Science - USA (2022-02-18)

RESEARCH ARTICLE SUMMARY

◥

ENVIRONMENTAL TOXINS

From cohorts to molecules: Adverse impacts

of endocrine disrupting mixtures

Nicolò Caporale†, Michelle Leemans†, Lina Birgersson†, Pierre-Luc Germain†, Cristina Cheroni†,
Gábor Borbély, Elin Engdahl, Christian Lindh, Raul Bardini Bressan, Francesca Cavallo,
Nadav Even Chorev, Giuseppe Alessandro DÕAgostino, Steven M. Pollard, Marco Tullio Rigoli,
Erika Tenderini, Alejandro Lopez Tobon, Sebastiano Trattaro, Flavia Troglio, Matteo Zanella,
Åke Bergman, Pauliina Damdimopoulou, Maria Jönsson, Wieland Kiess, Efthymia Kitraki,
Hannu Kiviranta, Eewa Nånberg, Mattias Öberg, Panu Rantakokko, Christina Rudén, Olle Söder,
Carl-Gustaf Bornehag‡, Barbara Demeneix‡, Jean-Baptiste Fini‡, Chris Gennings‡, Joëlle Rüegg‡,
Joachim Sturve‡, Giuseppe Testa‡

INTRODUCTION:Endocrine disrupting chemicals
(EDCs) are compounds that interfere with
physiological hormonal regulation. Humans
are pervasively exposed to many different EDCs,
and a growing body of evidence indicates that
early life exposure to such EDC mixtures can
induce changes in the human organism that
underlie increased susceptibility to diseases
throughout the life span, including neuro-
developmental disorders. Chemical regulation
is, however, entirely based on the risk assess-
ment of individual compounds, leaving the real-
life impact of chemical mixtures unexamined
and unregulated. This is relevant insofar as

cumulative exposure to multiple compounds

may be associated with adverse health outcomes

even when the concentrations of individual

chemicals fall below the regulatory dose.

RATIONALE:We set out to make the epidemi-

ological associations between exposure to

mixtures and health outcomes experimentally

tractable, defining molecular pathways and

dose responses that could be translated back

to actual human exposures and thereby refine

current risk assessment practices. As opposed

to previous studies that focused on single com-

pounds, we identified and tested an EDC mix-

ture associated with adverse neurodevelopmental

outcomes in the Swedish Environmental

Longitudinal, Mother and child, Asthma and

allergy (SELMA) pregnancy cohort (MIX N) by

integrating epidemiological data with experi-

mental toxicology and characterized real life–

relevant exposure.

RESULTS:Weusedweightedquantilesum(WQS)

regression to identify chemicals associated with

language delay in children and included those

chemicals in MIX N. MIX N was synthesized

following the relative proportions and total con-

centrations found in the SELMA cohort. It was

then tested in both in vitro and in vivo models.

In human fetal primary neural stem cells and

three-dimensional cortical brain organoids dif-

ferentiated from human pluripotent stem cells,

transcriptomic analysis showed that MIX N

interferes with hormonal pathways and dysregu-

lates expression of genes and biological path-

ways that are causally linked to autism spectrum

disorders. Data from experiments inXenopus

leavisandDanio rerio, in vivo models validated

by the Organisation for Economic Co-operation

and Development (OECD), confirmed thyroid

function as one of the key and unifying points

of vulnerability to MIX N and linked thyroid

disruption to neurodevelopmental effects mea-

sured as alterations in locomotor activity. The

resulting dose-response relationships were then

used to estimate a point of departure (POD),

which is the toxicological measure to estimate

no-effect concentration. This enabled us to apply

a similar mixture approach (SMACH) where

we (i) identified individuals in the SELMA

study who were sufficiently similarly exposed

compared with the experimental mixtures and

(ii) determined the proportion of the SELMA

children with exposure ranges of concern using

the POD as reference.

CONCLUSION:Integrating experimental and

epidemiological evidence, we established me-

chanistic and correlative evidence for neuro-

developmental adversities of an EDC mixture

associated with language delay. Using the gener-

ated experimental data in a risk assessment con-

cept, we found increased odds of language delay

in offspring of up to 54% of pregnant women.

These results emphasize the need to take mix-

tures into account during chemical testing and

risk assessment and provide an integrative frame-

work to guide risk assessment strategies.▪

RESEARCH

SCIENCEscience.org 18 FEBRUARY 2022•VOL 375 ISSUE 6582 735

The list of author affiliations is available in the full article online.

*Corresponding author. Email: [email protected]

(C.-G.B.); [email protected] (B.D.); joelle.ruegg@ebc.

uu.se (J.R.); [email protected] (G.T.)

†These authors contributed equally to this work.

‡These authors contributed equally to this work.

Cite this article as N. Caporaleet al.,Science 375 ,

eabe8244 (2022). DOI: 10.1126/science.abe8244

READ THE FULL ARTICLE AT

https://doi.org/10.1126/science.abe8244

Epidemiology Biostatistics Chemistry

Similar mixture approach

Brain organoids

Fetal progenitors

Danio rerio

Xenopus laevis

Sufficient similarity

MIX N

Weighted quantile

sum regression

SELMA cohort

EDC levels in urine,

blood and clinical data

Identification of EDCs of concern EDC mixture and synthesis

Identification of

molecular

mechanisms of action

Dose-response modeling

for benchmark dose

estimation

Determination of the human

population with exposure ranges

of concern

Chemical 1

Chemical 2

Mixture 2

Reference mixture

Mixture 1

Experimental biology

Phthalates

PFAS

Bisphenol-A

0 0.05 0.10 0.15 0.20 0.25

Diagram showing the integrative framework of the study.A mixture of EDCs was associated with adverse
neurodevelopmental outcomes in the SELMA pregnancy cohort and was tested in human in vitro and in in vivo
models to elucidate the molecular and functional impact of exposure. Experimental data were finally referred
back to the cohort for risk assessment by a similar mixture approach. PFAS, perfluoroalkyl substance.