2005, Park and Mallory-Smith 2005). Simply removing herbicide applications will not favor the herbicide-susceptible weed populations if herbicide-resistant weed populations is not at an ecological disadvantage.

Waterhemp Biology Amaranthus tuberculatus (Moq.) J.D. Sauer (tall waterhemp) and Amaranthus rudis J.D.

Sauer (common waterhemp) are native to the Midwest Corn Belt region. Common waterhemp was found west of the Mississippi River while tall waterhemp was found east of the Mississippi River (Steckel 2007). Botanists have come to the consensus that tall and common waterhemp have interbred into one polymorphic species referred to as Amaranthus tuberculatus, tall waterhemp, or more simply “waterhemp” (Pratt and Clark 2001). 4

Distinguishing phenotypic characteristics include; glaborous stems and leaves, lancelolate leaves, and a spikelet flower head (Sauer 1955). Waterhemp is a summer annual plant found in riparian areas, historically, but has successfully invaded crop fields starting in the 1980’s

(Sauer 1955, Steckel 2007). Waterhemp is a dioecious species with male and female flowers on separate plants (Sauer 1955) and thus must cross-pollinate which increases opportunity for genetically diverse offspring. Cross-pollination is advantageous for dioecious species as it ensures that favorable genotypes/phenotypes are passed on to future generations (Bram and

Quinn 2000) in part explains why herbicide resistance(s) mutations rapidly spread within a waterhemp population (Primack et al. 1989).

Waterhemp can grow to 3.7 m and produce over 1 million seeds in noncompetitive environments (Steckel 2007). The height is reduced to 1.5 m and seed production is reduced to 30 to 300 thousand seeds under competitive conditions (Steckel 2007, Nordby and

Hartzler 2004, Hartzler et al. 2004). Waterhemp expresses extended germination and emergence throughout the growing season (Hartzler et al. 1999). These characteristics help waterhemp successfully establish in agroecosystems.

Initially, waterhemp was susceptible to acetolactate synthase -inhibiting herbicides (ALS, EC

2.2.1.6), photosystem II-inhibiting herbicides (EC 1.10.3.9), and 5-enolpyruvyl-shikimate-3- phosphate synthase-inhibiting herbicides (EPSPS, EC 2.5.1.19) (Hinz and Owen 1997,

Anderson 1994, Zelaya and Owen 2000). Waterhemp has evolved resistance to these efficacious herbicides, leaving less efficacious herbicides to rely on for control (Heap 2018,

Bell et al. 2013). Understanding waterhemp biology is important to the success of the different methods of control. While individual tactics of an integrated weed management

(IWM) program will not be as efficacious as a herbicide, a multitude of IWM tactics can 5 provide excellent weed control (Swanton and Weise 1991). The basis of IWM depends greatly on weed biology. Tactics need to be executed at the appropriate time to ensure successful weed kill before viable seed is added back to the soil seed bank (Bhowmik 1997).

VLCFA-Inhibiting Herbicides Very long chain fatty acids (VLCFA) are fatty acids with acyl chains of 18 carbons or more

(Bach and Faure 2010). VLCFAs are essential for many plant structures and functions including signaling molecules, triacylglycerol, and cuticular waxes (Bach and Faure 2010,

Halsman and Kunst 2013). The lack of VLCFAs would be phytotoxic to the plant (Boger

2003). ). These herbicides inhibit the condensing enzymes of the elongation process that ultimately inhibits the production of VLCFAs (Boger 2003). The lack of VLCFAs prevents germination or early development of susceptible plants (Boger et al. 2000, Fuerst 1987).

The first VLCFA-inhibiting herbicides were introduced in the early 1960’s and are still being applied extensively to crop fields (Boger et al. 2000). Current VLCFA-inhibiting herbicides include s-metolachlor, acetochlor, pyroxasulfone, and dimethenamid. VLCFA-inhibiting herbicides are applied to primarily control grasses; however, there is efficacy on select small- seeded broadleaf weeds (Fuerst 1987, Boger 2003).

Despite their long-term use, there has only been one confirmed case of resistance to VLCFA- inhibiting herbicides in the United States (Heap 2018). Thus, this class of herbicides is increasingly relied on for controlling some herbicide-resistant weeds, including rigid ryegrass, waterhemp, and Amaranthus palmeri S. Wats. (Palmer amaranth) (Busi et al. 2012,

Kohrt and Sprague 2017). Since there has been no confirmation of waterhemp populations with evolved resistance to the VLCFA-inhibiting herbicides, these products are applied as 6 much as two times a growing season to provide residual waterhemp control in Zea mays L.

(corn) and Glycine max (L.) Merr. (soybean) (Schryver et al. 2017).

Understanding if multiple herbicide-resistant (MHR) waterhemp populations are still susceptible to the VLCFA-inhibiting herbicides will determine if these herbicides remain a viable weed control tool. Given that NTS resistance mechanisms may result in resistance to unrelated herbicides, if such mechanisms are present in multiple herbicide-resistant waterhemp, all herbicides could be rendered useless including the VLCFA-inhibiting herbicides (Busi and Powles 2016, Tranel et al. 2007). While evolution of resistance to the

VLCFA-inhibiting herbicides is not widespread despite recurrent applications for 50 years, it does not mean that resistance will not evolve (Gressel 2009).

Giant Ragweed Biology

Ambrosia trifida L. (giant ragweed) is a summer annual native to North America, inhabiting disturbed landscapes (Abul-Fatih and Bazzaz 1979). Distinguishing phenotypic characteristics include large circular cotyledons, un-lobed first true leaves, three to five lobed leaves after the first true leaves, and rough pubescence covering the entire plant (Bassett and

Crompton 1982). Giant ragweed is a monoecious plant with both male and female flowers on the same plant. The male flowers are found at the apical meristems, while the female flowers are found at the leaf axils (Bassett and Crompton 1982). The advantage of being monoecious is the plant does not need to cross-pollinate which ensures successful reproduction by just one plant. The disadvantage, however, is that genetic recombination potential is low since cross-pollination is not obligatory. However, giant ragweed populations have relatively high genetic diversity since plant pollen can beis wind-dispersed resulting in cross-pollination

(Brabham et al. 2011). 7

Giant ragweed was not an agronomic weed of importance until relatively recently (Regnier et al. 2016). Giant ragweed emerges in early March until early July and produces relatively low amounts of seeds per plant (Harrison et al. 2001). Early weed emergence in summer annual crops makes control more difficult as the weeds are often established before crop emergence

(Harrison et al. 2001). Giant ragweed heights can reach six meters and plants usually grow above the crop canopy (Abul-Fatih and Bazzaz 1979, Jurik 1991). Giant ragweed fecundity under noncompetitive environments is 5,000 seeds plant-1 and is reduced to 17 seeds plant-1 under competitive environments (Baysinger and Sims 1991). Giant ragweed seeds generally last in the soil seed bank for four years or less (Goplen 2015, Harrison et al. 2001, Harrison et al. 2003). Rodents, insects, and earthworms may consume approximately 50% of the seed produced annually (Abul-Faith and Bazzaz 1979, Harrison et al. 2001, Harrison et al. 2003).

While giant ragweed plants are very competitive and possess herbicide resistance traits, populations can be successfully managed by controlling the amount of seeds that replenish the soil seed bank. Herbicides are effective for weed control, however over-reliance will inevitability result in the evolution of herbicide resistance (Gressel and Segel 1978). Future evolved herbicide resistance(s) in giant ragweed populations will increase the complexity of control. Giant ragweed biology is well documented, thus the establishment of a successful

IWM program should not be difficult (Bhowmik 1997).

HPPD-Inhibiting Herbicides

4-hydroxyphenylpyruvate dioxygenase-inhibiting herbicides (HPPD) (EC 1.13.11.27) represent the most recent herbicide mechanism of action to be discovered and offered 8 commercially (Mitchell et al. 2001). HPPD-inhibiting herbicides have been applied recurrently in cereal crops to control glyphosate-resistant weeds (McMullan and Green

2011). At this time resistance to HPPD-inhibiting herbicides is not yet as widespread as other herbicides, and resistance has only been confirmed in waterhemp and Palmer amaranth

(Owen et al. 2015, Heap 2018). HPPD-inhibiting herbicides are used more frequently in the

Midwest compared to other parts of the United States which can explain why most of the

HPPD-resistant weeds are concentrated in Iowa, Illinois, and Nebraska (Gressel et al. 2016,

Heap 2018).

HPPD-inhibiting herbicides disrupt the biosynthesis of phenylquinones, including plastiquinones (Matringe et al. 2005). Plastiquiones are electron acceptors in photosystem II and also involved in carotenoid biosynthesis. Carotenoids serve as an accessory light- harvesting pigment and provide protection of scavenging reactive oxygen species formed in the chloroplast (Young 1991). Plastiquione and carotenoid biosynthesis inhibition result in the accumulation of high-energy reactive oxygen species due to the lack of photon processing and the accumulation of these reactive species will cause photo-oxidation leading to cell death (Reinboth et al. 1996). The symptomology of the HPPD-inhibiting herbicide application to sensitive plants is bleaching of the apical meristem tissue due to pigment damage (Ma et al. 2013).

While only two weeds species have evolved resistance to this herbicide group, the evolution of resistance will certainly occur in other agronomic weeds such as giant ragweed and

Conyza canadensis (L.) Cronquist (horseweed). Currently some waterhemp populations resistant to HPPD-inhibiting herbicides have multiple resistance to glyphosate, ALS-, and

PSII-inhibiting herbicides (Hausman et al. 2011, Green and McMullan 2011, Owen 2013). 9

Summary and Research Objectives Man has been battling weeds since the first plants were domesticated for consumption.

Organisms that are subjected to selection pressures will evolve in response to these pressures

(Darwin 1859). Recently, herbicides have been the primary selection agents to drive evolution in agricultural weeds (Owen and Zelaya 2005). The widespread distribution of herbicide-resistant weeds, such as waterhemp and giant ragweed, provide sufficient evidence of this phenomenon. Despite the distribution and frequency of herbicide-resistant weeds globally, herbicide usage has not decreased (Heap 2018, Kniss 2017). Weed management approaches that rely on only one tactic will eventually provide unsatisfactory results.

Herbicide efficacy is an exhaustible resource due to the high selection pressure imposed by the pesticide for resistant individuals in weed populations (Davis and Frisvold 2017).

Herbicide resistance mutations could come at a fitness cost to plants and decrease the likelihood of the weed successfully producing offspring. Increasing the number of herbicide resistance mechanisms possessed by a weed species could result in an increase of the incurred fitness penalty. If MHR waterhemp exhibit substantial fitness penalties attributable to herbicide resistance mechanisms, management practices could be manipulated to select for herbicide-susceptible waterhemp populations rather than MHR waterhemp populations.

Since the evolution of resistance to the VLCFA-inhibiting herbicides has not been confirmed in waterhemp, usage of these herbicides may increase to control herbicide-resistant waterhemp populations. However, if MHR waterhemp populations express cross-resistance to unrelated herbicides, the VLCFA-inhibiting herbicides could be detoxified without an altered target site. 10

Evolution of resistance to the HPPD-inhibiting herbicides is currently limited to two

Amaranthus species. The increased usage of these herbicides in recurrent corn rotation likely selected resistant individuals in the field. When a report of giant ragweed not being controlled by an HPPD-inhibiting herbicide arose, there was no reasonable doubt that such a population could not exist.

The objectives of this research are:

1) Determine if herbicide-resistant waterhemp have a fitness penalty compared to

herbicide-susceptible waterhemp and grow and develop at a different rate and if

multiple herbicide resistances increase the fitness costs.

2) Determine the VLCFA-inhibiting herbicide efficacy of several MHR waterhemp

populations and if VLCFA-inhibiting herbicides efficacy is reduced with increasing

levels of multiple herbicide resistance when compared to a herbicide-susceptible

waterhemp population.

3) Confirm the evolution of resistance to HPPD-inhibiting herbicides in the giant

ragweed population and determine the efficacy of herbicides with different sites-of-

action.

This thesis is organized as five chapters. The first chapter contains a general introduction and literature review of herbicide resistance, Amaranthus tuberculatus (Moq.) J.D. Sauer

(waterhemp) biology, very long chain fatty acid-inhibiting herbicides, Ambrosia trifida L

(giant ragweed) biology, 4-hydroxyphenylpyruvate dioxygenase-inhibiting herbicides, and a summary of research objectives. Chapters two, three, and four are the research manuscripts which are formatted for the submission to scientific journals and the organization of each manuscript is journal specific. The manuscript titled “Influence of multiple herbicide 11 resistance on growth and development in Amaranthus tuberculatus” is formatted for submission to Weed Research. The manuscript titled “Investigating efficacy of selected very long chain fatty acid-inhibiting herbicides on Amaranthus tuberculatus populations with evolved multiple herbicide resistances” is formatted for submission to Weed Science. The third research manuscript is titled “Investigating a putative HPPD-resistant Ambrosia trifida population” and is formatted for submission to Weed Technology. Chapter five consists of general conclusions and future research recommendations.

