Professional Documents
Culture Documents
Selective Control of Parasitic Nematodes Using Bioactivated Nematicides
Selective Control of Parasitic Nematodes Using Bioactivated Nematicides
Parasitic nematodes are a major threat to global food security, particularly as the
world amasses 10 billion people amid limited arable land1–4. Most traditional
nematicides have been banned owing to poor nematode selectivity, leaving farmers
with inadequate means of pest control4–12. Here we use the model nematode
Caenorhabditis elegans to identify a family of selective imidazothiazole nematicides,
called selectivins, that undergo cytochrome-p450-mediated bioactivation in
nematodes. At low parts-per-million concentrations, selectivins perform comparably
well with commercial nematicides to control root infection by Meloidogyne incognita,
a highly destructive plant-parasitic nematode. Tests against numerous phylogenetically
diverse non-target systems demonstrate that selectivins are more nematode-selective
than most marketed nematicides. Selectivins are first-in-class bioactivated nematode
controls that provide efficacy and nematode selectivity.
Global food demand will be increasingly difficult to meet as the human tioxazafen and cyclobutrifluram. Unfortunately, iprodione has already
population approaches 10 billion people1–4. There is a scarcity of arable been banned in Europe owing to its carcinogenic potential and risk
land for agricultural expansion, and land-conversion efforts are con- to aquatic life10. Meanwhile, the market release of a tioxazafen-based
strained by social and ecological factors. Maximizing production from seed treatment has been postponed owing to reports of skin irritation
currently cultivated land will be crucial to ensure global food security1–4. in individuals handling this nematicide. There is a pressing need for
Farmers rely on agrochemicals to maximize yields by controlling new nematicides with improved selectivity.
crop pathogens4,5. Plant-parasitic nematodes (PPNs) are especially
destructive pathogens that cause more than US $100 billion in crop
losses every year6,7. Synthetic nematicides have played an essential Selective imidazothiazole nematicides
part in PPN control for decades; however, concerns over environmental From a library of uncharacterized compounds that disrupt the growth
toxicity and human safety have justifiably prompted bans on the most of the free-living nematode C. elegans14, we identified three new
commonly used nematicides8,9. The lack of available agents to control imidazothiazole-containing molecules that selectively kill nematodes
PPNs is stark. Of the 20 key nematicides used in the twentieth century, (Fig. 1a). At low micromolar concentrations, these selective imidazothia-
only 4 are currently approved for use in the European Union and only 3 zole nematicides, which we call selectivin-A, selectivin-B and selectivin-C,
are used in the USA without restriction9–11. Although warranted, these killed all four of the free-living nematode species that we can easily cul-
withdrawals leave farmers with limited options and no control meas- ture in the laboratory. Moreover, these selectivins were able to kill eggs of
ures for several PPNs8,12. a nematode parasite of cattle and the infective juveniles of at least one out
Despite the need for more selective PPN control measures, only of three distinct species of root-knot nematode that infect crops (Fig. 1b).
seven non-fumigant synthetic nematicides have been developed These eight nematode species span five genera and two evolutionary
in the past 25 years9,13. These next-generation nematicides are flu- clades, which suggests that diverse nematode species, including para-
azaindolizine, fluensulfone, fluopyram, iprodione, spirotetramat, sites, are susceptible to the selectivins. By contrast, these compounds
1
The Donnelly Centre for Cellular and Biomolecular Research, University of Toronto, Toronto, Ontario, Canada. 2Department of Molecular Genetics, University of Toronto, Toronto, Ontario,
Canada. 3Davenport Research Laboratories, Department of Chemistry, University of Toronto, Toronto, Ontario, Canada. 4USDA-ARS Horticultural Crops Research Laboratory, Corvallis, OR, USA.
5
Division of Neurology and Program in Genetics and Genome Biology, The Hospital for Sick Children, Toronto, Ontario, Canada. 6Institute for Integrative Genome Biology, University of California,
Riverside, Riverside, CA, USA. 7Department of Botany and Plant Sciences, University of California, Riverside, Riverside, CA, USA. 8Department of Comparative Biology and Experimental
Medicine, Host–Parasite Interactions Program, Faculty of Veterinary Medicine, University of Calgary, Calgary, Alberta, Canada. 9Ajmera Transplant Centre, Toronto General Research Institute,
University Health Network, Toronto, Ontario, Canada. 10Department of Laboratory Medicine and Pathobiology and Immunology, University of Toronto, Toronto, Ontario, Canada. 11USDA-ARS
Mycology and Nematology Genetic Diversity and Biology Laboratory, Beltsville Agricultural Research Center, Beltsville, MD, USA. 12Department of Biological Sciences, Host Parasite Interactions
Program, Faculty of Science, University of Calgary, Calgary, Alberta, Canada. 13Department of Biochemistry, University of Toronto, Toronto, Ontario, Canada. 14Mediterranean Institute for Life
Sciences, Split, Croatia. 15Department of Pharmacology and Toxicology, University of Toronto, Toronto, Ontario, Canada. ✉e-mail: andy.burns@utoronto.ca; peter.roy@utoronto.ca
R2
Ce Cb Pp KR Co Mi Mc Mh
Sel-A 0 1
Selectivin R1
N
R3 Sel-B Relative
core structure N S Sel-C viability
0
1.6
3.1
6.3
12.5
25
50
100
0
1.6
3.1
6.3
12.5
25
50
100
0
1.6
3.1
6.3
12.5
25
50
100
0
1.6
3.1
6.3
12.5
25
50
100
0
1.6
3.1
6.3
12.5
25
50
100
0
5
15
45
0
5
15
45
0
5
15
45
N
Sel-A Cl
Concentration (μM)
N S
c
Fungi Fish Human cells Insect
N
Sel-B F Sc Ca Dr HEK HepaRG Dm (ad) Dm (lv)
N S
Sel-A 0 1
N Sel-B Relative
Sel-C Br Sel-C viability, growth
N S
or motility
12.5
12.5
12.5
12.5
12.5
0
1.6
3.1
6.3
25
50
100
0
1.6
3.1
6.3
25
50
100
0
1.6
3.1
6.3
25
50
100
0
1.6
3.1
6.3
25
50
100
0
1.6
3.1
6.3
25
50
100
0
5
15
45
0
5
15
45
Concentration (μM)
Fig. 1 | The selectivins. a, The selectivin core scaffold and structures of (Dm (lv)). Viability was assessed for D. rerio and D. melanogaster larvae. Cell
selectivin-A (Sel-A), selectivin-B (Sel-B) and selectivin-C (Sel-C). b, Selectivin growth was assessed for S. cerevisiae, C. albicans, HEK293 cells and HepaRG cells.
