Article

Selective control of parasitic nematodes

using bioactivated nematicides
­­­
https://doi.org/10.1038/s41586-023-06105-5 Andrew R. Burns1,2 ✉, Rachel J. Baker3, Megan Kitner4, Jessica Knox1,2, Brittany Cooke1,2,
Jonathan R. Volpatti2,5, Aditya S. Vaidya6,7, Emily Puumala2, Bruna M. Palmeira8,
Received: 18 May 2021
Elizabeth M. Redman8, Jamie Snider1, Sagar Marwah9,10, Sai W. Chung9,10,
Accepted: 20 April 2023 Margaret H. MacDonald11, Jens Tiefenbach1, Chun Hu1,2, Qi Xiao12, Constance A. M. Finney12,
Henry M. Krause1,2, Sonya A. MacParland9,10, Igor Stagljar1,2,13,14, John S. Gilleard8,
Published online: 24 May 2023
Leah E. Cowen2, Susan L. F. Meyer11, Sean R. Cutler6,7, James J. Dowling2,5, Mark Lautens3,
Check for updates Inga Zasada4 & Peter J. Roy1,2,15 ✉

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
population approaches 10 billion people1–4. There is a scarcity of arable
land for agricultural expansion, and land-conversion efforts are con-
strained by social and ecological factors. Maximizing production from
currently cultivated land will be crucial to ensure global food security1–4.
Farmers rely on agrochemicals to maximize yields by controlling
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
part in PPN control for decades; however, concerns over environmental
toxicity and human safety have justifiably prompted bans on the most
commonly used nematicides8,9. The lack of available agents to control
PPNs is stark. Of the 20 key nematicides used in the twentieth century,
only 4 are currently approved for use in the European Union and only 3
are used in the USA without restriction9–11. Although warranted, these
withdrawals leave farmers with limited options and no control meas-
ures for several PPNs8,12.
Despite the need for more selective PPN control measures, only
seven non-fumigant synthetic nematicides have been developed
in the past 25 years9,13. These next-generation nematicides are flu-
azaindolizine, fluensulfone, fluopyram, iprodione, spirotetramat,
tioxazafen and cyclobutrifluram. Unfortunately, iprodione has already
been banned in Europe owing to its carcinogenic potential and risk
to aquatic life10. Meanwhile, the market release of a tioxazafen-based
seed treatment has been postponed owing to reports of skin irritation
in individuals handling this nematicide. There is a pressing need for
new nematicides with improved selectivity.
Selective imidazothiazole nematicides
From a library of uncharacterized compounds that disrupt the growth
of the free-living nematode C. elegans14, we identified three new
imidazothiazole-containing molecules that selectively kill nematodes
(Fig. 1a). At low micromolar concentrations, these selective imidazothia-
zole nematicides, which we call selectivin-A, selectivin-B and selectivin-C,
killed all four of the free-living nematode species that we can easily cul-
ture in the laboratory. Moreover, these selectivins were able to kill eggs of
a nematode parasite of cattle and the infective juveniles of at least one out
of three distinct species of root-knot nematode that infect crops (Fig. 1b).
These eight nematode species span five genera and two evolutionary
clades, which suggests that diverse nematode species, including para-
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

102 | Nature | Vol 618 | 1 June 2023

 a b Free-living nematodes Parasitic nematodes

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
selectivin-A (Sel-A), selectivin-B (Sel-B) and selectivin-C (Sel-C). b, Selectivin
dose–response for the following eight nematode species: C. elegans (clade V)
(Ce); Caenorhabditis briggsae (clade V) (Cb); Pristionchus pacificus (clade V)
(Pp); Rhabditophanes spp. KR3021 (clade IV) (KR); Cooperia oncophora
(clade V) (Co); M. incognita (clade IV) (Mi); Meloidogyne chitwoodi (clade IV)
(Mc); and Meloidogyne hapla (clade IV) (Mh). The colour-coded scale denotes
viability relative to untreated control. c, Selectivin dose–response for the
following six phylogenetically distinct non-nematode model systems: Sc,
S. cerevisiae (Sc); Candida albicans (Ca); D. rerio (Dr); HEK293 cells (HEK);
HepaRG cells; D. melanogaster adults (Dm (ad)); and D. melanogaster larvae
(Dm (lv)). Viability was assessed for D. rerio and D. melanogaster larvae. Cell
growth was assessed for S. cerevisiae, C. albicans, HEK293 cells and HepaRG cells.
Motility was assessed for D. melanogaster adults. The colour-coded scale denotes
viability, growth or motility relative to untreated control, as appropriate for
the system tested. In b,c, dose–response data are an average across n biological
replicates (BRs) or technical replicates (TRs), with the following n values for each
system: Ce (n = 10 BRs for sel-A, 9 BRs for sel-B, 13 BRs for sel-C), Cb (n = 3 BRs),
Pp (n = 3 BRs), KR (n = 3 BRs), Co (n = 3 BRs), Mi (n = 6 BRs), Mc (n = 6 BRs), Mh
(n = 6 BRs), Sc (n = 3 BRs), Ca (n = 3 TRs), Dr (n = 3 BRs), HEK (n = 3 TRs), HepaRG
(n = 3 TRs), Dm (n = 3 BRs).

exhibited little-to-no activity against fungi, insects, fish or human cells
at concentrations that readily killed nematodes (Fig. 1c). Furthermore,
mice given a daily oral dose of 50 mg kg–1 selectivin-A for 5 days showed
no obvious pathologies relative to untreated mice. To our knowledge,
broad-spectrum nematicidal activity has not been previously described
for any compound containing the 6-phenylimidazo[2,1-b]thiazole scaf-
fold of selectivins. Thus, the selectivins represent a new structural class
for the development of selective and effective PPN control.
Selectivins have a new mode of action
As a first step towards understanding the mode of action of selectivins,
we tested selectivin-A and selectivin-C against eight mutant strains
of C. elegans that are each specifically resistant to one of the major
anthelmintic or nematicide classes14–22 (Extended Data Table 1). None
of the mutants were resistant to selectivins, which suggests that selec-
tivins have a distinct mode of action compared with the commercial
agents examined.
To further characterize the mode of action of the selectivins, we
performed a forward genetic screen of C. elegans mutants (generated
through random mutagenesis) that resist selectivin-induced lethality.
This approach has previously produced large numbers of mutants
that specifically resist many of the major classes of anthelmintics and
nematicides that are commercially used14,15,17,19,20,22,23. Despite screening
more than 10 million mutagenized genomes, selectivin-resistant worms
could not be generated (Extended Data Table 2). This result reinforces
the conclusion that selectivins kill nematodes using a mechanism that
is distinct from traditional nematode control agents.
Selectivins are bioactivated nematicides
Previous work using C. elegans has shown that metabolites of com-
pounds that are lethal to nematodes often accumulate in worm tissue24.
It is therefore possible that a nematicide could be transformed from an
innocuous parent molecule into a metabolic product that is lethal to
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-
mal CYP enzymes26, many of which are known to metabolize drugs27.
To test whether selectivins are bioactivated by C. elegans, we capital-
ized on previous work that achieved broad disruption of microsomal
CYP enzymes by compromising EMB-8 activity during larval develop-
ment, bypassing its essential role during embryogenesis27. EMB-8 is
the C. elegans P450 oxidoreductase (POR) enzyme that is a necessary
redox partner of all microsomal CYP enzymes27,28. Consistent with the
notion that selectivins are bioactivated nematicides, emb-8 knockdown
rendered C. elegans resistant to selectivin-induced lethality (Fig. 2a). By
contrast, the compound wact-11, which does not require bioactivation
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
100 µM selectivin-A for 6 h and then processed whole worm lysates
by reversed-phase high-performance liquid chromatography (HPLC)
coupled with a diode-array absorbance detector (HPLC-DAD) (Fig. 2b).
We identified five distinct peaks in the selectivin-A-treated lysates
that were absent from the untreated controls. The peak that eluted
at 7 min had the same retention time and absorbance spectrum as
the selectivin-A standard that was directly injected onto the column.
This peak is probably unmodified selectivin-A (parent molecule) that
persists in worm tissue following incubation. The remaining four peaks
eluted earlier in the HPLC run relative to selectivin-A, which suggested
that they are relatively less lipophilic. These peaks were absent from
the selectivin-A standard, and they were not found in the lysates of
dead worms incubated in selectivin-A. This result supports the idea
that they are bona fide selectivin-A metabolites and not accumulated
contaminants or spontaneous oxidation products (Fig. 2b,c). We named
these presumptive selectivin-A metabolites M1, M2, M3 and M4.
Consistent with our hypothesis that selectivin-A is bioactivated, the
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).
Unmodified selectivin-A is unlikely to be the active agent in vivo because
worms in which emb-8 was knocked down survived treatment with
100 µM selectivin-A even though the internal concentration was nearly
twice as high as that found in wild-type worms (Fig. 2a,d). Thus, selec-
tivins are bioactivated pro-nematicides.
Bioactivation to a toxic electrophile
To identify the toxic selectivin metabolite (or metabolites), we
collected HPLC fractions containing the four selectivin-A metabo-
lites and analysed their structures by electrospray ionization mass

Nature | Vol 618 | 1 June 2023 | 103

Article
a Wild-type emb-8 KD b mAU per mg dry weight
Sel-A
Sel-B 0 50 100 150 200
Sel-C 340
Sel-A 320
Wact-11

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

104 | Nature | Vol 618 | 1 June 2023

 a c
M1 Control 501.07 MS/MS 148.04 M3 Control 372.02 MS/MS
H NH2
HOOC N HOOC NH2
Relative abundance

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

(relative to untreated) (%)

– NACET
+ NACET
e Sulfoxide
Electrophilic
Glutathione γ-Glu-Cys Cys 150
carbon
Sel-A CYP metabolite

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 (%)

40 Sel-C Sel-C bioactivation

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

Individual C. elegans CYP RNAi KDs AUC (M1–M3) AUC (M1–M3)

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).

Nature | Vol 618 | 1 June 2023 | 105

Article
To identify the CYP enzyme (or enzymes) responsible for selectivin
bioactivation, we individually knocked down 60 out of 76 C. elegans
microsomal CYP enzymes by RNA interference (RNAi) and assayed for
selectivin-A resistance. cyp-35c1 knockdown conferred resistance to
selectivin-A, which suggested that it has a role in bioactivation (Fig. 3g
and Supplementary Table 2). We validated this result with a cyp-35c1
deletion mutant, which demonstrated resistance to all three selectivin
analogues (Fig. 3h). Furthermore, abundance of the metabolites M1, M2
and M3 were reduced by half in the tissues of cyp-35c1 deletion mutants
compared with wild-type worms (Fig. 3i), which provides support that
the reactive sulfoxide metabolite is produced by CYP-35C1. Metabolite
production was not completely eliminated in cyp-35c1 deletion mutants,
which suggested that additional CYP enzymes act redundantly to pro-
duce the sulfoxide metabolite. This redundancy may explain why we did
not identify cyp-35c1 in our genetic screen, which was carried out using L1
worms that are twice as sensitive to selectivin-A compared with the young
adults used for the RNAi screen and follow-up dose–response analyses
(Figs. 1b and 3h). To demonstrate that CYP-35C1 is able to produce the
reactive metabolite, we heterologously expressed it in budding yeast
and assayed for both the production of the reactive metabolite and
for selectivin-induced growth inhibition. Unlike wild-type yeast, yeast
expressing nematode CYP-35C1 readily produced the sulfoxide conju-
gates of selectivin-A and were sensitive to selectivin-induced cytotoxic-
ity (Fig. 3j–l, Extended Data Table 3 and Supplementary Table 1). These
results demonstrate that CYP-35C1 is both necessary and sufficient for
the production of the selectivin sulfoxide metabolite and further support
the conclusion that selectivins are bioactivated to a toxic electrophile.
Bioactivation is specific to nematodes
Given their selective nematicidal activity, we proposed that the bio-
activation of selectivins may be phylogenetically restricted to nema-
todes. To test the phylogenetic specificity of selectivin bioactivation,
we analysed selectivin-A metabolites in both the lysates and incubation
buffers of five distinct nematode species spanning four genera and two
evolutionary clades, and in three non-nematode species from diverse
phyla (Extended Data Fig. 3a,b). The non-nematode species tested were
the yeast Saccharomyces cerevisiae, the fly Drosophila melanogaster
and the fish Danio rerio. Low-molecular-weight thiol conjugates of the
sulfoxide metabolite were abundantly produced by all five of the nema-
tode species, including the PPN M. incognita (Extended Data Fig. 3a–c,
Extended Data Table 3 and Supplementary Table 1). By contrast, the three
non-nematode species produced significantly less of the conjugates
relative to nematodes (Extended Data Fig. 3a–c). Notably, the conjugates
were undetectable by HPLC-DAD and MS in yeast lysates and buffers,
and they are undetectable by HPLC-DAD in the lysates and buffers from
flies (Extended Data Fig. 3a–c and Extended Data Table 3). Only the
cysteine conjugate was detectable by MS at very low levels in fly lysates
(Extended Data Table 3). Fish produced more than fourfold less of the
conjugates than C. elegans, and more than twofold less than the other
nematode species tested (Extended Data Fig. 3a–c). Furthermore, the
glutathione conjugate (M2) readily accumulated in the tissues of all five
of the nematode species tested over the course of the incubation period,
but not in the tissues of the non-nematode species. This result suggests
that abundant sulfoxide metabolite production and oversaturation of
the detoxification pathway is common to nematodes but not to distinct
phyla (Extended Data Fig. 3a,d). Notably, tissue accumulation of the
selectivin-A parent compound in the non-nematode species was com-
parable to, or greater than, accumulation in C. elegans tissue (Extended
Data Fig. 3a, e). This result suggests that poor bioavailability is unlikely to
account for the lower levels of metabolite production by non-nematode
species and provides further evidence that the selectivins require
bioactivation to induce lethality. Taken together, these data indicate
that distinct nematode species, including PPNs, are capable of strongly
bioactivating the selectivins, and that bioactivation to acutely toxic
106 | Nature | Vol 618 | 1 June 2023
levels is phylogenetically restricted to nematodes. Future work will be
required to determine the long-term effects of the relatively low levels
of metabolite production observed in some non-nematode species.
Lead nematicides for PPN control
We were encouraged by the selectivity, broad-spectrum activity and new
mechanism of action of the selectivins. Therefore we performed an ana-
logue series screen to identify lead nematicides capable of effectively con-
trolling root infection by the root-knot nematode M. incognita, one of the
most destructive crop pathogens in the world36. Eleven analogues were
included in the series, along with all four of the marketed next-generation
soil-applied synthetic nematicides for which powder is commercially
available: fluensulfone, fluopyram, iprodione and tioxazafen. Selectivin
analogues were either purchased or synthesized (Methods). Tests were
performed against M. incognita in its soil environment with a tomato host
plant (Fig. 4a). Soil-based assays are vital for establishing the real-world
utility of candidate nematicides, as many compounds that are effective
in vitro lose activity in the soil because of poor mobility and/or degrada-
tion37. We assayed compounds by adding 20 ml of 45 µM compound in
water and between 1,000 and 2,500 infective second-stage juvenile ( J2)
nematode larvae to 90 g of potted soil. The final soil concentration was
approximately 2.5 ppm or 4.5 kg per hectare (kg ha–1), which is compa-
rable to the application rates of commercial nematicides37,38. After 24 h,
we planted 2–3-week-old tomato seedlings in the soil. Eight weeks later,
the per cent effectiveness of the compounds at reducing the number of
eggs in the roots was calculated relative to untreated controls (Methods).
Of the 15 compounds tested, selectivin analogues A, C and E, and the
four commercial nematicides, had per cent effectiveness values that dif-
fered significantly from zero effect (one-sample t-test P < 0.01) (Fig. 4a).
Fluensulfone and fluopyram demonstrated nearly 100% effectiveness
and were the best-performing nematicides. Consistent with previous
reports, iprodione was the least effective commercial nematicide (38%
effective)13. Selectivin-A and selectivin-E were the most effective selec-
tivin analogues (43% and 56% effective, respectively), and their efficacies
were not significantly different from that of tioxazafen (one-tailed t-test
P > 0.05) (Fig. 4a). Of note, we challenged the plants with a nematode
density that was at least twofold greater than what is considered high
density in fields39,40, which may account for why some of the nematicides
tested had incomplete efficacies in this assay. Treatment with the major-
ity of selectivins, including selectivin-A and selectivin-E, increased the
root weight relative to untreated control (Extended Data Fig. 4), which
is probably due to the reduced nematode burden.
To assess the selectivity of selectivin-A and selectivin-E for PPNs, we
tested them against a panel of phylogenetically diverse and experi-
mentally tractable non-target systems, including rhizobacteria, fungi,
plants, free-living nematodes (including the soil-beneficial nematode
Phasmarhabditis hermaphrodita), insects, fish and two different human
cell types (Fig. 4b,c). One of the human cell types is HepaRG, which is a
liver progenitor cell type that is a well-established model for assessing
drug metabolism and bioactivation in human liver41. We also included
in our tests all of the newly developed synthetic nematicides that are
purchasable, the natural product nematicide abamectin, the organo-
phosphate nematicide fenamiphos and the carbamate nematicide
oxamyl, which is still in use. Selectivin-E had the best selectivity profile
compared with all of the nematicides tested. Selectivin-E was inactive
against all of the non-target systems up to 100 µM, which was the high-
est concentration assayed. By contrast, every commercial nematicide
tested had unwanted toxicity in at least one of the non-target assays.
Notably, selectivin-E was non-lethal to the three free-living nematodes
included in our analyses, which suggested that it may be innocuous to
non-target soil-beneficial nematodes. Notably, the free-living nema-
todes did not produce a reactive sulfoxide metabolite of selectivin-E.
Instead, it is likely that they detoxify selectivin-E through hydroxyla-
tion and subsequent phosphoglucosidation (Extended Data Fig. 5).
 a [Aqueous] [Soil] b
Sel-E AB FE FS FP
Treatment (μM) (ppm) kg ha–1 R1 R2 R3 n

