Review
`
`Received: 3 December 2013
`
`Revised: 21 March 2014
`
`Accepted article published: 3 April 2014
`
`Published online in Wiley Online Library: 6 May 2014
`
`(wileyonlinelibrary.com) DOI 10.1002/ps.3790
`
`Resistance to acetyl-CoA
`carboxylase-inhibiting herbicides
`Shiv S Kaundun*
`
`Abstract
`
`1405
`
`Resistance to acetyl-CoA carboxylase herbicides is documented in at least 43 grass weeds and is particularly problematic in
`Lolium, Alopecurus and Avena species. Genetic studies have shown that resistance generally evolves independently and can
`be conferred by target-site mutations at ACCase codon positions 1781, 1999, 2027, 2041, 2078, 2088 and 2096. The level of
`resistance depends on the herbicides, recommended field rates, weed species, plant growth stages, specific amino acid changes
`and the number of gene copies and mutant ACCase alleles. Non-target-site resistance, or in essence metabolic resistance, is
`prevalent, multigenic and favoured under low-dose selection. Metabolic resistance can be specific but also broad, affecting
`other modes of action. Some target-site and metabolic-resistant biotypes are characterised by a fitness penalty. However, the
`significance for resistance regression in the absence of ACCase herbicides is yet to be determined over a practical timeframe.
`More recently, a fitness benefit has been reported in some populations containing the I1781L mutation in terms of vegetative
`and reproductive outputs and delayed germination. Several DNA-based methods have been developed to detect known ACCase
`resistance mutations, unlike metabolic resistance, as the genes remain elusive to date. Therefore, confirmation of resistance is
`still carried out via whole-plant herbicide bioassays. A growing number of monocotyledonous crops have been engineered to
`resist ACCase herbicides, thus increasing the options for grass weed control. While the science of ACCase herbicide resistance
`has progressed significantly over the past 10 years, several avenues provided in the present review remain to be explored for a
`better understanding of resistance to this important mode of action.
`© 2014 Society of Chemical Industry
`
`Supporting information may be found in the online version of this article.
`
`Keywords: acetyl-CoA carboxylase; ACCase; resistance mechanism; fitness; resistance detection; resistant crops
`
`1 INTRODUCTION
`Weeds compete with crops for light, water and soil nutrients. They
`are by far the most challenging pests in agricultural production
`systems.1 If uncontrolled, ensuing average yield losses are esti-
`mated at 35% for six major crops worldwide.2 The advent of her-
`bicides has contributed significantly to protecting crop yields and
`increasing farmers’ profitability.3 One such single-site herbicide
`mode of action introduced in the mid-1970s consists of inhibitors
`of acetyl-CoA carboxylase (ACCase).4 ACCase herbicides are mainly
`used for grass weed control in dicotyledonous crops, with a few
`compounds applied in small-grain cereal crops and rice.5 Given
`their convenience for managing grass weeds post-emergence,
`ACCase herbicides were quickly adopted as they also represented
`a marked improvement over the then commonly used method of
`selective grass weed control. Current overall annual sales exceed
`$US 1 billion and account for 5% of all commercial herbicides.6
`Over time, however, extensive and recurrent use of ACCase her-
`bicides has selected for resistance in key grass weed species
`encompassing 20 genera.7 Resistance is reported in 33 countries,
`especially in areas with intensive use of ACCase herbicides often
`applied as the sole method for grass weed control. Two compre-
`hensive reviews of the mechanisms and evolutionary dynamics of
`resistance to ACCase resistance have been carried out in 19948 and
`2005.9 The present analysis will therefore summarise the current
`understanding of resistance with emphasis on studies conducted
`over the last 8 years.
