Host- Bacterial Co-Metabolism Pyruvate Modulating Afb1 Toxicity in Caenorhabditis Elegans

Full text

Science of the Total Environment 900 (2023) 165809
Available online 26 July 2023
0048-9697/© 2023 Elsevier B.V. All rights reserved.
Bacteria pyruvate metabolism modulates AFB
1
toxicity in
Caenorhabditis elegans
Bowen Tang, Kathy S. Xue, Jia-Sheng Wang, Phillip L. Williams, Lili Tang
*
Department of Environmental Health Science, College of Public Health, University of Georgia, Athens, GA 30602, USA
ARTICLE INFO
Editor: Lidia Minguez Alarcon
Keywords:
Host-microbe-chemical interaction
Aatoxin B
1
C. elegans
Engineered bacteria
Pyruvate metabolic pathway
ABSTRACT
Aatoxin B
1
(AFB
1
), the most potent mycotoxin and Group 1 human carcinogen, continues to pose a signicant
public health burden, particularly in developing countries. Increasing evidence has shown the gut microbiota as a
key mediator of AFB
1
toxicity through multiple interactive host-microbiota activities. In our previous study we
observed that disturbances in bacterial pyruvate metabolism might have a signicant impact on AFB
1
in the host.
To further investigate the impact of the pyruvate pathway on AFB
1
toxicity in C. elegans, we engineered two
bacterial strains (triple-overexpressed and triple-knockout strains with aceB, lpd, and pB). Additionally, we
employed two mutant worm strains (pyk-1 and pdha-1 mutants) known to affect pyruvate metabolism. Our
results revealed that the co-metabolism of pyruvate by the host and bacterial strains synergistically inuences
AFB
1
toxicity. Remarkable, we found that bacterial pyruvate metabolism, rather than that of the host, plays a
pivotal role in modulating AFB
1
toxicity in C. elegans. Our study sheds light on the role of gut microbiota involved
in pyruvate metabolism in inuencing AFB
1
toxicity in C. elegans.
1. Introduction
Aatoxin B
1
(AFB
1
) is the most toxic and hepatocarcinogenic
mycotoxin among the know mycotoxins produced by Aspergillus avus
and A. parasiticus and remains a crucial environmental contamination in
many developing countries (Creppy, 2002). AFB
1
can be metabolically
activated in the liver, which may cause acute hepatic toxicity and im-
mune toxicity, forming DNA adducts and inducing gene mutations, and
has been thereby categorized as Group I human carcinogen (Baan et al.,
2009). AFB
1
is well-documented to be a causative agent of hepatocel-
lular carcinoma (HCC), growth suppression, immune system modula-
tion, and malnutrition (Smoke and Smoking, 2004).
Recent evidence suggested that dietary exposure to AFB
1
can disrupt
the composition and function of gut microbiome, which potentially
inuencing its toxic effects (Chang et al., 2020; Wang et al., 2016; Zhou
et al., 2018). However, the underlying mechanism of the complex re-
lationships among host, gut microbes, and AFB
1
toxicity remains largely
unknown. Further investigation is needed to unravel the complex re-
lationships and shed light on the potential involvement of the gut
microbiome in AFB
1
toxicity.
The nematode Caenorhabditis elegans is an important model organism
in biomedical and toxicological research, mainly due to its well-
characterized genome, short reproductive life cycle, a large number of
offspring, and ease of maintenance (Leung et al., 2008; Stiernagle,
1999). C. elegans is of great convenience for microbiome research
because worms have intestinal cells similar in structure to human in-
testinal cells (McGhee, 2007). Furthermore, the genetic manipulability
of C. elegans allows for precise genetic modications, which can be
complemented by conducting genetic screens and analysing bacterial
mutant collections. The integration of these advantages enables a
comprehensive exploration of microbial-host interactions and the un-
derlying mechanisms involved (Zhang et al., 2017).
In our previous study, we established a 3-way (microbiome-host-
chemical) high-throughput screen (HTS) platform, using C. elegans fed
with E. coli Keio collection on an integrated robotic platform COPAS
Biosort to explore the role of the gut microbiota in modulating AFB
1
toxicity in the context of microbiota-host interaction. We performed 2-
step screenings using 3985 Keio mutants and identied 73 E. coli mu-
tants that modulated C. elegans growth phenotype. Notably, among these
mutants, four genes (aceA, aceB, lpd, and pB) involved in the pyruvate
pathway were identied and subsequently conrmed to be more sensi-
tive to AFB
1
. Our results suggested that disturbances in bacterial
* Corresponding author at: 148 Environmental Health Science Building, 150 Green Street, Athens, GA 30602, USA.
E-mail addresses: bowentang@outlook.com (B. Tang), ksxue@uga.edu (K.S. Xue), jswang@uga.edu (J.-S. Wang), pwilliam@uga.edu (P.L. Williams), ltang@uga.
edu (L. Tang).
Contents lists available at ScienceDirect
Science of the Total Environment
journal homepage: www.elsevier.com/locate/scitotenv
https://doi.org/10.1016/j.scitotenv.2023.165809
Received 28 May 2023; Received in revised form 23 July 2023; Accepted 24 July 2023

Science of the Total Environment 900 (2023) 165809
2
pyruvate metabolism might have a signicant impact on AFB
1
in the
host (Tang et al., 2023).
