Possible link between autism and antidepressants use during pregnancy

Exposure to antidepressants in the womb leads to autistic-like behaviour in full-grown offspring in an animal model, according to a new study led by Duke-NUS Medical School.

Representative heat map of results from the three-chamber social interaction test with a familiar mouse and novel mouse. Control animals show preference for new mice, but antidepressant-treated animals show no preference for familiar or new mice.

SINGAPORE, 30 April 2019 – An international team led by Duke-NUS Medical School has found
a potential link between autistic-like behaviour in adult mice and exposure to a common
antidepressant in the womb. They also identified a treatment that helped improve memory loss
and social interactions, according to the new study published in the journal Molecular Brain.

Antidepressants are commonly prescribed for treating major depression and post-traumatic
stress disorder, including in pregnant women. One of the most commonly prescribed
antidepressants is fluoxetine, a serotonin reuptake inhibitor. Fluoxetine can cross the placenta
and is also detected in breast milk. Little is known about its safety during pregnancy, and not
enough studies have been conducted on its long-term effects on offspring.

“Many human association studies have been conducted to investigate connections between
antidepressant exposure during pregnancy and children with autism and attention deficit disorder (ADHD). But they have not been able to pinpoint a causal relationship,” stated Associate
Professor Hyunsoo Shawn Je, from Duke-NUS’ Neuroscience and Behavioural Disorders (NBD)
Programme, a senior and corresponding author of the study.

The team from Duke-NUS and their collaborators in South Korea and Singapore investigated
adult mice born to mothers treated with fluoxetine (sold under the brand names Prozac and
Sarafem) over a 15-day time period that corresponds to the second trimester in humans, in
comparison with those born to mothers given normal saline as controls. They found key
differences in behaviour. For example, the unexposed mice normally explored all three arms of a
Y-shaped maze over a ten-minute time period and, over the courses of multiple arm entries, mice usually enter a less recently visited arm, while the fluoxetine-exposed ones were less inclined to
explore unvisited arm.

In a second experiment, the mice were introduced to two juvenile mice, one after the other. When
the second new mouse was introduced, mice that were not exposed to fluoxetine were more
likely to only sniff the newly introduced mouse, recognizing that they had already met the first
mouse. But the fluoxetine-exposed group sniffed both mice, indicating that they had impaired
social novelty recognition.

The team then examined nerve signal transmission in the prefrontal cortex, a part of the brain
involved in moderating social behaviour. They found impaired transmission caused by an overactive serotonin receptor.
Treating fluoxetine-exposed mice with a compound that blocks the
receptor alleviated their behavioural problems and improved their working memory.

The team next wants to examine autistic children born to mothers treated with antidepressants
using positron emission tomography (PET) scans, an imaging technique used to observe
metabolic processes in the body. If they also show enhanced serotonin receptor activity in the
same area of the brain, the team plans to test whether FDA-approved serotonin receptor
blockers can normalize their behaviours.

“The consensus among experts is that the rise in the number of people diagnosed with autism
around the world is likely due to more awareness and testing rather than an increase in the
prevalence of autism,” noted Professor Patrick Casey, Senior Vice Dean for Research at DukeNUS.
“This collaborative study by our researchers offers a compelling case for a link between
autism and antidepressant exposure in the womb in an animal model, and a possible mechanism
that could potentially be exploited for future therapies.”

About Duke-NUS Medical School
Duke-NUS is a partnership between Duke University School of Medicine and the National
University of Singapore (NUS).

In 2005, with support from the Singapore government, NUS and Duke University, two academic
institutions with strong track records in research and education, committed to combine the unique
medical education curriculum at Duke University School of Medicine with the academic rigour
and rich resources offered by NUS, and to offer students an enriching and innovative medical
educational experience.

Duke-NUS is located on the main campus of the largest healthcare group in the country,
Singapore Health Services (SingHealth). This group collectively delivers multi-disciplinary care
among 42 clinical specialties across a large network of hospitals, national specialty centres and
polyclinics. Together, Duke-NUS and SingHealth constitute a leading, world class Academic
Medical Centre embodying the goal of delivering the highest levels of patient care, education and

For more information, please visit the website (link below).

For media enquiries, please contact:
Federico Graciano
Duke-NUS Medical School
Tel: +65 6601 3272
Email: [email protected]

Published: 30 Apr 2019

Contact details:

Duke-NUS Communications

Duke-NUS Medical School Singapore 8 College Road, Level 6 Singapore 169857

+65 6516 2585
News topics: 
Academic discipline: 
Content type: 

Reference/Note: Yu W, Yen YC, Lee YH, Tan S, Xiao Y, Lokman H, Ting AKT, Ganegala H, Kwon T, Ho WK and Je HS (2019). Prenatal selective serotonin reuptake inhibitor (SSRI) exposure induces working memory and social recognition deficits by disrupting inhibitory synaptic networks in male mice. Molecular Brain 2019 12:29. DOI: 10.1186/s13041-019-0452-5