`
`2 ACCase TARGET AND HERBICIDES
`ubiquitous,
`Acetyl-CoA
`carboxylase
`(EC
`6.4.1.2)
`is
`a
`biotin-dependent enzyme that catalyses the carboxylation of
`acetyl-CoA into malonyl-CoA using ATP as a source of energy
`and bicarbonate as a source of carbon.10,11 Catalysis is conducted
`in two steps: carboxylation of the biotin cofactor, followed by
`the transfer of the carboxyl group onto acetyl-CoA using the
`transcarboxylase activity of the enzyme. Biotin must visit both
`the biotin carboxylase and the carboxyltransferase sites, and as
`such a swinging arm model has been proposed to represent this
`translocation.12 Malonyl-CoA is required for de novo fatty acid
`synthesis in the plastid and for elongation of very-long-chain
`fatty acid and secondary plant metabolites such as flavonoids and
`suberins in the cytoplasm.13
`Plants have two different ACCases, namely cytoplasmic and
`plastidic.14 Both isoforms consist of three major
`functional
`domains: biotin carboxyl carrier (BCC), biotin carboxylase (BC)
`and carboxyl transferase (CT), which is further subdivided into
`𝛼 and 𝛽 subunits.15 In grasses of the Poaceae family, plastidic
`ACCase is homomeric, with BCC, BC and CT located on a single
`
`∗ Correspondence to: Shiv S Kaundun, Syngenta, Jealott’s Hill International
`Research Centre, Biological Sciences, Bracknell, Berkshire RG42 6EY, UK. E-mail:
`deepak.kaundun@syngenta.com
`
`Syngenta, Jealott’s Hill
`Bracknell, Berkshire, UK
`
`International Research Centre, Biological Sciences,
`
`Pest Manag Sci 2014; 70: 1405–1417
`
`www.soci.org
`
`© 2014 Society of Chemical Industry
`
`Ex. 1037
`RiceTec, Inc.
`Page 001
`
`PGR2021-00114
`
`

`

`www.soci.org
`
`SS Kaundun
`
`polypeptide.16 In contrast, the domains are encoded by different
`genes that are coordinately expressed to form a functional het-
`eromeric enzyme in most other plant species.17 Exceptions to this
`dichotomy include some members of the Geraniaceae family that
`have a homomeric plastidic ACCase characteristic of grasses, and
`some Brassica and Arabidopsis species that contain both a homo-
`meric and heteromeric ACCase in their chloroplasts.10,17,18 Active
`cytoplasmic and chloroplastic ACCases from grass weeds function
`as a homodimer as suggested by initial biochemical studies and
`confirmed recently by crystallography.19,20 The dimer is made up
`of two head-to-tail monomers generating two active sites.
`ACCase herbicides inhibit de novo fatty acid synthesis in sen-
`sitive grass weeds leading to rapid necrosis and plant death.21
`The compounds can be divided into three classes, namely ary-
`loxyphenoxypropionates (FOPs), cyclohexanediones (DIMs) and
`phenylpyrazolin (DEN), based on their chemical structures.22
`Currently, ten FOPs, nine DIMs and one DEN herbicide are com-
`mercially available for controlling both grass weeds and volunteer
`crops. The latest and leading ACCase herbicide is pinoxaden, intro-
`duced in 2006 for selective grass weed control in wheat, barley
`and triticale.23 Fenoxaprop-P-ethyl, launched some 30 years ago,
`remains an important product for use in small-grain cereal crops,
`rice and soybean.24
`FOP and DIM herbicides have long been suspected to affect lipid
`synthesis in plants.25 Convincing proof of disruption of plastidic
`ACCase activity was generated by pioneering work on isolated
`chloroplasts from corn26 and barley.27 Subsequent kinetic studies
`indicated that FOP and DIM herbicides showed nearly competitive
`inhibition with respect to the acetyl-CoA substrate.28,29 Addition-
`ally, FOP and DIM herbicides were suggested to have overlapping
`sites at the CT domain as they were mutually exclusive. Recently,
`the precise binding of haloxyfop, tepraloxydim and pinoxaden was
`determined using yeast CT domain as a surrogate.30– 32 Haloxy-
`fop and tepraloxydim were attached to the active site, particularly
`at the interface of the dimer. The DIM and FOP herbicides shared
`two main anchoring points but overall probed distinct regions
`of the dimer interface. Pinoxaden and tepraloxydim were bound
`at a very similar location in spite of their very different chemical
`structures.30,31 Contrary to tepraloxydim and especially pinoxaden,
`haloxyfop binding required large conformational changes.32 More
`precisely, movement of the side chains of tyrosine 1738 and pheny-
`lalanine 1956 is necessary to generate a large hydrophobic pocket
`for the pyridinyl ring of haloxyfop to sit in.