dose–response for the following eight nematode species: C. elegans (clade V) Motility was assessed for D. melanogaster adults. The colour-coded scale denotes
(Ce); Caenorhabditis briggsae (clade V) (Cb); Pristionchus pacificus (clade V) viability, growth or motility relative to untreated control, as appropriate for
(Pp); Rhabditophanes spp. KR3021 (clade IV) (KR); Cooperia oncophora the system tested. In b,c, dose–response data are an average across n biological
(clade V) (Co); M. incognita (clade IV) (Mi); Meloidogyne chitwoodi (clade IV) replicates (BRs) or technical replicates (TRs), with the following n values for each
(Mc); and Meloidogyne hapla (clade IV) (Mh). The colour-coded scale denotes system: Ce (n = 10 BRs for sel-A, 9 BRs for sel-B, 13 BRs for sel-C), Cb (n = 3 BRs),
viability relative to untreated control. c, Selectivin dose–response for the Pp (n = 3 BRs), KR (n = 3 BRs), Co (n = 3 BRs), Mi (n = 6 BRs), Mc (n = 6 BRs), Mh
following six phylogenetically distinct non-nematode model systems: Sc, (n = 6 BRs), Sc (n = 3 BRs), Ca (n = 3 TRs), Dr (n = 3 BRs), HEK (n = 3 TRs), HepaRG
S. cerevisiae (Sc); Candida albicans (Ca); D. rerio (Dr); HEK293 cells (HEK); (n = 3 TRs), Dm (n = 3 BRs).
HepaRG cells; D. melanogaster adults (Dm (ad)); and D. melanogaster larvae
exhibited little-to-no activity against fungi, insects, fish or human cells CYP enzymes by compromising EMB-8 activity during larval develop-
at concentrations that readily killed nematodes (Fig. 1c). Furthermore, ment, bypassing its essential role during embryogenesis27. EMB-8 is
mice given a daily oral dose of 50 mg kg–1 selectivin-A for 5 days showed the C. elegans P450 oxidoreductase (POR) enzyme that is a necessary
no obvious pathologies relative to untreated mice. To our knowledge, redox partner of all microsomal CYP enzymes27,28. Consistent with the
broad-spectrum nematicidal activity has not been previously described notion that selectivins are bioactivated nematicides, emb-8 knockdown
for any compound containing the 6-phenylimidazo[2,1-b]thiazole scaf- rendered C. elegans resistant to selectivin-induced lethality (Fig. 2a). By
fold of selectivins. Thus, the selectivins represent a new structural class contrast, the compound wact-11, which does not require bioactivation
for the development of selective and effective PPN control. for activity14, killed worms in an EMB-8-independent manner.
To identify selectivin metabolites, we incubated both young
adult and first larval stage (L1) worms in the presence or absence of
Selectivins have a new mode of action 100 µM selectivin-A for 6 h and then processed whole worm lysates
As a first step towards understanding the mode of action of selectivins, by reversed-phase high-performance liquid chromatography (HPLC)
we tested selectivin-A and selectivin-C against eight mutant strains coupled with a diode-array absorbance detector (HPLC-DAD) (Fig. 2b).
of C. elegans that are each specifically resistant to one of the major We identified five distinct peaks in the selectivin-A-treated lysates
anthelmintic or nematicide classes14–22 (Extended Data Table 1). None that were absent from the untreated controls. The peak that eluted
of the mutants were resistant to selectivins, which suggests that selec- at 7 min had the same retention time and absorbance spectrum as
tivins have a distinct mode of action compared with the commercial the selectivin-A standard that was directly injected onto the column.
agents examined. This peak is probably unmodified selectivin-A (parent molecule) that
To further characterize the mode of action of the selectivins, we persists in worm tissue following incubation. The remaining four peaks
performed a forward genetic screen of C. elegans mutants (generated eluted earlier in the HPLC run relative to selectivin-A, which suggested
through random mutagenesis) that resist selectivin-induced lethality. that they are relatively less lipophilic. These peaks were absent from
This approach has previously produced large numbers of mutants the selectivin-A standard, and they were not found in the lysates of
that specifically resist many of the major classes of anthelmintics and dead worms incubated in selectivin-A. This result supports the idea
nematicides that are commercially used14,15,17,19,20,22,23. Despite screening that they are bona fide selectivin-A metabolites and not accumulated
more than 10 million mutagenized genomes, selectivin-resistant worms contaminants or spontaneous oxidation products (Fig. 2b,c). We named
could not be generated (Extended Data Table 2). This result reinforces these presumptive selectivin-A metabolites M1, M2, M3 and M4.
the conclusion that selectivins kill nematodes using a mechanism that Consistent with our hypothesis that selectivin-A is bioactivated, the
is distinct from traditional nematode control agents. abundance of all four metabolites in worm tissue was substantially
reduced following emb-8 knockdown, whereas the abundance of the
unmodified selectivin-A parent increased by almost twofold (Fig. 2b,d).