In this study, we aimed to gain a deeper understand the impact of the
pyruvate pathway on AFB
1
toxicity in C. elegans. To achieve this, we
employed genetic engineering techniques to create two bacterial strains:
one with triple overexpression (aceB, lpd, and pB) and the other with
triple knockout of these genes. Additionally, we utilized two mutant
worm strains, pyk-1 and pdha-1 mutants, which are known to disrupt
pyruvate metabolism. By employing these genetic modications, we
were able to directly examine the impact of specic bacterial alterations
on the interaction between pyruvate pathway and AFB
1
toxicity.
2. Method and materials
2.1. Chemicals
Aatoxin B
1
(AFB
1
), sodium pyruvate and Pyruvate Assay Kit
(MAK071) were purchased from Sigma-Aldrich (St. Louis, MO). Stock
solutions at 1 M were prepared in DMSO. Fresh working solutions at
different concentrations were diluted from the stock using K-medium
(32 mM KCl and 51 mM NaCl) (Williams and Dusenbery, 1990) con-
taining E. coli (wild type or engineered type) as a food source at 1 mg/mL
(Brenner, 1974).
2.2. Nematode and bacterial strains
The nematodes, N2, VC3879 pyk-1 (gk3762), and VC4479 pdha-1
(gk5568), were purchased from the Caenorhabditis Genetics Center
(Minneapolis, MN, USA). All C. elegans strains were maintained at 20 ◦C
on solid nematode growth medium (NGM) plates seeded with OP50
(Brenner, 1974) or engineered bacteria.
Escherichia coli Keio collection, including the parent strain BW25113
and knockout strains △aceB, △lpd, and △pB), as the food source for
C. elegans were purchased from the NBRP at the National Institute of
Genetics, Shizuoka, Japan. Wild-type E. coli strain (BW25113) was
grown in Luria Broth (LB) medium (tryptone 10 g/L, yeast extract 5 g/L,
and NaCl 5 g/L) while E. coli Keio mutants were grown in LB containing
25
μ
g/L kanamycin. All the bacteria were grown for ~18 h at 37 ◦C to
reach OD600 value to 1.0–1.5 and washed with K-medium 3 times
before being diluted to 1 mg/mL as a food source for nematodes.
2.3. High-throughput screening for growth in C. elegans
High-throughput assay testing for growth was performed using
COPAS Biosort (Boyd et al., 2010b; Tang et al., 2019; Tang et al., 2020).
Isolated C. elegans eggs with bleach‑sodium hydroxide were hatched in
K-medium overnight to obtain synchronized L1 growth-arrested larvae
(Lewis and Fleming, 1995). A total of 50 age-synchronized L1-stage
worms were sorted and dispensed into each well containing 90
μ
L of test
solutions which contained food source and the test chemical. After in-
cubation for 48 h at 20 ◦C, the plates were loaded onto COPAS Biosort
for growth measurement. Prior to measurement, 10
μ
L formalin (v/v, 10
%) were added to each well to kill and straighten the animals for proper
measurement. The TOF values for each event were recorded as size in-
dicators for each worm.
2.4. Plasmid construction for triple overexpression of targeted genes in
BW25113
All DNA manipulations were performed based on the standard mo-
lecular cloning protocols (Sambrook et al., 1989). E. coli competent cells,
XL1-Blue (Stratagene), were used for plasmid construction. All primers
used in this study were listed in Supplementary Table 1. Gene products
(pB) amplied by PCR were digested with the corresponding enzymes
and ligated into similarly digested medium-copy-number plasmid pCS27
between SalI and MluI to form pCS27-pB. The amplied aceB and lpd
from E. coli were digested with Acc65I and NdeI, NdeI and SalI,
respectively, and then integrated into pCS27-pB to form pCS27-aceB-
lpd-pB. 2 % of the overnight cell culture was inoculated into 50 mL LB
medium in 125 mL asks and incubated for 2 h at 37 ◦C and 280 rpm.
Isopropyl β-D-1-thiogalactopyranoside (IPTG) was added at a nal
concentration of 0.5 mM, and cell cultures were switched to 30 ◦C for
induction. LB medium with kanamycin (25
μ
g/mL) was used for plasmid
propagation and cell inoculation.
2.5. Gene disruption for triple knockout of targeted genes in BW25113
The disruption of aceB, lpd, and pB in E. coli was conducted by P1
phage-based transduction (Thomason et al., 2007). Specically, P1 ly-
sates for target knockouts were prepared from the Keio collection
strains. The transduced cells were screened on an LB-agar plate with
kanamycin and 100 mM sodium citrate. Kanamycin cassette on Keio
mutants was anked by the FLP recombinase site, thus allowing for the
excision of the cassette. The kanamycin cassette on the gene of interest
was removed by electroporation of pCP20, containing FLP recombinase,
into target strains (Datsenko and Wanner BLJPotNAoS., 2000). Trans-
formants were selected for ampicillin resistance at 30 ◦C and then raised
at 42 ◦C to remove the pCP20 plasmid. Colonies were veried by the
inability to grow on LB plates with the kanamycin. Knockout strains
were conrmed by colony PCR.