Mock
Fluensulfone 38 2.5 4.5 – – – 6 7.76 × 10–15
Fluopyram 28 2.5 4.5 – – – 6 1.44 × 10–15

Tioxazafen 45 2.3 4.1 – – – 6 0.004

15 μM
Sel-E 45 2.5 4.5 Cl H Me 6 0.001

Sel-A 45 2.4 4.3 Cl H H 6 0.009

45 μM
Sel-C 45 2.8 5.0 Br H H 6 0.007
Iprodione 34 2.5 4.5 – – – 6 0.008

Sel-K 45 3.4 6.1 I H Me 6 0.025

Sel-G 45 2.3 4.1 F H Me 4 0.167 Sel-A IP OX SP TX
Sel-B 45 2.2 4.0 F H H 6 0.066

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 (%)

c Rhizobacteria Fungus Non-target nematodes Insect Fish Human cells

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
nematode control. a, Per cent effectiveness of 11 selectivin analogues and 4
commercial nematicides at preventing tomato plant root infection by
M. incognita (Methods). For each analogue, aqueous molar concentration, parts
per million soil concentration and kg ha–1 values are indicated. The R-groups for
each analogue are also indicated (see Fig. 1a for the core structure and R-group
positions of selectivins). P values at the far right of the graph were obtained
from two-tailed one-sample t-tests comparing the mean per cent effectiveness
of each treatment with zero effect. Error bars are s.e.m. b, Arabidopsis thaliana
seedling establishment and greening in response to sel-E, sel-A and the
following eight commercial nematicides: abamectin (AB), fenamiphos (FE),
fluensulfone (FS), fluopyram (FP), iprodione (IP), oxamyl (OX), spirotetramat
(SP) and tioxazafen (TX). A representative image taken from one of three
independent trials is shown. c, Dose–response of phylogenetically diverse
non-target systems treated with sel-E, sel-A and eight commercial nematicides.
The phenotypic outcome of each treatment, relative to untreated control is
denoted by the colour-coded scales. The following systems were treated:
P. simiae, C. albicans, C. elegans, P. pacificus, P. hermaphrodita (P. herm.),
D. melanogaster; D. rerio, human HEK293 cells and human HepaRG cells. Dose–
response data are an average across: P. simiae (n = 3 TRs), C. albicans (n = 3 TRs),
C. elegans (n = 10 BRs for sel-A, 4 BRs for sel-E, 6 BRs for all other compounds),
P. pacificus (n = 3 BRs for sel-A, 4 BRs for sel-E, 6 BRs for all other compounds),
P. herm. (n = 3 BRs), D. melanogaster (n = 3 BRs), D. rerio (n = 3 BRs), HEK293 (n = 3
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
marketed next-generation nematicides, but it was more selective for
nematodes than the majority of the commercial nematicides tested
(Fig. 4b,c). Selectivin-A and selectivin-E have lipophilic properties
(logP value of around 4) and are only moderately water soluble. These
results indicate that they will not readily leach into ground water and
are unlikely to be taken up systemically by plants42, thereby limiting resi-
dues in fruits and other plant products ingested by consumers. Taken
together, these data suggest that our best-performing lead nematicides
can effectively and selectively control nematode pests. Going forwards,
it will be important to test the selectivins against additional organisms
that affect soil health and plants of commercial interest.
CYP4731A3 bioactivates selectivin-A
To identify the M. incognita CYP enzyme (or enzymes) that bioactivate
selectivins, we exploited the yeast-based CYP expression system that
we used to validate C. elegans CYP-35C1 as a selectivin-bioactivating
enzyme (Fig. 3j–l). Using this system, we individually expressed 19 dis-
tinct M. incognita CYP enzymes in separate yeast strains and measured
their growth in the presence or absence of either 50 µM selectivin-A
or selectivin-E. These 19 CYP enzymes are a representative subset of a
larger set of 31 CYP enzymes identified through bioinformatics assays
from the most recent iteration of the M. incognita genome (Methods
and Supplementary Tables 3,4). Selectivin-A-treated yeast cells express-
ing M. incognita CYP4731A3 grew slower than untreated cells (Fig. 5a,b
and Supplementary Table 5). By contrast, yeast carrying the empty
vector were relatively insensitive to selectivin-A treatment (Fig 5a,b and
Supplementary Table 5). Yeast expressing CYP4731A3 produced sulfox-
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.
Notably, CYP4731A3 has a high level of basal expression in M. incognita
tissue, falling within the top 3.5% of all expressed genes, and it is the
third most highly expressed CYP of the 31 CYP enzymes we identified
in the genome43 (Supplementary Table 6). Furthermore, CYP4731A3

Nature | Vol 618 | 1 June 2023 | 107

Article
a c mAU per mg dry weight against non-nematode phyla, this new class of nematicides has the
3
potential to selectively control a wide variety of PPNs. Unlike several
CYP4731A3 0 250 500
new nematicides on the market, which are repurposed fungicides or
Sel-A 320
P < 0.01 M1 M2 M3
insecticides9,13, the selectivins are ‘true nematicides’ that selectively
2 300
–log10(P value)

+ 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

0 5,000 10,000 15,000

[Sel-A] (μM)
Fig. 5 | M. incognita CYP4731A3 bioactivates selectivin-A. a, Volcano plot of
the yeast-based M. incognita CYP expression screen data. The x axis shows the
log 2-transformed fold difference in growth between the EV control and 19
M. incognita CYP-expressing yeast strains resulting from treatment with 50 µM
sel-A. The y axis shows log10 -transformed P values obtained from unpaired
two-tailed Student’s t-tests comparing mean relative growth values between
CYP-expressing strains and EV control. Each data point corresponds to an
individual yeast strain expressing a specific M. incognita CYP enzyme. The data
point above the orange dotted line has a P value of 0.0034 and corresponds to
the strain expressing M. incognita CYP4731A3. Raw screening data can be found
in Supplementary Table 5. n = 6 BRs. b, Sel-A dose–response for yeast expressing
CYP4731A3 and yeast carrying an EV that does not express CYP4731A3.
Dose–response data are an average across 13 and 11 BRs for EV and CYP4731A3-
expressing yeast, respectively. c, Biomass-normalized and background-
corrected HPLC chromatograms of lysates from yeast expressing CYP4731A3
or EV control yeast treated with 100 µM sel-A. Unmodified sel-A and metabolites
M2 and M3 are indicated. MS data for HPLC-purified M1–M3 fractions can be
found in Extended Data Table 3 and Supplementary Table 1. d, HPLC-DAD
quantification of total M2 and M3 in the lysates of yeast expressing M. incognita
CYP4731A3 and yeast carrying EV that does not express CYP4731A3 (n = 4 BRs).
AUC is area under the curve for M2 and M3 combined, calculated at 260 nm.
Data are presented as the mean ± s.e.m. The P value was obtained from an
unpaired one-tailed Student’s t-test comparing the means of EV control and
CYP4731A3-expressing yeast. M1 was not detectable.
expression is induced when nematodes were treated with commercial
nematicides43. Taken together, these results suggest that CYP4731A3
may bioactivate selectivin-A in vivo. None of the 19 CYP-expressing
yeast strains treated with selectivin-E had perturbed growth (Sup-
plementary Table 5). We speculate that selectivin-E may be bioacti-
vated either by an untested CYP enzyme or by a combination of CYP
enzymes in a step-wise fashion. Alternatively, selectivin-E may have
a CYP-independent mechanism of action. Regardless, we identified
a target CYP enzyme for the lead nematicide selectivin-A, which will
facilitate lead optimization, and we established CYP-mediated bioac-
tivation as a new approach to achieve selective killing of nematodes.
Discussion
Here we reported our discovery of the nematicidal utility and mode
of action of the selectivins, which is a nematicide class that can kill
through metabolic bioactivation in nematodes. From a small set of ana-
logues, we identified promising lead compounds that can effectively
control the PPN M. incognita. Given the broad-spectrum activity of
the selectivins against diverse nematode species, and their innocuity
AUC (M2 + M3) fungi, to be applied to soil or seeds as part of an integrated pest manage-
ment strategy44–46. Ideally, new nematicides will complement, rather
than antagonize, these organisms. It is notable that our lead nematicide
selectivin-E had no effect on either the PGPR Pseudomonas simiae or
the soil-beneficial gastropod-killing nematode P. hermaphrodita, which
suggests that it could be used to control PPNs in combination with
soil-beneficial microbes without compromising their utility. Our other
lead nematicide, selectivin-A, was also innocuous against P. simiae and
two distinct fungal species. However, despite demonstrating greater
nematode selectivity than the majority of the commercial nemati-
cides examined, selectivin-A was lethal to P. hermaphrodita and other
free-living nematodes. This property of selectivin-A does not preclude
it from commercial development given that several next-generation
commercial nematicides are also lethal to free-living nematodes
(for example, abamectin, fluopyram and tioxazafen). However, its
potentially adverse effects on beneficial soil-dwelling nematode com-
munities should be minimized. In principle, one way to mitigate the
negative effects of selectivin-A on beneficial nematodes is to formulate
it as a seed treatment, thereby confining its activity to a concentrated
zone around the seed and developing seedling. Indeed, both abamectin
and tioxazafen are formulated as seed treatments47,48.
Our work demonstrates that worm-bioactivated compounds can be
broadly effective against disparate worm species. This result is perhaps
unexpected because broad-spectrum activity of a worm-activated
compound will depend on low interspecies and intraspecies variability
in enzyme expression and activity. Indeed, drug metabolism commonly
varies among individuals of the same species49. Thus, it would not be
surprising if a pro-nematicide approach was overlooked by industrial
discovery programmes. Although we focused on PPN control, this
approach could also be applied to anthelmintics for the treatment of
worms that parasitize humans, livestock and companion animals50.
Online content
Any methods, additional references, Nature Portfolio reporting summa-
ries, source data, extended data, supplementary information, acknowl-
edgements, peer review information; details of author contributions
and competing interests; and statements of data and code availability
are available at https://doi.org/10.1038/s41586-023-06105-5.
1. Tilman, D., Balzer, C., Hill, J. & Befort, B. L. Global food demand and the sustainable
intensification of agriculture. Proc. Natl Acad. Sci. USA 108, 20260–20264 (2011).
2. Hunter, M. C., Smith, R. G., Schipanski, M. E., Atwood, L. W. & Mortensen, D. A. Agriculture
in 2050: recalibrating targets for sustainable intensification. Bioscience 67, 386–391
(2017).
3. Cunningham, S. A. et al. To close the yield-gap while saving biodiversity will require
multiple locally relevant strategies. Agric. Ecosyst. Environ. 173, 20–27 (2013).