Selectivins are bioactivated nematicides Unmodified selectivin-A is unlikely to be the active agent in vivo because
Previous work using C. elegans has shown that metabolites of com- worms in which emb-8 was knocked down survived treatment with
pounds that are lethal to nematodes often accumulate in worm tissue24. 100 µM selectivin-A even though the internal concentration was nearly
It is therefore possible that a nematicide could be transformed from an twice as high as that found in wild-type worms (Fig. 2a,d). Thus, selec-
innocuous parent molecule into a metabolic product that is lethal to tivins are bioactivated pro-nematicides.
worms. Microsomal cytochrome P450 (CYP) enzymes are commonly
involved in the oxidative biotransformation of inert pro-drugs into
bioactive metabolites25. The C. elegans genome encodes 76 microso- Bioactivation to a toxic electrophile
mal CYP enzymes26, many of which are known to metabolize drugs27. To identify the toxic selectivin metabolite (or metabolites), we
To test whether selectivins are bioactivated by C. elegans, we capital- collected HPLC fractions containing the four selectivin-A metabo-
ized on previous work that achieved broad disruption of microsomal lites and analysed their structures by electrospray ionization mass
standard
300
Sel-A
0
1.6
3.1
6.3
12.5
25
50
100
0
1.6
3.1
6.3
12.5
25
50
100
280
Concentration (μM)
260
0 1
Relative viability
M4 340
M1 M2 M3 Sel-A
320
c P = 0.0019
+ sel-A
WT L1
300
20,000 Live worms 280
Dead worms
10,000 260
3,000 340
Wavelength (nm)
Sel-A
AUC
320
P = 0.0015
P = 0.0004
P = 0.0265
P = 0.0027
Dead WT L1
2,000
+ sel-A
300
1,000 280
260
0
Sel-A M1 M2 M3 M4
Analyte M4 340
M1 M2 M3 Sel-A 320
WT adults
+ sel-A
300
d 2,000 WT 280
emb-8 KD
P = 0.0415 260
1,500
P = 3.543 × 10–6
10–5
10–5
P = 2.539 × 10–5
340
emb-8 KD adults
AUC
1,000 Sel-A
P = 2.146 ×
P = 2.545 ×
320
+ sel-A
300
500 280
260
0
Sel-A M1 M2 M3 M4
3
4
5
9
0
9
0
1
0
1
2
2.
2.
2.
4.
5.
6.
7.
7.
2.
2.
2.
Analyte
Retention time (min)
Fig. 2 | Selectivins are bioactivated nematicides. a, Selectivin and wact-11 is the sel-A standard injected directly onto the column and is not biomass-
dose–response for wild-type (WT) and emb-8 knockdown (KD) adult worms. normalized or background-corrected. Unmodified sel-A and metabolites M1–M4
WT dose–response data are an average across 7 BRs for sel-A and 4 BRs for the are indicated. c, HPLC-DAD quantification of unmodified sel-A and M1–M4 in the
other three compounds. emb-8 KD dose–response data are an average across 3 lysates of live and dead L1 worms (n = 3 BRs with 200,000 worms per replicate).
BRs. b, HPLC chromatograms of lysates from live and dead L1 worms and from d, HPLC-DAD quantification of sel-A and M1–M4 in the lysates of WT and emb-8
WT and emb-8 KD adults treated with 100 µM sel-A. The chromatograms shown KD adults (n = 3 BRs with 1,250 worms per replicate). In both c and d, the AUC is
are the product of dry weight normalization of raw chromatograms followed the area under the curve for the given analyte peak, calculated at 260 nm for
by subtraction of the absorbance intensity from paired untreated controls. M1–M3 and at 300 nm for M4. Data are presented as the mean ± s.e.m. For each
Absorbance intensity is in milliabsorbance units (mAU) per mg of dry weight, analyte, P values were obtained from unpaired one-tailed Student’s t-tests
and the absorbance wavelength (y axis) is in nanometres. The top chromatogram comparing the means of live and dead worms (c) or WT and emb-8 KD worms (d).
spectrometry (ESI-MS) followed by MS/MS fragmentation. Based on our and Extended Data Fig. 2). The 1H-13C heteronuclear single quantum
analysis of the M4 fraction, we speculated that M4 is a disulfide of two coherence spectrum indicated a carbon (C13) at 64.1 ppm containing
4-(4-chlorophenyl)-1H-imidazole-2-thiol monomers (Extended Data two diastereotopic protons at 3.82 and 4.67 ppm, and a carbon (C14) at
Fig. 1a,b, Extended Data Table 3 and Supplementary Table 1). Alkylating 60.9 ppm containing a proton at 6.00 ppm. From the 1H-1H correlation
lysis resulted in the presumptive alkylated monomer, but not the spectroscopy spectrum, a correlation between the protons on C13 and
disulfide, which indicated that the monomer forms in vivo, whereas C14 was observed, which suggested that these carbons are directly con-
the disulfide forms after lysis (Extended Data Fig. 1c–e and Supplemen- nected. Connectivity of the glutathione molecule through sulfur to C14
tary Table 1). Dose–response analysis with the imidazole-thiol mono- was confirmed by a heteronuclear multiple bond correlation between the
mer, which is commercially available, revealed that it was unable to kill proton on C14 and C17 and vice versa. Therefore, we identified metabolite
worms up to a concentration of 100 µM, despite readily accumulating M2 as a glutathione conjugate of an electrophilic sulfoxide metabolite of
in worm tissue (Extended Data Fig. 1f–h). Therefore, the imidazole-thiol selectivin-A. As is typical for reactive electrophiles, it was not possible
is not responsible for the lethality of selectivins. to obtain a pure isolate of the unconjugated metabolite. We therefore
Abundant masses in the M1, M2 and M3 fractions, and respective inferred its structure from the glutathione conjugate, which is a com-
MS/MS fragmentations, were consistent with γ-glutamylcysteine, glu- mon practice in drug bioactivation studies29,30. We speculated that M1
tathione and cysteine conjugates of an electrophilic sulfoxide metabo- and M3 are γ-glutamylcysteine and cysteine conjugates of the same
lite of selectivin-A, respectively (Fig. 3a–c, Extended Data Table 3 and sulfoxide metabolite, respectively. Alkylating lysis did not lower the
Supplementary Table 1). NMR characterization of a HPLC-purified abundance of the sulfoxide conjugates in worm lysates, which indicated
M2 fraction was consistent with the glutathione sulfur reacting at the that they are produced in vivo (Extended Data Fig. 1c,e). Thus, our data
electrophilic β-carbon of the proposed sulfoxide metabolite (Fig. 3d suggest that worms have a canonical detoxification pathway in which
Relative abundance
501.07 148.04 COOH 372.02 148.04 224.99
$
1.0 $ $ 1.0
$ $ $
O
277.09 224.99 S γ-Glu-Cys 224.99 S
Cys
N N
224.99 Cl Cl
N S 374.02
!"# !"# N S
!"#
0.5 !"#
503.06 !"# 277.09 !"#
0.5
O 501.07 O 148.04
Metabolite M1 Metabolite M3
[M1+H]+ = 501.07 [M3+H]+ = 372.02
372.02
!0 ! ! !0 ! !