2.6. Qualication of pyruvate level
To measure the pyruvate level in C. elegans (N2, pyk-1, and pdha-1),
worms were washed with K-medium 3 times and extracted with meth-
anol/water (4:1) using a FastPrep-24 (MP Biomedicals) for ve cycles of
60 s. The tubes were then centrifuged at 3000g for 20 min to separate the
supernatant. The extraction process was repeated twice, and the su-
pernatants were combined for pyruvate analysis. Three types of Bacteria
(BW25113, triple overexpression, and triple knockout) at exponential
growth phase (OD600 =0.5) were collected and centrifuged at 1500g
for 1 min. The supernatant was used for pyruvate measurement. Pyru-
vate levels were estimated using the ELISA kit from Abcam (MA, USA)
according to the manufacturer instruction.
2.7. Statistical analysis
All data were processed with the statistical program R, version 4.0.2.
Bacterial genes which induced statistically signicant changes in worms
in response to AFB
1
in exposed groups (worms fed E. coli Keio mutants)
compared to controls (worms fed the wild-type strain BW25113) were
identied using one-way analysis of variance (ANOVA) with Tukey’s t-
test. A p-value of <0.05 was considered statistically signicant. To
model decreasing activity of growth with increasing dose, the Hill
expression Hill=vm*dosex
kmx+dosex was subtracted from V
m
+a
0
. EC
50
(the 50 %
effective concentration of a toxicant) and corresponding 95 % CI (con-
dence intervals) for each cohort were calculated by tting the mean
values of parameters (TOF for growth) to the decreasing Hill equation
(Eq. (1)) (Boyd et al., 2010a). The V
m
+a
0
refers to the worm response
after exposure for the non-treated nematodes and is the maximum value
the expression can have. The value of the expression decreases to the
lower asymptote, a
0
, as chemical concentration increases. The lower
asymptote indicates a maximum possible decrease in the response. The
parameter, Km, is the estimated concentration that decreases the
response to halfway between Vm +a
0
and a
0
, which is the EC
50
. At last,
x is the shape parameter that governs the steepness of the decrease.
Decreasing Hill =a0+Vm
1+(dose
Km )x(1)
B. Tang et al.

Science of the Total Environment 900 (2023) 165809
3
3. Results
3.1. Bacterial pyruvate metabolism plays an important role in modulating
AFB
1
toxicity
In our previous study, we utilized a genetic screen with the Keio
collection and subsequent enrichment analysis, which revealed the
involvement of four bacterial genes (aceA, aceB, lpd, and pB) in
modulating AFB
1
toxicity in C. elegans through the pyruvate biosynthesis
pathway (Fig. 1). To investigate the impact of bacterial pyruvate
metabolism on AFB
1
toxicity in the host, we performed genetic engi-
neering on two bacterial strains derived from the wild-type BW25113:
one strain with triple knockouts of the three genes, and another strain
with triple overexpression. According to the bacterial metabolic
pathway (Fig. 1), aceB, lpd, and pB play crucial roles in regulating the
conversion of pyruvate into either acetyl Co-A or formate. We hypoth-
esis that the triple knockout of aceB, lpd, and pB leads to an increase in
pyruvate production, whereas triple overexpression results in a reduc-
tion in pyruvate levels. This is conrmed by measuring the pyruvate
levels in the bacteria medium (Fig. 2A). Compared to BW25113, the
pyruvate level in the culture medium was 11.6 % higher in triple
knockout bacteria and 18.4 % lower in triple overexpression bacteria.
The wild-type worm N2 was fed the two types of engineered bacteria
(triple knockouts and triple overexpression) as well as wild-type bacteria
(BW25113), and subsequently exposed to AFB
1
at concentrations
ranging from 0.1 to 10
μ
mol/L
,
and their body length was measured
afterward (Fig. 2B). Animals fed the pyruvate-downregulated bacterial
showed reduced sensitivity to AFB
1
with longer body length at each
concentration (p <0.05) compared to those fed the wild-type bacteria.
Conversely, animals fed the pyruvate-upregulated triple-knockout bac-
teria exhibited increased sensitivity to AFB
1
, with shorter body lengths
at each concentration (p <0.05). The calculated EC
50
values for the
three bacteria are showed in Table 1. Triple overexpression bacterial has
higher EC50 (8.05
μ
M), the EC50 of BW25113 and Triple knockout
bacterial was 3.92 and 3.12
μ
M, separately. To assess the potential
impact of the engineered bacterial strains on the response of worms, the
effects of these strains on the growth of C. elegans were also examined in
the absence of AFB
1
(Supplemental Fig. 1), the results indicated no
signicant difference in the body length, compared to wild-type bacte-
rial (p >0.05). These ndings suggest that pyruvate in bacteria
potentially plays a role in modulating host response to AFB
1
.
3.2. Host pyruvate metabolism involved in modulating AFB
1
toxicity
To gain a deeper understanding of pyruvate’s role in the host, we
conducted further investigations to determine if the host’s pyruvate
metabolism is involved in modulating AFB
1
toxicity. Two mutant worm
strains, pyruvate kinase-decient (pyk-1) worms and pyruvate
dehydrogenase-decient (pdha-1) worms were employed. According to
pyruvate metabolic pathway in C. elegans (Fig. 3), knocking out the
pdha-1 gene resulted in increased pyruvate levels, while knocking out
the pyk-1 gene led to decreased pyruvate levels. As shown in Fig. 4, pyk-1
knockout worms showed 8 % lower (p >0.05) pyruvate level while
pdha-1 worms showed 23.2 % (p <0.01) higher level of pyruvate,
compared to the wild-type N2.