108 | Nature | Vol 618 | 1 June 2023

4. Popp, J., Petö, K. & Nagy, J. Pesticide productivity and food security. A review. Agron.
Sustain. Dev. 33, 243–255 (2013).
5. Oerke, E.-C. Crop losses to pests. J. Agric. Sci. 144, 31–43 (2006).
6. Singh, S. K., Hodda, M. & Ash, G. J. Plant-parasitic nematodes of potential phytosanitary
importance, their main hosts and reported yield losses. EPPO Bull. 43, 334–374 (2013).
7. Abad, P. et al. Genome sequence of the metazoan plant-parasitic nematode Meloidogyne
incognita. Nat. Biotechnol. 26, 909–915 (2008).
8. Jones, R. K. in Nematology in South Africa: A View from the 21st Century (eds Fourie, H. et al.)
129–150 (Springer, 2017).
9. Desaeger, J., Wram, C. & Zasada, I. New reduced-risk agricultural nematicides—rationale
and review. J. Nematol. 52, e2020-911 (2020).
10. EU Pesticides Database, https://ec.europa.eu/food/plant/pesticides/eu-pesticides-database/
start/screen/active-substances (European Commission).
11. NPIC Product Research Online (NPRO), http://npic.orst.edu/NPRO/# (National Pesticide
Information Center, accessed 25 July 2022).
12. Jordan, S., Nischwitz, C., Ramirez, R. & Gordillo, L. F. Managing the spread of alfalfa stem
nematodes (Ditylenchus dipsaci): the relationship between crop rotation periods and
pest reemergence. Nat. Resour. Model. 30, e12083 (2017).
13. d’Errico, G., Giacometti, R., Roversi, P. F., d’Errico, F. P. & Woo, S. L. Mode of action and
efficacy of iprodione against the root-knot nematode Meloidogyne incognita. Ann. Appl.
Biol. 171, 506–510 (2017).
14. Burns, A. R. et al. Caenorhabditis elegans is a useful model for anthelmintic discovery.
Nat. Commun. 6, 7485 (2015).
15. Culetto, E. et al. The Caenorhabditis elegans unc-63 gene encodes a levamisole-
sensitive nicotinic acetylcholine receptor α subunit. J. Biol. Chem. 279, 42476–42483
(2004).
16. Hu, Y., Xiao, S. H. & Aroian, R. V. The new anthelmintic tribendimidine is an l-type
(levamisole and pyrantel) nicotinic acetylcholine receptor agonist. PLoS Negl. Trop. Dis.
3, e499 (2009).
17. Driscoll, M., Dean, E., Reilly, E., Bergholz, E. & Chalfie, M. Genetic and molecular analysis
of a Caenorhabditis elegans β-tubulin that conveys benzimidazole sensitivity. J. Cell Biol.
109, 2993–3003 (1989).
18. Dent, J. A., Smith, M. M., Vassilatis, D. K. & Avery, L. The genetics of ivermectin resistance
in Caenorhabditis elegans. Proc. Natl Acad. Sci. USA 97, 2674–2679 (2000).
19. Kaminsky, R. et al. A new class of anthelmintics effective against drug-resistant nematodes.
Nature 452, 176–180 (2008).
20. Guest, M. et al. The calcium-activated potassium channel, SLO-1, is required for the action
of the novel cyclo-octadepsipeptide anthelmintic, emodepside, in Caenorhabditis elegans.
Int. J. Parasitol. 37, 1577–1588 (2007).
21. Kawasaki, I., Jeong, M. H., Oh, B. K. & Shim, Y. H. Apigenin inhibits larval growth of
Caenorhabditis elegans through DAF-16 activation. FEBS Lett. 584, 3587–3591 (2010).
22. Rand, J. B. & Russell, R. L. Choline acetyltransferase-deficient mutants of the nematode
Caenorhabditis elegans. Genetics 106, 227–248 (1984).
23. Hu, Y., Platzer, E. G., Bellier, A. & Aroian, R. V. Discovery of a highly synergistic anthelmintic
combination that shows mutual hypersusceptibility. Proc. Natl. Acad. Sci. USA 107,
5955–5960 (2010).
24. Burns, A. R. et al. A predictive model for drug bioaccumulation and bioactivity in
Caenorhabditis elegans. Nat. Chem. Biol. 6, 549–557 (2010).
25. Ortiz de Montellano, P. R. Cytochrome P450-activated prodrugs. Future Med. Chem. 5,
213–228 (2013).
26. Leung, M. C. K., Goldstone, J. V., Boyd, W. A., Freedman, J. H. & Meyer, J. N. Caenorhabditis
elegans generates biologically relevant levels of genotoxic metabolites from aflatoxin
B1 but not benzo[a]pyrene in vivo. Toxicol. Sci. 118, 444–453 (2010).
27. Harlow, P. H., Perry, S. J., Stevens, A. J. & Flemming, A. J. Comparative metabolism of
xenobiotic chemicals by cytochrome P450s in the nematode Caenorhabditis elegans.
Sci Rep. 8, 13333 (2018).
28. Porter, T. D. New insights into the role of cytochrome P450 reductase (POR) in microsomal
redox biology. Acta Pharm. Sin. B 2, 102–106 (2012).
29. Kalgutkar, A. S. et al. Reactive metabolite trapping studies on imidazo- and
2-methylimidazo[2,1-b]thiazole-based inverse agonists of the ghrelin receptor. Drug Metab.
Dispos. 7, 1375–1388 (2013).
30. Ryan, E. et al. Evidence for the in vitro bioactivation of aminopyrazole derivatives: trapping
reactive aminopyrazole intermediates using glutathione ethyl ester in human liver
microsomes. Chem. Res. Toxicol. 28, 1747–1752 (2015).
31. Cooper, A. J. L. & Hanigan, M. H. Metabolism of glutathione S-conjugates: multiple
pathways. Compr. Toxicol. https://doi.org/10.1016/B978-0-12-801238-3.01973-5 (2018).
32. Ferguson, G. D. & Bridge, W. J. The glutathione system and the related thiol network in
Caenorhabditis elegans. Redox Biol. 24, 101171 (2019).
33. Blum, R., Meyer, K. C., Wünschmann, J., Lendzian, K. J. & Grill, E. Cytosolic action of
phytochelatin synthase. Plant Physiol. 153, 159–169 (2010).
34. Essig, Y. J., Webb, S. M. & Stürzenbaum, S. R. Deletion of phytochelatin synthase modulates
the metal accumulation pattern of cadmium exposed C. elegans. Int. J. Mol. Sci. 17, 257
(2016).
35. Giustarini, D., Milzani, A., Dalle-Donne, I., Tsikas, D. & Rossi, R. N-Acetylcysteine ethyl
ester (NACET): a novel lipophilic cell-permeable cysteine derivative with an unusual
pharmacokinetic feature and remarkable antioxidant potential. Biochem. Pharmacol. 84,
1522–1533 (2012).
36. Trudgill, D. L. & Blok, V. C. Apomictic, polyphagus root-knot nematodes: exceptionally
successful and damaging biotrophic root pathogens. Annu. Rev. Phytopathol. 39, 53–77
(2001).
37. Oka, Y. From old-generation to next-generation nematicides. Agronomy 10, 1387 (2020).
38. Devguard 500SC Label,https://za.uplonline.com/download_links/
awyXPTiWs8OT3dGCErYdw3nnA3nLnwxpCpUn7jzb.pdf (deVGen/UPL).
39. Asif, M., Rehman, B., Parihar, K., Ganai, M. A. & Siddiqui, M. A. Effect of various physico-
chemical factors on the incidence of root knot nematode Meloidogyne spp. infesting
tomato in District Aligarh (Uttar Pradesh) India. J. Plant Sci. 10, 234–243 (2015).
40. Barker, K. R., Schmitt, D. P. & Imbriani, J. L. in An Advanced Treatise on Meloidogyne
Volume II: Methodology (eds Barker, K. R. et al.) 135–148 (North Carolina State University
Department of Plant Pathology and the United States Agency for International
Development, 1985).
41. Marion, M. J., Hantz, O. & Durantel, D. in Hepatocytes: Methods and Protocols (ed. Maurel, P.)
261–272 (Humana Press, 2010).
42. Rocher, F. et al. Salicylic acid transport in Ricinus communis involves a pH-dependent
carrier system in addition to diffusion. Plant Physiol. 150, 2081–2091 (2009).
43. Wram, C. L., Hesse, C. N. & Zasada, I. A. Transcriptional response of Meloidogyne incognita
to non-fumigant nematicides. Sci Rep. 12, 9814 (2022).
44. Dara, S. K. The new integrated pest management paradigm for the modern age. J. Integr.
Pest. Manag. 10, 12 (2019).
45. Cardarelli, M., Woo, S. L., Rouphael, Y. & Colla, G. Seed treatments with microorganisms
can have a biostimulant effect by influencing germination and seedling growth of crops.
Plants 11, 259 (2022).
46. Migunova, V. D. & Sasanelli, N. Bacteria as biocontrol tool against phytoparasitic nematodes.
Plants 10, 389 (2021).
47. Slomczynska, U. et al. in Discovery and Synthesis of Crop Protection Products (eds.
Maienfisch, P. & Stevenson, T.) 129–147 (American Chemical Society, 2015).
48. Jensen, J. P., Kalwa, U., Pandey, S. & Tylka, G. L. Avicta and clariva affect the biology of
the soybean cyst nematode, Heterodera glycines. Plant Dis. 102, 2480–2486 (2018).
49. Ahmed, S., Zhou, Z., Zhou, J. & Chen, S.-Q. Pharmacogenomics of drug metabolizing
enzymes and transporters: relevance to precision medicine. Genomics Proteomics
Bioinformatics 14, 298–313 (2016).
50. Nixon, S. A. et al. Where are all the anthelmintics? Challenges and opportunities on the
path to new anthelmintics. Int. J. Parasitol. Drugs Drug Resist. 14, 8–16 (2020).
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this
article under a publishing agreement with the author(s) or other rightsholder(s); author
self-archiving of the accepted manuscript version of this article is solely governed by the
terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to Springer Nature Limited 2023
Nature | Vol 618 | 1 June 2023 | 109
Article
Methods
Free-living nematode strains and culture methods
The C. elegans WT N2 strain, the MJ69 strain harbouring the temperature-
sensitive emb-8(hc69) allele and the RB1993 strain harbouring the
cyp-35c1(ok2628) deletion allele were obtained from the C. elegans
Genetics Center (CGC, University of Minnesota). C. briggsae strain
AF16 and P. pacificus strain PS312 were both obtained from the CGC.
Rhabditophanes spp. KR3021 was obtained from Marie-Anne Félix (Insti-
tute of Biology of the Ecole Normale Supérieure). P. hermaphrodita
strain B178 was obtained from R. Rae (Liverpool John Moores University).
The following anthelmintic-resistant or nematicide-resistant C. elegans
strains were used for experiments: VC731 (unc-63(ok1075)), PR1152
(cha-1(p1152)), RP2674 (sdhc-1(tr393)), CB3474 (ben-1(e1880)), RB2119
(acr-23(ok2804)), NM1968 (slo-1(js379)), CF1038 (daf-16(mu86)), and
DA1316 (avr-14(ad1302);avr-15(vu227);glc-1(pk54)). With the exception
of RP2674, which was obtained from a genetic screen performed in our
laboratory, all of these strains were obtained from the CGC. With the
exception of C. elegans strain MJ69, all nematode species and strains
were cultured and propagated at 20 °C using standard techniques51.
C. elegans strain MJ69 was propagated at the permissible temperature
of 15 °C. We found that using solid NGM agar plates with thin bacte-
rial lawns and high nematode densities promoted growth of Rhabdi-
tophanes spp. KR3021.
Commercial chemical sources
Selectivin-A, selectivin-B, selectivin-G, selectivin-J, selectivin-K and
tioxazafen were purchased from ChemBridge. Selectivin-C was pur-
chased from Vitas-M and MilliporeSigma. Selectivin-D was purchased
from MolPort. 4-(4-Chlorophenyl)-1H-imidazole-2-thiol was purchased
from Life Chemicals. Abamectin, fenamiphos, fluopyram, iprodione,
oxamyl, spirotetramat and iodoacetamide (IAA) were purchased from
MilliporeSigma. Fluensulfone was purchased from Cayman Chemical.
C. elegans dose–response experiments
A HB101 bacterial suspension in liquid NGM was prepared by concen-
trating a saturated overnight HB101 culture twofold with liquid NGM
(see ref. 14 for the NGM recipe). Forty microlitres of this bacterial sus-
pension was added to each well of a 96-well flat-bottom culture plate,
after which approximately 25 synchronized L1 worms, in 10 µl of M9
buffer (see ref. 52 for the recipe), were added to each well. The synchro-
nized L1 worms were obtained from an embryo preparation performed
the previous day (see ref. 52 for the protocol). For the L1 assays, 0.5 µl
of chemical solution (or DMSO alone) was immediately added to the
wells using a multichannel pipette; the final DMSO concentration was
1% (v/v). The culture plate was sealed with Parafilm and placed in a box
with several wet paper towels to prevent evaporation. The worms were
incubated for 3 days at 20 °C with shaking at 200 r.p.m., and the num-
ber of viable animals was counted. A dead worm was considered any
worm that failed to move after vigorous agitation of the plate and that
appeared morphologically ‘dead’, that is, clear appearance and unre-
solved internal structures. Although the counts were performed after
3 days of incubation in the chemical, it was noted that the L1 worms
were dead within 24 h of the addition of the chemicals.
For the emb-8 knockdown dose–response assays, the MJ69 strain was
used and worms were fed HT115(DE3) Escherichia coli harbouring a RNAi
feeding vector expressing dsRNA targeting the emb-8 gene. The HT115
suspension was made by concentrating an overnight bacterial culture
grown at 20 °C, with an OD600 of 0.8, 5-fold with liquid NGM containing
1 mM IPTG and 100 µg ml–1 carbenicillin. The HT115 cells were induced
with 1 mM IPTG for 1 h at 20 °C before concentrating with NGM. Syn-
chronized L1 worms were grown for 2 days at the non-permissive tem-
perature of 25 °C until they reached young adulthood, at which point the
chemical was added to the wells using a multichannel pipette. Viability
was scored 1 day later. The WT control experiments were performed
in an identical manner; however, WT worms fed HT115(DE3) E. coli
carrying the empty RNAi vector L4440 were used. Dose–response
assays with the CYP-35C1 deletion mutant were carried out the same
way as for the WT control assays.
The dose–response experiments for the anthelmintic-resistant and
nematicide-resistant mutants were carried out as described for the L1
dose–response assays. One notable exception is the aldicarb-resistant
strain PR1152. This strain grows slowly, and so the viability counts were
performed 5 days after addition of the chemical as opposed to 3 days
to allow the DMSO control worms to reach adulthood.
At least three biological replicates were performed for each dose–
response assay. For each biological replicate, two technical replicates
were performed and the numbers of viable animals for each technical
replicate were combined (that is, about 50 worms assayed per con-
centration). The number of viable worms at each concentration was
divided by the corresponding DMSO control value to give the rela-
tive viability for each concentration. The relative viability values were
then averaged across the biological replicates. Where appropriate, the
lethal concentration 50 (LC50) values were calculated using GraphPad
Prism (v.6.0). The concentration values were log-transformed and a
four-parameter logistic curve was fitted to the dose–response data
by nonlinear regression, from which the LC50 values were extracted.
C. briggsae, P. pacificus and Rhabditophanes spp. KR3021 dose–
response experiments
Dose–response assays and quantification for C. briggsae, P. pacificus
and Rhabditophanes spp. KR3021 were carried out as described for the
C. elegans L1 dose–response assays; however, the selectivin-induced
phenotypes in P. pacificus and Rhabditophanes, even at the highest
concentrations, were a combination of lethality and larval arrest with
severe sickness. Therefore, for these dose–response assays, the num-
ber of animals that reached the L2 stage or older was quantified as
opposed to the number of viable worms. The arrested animals appeared
extremely sick and would probably die before reaching reproductive
adulthood. Therefore, this arrested phenotype was considered to be
practically analogous to death.
P. hermaphrodita dose–response experiments
In brief, 24-well solid-medium chemical assay plates were prepared
as previously described52. Each well of the assay plates contained 1 ml
of solid MYOB medium with dissolved chemical at the appropriate
concentration or DMSO alone for untreated solvent control. The final
DMSO concentration was 0.5% in every well. After thoroughly drying
the plates in a sterile flow hood, the wells were seeded with 25 µl of an
overnight culture of OP50 E. coli and allowed to sit at room temperature
for 24 h. Using glass pipettes, P. hermaphrodita B178 nematodes (mixed
stage) were washed off of several starved 10-cm culture plates with M9
buffer (see ref. 52 for the recipe) and collected in clean glass culture
tubes. Larger worms were allowed to settle to the bottom of the tubes
by gravity and young larvae (mostly L1 to L2 stage) were collected into
separate clean glass culture tubes. The nematode concentration was
adjusted to about 15 worms in 20 µl of M9 buffer. Next 20 µl of worm
suspension was added to each well of the 24-well chemical assay plates.
The M9 buffer was allowed to soak into the agar substrate and then the
plates were sealed with Parafilm and stored upside down in a Tupper-
ware box at 20 °C for 3 days. After 3 days, the number of moving worms
in each well was counted. The number of moving worms at each chemi-
cal concentration was divided by the corresponding DMSO control
value to give the relative motility for each concentration. The relative
motility values were then averaged across three biological replicates.
C. oncophora dose–response experiments
Fresh cattle faeces containing eggs of an ivermectin-resistant strain
of C. oncophora were supplied by D. Colwell and D. Gray (Lethbridge
Research Station, Agriculture and Agri-Food Canada). Established
methods were used to carry out the experimental cattle infections53,
and these methods were performed in accordance with the Canadian
Council of Animal Care regulations (licence numbers AC13-0157 (Uni-
versity of Calgary) and 1730 (Agriculture AgriFood Canada Research
Station, Lethbridge)). Cattle faeces containing C. oncophora eggs were
anaerobically stored at room temperature for a maximum of 6 days
before use. Eggs were isolated from faeces using a standard saturated
salt flotation method54 immediately before the egg hatch assay. Next
80 µl of distilled and deionized water was added to each well of a 96-well
culture plate, and then 1 µl of chemical at the appropriate concentration
in DMSO was added to each well using a multichannel pipette. Approxi-
mately 50 eggs were added per well in 20 µl of water for a final volume
of 100 µl in each well; the final DMSO concentration was 1% (v/v). The
eggs were incubated in the chemicals for 2 days at room temperature,
after which hatching was stopped by the addition of 1 µl iodine tincture
to each well. The number of hatched larvae was counted at each con-
centration, and eggs that failed to hatch were scored as dead. Relative
viability values were calculated by dividing the fraction of eggs that
hatched at each concentration by the fraction of eggs that hatched
in the corresponding DMSO control well. Three biological replicates
were performed for each dose–response experiment, and the relative
viability values were averaged across the biological replicates. The
average hatch rate for the DMSO control wells was greater than 93%
for both biological replicates.
Meloidogyne spp. in vitro dose–response experiments
M. incognita (Kofoid & White) Chitwood Race 1 (originally isolated in
Maryland) maintained on pepper (Capsicum annuum L.) cv. PA-136
and M. chitwoodi (originally isolated from Washington) maintained
on tomato (Solanum lycopersicum) were grown in a greenhouse
as previously described55,56. M. hapla (originally isolated from a muck
soil sample in Ste-Clotilde, Quebec, Canada) were maintained on
tomato (S. lycopersicum ‘Rutgers’). Infective J2 larvae were collected
as described in ref. 57. The microwell dose–response experiments were
conducted in 96-well polystyrene plates, similar to previously described
protocols55,58. In brief, each well received approximately 35–50 J2 larvae
in 10 µl of sterile distilled water (SDW), followed by the addition of 190 µl