`While most grass weeds are controlled with ACCase herbicides,
`some Vulpia, Poa and Festuca species are inherently tolerant
`owing to an insensitive ACCase resulting from a fixed leucine
`residue at codon position 1781 (A. myosuroides equivalent, as
`is conventional for weed ACCases) which has been confirmed
`for the last two species.33– 35 Selectivity in monocotyledonous
`crops is in most cases provided by the use of safeners allowing
`faster and higher levels of herbicide detoxification in the crop
`versus the target weeds.36 Broadleaf species on the other hand
`are innately less sensitive owing to major differences in plastidic
`isoforms between grass and dicotyledonous species.37 A more
`comprehensive review of the ACCase target and herbicides can
`be found in several earlier publications.10,12,16,17,22
`
`3 OCCURRENCE, EVOLUTION AND SPREAD
`OF RESISTANCE
`The first case of resistance to ACCase herbicides was reported in
`1982 in a Lolium rigidum population from an Australian wheat
`
`field.38 The number of resistance cases both in terms of species
`and acreages has increased steadily over the last 30 years.7 Resis-
`tance is particularly widespread in L. rigidum throughout the Aus-
`tralian wheat belt, in practically every single cereal farm infested
`with A. myosuroides in the United Kingdom39 and in A. fatua in large
`areas in Western Canada.40 Other affected grass weed populations
`described recently include Phalaris minor from India,41 Greece42
`and Iran,43 Phalaris paradoxa from Italy44 and Israel,45 Sorghum
`halepense from Greece46 and the United States,47 Brachiaria plan-
`taginea from Brazil,48 Digitaria sanguinalis from France,49 Apera
`spica-venti from Central and Eastern Europe50 and Alopecurus
`japonicus51 and Beckmannia syzigachne52 from China, along with
`many other species from various regions as summarised in the
`International Survey of Herbicide Resistant Weeds website.7 The
`areas concerned nevertheless remain an underrepresentation of
`the real status of resistance to ACCase herbicides as there are
`some weed species and populations such as Lolium spp., Scle-
`rochloa kengiana and Fimbristilis miliacea from North Africa, China
`and Vietnam respectively that are not documented but are highly
`suspected to be affected by resistance.
`It is noteworthy that, where comprehensive surveys have been
`carried out repeatedly within the same agricultural zones, a signifi-
`cant escalation of resistance has been observed in a relatively short
`period of time. This is exemplified by an increase of 15% and 28%
`in resistance in wild oats and ryegrass in Western Canada (41%)
`and Australia (96%) respectively.40,53 Overall, resistance tends to be
`widespread with regard to the ACCase herbicides used earlier, pro-
`gressively increasing to the ones introduced more recently.44,54– 56
`For instance, resistance to clethodim, the most effective ACCase
`herbicide,57 has risen from 0.5 to 8 and 65% in three different ran-
`dom L. rigidum surveys carried out in Western Australia in 1998,
`2003 and 2010 respectively.53,58,59 Similar trends are observed for
`tepraloxydim, which is increasingly being used to control some
`resistant black-grass populations in the United Kingdom.39 Equally,
`where ACCase herbicides have not been used extensively, as is the
`case with European A. spica-venti, resistance has been recorded
`in only 2% of a total of 250 fields tested.50 Interestingly, submit-
`ted to very similar selection pressures in Australia, very high and
`lower levels of resistance to clodinafop-propargyl, pinoxaden and
`sethoxydim are observed in L. rigidum and A. fatua respectively,
`thus reflecting differences in the species’ abilities for evolving resis-
`tance to ACCase herbicides.59,60
`With a view to further investigate the mode of evolution of
`resistance to ACCase herbicides, Cavan et al.61 used anonymous
`simple sequence repeat (SSR) markers to examine the genetic
`profiles of four different A. myosuroides patches that had sur-
`vived a herbicide application within a small agricultural field. The
`genotypes were markedly diverse among the four sites allowing
`the conclusion of localised evolution of resistance, even within
`short distances. Similar inferences could be made from several
`black-grass studies at the local, country and regional levels based
`on signature ACCase sequences62 and known target-site resis-
`tance mutations.63– 66 Overall, the different studies have shown
`that there is enough standing genetic variation within the grass
`populations for resistance to evolve independently.62,63,65– 67 How-
`ever, once resistance has reached a certain level, it can spread very
`quickly within a field upon continuous ACCase herbicide selection
`pressure68 or even adjacent sites as demonstrated in two stud-
`ies on black-grass and ryegrass from France and Australia.69,70 In
`the latter case, resistance to ACCase herbicides in a neighbour-
`ing organic farm was as high as 2% while being 21% in an adja-
`cent conventional field.69 Resistance spreading via pollen is faster
`
`wileyonlinelibrary.com/journal/ps
`
`© 2014 Society of Chemical Industry
`
`Pest Manag Sci 2014; 70: 1405–1417
`
`1406
`
`Ex. 1037
`RiceTec, Inc.