Fig. 4B demonstrates that the pyk-1 mutant worm resulted in reduced
sensitivity to AFB
1
as indicated by body length, in a concentration-
dependent manner with the highest EC
50
value (4.60
μ
M, Table 2),
compared to the wild-type worm and pdha-1 mutant worm (p <0.05).
Conversely, the pdha-1 mutant worm led to increased sensitive to AFB
1
,
in a concentration-dependent manner, with the lowest EC
50
value (2.38
μ
M, Table 2), compared to the wild-type worm and pyk-1 mutant worm
(p <0.05).
3.3. Host-bacteria co-metabolism synergistic modulating AFB
1
toxicity
To further investigate the effect of host-bacterial interaction on AFB
1
toxicity and determine whether that of the host or the microbe, plays a
more important role in modulating host response to AFB
1
, we examined
the worms’ body length change in response to AFB
1
by using both
engineered bacteria strains and mutant worm strains. As mentioned
above, knocking out pyk-1 worm and triple-overexpression bacterial
strain led to decrease pyruvate level, while the pdha-1 mutant worm and
triple-knockout bacterial strain resulted in increasing production of
pyruvate. As expected, the response to AFB
1
has signicantly difference,
in both a concentration- and bacteria-dependent manner, compared to
those fed with wild-type bacterial (Fig. 5).
When pyruvate level was downregulated in both bacteria (triple-
overexpression) and host (pyk-1 mutant worms), pyk-1 mutant worms
became less sensitive to AFB
1
as indicated by body length, compared to
those fed the wild-type bacteria, even when compared to their untreated
controls, at the lowest concentration. At the highest concentration (10
μ
M), pyk-1 mutant worms fed the triple-overexpressed bacteria were
1.39-fold more resistant to AFB
1
compared to those fed the wild-type
bacteria, suggesting the protective effects of decreased pyruvate level
in triple-overexpressed bacteria. However, the worm’s growth was
signicantly inhibited fed with triple-knockout bacterial, compared to
those fed the wild-type bacteria. On the other hand, when pyruvate level
was upregulated in both bacteria (triple-knockout) and host (pdha-1
mutant worms) (Fig. 5A), pdha-1 mutant worms exhibited greater size
reduction. At the highest (10
μ
M) concentrations, animals became 1.34-
fold more sensitive to AFB
1
than worms fed the wild-type bacteria,
suggesting the synergistic effects of increased pyruvate level in the
triple-overexpressed bacteria.
For determining whether the host or the bacterial pyruvate plays a
more important role in modulating host response to AFB
1
, we examined
worms body length change in response to AFB
1
using an antagonistic
mechanism by increasing pyruvate metabolism in bacteria while
decreasing that in the host, and vice versa. Fig. 5B showed that when
pyruvate levels were upregulated in bacteria (triple-knockout) but
downregulated in worms (pyk-1 mutants), animals’ growth was signif-
icantly inhibited, compared to those fed the wild-type and triple-
overexpression bacteria. However, when pyruvate levels were
decreased in bacteria (triple-overexpression) but increased in worms
(pdha-1 mutants), worms’ growth was signicantly promoted, compared
to those fed the wild-type and triple-knockout bacteria. This indicates
Fig. 1. Disrupted bacterial metabolism based on KEGG enrichment analysis.
PFL (pB), pyruvate formate lyase; lpd, lipoamide dehydrogenase, pyruvate
dehydrogenase E3 monomer; PDH (aceB), pyruvate dehydrogenase complex;
MS (aceB), malate synthase; OAA, oxaloacetate; GOX, glyoxylate.
B. Tang et al.

Science of the Total Environment 900 (2023) 165809
4
that pyruvate metabolism in the bacteria, rather than that in the host,
plays an essential role in modulating AFB
1
toxicity in C. elegans.
3.4. Pure pyruvate supplement modulating AFB
1
toxicity
To further conrm the effects of pyruvate in modulating AFB
1
toxicity, 1 mM pyruvate was supplemented to worm mutants fed
different bacteria, whose growth was then examined after exposure to 1
μ
M AFB
1
. All the animals became more sensitive to AFB
1
when sup-
plemented with pyruvate (Fig. 6). Animals supplemented with pyruvate
showed increased sensitivity to AFB
1
with shorter body length (p <0.01)