SDW or of 190 µl of SDW containing dissolved chemical or DMSO alone.
The chemicals were tested at 5, 15 and 45 µM concentrations, and the
final concentration of DMSO in each well was 0.5% (v/v). The wells were
covered with a plastic adhesive strip, and the lids of the plates were
sealed with Parafilm. The nematodes were incubated in chemicals for
2 days at 26 °C, at which point the J2 larvae were rinsed twice with SDW
and incubated in the second SDW rinse for an additional day. After
incubation, the fraction of viable nematodes at each concentration
was calculated by dividing the number of motile nematodes by the
total number of nematodes in the well. Relative viability was calculated
by dividing the fraction of viable nematodes at a given treatment con-
centration by the fraction of viable nematodes in the untreated DMSO
control sample. The average relative viability was calculated across six
biological replicates.
WT S. cerevisiae (budding yeast) dose–response experiments
A saturated culture of the yeast strain RY0568 was diluted to an OD600
of 0.015 with fresh YPD medium (see ref. 59 for the recipe). Next 100 µl
of this dilute yeast suspension was added to each well of a 96-well plate.
The yeast was grown for 4 h at 30 °C with shaking at 140 r.p.m. Using a
multichannel pipette, 1 µl of chemical solution was added to each well
to achieve the desired final concentrations. The final DMSO concentra-
tion was 1% (v/v). The microwell plate was sealed with a clear plastic
adhesive strip and then loaded into a TECAN plate reader set at 30 °C.
National Instruments LabVIEW 2013 software (v.13.0) was used for data
acquisition. The OD600 of each well was measured every 15 min over a
48-h period, and the plate was shaken intermittently throughout the
run. The growth data were background-corrected by subtracting the
absorbance value at the first time point from all absorbance values
of the growth curve. Using GraphPad Prism (v.6.0), four-parameter
logistic curves were fitted to the growth data by nonlinear regression.
The time point at which absorbance was equal to 95% of the absorbance
maximum of the curve was determined for each untreated control, and
the AUC was calculated up to that time point for both the untreated
control and the corresponding chemical-treated samples. The AUC
for each treatment concentration was divided by the AUC for the cor-
responding untreated control, resulting in a relative growth value for
each concentration tested. At least three biological replicates were
performed for each dose–response experiment, and the relative growth
values were averaged across the replicates.
Bioinformatics identification of M. incognita CYP proteins
WormBase ParaSite60 (v.WBPS16) BioMart was used to query the
M. incognita v3 genome (NCBI BioProject identifier PREJEB8714)61
and to extract all protein sequences containing a CYP superfamily
domain (InterPro identifier IPR036396). Incomplete sequences and
those shorter than 450 amino acids in length were removed. This
resulted in a set of 31 proteins, all of which contained a CYP, E-class,
group I domain (InterPro identifier IPR002401). These sequences were
aligned using Clustal Omega62 and organized into homologous group-
ings based on protein sequence identity (Supplementary Table 3). In
total, 19 M. incognita CYP enzymes that were representative of the
diversity of sequences within the family were selected for cDNA syn-
thesis and further analysis. All 19 CYP sequences were codon-optimized
for yeast expression before synthesis (see Supplementary Table 4 for
the sequences).
CYP-expressing yeast strain construction
The C. elegans cyp-35c1 and emb-8 cDNA sequences were obtained from
WormBase63 and were codon-optimized for yeast expression (see Sup-
plementary Table 4 for the sequences). The codon-optimized cDNAs
were synthesized by Integrated DNA Technologies. Using common
molecular biology techniques, cyp-35c1 and emb-8 cDNA were cloned
into the American Type Culture Collection (ATCC) p426 GAL1 expres-
sion vector (URA3(-) rescuing marker; 2µ origin of replication) and
the ATCC p425 GAL1 expression vector (LEU2(-) rescuing marker; 2µ
origin of replication), respectively. S. cerevisiae BY4741 (MATa his3Δ1
leu2Δ0 met15Δ0 ura3Δ0) was co-transformed with the cyp-35c1 and
emb-8 expression vectors and co-transformants were selected for on
URA(-) and LEU(-) SD medium59.
The 19 M. incognita CYP-expression vectors were constructed by
Twist Bioscience using the ATCC p416 GAL1 expression vector (URA3(-)
rescuing marker; CEN origin of replication). The 19 CYP expression
vectors were individually transformed into S. cerevisiae BY4741 (MATa
his3Δ1 leu2Δ0 met15Δ0 ura3Δ0) and transformants were selected for
on URA(-) SD medium59.
Dose–response experiments with CYP-expressing yeast
Yeast SD medium supplemented with the appropriate amino acids or
nucleobases was used for yeast growth and dose–response assays59.
The CYP-35C1 and EMB-8 co-expressing strain, as well as a strain car-
rying both of the empty p425 GAL1 and p426 GAL1 expression vectors,
were grown in URA(-) and LEU(-) selection medium. The M. incognita
CYP4731A3-expressing strain, as well as a strain carrying an empty p416
GAL1 expression vector, were grown in URA(-) selection medium. The
following methods apply to the CYP-35C1 and EMB-8 co-expressing
strain, the M. incognita CYP4731A3-expressing strain and to both empty
vector (EV) strains. In brief, 5 ml of an overnight saturated culture was
grown in appropriate selection medium containing 2% raffinose at 30 °C
with shaking. The culture was diluted to an OD600 of 0.03 in appropri-
ate selection medium containing 2% raffinose. Yeast were grown for
4 h at 30 °C with shaking. To induce expression, galactose was added
to a final concentration of 2%. Next 100 µl of yeast suspension was
Article
added to each well of a 96-well culture plate. Yeast were grown for 4 h at
30 °C with shaking at 140 r.p.m. A multichannel pipette was used to add
1 µl of chemical dissolved in DMSO at the appropriate concentration
to each well (1% DMSO v/v). DMSO alone was added to the untreated
control wells. The plate was sealed with a clear plastic adhesive strip
and loaded into a TECAN plate reader set at 30 °C. The OD600 of each
well was measured every 15 min over a 48-h period, and the plate was
shaken intermittently throughout the run. Dose–response assays were
repeated at least three times (biological replicates). Data analysis was
performed in an identical manner to that described for the WT yeast
dose–response experiments (see above).
Yeast-based screen for M. incognita CYP enzymes that
bioactivate selectivin-A
Selectivin-A and selectivin-E sensitivity assays with the 19 M. incognita
CYP-expressing yeast strains were carried out as described for the
dose–response experiments with CYP-expressing yeast (see above),
with the following two exceptions: (1) overnight cultures were diluted
to 0.05 instead of 0.03 and grown for 2 h instead of 4 h before adding
galactose; (2) after the addition of galactose, yeast cells were grown
for 2 h instead of 4 h before addition of chemical. A strain carrying the
empty p416 GAL1 expression vector was included as the EV control. Each
strain was treated with 50 µM selectivin-A, 50 µM selectivin-E or DMSO
alone as an untreated control (1% v/v). All experiments were repeated six
times (biological replicates). Data analysis was performed in an identi-
cal manner to that described for the WT yeast dose–response experi-
ments, and relative growth values for each strain and each condition
were calculated the same way. Average relative growth values for each
strain were normalized to the EV control, and then log2-transformed to
give log2 fold difference in relative growth. For each strain, a P value was
obtained from an unpaired two-tailed Student’s t-test comparing the
mean relative growth values between the CYP-expressing strain and the
EV control. The P values were log10-transformed. The log-transformed
data were plotted as a volcano plot. A P value less than 0.01 indicates
a significant difference.
C. albicans dose–response experiments
Compound potency against C. albicans (SN95) was assessed using two-
fold dose–response assays following a standard protocol. In brief, YPD
medium was inoculated with about 1 × 103 cells per ml from saturated
overnight cultures. Assays were performed in 384-well, flat-bottom
microtitre plates (Corning) in a final volume of 40 µl per well. Each drug
was added to wells in a twofold concentration gradient from 100 µM to
0 µM. Plates were then incubated in the dark at 30 °C under static condi-
tions for 48 h. End point growth was quantified by measuring the OD600
using a Molecular Devices SpectraMax Plus 384 Microplate Reader with
SoftMax Pro 7 software, and the results were corrected for background
medium. Relative fungal growth for each compound treatment was
defined by normalizing to the levels observed in untreated controls.
All dose–response assays were performed in technical triplicate and
relative growth levels were averaged across replicates.
P. simiae dose–response experiments
P. simiae WSC417 was used for all experiments. Liquid LB medium was
inoculated with around 1 × 105 cells per ml from saturated overnight
cultures. Assays were performed in 96-well flat-bottom microtitre
plates in a final volume of 100 µl per well. Using a multichannel pipette,
the chemical was added to each well at the appropriate concentration
in 0.5 µl of DMSO for a final DMSO concentration of 0.5% v/v. DMSO
alone was added to the untreated control wells. The OD600 of each well
was measured using a BMG Labtech POLARStar Omega spectropho-
tometer with Omega software (v.1.20), and the plates were then incu-
bated at 28 °C with shaking at 200 r.p.m. for 16 h. End point growth
was quantified by measuring OD600 after the 16 h of incubation and
then background-corrected by subtracting the corresponding OD600
values from the 0-h time point. The relative growth for each chemical
treatment was defined by normalizing the corrected OD600 value to
that of the corresponding untreated DMSO control. All dose–response
assays were performed in technical triplicate and relative growth levels
were averaged across replicates.
D. rerio (zebrafish) dose–response experiments
Fish were maintained at 28.5 °C on a 14–10-h light–dark cycle and staged
according to hours post fertilization (h.p.f.). For each biological rep-
licate, eggs from AB WT fish were collected at 4 h.p.f. At 3 days post
fertilization (d.p.f.), both male and female embryos were arrayed in
24-well culture plates at 10 per well, which is a commonly used sample
size for this type of dose–response experiment. Next 5 µl of the chemi-
cal dissolved in DMSO at the appropriate concentration was added
to 1 ml of E3 medium and then intensively vortexed to mix. Water was
removed from the embryos in the wells and 1 ml of chemical-treated
water was transferred to each of the wells. The DMSO control wells
contained DMSO alone. The final DMSO concentration in every well
was 0.5% (v/v). The culture plates were sealed with Parafilm, wrapped
in aluminium foil and incubated at 28.5 °C for 72 h. After incubation,
viability was assessed. Using a stereomicroscope, lethality was scored
either by appearance of whole-body necrosis or the absence of both a
heartbeat and touch-evoked response. Relative viability was calculated
by dividing the number of viable embryos in the treatment wells by the
number of viable embryos in the DMSO control well. Three biological
replicates were performed for each dose–response experiment, and
the relative viability values were averaged across the three replicates.
The zebrafish ethics protocol is 100005273, approved by the Animal
Care Committee at The Hospital for Sick Children, Canada. Fish were
chosen at random for inclusion in the dose–response assays. The assays
were blinded such that the experimenter did not have any knowledge
of the compounds being tested.
HEK293 cell dose–response experiments
HEK293 cells (Thermo Fisher, R71007) were seeded into 96-well plates
at 5,000 cells per well in 100 µl total volumes of DMEM, 10% FBS and
1% penicillin–streptomycin medium and grown overnight at 37 °C in
the presence of 5% CO2. Compounds (0.5-µl volumes from appropriate
source plates) were then added to cells, and growth was continued for
an additional 48 h. Following growth, 10 µl of CellTiter-Blue viability
reagent (Promega) was added to each well, and plates were incubated
for an additional 4 h at 37 °C in the presence of 5% CO2. Fluorescence
measurements (560 nm excitation/590 nm emission) were then per-
formed using a CLARIOstar plate reader (BMG Labtech) to quantify
reagent reduction and to estimate cell viability. Fluorescence measure-
ments were corrected for background from medium. Relative growth
was calculated by dividing the corrected fluorescence in the treatment
wells by that measured in the corresponding DMSO control well. The
final values are an average of at least three technical replicates. The
HEK293 cells were authenticated by STR analysis at the Toronto Hospital
for Sick Children authentication facility, and recent testing confirmed
that the cell line was negative for mycoplasma contamination.
HepaRG cell dose–response experiments
HepaRG cells (Thermo Fisher, Gibco, cryopreserved HPRGC10) were
seeded in a 96-well plate at 20,000 cells per well in 100 µl of DMEM-F12
(Gibco) with 10% FBS (Invitrogen), 1× penicillin–streptomycin (Sigma),
2 mM l-glutamine (Sigma), 1× ITS-A (Gibco) and 40 ng ml–1 dexametha-
sone (BioShop). Subsequently, the medium was refreshed, and a dilu-
tion series of the test compound was added to each well with a final
total volume of 100 µl. After 48 h, CellTiter-Blue viability dye (Promega)
was added for 3 h, and viability was quantified using plate reader as
per the manufacturer’s protocol (Promega). Fluorescence was meas-
ured at 560 nm excitation/590 nm emission and corrected for back-
ground from the medium. Relative growth was calculated by dividing
the corrected fluorescence in the treatment wells by that measured in
the corresponding DMSO control well. The final values are an average
of at least three technical replicates. The HepaRG cells were authenti-
cated by SNP genotyping and confirmed to be negative for mycoplasma
contamination by the commercial supplier. The HepaRG cells were
used directly from the supplier.
D. melanogaster (fruit fly) dose–response experiments
Fly food in agar substrate was prepared by mixing together 100 ml of
unsulfured molasses, 100 ml of cornmeal, 41.2 g of baker’s yeast and
14.8 g of agar into 1,400 ml of distilled deionized water and boiling
for 30 min. The medium was allowed to cool to 56 °C, at which point
5 ml was added by syringe to plastic cylindrical fly vials. Next 10 µl of
chemical, or DMSO alone, was added to the medium in each vial. The
chemicals were mixed into the medium by mechanical mixing using
a pipette. The final DMSO concentration was 0.2% (v/v). The medium
was allowed to solidify at room temperature (about 22 °C) overnight.
The following day (day 0), 8 pairs of male and female w1118 flies were
added to each vial so that there were 16 flies in total per vial. The vials
were stored at room temperature for 7 days, at which point the number
of motile flies was counted. Fly motility was scored as any observable
movement after the vial had been vigorously jostled. Relative motility
was calculated by dividing the number of motile flies in the treatment
vials by the average number of motile flies in two DMSO control vials.
On day 8, the 16 parental flies were removed from the vials and the
progeny larvae were allowed to continue to grow and hatch into adult
flies. To assess larval viability, hatched flies were counted and discarded
on days 10, 12, 14, 16, 18 and 20. The counts were summed. Relative
viability was calculated by dividing the number of hatched flies in the
treatment vials by the average number of hatched flies in the two DMSO
control vials. The final relative motility and relative viability values are
an average across three biological replicates.
A. thaliana greening experiments
Greening experiments were performed with A. thaliana seeds of WT
Col-0; seeds were surface-sterilized in bleach and plated onto 0.5× MS,
0.5% sucrose agar medium supplemented with compounds of interest
at 5, 15 and 45 µM concentrations (0.2% DMSO (v/v/)). After 4 days of
stratification at 4 °C, plates were transferred to a growth chamber
(16 h–8 h, 150 µE m–2) and greening was recorded after 4 days. Pictures
were recorded by camera (SONY a7s) with a FE1.8/55 lens (FE 55 mm
F1.8 ZA; SEL55F18Z). Three biological replicates were performed for
each treatment.
Selectivin-A mouse studies
Female C57Bl/6 mice aged 6–8 weeks (bred and maintained at the
animal care facility, Department of Biological Sciences, University of
Calgary, Canada) were used. All animal experiments were approved
by the University of Calgary’s Life and Environmental Sciences Ani-
mal Care Committee (protocol AC17-0082). All protocols for animal
use and euthanasia were in accordance with the Canadian Council for
Animal Care (Canada). Infected mice were orally gavaged with 200
third stage Heligmosomoides polygyrus larvae (maintained in-house.
Original stock was a gift from A. Shostak, University of Alberta, Canada)
and euthanized on day 22 after infection. Each group (treated versus
non-treated) had a minimum of 7 mice. Experiments of this type
commonly use five mice per group, and we chose a seven-per-group
minimum to buffer against variability in the effects of drug treatment.
Mice were littermates. Mice were treated orally with 5 daily doses of
selectivin-A (50 mg kg–1 resuspended in DMSO). Control mice were
given DMSO only as a control. Mice were chosen at random for inclu-
sion in either the treatment or control groups. To reduce human error
with drug delivery, all mice in one cage (from two to five) were on the
same treatment. There were two cages per treatment group. Mice were
monitored daily for visible signs of treatment-induced pathology. The
study was blinded. One experimenter was responsible for the drug
treatments and a second experimenter was responsible for assessing
any pathologies without knowing which mice were in the treatment
or control group.
Incubations for HPLC analysis
For the C. elegans and C. briggsae L1 incubations, synchronized hatch-
lings were obtained from an embryo preparation of gravid adults52.
In total, 200,000 hatchlings in 500 μl of M9 buffer (see ref. 52 for
the M9 buffer recipe) were treated with 100 μM selectivin-A, 100 µM
4-(4-chlorophenyl)-1H-imidazole-2-thiol (that is, imidazole-thiol metab-
olite (ITM)), 100 µM selectivin-E or DMSO alone for control purposes.
The final concentration of DMSO in all samples was 1% (v/v). Before
the incubations, the hatchlings used for the dead worm controls were
heat-killed at 95 °C for 45 min. The incubations were carried out in
standard 1.5 ml microcentrifuge tubes on a nutating shaker at 20 °C for