0!
&5#
&5#
!0
!0
!0
z
0
5
0
5
0
44%
&0!
&0!
/z!
0!
(&m/
%37
%37
%37
%37
$1%4
$1&9
24
'&29
34
50
50
50
50
14
19
24
29
34
39
'%
(%
!
!
d O NH2
b M2 Control 558.09 MS/MS 205.06 H
O H NH2 HO C33 C19 N C22 C25 OH
558.09 205.06 N C34 N C18 C21 C24 C26
Relative abundance
HOOC N COOH
$
1.0 $ $ H
H
O
224.99 S GSH O C17 O O
334.11 S
N
Cl
!"#
0.5 560.09
!"# !"# 224.99 N S 129.7 127.5 115.8
334.11 C1 C6 60.9 COSY
O 558.09 C11 C14
Metabolite M2 132.8
N
[M2+H]+ = 558.09 Cl C2 C5 C7150.0 C13 HMBC
134.5 C9 64.1
!0 ! ! C3 C4 N
153.2
S
129.7 127.5
! /z
%7
'2
%7
'2
!! 0
#!0
!0
!0
!0
!0
##55
#&56
##55
#&56
%20
%25
&3! 0
&3#5
'!!40
'#45
#!50
##m
O
f
Viability
M2 S M1 S M3 S
100
enzymes GST PCS GGT N
N N N N
Cl Cl Cl Cl Cl
N S N N S N S
50
CYP- S N S
35C1 O GSH O O O 0
Sel-A Sel-B Sel-C
Treatment
g h 1 j 1
Relative viability
Relative growth
WT cyp-35c1Δ EV CYP-35C1
50 l
Sel-A Sel-A
Sel-B Sel-B No selectivin
CYP-35C1
Moving animals (%)
0
25
50
25
50
100
100
12.5
12.5
0
0
25
50
25
50
6.3
6.3
100
100
12.5
12.5
0 0 EV
30
Concentration (μM) Concentration (μM)
20 i k
WT Selectivin
EV P = 1.492 × 10–6
10 Z-score > 3 P = 0.0033 Selectivin sulfoxide
cyp-35c1Δ + CYP-35C1 CYP-35C1
0
0 2,000 4,000 6,000 8,000 0 20,000 40,000 60,000
13A01
13A02
13A03
13A04
13A05
13A06
13A07
13A11
13B01
13B02
14A01
14A04
14A05
23A01
25A01
25A02
25A03
25A04
25A05
25A06
29A02
31A02
32A01
33B01
33E01
33E02
33E03
34A01
34A02
34A03
34A04
34A05
34A06
34A07
34A08
34A09
34A10
35A02
35A03
35A04
35A05
35B01
35B03
36A01
37A01
37B01
43A01
33C01
33C02
33C03
33C04
33C06
33C07
33C08
33C09
33C11
33D01
33D03
35C01
35D01
Fig. 3 | Selectivins are bioactivated to a toxic electrophile. a–c, MS data identifies CYP-35C1. Z-scores were calculated using mean and s.d. of 50 mock
for metabolite M1 (a) M2 (b) and M3 (b) HPLC fractions and untreated control experiments. Raw data are provided in Supplementary Table 2. h, Selectivin
fractions, together with MS/MS fragmentation data for selected masses. dose–response for C. elegans WT and cyp-35c1 deletion (cyp-35c1Δ) mutants.
Structures are shown for γ-glutamylcysteine (γ-Glu-Cys), glutathione (GSH) WT dose–response data are an average across 7 BRs for sel-A and 4 BRs for the
and cysteine conjugates of a sel-A sulfoxide metabolite, along with fragment other two compounds. cyp-35c1Δ mutant dose–response data are an average
structures. d, 1H-NMR and 13C-NMR characterization of HPLC-purified M2. 1H-1H across 3 BRs. i, HPLC-DAD quantification of total M1–M3 in the lysates of
correlation spectroscopy (COSY) and 1H-13C heteronuclear multiple bond C. elegans WT and cyp-35C1Δ mutants (n = 3 BRs). j, Selectivin dose–response
correlation (HMBC) values are indicated. Carbon chemical shifts are in green. for CYP-35C1-expressing yeast and empty vector (EV) control. Dose–response
NMR spectra are provided in Extended Data Fig. 2. e, A proposed pathway for data are an average across 3 BRs. k, HPLC-DAD quantification of total M1–M3 in
sel-A sulfoxidation and low-molecular-weight thiol conjugation and CYP the lysates of CYP-35C1-expressing yeast and EV control (n = 4 BRs). l, Selectivin
enzymes that probably mediate S-oxidation. GSH reacts with the electrophilic bioactivation occurs in CYP-35C1-expressing yeast not in EV control. In i and k,
sulfoxide metabolite spontaneously or through glutathione-S-transferase AUC is area under the curve at 260 nm for M1–M3 combined. Data are presented
(GST), and phytochelatin synthase (PCS) and γ-glutamyl transferase (GGT) as the mean ± s.e.m. P values are from unpaired one-tailed Student’s t-tests
convert it to γ-Glu-Cys and cysteine. f, Viability of selectivin-treated L1 worms comparing the means of WT and cyp-35C1Δ worms (i) or EV control and CYP-35C1-
with or without 2.5 mM NACET (n = 5 BRs). Data are presented as the mean ± s.d. expressing yeast (k).