compared to those without pyruvate supplement. In wild-type worms
N2, the addition of pyruvate caused a 1.1, 2, and 1.5-fold increase in
sensitivity to AFB
1
when the animals were fed triple-overexpressed,
wild-type, and triple-knockout bacteria, respectively. In pyruvate
kinase-decient worms (pyk-1), pyruvate supplement induced 1.3- and
1.2-fold increased sensitivity to AFB
1
when worms were fed wild-type
and triple-overexpressed bacteria, respectively. In pyruvate
dehydrogenase-decient worms (pdha-1), worms fed triple-knockout
and wild-type bacteria suffered 1.45- and 2.13-fold increased sensi-
tivity to AFB
1
, respectively. To summarize, our nding suggested that
pyruvate plays an important role in modulating AFB
1
toxicity in
C. elegans.
4. Discussion
The gut microbiota plays a signicant role in the toxicity of envi-
ronmental contaminants (Claus et al., 2016). In our previous study, we
established a 3-way high-throughput screening system involving mi-
crobes, worms and chemicals. Through this system, we identied four
bacteria genes (aceA, aceB, lpd, and pB) involved in the pyruvate
metabolism that displayed signicant contribution to induce different
growth rates in response to AFB
1
in C. elegans (Tang et al., 2023).To
further explore the inuence of the pyruvate pathway in modulating
AFB
1
toxicity in C. elegans, we engineered two bacteria strains (triple-
overexpressed and triple-knockout with aceB, lpd, and pB) together
with two mutant worm strains (pyk-1 and pdha-1 mutants). Our ndings
indicate that co-metabolism of pyruvate by the host and bacterial syn-
ergistically modulated AFB
1
toxicity. Specically, and the bacterial py-
ruvate metabolism, rather than that of the host, plays a crucial role in
modulating AFB
1
toxicity in C. elegans.
Pyruvate metabolism is a crucial process in cellular respiration and
plays a central role in energy production and various metabolic path-
ways. Pyruvate, a three‑carbon molecule, is derived from glucose
metabolism through glycolysis. Once formed, pyruvate can be further
metabolized through different pathways depending on the cell’s energy
demands and metabolic conditions (Gray et al., 2014). Studies have
demonstrated that mycotoxins, including AFB
1
, impaired ATP avail-
ability and resulted in decreased pyruvate levels (Baldissera et al., 2018;
Po´
or et al., 2014). However, the mechanisms of pyruvate’s effects on
AFB
1
toxicity remain largely unknown. Research has shown that certain
metabolic pathways can modulate the toxicity of AFB
1
, including those
involving pyruvate. Our results showed that addition of pyruvate
increased worms’ sensitivity to AFB
1
by fed with triple- knockout bac-
teria and supplementing pyruvate (1 mM) to worms. This might be
related to oxidative stress and the activity of cytochrome P450. The
Fig. 2. (A) Pyruvate levels in bacteria medium. (B) The effects of different bacteria on growth of wild type C. elegans (N2) in response to AFB
1
. Raw data were
normalized to the TOF value for untreated animals, shown as a percentage of the TOF value for untreated animals. Results are presented as mean of 5 replicates and
error bars represent standard error. *p <0.05, **p <0.01 determined by Student’s t-test.
Table 1
The estimated EC
50
values and corresponding 95 % condence interval on
growth in wild type C. elegans fed on different bacteria mutants to AFB
1
exposure.
BW25113 Triple knockout Triple overexpression
N2 EC
50
(
μ
M) 3.92 3.12 8.05
95 % CI (2.98–4.86) (0–6.83) (1.88–14.22)
Fig. 3. Pyruvate metabolism in C. elegans. Modied based on C. elegans meta-
bolism pathway in KEGG. In pyruvate kinase mutant worms (pyk-1), pyruvate
level is decreased while in pyruvate dehydrogenase mutant worms (pdha-1),
pyruvate level is increased.
B. Tang et al.

Science of the Total Environment 900 (2023) 165809
5
pyruvate was found to induce hydrogen peroxide accumulation and
cause excessive ROS production in C. elegans (Mouchiroud et al., 2011),
which further promoted DNA damage and cell death induced by AFB
1
.
Although in one study found that pyruvate supplement at 100 mM in
worms’ diet inducing stress resistance and cellular defence and promote
longevity in worms (Tauffenberger et al., 2019). However, lower con-
centrations of pyruvate (10 mM) did not induce this phenomenon
(Tauffenberger et al., 2019). To avoid the above effects, a pyruvate
supplement at a lower concentration (1 mM) was selected in this study.
In addition, we also found pyruvate supplement alone at such concen-
tration did not cause any different growth rate in C. elegans in the
absence of AFB
1
.
In C. elegans, Pyruvate kinase (PYK) catalyses irreversibly transfer
phosphor from phosphoenolpyruvate (PEP) to ADP to form pyruvate and
ATP. Another important enzyme complex involved in pyruvate meta-
bolism is pyruvate dehydrogenase (PDH), which converts of pyruvate
into citric acid cycle to generate energy (Wang et al., 2002). Our result
Fig. 4. (A) Pyruvate levels in C. elegans. (B) The effects of BW25113 on growth of wild type and mutant C. elegans in response to AFB
1
. Raw data were normalized to
the TOF value for untreated animals, shown as a percentage of the TOF value for untreated animals. Results are presented as mean of 5 replicates and error bars
represent standard error. *p <0.05, **p <0.01 determined by Student’s t-test.
Table 2
The estimated EC
50
values and corresponding 95 % condence interval on
growth in wild type and mutant C. elegans fed on BW25113 in response to AFB
1
exposure.