6 h. After 6 h of incubation, the worms were transferred to the wells of
a Pall AcroPrep 96-well filter plate (0.45 μm wwPTFE membrane, 1 ml
well volume), the buffer was drained from the wells by vacuum and
the worms were subsequently washed once with 600 μl of M9 buffer.
After washing, the worms were re-suspended in 50 μl of M9 buffer
using low-binding tips, transferred to a new 1.5-ml microcentrifuge
tube and stored frozen at –80 °C. In some cases, the incubation buffer
was also collected and frozen at –80 °C. For the samples alkylated
with IAA, worms were resuspended in 50 µl of M9 buffer containing
50 mM IAA and allowed to incubate for 40 min at room temperature
before freezing at –80 °C. The P. pacificus L1 incubations were carried
out using the same methods as above; however, 100,000 hatchlings
were used. The Rhabditophanes incubations were carried out in a
similar way; however, a mixed population of 50,000 L1 and L2 larvae
obtained through gravity separation of adults and larvae were used for
incubations. The M. incognita incubations were carried out the same
way but with three differences: (1) 25,000 J2 larvae were used; (2) incu-
bations and washes were performed using distilled water instead of M9
buffer; (3) after incubation and washing, worms were resuspended in
25 µl of M9 buffer. M. incognita culture and J2 collection were done as
previously described64.
For the WT and cyp-35c1Δ mutant C. elegans young adult incuba-
tions, a culture of HT115 bacteria harbouring the empty L4440 RNAi
feeding vector was grown to an OD600 of 0.6 in LB medium supple-
mented with 100 µg per ml ampicillin, and then induced with 1 mM
IPTG for 1 h at 20 °C until the culture reached an OD600 of about 0.8.
After induction, the culture was concentrated 5-fold with liquid NGM
supplemented with 1 mM IPTG and 100 μg ml–1 carbenicillin (see ref. 14
for the NGM recipe). Next 2 ml of this bacterial suspension was added
to a 15-ml conical tube. For the treated samples, selectivin-A was added
to the suspension to a final concentration of 100 µM and 1% DMSO
(v/v). For the untreated control samples, DMSO was added alone to a
1% (v/v) final concentration. Next 1,250 L1 hatchlings, obtained from
an embryo preparation of gravid adults52, were added to each tube in
0.5 ml of M9 buffer. The worms were incubated on a nutating shaker
for 2 days at 25 °C until they reached the young adult stage, at which
point selectivin-A was added to a final concentration of 100 µM and
1% DMSO (v/v). DMSO alone was added to the untreated control sam-
ples. The worms were incubated for 6 h at 25 °C. After incubation, the
worms were pelleted by centrifugation and washed three times with
M9 buffer to remove the bacteria. After the final wash, the buffer was
aspirated down to 0.5 ml, and using low-binding tips, the worms were
transferred to the wells of a Pall AcroPrep 96-well filter plate (0.45 μm
wwPTFE membrane, 1 ml well volume). The buffer was drained from the
wells by vacuum. The worms were re-suspended in 50 μl of M9 buffer
using low-binding tips, transferred to a new 1.5-ml microcentrifuge
tube and stored frozen at –80 °C. The emb-8 knockdown incubations
were performed in an identical manner to the WT young adult dose–
response experiments, except the MJ69 strain was used in place of the
Article
WT strain, and worms were fed HT115(DE3) E. coli harbouring an RNAi
feeding vector expressing dsRNA targeting the emb-8 gene.
For the WT S. cerevisiae incubations, an overnight culture of the yeast
strain RY0568 was diluted to an OD600 of 0.1 in 5 ml of YPD medium and
grown at 30 °C for 3 h until the OD600 reached 0.3 (see ref. 59 for YPD
recipe). This culture was diluted to an OD600 of 0.2 in 2.5 ml of YPD and
selectivin-A was added at a concentration of 100 µM (1% DMSO (v/v)).
DMSO was added alone to the untreated control samples. The culture
was incubated for 6 h at 30 °C until the OD600 reached 3.2. The cells were
transferred to a 15-ml conical tube and pelleted by centrifugation at
1,000g for 5 min. Next 2 ml of the medium was collected and frozen at
–80 °C, leaving the cells in 0.5 ml of medium. The cells were washed
twice with distilled and deionized water. After the second wash and
spin, all of the water was removed and the cells were re-suspended
in 0.5 ml of M9 buffer. The cells were pelleted again, and all of the M9
buffer was removed by aspiration. The cell pellet was frozen at –80 °C.
For the CYP-expressing (and EV control) yeast strains, saturated cul-
tures were grown in the appropriate SD selection medium containing 2%
raffinose. The saturated cultures were diluted to an OD600 of 0.3 in the
appropriate SD selection medium containing 2% raffinose to achieve a
4.95 ml final volume. The yeast was grown in glass test tubes for 4 h at
30 °C on a rotating wheel. Next 550 µl of 20% galactose was added to
the culture to induce protein expression, and cells were grown for 4 h
at 30 °C on a rotating wheel. After that, 55 µl of 10 mM selectivin-A in
DMSO, or DMSO alone, was added to the cultures. The cultures were
gently vortexed and then incubated at 30 °C for 8 h on a rotating wheel.
The cells were transferred to 15-ml conical tubes and pelleted at 1,000g
for 5 min. Then 5 ml of incubation buffer was removed and the cells were
transferred to 1.5-ml microcentrifuge tubes. The tubes were centrifuged
at 1,000g for 2 min, and all of the medium was aspirated. The cell pellet
was resuspended in 0.5 ml of ddH2O. The cells were pelleted again at
1,000g for 2 min, the medium was aspirated and the cells were resus-
pended in 0.5 ml of M9 buffer. The cells were pelleted again, all of the
medium was removed and the cell pellet was stored frozen at –80 °C.
For the zebrafish incubations, 50 AB WT zebrafish aged 1 d.p.f. were
transferred into 0.5 ml of E3 medium in 24-well culture plates and
allowed to grow to 5 d.p.f. At 5 d.p.f., they were transferred into fresh
0.5 ml E3 medium in 1.5-ml microcentrifuge tubes and either 100 µM
selectivin-A or DMSO alone was added (1% DMSO (v/v)), as well as 0.05%
gelatin to break the surface tension of the water. The fish were incu-
bated for 6 h at 20 °C on a nutating shaker. After 6 h of incubation, the
fish were transferred to the wells of a Pall AcroPrep 96-well filter plate
(0.45 μm wwPTFE membrane, 1 ml well volume) using low-binding
tips with the ends cut to allow the fish to pass through the opening.
The buffer was drained from the wells by vacuum, and the fish were
subsequently washed once with 600 μl of M9 buffer. After washing,
the fish were re-suspended in 100 μl of M9 buffer using low-binding
tips with the ends cut to allow the fish to pass through the opening,
and transferred to a new 1.5-ml microcentrifuge tube. The fish were
quickly pelleted by centrifugation and 50 µl of M9 buffer was removed.
The fish were then frozen at –80 °C. The incubation buffer was also
collected and frozen at –80 °C.
For the fly larvae incubations, three third instar w1118 fly larvae were
transferred into 0.5 ml of PBS buffer in 1.5 ml microcentrifuge tubes.
Either 100 µM selectivin-A or DMSO alone was added (1% DMSO (v/v)).
The fly larvae were incubated for 6 h at 20 °C on a nutating shaker.
After 6 h of incubation, the larvae were transferred to the wells of a Pall
AcroPrep 96-well filter plate (0.45 µm wwPTFE membrane, 1 ml well
volume) using low-binding tips with the ends cut to allow the larvae
to pass through the opening. The buffer was drained from the wells by
vacuum, and the larvae were subsequently washed once with 600 μl
of M9 buffer. After washing, the larvae were transferred into 100 µl of
M9 buffer using a fine hair brush. The volume of M9 was adjusted to
50 µl, and the larvae were frozen at –80 °C. The incubation buffer was
also collected and frozen at –80 °C.
A 100 µM concentration was chosen for the incubations because
it is the highest concentration that remains soluble in the buffer and
therefore maximizes the signal-to-noise ration. A 6-h incubation period
was chosen because C. elegans nematodes are obviously affected by
selectivin-A at this time point but they are not yet dead (by 12 h they are
dead). The incubations were performed in at least biological triplicates.
Incubation buffer de-salting and drying
Worm, fish and fly incubation buffers were de-salted using a 96-well
HyperSep C8 solid phase extraction (SPE) plate with 1 ml well volume
and 100 mg bed weight (Thermo Scientific). The SPE plates were cou-
pled to a vacuum manifold and vacuum pump, and a flow rate of around
1 ml min–1 was used. The columns were activated with 3 ml of 100%
acetonitrile and then washed with 3 ml of distilled deionized water.
Incubation buffers were added to each well and passed through the
columns by vacuum. The columns were washed once with 1 ml of water.
Three sequential elutions were done using 250 µl of 20% acetonitrile,
250 µl of 50% acetonitrile and 250 µl of 100% acetonitrile. The eluates
were dried in an Eppendorf Vacufuge concentrator and then stored
frozen at –80 °C for downstream HPLC analysis.
Yeast incubation buffers were de-salted using Sep-Pak light C8 car-
tridges (Waters). The columns were activated, loaded, washed and
eluted in the same manner described above for the SPE plates; however,
solvent and sample was passed through the cartridges using a syringe
and manual force. A flow rate of about 1 ml min–1 was used. The eluates
were dried and stored frozen at –80 °C for downstream HPLC analysis.
Sample lysis for HPLC analysis
Free-living worm and fish lysis. Frozen worm and fish samples were
thawed, and 50 µl of 2× lysis buffer (20 mM Tris-HCl pH 8.3, 0.2% SDS
and 240 µg ml–1 proteinase K) was added to the tubes. The tubes were
incubated for 80 min in a 56 °C water bath with vigorous vortexing
every 15 min. The samples were then bath sonicated for 20 min at room
temperature using a Branson 1510 bath sonicator. The lysates were
stored frozen at –80 °C until processing by HPLC.
M. incognita J2 lysis. J2 larvae were thawed, and 25 µl of 2× lysis buffer
(20 mM Tris-HCl pH 8.3, 0.2% SDS and 1,600 µg ml–1 proteinase K) was
added to the tubes. The tubes were incubated for 3 h in a 65 °C water
bath with vigorous vortexing every 20 min. The samples were then
bath sonicated for 30 min using a Branson 1510 bath sonicator with
the heater turned on. The lysates were stored frozen at –80 °C until
processing by HPLC.
WT yeast lysis. Frozen yeast pellets were thawed and re-suspended in
40 µl of M9 buffer containing 1 M sorbitol and 300 U ml–1 zymolase. The
samples were incubated at 37 °C for 60 min with vortexing every 10 min,
after which 50 µl of 2× yeast lysis buffer (20 mM Tris-HCl pH 8.3, 0.2%
SDS and 720 µg ml–1 proteinase K) was added to the tubes. The tubes
were incubated for 2 h at 56 °C with vigorous vortexing every 10 min.
The samples were then bath sonicated for 20 min at room temperature
using a Branson 1510 bath sonicator. The lysates were stored frozen at
–80 °C until processing by HPLC.
CYP-expressing yeast lysis. Yeast cell pellets were thawed. With the
exception of CYP-35C1-expressing yeast incubated in selectivin-A,
cells were resuspended in 70 µl of M9 buffer containing 1 M sorbitol
and 300 U ml–1 zymolase. CYP-35C1-expressing cells incubated in
selectivin-A were resuspended in 42 µl of the M9, sorbitol and zymo-
lase solution (the lower volume was to account for the lower biomass
resulting from growth inhibition). The cells were then incubated at 37 °C
for 1 h with vortexing every 10 min. Next 2× lysis buffer (20 mM Tris-HCl
pH 8.3, 0.2% SDS and 720 µg ml–1 proteinase K) was added to the tubes
(50 µl for CYP-351-expressing cells incubated in selectivin-A and 100 µl
for all other cells), and the cells were incubated for 2 h with vortexing
every 10 min. The samples were then bath sonicated for 20 min at room
temperature using a Branson 1510 bath sonicator. The lysates were
stored frozen at –80 °C until processing by HPLC.
Fly larvae lysis. Frozen larvae were thawed, and 50 µl of fly lysis buffer
(20 mM Tris-HCl pH 8.3, 0.2% SDS and 720 µg ml–1 proteinase K) was
added to the tubes. The larvae were incubated for 30 min at 56 °C and
were then crushed with a pestle to promote lysis (the exoskeleton
remained intact). The samples were incubated at 56 °C for an additional
1.5 h with vigorous vortexing every 10 min. The samples were then bath
sonicated for 20 min at room temperature using a Branson 1510 bath
sonicator. The lysates were stored frozen at –80 °C until processing
by HPLC.
Dried incubation buffer preparation. Dried incubation buffers were
lysed in the same way as the free-living worm pellets; however, 50 µl of
M9 buffer was added to the dried buffer before proceeding through
the lysis steps.
HPLC method
The frozen lysates were thawed, and 50 µl of acidified acetonitrile solu-
tion (50% acetonitrile, 0.2% acetic acid) was added to the thawed lysates.
For the fly samples, three separate lysates were pooled for each repli-
cate to minimize variability associated with having only three larvae
per sample. The samples were mixed by vortexing for approximately
10 s and then centrifuged at 17,949g for 1 min. After centrifugation,
50 μl of lysate was injected onto a 4.6 × 150 mm Zorbax SB-C8 column
(5 µm particle size) and eluted over 8.65 min with the solvent and flow
rate gradients shown in Extended Data Table 4. UV-vis absorbance was
measured every 2 nm between 190 and 602 nm. HPLC was performed
using a HP 1050 system equipped with HP Chemstation for LC 3D soft-
ware (v.A.10.02), an autosampler, a vacuum degasser and a variable
wavelength DAD. The column was maintained at around 22 °C.
HPLC-DAD quantification
Raw absorbance data were exported to .csv files using HP Chemsta-
tion for LC 3D (v.A.10.02), and Microsoft Excel for Mac (v.16.16.27) was
used for all chromatogram normalizations, background corrections
and AUC determinations. The dry weight of biomaterial in each of the
yeast, worm and fish samples was determined by calculating the AUC
at 300 nm for the major peak of endogenous content, which eluted
between 1 and 1.5 min, and then deriving the dry weight in mg using
standard curves of AUC300 versus dry weight generated using known
dry weights for each of the test organisms. The dry weight of fly larvae
biomaterial was derived in the same way; however, AUC220 was used
instead of AUC300 as it scaled more linearly. To control for differences in
biomass, the absorbance intensities of the raw chromatograms, in mil-
liabsorbance units (mAU), were divided by the dry weight of the sample
resulting in mAU per mg-of-dry-weight values. Every selectivin-A and
ITM-treated sample was paired with an untreated DMSO control sample
prepared and processed at the same time and in the same way. To correct
for background absorbance from endogenous material, the untreated
DMSO control mAU per mg-of-dry-weight values were subtracted from
the corresponding treated sample mAU per mg-of-dry-weight values
to give background-corrected chromatograms. The heat-mapped
chromatograms shown in the main figures are these normalized and
corrected chromatograms. The heat-mapped chromatograms were
generated using custom Python scripts. The AUC for the analyte peaks
was computed using the normalized and corrected chromatograms,
and an average across at least three biological replicates was calcu-
lated. A peak was considered not detectable if an absorbance intensity
maximum could not be found at the expected retention time. If a peak
was not detectable, it was given an AUC value of zero. AUCs for M1, M2,
M3 and ITM-alk were calculated at 260 nm, whereas the AUC for M4
was calculated at 300 nm. The same quantification scheme was used
for the incubation buffer samples, with two minor exceptions for the
yeast incubation buffers. (1) A YPD blank sample prepared in the same
way as the experimental yeast buffers was processed by HPLC, and the
absorbance intensity values were subtracted from the experimental
intensities to account for absorbing material from the YPD medium.
This was done before dry weight normalization and subtraction of
the paired DMSO control absorbance intensities. (2) As only 2 out of
2.5 ml (4/5) of yeast incubation buffer was processed, the dry weight
values used to normalize the yeast buffer data were multiplied by 4/5.
Average AUC values were calculated across at least three experimental
replicates.
Metabolite MS
HPLC fractions containing either the selectivin-A metabolites M1, M2,
M3 and M4 and ITM-alk, or the selectivin-E metabolites E-M1, E-M2 and
E-M3 were collected from three separate lysates, combined and dried
using a SpeedVac concentrator. The identical fractions from DMSO
control lysates were also collected and dried. The dried fractions were
re-suspended in a minimal volume of 1:1 (v/v) methanol:0.1% aqueous
formic acid. Electrospray ionization MS (ESI-MS) analyses were carried
out using a 6538 UHD model quadrupole time-of-flight mass analyser
equipped with an atmospheric pressure ESI source and a 1260 Infinity
model HPLC system (Agilent Technologies). Samples were analysed by
loop injection with a mobile phase composed of 1:1 (v/v) methanol:0.1%
aqueous formic acid and a flow rate of 0.25 ml min–1. Mass spectra were
recorded in the 2 GHz mode, and the high-resolution MS analyses for
molecular formula determinations were obtained using external cali-
bration. Tandem MS/MS analyses were obtained by collision-induced
dissociation using the targeted MSn function of the acquisition soft-
ware. MS/MS spectra were recorded sequentially at three different
fragmentation voltages (10, 20 and 30 V), and the resulting spectrum
was composed of the average of those three collision energies. Agilent
MassHunter Data Acquisition software (v.B.05.01) was used for MS data
acquisition. Primary data analysis and export were done using Agilent
MassHunter Qualitative Analysis software (v.B.06.00 SP1). Downstream
analysis and MS plot generation (using centroid data) were done using
Microsoft Excel for Mac (v.16.16.27).
Metabolite M2 purification and characterization by NMR
HB101 E. coli cells from an 8-litre saturated culture were pelleted by
centrifugation and resuspended in 400 ml of liquid S-medium (see
ref. 51 for the recipe). The bacterial suspension was transferred to a
1-litre glass Erlenmeyer flask. Next, 6 million synchronized C. elegans L1
larvae were added to the bacterial suspension in 12 ml of M9 buffer (see
ref. 52 for the recipe). The synchronized L1 larvae were obtained from
a large-scale embryo preparation of gravid adults performed the day
before (see ref. 52 for the protocol). Worms were incubated at 20 °C on
an orbital shaker at 200 r.p.m. for 3.5 days until they reached adulthood.
Worms were then washed 7 times with M9 buffer, resuspended in 300 ml
of M9 buffer and transferred to a 500 ml glass Erlenmeyer flask. The
worms were allowed to settle to the bottom of the flask (about 20 min
at room temperature), and 3 ml of 10 mM selectivin-A was added to the
worm suspension (note that the selectivin-A used for this incubation