`Page 002
`
`PGR2021-00114
`
`

`

`1407
`
`Resistance to ACCase herbicides
`
`www.soci.org
`
`for allogamous than for autogamous species, as represented by
`A. myosuroides and A. fatua respectively.71,72 A less frequent form
`of resistance spread, but still worth underlining, is via seed import
`from regions with high infestation of resistant grass weeds. For
`instance, analysis of a 20 kg batch of Australian wheat exported
`to Japan in 2006 identified as many as 4673 L. rigidum seeds, 35%
`of which were resistant to diclofop-methyl.73
`Importantly, once resistance to ACCase herbicides has been
`established, it is difficult to overcome even with the use of other
`herbicide modes of action and alternative methods of weed con-
`trol combined. A black-grass study conducted over a 6-year period
`has shown that while different cropping systems and other herbi-
`cide modes of action can result in an overall decline in the weed
`seed bank, the frequency of ACCase-resistant individuals within
`the field did not decrease over this timeframe.74 Similar observa-
`tions were made in a wheat field from south-eastern Italy infested
`with Lolium multiflorum resistant to ACCase herbicides.75
`
`4 MECHANISM OF RESISTANCE
`The knowledge of resistance mechanisms is important for the
`design of effective weed control strategies to manage and delay
`the onset of resistance to ACCase herbicides. Understandably, an
`overwhelming number of studies have been conducted on the
`most problematic Lolium, Alopecurus and Avena species. Nonethe-
`less, some minor species such as Phalaris spp.,44,76 Eleusine indica,77
`B. syzigachne52 and Rottboellia cochinchinensis,78 to name but a
`few, have also been investigated recently. As with all other her-
`bicide modes of action, resistance to ACCase herbicides can be
`divided into target- and non-target-based mechanisms.
`
`4.1 Target-site resistance
`Target-site resistance is essentially caused by single amino acid
`changes in the carboxyltransferase domain impacting on the
`effective binding of ACCase herbicides.57,79 It is therefore not sur-
`prising that some earlier studies have shown that target-site resis-
`tance is more or less inherited as a monogenic trait.68,80– 82 Excep-
`tions to this rule are threefold increases in ACCase-specific activity
`in two resistant S. halepense and Leptochloa chinensis populations
`from the United States and Thailand respectively.83,84 These two
`exceptions could be dismissed on the basis of intrinsic differences
`between sensitive and resistant populations being compared. In
`fact, the involvement of a resistant ACCase was suggested in a
`L. multiflorum population more than 20 years ago.85 However, it
`was not until the early 2000s before a first potential target-site
`resistance mutation was uncovered in a Setaria viridis86 and a
`L. rigidum87 population. At the last major review on ACCase
`herbicides, five different resistance codons were described mainly
`in A. myosuroides and classified into two main categories: FOP spe-
`cific (2027, 2041 and 2096) and FOP/DIM (1781 and 2078).9 The dis-
`covery was facilitated by the relatively conserved nature of ACCase
`in A. myosuroides88,89 compared with species such as Lolium spp.,
`in which several other amino acid changes not implicated in resis-
`tance are often present alongside the resistance mutations.90– 94
`Over the last 8 years, around 30 different mechanism studies
`involving a dozen weed species have also identified target-site
`resistance to at least one ACCase herbicide (see supporting infor-
`mation Table S1). The precise amino acid changes implicated
`could often be determined owing to the relatively conserved
`nature of plastidic ACCase and universal PCR methodologies for
`analysing the CT binding domain in grasses (Table 1).35 Most stud-
`ies have detected the same five mutations described earlier in
`
`A. myosuroides. Additionally, two other resistance codons at posi-
`tions 1999 and 2088 were uncovered, principally in Lolium spp.