N2 pyk-1 pdha-1
BW25113 EC
50
(
μ
M) 3.92 4.60 2.38
95 % CI (2.98–4.86) (0.95–8.25) (0.66–4.10)
Fig. 5. The effects of BW25113 and engineered bacteria on growth of mutant C. elegans (pyk-1 and pdha-1) in response to AFB
1
. (A) pyk-1 and pdha-1 worms treated
with triple overexpression and triple knockout bacteria, respectively. (B) pyk-1 and pdha-1 worms treated with triple knockout and triple overexpression bacteria,
respectively. Raw data were normalized to the TOF value for untreated animals, shown as a percentage of the TOF value for untreated animals. Results are presented
as mean of 5 replicates and error bars represent standard error. *p <0.05, ***p <0.001 determined by Student’s t-test.
B. Tang et al.

Science of the Total Environment 900 (2023) 165809
6
indicated that pyk-1 and pdha-1 mutant worms, which knocking out pyk
and pdh genes, affected the pyruvate content in C. elegans, consequently
synergistically modulating AFB
1
toxicity in combination with bacterial
pyruvate pathway. Interestingly, despite the pyruvate level in pyk-1
knockout worms didn’t show a signicant reduction compared to
wild-type worms. However, the pyk-1 mutant worm still exhibited
reduced sensitivity to AFB
1.
This observation suggests that the decreased
sensitivity may be attributed to the role of PYK itself. Previous studies
have reported that PYK can react with AFB
1
-dialdehyde, which is
derived from the rapid hydrolysis of AFBO to AFB
1
-dihydrodiol in AFB
1
phase II detoxication pathway, thereby leading to increased cytotox-
icity (Hayes et al., 1993; Neal et al., 1981). Furthermore, PYK has
demonstrated efcient bind AFB
1
and signicant enhancement of AFB
1
activation in the nucleus (McLean and Dutton, 1995).
AFB
1
, the most toxic and hepatocarcinogenic mycotoxin known,
continues to pose a signicant public health burden, particularly in
developing countries. (Benkerroum, 2009; Liu and Wu, 2010; Rasheed
et al., 2021). In 2004, it was estimated that ~4.5 billion people in
developing countries were at risk for chronic and uncontrolled AFB
1
exposure (Williams et al., 2004). To address this issue, many countries
have implemented strict regulations to minimize AFB
1
levels in food.
However, these regulations have led to substantial annual losses in the
agricultural industry. In the United States alone, losses due to aatoxins,
including AFB
1
, have been estimated to range from $52.1 million to
$1.68 billion annually (Mitchell et al., 2016). Current strategies for AFB
1
control primarily focus on decontamination by degrading or removing
AFB
1
from crops, as well as prevention by implementing appropriate
management systems to reduce AFB
1
contamination in the eld and
during storage (Benkerroum, 2009). However, these methods are
insufcient when AFB
1
is already absorbed by the human body. In recent
years, probiotic bacteria have emerged as potential detoxication tools
for mycotoxins by inhibiting intestinal absorption, limiting the entry of
toxicants into the systemic circulation, and facilitating their excretion
(Paulina et al., 2021). The results of our study provide signicant in-
sights into microbiome-based intervention strategies for AFB
1
exposure.
By investigating the role of the gut microbiota, specically the bacteria
involved in pyruvate metabolism, in modulating AFB
1
toxicity in
C. elegans, our ndings contribute to the growing body of knowledge on
the potential of microbiota-based approaches to mitigate the adverse
effects of AFB
1
. Understanding the interplay between the gut microbiota
and AFB
1
toxicity can pave the way for the development of novel in-
terventions that target the microbiome and its interactions with
mycotoxins.
The main limitation of this study is the differences in AFB
1
meta-
bolism between C. elegans and mammalian models. AFB
1
requires
metabolic activation by CYP enzymes to form ultimate epoxide
metabolites, which bind to DNA molecules, and resulting in DNA dam-
age. AFB
1
can be activated in mammals by CYP1, CYP2, and CYP3 family
enzymes, however, C. elegans lacks CYP1 family enzymes (Gotoh, 1998;
Leung et al., 2010). This will decrease the sensitivity of C. elegans
response to the toxicity of AFB
1
, though there was a study found that
C. elegans can metabolize AFB
1
into DNA-binding metabolites and that
this activation is CYP dependent (Leung et al., 2010). Differences in
study models, such as variations in species, have been challenges for the
research eld of toxicology and risk assessment, especially for extrap-
olations from lab study ndings to human population risks(Queiros
et al., 2019).
5. Conclusion
Our previous study identied four bacterial genes (aceA, aceB, lpd,
and pB) that related to pyruvate metabolism pathways play a signi-
cant role in modulating AFB
1
toxicity in C. elegans. To further investigate
the impact of the pyruvate pathway on AFB
1
toxicity in C. elegans, we
engineered two bacterial strains (triple-overexpressed and triple-
knockout strains with aceB, lpd, and pB) along with two mutant
worm strains (pyk-1 and pdha-1 mutants). Our ndings demonstrate
that the co-metabolism of pyruvate by both the host and bacterial strains
synergistically inuences AFB
1
toxicity. Notably, bacterial pyruvate
metabolism, rather than that of the host, plays a pivotal role in modu-
lating AFB
1
toxicity in C. elegans. Our study sheds light on the role of gut
microbiota, particularly bacteria involved in pyruvate metabolism, in
inuencing AFB
1
toxicity in C. elegans. These ndings add to the
expanding understanding of the potential of microbiota-based strategies
in alleviating the detrimental impacts of AFB
1
.