was synthesized in-house). The flask was shaken moderately by hand
for 1 min to fully dissolve selectivin-A. The worms were then incubated
at 20 °C with shaking at 200 r.p.m. for 6.5 h. The worms were allowed
to settle on ice for 30 min. Using a glass pipette, 250 ml of incubation
buffer was collected and distributed to 6 chilled 50-ml plastic conical
tubes (note that there were still many worms in the incubation buffer
at this stage). The tubes were centrifuged for 2 min at 800g and then
placed on ice for 30 min to allow the worms to completely settle to the
bottom of the tubes. Using a glass pipette, 40 ml of incubation buffer
was then collected from each tube, being careful not to aspirate any
worms, and combined into a clean 1-litre glass bottle. The incubation
buffer was frozen at −80 °C.
Article
Sep-Pak C8 35CC cartridges with 10 g sorbent (Waters) were used
to de-salt and concentrate the incubation buffer. Five separate car-
tridges were each activated with 50 ml 100% acetonitrile (flow rate of
about 2 ml min–1). The cartridges were then washed with 50 ml ddH2O.
Next, 48 ml of thawed incubation buffer was loaded onto each col-
umn. Columns were washed with 50 ml ddH2O. Columns were eluted
sequentially with 10 ml 20% acetonitrile, 10 ml 50% acetonitrile and
10 ml 100% acetonitrile. The eluates from each column were collected
in 5 × 100-ml glass bottles. The 150 ml of eluate was distributed evenly
across 24 × 20-ml scintillation vials and stored at 4 °C overnight. The
next day, the eluates were dried using a Genevac centrifugal evaporator
(about 3 h at 55 °C). The dried eluate from the 24 vials was dissolved in
2.5 ml of a 5% acetonitrile and 0.1% acetic acid solution (about 1.5 ml of
dissolved eluate remained after dissolution and transfers). The com-
ponents of the eluate solution were separated using a HP 1050 HPLC
system equipped with an autosampler, vacuum degasser and variable
wavelength DAD, and the fraction containing metabolite M2 was col-
lected. For each HPLC run, 50 μl of the eluate solution was injected onto
a 4.6 × 150 mm Zorbax SB-C8 column (5 µm particle size) and eluted
over 30 min with the solvent and flow rate gradients shown in Extended
Data Table 4. UV-vis absorbance was measured every 2 nm between 190
and 602 nm. The column was maintained at about 22 °C. Thirty separate
HPLC runs were required to process all of the eluate solution. The M2
metabolite eluted between 20.3 and 21 min. M2 metabolite fractions
were collected and split into two pre-weighed 20 ml glass scintillation
vials. The combined fractions were dried using a Genevac centrifu-
gal evaporator (about 4 h at 40 °C). Before Genevac drying, a small
amount of the M2 fraction was taken for purity analysis by HPLC-DAD.
Purity was estimated by dividing area under the M2 curve at 250 nm
by the total peak area across the run. The purity of M2 was 95.4%. The
yield of M2 was estimated to be 2 mg by subtracting the weight of the
empty vials from the weight of the vials containing dried metabolite
and multiplying by 0.954.
NMR data for purified M2 was obtained at 293 K on a Varian Mer-
cury 300 MHz, Varian Mercury 400 MHz, Bruker Advance III 400 MHz,
Agilent DD2 500 MHz equipped with a 5 mm Xses cold probe, Agilent
DD2 600 MHz or Agilent DD2 700 MHz. 1H spectra were referenced to
the residual solvent signal (CDCl3 = 7.26 ppm, DMSO-d6 = 2.50 ppm,
CD3OD = 3.31 ppm). 13C{1H} spectra were referenced to the residual sol-
vent signal (CDCl3 = 77.16 ppm, DMSO-d6 = 39.52 ppm, CD3OD = 49.00).
Data for 1H-NMR are reported as follows: chemical shift (δ ppm), mul-
tiplicity (s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet)
and coupling constant (Hz), integration. Agilent VnmrJ software (v.4.2)
was used for data acquisition, and MestreNova software (v.14.3.1) was
used for data processing. All NMR spectra were Fourier transformed
with exponential line broadening, phased and baseline corrected.
NACET experiments
Saturated HB101 bacterial cultures were concentrated twofold with
either regular liquid NGM or liquid NGM containing 2.5 mM NACET. In
the wells of a 96-well culture plate, 25 synchronized L1 worms in 10 µl
of M9 buffer were added to 40 µl of the bacterial suspensions. The
synchronized L1 worms were obtained from an embryo preparation
performed the previous day (see ref. 52 for the protocol). The culture
plate was sealed with Parafilm and placed in a box with several wet
paper towels. The worms were incubated for 4 h at 20 °C, after which
100 µM of nematicide, or DMSO alone, was added independently to
NACET-containing or NACET-free wells. The plates were re-sealed with
Parafilm, replaced in the box with wet paper towels and allowed to incu-
bate for an additional 24 h at 20 °C before assessing lethality. Lethality
was assessed as described above for the C. elegans L1 dose–response
assays. A concentration of 2.5 mM NACET was chosen because it is
the highest concentration of NACET that can be used without sub-
stantially slowing worm growth or killing the worms. Two technical
replicates were performed for each condition, and the number of living
worms was summed across the two replicates. The number of viable
worms for each condition was divided by the corresponding DMSO
control value to give the relative viability for each condition. This
was repeated five separate times to give five independent biological
replicates.
C. elegans cyp RNAi screen
RNAi-mediated knockdown of 60 out of 76 C. elegans microsomal CYP
enzymes was performed in 96-well culture plates using methods identi-
cal to those described for the emb-8 RNAi dose–response assays, but
with the following three modifications: (1) WT worms were used instead
of the MJ69 strain; (2) synchronized L1 larvae were grown for 2.5 days at
20 °C instead of 2 days at 25 °C; (3) a single selectivin-A concentration
of 25 µM was used for the screen. Each well had bacteria expressing
a single dsRNA targeting a single cyp, and around 25 L1 larvae were
added to each well. After the worms had fed on RNAi-expressing bac-
teria for 2.5 days and had reached adulthood, 25 µM selectivin-A (1%
DMSO (v/v)) was added to each well. After 24 h of chemical exposure,
the percentage of worms in the well that were moving was quantified.
Each cyp RNAi knockdown was tested in technical duplicate and an
average per cent-moving value was calculated. Fifty control repli-
cates using bacteria carrying the empty L4440 RNAi vector were also
included for control purposes. Using the mean and standard deviation
from the control replicates, Z-scores were calculated for the average
per cent-moving values for each cyp knockdown. A Z-score greater than
three was considered to be significant. The raw data for the C. elegans
cyp RNAi knockdown screen can be found in Supplementary Table 2.
Attempted oxidation of selectivin-A
In an attempt to generate the sulfoxide metabolite of selectivin-A in the
laboratory, we tried three different oxidation methods: (1) hydrogen
peroxide as oxidant: H2O2 (1 equiv.), acetone/H2O (5:1) 0 °C to room
temperature; (2) oxone as oxidant: oxone (5 equiv.), H2O/methanol,
room temperature; (3) m-CPBA as oxidant: m-CPBA (2.5 equiv.), dichlo-
romethane, room temperature. Using hydrogen peroxide as the oxidant
resulted in the recovery of starting material alone (that is, selectivin-A),
and cleavage of the imidazole ring occurred when oxone was used as
the oxidizing agent. When m-CPBA was used as the oxidant, a com-
plex mixture was observed and purification only resulted in further
decomposition.
Synthesis of selectivin-A, selectivin-E, selectivin-F, selectivin-H
and selectivin-I
For each selectivin analogue, the appropriate 2-bromo-acetophenone
was synthesized from the corresponding commercially available aceto-
phenone according to procedures in the literature65. The imidazo[2,1-b]
thiazoles were prepared according to a previously published pro-
cedure66, but with modifications. To a 2 dram vial, the appropriate
α-bromoketone (1 mmol, 1 equiv.), the appropriate 2-aminothiazole
(1.3 mmol, 1.3 equiv.) and ethanol (1.6 ml) were added, and the reaction
mixture was stirred at reflux until disappearance of the α-bromoketone
was evident by thin-layer chromatography. The mixture was concen-
trated then purified by column chromatography using the given eluent
to provide the imidazo[2,1-b]thiazole.
6-(4-Chlorophenyl)imidazo[2,1-b]thiazole (selectivin-A). At 16 mmol
scale. Purified using pentanes–ethyl acetate (13:7 to 10:10 (v/v)), fol-
lowed by trituration with methanol. White solid (1.0 g, 28%). 1H-NMR
(CDCl3, 400 MHz) δ 7.72 (d, J = 8.6 Hz, 2H), 7.67 (s, 1H), 7.38 (d, J = 4.5 Hz,
1H), 7.34 (d, J = 8.6 Hz, 2H), 6.80 (d, J = 4.5 Hz, 1H). 13C-NMR (CDCl3,
100 MHz) δ 150.4, 146.7, 133.1, 132.6, 128.9, 126.5, 118.6, 112.8 and 108.2.
6-(4-Chlorophenyl)-2-methylimidazo[2,1-b]thiazole (selectivin-E).
At 10 mmol (2.3 g) scale. Purified using pentanes–ethyl acetate (15:5 to
10:10 (v/v)). Pale orange solid (30%). 1H-NMR (CDCl3, 500 MHz): 7.73
(d, J = 8.5 Hz, 2H), 7.59 (s, 1H), 7.34 (d, J = 8.5 Hz, 2H), 7.13 (q, J = 1.4 Hz,
1H), 2.42 (d, J = 1.4 Hz, 3H). 13C{1H}-NMR (CDCl3, 125 MHz): 149.9, 145.5,
132.9, 132.8, 128.9, 127.0, 126.4, 115.2, 107.9 and 14.2.
6-(4-Fluorophenyl)-3-methylimidazo[2,1-b]thiazole (selectivin-F).
Purified using pentanes–ethyl acetate (15:5 (v/v)). Brown solid (38%,
MP = 109–114 °C). 1H-NMR (CDCl3, 500 MHz): 7.83–7.78 (m, 2H), 7.57
(s, 1H), 7.12–7.05 (m, 2H), 6.42 (q, J = 1.3 Hz, 1H), 2.43 (d, J = 1.3 Hz, 3H).
13
C{1H}-NMR (CDCl3, 125 MHz): 162.4 (d, J = 246.2 Hz), 149.9, 146.9, 130.4,
127.9 and 127.0 (d, J = 8.0 Hz), 115.7 (d, J = 21.6 Hz), 107.0, 105.8 and 13.5.
19 1
F{ H}-NMR (CDCl3, 375 MHz): –115.0. Infrared (neat): 3,134, 2,965, 2,926,
2,883, 1,750, 1,475, 1,375, 1,155, 1,092, 1,009, 831, 755 and 692. Mass:
DART+, calculated for C12H10N2FS 233.05432 [M+H]+, found 233.05424.
6-(4-Bromophenyl)-3-methylimidazo[2,1-b]thiazole (selectivin-H).
Purified using pentanes–ethyl acetate (16:4 to 15:5 (v/v)). Orange solid
(33%). The spectral data were in accordance with literature62. 1H-NMR
(CDCl3, 500 MHz): 7.73–7.69 (m, 2H), 7.61 (s, 1H), 7.53–7.49 (m, 2H),
6.42 (q, J = 1.3 Hz, 1H), 2.42 (d, J = 1.3 Hz, 3H). 13C{1H}-NMR (CDCl3, 125
MHz): 150.1, 146.8, 133.4, 131.9, 127.8, 126.8, 121.2, 107.1, 106.3 and 13.5.
6-(4-Bromophenyl)-2-methylimidazo[2,1-b]thiazole (selectivin-I).
Purified using pentanes–ethyl acetate (16:4 to 8:12 (v/v)). White solid
(32%, MP = 235–240 °C). 1H-NMR (CDCl3, 500 MHz): 7.69–7.64 (m, 2H),
7.61 (s, 1H), 7.52–7.47 (m, 2H), 7.13 (q, J = 1.4 Hz, 1H), 2.42 (d, J = 1.5 Hz, 3H).
13
C{1H}-NMR (CDCl3, 125 MHz): 150.0, 145.5, 133.2, 131.9, 127.0, 126.7,
121.0, 115.2, 108.0 and 14.2. Infrared (neat): 3,134, 2,965, 2,926, 2,883,
1,750, 1,475, 1,375, 1,155, 1,092, 1,009, 831, 755 and 692. Mass: DART+,
calculated for C12H10N2SBr 292.97426 [M+H]+, found 292.97416.
Work-up and isolation of compounds were performed using stand-
ard benchtop techniques. All commercial reagents were purchased
from chemical suppliers (Sigma-Aldrich, Combi-Blocks, Alfa Aesar
or Strem Chemicals) and used without further purification. Dry sol-
vents were obtained using standard procedures (THF was distilled over
sodium–benzophenone, dichloromethane was distilled over calcium
hydride). Reactions were monitored using thin-layer chromatogra-
phy on EMD Silica Gel 60 F254 plates. Visualization was performed
under UV light (254 nm) or using potassium permanganate (KMnO4)
or I2 stain. Flash column chromatography was performed on Siliaflash
P60 40–63 µm silica gel purchased from Silicycle. NMR characteriza-
tion data were obtained at 293 K on a Varian Mercury 300 MHz, Varian
Mercury 400 MHz, Bruker Advance III 400 MHz, Agilent DD2 500 MHz
equipped with a 5 mm Xses cold probe or Agilent DD2 600 MHz. 1H spec-
tra were referenced to the residual solvent signal (CDCl3 = 7.26 ppm,
DMSO-d6 = 2.50 ppm). 13C{1H} spectra were referenced to the residual
solvent signal (CDCl3 = 77.16 ppm, DMSO-d6 = 39.52 ppm). Data for
1
H-NMR are reported as follows: chemical shift (δ ppm), multiplicity
(s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet), cou-
pling constant (Hz) and integration. NMR spectra were recorded at the
University of Toronto Department of Chemistry NMR facility. Infrared
spectra were recorded on a Perkin-Elmer Spectrum 100 instrument
equipped with a single-bounce diamond/ZnSe ATR accessory in the
solid state and are reported in wavenumber (cm–1) units. Melting point
ranges were done on a Fisher-Johns melting point apparatus and are
reported uncorrected. High resolution mass spectra were recorded at
the Advanced Instrumentation for Molecular Structure in the Depart-
ment of Chemistry at the University of Toronto.
M. incognita tomato plant root infection experiments
A M. incognita population originally collected from grape (Vitis vinif-
era) in Parlier, California, was used in all experiments, and they were
maintained on tomato plants (S. lycopersicum ‘Rutgers’) as previously
described56. Infective M. incognita J2 larvae were collected as previously
described56. For the infection assays, 90 g of soil (1:1 sand:loam mix)
was added to each compartment of several 6-compartment plastic
garden packs. The soil was drenched with 18 ml of deionized water
containing dissolved chemical or DMSO alone. Depending on the
batch, either 1,000 or 2,500 infective J2 larvae were then added to
the soil in 2 ml of water for a total water volume of 20 ml. All of the tests
done in the same batch and on the same day had the same number of
J2 larvae added to the soil so that the DMSO control and treatment
samples had the same number of J2 larvae added. The final aqueous
concentration of the selectivin analogues was 45 µM. This concentra-
tion was chosen because the analogues identified from our primary
screen are lethal to root-knot nematodes at this concentration in vitro
(Fig. 1b). This concentration corresponds to around 4.5 kg of active
ingredient per ha of soil. Despite the label rates being lower than
4.5 kg ha–1 for a subset of the commercial nematicides, we used the
same 4.5 kg of active ingredient per ha application rate for all of the
commercial nematicides tested so that the results would be compara-
ble among the commercial controls and the selectivin analogues. The
final aqueous DMSO concentration varied from 0.05% to 0.4% (v/v)
depending on the stock concentration of the chemical and the final
concentration of the chemical tested. The highest DMSO concentra-
tion of 0.4% (v/v) was used for all DMSO controls. Importantly, we
observed that 0.4% DMSO did not inhibit root infection compared
with water control. There were four DMSO controls in each batch. The
J2 larvae were incubated in the soil and chemical for 24 h, after which
2–3-week-old tomato seedlings were transplanted into the soil (one
plant per compartment). Inoculated plants were grown for 8 weeks in a
greenhouse as previously described56 under long-day conditions (16 h
photoperiod) with 26/18 °C day/night temperatures. After 8 weeks,
the plants were destructively collected. The tops were removed and
discarded, and roots were gently washed with water to remove adher-
ing soil. Eggs were extracted by placing rinsed roots in 0.6% sodium
hypochlorite and agitating at 300 r.p.m. for 3 min. Roots were then
rinsed over nested 250 and 25.4 μm sieves, with eggs collected from
the latter and suspended in water. Roots were dried in a 65 °C oven for
at least 72 h, after which dry roots were weighed. The number of eggs
from each plant root was counted on a dissection microscope using a
haemocytometer, and the number of eggs per milligram of root was
calculated by dividing the total egg number by the mass of the dried
root material. Quantifying the number of eggs per unit of root weight
is a common method for assessing plant susceptibility to parasitic
nematodes67–71. Furthermore, this normalization step controls for
lower egg numbers in the untreated DMSO control samples that result
from stunted root growth owing to nematode infection. The eggs per
milligram of root value for each treatment replicate was normalized to
the average of the four DMSO controls for a given batch. To calculate
per cent effectiveness, the normalized values were subtracted from 1
and then multiplied by 100. An average per cent effectiveness value
was then calculated across at least four replicates (zero eggs = 100%
effectiveness; same number of eggs relative to control = 0% effective-
ness; more eggs relative to control = less than 0% effectiveness). Impor-
tantly, although treatment with our two lead nematicides selectivin-A
and selectivin-E increased root weight relative to the no-compound
controls (as expected for effective nematicides), the root weight
values were not among the highest measured across the 15 different
treatments (Extended Data Fig. 4). Therefore, it is unlikely that the
relatively high per cent effectiveness values for these compounds
are simply due to higher root weight values. Parts-per-million soil
concentrations were calculated as µg active ingredient per g of soil.
Kilograms per hectare values were calculated assuming a soil depth
of 15 cm and a soil density of 1,200 kg m–3.
Forward genetic screens for selectivin-resistant mutants
Forward genetic screens were carried out as previously described14.
In brief, WT parental (P0) worms were mutagenized in 50 mM ethyl
methanesulfonate for 4 h. Synchronized L1 larvae from either the F1
(progeny) or F2 (grand progeny) generations were dispensed onto
Article
10 cm MYOB agar plates (see ref. 52 for how to prepare MYOB agar
medium) containing a 100% penetrant lethal dose of the nematicide.
Worms were plated at a density of about 20,000 L1 larvae per plate. The
plates were scanned by eye for viable worms. In total, 5 million F1 larvae
(that is, 10 million independently mutagenized haploid genomes)
and 50,000 F2 larvae (that is, 100,000 independently mutagenized
haploid genomes) were screened for selectivin-A resistance. Next,
75,000 F1 larvae (that is, 150,000 independently mutagenized haploid
genomes) and 25,000 F2 larvae (that is, 50,000 independently muta-
genized haploid genomes) were screened for selectivin-C-resistant
mutants. Therefore, a total of 10.3 million mutagenized genomes
were screened.
Statistical information
HPLC analyses of unmodified selectivin parent and metabolites were
performed at least three times for each strain or organism, and AUC
values for each analyte were calculated and averaged across the repli-
cates. Data are presented as the mean ± s.e.m. P values were calculated
using unpaired one-tailed Student’s t-tests assuming equal standard
of deviation. In Extended Data Fig. 3c, the pairwise P values are as
follows: Ce/Cb = 0.0028, Ce/Pp = 0.0011, Ce/KR = 0.0028, Ce/Mi = 0.0023,
Ce/Sc = 0.0001, Ce/Dm = 0.0001, Ce/Dr = 0.0001, Cb/Pp = 0.4160,
Cb/KR = 0.4680, Cb/Mi = 0.3311, Cb/Sc = 0.0008, Cb/Dm = 0.0009,
Cb/Dr = 0.0069, Pp/KR = 0.1971, Pp/Mi = 0.0787, Pp/Sc = 2.315 × 10–6,
Pp/Dm = 2.124 × 10–6, Pp/Dr = 0.0004, KR/Mi = 0.1555, KR/Sc = 3.380 × 10–7,
