exposure (100X:P= 0.0006; 1000X:P=
0.0033) (Figs. 1C and 5D and table S6).
To confirm the impact of the mixture on
the thyroid system in an additional toxicol-
ogy-relevant in vivo model, we assessed the
expression of TH-related genes in zebrafish
larvae (Danio rerio). Zebrafish embryos were
exposed to MIX N (0.01X to 100X) for 48 hours
(Fig. 5F), and gene expression was determined
by RT-qPCR. Exposure to the mix led to signi-
ficantly decreased expression of the thyroid
hormone receptorsthra(P= 0.0006) andthrb
(P= 0.0010) at 100X concentration (Figs. 1C and
5G and table S6). To investigate whether en-
docrine systems other than the thyroid (estrogen,
androgen, and glucocorticoid) were affectedby the mixture in vivo, we probed a panel of
candidate genes (table S6) and found them
to be unaffected.
By analogy with the in vitro models, single-
compound exposure to BPA was performed in
the in vivo systems (fig. S6, A to H). InXenopus
embryos, the XETA test showed a stronger ef-
fect with MIX N versus BPA (10X versus 1000X,Caporaleet al.,Science 375 , eabe8244 (2022) 18 February 2022 8 of 15
Fig. 5. MIX N exposure disrupts TH
signaling and behavior both inX. laevis
andD. rerio.(A) The experimental setup for
X. laevis.(B) Thyroid-disrupting effects were
assessed using XETA. Representative images
of tadpoles exposed to T3 or T3 and 1000X
MIX N are shown. Scale bar is 500mm.
(C) Specific fluorescence in tadpole heads
was quantified as relative fluorescence units
normalized against T3. A pool of three
independent experiments with >11 tadpoles per
concentration per experiment (mean ± SEM)
was analyzed. #### indicatesP< 0.0001 for
T3 versus control using StudentÕsttest.
(D) RT-qPCR after brain dissection for TH
early-response target genes (oatp1c1,klf9, and
mct8) in wild-type tadpoles exposed to the
protocol shown in (A). Shown is a pool of three
to nine independent experiments (mean ±
mininum or maximum response with two to
five replicates per concentration per experi-
ment; foroatp1c1andklf9, a pool of nine
experiments for the concentrations 1X, 10X,
and 100X and a pool of six experiments for
1000X, with each exposure compared with the
appropriate control; formct8, a pool of six
experiments for 1X, 10X, and 100X and a pool
of three experiments for 1000X, with each
exposure compared with the appropriate con-
trol). (E) Mobility of exposed tadpoles was
measured for 10 min in alternating 30-s light
and 30-s dark cycles. Distance traveled was
analyzed over time. A pool of five to seven
independent experiments is shown (mean ±
SEM, 9 to 14 tadpoles per concentration per
experiment, with a pool of seven experiments
for 1X, 10X, and 100X and a pool of five
experiments for 1000X). (F) The experimental
setup forD. rerio.(G)D. rerioembryos were
exposed to MIX N [protocol shown in (F)], and
RT-qPCR was performed on pooled whole
larvae for genes that included thyroid receptor
(thrb). Shown is a pool of three independent
experiments (three replicates per concentra-
tion per experiment). (H) The mobility of
exposed zebrafish embryos was measured for
40 min at 120 hours postfertilization in
alternating 5-min dark and 5-min light cycles.
Distance traveled was analyzed over time.
Shown is a pool of three independent
experiments as mean ± SEM, with six to eight
larvae per concentration per plate and three
plates per experiment. For box plots in (D) and (G), boxes and whiskers represent the interquartile range and minimum and maximum values, respectively.
RESEARCH | RESEARCH ARTICLE