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.scitotenv.2023.165809.
Funding sources
This work was supported by the USDA/NIFA under Grant 2022-
67017-36237. N2 worms were provided by the CGC, funded by the NIH
Ofce of Research Infrastructure Programs (P40 OD010440).
CRediT authorship contribution statement
Bowen Tang: Methodology, Investigation, Data curation, Formal
analysis, Writing – original draft. Kathy S. Xue: Methodology, Valida-
tion, Resources, Writing – review & editing, Visualization. Jia-Sheng
Wang: Validation, Supervision, Writing – review & editing. Phillip L.
Williams: Validation, Resources, Writing – review & editing. Lili Tang:
Supervision, Project administration, Funding acquisition, Writing –
Fig. 6. The effects of AFB
1
(1
μ
M) in C. elegans mutants fed different bacteria strains, with or without pyruvate supplement (1 mM). Mean TOF values were obtained
directly from COPAS Biosort without any normalization. Results are presented as mean of 5 replicates and error bars represent standard error. ***p <0.01
determined by Student’s t-test.
B. Tang et al.

Science of the Total Environment 900 (2023) 165809
7
review & editing, All authors have read and agreed to the published
version of the manuscript.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Data availability
Data will be made available on request.
References
Baan, R., Grosse, Y., Straif, K., Secretan, B., El Ghissassi, F., Bouvard, V., et al., 2009.
A review of human carcinogens—part F: chemical agents and related occupations,
10, pp. 1143–1144.
Baldissera, M.D., Souza, C.F., Zeppenfeld, C.C., Descovi, S., da Silva, A.S.,
BJFP, Baldisserotto, et al., 2018. Changes in the Cerebral Phosphotransfer Network
Impair Energetic Homeostasis in an Aatoxin B 1-contaminated Diet, 44,
pp. 1051–1059.
Benkerroum, N., 2009. Aatoxins: producing-molds, structure, health issues and
incidence in Southeast Asian and Sub-Saharan African countries. Int. J. Environ. Res.
Public Health 17 (4), 1215.
Boyd, W.A., McBride, S.J., Rice, J.R., Snyder, D.W., Freedman, J.H., 2010a. A high-
throughput method for assessing chemical toxicity using a Caenorhabditis elegans
reproduction assay. Toxicol. Appl. Pharmacol. 245, 153–159.
Boyd, W.A., Smith, M.V., Kissling, G.E., Freedman, J.H., 2010b. Medium-and high-
throughput screening of neurotoxicants using C. elegans. Neurotoxicol. Teratol. 32,
68–73.
Brenner, S., 1974. The genetics of Caenorhabditis elegans. Genetics 77, 71–94.
Chang, J., Wang, T., Wang, P., Yin, Q., Liu, C., Zhu, Q., et al., 2020. Compound probiotics
alleviating aatoxin B1 and zearalenone toxic effects on broiler production
performance and gut microbiota. Ecotoxicol. Environ. Saf. 194, 110420.
Claus, S.P., Guillou, H., Ellero-Simatos, S., 2016. The gut microbiota: a major player in
the toxicity of environmental pollutants? npj Biolms Microbiomes 2, 16003.
Creppy, E.E., 2002. Update of survey, regulation and toxic effects of mycotoxins in
Europe. Toxicol. Lett. 127, 19–28.
Datsenko, K.A., Wanner BLJPotNAoS., 2000. One-step Inactivation of Chromosomal
Genes in Escherichia coli K-12 Using PCR Products, 97, pp. 6640–6645.
Gotoh, O., 1998. Divergent structures of Caenorhabditis elegans cytochrome P450 genes
suggest the frequent loss and gain of introns during the evolution of nematodes. Mol.
Biol. Evol. 15, 1447–1459.
Gray, L.R., Tompkins, S.C., Taylor, E.B., 2014. Regulation of pyruvate metabolism and
human disease. Cell. Mol. Life Sci. 71, 2577–2604.
Hayes, J.D., Judah, D.J., Neal GEJCr., 1993. Resistance to Aatoxin B1 is Associated
With the Expression of a Novel Aldo-keto Reductase Which has Catalytic Activity
Towards a Cytotoxic Aldehyde-containing Metabolite of the Toxin, 53,
pp. 3887–3894.
Leung, M.C., Williams, P.L., Benedetto, A., Au, C., Helmcke, K.J., Aschner, M., et al.,
2008. Caenorhabditis elegans: an emerging model in biomedical and environmental
toxicology. Toxicol. Sci. 106, 5–28.
Leung, M.C., Goldstone, J.V., Boyd, W.A., Freedman, J.H., Meyer, J.N., 2010.
Caenorhabditis elegans generates biologically relevant levels of genotoxic
metabolites from aatoxin B1 but not benzo[a]pyrene in vivo. Toxicol. Sci. 118,
444–453.
Lewis, J.A., Fleming, J.T., 1995. Basic culture methods. In: Methods in Cell Biology, 48.
Elsevier, pp. 3–29.
Liu, Y., Wu, F., 2010. Global burden of aatoxin-induced hepatocellular carcinoma: a
risk assessment. Environ. Health Perspect. 118 (6), 818–824.
McGhee, J.D., 2007. The C. elegans intestine. In: WormBook: The Online Review of C.