CHAPTER 2: INFLUENCE OF MULTIPLE HERBICIDE RESISTANCE ON GROWTH AND DEVELOPMENT IN AMARANTHUS TUBERCULATUS A paper to be submitted to Weed Research

Eric A. L. Jones and Micheal D. K. Owen*

*First and second authors: Graduate Assistant and University Professor Emeritus,

Department of Agronomy, Iowa State University, Ames, IA 5001-1010.

Abstract

Herbicide resistance mutations should come at a cost to plants. However, in some agroecosystems herbicide-resistant weed populations are becoming more prevalent than herbicide-susceptible weed populations. More specifically, multiple-herbicide-resistant

(MHR) Amaranthus tuberculatus (waterhemp) populations, once considered an oddity, are now the normality. The objective of this study was to determine if MHR waterhemp populations grow and develop differently when compared to a herbicide-susceptible waterhemp population. The hypothesis was that MHR waterhemp populations grow and develop at a lesser rate than the herbicide-susceptible waterhemp population. Three-, four,- and five-way herbicide-resistant and herbicide-susceptible waterhemp populations were grown in the field for 20 weeks in 2016 and 2017. Plant height and presence of flowers were recorded weekly for 11 weeks. Shoot biomass and seed production were recorded at plant maturity. Significant differences (P < 0.0001) in height were detected for all populations each week in 2016 and 2017. There was no consistent clustering of MHR populations across the different weeks and years. It was concluded that the relative growth rate did not differ across waterhemp populations. Gender and population were a significant effect (P < 0.0001) on 13 flowering date, although 1.5 weeks was the longest difference between flowering dates and was not considered biologically important. Accumulated biomass was not significantly different (P = 0.84) across all waterhemp populations, but there was significant difference (P

< 0.0001) for gender and year. While seed production was statistically different for waterhemp populations (0.0010) there were no consistent patterns of seed production with regard to waterhemp populations. The results of this experiment provided evidence that

MHR waterhemp populations do not grow and develop differently than herbicide-susceptible waterhemp populations under field conditions. MHR waterhemp plants are not at an ecological disadvantage compared with herbicide-susceptible waterhemp plants. Thus, it can be concluded that herbicide-free environments will not negatively impact MHR waterhemp populations.

Introduction

Amaranthus tuberculatus (Moq.) J.D. Sauer (waterhemp) is a native, small-seeded, summer annual plant that historically grew along riverbanks and disturbed areas (Sauer 1955). Prior to conservation tillage, conventional tillage effectively controlled many small-seeded weeds such as waterhemp. Due to changes from conventional tillage to conservation tillage, waterhemp adapted and thrived in the agricultural fields of the Midwest United States

(Steckel 2007). Herbicides controlled the weeds that flourished in the minimally-tilled soil, allowing conservation tillage to be successful (Buhler 1995). However, intensive herbicide use selected individual waterhemp plants in the field that have naturally occurring herbicide resistance mutations and a gradual shift from herbicide-susceptible to herbicide-resistant waterhemp populations occurred (Heap 2018). After a herbicide has lost efficacy due to the 14 evolution of resistant weed populations, farmers will choose another herbicide thus continuing the “herbicide selection cycle” (Young 2006).

Evidence of this herbicide selection cycle with waterhemp is illustrated with the intensive usage of acetolactate synthase (ALS) (EC 2.2.1.6) inhibitor herbicides (Herbicide Group

[HG] 2) in the late 1980’s and early 1990’s resulting in a majority of waterhemp populations evolving resistance to the ALS-inhibitor herbicides (Young 2006). The commercial introduction of glyphosate-resistant (e.g., Roundup Ready) soybeans, maize, and cotton coincided with the escalation of waterhemp resistance to the ALS inhibitors (Hinz and Owen

1997). Glyphosate (HG 9) controlled waterhemp and could be applied postemergence without injury to glyphosate-resistant crops; with time, farmers applied glyphosate more frequently and at increasingly higher rates to control weeds (Young 2006) and facilitated the evolution of glyphosate resistance in waterhemp (Powles 2008). As a result, other herbicides were applied with or in lieu of glyphosate to control the glyphosate-resistant waterhemp populations. The list of herbicides used to help control glyphosate-resistant waterhemp included: protoporphyrinogen oxidase (EC 1.3.3.4) (PPO) inhibitors (HG 14), 4- hydroxylphenylpyruvate dioxygenase (EC 1.13.11.27) (HPPD) inhibitors (HG 27), photosystem II (PSII) (EC 1.10.3.9) inhibitors (HG 5), growth regulators (HG 4), and glutamine synthase inhibitors (EC 6.3.1.2) (HG 10). Waterhemp populations now exist that are resistant to six herbicide groups (HG 2, 4, 5, 9, 14, and 27) (Bernards et al., 2012), and herbicide resistance profiles ranging from resistance to one herbicide group to five herbicide groups have been reported (Owen 2013, Evans 2016, Heap 2018).

Herbicide resistance may impart a fitness cost to weeds (Vilva-Aiub et al., 2015). The fitness cost of herbicide resistance can be defined as a reduction in plant productivity in a herbicide- 15 free environment caused by pleiotropic effects of the herbicide resistance mutation (Vila-

Aiub et al., 2015). An example of a fitness penalty is the reduced success of producing offspring that contribute to the next generation by a particular weed population relative to other weed populations (Primack et al., 1989). The fitness cost may be the result of mutations that impede normal plant function, causing a diversion of resources from growth and development processes to the evolved resistance mechanism, and pleiotropic effects from resistance allele in absence of the herbicide (Vila-Aiub et al., 2009). Herbicide resistance costs are measured in herbicide-free environments by controlling for differences in genetic backgrounds, comparison of phenotypes, ability to contribute viable offspring to the next generation, and changes of the herbicide resistance frequency over time to determine if herbicide-susceptible weeds will be favored over herbicide-resistant weeds (Vila-Auib et al.,

2011, Wu et al., 2017)

Evolving additional herbicide resistance mechanisms may impart additional fitness costs to the plant. Conceivably the growth and development of multiple herbicide-resistant (MHR) waterhemp populations might occur at a lesser rate when compared to populations with a single herbicide resistance trait or herbicide-susceptible waterhemp populations. Thus, the objective of the study was to determine if MHR waterhemp grow and develop at a different rate than herbicide-susceptible waterhemp in the absence of herbicides and if multiple herbicide resistances increase the fitness cost to a waterhemp population. The hypothesis of the study is that multiple herbicide resistance traits in waterhemp will cause greater reductions in growth and development when compared to herbicide-susceptible waterhemp populations and populations with fewer herbicide resistances.

Materials and Methods

Plant material

MHR waterhemp populations were selected from a collection of Iowa waterhemp populations (Owen 2013). The collection of waterhemp populations was screened with five herbicides in the greenhouse to determine the frequency and distribution of herbicide- resistant waterhemp in Iowa (Owen 2013). The sites of action in the herbicide efficacy screen included; HG2 (e.g., imazethapyr), HG 5 (e.g., atrazine), HG 9 (e.g., glyphosate) (EC

2.5.1.19), HG 14 (e.g., lactofen), and HG 27 (e.g., mesotrione). The herbicide resistance profiles of the selected waterhemp populations for the study included: herbicide-susceptible,

3-way herbicide-resistance (HG 2, 5, and 27), 4-way herbicide-resistance (HG 2, 5, 9, and

27), and 5-way herbicide-resistance (HG 2, 5, 9, 14, and 27) (Table 1). The herbicide- susceptible population was collected along the edge of a wooded area adjacent to an agricultural field at Curtiss Farm, Ames, IA in 2006 and had no known herbicide history.

Two populations of each resistance profile were selected for the study except for the herbicide-susceptible population.

The selected waterhemp populations were planted in the field in 2015 for a seed increase study (Jones and Owen, unpublished data). The seeds collected from the 2015 study were used as the plant material for the 2016 and 2017 experiments. The herbicide-susceptible and herbicide–resistant waterhemp population seeds were kept in cold storage at 6°C for one year. Seeds from each population were placed in a petri dish with a small amount of water and kept at 6°C for two weeks to break dormancy. The petri dishes were then placed into a dryer without lids for 48 hours at 45°C. The seeds were then sown into Ray Leach Cone- tainers™ (Stuewe & Sons, Inc., Tangent, OR, USA) containing a 4:1 ratio of Sunshine Mix 17

#1/LC1 potting soil and sand with approximately 1 g of Osmocote Flower Food Granules

(14-14-14) at the beginning of May in 2016 and 2017. Plants were maintained in the greenhouse at 24°C and watered as needed. Sunlight was supplemented with 600-1,000 µmol m-2 s-1 PFFD of artificial light set to a 14-hr photoperiod.

Plants were transplanted to the field when a majority of the plants reached 12 cm in height at the end of May in 2016 and 2017. Plants were watered in the field as needed for one week to ensure establishment. Plants that were lost during transplanting were replaced with similar- sized plants during the first week after transplanting. Plants remained in the field for approximately 20 weeks in both years.

Field description

The field for the experiments was located at the USDA Plant Introduction Station in Ames,

IA in 2016 and 2017. Temperature and precipitation data for the 2016 and 2017 seasons were gathered from the Iowa Mesonet station located at the Plant Introduction Station (Tables 2 and 3). The soil types of the field are a Clarion loam (fine-loamy, mixed, superactive, mesic typic hapludolls) and sewage lagoon with a pH 7 and 2.3% organic matter. The experimental design was a randomized complete block with six replications. Strips of 1 x 90 m black plastic mulch were placed 3 meters apart for each replication and Avena sativa L. (oats) was sown between strips to act as a smother crop to suppress volunteer weeds. Weeds that were not suppressed by the oats were hand-rogued throughout the growing season. Plots were spaced 5 meters apart in 2016 and 9 meters apart in 2017 and placed within the black plastic mulch. Each plot contained six waterhemp plants of unknown gender. Waterhemp is a dioecious species and segregates approximately 50:50 male and female so it was expected that each plot would have 3 male and 3 female plants. Six holes in a 2x3 arrangement equally 18 spaced 25 cm apart were dug with a bulb planter into the black plastic mulch for each plot.

Due to the eventual large size of the plants, a 1-meter long wooden lath was placed in the middle of each plot and plants were fasten to the lath for support as needed.

Morphological measurements

Plants were not measured for one week to allow for acclimation to the field conditions. After the acclimation period, morphological traits were recorded for the first 11 weeks.

Morphological measurements included measuring plants weekly to calculate relative growth rate, and determining the development of reproductive structures. Plant height was determined by measuring the plants from the soil surface to the apical meristem weekly. The relative growth rate was extrapolated from the weekly plant heights using the equation:

F = y0 + a / (1 + exp(-(x-x0) / b)) where y0 equals the initial plant height, a equals the maximum plant height, x equals the week, and b equals the growth rate throughout the 11 weeks. Visual inspections assessed if flowers were present, which also determined the plant gender. Measurements taken after plant harvest were plant biomass and female plant seed production. Plants were excised at the soil surface and placed into a Delnet® Pollination™ bag (DelStar Technologies, Inc.,

Middletown, DE, USA) to minimize biomass and seed losses. Harvested plants were weighed in the field to obtain a wet biomass weight and then dried for 24 hours at approximately 50° C and dry biomass weight was determined.

Female plants were threshed by hand to remove seeds from the florets and seeds were separated from plant residues using sieves and a South Dakota Seed Blower (Seedburo

Equipment Company, Chicago, IL, USA). Visual inspections, as described by Sawma and 19

Mohler (2002), during the cleaning process determined if seeds were viable or nonviable.

Nonviable seeds where separated out along with the plant residue. Semi-cleaned seeds were rubbed between pieces of rubber sheeting which supplied enough friction and pressure to remove the remaining florets. Samples were then processed with the South Dakota Seed

Blower to further remove plant residue. The total number of seeds produced by each female plant extrapolated by determining the mass of five sub-samples of 100 seeds for each population (Bertucci et al., 2018, Sellers et al., 2003). The total number of seeds produced was derived from the equation:

T = (W / S) * 100 where W equals the total seed mass, S equals the average mass of the 5 100-seed sub- samples, and T equals the extrapolated number of seeds produced.

Statistical Analysis

All statistical analyses were performed using Statistical Analysis Software, SAS 9.3 (SAS

Institute, Inc., NC, USA). Year, plant height, flowering date, accumulated plant biomass, and seed production data were subject to ANOVA using the GLIMMIX procedure. Means and interactions of significant effect were separated using Fisher’s protected LSD test at P ≤ 0.05.

Results

Relative growth rate

Year, week, and waterhemp population were significant effects for the relative growth rate (P

< 0.0001). Plant gender was not significant for relative growth rate (P = 0.28). Plants on average grew 17.8 and 19.4 cm per week and reached an average final height of 190 and 20

203.4 cm in 2016 and 2017, respectively (Figures 1 and 2). Despite the significant differences. there was no clustering or deviations away from the general trend line by specific waterhemp populations. Thus, the conclusion is there was no difference in relative growth amongst the waterhemp populations.