g, A C. elegans CYP RNAi screen for suppression of sel-A-induced lethality
glutathione is enzymatically conjugated to the potentially toxic electro- multiple oxidation schemes (Methods). Instead, we tested whether
philic sulfoxide metabolite, thereby sequestering it from cellular nucleo- the exogenous addition of N-acetylcysteine ethyl ester (NACET),
philes. The conjugated glutathione is then enzymatically converted which is enzymatically converted into cysteine for use in the synthe-
to γ-glutamylcysteine and cysteine, which results in metabolites that sis of new glutathione35, could suppress selectivin-induced lethality.
are readily excretable31–34 (Fig. 3e). Alternatively, but not in a mutually NACET has been used to block the damage caused by acetaminophen
exclusive manner, the sulfoxide metabolite could spontaneously react overdose, which results from the build-up of a reactive electrophilic
with glutathione and the other low-molecular-weight thiols without the metabolite that depletes glutathione and reacts with proteins to cause
aid of enzyme catalysts. Regardless of the mechanism, the accumula- hepatotoxicity35. We reasoned that the sulfoxide metabolite may kill
tion of the sulfoxide conjugates in worm tissue suggests that sulfoxide nematodes through a similar mechanism. Administration of 2.5 mM
metabolite production outpaces the capacity of worms to detoxify them. NACET suppressed selectivin-induced lethality, which indicated
We could not directly test whether the sulfoxide metabolite that the sulfoxide metabolite is the active nematicidal agent in vivo
induces lethality because it could not be synthesized despite trying (Fig. 3f).
Mock
Fluensulfone 38 2.5 4.5 – – – 6 7.76 × 10–15
Fluopyram 28 2.5 4.5 – – – 6 1.44 × 10–15
15 μM
Sel-E 45 2.5 4.5 Cl H Me 6 0.001
45 μM
Sel-C 45 2.8 5.0 Br H H 6 0.007
Iprodione 34 2.5 4.5 – – – 6 0.008
Mock
Sel-H 45 2.9 5.2 Br Me H 6 0.194
Sel-J 45 3.3 5.9 I H H 4 0.386
Sel-F 45 2.3 4.1 F Me H 6 0.411
15 μM
Sel-D 45 2.5 4.5 Cl Me H 4 0.274
Sel-I 45 2.9 5.2 Br H Me 6 0.796
45 μM
–50 0 50 100
Treatment effectiveness (%)
Nematicide P. simiae C. albicans C. elegans P. pacificus P. herm. D. melanogaster D. rerio HEK293 HepaRG
Sel-E
Sel-A
Abamectin
Fenamiphos
Fluensulfone
Fluopyram
Iprodione
Oxamyl
Spirotetramat
Tioxazafen
0
1.56
3.13
6.25
12.5
25
50
100
0
1.56
3.13
6.25
12.5
25
50
100
0
1.56
3.13
6.25
12.5
25
50
100
0
1.56
3.13
6.25
12.5
25
50
100
0
5
15
45
0
5
15
45
0
5
15
45
0
1.56
3.13
6.25
12.5
25
50
100
0
1.56
3.13
6.25
12.5
25
50
100
0
1.56
3.13
6.25
12.5
25
50
100
Concentration (μM)
0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
Relative Relative Relative Relative Relative Relative Relative Relative Relative Relative
growth growth viability viability motility motility viability viability growth growth
(adult) (larval)
Fig. 4 | Selectivin-A and selectivin-E are lead nematicides for plant-parasitic (SP) and tioxazafen (TX). A representative image taken from one of three
nematode control. a, Per cent effectiveness of 11 selectivin analogues and 4 independent trials is shown. c, Dose–response of phylogenetically diverse
commercial nematicides at preventing tomato plant root infection by non-target systems treated with sel-E, sel-A and eight commercial nematicides.
M. incognita (Methods). For each analogue, aqueous molar concentration, parts The phenotypic outcome of each treatment, relative to untreated control is
per million soil concentration and kg ha–1 values are indicated. The R-groups for denoted by the colour-coded scales. The following systems were treated:
each analogue are also indicated (see Fig. 1a for the core structure and R-group P. simiae, C. albicans, C. elegans, P. pacificus, P. hermaphrodita (P. herm.),
positions of selectivins). P values at the far right of the graph were obtained D. melanogaster; D. rerio, human HEK293 cells and human HepaRG cells. Dose–
from two-tailed one-sample t-tests comparing the mean per cent effectiveness response data are an average across: P. simiae (n = 3 TRs), C. albicans (n = 3 TRs),
of each treatment with zero effect. Error bars are s.e.m. b, Arabidopsis thaliana C. elegans (n = 10 BRs for sel-A, 4 BRs for sel-E, 6 BRs for all other compounds),
seedling establishment and greening in response to sel-E, sel-A and the P. pacificus (n = 3 BRs for sel-A, 4 BRs for sel-E, 6 BRs for all other compounds),
following eight commercial nematicides: abamectin (AB), fenamiphos (FE), P. herm. (n = 3 BRs), D. melanogaster (n = 3 BRs), D. rerio (n = 3 BRs), HEK293 (n = 3
fluensulfone (FS), fluopyram (FP), iprodione (IP), oxamyl (OX), spirotetramat TRs for sel-A, 3 TRs for FE, 4 TRs for all other compounds), HepaRG (n = 3 TRs).