`and Avena spp. Fourteen allelic variants have thus far been impli-
`cated in resistance, namely I1781L/V/A/T, W1999C/L/S, W2027C,
`I2041N/V, D2078G, C2088R and G2096A/S.57,92,95– 97 The frequency
`of the mutations varied according to the weed species and regions
`examined, and appeared to be governed by the local herbicide
`selection pressure applied.65 For example, the I1781L mutation is
`overwhelmingly present in black-grass in the United Kingdom and
`France, while the G2096A is predominant in Germany.98 Analysis of
`a large number of Lolium spp. populations from the United King-
`dom and Australia showed a predominance of the D2078G and
`I2041N mutations respectively,99,100 potentially reflecting the alter-
`nating use of FOPs and DIMs in small-grain cereal and dicotyle-
`donous break crops in the United Kingdom and a more prolonged
`FOP use in continuous wheat cropping systems in Australia. Impor-
`tantly, DNA analysis from a few hundred herbarium blackgrass
`plants predating any synthetic herbicide use revealed the pres-
`ence of an I1781L mutation in one of the individuals tested,101
`thus indicating that some ACCase mutations are intrinsically more
`prevalent than initially assumed.102
`One of the issues with several resistance mechanism studies is
`with regard to heterogeneous weed populations being compared
`with an unrelated sensitive population to estimate the resistance
`indices associated with the different target-site mutations. Thus,
`resistance was in some cases diluted or overestimated owing to the
`presence of wild and heterozygous mutant individuals and addi-
`tional underlying non-target-site resistance present in the popu-
`lations. To address this problem, a yeast-gene replacement assay
`was developed, allowing comparison of wild and mutant strains
`differing only at the single mutated amino acid position being
`investigated.103– 105 In addition to being very quick, this method
`does not require actual weed populations for testing. The down-
`side is that some mutations such as the C2088R and G2096A
`are not viable with this approach, and, as with enzyme assays,
`translation to whole plants is not always straightforward, espe-
`cially for mutations that carry a moderate level of resistance. For
`instance, Jang et al.103 estimated slightly higher levels of resis-
`tance to sethoxydim relative to pinoxaden for the W2027C muta-
`tion in the yeast-gene replacement assay. Conversely, whole-plant
`herbicide and molecular data suggested that the efficacy of the
`DEN, but not the DIMs, was affected by the W2027C mutation in
`black-grass and Japanese foxtail.56,106,107 Alternatively, individual
`plants from the same population and genetic background could
`be genotyped for wild-type and mutant alleles prior to carrying
`out dose responses with a range of herbicides, thus overcoming
`the issue posed by confounding effects of target-site resistance
`and non-target-site resistance. In this manner, the importance of
`mutations at codon positions 1781, 1999, 2078 and 2088 could be
`more precisely assessed.91,92,96,97
`Contrary to earlier classifications,9 numerous recent investiga-
`tions have clearly shown that, with the exception of the D2078G
`and C2088R mutations that confer broad resistance to all her-
`bicides tested,76,77,91,97,106,108– 110 the levels of resistance depend
`on specific amino acid changes, the number of resistant alleles,
`weed species, plant growth stages and recommended field rates
`of herbicides. For instance, plants with the I2041N were found
`to be sensitive to cycloxydim in A. myosuroides but resistant in
`P. paradoxa.45,106 Clethodim at the Australian field rate was found
`to control L. rigidum plants that were heterozygous for the I1781L
`mutation but was mostly ineffective on homozygous LL1781
`mutant individuals.109 By contrast, I1781L mutant A. myosuroides
`
`Pest Manag Sci 2014; 70: 1405–1417
`
`© 2014 Society of Chemical Industry
`
`wileyonlinelibrary.com/journal/ps
`
`Ex. 1037
`RiceTec, Inc.