Flowering date

Waterhemp gender and population flowering date were significant (P < 0.0001), while year

(P = 0.16) was not a significant effect for population flowering date. Female flowers developed 1.5 weeks later than the male flowers in both years. Across all waterhemp populations, female flowers began to develop 7 weeks after transplanting and male flowers began to develop at 5.5 weeks (Figure 3). This result was expected, as male plants need to produce pollen before female flowers are receptive to increase likelihood of successful pollination (Bawa 1980, Gross and Soule 1981). Female flowers usually require more resources to develop than male flowers (Gross and Soule 1981). Research has shown that flowering synchronization does not negatively affect successful reproduction and impart a strong selection factor (Wu and Owen 2014, Ollerton and Lack 1990). Flowering dates separated by one week were unlikely of biological importance.

Biomass

No differences in dry biomass were detected amongst the different waterhemp populations (P

= 0.84). Plants weighed 0.54 kg, on average. It was concluded that the MHR waterhemp populations accumulated biomass similarly and were not different than the herbicide- susceptible waterhemp population. However, there were significant differences of accumulated biomass by year and amongst plant gender (P < 0.0001). On average, female 21 plants accumulated 0.35 kg more biomass when compared to the male plants across all populations (Figures 4). Female plants accumulated more biomass per plant in 2017 (0.82 kg) when compared to 2016 (0.62 kg). Biomass per male plant was not significantly different for the waterhemp populations in 2016 (0.31 kg) and 2017 (0.40 kg) (Figures 4 and 5). The differences in accumulated biomass between male and female plants was expected. The larger size of female plants is the consequence of continuing to grow and accumulate resources throughout the season to ensure successful seed production (Bram and Quinn 2000,

Gross and Soule 1981). After pollen dispersal, male plants usually senesce.

Seed production

The mass of the five 100-seed sub-samples was statistically different for all waterhemp populations in 2016 and 2017 (P < 0.0001) (Figure 5). After the total mass of seeds produced was transformed into total number of seeds produced, the differences were statistically different across the waterhemp populations (P = 0.0010) (Figure 6). Across all waterhemp populations, the total number of seeds produced was not statistically different between years

(P = 0.67); 670,000 and 630,000 seeds were produced by each female plant in 2016 and

2017, respectively. The waterhemp plants were probably not producing seeds at a maximum capability due to the high level of intraspecific competition imparted by having six vigorously growing waterhemp plants per plot. Despite the statistical differences of the number of seeds produced, no specific MHR profile consistently produced the most or least amount of seeds (P = 0.84) (Figure 7). The total number of seeds produced by all MHR waterhemp populations did not support the hypothesis that any MHR profile incurred a fitness penalty when compared to the herbicide-susceptible waterhemp population.

Discussion

Plants were assumed to have retained at least 90% or more of total seeds produced when harvest and shattered seeds were not collected (Schwartz et al., 2016). Early seed shattering is a weedy plant trait that allows for seeds to enter the soil seed bank for successful future germination by avoiding removal by humans or machines (Baker 1956, Burton et al., 2016).

However, early seed shattering is not associated with waterhemp. Regardless, future research should account for shattered seeds in the total amount of seeds produced to better determine seed production for each waterhemp population and assess if a fitness penalty may occur.

Populations 3A and 3B did not germinate as well as other waterhemp populations (data not shown). The low germination could be a putative fitness penalty not explored by this study although it could be a factor attributable to seed handling and storage. The 3-way herbicide- resistant populations may incur a putative fitness penalty that reduces seed viability and germination. While, the 3-way herbicide-resistant waterhemp plants grew and developed similarly to all other waterhemp populations, low germinating populations would be at a disadvantage with populations that exhibit much higher germination. However, the most common MHR populations that inhabit Iowa are 3-way herbicide-resistant which suggests that the populations may not incur a fitness penalty (Owen 2013). The seed production results of this experiment show that each waterhemp population can produce several hundred thousand seeds. The germination of seeds and seed shattering characteristics need to be further studied to better determine if the MHR waterhemp populations incur a putative fitness penalty attributable to herbicide resistances. 23

Environmental conditions for the 2016 and 2017 growing seasons were dissimilar as depicted by the significant effect imposed by year for relative growth rate, accumulated biomass, and seed production (P < 0.0001) (Tables 2 and 3). The resistance profile of the waterhemp population was also a significant effect for relative growth rate, flowering date, and seed production. However, given the variability comparing specific populations or MHR profiles, it is difficult to determine if any of the MHR waterhemp populations incurred a fitness penalty (Vila-Auib 2005). Differences in flowering date and biomass detected between genders is biologically important for waterhemp, however. The morphological differences between male and female plants are likely sex-specific selection to ensure successful reproduction (Bram and Quinn 2000).

Criticism arises when fitness studies do not account for genetic background differences within separate plant populations (Wu et al., 2017, Vila-Auib et al., 2015, Giacomini et al.,

2014). Differences in genetic background can cause populations to grow differently which could be falsely attributed to a herbicide resistance mechanism (Vila-Auib et al., 2015). Two waterhemp populations of each resistance profile were studied to determine if populations of the same herbicide resistance profile had segregating morphological traits. This provided more statistical power, allowing greater confidence in the finding of no fitness penalties.

The specific herbicide resistance mechanisms for each waterhemp population were not identified. Therefore, the waterhemp populations with the same herbicide resistance profiles may contain different resistance mechanisms for a specific herbicide group. Different resistance mechanisms may impart different fitness costs on weeds (Vila-Aiub et al., 2015).

Understanding the specific herbicide resistance mechanism for the waterhemp populations would likely provide evidence describing which herbicide resistance mechanism(s) are less 24 costly to the plant and thus are more likely to evolve within waterhemp populations (Wu et al., 2017). Future research should investigate the resistance mechanisms of the selected MHR waterhemp populations to determine the similarities or dissimilarities between the respective resistance profiles.

Conclusions

The results of the experiments fail to reject the null hypothesis since MHR waterhemp populations do not grow and develop differently when compared to herbicide-susceptible waterhemp populations. It is safe to conclude that MHR populations are not at an ecological disadvantage and thus do not exhibit a fitness penalty attributable to mutations that code for herbicide resistance. Herbicide-free environments will not favor herbicide-susceptible waterhemp populations over herbicide-resistant waterhemp populations. The lack of fitness penalties will likely support the prevalence of MHR waterhemp populations in the agroecosystem.

References

Baker HG (1956) Characters and modes of origin of weeds. Pages 147-172 in Baker HG,

Stebbins GL, eds. The genetics of Colonizing Species. New York: Academic Press.

Bernards ML, Crespo RJ, Kruger GR, Gaussoin R, Tranel PJ (2012) A waterhemp

(Amaranthus tuberculatus) population resistant to 2,4-D. Weed Sci. 60, 379-384

Bertucci MB, Bartley P, Jennings KM, Monks DW, Jackson B (2018) Automated seed

counts and verification of seed production estimates of Palmer amaranth using a

computerized particle analyzer. Abs. WSSA 58, 12 25

Bram MR and Quinn JA (2000) Sex expression, sex-specific traits, and the effects of salinity

on growth and reproduction of Amaranthus cannabinus (Amaranthaceae), a dioecious

annual. Am. J. Bot. 87, 1609-1618

Buhler DD (1995) Influence of tillage systems on weed population dynamics and

management in corn and soybean in the Central USA. Crop Sci. 35, 1247-1258

Burton NR, Beckie HJ, Willenborg CJ, Shirtliffe SJ, Schoenau JJ, Johnson EN (2016)

Evaluating seed shatter of economically important weed species. Weed Sci. 64, 673-

682

Evans CM (2016) Characterization of a novel five-way-resistant population of waterhemp

(Amaranthus tuberculatus) M.S. dissertation. Champaign, IL: University of Illinois.

Darwin C (1859) On the origin of species by means of natural selection or the preservation of

favoured races in the struggle for life. 6th ed. London, England: Penguin Books.

Giacomini D, Westra P, Ward SM (2014) Impact of genetic background in fitness cost

studies: an example from glyphosate-resistant Palmer amaranth

Grosse KL and Soule JD (1981) Differences in biomass allocation to reproductive and

vegetative structures of male and female plants of a dioecious, perennial herb, Silenne

alba (Miller) Krause. Am. J. Bot. 68, 801-807

Ollerton J and Lack AJ (1992) Flowering phenology: an example of relaxation of natural

selection? Trends Ecol. Evol. 7, 274-276

Owen MDK (2013) Pest resistance: Overall principles and implications on evolved herbicide

resistance in Iowa. ICM Conference 25, 125-136 26

Powles SB (2008) Evolved glyphosate-resistant weeds around the world: lessons to be learnt.

Pest Manag. Sci. 64, 360-365

Primack RB and Kang H (1989) Measuring fitness and natural selection in wild plant

populations. Annu. Rev. Ecol. Evol. Syst. 20, 367-396

Sauer JD (1955) Revision of the dioecious amaranths. Mandroño 13, 5-46

Sawma JT and Mohler CL (2002) Evaluating seed viability by an unimbibed seed crush test

in comparison with the tetrazolium test. Weed Technol. 16, 781-786

Sellers BA, Semda RJ, Johnson WG, Kendig JA, Ellersieck MR (2003) Comparative growth

of six amaranthus species in Missouri. Weed Sci. 51, 329-333

Steckel LE (2007) The dioecious Amaranthus spp.: here to stay. Weed Technol. 21, 567-570

Vila-Auib MM, Gundel PE, Preston C (2015) Experimental methods for estimation of plant

fitness costs associated with herbicide-resistance genes. Weed Sci. 63, 203-216

Vila-Auib MM, Neve P, Powles SB (2009) Fitness costs associated with evolved herbicide

resistance alleles in plants. New Phytol. 184, 751-767

Wu C, Davis AS, Tranel PJ (2017) Limited fitness costs of herbicide-resistance traits in

Amaranthus tuberculatus facilitate resistance evolution. Pest Manag. Sci. 74, 293-301

Wu C and Owen MDK (2014) When is the best time to emerge: reproductive phenology and

success of natural common waterhemp (Amaranthus rudis) cohorts in the Midwest

United States? Weed Sci. 62, 107-117

Young BG (2006) Changes in herbicide use patterns and production practices resulting from

glyphosate-resistant crops. Weed Technol. 20, 301-307 27

Tables Table 1. Herbicide resistance profiles and locations of waterhemp populations collected in Iowa during 2011 included in the growth and development studies established at the USDA Plant Introduction Station, Ames, Iowa.a Classification Abbreviation Herbicide Location Resistance Profilea

Herbicide- C Susceptible Story County, IA Susceptible

3-Way 3A ALS, PSII, HPPD Henry County, IA Resistance

3-Way 3B ALS, PSII, HPPD Cherokee County, Resistance IA

4-Way 4A ALS, PSII, Monona County, Resistance EPSPS, HPPD IA

4-Way 4B ALS, PSII, Plymouth Resistance EPSPS, HPPD County, IA

5-Way 5A ALS, PSII, Not Recorded Resistance EPSPS, HPPD, PPO 5-Way 5B ALS, PSII, Woodbury Resistance EPSPS, HPPD, County, IA PPO a Abbreviations: ALS, acetolactate synthase; PSII, photosystem II; HPPD, 4- hydroxyphenylpyruvate dioxygenase; EPSPS, 5-enolpyruvylshikimate-3-phosphate synthase;

PPO, protoporphrinogen oxidase

Table 2. Temperature, growing degree days and precipitation in 2016, Ames, IA.

Table 3. Temperature, growing degree days and precipitation in 2017, Ames, IA.

Figures

Figure 1. Growth of male and female waterhemp plants of 3A (■), 3B (□), 4A (▲), 4B (Δ), 5A (●), 5B (○), and C (X) populations a at the USDA Plant Introduction Station, Ames, Iowa in 2016. b, c a 3A,B = 3-way herbicide resistance, 4A,B = 4-way herbicide resistance, 5A,B = 5-way herbicide resistance, C = herbicide susceptible. b The first week measurements were recorded on June 8th, 2016. c Year, week, and population were a significant effect on growth rate (P < 0.0001), while the gender was not a significant effect on growth rate, male and female waterhemp plants were pooled (P = 0.28).

Figure 2. Season-long growth for both male and female waterhemp plants of 3A (■), 3B (□), 4A (▲), 4B (Δ), 5A (●), 5B (○), and C (X) populationsa during the 2017 season at the USDA Plant Introduction Station, Ames, Iowa. b, c a 3A,B = 3-way herbicide resistance, 4A,B = 4-way herbicide resistance, 5A,B = 5-way herbicide resistance, C = herbicide susceptible. b The first week measurements were recorded on May 30th, 2017. c Year, week, and population were a significant effect on growth rate (P < 0.0001), while the gender was not a significant effect on growth rate, male and female waterhemp plants were pooled (P = 0.28).

8 A AB A ABC 7 ABC ED DC BCD E CDE 6 EF E F F 5

4 Weeks 3

0 C F 3A F 3B F 4A F 4B F 5A F 5B F C M 3A M 3B M 4A M 4B M 5A M 5B M Population

Figure 3. Weeks required to flower for female (F) and male (M) waterhemp plants a at the USDA Plant Introduction Station, Ames, Iowa. b, c, d a 3A,B = 3-way herbicide resistance, 4A,B = 4-way herbicide resistance, 5A,B = 5-way herbicide resistance, C = herbicide susceptible. b Population and gender were significant effects weeks required to flower (P < 0.0001)Year was not a significant effect on flowering date (P = 0.16), thus the data from 2016 and 2017 was pooled. c Means presented within each column with no common letter(s) are significantly different according to Fisher’s protected LSD test where P ≤ 0.05. d The error bars represent the standard error of each population mean

Figure 4. Accumulated biomass of female (F) and male (M) waterhemp plants in 2016 (gray) and 2107 (white) at the USDA Plant Introduction Station, Ames, Iowa. a, b, c a Means presented within each column with no common letter(s) are significantly different according to Fisher’s protected LSD test where P ≤ 0.05. b Year and gender were significant effects (P < 0.0001), while population was not a significant effect (P = 0.84) on biomass, thus population data was pooled. c The error bars represent the standard error of each population mean.