Elegans Biology [Internet].
McLean, M., Dutton, M.F.J.P., 1995. Therapeutics. In: Cellular Interactions and
Metabolism of Aatoxin: An Update, 65, pp. 163–192.
Mitchell, N.J., Bowers, E., Hurburgh, C., Wu, F., 2016. Potential economic losses to the
US corn industry from aatoxin contamination. Food Addit. Contam. Part A Chem.
Anal. Control Expo. Risk Assess. 33, 540–550.
Mouchiroud, L., Molin, L., Kasturi, P., Triba, M.N., Dumas, M.E., Wilson, M.C., et al.,
2011. Pyruvate Imbalance Mediates Metabolic Reprogramming and Mimics Lifespan
Extension by Dietary Restriction in Caenorhabditis Elegans, 10, pp. 39–54.
Neal, G., Metcalfe, S., Legg, R., Judah, D., Green, J.J.C., 1981. Mechanism of the
Resistance to Cytotoxicity Which Precedes Aatoxin B 1 Hepatocarcinogenesis, 2,
pp. 457–461.
Paulina, S., Edyta, J.-K., Michał, W., Monika, A., Marek, R.L.J.F., 2021. Probiotics as a
biological detoxication tool of food chemical contamination: a review. Food Chem.
Toxicol. 112306.
Po´
or, M., Veres, B., Jakus, P.B., Antus, C., Montsk´
o, G., Zrínyi, Z., et al., 2014. Flavonoid
Diosmetin Increases ATP Levels in Kidney Cells and Relieves ATP Depleting Effect of
Ochratoxin A, 132, pp. 1–9.
Queiros, L., Pereira, J.L., Goncalves, F.J.M., Pacheco, M., Aschner, M., Pereira, P., 2019.
Caenorhabditis elegans as a tool for environmental risk assessment: emerging and
promising applications for a “nobelized worm”. Crit. Rev. Toxicol. 49, 411–429.
Rasheed, H., Xu, Y., Kimanya, M.E., Pan, X., Li, Z., Zou, X., et al., 2021. Estimating the
Health Burden of Aatoxin Attributable Stunting Among Children in Low Income
Countries of Africa, 11, pp. 1–11.
Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual.
Cold Spring Harbor Laboratory Press.
Smoke, T., Smoking, I., 2004. IARC Monographs on the Evaluation of Carcinogenic Risks
to Humans, 1. IARC, Lyon, pp. 1–1452.
Stiernagle TJCe, 1999. Maintenance of C. Elegans, 2, pp. 51–67.
Tang, B., Tong, P., Xue, K.S., Williams, P.L., Wang, J.S., Tang, L., 2019. High-throughput
assessment of toxic effects of metal mixtures of cadmium(Cd), lead(Pb), and
manganese(Mn) in nematode Caenorhabditis elegans. Chemosphere 234, 232–241.
Tang, B., Williams, P.L., Xue, K.S., Wang, J.-S., Tang, L., 2020. Detoxication
mechanisms of nickel sulfate in nematode Caenorhabditis elegans. Chemosphere
260, 127627.
Tang, B., Xue, K.S., Wang, J.-S., Williams, P.L., Tang, L., 2023. Host-microbiota affects
the toxicity of aatoxin B1 in Caenorhabditis elegans. Food Chem. Toxicol. 176,
113804.
Tauffenberger, A., Fiumelli, H., Almustafa, S., PJJCd, Magistretti, 2019. Lactate and
pyruvate promote oxidative stress resistance through hormetic ROS signaling. Cell
Death Dis. 10, 1–16.
Thomason, L.C., Costantino, N., DLJCpimb, Court, 2007. E. Coli Genome Manipulation
by P1 Transduction, 79, 1.17. 1-1.17. 8.
Wang, H., Chu, W., Das, S.K., Ren, Q., Hasstedt, S.J., Elbein, S.C.J.D., 2002. Liver
Pyruvate Kinase Polymorphisms are Associated With Type 2 Diabetes in Northern
European Caucasians, 51, pp. 2861–2865.
Wang, J., Tang, L., Glenn, T.C., Wang, J.S., 2016. Aatoxin B1 induced compositional
changes in gut microbial communities of male F344 rats. Toxicol. Sci. 150, 54–63.
Williams, J.H., Phillips, T.D., Jolly, P.E., Stiles, J.K., Jolly, C.M., Aggarwal, D., 2004.
Human aatoxicosis in developing countries: a review of toxicology, exposure,
potential health consequences, and interventions. Am. J. Clin. Nutr. 80, 1106–1122.
Williams, P.L., Dusenbery, D.B., 1990. Aquatic toxicity testing using the nematode,
Caenorhabditis elegans. Environ. Toxicol. Chem. 9, 1285–1290.
Zhang, J., Holdorf, A.D., Walhout, A.J.M., 2017. C. elegans and its bacterial diet as a
model for systems-level understanding of host–microbiota interactions. Curr. Opin.
Biotechnol. 46, 74–80.
Zhou, J., Tang, L., Wang, J., Wang, J.S., 2018. Aatoxin B1 disrupts gut-microbial
metabolisms of short-chain fatty acids, long-chain fatty acids, and bile acids in male
F344 rats. Toxicol. Sci. 164, 453–464.
B. Tang et al.