`Page 003
`
`PGR2021-00114
`
`

`

`www.soci.org
`
`SS Kaundun
`
`Table 1. List of monocotyledonous species with published carboxyltransferase domain sequences (>90% coverage of CT domain)
`
`Species
`
`Accession number
`
`Source
`
`Length
`
`Alignment lengtha
`
`% Identitya
`
`Aegilops tauschii
`Alopecurus japonicus
`Alopecurus myosuroides
`Avena fatua
`Beckmannia syzigachne
`Brachypodium distachyon
`Echinochloa crus-galli
`Lolium multiflorum
`Lolium rigidum
`Phalaris minor
`Phalaris paradoxa
`Setaria italica
`Setaria viridis
`Sorghum bicolor
`Triticum aestivum
`Triticum turgidum
`Triticum urartu
`Zea mays
`
`EU660897.1
`JQ068820.1
`AJ310767.1
`JF785552.1
`KF501579.1
`XM_003581327.1
`HQ395758.1
`AY710293.1
`DQ184646.1
`AY196481.1
`AM745339.1
`AF294805.1
`AM408428.1
`XM_002446133.1
`EU660900.1
`EU660898.1
`EU660896.1
`U58598.1
`
`Genomic DNA
`mRNA
`mRNA
`Genomic DNA
`Genomic DNA
`mRNA
`mRNA
`Genomic DNA
`Genomic DNA
`Genomic DNA
`Genomic DNA
`mRNA
`Genomic DNA
`mRNA
`Genomic DNA
`Genomic DNA
`Genomic DNA
`mRNA
`
`123 798
`7589
`7589
`2039
`10 174
`7553
`7527
`3044
`1563
`2027
`1520
`7630
`12 934
`7537
`97 428
`165 764
`98 890
`5442
`
`1647
`1638
`1638
`1647
`1638
`1647
`1643
`1644
`1563
`1647
`1520
`1642
`1642
`1640
`1647
`1647
`1647
`1640
`
`91.2
`99.51
`100
`93.26
`96.52
`90.29
`87.89
`92.52
`92.58
`94.05
`94.01
`88.31
`88.31
`88.11
`91.5
`91.62
`91.5
`87.56
`
`*Alignment length and sequence identity with respect to the carboxyltransferase domain of Alopecurus myosuroides accession AJ310767.
`
`and L. multiflorum plants were mostly controlled at the Euro-
`pean field rates.106,108,111 Similarly, while the W1999C mutation
`impacted highly on the efficacy of fenoxaprop-P-ethyl, it was
`found to be sensitive to clodinafop-propargyl and sethoxydim
`in Avena sterilis.105 Likewise, the W1999S mutation conferred
`high levels of resistance to FOP and DEN herbicides and partial
`resistance to sethoxydim and cycloxydim while being sensitive to
`clethodim and tepraloxydim.92 Variable levels of resistance were
`also associated with the I1781T mutation in the heterozygous
`state.96 The 1781 threonine allelic variant impacted moderately
`on clodinafop-propargyl and cycloxydim but was sensitive to
`pinoxaden, tepraloxydim and clethodim. It is noteworthy that
`Hordeum and a few Bromus species,35,109 which are fixed for the
`C2088F mutation and also present in some individual ryegrass and
`wild oat plants (data not published), are sensitive to all effective
`ACCase herbicides. Thus, conclusions on the importance of a novel
`allelic variant at a known resistance codon such as the I1781A95
`or W1999L108,112 or any other position, can only be established via
`proper rigorous cosegregation or dose response studies based
`on wild and mutant subpopulations sharing the same genetic
`background.