Figure 5. The average mass of five 100 seed sub-samples from each waterhemp population a for 2016 (gray columns) and 2017 (white columns) at the USDA Plant Introduction Station, Ames, Iowa .b, c, d a 3A,B = 3-way herbicide resistance, 4A,B = 4-way herbicide resistance, 5A,B = 5-way herbicide resistance, C = herbicide susceptible. b Means presented within each column with no common letter(s) are significantly different according to Fisher’s protected LSD test where P ≤ 0.05. c Year and population were significant effects (P < 0.0001) on the mass of 100 seeds. d The error bars represent the standard error of each population mean.

1E+06 A 9E+05 AB ABC 8E+05 BCD 7E+05 BC CD 6E+05 5E+05 D 4E+05

Seeds per plantper Seeds 3E+05 2E+05 1E+05 0E+00 S 3A 3B 4A 4B 5A 5B Population

Figure 6. Fecundity of waterhemp populations a at the USDA Plant Introduction Station, Ames, Iowa in 2016 and 2017. b, c, d a 3A,B = 3-way herbicide resistance, 4A,B = 4-way herbicide resistance, 5A,B = 5-way herbicide resistance, C = herbicide susceptible. b Means presented within each column with no common letter(s) are significantly different according to Fisher’s protected LSD test where P ≤ 0.05. c Population was a significant effect (P = 0.0010) while year was not a significant effect (P = 0.64), thus the data from 2016 and 2017 was pooled d The error bars represent the standard error of each population mean.

Figure 7. Fecundity of waterhemp populations a at the USDA Plant Introduction Station, Ames, Iowa in 2016 and 2017. b, c, d a 3A,B = 3-way herbicide resistance, 4A,B = 4-way herbicide resistance, 5A,B = 5-way herbicide resistance, C = herbicide susceptible. b Means presented within each column with no common letter(s) are significantly different according to Fisher’s protected LSD test where P ≤ 0.05. c Population was a significant effect (P = 0.0010) while year was not a significant effect (P = 0.64), thus the data from 2016 and 2017 was pooled d The error bars represent the standard error of each population mean.

CHAPTER 3. INVESTIGATING THE EFFICACY OF SELECTED VERY LONG CHAIN FATTY ACID-INHIBITING HERBICIDES ON AMARANTHUS TUBERCULATUS POPULATIONS WITH EVOLVED MULTIPLE HERBICIDE RESISTANCES A paper to be submitted to Weed Science Investigating the efficacy of selected very long chain fatty acid-inhibiting herbicides on

Amaranthus tuberculatus populations with evolved multiple herbicide resistances

Eric A. L. Jones and Micheal D. K. Owen*

*First and second authors: Graduate Assistant and University Professor Emeritus,

Department of Agronomy, Iowa State University, Ames, IA 5001-1010.

Abstract

Very long chain fatty acid (VLCFA)-inhibiting herbicides have been applied to maize and soybean fields in Iowa since the 1960s. There are no confirmed weed populations in Iowa with evolved resistance to the VLCFA inhibiting herbicides. Recently, VLFCA-inhibiting herbicides have been applied more extensively to control multiple herbicide-resistant

Amaranthus tuberculatus (Moq.) Sauer (waterhemp) populations. Waterhemp has evolved resistance to six herbicide sites-of-action (herbicide group [HG] 2, 4, 5, 9, 14, and 27). The hypothesis is, that VLCFA-inhibiting herbicides will be less efficacious with increasing herbicide resistances within a waterhemp population. Thus, the objective of this study was to determine if multiple herbicide resistance in waterhemp will decrease VLCFA-inhibiting herbicide control efficacy. Dose-response assays were conducted in the field and in a germination chamber to determine the efficacy of three VLCFA-inhibiting herbicides, acetochlor, s-metolachlor, and flufenacet, on selected multiple herbicide-resistant waterhemp populations. Multiple herbicide-resistant waterhemp populations from Grundy (HG 2, 5, 9 37 and 27) and Story Counties, Iowa (HG 2, 14, and 27) fields responded similarly to the herbicides. In the germination chamber, 3-way (HG 2, 5, and 27), 4-way (HG 2, 5, 9, and

27), and 5-way (HG 2, 5, 9, 14, and 27) multiple herbicide-resistant waterhemp populations and a herbicide-susceptible population responded to the herbicide treatments similarly.

However, there were differences in efficacy between the herbicides. Acetochlor achieved the highest control efficacy followed by s-metolachlor then flufenacet.

Introduction

Very long chain fatty acids (VLCFA) are fatty acids with acyl chains of 18 carbons or longer and are major constituents of precursors for triacylglycerol, cell membranes, signaling molecules, and cuticular waxes (Bach and Faure 2010). These molecules are of great importance for plant growth and development. VLCFA-inhibiting herbicides interfere with the elongation of C18 chains, specifically elongation steps from C18:0 to C20:0 and C20:0 to

C22:0, which causes phytotoxicity in sensitive plants (Böger 2003, Tanetani et al. 2009).

VLCFA-inhibiting herbicides are relatively old chemistries, for example, alachlor was first used commercially in 1969 (Hamm 1974). VLCFA-inhibiting herbicides are generally considered to be more efficacious on grass, weeds but they are active on some broadleaf weeds (Busi 2014, Hamm 1974). The VLCFA-inhibiting herbicides must impact several elongases that have differing physiological functions. However, mutations to the elongases that would provide resistance to the VLCFA-inhibiting herbicides can be lethal, which may explain why resistance to these herbicides is extremely rare (Trenkamp et al. 2004, Halsam and Kunst 2013).

VLCFA-inhibiting herbicide resistance has been confirmed in only five grass weed species globally, Alopecurus myosuroides Huds. (blackgrass), Avena fatua L. (wild oat), 38

Echinochloa crus-galli (L.) P. Beauv. (barnyardgrass), Lolium multiflorum Lam. (Italian ryegrass), and Lolium rigidum Gaud. (rigid rygrass) (Heap 2018). While resistance to

VLCFA-inhibiting herbicides is not common, the existence of resistant biotypes should suggest that resistance can evolve as herbicides are the most consistent selector for weed evolution in agriculture (Gressel 2009).

Amaranthus tuberculatus (Moq.) Sauer (waterhemp) is an important weed in the Midwest

United States and has rapidly evolved resistance to a number of herbicides used in corn and soybean (Owen 2013, Crespo 2016). The VLCFA-inhibiting herbicides are generally efficacious on herbicide-resistant waterhemp populations (Hausman et al. 2013, Steckel et al.

2002). Many weeds including waterhemp have evolved non-target-site resistance to herbicides by the facilitation of modified mixed function oxidases such as cytochrome P450- monooxygenase or glutathione-S transferases (Yuan et al. 2007, Délye 2013). Such xenobiotic-degrading enzymes can provide cross-resistance to other herbicides (Letouze and

Gasquez 2003). It is possible that the mixed function oxidase have the capacity to metabolize many xenobiotic compounds. This phenomenon of oxidases efficiently metabolizing herbicides is reported in multiple herbicide-resistant (MHR) Lolium rigidum (rigid ryegrass),

Avena fatua (wild oat), and Alopercurus myosuroides (blackgrass) populations (Yu et al.

2009, Cummins et al. 1999) and it is possible that MHR waterhemp populations could also possess similar enzymes that efficiently metabolize herbicides. Thus, resistance to the

VLCFA-inhibiting herbicides in waterhemp could evolve without mutations to the target site.

Research describing the VLCFA-inhibiting herbicides efficacy on MHR waterhemp populations is important as MHR waterhemp populations are becoming more common

(Tranel et al. 2011). If MHR waterhemp populations can efficiently metabolize the VLCFA- 39 inhibiting herbicides, another useful weed management tool will be lost. The objectives of this research was to determine the efficacy of selected VLCFA-inhibiting herbicides on several MHR waterhemp populations. The hypothesis is that waterhemp populations with resistance to more herbicide groups will be less susceptible to VLCFA-inhibiting herbicides.

Materials and Methods

Field dose-response experiments. The VLCFA-inhibiting herbicide dose-response experiments were conducted in Grundy County, Iowa (2016) and Story County, Iowa (2017).

The soil types for the Grundy County site were a Tama silty clay loam (fine-silty, mixed, superactive, mesic typic argiudolls) and Sawmill-Garwin complex (fine-silty, mixed, superactive, mesic cumulic and typic endoaquolls) with a pH of 7.1 and 4.0% organic matter.

The soil types for the Story county site were a Clarion loam (fine-loamy, mixed, superactive, mesic typic hapludolls) and Nicolett loam (fine-loamy, mixed, superactive, mesic aquic hapludolls) with a pH of 6.8 and 3.0% organic matter. The herbicide resistance profile of the

Grundy County waterhemp population was 4-way resistance to herbicide groups [HG] 2, 5,

9, and 27 and the Story County population was 3-way resistance to HG 2, 14, and 27. The experiment design was a randomized complete block with each treatment replicated three times. Plots were 3.0 x 7.6 m in size and herbicides were applied after crop planting using

-1 140 L ha carrier volume through a CO2-powered backpack sprayer with TeeJet AI110015 nozzles (Spraying Systems Co., Wheaton, IL, USA). Experiments were conducted with crop competition (corn) and no crop competition (fallow) in 2016 and 2017, respectively. The soybeans adjacent to the fallow study in Story County were used to describe application and evaluation timings. Herbicides were applied at 0.25x, 0.5x, 1x, 2x, and 4x where the 1x amount was based on the maximum labeled rate for the respective herbicide (Table 1). Both 40 experiment sites received approximately 2.5 cm of rain within 7 days after treatment to facilitate herbicide activation. Visual efficacy evaluations were conducted at 21 and 42 days after crop emergence (DAE) using a rating scale from 0% to 100% where 0% equaled no control and 100% equaled complete waterhemp control.

Germination chamber dose-response experiment. A germination chamber dose-response assay was used to assess the susceptibility of MHR waterhemp populations to VLCFA- inhibiting herbicides. The VLCFA-inhibiting herbicides used in the field experiments were also used in the germination chamber experiment (Table 1). The experimental design was completely randomized and each treatment was replicated twice. The experiments were repeated three times.

Dose-response assays were conducted by adding 6 ml of different herbicide concentrations to the petri dishes containing germination paper. Herbicide concentrations were established based on a 3.16 log scale from 0.1 to 10.0 active ingredient parts per million (ppm) along with an untreated control (Tranel, personal communication). 20 stratified seeds were placed on the treated germination paper inside of the petri dishes and placed in the germination chamber. The germination chamber (Percival Scientific, Inc., Perry, IA, USA) was adjusted to 14-hour photoperiods with 25° C during the light and 20° C during the dark. Light was supplemented by mercury halide light providing 600-1,000 µmol m-2 s-1 photosynthetic photon flux density.

Percent of control was determined by dividing the number of treated seeds that germinated by the number of germinated untreated seeds 2 days after treatment (DAT). The evaluation time was select as approximately 90% germination had occurred in the untreated controls and 41 radicals were visible. Efficacy evaluations were made using a 0% to 100% rating scale where

0% equaled no control and 100% equaled complete control.

Plant Material. Multiple herbicide-resistant waterhemp populations were selected from waterhemp populations collected across Iowa in 2011 (Owen 2013). Waterhemp populations with resistance to 4X rates of the respective herbicides were selected for the growth chamber experiment (Table 1). The herbicide resistance profiles of the selected MHR waterhemp populations included 3-way herbicide-resistance (HG 2, 5, and 27), 4-way herbicide- resistance (HG 2, 5, 9, and 27), and 5-way herbicide-resistance (HG 2, 5, 9, 14, and 27)

(Table 2). Two populations of each MHR phenotype were selected to assess if populations expressed similar responses to the VLCFA-inhibiting herbicides. The herbicide-susceptible waterhemp population was collected along the edge of a wooded area adjacent to an agricultural field at Curtiss Farm, Ames, IA in 2006.

The selected waterhemp populations were planted in the field as part of a waterhemp biology study in 2015 (Jones and Owen, unpublished data) and seeds from this study were used for the VCLFA-inhibiting herbicide dose-response assay conducted in the germination chamber.

The herbicide-susceptible and herbicide–resistant waterhemp population seeds were kept in cold storage at 6°C for one year. Seeds from each population were placed in a petri dish with a small amount of water and kept at 6°C for two weeks to break dormancy. The petri dishes were then placed into a dryer without lids for 48 hours at 45°C. Seeds were kept in cold storage until needed.