KR/Dm = 8.059 × 10–8, KR/Dr = 0.0017, Mi/Sc = 4.109 × 10–5, Mi/Dm =
3.816 × 10–5, Mi/Dr = 0.0041, Sc/Dm = 0.3169, Sc/Dr = 0.0044 and
Dm/Dr = 0.0046. In Extended Data Fig. 3d, the pairwise P values
are as follows: Ce/Cb = 0.0004, Ce/Pp = 0.0021, Ce/KR = 0.0323,
Ce/Mi = 0.1166, Ce/Sc = 0.0007, Ce/Dm = 0.0007, Ce/Dr = 0.0001,
Cb/Pp = 3.586 × 10 –5, Cb/KR = 1.407 × 10 –5, Cb/Mi = 4.775 × 10 –6,
Cb/Sc = 0.0022, Cb/Dm = 0.0022, Cb/Dr = 0.0006, Pp/KR = 3.505 × 10–6,
Pp/Mi = 8.507 × 10–6, Pp/Sc = 2.618 × 10–9, Pp/Dm = 2.618 × 10–9, Pp/Dr =
4.158 × 10 –11 , KR/Mi = 0.0003, KR/Sc = 4.242 × 10 –7, KR/Dm =
4.242 × 10–7, KR/Dr = 1.067 × 10–8, Mi/Sc = 1.719 × 10–5, Mi/Dm = 1.719 × 10–5,
Mi/Dr = 1.093 × 10–6, Sc/Dm = ND, Sc/Dr = ND, Dm/Dr = ND. For the
yeast-based M. incognita CYP expression screen, assays with the
CYP-expressing strains and the EV control strain were replicated six
times. P values were obtained from unpaired two-tailed Student’s
t-test comparing the mean relative growth values between the
CYP-expressing strains and the EV control. For the C. elegans cyp RNAi
knockdown screen, each cyp knockdown was replicated twice, and
no-knockdown controls (that is, mock experiments) were replicated 50
times. Using the mean per cent-moving value and s.d. from the control
replicates, Z-scores were calculated for each cyp knockdown. A cyp
knockdown resulting in an average per cent-moving value with a Z-score
greater than three was considered to be significant. For the selectivin
analogue series, soil-based root-infection assays were performed at
least four times. To determine whether chemical treatment resulted
in a per cent effectiveness greater than zero, P values were calculated
using two-tailed one-sample t-tests comparing the treatment mean
with zero. A P value < 0.01 indicates a significant difference. Per cent
effectiveness data are presented as the mean ± s.e.m. To determine
whether selectivin-E and selectivin-A mean per cent effectiveness val-
ues differ from that of tioxazafen, unpaired one-tailed Student t-tests
were performed assuming equal s.d. P values > 0.05 indicate no sig-
nificant difference in effectiveness. No statistical methods were used
to predetermine sample sizes. Blinding was performed for the follow-
ing experiments: mouse studies, soil-based plant-parasitic nematode
assays, A. thaliana assays, fish assays, human cell assays, C. albicans
assays, in vitro M. incognita and M. chitwoodi assays, MS and NMR. All
other experiments were not blinded. In all experiments, test subjects
(organisms, cells, among others) were chosen at random for experimen-
tation. GraphPad Prism (v.6.0) and Microsoft Excel for Mac (v.16.16.27)
were used for all statistical analyses.
Reporting summary
Further information on research design is available in the Nature Port-
folio Reporting Summary linked to this article.
Data availability
Data are available for Figs. 1– 5 and Extended Data Figs. 1, 3 and 4. Any
other data that support the conclusions of this manuscript are available
upon request. Source data are provided with this paper.
Code availability
The Python and R code used to generate the dose–response heatmaps
and the HPLC chromatogram heatmaps have been published72.
51. Lewis, J. A. & Fleming, J. T. in Caenorhabditis elegans: Modern Biological Analysis of an
Organism (eds Epstein, H. F. & Shakes, D. C.) 3–29 (Academic Press, 1995).
52. Burns, A. R. et al. High-throughput screening of small molecules for bioactivity and target
identification in Caenorhabditis elegans. Nat. Protoc. 1, 1906–1914 (2006).
53. Demeler, J., Kuttler, U. & von Samson-Himmelstjerna, G. Veterinary parasitology
adaptation and evaluation of three different in vitro tests for the detection of resistance to
anthelmintics in gastro intestinal nematodes of cattle. Vet. Parasitol. 170, 61–70 (2010).
54. Bartley, D. J. et al. A survey of anthelmintic resistant nematode parasites in Scottish sheep
flocks. Vet. Parasitol. 117, 61–71 (2003).
55. Jindapunnapat, K., Reetz, N. D., Macdonald, M. H., Bhagavathy, G. & Meyer, S. L. F. Activity
of vetiver extracts and essential oil against Meloidogyne incognita. J. Nematol. 50,
147–162 (2018).
56. Wram, C. L. & Zasada, I. A. Short-term effects of sublethal doses of nematicides on
Meloidogyne incognita. Phytopathology 109, 1605–1613 (2019).
57. Meyer, S. L. F., Chauhan, K. R. & MacDonald, M. H. Evaluation of roselle (Hibiscus
sabdariffa) leaf and pomegranate (Punica granatum) fruit rind for activity against
Meloidogyne incognita. Nematropica 46, 85–96 (2016).
58. Meyer, S. L. F. et al. Plantago lanceolata and Plantago rugelii extracts are toxic to
Meloidogyne incognita but not to certain microbes. J. Nematol. 38, 333–338 (2006).
59. Sherman, F. Getting started with yeast. Methods Enzymol. 350, 3–41 (2002).
60. Howe, K. L. et al. WormBase ParaSite—a comprehensive resource for helminth genomics.
Mol. Biochem. Parasitol. 215, 2–10 (2016).
61. Blanc-Mathieu, R. et al. Hybridization and polyploidy enable genomic plasticity without
sex in the most devastating plant-parasitic nematodes. PLoS Genet. 13, e1006777 (2017).
62. Sievers, F. et al. Fast, scalable generation of high-quality protein multiple sequence
alignments using Clustal Omega. Mol. Syst. Biol. 7, 539 (2011).
63. WormBase release WS283, http://www.wormbase.org (Alliance of Genome Resources,
accessed 16 December 2021).
64. Wram, C. L. & Zasada, I. Differential response of Meloidogyne, Pratylenchus, Globodera,
and Xiphinema species to the nematicide fluazaindolizine. Phytopathology. 110,
2003–2009 (2020).
65. Chundawat, T. S., Kumari, P., Sharma, N. & Bhagat, S. Strategic synthesis and in vitro
antimicrobial evaluation of novel difluoromethylated 1-(1,3-diphenyl-1H-pyrazol-4-yl)-
3,3-difluoro-1,3-dihydro-indol-2-ones. Med. Chem. Res. 25, 2335–2348 (2016).
66. Pyl, T., Giebelmann, R. & Beyer, H. Über bicyclische heterocyclen mit gemeinsamem
stickstoffatom, I. Zur kenntnis der imidazo[2.1-b]thiazole. Justus Liebigs Ann. Chem. 643,
145–153 (1961).
67. Huang, G. et al. Engineering broad root-knot resistance in transgenic plants by RNAi
silencing of a conserved and essential root-knot nematode parasitism gene. Proc. Natl
Acad. Sci. USA 103, 14302–14306 (2006).
68. Mota, F. C. et al. New sources of resistance to Meloidogyne incognita race 3 in wild cotton
accessions and histological characterization of the defence mechanisms. Plant Pathol.
62, 1173–1183 (2013).
69. Basso, M. F. et al. MiDaf16-like and MiSkn1-like gene families are reliable targets to develop
biotechnological tools for the control and management of Meloidogyne incognita. Sci
Rep. 10, 6991 (2020).
70. Hajihassani, A., Rutter, W. B. & Luo, X. Resistant pepper carrying N, Me1, and Me3 have
different effects on penetration and reproduction of four major Meloidogyne species.
J. Nematol. 51, 1–9 (2019).
71. de Souza, J. D. A. et al. Knocking-down Meloidogyne incognita proteases by plant-
delivered dsRNA has negative pleiotropic effect on nematode vigor. PLoS ONE 8, e85364
(2013).
72. Burns, A. R. and Roy, P. J. Python and R scripts for the generation of dose–response heatmaps
and heatmaps of HPLC chromatograms. Zenodo https://doi.org/10.5281/zenodo.7731172
(2023).
Acknowledgements Free-living nematode strains were provided by the CGC (University of
Minnesota), M.-A. Félix (Institute of Biology of the Ecole Normale Supérieure, Paris, France) and
R. Rae (Liverpool John Moores University, Liverpool, UK). M. hapla was provided by B. Mimee at
Agriculture and Agri-food Canada in Saint-Jean-sur-Richelieu, Quebec, Canada. MS analyses
were performed by staff at the Advanced Instrumentation for Molecular Structure Mass
Spectrometry Laboratory in the Department of Chemistry at the University of Toronto, Canada.
NMR analyses were performed by staff at the CSICOMP NMR Facility in the Department of
Chemistry at the University of Toronto, Canada. Many thanks to D. Colwell and D. Gray at the
Lethbridge and Agri-Food Canada Research Centre for supplying faeces from infected calves
for C. oncophora egg collection; K. Yoshioka at the University of Toronto, Canada, for P. simiae
WSC417; J. Goldstone from the Woods Hole Oceanographic Institution for providing the
M. incognita cyp gene names; B. Ciruna for the gift of NACET; G. Brown for the culture of
S. cerevisiae; B. Derry for helpful comments on the manuscript; and N. Robbins from L. Cowen’s
laboratory for comments on the work. The following grants supported this work: CIHR grants
(313296 and 173448), an NSERC i2i grant (555963-20), a David Dime Grant, and a Canada
Research Chair (Tier 1) to P.J.R.; a NSERC grant (RGPIN-2020-04168) to M.L.; a CIHR Foundation
grant (FDN-154288) and a Canada Research Chair (Tier 1) to L.E.C.; an EvoFunPath fellowship
from the NSERC CREATE program (555337-2021) to E.P.; a CIHR project grant (303157) to I.S.
C.A.M.F. is supported by the Alberta Innovates Technology Fund (G2016000681). J.S.G. is
supported by a NSERC discovery grant (RGPIN-2021-02489), and by the NSERC-CREATE
Host–Parasite Interactions Program graduate training programme at the University of Calgary.
Author contributions Unless otherwise noted, all experiments were carried out in the laboratory
of P.J.R. All free-living nematode dose–response experiments were carried out by A.R.B. A.R.B.
performed the HPLC-based metabolism experiments and analysed all related MS data. A.R.B.
performed the NACET experiments. B.M.P., E.M.R. and A.R.B. performed the C. oncophora
dose–response experiments in J.S.G.’s laboratory. S. cerevisiae work was done by A.R.B., J.K. and
B.C. M. incognita bioinformatics was done by J.K. B.C. performed the yeast-based M. incognita
CYP expression screen, and data were analysed by A.R.B. J.S. performed the HEK293 cell dose–
response assays in I.S.’s laboratory. J.T. (in the laboratory of H.M.K.) and J.R.V. (in the laboratory
of J.J.D.) performed the zebrafish dose–response assays. S.M. and S.W.C. performed the
HepaRG cell dose–response experiments in S.A.M.’s laboratory. E.P. performed the C. albicans
assays in L.E.C.’s laboratory. In the laboratory of S.R.C., A.S.V. performed the Arabidopsis
greening assays. A.R.B performed the P. simiae dose–response experiments. C.A.M.F. and Q.X.
carried out the mouse experiments. A.R.B. performed the metabolite purification for NMR
analyses. R.J.B. carried out the selectivin syntheses and characterization in M.L.’s laboratory,
and helped with metabolite NMR data analyses. A.R.B. performed the C. elegans cyp RNAi
screen. M.H.M. carried out the in vitro M. incognita assays in S.L.F.M.’s laboratory. M.K. carried
out the in vitro M. chitwoodi assays and the M. incognita soil-based reproduction assays
in tomato plants in the laboratory of I.Z. A.R.B. and J.K. carried out the M. hapla assays.
D. melanogaster experiments were carried out by C.H. and A.R.B. in the laboratories of P.J.R.
and H.M.K. The genetics screens for resistant mutants were performed by A.R.B. The project
was conceived by A.R.B. and P.J.R. The manuscript was written by A.R.B. and P.J.R.
Competing interests L.E.C. is a co-founder and shareholder in Bright Angel Therapeutics,
a platform company for development of new antifungal therapeutics. L.E.C. is a consultant
for Boragen, a small-molecule development company focused on leveraging the specific
chemical properties of boron chemistry for crop protection and animal health. L.E.C. is a
Science Advisor for Kapoose Creek, a company that harnesses the therapeutic potential of
fungi. Mention of trade names or commercial products in this publication is solely for the
purpose of providing specific information and does not imply recommendation or endorsement
by the US Department of Agriculture. USDA is an equal opportunity provider and employer.
The University of Toronto is pursuing patent protection related to the use of selectivions as
nematacides for which two national phase applications, US Patent Application 17/624,629, and
Canadian Patent Application 3,146,085, are currently under prosecution and list A.R.B., M.L.,
R.B. and P.J.R. as inventors. A patent application, listing A.R.B., J.K., B.C., and P.J.R. as inventors,
is under preparation for the use of yeast-based expression systems for the identification of
bioactivated nematacides, insecticides, pesticides and drug leads. The patent application will
be soon filed listing the inventors and the University of Toronto as applicants.
Additional information
Supplementary information The online version contains supplementary material available at
https://doi.org/10.1038/s41586-023-06105-5.
Correspondence and requests for materials should be addressed to Andrew R. Burns or
Peter J. Roy.
Peer review information Nature thanks Johan Desaeger, Timothy Geary, Fredd Vergara and the
other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer
reports are available.
Reprints and permissions information is available at http://www.nature.com/reprints.
Article
Extended Data Fig. 1 | The selectivin-A imidazole-thiol metabolite is not
nematicidal. a, Mass spectrometry (MS) data for the M4 metabolite HPLC
fraction and untreated control fraction, as well as MS/MS data for the 418.99
mass. The proposed M4 disulfide structure is shown, along with the protonated
imidazole-thiol monomer structure resulting from fragmentation. b, A proposed
metabolic pathway producing the imidazole-thiol metabolite (ITM) which
oxidizes to form M4/ITM disulfide (ITM-DiS). Imidazothiazole ring opening is
likely CYP-dependent. c, Biomass-normalized and background-corrected
HPLC chromatograms for lysates of selectivin-A-treated L1 worms lysed ±
iodoacetamide (IAA). Peaks corresponding to selectivin-A (sel-A), metabolites
M1-M4, and alkylated ITM (ITM-alk) are indicated. d, MS data for the ITM-alk
HPLC fraction and untreated control fraction. e, HPLC-DAD quantification of
selectivin-A, M1-M4, and ITM-alk in the lysates of selectivin-A-treated L1 worms
lysed ± IAA (n = 3 biological replicates with 200,000 worms per replicate).
Area under the curve (AUC) for metabolites M1-M3 and ITM-alk was calculated
at 260 nm, and at 300 nm for M4. Data are presented as mean ± SEM. For each
analyte, unpaired one-tailed Student’s t-tests were performed comparing
the means of the ± IAA conditions. P-values are shown. f, Dose-response for
C. elegans L1s treated with selectivins or commercially-sourced ITM. g, Biomass-
normalized and background-corrected HPLC chromatograms for lysates
of L1 worms incubated in commercially-sourced ITM, lysed ± IAA. The top
chromatogram is the ITM standard injected directly onto the column, and is
not normalized or corrected. ITM oxidizes to ITM-DiS prior to HPLC analysis.
ITM-DiS and ITM-alk peaks are indicated. h, HPLC-DAD quantification of
M4/ITM-DiS and ITM-alk in the lysates of 100 µM selectivin-A-treated or 100 µM
ITM-treated L1 worms, lysed ± IAA (n = 3 biological replicates with 200,000
worms per replicate). AUC was calculated at 260 nm. Data are presented as
mean ± SEM.
Extended Data Fig. 2 | NMR spectra for HPLC-purified selectivin-A
metabolite M2. a, 1H NMR spectra. The structure of M2 with all carbon and
heteroatom assignments is given. The top number in each multiplet box
corresponds to the carbon that the proton is connected to. 1H NMR (700 MHz,
cd3od) δ 8.02 (s, 1H), 7.82 – 7.78 (m, 2H), 7.42 – 7.37 (m, 2H), 5.98 (dd, J = 7.9,
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),
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),
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,
6.6 Hz, 1H), 2.16 – 2.05 (m, 2H), 1.97 (s, 3H). b, 1H-1H COSY NMR spectra. c, 1H-13C
HSQC NMR spectra. d, 1H-13C HMBC NMR spectra.
Article
Extended Data Fig. 3 | Selectivin bioactivation is nematode-specific.
a and b, Biomass-normalized and background-corrected HPLC chromatograms
of lysates (a) and incubation buffers (b) from C. elegans (Ce), C. briggsae (Cb),
P. pacificus (Pp), Rhabditophanes sp. KR3021 (KR), M. incognita (Mi),
D. melanogaster (Dm), and D. rerio (Dr) larvae, as well as S. cerevisiae (Sc) cells,
incubated in 100 µM selectivin-A. Selectivin-A and metabolite peaks are
indicated. c, Total detectable production of M1 + M2 + M3 in lysate and buffer
for each of the seven species tested. d, M2 tissue accumulation in each species
tested. e, Unmodified selectivin-A tissue accumulation in each of the test
species. AUC is area under the curve at 260 nm. Data are presented as mean ±
SEM. For c and d, unpaired one-tailed Student’s t-tests were performed for all
pairwise comparisons of means. The means sharing a letter are statistically
indistinguishable (p > 0.01). The p-values for each comparison in c and d can be
found in the Statistical Information subsection of the Methods. For e, unpaired
one-tailed Student’s t-tests were performed relative to the C. elegans mean.
P-values are shown. For c-e, n = 4 biological replicates for Ce, Cb, Pp, and Dr, and
n = 3 biological replicates for KR, Mi, Sc, and Dm.
Extended Data Fig. 4 | Selectivin treatment generally increases the root
weights of tomato plants challenged with nematode infection. The effects
of 11 selectivin analogs and 4 commercial nematicides on the root weights of
tomato plants challenged with M. incognita infection is shown. Root weight
values are percent of untreated control. For each analog, aqueous molar
concentration, parts-per-million soil concentration, and kilograms-per-hectare
values are shown. The R-groups for each selectivin analog are indicated (see
Fig. 1a for the selectivin core scaffold and R-group positions). Data are presented
as mean ± SEM.
Article
Extended Data Fig. 5 | Free-living nematodes likely detoxify selectivin-E via
hydroxylation and subsequent phosphoglucosidation. a, Structure of
selectivin-E (sel-E) with exact mass. b-c, Raw HPLC chromatograms for lysates
of C. elegans and P. pacificus L1 worms incubated in 100 µM selectivin-E or
DMSO alone (untreated). The top chromatogram is the selectivin-E standard
injected directly onto the column. The unmodified selectivin-E parent
compound and selectivin-E metabolites E-M1, E-M2, and E-M3 are indicated.
c, Mass spectrometry (MS) data for the E-M1, E-M2, and E-M3 HPLC fractions
and corresponding untreated control fractions. Raw MS counts are shown for
the most abundant masses identified in the mass spectra that are not detectable
in the corresponding untreated control fractions. Raw counts were taken from
centroid plots of the raw mass spectra. Counts below 1 × 103 were considered
not detectable (nd). Based on the presumptive masses of the three metabolites
we predict that E-M1 is a hydroxylated metabolite of selectivin-E, E-M2 is an
O-linked glucoside of selectivin-E, and E-M3 is an O-linked phosphoglucoside.
Molecular formulas are indicated for the predicted metabolite structures, as
well as exact masses, P. pacificus experimental accurate masses, and the ppm
difference between the two. A ppm difference of less than 5 ppm suggests that
the molecular formulae are correct. Ce, C. elegans; Pp, P. pacificus.
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.