`Interestingly, in hexaploid weed species such as Echinochloa
`spp. and Avena spp., all three homeologous ACCase genes were
`found to be expressed.105,113,114 Individual plants from a single Aus-
`tralian A. fatua population could carry one, two or all three of
`the I1781L, D2078G and C2088R mutations detected.114 The indi-
`vidual mutations endowed relatively lower levels of resistance in
`the hexaploid species compared to diploid species. This poten-
`tial dilution effect could explain the relatively slower evolution of
`ACCase resistance in Avena spp. compared to diploid species.53,60
`It could also account for the difference in the levels of resistance
`computed for pinoxaden with regard to the I2041N mutation in
`Avena fatua on the one side110 and Alopecurus,56 Lolium108 and
`Phalaris45 species on the other. Additionally, wheat mutagene-
`sis studies have shown that the level of resistance to ACCase
`herbicides depends on the specific A, B or D genome where the
`
`mutation is located, further adding to the complexity of resistance
`in hexaploid species.115
`Yeast ACCase crystal structures in complex with FOP, DIM and
`DEN herbicides revealed that out of the seven codons involved
`in target-site resistance, only the 1781, 1999 and 2041 amino
`acid residues were directly implicated in the binding of the
`herbicides.103 Taking advantage of the conserved nature of
`amino acid sequences around the vicinity of the CT domain
`binding site, homology models were built for Setaria italica and
`A. myosuroides with a view to rationalising the importance of
`I1781L, W2027C, I2041N and D2078G mutations on ACCase her-
`bicide efficacy.116– 120 Molecular docking and molecular dynamic
`simulations indicated that the W2027C mutation for example,
`though remote, caused conformational changes in the binding
`site of FOP herbicides.117 In particular, significant changes were
`associated with phenylalanine 377, tyrosine 161 and tryptophan
`346, which are critical for FOP binding. Consequently, the pi–pi
`interaction between the herbicides and phenylalanine 377 and
`tyrosine 161 was decreased, accounting for the molecular basis of
`resistance caused by the W2027C mutation (Fig. 1).
`
`4.2 Non-target-site resistance
`Non-target-site resistance (NTSR) is now increasingly recognised
`as being the predominant resistance mechanism to ACCase
`herbicides.121 Several recent large-scale black-grass surveys have
`shown that most resistant individuals did not contain a known
`target-site mutation.65,122 Similar observations were made in
`Lolium multiflorum populations from the United Kingdom, based
`on both molecular and glasshouse biological studies.123 Addi-
`tionally, when investigated in detail, non-target-site resistance
`to at least one ACCase herbicide is often present in populations
`containing target-site resistance.92,96,124,125 Also, it has often been
`assumed that NTSR confers lower levels of resistance that can
`sometimes be controlled when plants are treated at an early
`growth stage. Over time however, the build-up of NTSR resis-
`tance has reached very high levels, especially to earlier ACCase
`
`wileyonlinelibrary.com/journal/ps
`
`© 2014 Society of Chemical Industry
`
`Pest Manag Sci 2014; 70: 1405–1417
`
`1408
`
`Ex. 1037
`RiceTec, Inc.
`Page 004
`
`PGR2021-00114
`
`

`

`Resistance to ACCase herbicides
`
`www.soci.org
`
`1409
`
`procedures. Alternatively, metabolic resistance has been inferred
`indirectly from the use of synergists that inhibit detoxifying
`enzymes involved in ACCase herbicide metabolism, the absence
`of known target-site mutations and differential responses to
`closely metabolisable and non-metabolisable FOP and DIM
`herbicides.91,122,125,136 The synergists used include compounds
`such as amitrole, 1-aminobenzotriazole (ABT), piperonyl butoxide
`(PBO) and malathion that impact on P450 enzymes.137 However,
`the data from these approaches should be taken with caution
`as resistance could be due to target-site mutations yet to be
`uncovered and also because the ability to augment herbicide
`activity can be synergist, herbicide, population and species spe-
`cific. For instance, ABT, but not malathion or tetcyclasis, was
`shown to improve the efficacy of diclofop-methyl in a resistant
`L. rigidum population.133 The activity of tralkoxydim, also affected
`by metabolic resistance in this population, was unaltered by any
`of the synergists used. Similarly, none of the three cytochrome
`P450 inhibitors, i.e. ABT, malathion or tetcyclasis, could reverse