Statistical analysis. All statistical analyses were performed using Statistical Analysis

Software, SAS 9.3 (Statistical Analysis Systems version 9.3, SAS Institute, Cary, NC). Data for field and lab experiments were analyzed using the GLIMMIX procedure in SAS. Means 42 and interactions of significant effects were separated using Fisher’s protected LSD test at P ≤

0.05.

Dose-response curves for field and germination chamber experiments were created using a logarithmic curve equation:

Y = Y0 + A * ln(X)

where Y0 and A were constants, Y = % of control, and X = herbicide rate or concentration.

Deviations from the curve are described by r2. The two-parameter logarithmic model estimated the herbicide dose that caused a 50% reduction in germination (LD50) and a 90% reduction in germination (LD90) for each herbicide and waterhemp population. The data points from the experiments were used to derive the model equations on SigmaPlot 12.5

(Systat Software, Inc. version 12.5, San Jose, CA). Resistance ratios (R/S) were calculated by dividing the LD50 of the resistant (R) populations by the LD50 of the susceptible (S) population, respectively.

Results and Discussion

Responses of multiple herbicide-resistant waterhemp populations to VLCFA-inhibiting herbicides under field conditions. Herbicide, rate, and evaluation date were significant for herbicide efficacy (P < 0.0001) but waterhemp population was not significant (P = 0.90).

Acetochlor provided the highest efficacy followed by s-metolachlor and flufenacet 21 and 42

DAE for the Grundy and Story county sites (Figures 1-3). The acetochlor LD50 could not be calculated as the rates used in the experiments were not low enough to achieve a 50% control. All herbicides at the maximum labeled rate provided efficacy above 80% at 21 DAE.

These results suggest that MHR waterhemp populations respond similarly to VLCFA- 43 inhibiting herbicides. Thus, it is not likely that resistance to VLCFA-inhibiting herbicides has evolved in the two MHR waterhemp populations included in the field experiments.

Reponses of multiple herbicide-resistant waterhemp populations to VLCFA-inhibiting herbicides under germination chamber conditions. Herbicide treatment, rate, and waterhemp population were significant effects for efficacy (P < 0.0001). Similar to the field experiments, acetochlor provided the highest efficacy followed by s-metolachlor and flufenacet (Figures 4-6). Resistance ratios for acetochlor remained under 2 for all MHR populations. Resistance ratios exceeded 4 for one 4-way (4B) and one 5-way (5A) MHR waterhemp populations when treated with s-metolachlor. The other 4-way (4A) and 5-way

(5B) MHR waterhemp populations expressed R/S ratios less than 2. Resistance ratios for flufenacet could not be calculated, as the herbicide concentrations used were too low to establish an LD50 or LD90. Resistance ratios which are statistically greater than 1 usually suggest that the population has evolved herbicide resistance (Burgos et al. 2013) The low R/S ratios from this experiment provide evidence that it is unlikely the MHR waterhemp populations have evolved resistance to the VLCFA-inhibiting herbicides (Figures 4-6).

The results of the experiments fail to reject the null hypothesis as all MHR and herbicide- susceptible waterhemp populations included in these experiments responded to the VLCFA- inhibiting herbicides the same. The VLCFA-inhibiting herbicides will likely continue to be useful tools for waterhemp control. However, total reliance on this herbicide group should be avoided, even if the initial frequency of resistance is low (Busi 2014, Heap 2014). VLCFA- inhibiting herbicides are now recommended more frequently to control herbicide-resistant waterhemp (Meyer et al. 2015, Hay 2015, Kohrt and Sprague 2017). Recommendations include “layering” the herbicide applications, by applying a VLCFA-inhibiting herbicide 44 preemergence and again postemergence (Steckel et al. 2002, Jhala et al. 2015, Kohrt and

Sprague 2017). The concurrent use of any selection agent may select for resistant weed phenotypes (Heap 2014). Thus, layering VLCFA-inhibiting herbicides increases the risk of selecting for resistance in waterhemp. Glyphosate resistance was hypothesized to never evolve in weed species (Bradshaw et al. 1997), however the current status of 38 different glyphosate-resistant weed species disproves that hypothesis (Heap 2018).

Future research should investigate the mechanisms of resistance in each waterhemp population to determine if target site, non-target site (metabolic resistance), or both exist. The hypothesis of this research was that MHR waterhemp populations had the ability to metabolize xenobiotics non-selectively, which could render the VLCFA-inhibiting herbicides non-efficacious. Knowledge of the herbicide resistance mechanisms will better explain if target or non-target site resistance mechanisms may help understand why the VLCFA- inhibiting herbicides are still efficacious and why resistance has not evolved.

Literature Cited

Böger P (2003) Mode of Action for Chloroacetamides and Functionally Related Compounds.

J. Pestic. Sci. 28:324-329

Böger P, Matthes B, Scmalfuß J (2000) Towards the primary target of chloroacetamides-new

findings pave the way. Pest Manag. Sci. 56:497-508

Bradshaw LD, Padgette SR, Kimball SL, Wells BH (1997) Perspectives on glyphosate

resistance. Weed Technol. 11:189:198 45

Burgos NR, Tranel PJ, Streibig JC, Davis VM, Shaner D, Norsworthy JK, Ritz C (2013)

Review: Confirmation of resistance to the herbicides and evaluation of resistance

levels. Weed Sci. 61:4-20

Busi R (2014) Resistance to herbicides inhibiting the biosynthesis of very-long-chain fatty

acids. Pest Manag. Sci. 70:1378-1384

Cummins I, Cole DJ, Edwards R (1999) A role for glutathione transfereases functioning as

glutathione peroxidases in resistance to multiple herbicides in black-grass. Plant J.

18:285-292

Délye C (2013) Unravelling the genetic bases of non-targer-site-based resistance (NTSR) to

herbicides: a major challenge for weed science in the forthcoming decade. Pest

Manag. Sci. 69:176-187

Gressel J (2009) Evolving understanding of the evolution of herbicide resistance. Pest

Manag. Sci. 65:1164-1173

Halsam TM, Kunst L (2013) Extending the story of very-long-chain fatty acid elongation.

Plant Sci. 210:93-107

Hamm PC (1974) Discovery, development, and current status of the chloroacetamide

herbicides. Weed Sci. 22:541-545

Hausman NE, Tranel PJ. Riechers DR, Maxwell DJ, Gonzini LC, Hager AG (2013)

Responses of an HPPD inhibitor-resistant waterhemp (Amaranthus tuberculatus)

population to soil-residual herbicides. Weed Technol. 27:704-411 46

Hay MM (2015) Control of Palmer amaranth (Amaranthus palmeri) and common waterhemp

(Amaranthus rudis) in double crop soybean and with very long chain fatty acid

inhibitor herbicides. M.S. dissertation. Manhattan, KS: Kansas State University.

Heap I (2014) Herbicide Resistant Weeds. Pages 281-301 in Integrated Pest Management

Pesticide Problems Vol. 3. New York: Springer.

Heap I (2018). International Survey of Herbicide Resistant weeds.

www.weedscience.org/in.asp (Accessed January 10, 2018)

Jhala AJ, Malik MS, Willis JB (2015) Weed control and crop tolerance of micro-

encapsulated acetochlor applied sequentially in glyphosate-resistant soybean. Can. J.

Plant Sci. 95:973-981

Kohrt JR, Sprague CL (2017) Herbicide management strategies in field corn for a three-way

herbicide-resistant Palmer amaranth (Amaranthus palmeri) population. Weed Sci.

31:363-372

Letouzé A, Gasquz, J (2003) Enhanced activity of several herbicide-degrading enzymes: a

suggested mechanism responsible for multiple resistance in blackgrass (Alopecurus

myosuriodes Huds.). Agronomie 23:601-608

Meyer CJ, Norsworthy JK, Young BG, Steckel LE, Bradley KW, Johnson WG, Loux MM,

Davis VM, Kruger GR, Bararpour MT, Ikley JY, Spaunhorst DJ, Butts TR (2015)

Herbicide program approaches for managing glyphosate-resistant Palmer amaranth

(Amaranthus palmeri) and waterhemp (Amaranthus tuberculatus) in future soybean-

trait technologies. Weed Technol. 29:716-729 47

Owen MDK (2013) Pest resistance: Overall principles and implications on evolved herbicide

resistance in Iowa. ICM Conference 25:125-136

Steckel LE, Sprague CL, Hager AG (2002) Common waterhemp (Amaranthus rudis) control

in corn (Zea mays) with preemergence and sequential applications of residual

herbicides. Weed Technol. 16:755-761

Tanetai Y, Kaku K, Kawai K, Fujiola T, Shimizu T (2009) Action mechanism of a novel

herbicide, pyroxasulfone. Pest Biochem. Physiol. 95:47-55

Tranel PJ, Bell MS, Hager AG, Riggins CW (2011) Herbicide Resistances in Amaranthus

tuberculatus: A Call For New Options. J. Agric. Food Chem. 59:5805-5812

Trenkamp S, Martin W, Tietjen K (2004) Specific and differential inhibition of very-long-

chain fatty acid elongases from Arabidopsis thaliana by different herbicides. PNAS

101:11903-11908

Yu Q, Abdallah I, Han H, Owen M, Powles S (2009) Distinct non-target site mechanisms

endow resistance to glyphosate, ACCase and ALS-inhibiting herbicides in multiple

herbicide-resistant Lolium rigidum. Planta 230:713-723

Yuan JS, Tranel PJ, Stewart CN (2007) Non-target-site herbicide resistance: a family

business. Trends in Plant Science 12:6-13

Tables

Table 1. Very long chain fatty acid-inhibiting herbicides applied at Grundy and Story

County, Iowa experiment sites in 2016 and 2017.

Common Name Trade Name Application Manufacturer Rates (g ai ha-1)

Acetochlor Harness 7 EC .25x (670) Monsanto, Inc., .5x (1350) St. Louis, MO, 1x (2690) USA 2x (5380) 4x (10,760)

S-metolachlor Dual II Magnum .25x (540) Syngenta Crop 7.64 EC .5x (1070) Protection, Inc., 1x (2140) Greensboro, NC, 2x (4280) USA 4x (6420)

Flufenacet Cadou 4 SC .25x (210) Bayer .5x (420) CropScience, 1x (840) Inc., Research 2x (1680) Triangle Park, 4x (3360) NC,USA

Table 2. Multiple herbicide-resistant waterhemp populations used in the germination chamber very long chain fatty acid-inhibiting herbicide dose-response experiment. Classification Abbreviation Herbicide Location Resistance Profilea

Herbicide- C Susceptible Story County, IA Susceptible

3-Way 3A ALS, PSII, HPPD Henry County, IA Resistant

3-Way 3B ALS, PSII, HPPD Cherokee County, Resistant IA

4-Way 4A ALS, PSII, Monona County, Resistant EPSPS, HPPD IA

4-Way 4B ALS, PSII, Plymouth Resistant EPSPS, HPPD County, IA

5-Way 5A ALS, PSII, Not Recorded Resistant EPSPS, HPPD, PPO 5-Way 5B ALS, PSII, Woodbury Resistant EPSPS, HPPD, County, IA PPO a Abbreviations: ALS, acetolactate synthase; PSII, photosystem II; HPPD, 4- hydroxyphenylpyruvate dioxygenase; EPSPS, 5-enolpyruvylshikimate-3-phosphate synthase; PPO, protoporphrinogen oxidase

Figures

Figure 1. Main plot: Evaluation of acetochlor control for Grundy County (●) in 2016 and a b Story County waterhemp populations (▲) in 2017 21 DAE under field conditions. LD50 2 and LD90 values were calculated from the efficacy curves Y = 89.5 + 12.7 * ln(X), r = .97, and Y = 84.5 + 7.8 * ln(X), r2 = .89, for the Grundy and Story County populations in 2016 and 2017, respectively. c

Inset: Rate of acetochlor to achieve the lethal dose to control 50 (LD50) and 90 (LD90) percent of waterhemp plants at both Grundy and Story County, Iowa sites. LD50 and LD90 are reported in the multiplicative of the maximum labeled rate and (g active ingredient ha-1). aThe Grundy County waterhemp population evolved resistance to herbicide groups 2, 5, 9, and 27, and the Story County waterhemp population evolved resistance to herbicide groups 2, 14, and 27. bAbbreviations: DAE, days after emergence. c Herbicide treatment, rate, and evaluation date were significant effects (P < 0.0001), while the population was not a significant effect (P = 0.90) on waterhemp control. 51

Figure 2. Main plot: Evaluation of s-metolachlor control of Grundy (●) and Story County a b waterhemp populations (▲) 21 DAE under field conditions. LD50 and LD90 values were calculated from the efficacy curves Y = 73.9 + 20.9 * ln(X), r2 = .99, and Y = 70.5 + 23.7 * ln(X), r2 = .94, for the Grundy and Story County populations in 2016 and 2017, respectively. c

Inset: Rate of s-metolachlor to achieve the lethal dose to control 50 (LD50) and 90 (LD90) percent of waterhemp plants at both Grundy and Story County, Iowa sites. LD50 and LD90 are reported in the multiplicative of the maximum labeled rate and (g active ingredient ha-1). aThe Grundy County waterhemp population evolved resistance to herbicide groups 2, 5, 9, and 27 and the Story County waterhemp population evolved resistance to herbicide groups 2, 14, and 27. bAbbreviations: DAE, days after emergence. c Herbicide treatment, rate, and evaluation date were significant effects (P < 0.0001), while the population was not a significant effect (P = 0.90) on waterhemp control.