A description of all covariates tested

A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons
A full description of the statistical parameters including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient)
AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals)

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.

Software and code

Policy information about availability of computer code
Data collection HPLC-DAD data | HP Chemstation for LC 3D (v. A.10.02)
MS data | Agilent MassHunter Data Acquisition (v. B.05.01)
NMR data | Agilent VnmrJ (v. 4.2)
S. cerevisiae data | National Instruments LabVIEW 2013 (v. 13.0)
C. albicans data | SoftMax Pro 7
P. simiae data | BMG Labtech Omega (v. 1.20)
All other data were collected manually

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

nature research | reporting summary

Policy information about availability of data
All manuscripts must include a data availability statement. This statement should provide the following information, where applicable:
- Accession codes, unique identifiers, or web links for publicly available datasets
- A list of figures that have associated raw data
- A description of any restrictions on data availability

Data and materials availability:

A data availability statement is included in the manuscript. Raw data are included in Supplementary Data Files 1-6 and in the Source Data files. Any other materials
can be made available per request through the corresponding authors ARB or PJR. There are no restrictions on data availability. A list of figures that have associated
raw data is provided in the data availability statement.

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

Life sciences study design

All studies must disclose on these points even when the disclosure is negative.
Sample size No statistical methods were used to predetermine sample size. Nearly throughout, at least three biological (independent) replicates were
performed for each experiment. The number of replicates for each experiment are indicated in the Methods section and in the figure legends.
Sample sizes were chosen as the largest values allowing for error calculations and significance testing while also maintaining the feasibility of
data acquisition and analysis. Overall, variability in the data was low and we were interested in large effect sizes.

Data exclusions No data were excluded from the trials.

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.

Reporting for specific materials, systems and methods

We require information from authors about some types of materials, experimental systems and methods used in many studies. Here, indicate whether each material,
system or method listed is relevant to your study. If you are not sure if a list item applies to your research, read the appropriate section before selecting a response.

Materials & experimental systems Methods

n/a Involved in the study n/a Involved in the study
Antibodies ChIP-seq
Eukaryotic cell lines Flow cytometry
Palaeontology and archaeology MRI-based neuroimaging
Animals and other organisms
Human research participants
April 2020

Clinical data
Dual use research of concern

2
Eukaryotic cell lines

nature research | reporting summary

Policy information about cell lines
Cell line source(s) HEK293 cells were purchased from ThermoFisher (Cat# R71007). HepaRG cells were purchased from ThermoFisher (Gibco
Cryopreserved HPRGC10).

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)

Animals and other organisms

Policy information about studies involving animals; ARRIVE guidelines recommended for reporting animal research
Laboratory animals For the mouse studies, female C57Bl/6 mice aged 6-8 weeks (bred and maintained at the animal care facility, Department of
Biological Sciences, University of Calgary, Canada) were used. For the D. rerio (zebrafish) assays, both male and female AB wild-type
fish were used at 3 dpf.

Wild animals The study did not involve wild animals.

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