Selectivin-A was active against free-living nematodes, as were several enzyme (Fig. 3j–l). Using this system, we individually expressed 19 dis-
marketed next-generation nematicides, but it was more selective for tinct M. incognita CYP enzymes in separate yeast strains and measured
nematodes than the majority of the commercial nematicides tested their growth in the presence or absence of either 50 µM selectivin-A
(Fig. 4b,c). Selectivin-A and selectivin-E have lipophilic properties or selectivin-E. These 19 CYP enzymes are a representative subset of a
(logP value of around 4) and are only moderately water soluble. These larger set of 31 CYP enzymes identified through bioinformatics assays
results indicate that they will not readily leach into ground water and from the most recent iteration of the M. incognita genome (Methods
are unlikely to be taken up systemically by plants42, thereby limiting resi- and Supplementary Tables 3,4). Selectivin-A-treated yeast cells express-
dues in fruits and other plant products ingested by consumers. Taken ing M. incognita CYP4731A3 grew slower than untreated cells (Fig. 5a,b
together, these data suggest that our best-performing lead nematicides and Supplementary Table 5). By contrast, yeast carrying the empty
can effectively and selectively control nematode pests. Going forwards, vector were relatively insensitive to selectivin-A treatment (Fig 5a,b and
it will be important to test the selectivins against additional organisms Supplementary Table 5). Yeast expressing CYP4731A3 produced sulfox-
that affect soil health and plants of commercial interest. ide conjugates of selectivin-A, whereas the empty vector control strain
did not (Fig. 5c,d, Extended Data Table 3 and Supplementary Table 1).
These data demonstrate that CYP4731A3 can bioactivate selectivin-A.
CYP4731A3 bioactivates selectivin-A Notably, CYP4731A3 has a high level of basal expression in M. incognita
To identify the M. incognita CYP enzyme (or enzymes) that bioactivate tissue, falling within the top 3.5% of all expressed genes, and it is the
selectivins, we exploited the yeast-based CYP expression system that third most highly expressed CYP of the 31 CYP enzymes we identified
we used to validate C. elegans CYP-35C1 as a selectivin-bioactivating in the genome43 (Supplementary Table 6). Furthermore, CYP4731A3
+ sel-A
target nematodes.
EV
280
Wavelength (nm)
260 Nematicides are deployed in metric tonne quantities to protect
crops, therefore they must have scalable synthesis routes and be inex-
1
M2 M3 Sel-A 320 pensive to produce. The synthesis scheme we chose to produce selec-
CYP4731A3
300
tivin analogues satisfies many of the criteria for a large-scale industrial
+ sel-A
280
0 260 process. Selectivin synthesis using this method requires only two steps,
–0.5 0 0.5 two easily accessible solvents, moderate reaction temperatures and
log2(fold difference in growth) pressures, and no transition metal catalysts. Furthermore, the synthesis
0
1
2
3
4
5
9
0
1
2.
2.
2.
2.
2.
2.
6.
7.
7.
relative to EV control
Retention time (min) of selectivin-A and selectivin-E uses inexpensive, easily accessible and
b d relatively non-toxic starting reagents. The yields we report here are
Relative growth
0 1 EV
P = 1.821 × 10–9 two-step yields; intermediates were not purified or isolated between
steps, which is advantageous for large-scale synthesis.
EV In recent years, it has become more common for soil-beneficial
CYP-4731A3
+ CYP4731A3 microbes such as entomopathogenic bacteria, fungi and nematodes, as
well as plant-growth-promoting rhizobacteria (PGPR) and mycorrhizal
0
12.5
25
50
100
Extended Data Fig. 1 | The selectivin-A imidazole-thiol metabolite is not Area under the curve (AUC) for metabolites M1-M3 and ITM-alk was calculated
nematicidal. a, Mass spectrometry (MS) data for the M4 metabolite HPLC at 260 nm, and at 300 nm for M4. Data are presented as mean ± SEM. For each
fraction and untreated control fraction, as well as MS/MS data for the 418.99 analyte, unpaired one-tailed Student’s t-tests were performed comparing
mass. The proposed M4 disulfide structure is shown, along with the protonated the means of the ± IAA conditions. P-values are shown. f, Dose-response for
imidazole-thiol monomer structure resulting from fragmentation. b, A proposed C. elegans L1s treated with selectivins or commercially-sourced ITM. g, Biomass-
metabolic pathway producing the imidazole-thiol metabolite (ITM) which normalized and background-corrected HPLC chromatograms for lysates
oxidizes to form M4/ITM disulfide (ITM-DiS). Imidazothiazole ring opening is of L1 worms incubated in commercially-sourced ITM, lysed ± IAA. The top
likely CYP-dependent. c, Biomass-normalized and background-corrected chromatogram is the ITM standard injected directly onto the column, and is
HPLC chromatograms for lysates of selectivin-A-treated L1 worms lysed ± not normalized or corrected. ITM oxidizes to ITM-DiS prior to HPLC analysis.
iodoacetamide (IAA). Peaks corresponding to selectivin-A (sel-A), metabolites ITM-DiS and ITM-alk peaks are indicated. h, HPLC-DAD quantification of
M1-M4, and alkylated ITM (ITM-alk) are indicated. d, MS data for the ITM-alk M4/ITM-DiS and ITM-alk in the lysates of 100 µM selectivin-A-treated or 100 µM
HPLC fraction and untreated control fraction. e, HPLC-DAD quantification of ITM-treated L1 worms, lysed ± IAA (n = 3 biological replicates with 200,000
selectivin-A, M1-M4, and ITM-alk in the lysates of selectivin-A-treated L1 worms worms per replicate). AUC was calculated at 260 nm. Data are presented as
lysed ± IAA (n = 3 biological replicates with 200,000 worms per replicate). mean ± SEM.