Figure 3. Main plot: Evaluations of flufenact control of Grundy (●) and Story County a b waterhemp populations (▲) 21 DAE under field conditions. LD50 and LD90 values were calculated from the efficacy curves Y = 55.8 + 28.0 * ln(X), r2 = .91, and Y = 82.1 + 18.2 * ln(X), r2 = .96, for the Grundy and Story County populations in 2016 and 2017, respectively. c

Inset: Rate of flufenacet to achieve the lethal dose to control 50 (LD50) and 90 (LD90) percent of waterhemp plants at both Grundy and Story County, Iowa sites. LD50 and LD90 are reported in the multiplicative of the maximum labeled rate and (g active ingredient ha-1). aThe Grundy County waterhemp population evolved resistance to herbicide groups 2, 5, 9, and 27 and the Story County waterhemp population evolved resistance to herbicide groups 2, 14, and 27. bAbbreviations: DAE, days after emergence. c Herbicide treatment, rate, and evaluation date were significant effects (P < 0.0001), while the population was not a significant effect (P = 0.90) on waterhemp control.

Figure 4. Main plot: Evaluations of acetochlor control of 3A (■), 3B (□), 4A (▲), 4B (Δ), 5A (●), 5B (○), and C (X) waterhemp populations a 2 DAT b under germination chamber conditions. Lethal dose concentrations to control 50 (LD50) and 90 (LD90) percent of the waterhemp plants were calculated from the efficacy curves Y = 49.6 + 22.4 * ln(X), r2 = .75, Y = 62.3 + 16.2 * ln(X), r2 = .78, Y = 41.1 + 20.6 * ln(X), r2 = .86, Y = 52.9 + 20.3 * ln(X), r2 = .94, Y = 55.2 + 18.9 * ln(X), r2 = .77, Y = 44.0 + 20.0 * ln(X), r2 = .77, Y = 45.8 + 18.6 * ln(X), r2 = .81 for the 3A, 3B, 4A, 4B, 5A, 5B, and C populations, respectively. c

Inset: Concentration (ppm) of acetochlor required to achieve the LD50 and LD90 of the waterhemp plants. a 3A,B = evolved resistance to herbicide groups (HG) 2, 5, and 27, 4A,B = evolved resistance to HG 2, 5, 9, and 27, 5A,B = evolved resistance to HG 2, 5, 9, 14, and 27, C = herbicide susceptible. b Abbreviations: DAT, days after treatment. c Herbicide treatment, rate, and population were significant effects (P < 0.0001) on waterhemp control.

Figure 5. Main plot: Evaluations of s-metolachlor control of 3A (■), 3B (□), 4A (▲), 4B (Δ), 5A (●), 5B (○), and C (X) waterhemp populations a 2 DAT b under germination chamber conditions. Lethal dose concentrations to control 50 (LD50) and 90 (LD90) percent of the waterhemp plants were calculated from the efficacy curves Y = 66.3 + 15.7 * ln(X), r2 = .96, Y = 55.5 + 16.2 * ln(X), r2 = .91, Y = 37.5 + 13.1 * ln(X), r2 = .83, Y = 37.5 + 13.1 * ln(X), r2 = .69, Y = 43.5 + 18.3 * ln(X), r2 = .72, Y = 39.7 + 14.2 * ln(X), r2 = .66, Y = 41.2 + 14.6 * ln(X), r2 = .84 for the 3A, 3B, 4A, 4B, 5A, 5B, and C populations, respectively. c

Inset: Concentration (ppm) of s-metolachlor required to achieve the LD50 and LD90 of the waterhemp plants. a 3A,B = evolved resistance to herbicide groups (HG) 2, 5, and 27, 4A,B = evolved resistance to HG 2, 5, 9, and 27, 5A,B = evolved resistance to HG 2, 5, 9, 14, and 27, C = herbicide susceptible. b Abbreviations: DAT, days after treatment. c Herbicide treatment, rate, and population were significant effects (P < 0.0001) on waterhemp control.

Figure 6. Main plot: Evaluation of flufenacet control of 3A (■), 3B (□), 4A (▲), 4B (Δ), 5A (●), 5B (○), and C (X) waterhemp populations a 2 DAT b under germination chamber conditions. Lethal dose concentrations to control 50 (LD50) and 90 (LD90) percent of the waterhemp plants were calculated from the efficacy curves Y = 17.2 + 3.0 * ln(X), r2 = .09, Y = 20.6 + 2.4 * ln(X), r2 = .07, Y = 16.2 + -2 * ln(X), r2 = .56, Y = 20.6 + .008 * ln(X), r2 = .11, Y = 31.0 + 3.8 * ln(X), r2 = .17, Y = 17.5 + 2.7 * ln(X), r2 = .19, Y = 20.2 + 1.5 * ln(X), r2 = .59 for the 3A, 3B, 4A, 4B, 5A, 5B, and C populations, respectively. c

Inset: Concentration (ppm) of flufenacet required to achieve the LD50 and LD90 of the waterhemp plants. a 3A,B = evolved resistance to herbicide groups (HG) 2, 5, and 27, 4A,B = evolved resistance to HG 2, 5, 9, and 27, 5A,B = evolved resistance to HG 2, 5, 9, 14, and 27, C = herbicide susceptible. b Abbreviations: DAT, days after treatment. c Herbicide treatment, rate, and population were significant effects (P < 0.0001) on waterhemp control. 56

CHAPTER 4. INVESTIGATING A PUTATIVE HPPD-INHIBITING HERBICIDE- RESISTANT AMBROSIA TRIFIDA POPULATION A paper to be submitted to Weed Technology Investigating a Putative HPPD-Resistant Ambrosia trifida Population

Eric A. L. Jones and Micheal D. K. Owen*

*First and second authors: Graduate Assistant and University Professor Emeritus,

Department of Agronomy, Iowa State University, Ames, IA 5001-1010.

Abstract

An Ambrosia trifida L. (giant ragweed) population in northeast Iowa was reported in 2015 to be poorly controlled by mesotrione, a 4-hydroxyphenylpyruvate dioxygenase-inhibiting herbicide (HPPD) (EC 1.13.11.27). The farmer reported poor control of the giant ragweed only several years after the first HPPD-inhibiting herbicide application. If the evolution of resistance to the HPPD-inhibiting herbicides is confirmed, this will be a novel resistance in giant ragweed. An initial greenhouse screen with a field rate (105 g ai ha-1) of mesotrione applied to plants 7 cm in height resulted in differential responses to the herbicide ranging from plant death to survival. The variable response of the giant ragweed population suggested the population was in transition from a sensitive population to a resistant population. An on-farm field experiment examining HPPD-inhibiting herbicides applied postemergence were not statistically different from each other (P = 0.79), and the average efficacies were mesotrione (74%), tembotrione (81%), and topramezone (70%). However, a tembotrione dose-response assay conducted under greenhouse conditions provided results that the putative HPPD-resistant giant ragweed population responded to the herbicide 57 applications the same as a HPPD-susceptible giant ragweed population (P = 0.27). The results of this research suggest that the northeast Iowa giant ragweed population of is in a transition state moving from susceptible to resistant to the HPPD-inhibiting herbicides.

Introduction

Giant ragweed (Ambrosia trifida L.) is a highly competitive summer annual broadleaf weed in the Asteraceae family. Unlike many agronomic weeds, giant ragweed exhibits relatively low fecundity and seeds are short-lived in the soil seedbank (Harrison et al. 2001). Traits that promote the weediness of this species are early emergence, rapid early-season growth, ability to adapt to adverse environmental conditions, and ability to reach heights above the crop canopy (Abul-Fatih and Bazzaz 1979). Relatively recently, giant ragweed has been colonizing crop fields and considered one of the most difficult weeds to control by means of chemical inputs (Regnier et al. 2016).

Weed management in Midwest crop fields relies heavily on chemical control. Tillage and herbicides are the strongest selection agents for evolution of weed populations and recurrent herbicide applications will cause the weed population to shift from herbicide-susceptible to herbicide-resistant (Zelaya and Owen 2005). Giant ragweed populations have been confirmed with single or multiple resistance to acetolactate synthase (ALS) (EC 2.2.1.6) inhibitors and 5-enolpyruvylshikimate-3-phosphate synthase inhibitors (EPSP) (EC 2.5.1.19)

(Heap 2018).

A population of giant ragweed in northeast Iowa was reported in 2015 to be poorly controlled by mesotrione, a 4-hydroxyphenylpyruvate dioxygenase-inhibiting herbicide (HPPD) (EC

1.13.11.27). Soil samples containing giant ragweed seeds were collected from the Floyd 58

County, Iowa field in the spring 2016 and were transferred to the greenhouse to initiate germination. An initial non-replicated screen suggested that the population displayed differential responses to mesotrione. There has not been a confirmed case of a giant ragweed population with evolved resistance to the HPPD-inhibiting herbicides.

HPPD-inhibiting herbicides offer high efficacy for many broadleaf weeds and some grass weeds while not imposing harm to the crop (Beaudegnies et al. 2009). Resistance to HPPD- inhibiting herbicides has been reported in two species, waterhemp (Amaranthus tuberculatus

Moq. J.D. Sauer) and Palmer amaranth (Amaranthus palmeri S. Wats.) (Heap 2018). The resistance to these herbicides is not widespread across the United States and has only been confirmed in Iowa, Illinois, Nebraska, and Kansas (Heap 2018). HPPD-inhibiting herbicide usage has increased in maize to control glyphosate-resistant weeds (Green and McMullan

2011, Green 2012). Increased reliance on HPPD-inhibiting could result in the evolution of the HPPD-inhibiting herbicides become more common in other weed species (Gressel 2009,

Gressel et al. 2016, Green 2012). This instance of a putative HPPD-resistant giant ragweed population has been the only known case reported to date. Therefore, the objectives of this study were to confirm the evolution of HPPD-inhibiting herbicide resistance in a giant ragweed population and determine the efficacy of herbicides with different modes of action.

The hypothesis is the giant ragweed population has evolved resistance to HPPD-inhibiting herbicides.

Material and Methods

Field Experiments. HPPD inhibiting herbicides and non-HPPD-inhibiting herbicides. Field experiments were conducted in the summer 2016 and 2017 to determine if a Floyd County,

Iowa giant ragweed population demonstrated the differential response to HPPD-inhibiting 59 herbicides exhibited in the initial greenhouse screen. The soil type for the Floyd County field site was a Clyde silty clay loam (fine-loamy, mixed, superactive, mesic typic endoaquolls) with a pH of 7.3 and 4.5% organic matter. The design of the experiment was a randomized complete block with each treatment replicated three times. Plots were 3.0 x 7.6 m in size and herbicides were applied when giant ragweed was approximately 7 cm in height using 140 L

-1 ha carrier volume through a CO2-powered backpack sprayer with TeeJet TT110015 nozzles.

Herbicides in the experiment were applied post emergence (POST) with labeled adjuvants and at maximum labeled rates (Table 1). Visual efficacy evaluations were conducted 7, 14, and 21 days after treatment (DAT) on a scale from 0% to 100% where 0% represents no control and 100% represents complete control. Giant ragweed population density assessments were conducted 21 DAT with 0.3 x 0.3 m squares arbitrarily placed in each plot where an HPPD-inhibiting herbicide was applied. Percent control was calculated by using the formula:

C = T / U where C equals the percent control, T equals the number of plants in the treated plots, and U equals the number plants in the untreated plots.

Dose-response field experiment. The dose-response experiment under field conditions was conducted in 2017. Mesotrione, tembotrione, and topramezone were applied at 0.25x, 0.5x,

1x, 2x, and 4x as previously described where the 1x amount was based on the maximum labeled rate for the respective herbicide along with labeled adjuvants. Visual efficacy evaluations were conducted 7, 14, and 21 DAT on a scale from 0% to 100% where 0% represents no control and 100% represents complete control. 60

Greenhouse Dose-Response Experiment. Giant ragweed seedlings were collected from the

Floyd County, IA field and the known HPPD-susceptible population was collected from

Curtiss Farm, Ames, IA. Seedlings were transplanted into Ray Leach Cone-tainers™

(Stuewe & Sons, Inc., Tangent, OR, USA) containing a 4:1 ratio of Sunshine Mix #1/LC1 potting soil and sand with approximately 1 g of Osmocote Flower Food Granules (14-14-14).

Plants were maintained in the greenhouse at 24°C and watered as needed. Sunlight was supplemented with 600-1,000 µmol m-2 s-1 PFFD of artificial light set to a 14-hr photoperiod.

Tembotrione was applied in a spray chamber when plants were approximately 7 cm in height

-1 using 191.8 L ha carrier volume through a CO2-powered spray chamber with TeeJet

80015EVS nozzles. The range of herbicide rates were 1 to 10,000 g ai ha-1 equally spaced along a 3.16 logarithmic scale. All treatments contained methylated seed oil (1% v/v) and ammonium sulfate (146 g L-1). Plants were arranged in a randomized complete block design with 10 replications and the experiment was conducted twice. Plants were cut at the soil surface 21 DAT. Plant tissue was placed in a dryer at 65° C for 48 hours and dry weight measured. Dry weight reduction was calculated using the formula:

D = DT / DU where D equals the reduction in dry weight, DT equals the dry weight of the treated plant, and DU equals the dry weight of the untreated plant.