`fenoxaprop-P-ethyl metabolism in a black-grass population
`from the United Kingdom.138 Conversely, malathion in mixture
`with fenoxaprop-P-ethyl and pinoxaden allowed suppression of
`A. fatua populations suspected to be characterised by metabolic
`resistance.125
`Metabolic resistance to ACCase herbicides can result from con-
`stitutively overexpressed enzymes or be induced by external fac-
`tors. In particular, elevated levels of cyp P450 enzymes involved
`in phase 1 and of GST and O-glucosyl transferases operating in
`phase II detoxification were identified in the Peldon black-grass
`population.139 A similar observation was recorded in several other
`black-grass populations with respect to GST enzymes.140 Resis-
`tance could also be induced via the use of safeners such as mefen-
`pyr diethyl in increasing the peroxidase protective activities of phi
`and lambda gluthathione transferases.141 The level of GST involved
`in metabolism can also vary according to plant growth stages and
`environmental conditions.142
`Metabolic resistance is favoured under low-dose selection of
`minor genes that individually confer low levels of resistance
`but when accumulated confer significant levels of resistance
`to ACCase herbicides. This was elegantly demonstrated under
`glasshouse conditions by the recurrent selection at low doses
`of diclofop-methyl of progressively recalcitrant individuals from
`an initially sensitive L. rigidum population.143,144 Resistance to
`practical field rates of diclofop-methyl was attained after three
`generations, starting from a genetic pool of a few hundred sen-
`sitive plants only. A comparable scenario is thought to function
`under field conditions because not all plants receive effective rates
`of ACCase herbicides owing to suboptimal spray conditions, shad-
`ing upon high plant densities and staggered seed germination.145
`Subsequent studies confirmed metabolism of diclofop-methyl
`to the acid equivalent followed by further degradation into
`non-toxic metabolites similar to what is achieved in wheat.146
`Worryingly, metabolism-based resistance acquired via low-dose
`selection could also endow resistance to two other modes of
`action.145 This gives further credence to earlier observations of
`NTSR selected by ACCase herbicides affecting the acetolactate
`synthase herbicides iodo-/mesosulfuron in A. myosuroides.56,147
`While metabolic resistance selected by ACCase herbicides is clearly
`demonstrated to cut across herbicide modes of action, it can also
`be compound-specific within and between ACCase subclasses.
`Non-target-site resistance to fenoxaprop-P-ethyl was found to
`be more prevalent than clodinafop-propargyl and pinoxaden in
`French black-grass populations.56 Likewise, several ryegrass and
`
`Π- Π
`
`Figure 1. Impact of the W2027C (equivalent to W374C in the diagram)
`mutation on the binding of FOP herbicides exemplified here with diclofop.
`Overlay between wild and mutant diclofop and its corresponding W374C
`complex. The wild-type complex is in grey and cyan, while the mutant
`W374C complex is in blue and yellow (extracted from Xu et al.117).
`
`herbicides such as diclofop-methyl.91,126 In some cases the level
`of resistance conferred by NTSR has even surpassed those of the
`most common target-site mechanisms. In a black-grass popu-
`lation for example, NTSR was found to be a significantly bigger
`contributor to resistance to clodinafop-propargyl and pinoxaden
`than target-site mutations at position 1781.96
`Non-target-site resistance to herbicides encompasses a range
`of diverse mechanisms including reduced penetration, impaired
`translocation, sequestration and enhanced metabolism of the
`toxophores.79 The suggestion in the 1990s of the ability of
`some weed populations to resist ACCase herbicides via mem-
`brane repolarisation could not be substantiated by subsequent
`experiments.127,128 Similarly, the few anecdotal reports of resis-
`tance caused by reduced ACCase herbicide penetration and
`sequestration need further confirmation.129,130 More recently, the
`involvement of phi and lambda classes of GST has been reported
`in some multiple-herbicide-resistant black-grass and ryegrass
`populations.131,132 Resistance was suggested to be endowed
`via the scavenging peroxidase activities of these specific GST
`enzymes as well as a concomitant production of protective
`flavonoids to counteract the free noxious radicals generated by
`the herbicide action. This hypothesis was further strengthened
`with heterologous expression of GSTF1 in Arabidopsi