Extended Data Fig. 2 | NMR spectra for HPLC-purified selectivin-A 2.2 Hz, 1H), 4.74 (dd, J = 7.6, 5.3 Hz, 1H), 4.66 (dd, J = 14.9, 8.0 Hz, 1H), 3.86 (s, 2H),
metabolite M2. a, 1H NMR spectra. The structure of M2 with all carbon and 3.81 (dd, J = 14.9, 2.2 Hz, 1H), 3.62 (t, J = 6.2 Hz, 1H), 3.23 (dd, J = 14.2, 5.3 Hz, 1H),
heteroatom assignments is given. The top number in each multiplet box 3.08 (dd, J = 14.1, 7.6 Hz, 1H), 2.54 (ddd, J = 15.9, 7.9, 6.6 Hz, 1H), 2.48 (dt, J = 15.8,
corresponds to the carbon that the proton is connected to. 1H NMR (700 MHz, 6.6 Hz, 1H), 2.16 – 2.05 (m, 2H), 1.97 (s, 3H). b, 1H-1H COSY NMR spectra. c, 1H-13C
cd3od) δ 8.02 (s, 1H), 7.82 – 7.78 (m, 2H), 7.42 – 7.37 (m, 2H), 5.98 (dd, J = 7.9, HSQC NMR spectra. d, 1H-13C HMBC NMR spectra.
Article
Extended Data Fig. 3 | Selectivin bioactivation is nematode-specific. species. AUC is area under the curve at 260 nm. Data are presented as mean ±
a and b, Biomass-normalized and background-corrected HPLC chromatograms SEM. For c and d, unpaired one-tailed Student’s t-tests were performed for all
of lysates (a) and incubation buffers (b) from C. elegans (Ce), C. briggsae (Cb), pairwise comparisons of means. The means sharing a letter are statistically
P. pacificus (Pp), Rhabditophanes sp. KR3021 (KR), M. incognita (Mi), indistinguishable (p > 0.01). The p-values for each comparison in c and d can be
D. melanogaster (Dm), and D. rerio (Dr) larvae, as well as S. cerevisiae (Sc) cells, found in the Statistical Information subsection of the Methods. For e, unpaired
incubated in 100 µM selectivin-A. Selectivin-A and metabolite peaks are one-tailed Student’s t-tests were performed relative to the C. elegans mean.
indicated. c, Total detectable production of M1 + M2 + M3 in lysate and buffer P-values are shown. For c-e, n = 4 biological replicates for Ce, Cb, Pp, and Dr, and
for each of the seven species tested. d, M2 tissue accumulation in each species n = 3 biological replicates for KR, Mi, Sc, and Dm.
tested. e, Unmodified selectivin-A tissue accumulation in each of the test
Extended Data Fig. 4 | Selectivin treatment generally increases the root concentration, parts-per-million soil concentration, and kilograms-per-hectare
weights of tomato plants challenged with nematode infection. The effects values are shown. The R-groups for each selectivin analog are indicated (see
of 11 selectivin analogs and 4 commercial nematicides on the root weights of Fig. 1a for the selectivin core scaffold and R-group positions). Data are presented
tomato plants challenged with M. incognita infection is shown. Root weight as mean ± SEM.
values are percent of untreated control. For each analog, aqueous molar
Article
Extended Data Fig. 5 | Free-living nematodes likely detoxify selectivin-E via in the corresponding untreated control fractions. Raw counts were taken from
hydroxylation and subsequent phosphoglucosidation. a, Structure of centroid plots of the raw mass spectra. Counts below 1 × 103 were considered
selectivin-E (sel-E) with exact mass. b-c, Raw HPLC chromatograms for lysates not detectable (nd). Based on the presumptive masses of the three metabolites
of C. elegans and P. pacificus L1 worms incubated in 100 µM selectivin-E or we predict that E-M1 is a hydroxylated metabolite of selectivin-E, E-M2 is an
DMSO alone (untreated). The top chromatogram is the selectivin-E standard O-linked glucoside of selectivin-E, and E-M3 is an O-linked phosphoglucoside.
injected directly onto the column. The unmodified selectivin-E parent Molecular formulas are indicated for the predicted metabolite structures, as
compound and selectivin-E metabolites E-M1, E-M2, and E-M3 are indicated. well as exact masses, P. pacificus experimental accurate masses, and the ppm
c, Mass spectrometry (MS) data for the E-M1, E-M2, and E-M3 HPLC fractions difference between the two. A ppm difference of less than 5 ppm suggests that
and corresponding untreated control fractions. Raw MS counts are shown for the molecular formulae are correct. Ce, C. elegans; Pp, P. pacificus.
the most abundant masses identified in the mass spectra that are not detectable
Extended Data Table 1 | Anthelmintic/nematicide-resistant mutants are sensitive to the selectivins
Selectivin-A and selectivin-C LC50 data for wild-type worms and nine distinct nematicide/anthelmintic-resistant mutant strains. LC50 is the concentration at which 50% of the nematodes are
dead. OP, organophosphate; LEV, levamisole; APA, aminophenylamidine; THP, tetrahydropyrimidine; AAD, aminoacetonitrile derivative; BZ, benzimidazole.