Statistical Analysis. All statistical analyses were performed using Statistical Analysis

Software, SAS 9.3 (Statistical Analysis Systems version 9.3, SAS Institute, Cary, NC). Data for control efficacy for both field and lab experiments were subject to ANOVA using the

GLIMMIX procedure in SAS. Means and interactions of significant effect were separated using Fisher’s protected LSD test at P ≤ 0.05. 61

Dose-response curves for the experiments were created using a three-parameter exponential decay curve equation was used to calculate herbicide dose that caused a 50% reduction in dry weight (GR50) and the 80% reduction in dry weight (GR80):

Y = Y0 + A * exp (-B*X)

where Y0, A, and B are constants, Y = percent of control, and X = herbicide rate. Deviations from the curve are described by r2. The data points from experiments were used to derive the model equations on SigmaPlot 12.5 (Systat Software, Inc. version 12.5, San Jose, CA).

Resistance ratios (R/S) were calculated by dividing the GR50 of the putative resistant giant ragweed population by the GR50 of the susceptible population.

Results and Discussion

HPPD-inhibiting herbicide efficacy assays under field conditions. Herbicide treatment and evaluation date were not significant effects (P = 0.62, 0.44) for the visual efficacy evaluation in 2016. Giant ragweed control by an HPPD-inhibiting herbicide was never greater than 70% (Table 2). Giant ragweed population densities provided further evidence that the HPPD-inhibiting herbicides provided similar efficacies. Stand counts revealed the efficacies for mesotrione, tembotrione, and topramezone were 74%, 81%, and 70% respectively (Figure 1). Collectively, the control of the giant ragweed was poor. Variability between plots could denote that the distribution of resistant plants was not uniform in the field. These results provide evidence that evolution of HPPD-inhibiting herbicide resistance may be in early stages for the Floyd County, Iowa giant ragweed population.

Giant ragweed efficacy with non-HPPD-inhibiting herbicides under field conditions.

Herbicide treatment and evaluation date were significant effects (P < 0.0001, 0.0009) for 62 efficacy. Herbicide treatments achieved higher efficacy 14 and 21 DAT when compared to 7

DAT (Table 3). Herbicide efficacy was not significantly different between 14 and 21 DAT evaluations. 2,4-D, dicamba, and dicamba + diflufenzoypr did not provide giant ragweed control until 14 DAT, but provided the highest efficacy 21 DAT (Table 3). Atrazine did not provide efficacy greater than 75% until it was tank-mixed with 2,4-D, dicamba, and dicamba

+ diflufenzoypr (Table 3). Glufosinate provided control greater than 85% 7 DAT. However, without any residual control, by 14 and 21 DAT new giant ragweed seedlings had emerged.

Glyphosate provided only 43% control 14 DAT. Evolution of glyphosate resistance is apparent in the Floyd County giant ragweed population due the low efficacy. Overall, the results of this experiment provided further evidence that giant ragweed is hard to control with herbicides, as no herbicide provided efficacy over 90% (Table 3).

The experiments at the Floyd County field in 2017 were compromised and data could not be collected. A maximum-labeled rate of acetochlor (2690 g ai ha-1) was applied to the experimental area to minimize grass and small-seeded broadleaf pressure. Acetochlor is not labeled to control giant ragweed alone, however due to environmental impacts, a majority of the giant ragweed plants were controlled by the herbicide application. Plots could not be relocated as the farmer applied herbicides that were efficacious to giant ragweed on the rest of the field. Thus, the experiments were terminated.

HPPD-inhibiting herbicide dose-response assay under greenhouse conditions. Herbicide rate was a significant effect (P < 0.0001) on the reduction of plant dry biomass. The putative

HPPD-resistant giant ragweed population responded to the herbicide treatment the same as a

HPPD-susceptible giant ragweed population (P = 0.27) (Figure 2). These results suggest that resistance to the HPPD-inhibiting herbicides has not evolved in the Floyd County giant 63 ragweed population or the collected seedlings used in the experiment were the susceptible phenotype.

-1 GR50 values were 0.68 and 1.48 g ai ha for the Story and Floyd County populations, respectively (Figure 2). The R/S ratio was 2.17, indicating there is a differential response at the lower rates between the two giant ragweed populations. R/S ratios need to be significantly different than 1 to suggest the evolution of herbicide resistance (Burgos et al.

2013). However, the GR80 herbicide rates were similar for the Story and Floyd County populations. GR80 corresponded with plant death and the herbicide rates were well below the maximum-labeled rate (Figure 2). This result provides further evidence that HPPD-inhibiting herbicides are efficacious for both giant ragweed populations and the Floyd County giant ragweed population in early stages of evolving resistance to the HPPD-inhibiting herbicides.

The lack of differential responses between the putative resistant and susceptible giant ragweed populations could be imparted by samples containing only the susceptible phenotype from the field. The field experiment could have been established in an area of the field where the HPPD-resistant plants existed and thus the progeny in the immediate area were resistant. Seeds were collected from plants surviving herbicide application, but would not germinate in the greenhouse or germination chamber. Future work should focus on using seeds from the surviving plants or collecting seeds were the surviving plans were in the field.

Additional research could investigate how to make germinating giant ragweed in controlled environments less difficult and develop a protocol for the weed science community to utilize.

Differential responses of the Floyd and Story County giant ragweed populations could have been confounded by the cone-tainers where the plants were maintained. The cone-tainers may have been too small to sustain the large giant ragweed plants. Untreated plants were 64 prematurely senescing due to root growth inhibition. It is possible treated putative resistant plants prematurely senesced as well. This phenomenon could confound differential responses from putative resistant and susceptible giant ragweed populations. The results of the greenhouse dose-response assay are inconclusive. The experiment may yield different results if the giant ragweed plants were maintained in larger pots.

While giant ragweed may be susceptible to a specific herbicide site-of-action, control failures can occur by variables such as herbicide dose, adjuvant choice, adverse environmental conditions, and application timing (Boutsalis 2001). HPPD-inhibiting herbicides should control giant ragweed populations according to the herbicide labels. The results from this study provides evidence that the Floyd County giant ragweed population could be in the early stages of evolving resistances to the HPPD-inhibiting herbicides.

Literature Cited

Abul-Fatih HA and Bazzaz FA (1979) The biology of Ambrosia trifida L. I. influence of

species removal on the organization of the plant community. New Phytol. 83:813-816

Beaudegnies R, Edmunds AJF, Torquil FEM, Hall RG, Hawkes TR, Mitchell Glynn,

Schaetzer J, Wendeborn S, Wibley J (2009) Herbicidal 4-hydroxyphenylpyruvate

dioxygenase inhibitors-A review of the triketone chemistry story from a Syngenta

perspective. Bioorg. Med. Chem. 17:4134-4152

Boutsalis P (2001) Sygenta quick-test: a rapid whole-plant test for herbicide resistance. Weed

Technol. 15:257-263 65

Burgos NR, Tranel PJ, Streibig JC, Davis VM, Shaner D, Norsworthy JK, Ritz C (2013)

Review: Confirmation of resistance to the herbicides and evaluation of resistance

levels. Weed Sci. 61:4-20

Green JM (2012) The benefits of herbicide-resistant crops. Pest Manag. Sci. 86:1323-1331

McMullan PM and Green JM (2011) Identification of a tall waterhemp (Amaranthus

tuberculatus) biotype resistant to HPPD-inhibiting herbicides, atrazine, and

thifensulfuron. Weed Technol. 25:514-518

Gressel J (2009) Evolving understanding of the evolution of herbicide resistance. Pest

Manag. Sci. 65:1164-1173

Gressel J, Gassmann AJ, Owen MDK (2016) How well will stacked transgenic pest/herbicide

resistances delay pest from evolving resistance? Pest Manag. Sci. 73:22-34

Harrison SK, Regnier EE, Schmoll JT, Webb JE (2001) Competition and fecundity of giant

ragweed in corn. Weed Sci. 49:224-229

Heap I (2018) International survey of herbicide resistant weeds. www.weedscience.org

(accessed January 10, 2018)

Owen MDK, Zelaya IA (2005) Herbicide-resistant crops and weed resistance to herbicides.

Pest Manag. Sci. 61:301-311

Tables Table 1. Herbicide treatments applied post emergence to giant ragweed at the maximum labeled rate used in the field experiment to determine efficacy of Floyd County, Iowa in 2016. Common name Trade name Application Rates Manufacturer (g ai or ae ha-1)

2,4-D LV Weedone 4 SL (267) Nufarm Americas, Inc., Alsip, IL

Atrazine Aatrex 4 SC (2240) Syngenta Crop Protection, Inc., Greensboro, NC 2,4-D LV + atrazine (1370)

Dicamba Clarity 4 SL (560) BASF Crop Protection, Inc., Research Triangle Park, NC, USA

Dicamba + Status (392) BASF Crop Protection, Inc., Research Triangle diflufenzopyr Park, NC, USA

Dicamba + atrazine Marksman 3 F (1570) BASF Crop Protection, Inc., Research Triangle Park, NC, USA

Glufosinate Liberty 280 SL (450) Bayer CropScience, Inc., Research Triangle Park, NC, USA

Glyphosate Roundup (1460) Monsanto, Inc., St. Powermax 5 L Louis, MO

Mesotrione Callisto 4 EC (105) Syngenta Crop Protection, Inc., Greensboro, NC

Tembotrione Laudis (92) Bayer CropScience, Inc., Research Triangle Park, NC, USA

Tompramezone Impact 3 SC (25) AMVAC Chemical Corporation, Inc., Los Angeles, CA

Table 2. Efficacy of HPPD-inhibiting herbicides applied post emergence to giant ragweed on the Floyd County, Iowa at 7, 12, and 21 DAT in 2016. a,b,c % Control % Control % Control Herbicide 7 DAT 14 DAT 21 DAT Mesotrione 57 a 68 a 58 a

Tembotrione 57 a 62 a 67 a

Topramezone 50 a 60 a 40 a a Abbreviations: DAT, days after treatment b Means presented within each column with no common letter(s) are significantly different according to Fisher’s protected LSD test where P ≤ 0.05 c Herbicide treatment and evaluation date were not significant effects (P = 0.62, 0.44) on giant ragweed control.

Table 3. Efficacy of herbicides applied post emergence to giant ragweed under field conditions on the Floyd County, Iowa at 7, 12, and 21 DAT in 2016.a,b,c

% Control % Control % Control Herbicide 7 DAT 14 DAT 21 DAT

2,4-D 0 g 62 cde 50 cde Atrazine 78 abc 75 abc 68 abcd 2,4-D + atrazine 63 bcde 72 abcd 73 abcd Dicamba 0 g 80 abc 87 ab Dicamba + atrazine 67 abcde 88 a 90 a Dicamba + 0 g 60 cde 77 abc diflufenzopyr Glyphosate 30 f 43 ef 30 f Glufosinate 87 ab 0 g 0 g

a Abbreviations: DAT, days after treatment. b Means presented within each column with no common letter(s) are significantly different according to Fisher’s protected LSD test where P ≤ 0.05. c Herbicide treatment and evaluation date were significant effects (P < 0.0001, P = 0.0009) on giant ragweed control.

Figures

Untreated Mesotrione Tembotrione Topramezone

100

)

1 -

2 2 90 80 70 60 50 40 30 20 10 Plant density (Plants meter (Plants density Plant 0

Figure 1. Floyd County, Iowa giant ragweed population density counts 21 DAT for HPPD- inhibiting herbicides applied post emergence at the maximum-labeled rate. a, b, c a Abbreviations: Days after treatment, DAT. b Means presented within each column with no common letter(s) are significantly different according to Fisher’s protected LSD test where P ≤ 0.05. c The error bars represent the standard error of each treatment mean.

Figure 2. Main plot: Tembotrione dose-response curves for the Story (●) and Floyd (▲)

County, Iowa giant ragweed populations 21 days after treatment under greenhouse

a, b conditions. Herbicide rates that caused growth reduction by 50 (GR50) and 80 (GR80) percent were calculated from the efficacy curves Y = 19.2 + 80.5 * exp(-1.5 * X), r2 = 0.96, and Y = 12.6 + 86.7 * exp(-0.6 * X), r2 = .99, for the Story and Floyd County populations, respectively.

Inset: Rate of tembotrione required to achieve the GR50 and GR80 of giant ragweed plants from both Floyd [R] and Story County [S], Iowa sites. 71 a Herbicide rate was a significant effect (P < 0.0001), while the population was not a significant effect (P = 0.27) on giant ragweed control. b Story County = HPPD-inhibiting herbicide-susceptible giant ragweed population, Floyd

County = Putative HPPD-inhibiting herbicide-resistant giant ragweed population.

CHAPTER 5. FINAL CONCLUSIONS

Over 60 years ago, Harper (1956) predicted that weeds evolving resistance to herbicides before a herbicide-resistant weed population had been identified. The first herbicide-resistant weed population was confirmed in 1968 (Ryan 1970). Historically, herbicide-resistant weeds have been an oddity, however now herbicide-susceptible weeds are becoming an oddity. The current situation of 487 unique cases of herbicide-resistant weed globally is a good indication that the evolution and spread of herbicide resistance is unlikely to cease (Heap 2018).

Waterhemp and giant ragweed control will only increase in cost and complexity if new herbicide resistance(s) evolve.

The first objective of this research was to determine if select multiple herbicide-resistant waterhemp populations incurred a fitness penalty. Measurements of relative growth rate, flowering, accumulated biomass, and seed production provided evidence that multiple herbicide-resistant waterhemp grows and develops the same as herbicide-susceptible waterhemp. Since multiple herbicide-resistant waterhemp is not at an ecological disadvantage in herbicide-free field conditions, the populations are likely to remain within the agroecosystem.

Target-site and non-target-site resistance mechanisms have both been demonstrated to incur a fitness penalty (Vila-Auib et al. 2005, Menchari et al. 2008). Additional research is needed to understand the mechanisms of resistance in each multiple herbicide-resistant waterhemp populations. This knowledge would provide more evidence that neither target- or non-target- site resistance mechanism(s) in waterhemp incurs a fitness penalty. 73

The second objective of this research was to determine if the very long chain fatty acid

(VLCFA)-inhibiting herbicide efficacy on multiple herbicide-resistant waterhemp populations differed when compared to a herbicide-susceptible waterhemp population. Dose- response assays in the field and greenhouse confirmed that the multiple herbicide-resistant and herbicide-susceptible waterhemp populations respond to VLCFA-inhibiting herbicides the same. Acetochlor provided the highest efficacy followed by s-metolachlor then and flufenacet. VLCFA-inhibiting herbicides are currently a viable means to control waterhemp.

However, non-target-site herbicide resistance usually endows cross-resistance to unrelated herbicides (Beckie and Tardif 2012). If none of the multiple herbicide-resistant waterhemp populations possess a non-target-site resistance mechanism(s), it could explain why the

VLCFA-inhibiting herbicide efficacy remained high and none of the populations expressed a differential response to herbicide application.

The third objective of this research was to confirm the evolution of resistance to 4- hydroxyphenylpyruvate dioxygenase (HPPD)-inhibiting herbicides in the giant ragweed population and determine the efficacy of herbicides within different groups. The 1x rate of

HPPD-inhibiting herbicides did not achieve efficacy greater than 75%. All other herbicides applied in the field, besides glyphosate, provided at least 80% efficacy. However, dose- response assays results demonstrated that the putative HPPD-resistant giant ragweed population and HPPD-susceptible giant ragweed population responded the same to HPPD- inhibiting herbicide application. The initial findings of the research suggested that the Floyd

Country giant ragweed population is in a transition stage from susceptible to resistant.

Additional research needs to be conducted on the putative HPPD-resistant and HPPD- susceptible giant ragweed populations. Field dose-response assays need to be conducted in 74 the field to determine if the both populations exhibits response to herbicide applications similarly to the responses in the greenhouse. Additional site years will provide more statistical power to the research as well.

BIBLIOGRAPHY

Abul-Fatih HA and Bazzaz FA (1979) The biology of Ambrosia trifida L. I. influence of

species removal on the organization of the plant community. New Phytol. 83:813-816

Bassett IJ and Crompton CW (1982) The biology of Canadian weeds. 55. Ambrosia trifida L.

Can. J. Plant Sci. 62:100-3-1010

Baysinger JA and Sims BA (1991) Giant Ragweed (Ambrosia trifida) interference in

soybeans (Glycine max). Weed Sci. 39:358

Beckie HJ, Tardif FJ (2012) Herbicide cross resistance in weeds. Crop Protect. 35:15-28

Bell MS, Hager AG, Tranel PJ (2013) Multiple resistance to herbicides from four site-of-

action groups in waterhemp (Amaranthus tuberculatus). Weed Sci. 61:460-468

Bhowmik PC (1997) Weed biology: importance to weed management. Weed Sci. 45:249-356

Böger P (2003) Mode of action for chloroacetamides and functionally related compounds. J.

Pestic. Sci 28:324-329

Böger P, Matthes B, Scmalfuß J (2000) Towards the primary target of chloroacetamides-new

findings pave the way. Pest Manag. Sci. 56:497-508

Brabham CB, Gerber CK, and Johnson WG (2011) Fate of glyphosate-resistant giant

ragweed (Ambrosia trifida) in the presence and absence of glyphosate. Weed Sci.

59:506-511

Bram MR and Quinn JA (2000) Sex expression, sex-specific traits, and the effects of salinity

on growth and reproduction of Amaranthus cannabinus (Amaranthaceae), a dioecious

annual. Am J Bot 87: 1609-1618 76

Busi R and Powles SB (2016) Cross-resistance to prosulfocarb + S-metolachlor and

pyroxasulfone selected either herbicide in Lolium rigidum. Pest Manag. Sci. 72:1664-

1672

Busi R, Gaines TA, Walsh MJ, Powles SB (2012) Understanding the potential for resistance

evolution to the new herbicide pyroxasulfone: field selection at high doses versus

recurrent selection at low doses. Weed Research 52:489-499

Coustau C, Christine C, ffrench-Constant R (2000) Resistance to xenobiotics and parasites:

can we count the cost? Trends in Ecol. Evolut. 15:378-383

Darmency H, Menchari Y, Le Corre, Délye C (2014) Fitness cost due to herbicide resistance

may trigger genetic background evolution. Evolution 69:271-278

Darwin C (1859) On the origin of species by means of natural selection or the preservation of

favoured races in the struggle for life. 6th ed. London, England: Penguin Books.

Davis AS and Frisvold GB (2017) Are herbicides a once in a century method of weed

control? Pest Manag. Sci. 72:2209-2220

Délye C, Jasieniuk M, Le Corre V (2013) Deciphering the evolution of herbicide resistance

in weeds. Trends Genet. 29:649-658

Fuerst PE (1987) Understanding the mode of action of the chloroacetamide and

thiocarbamate herbicides. Weed Technol. 1:270-277

Goplen JJ (2015) Giant ragweed (Ambrosia trifida) seed bank dynamic and management.

M.S. dissertation. Minneapolis, MN: University of Minnesota. 77

Gressel J (2009) Evolving understanding of the evolution of herbicide resistance. Pest

Manag. Sci. 65:1164-1173

Gressel J and Segel LA (1978) The paucity of plants evolving genetic resistance to

herbicides: possible reasons and implications. J. theor. Biol. 75:349-371

Gressel J, Gassman AJ, Owen MDK (2016) How well will stacked transgenic pest/herbicide

resistances delay pests from evolving resistance? Pest Manag. Sci. 73: 22-34.

Harper JL (1956) The evolutions of weeds in relations to resistance to herbicides.

Proceedings of the 3rd British Weed Control Conference, Nov. 5-8, British Weed

Control Council, Farnham, UK. 179-188

Harrison SK, Regnier EE, Schmoll JT (2003) Postdispersal predation of giant ragweed

(Ambrosia trifida) seed in no-tillage corn. Weed Sci. 51:955-964

Harrison SK, Regnier EE, Schmoll JT, Webb JE (2001) Competition and fecundity of giant

ragweed in corn. Weed Sci. 49:224-229

Hartzler RG, Battles BA, Nordby DE (2004) Effect of common waterhemp (Amaranthus

rudis) emergence date on growth and fecundity in soybean. Weed Sci. 52:242-245

Hartzler RG, Buhler DD, Stoltenberg DE (1999) Emergence characteristics of four annual

weed species. Weed Sci. 47:578-584

Heap I (2018) International survey of herbicide resistant weeds. www.weedscience.org

(accessed January 10, 2018)

Hinz JRR, Owen MDK (1997) Acetolactate synthase resistance in a common waterhemp

(Amaranthus rudis) population. Weed Technol. 11:13-18 78

Kniss AR (2017) Long-term trends in the intensity and relative toxicity of herbicide use. Nat.

Comms. 8: 14865

Kohrt JR, Sprague CL (2017) Herbicide management strategies in field corn for a three-way

herbicide-resistant Palmer amaranth (Amaranthus palmeri) population. Weed Sci.

31:363-372

Liebman M, Gallandt ER (1997) Many little hammers: ecological management of crop-weed

interactions. P. 291-343. In L.E. Jackson (ed.) Ecology in agriculture. Academic

Press, San Diego, CA.

Ma R, Kaundun SS, Tranel PJ, Riggins CW, McGinness DL, Hager AG, Hawkes T,

McIndoe E, Riechers DE (2013) Distinct detoxification mechanisms confer resistance

to mesotrione and atrazine in a population of waterhemp. Plant Physiol. 163:363–377

Matringe M, Sailland A, Pelissier B, Rolland A, Zink O (2005) p-Hydroxyphenylpyruvate

dioxygenase inhibitor-resistant plants. Pest Manag. Sci. 61:269-276

McMullan PM and Green JM (2011) Identification of a tall waterhemp (Amaranthus

tuberculatus) biotype resistant to HPPD-inhibiting herbicides, atrazine, and

thifensulfuron. Weed Technol. 25:514-518

Menchari Y, Chauvel B, Darmency H, Déyle C (2008) Fitness costs associated with three

mutant acetyl-coenzyme A carboxylase alleles endowing herbicide resistance in

black-grass Alopecurus myosuroides. J. Appl. Ecol. 45:939-947

Nordby DE and Hartzler RG (2004) Influence of corn on common waterhemp (Amaranthus

rudis) growth and fecundity. Weed Sci. 52:255-259 79

Owen MDK (2013) Pest resistance: Overall principles and implications on evolved herbicide

resistance in Iowa. ICM Conference 25:125-136

Owen MDK, Beckie HJ, Leeson JY, Norsworthy JK, Steckel, LE (2015) Integrated pest

management and weed management in the United States and Canada. Pest Manag.

Sci. 71: 357-376.

Owen MDK, Zelaya IA (2005) Herbicide-resistant crops and weed resistance to herbicides.

Pest Manag. Sci. 61:301-311

Park KW and Mallory-Smith CA (2005) Multiple herbicide resistance in downy brome

(Bromus tectorum) and its impact on fitness. Weed Sci. 53:780-786

Powles, SB and Yu Q (2010) Evolution in action: plants resistant to herbicides. Ann Rev

Plant Biol 61:317-347

Pratt, DB and Clark, LG (2001) Amaranthus rudis and A. tuberculatus—one species or two?

J. Torrey Bot. Soc. 128:282–296

Primack RB and Kang H (1989) Measuring fitness and natural selection in wild plant

populations. Annu. Rev. Ecol. Evol. Systs. 20:367-396

Regnier EE, Harrison SK, Loux MM, Holloman C, Venkatesh R, Diekmann F, Taylor R,

Ford RA, Stoltenberg DE, Hartzler RG, Davis AS, Schutte BJ, Cardina J, Mahoney

KJ, Johnson WG (2016) Certified crop advisors’ perceptions of giant ragweed

(Ambrosia trifida) distribution, herbicide resistance, and management in the corn belt.

Weed Sci. 64:361-377 80

Reinbothe S, Reinbothe C, Apel K, Lebedev N (1996) Evolution of chlorophyll biosynthesis-

the challenge to survive photooxidation. Cell 86:703-705

Ryan GF (1970) Resistance of common groundsel to simazine and atrazine. Weed Sci.

18:614-616

Sauer JD (1955) Revision of the dioecious amaranths. Mandroño 13:5-46

Schryver MG, Soltani N, Hooker DC, Robinson DE, Tranel PJ, Sikkema PH (2017) Control

of glyphosate-resistant common waterhemp (Amaranthus tuberculatus var. rudis) in

soybean in Ontario. Weed Technol. 37:811-821

Soltani N, Vyn JD, Swanton CJ, Weaver S, Sikkema PH (2008) Growth and fitness of

triazine-susceptibe and triazine-resistant common waterhemp (Amaranthus

tuberculatus var. rudis) Weed Biol. Manag. 8:73-77

Steckel LE (2007) The dioecious Amaranthus spp.: here to stay. Weed Technol. 21:567-570

Swanton CJ and Weise SF (1991) Integrated weed management: the rationale and approach.

Weed Technol. 5:657-663

Vila-Aiub MM, Neve P, Powles SB (2005) Resistance cost of a cytochrome P450 herbicide

metabolism mechanism but not an ACCase target site mutation in a multiple resistant

Lolium rigidum population. New Phytol 167:787–796

Vila-Auib MM, Neve P, Powles SB (2009) Fitness costs associated with evolved herbicide

resistance alleles in plants. New Phytol. 184:751-767

Wu C, Davis AS, Tranel PJ (2017) Limited fitness costs of herbicide-resistance traits in

Amaranthus tuberculatus facilitate resistance evolution. Pest Manag. Sci. 74:293-301 81

Young AJ (1991) The photoprotective role of carotenoids in higher plants. Physiol. Plant

83:702-708

Yuan JS, Tranel PJ, Stewart CN (2007) Non-target-site herbicide resistance: a family

business. Trends Plant Sci. 12:6-13

Zelaya IA and Owen MDK (2005) Differential response of Amaranthus tuberculatus (Moq

ex DC) JD Sauer to glyphosate. Pest Manag. Sci. 61:936-950