Article
Extended Data Table 2 | Genetic resistance to selectivins is difficult to achieve
The number of genomes screened and the number of resistant mutants obtained from genetic screens for mutants that resist selectivin-A, selectivin-C, and four additional commercial nemati-
cides/anthelmintics. Wact-11 is a structural analog of the commercial nematicide fluopyram. Wact-11 and fluopyram share the same nematicidal mode-of-action, i.e. inhibition of mitochondrial
complex II, and wact-11 resistant mutants are also resistant to fluopyram14.
Extended Data Table 3 | Summary of MS data for selectivin-A metabolites from lysates
Raw MS counts for the m/z values corresponding to the selectivin-A (sel-A) metabolites M1, M2, M3, and M4 in HPLC fractions taken from lysates of five nematode species and three non-nematode
species incubated in selectivin-A. Raw counts were taken from centroid plots of the raw mass spectra. Counts below 1 x 103 were considered not detectable (nd). Ce, C. elegans; Cb, C. briggsae; Pp,
P. pacificus; KR, Rhabditophanes sp. KR3021; Mi, M. incognita; Sc WT, S. cerevisiae wild-type strain; Sc+35C1, S. cerevisiae expressing Ce-CYP-35C1; Sc + 4731A3, S. cerevisiae expressing
Mi-CYP4731A3; Dm, D. melanogaster; Dr, D. rerio. The KR, Mi, Sc + 35C1, and Sc + 4731A3 M4 metabolite fractions were not processed by MS.
Article
Extended Data Table 4 | HPLC solvent and flow rate gradients
Solvent and flow rate gradients for the HPLC methods used in this study. Solvent A is 4.9:95:0.1 (ACN:H2O:acetic acid). Solvent B is 95:4.9:0.1 (ACN:H2O:acetic acid).
nature research | reporting summary
Corresponding author(s): Peter J. Roy and Andrew R. Burns
Last updated by author(s): March 19, 2023
Reporting Summary
Nature Research wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency
in reporting. For further information on Nature Research policies, see our Editorial Policies and the Editorial Policy Checklist.
Statistics
For all statistical analyses, confirm that the following items are present in the figure legend, table legend, main text, or Methods section.
n/a Confirmed
The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement
A statement on whether measurements were taken from distinct samples or whether the same sample was measured repeatedly
The statistical test(s) used AND whether they are one- or two-sided
Only common tests should be described solely by name; describe more complex techniques in the Methods section.
For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted
Give P values as exact values whenever suitable.
For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings
For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes
Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated
Our web collection on statistics for biologists contains articles on many of the points above.
Data analysis HPLC-DAD data | HP Chemstation for LC 3D (v. A.10.02), Microsoft Excel for Mac (v. 16.16.27), GraphPad Prism (v. 6.0), and a custom python
script for generating chromatogram heatmaps (https://doi.org/10.5281/zenodo.7731172) using Python (v. 3.9.13) and Anaconda Spyder (v.
5.3.3)
MS data | Agilent MassHunter Qualitative Analysis (v. B.06.00 SP1), Microsoft Excel for Mac (v. 16.16.27), and ChemDraw Professional (v.
18.2.0.37)
NMR data | MestreNova (v 14.3.1)
April 2020
All other data were analyzed using Microsoft Excel for Mac (v. 16.16.27) and/or GraphPad Prism (v. 6.0).
Heatmaps of dose-response data were generated using a custom R script (https://doi.org/10.5281/zenodo.7731172) using R (v. 4.2.2).
Chemical structures were drawn using ChemDraw Professional (v. 18.2.0.37).
For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors and
reviewers. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Research guidelines for submitting code & software for further information.
1
Data
Field-specific reporting
Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.
Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences
For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf
Replication Statistical analyses were performed as described in the manuscript. At least three independent replicates were performed for nearly all
experiments and variability was relatively low between replicates. All measurements were reproducible. Covariates such as temperature,
humidity, light exposure, age, developmental stage, media/buffer composition, etc. were held constant to minimize variability.
Randomization In all experiments, test subjects (organisms, cells, etc.) were chosen at random for experimentation and there was no attempt at cherry
picking or biasing the samples.
Blinding Blinding was performed for the following: mouse studies, soil-based plant-parasitic nematode assays, arabidopsis assays, fish assays, human
cell assays, C. albicans assays, in vitro M. incognita and M. chitwoodi assays, mass spectrometry, and NMR. All other experiments were not
blinded because only one person was available for experimentation and the experiments are quantitative in nature thereby reducing the risk
of subjective bias.
Clinical data
Dual use research of concern
2
Eukaryotic cell lines
Authentication HEK293 cells were authenticated by STR analysis at the Toronto Hospital for Sick Children authentication facility. HepaRG
cells were used directly from ThermoFisher, and were authenticated by the commercial supplier via SNP genotyping.
Mycoplasma contamination HEK293 cells are tested regularly for mycoplasma contamination. Tests performed a short time before experimentation
confirmed that the cell line is negative for mycoplasma contamination. Mycoplasma contamination of the HepaRG cell line
was confirmed negative by the commercial supplier (ThermoFisher), and the cells were used directly from the supplier.
Commonly misidentified lines Neither the HEK293 nor the HepaRG cell line is on the list of commonly misidentified cell lines.
(See ICLAC register)
Field-collected samples Cooperia oncophora eggs were collected from the faeces of infected calves in accordance with the Canadian Council of Animal Care
regulations; Licenses #AC13-0157 (University of Calgary) and #1730 (Agriculture AgriFood Canada Research Station, Lethbridge).
Ethics oversight For the mouse studies, all experiments were approved by the University of Calgary’s Life and Environmental Sciences Animal Care
Committee (protocol AC17-0082). All protocols for animal use and euthanasia were in accordance with the Canadian Council for
Animal Care (Canada). For the fish studies, ethics protocol 100005273 was approved by the Animal Care Committee at The Hospital
for Sick Children, Canada.
Note that full information on the approval of the study protocol must also be provided in the manuscript.
April 2020