Interview conducted by April Cashin-Garbutt, BA Hons (Cantab)
How did your research into stress in adolescence and mental illness originate?
As a psychiatrist I am interested in how stress may play a role in the pathology of adult mental disorders. You may know that most of the major mental illnesses, such as schizophrenia, mood-disorder, substance abuse or even anxiety disorder, become prominent in early adulthood. Consequently, most clinicians are interested in what happens in adolescence.
How does the brain develop during adolescence?
The basic neuronal architecture exists even in childhood but a dramatic re-organisation of the network occurs in adolescence. There are at least two major events. One event is that during adolescence a large amount of myelination occurs. The brain is made up of neurons. Efficiency of neuronal transmission is determined by myelination. Interestingly, before our paper, one paper published in Science last year indicated isolation stress may affect adult behaviour by affecting myelination.
For more information please read the following paper.
Neurodevelopmental mechanisms of schizophrenia: understanding disturbed postnatal brain maturation through neuregulin-1-ErbB4 and DISC1.
Jaaro-Peled H, Hayashi-Takagi A, Seshadri S, Kamiya A, Brandon NJ, Sawa A.
Trends Neurosci. 2009 Sep;32(9):485-95. doi: 10.1016/j.tins.2009.05.007. Epub 2009 Aug 26. Review.
What did your research involve?
Firstly we designed transgenic mice. We believed that isolation stress should be very important. Our team entered a collaboration with a Japanese investigator – Dr. Nabeshima. The isolation stress paradigm is very unique so I asked him to have co-correspondence, so that whenever unique questions came to us on the isolation stress paradigm, Nabeshima could answer them.
Why did you choose to carry out your study on mice?
Our focus was on how a stressful event in adolescence may affect adult brain function and potentially lead to dysfunction and disease. The research thus needed to cover many years – from adolescence to adulthood. If we wished to address this question via a clinical study we would have to design a longitudinal study. But longitudinal studies are very expensive and time consuming. So to have a proof of concept we instead used mice. This is because 20 human years corresponds to 3 months in a mouse.
If we wish to test a possible hypothesis that requires 20 years observation and maintenance of the people it would be very difficult, however, by using mice we can see such pathological trajectories in only 3 months.
What did your research find?
The most important point in this paper is that when we test a neurochemical or behavioural change in transgenic mice without any stressors we didn’t observe any robust change. Isolation stressors did not lead to a big behavioural change in wildtype mice, however, if we combine we observe a synergistic effect and then we see the phenotype.
First of all what our research found was a very interesting synergistic interaction of genetic risk factor and isolation stress in adolescence. Secondly, such adolescent stressors led to long-term behavioural changes in adults.
Do you think these results would translate in humans?
This is another challenge but we have several ways to do that. We used RU486 – a glucocorticoid receptor antagonist. This compound was used for mice to make a proof of concept in this study. However, this compound can be used for the clinical trial study.
We now have a very clear proof of concept. The next thing we need to do before we utilise this concept in clinical patient care is a clinical study. One possible way to do that is to provide the patient with psychotic depression with this compound RU486 and then measure before and after the change in dopamine by Positron Emission Tomography (PET). Then we can translate mouse observation to humans. This is under discussion with one of my close colleagues Dr. Smith at Johns Hopkins.
What impact do you think your research will have on the prevention and treatment of mental illnesses?
People always ask this question! Essentially we need time. The result from the preclinical model is very clean, however, to apply this to the clinical setting we need a clinical study by using PET scans and other clinical means. After that we may be able to apply this to the patient care, however, there are two important points that need to be made.
RU486 that has been used in this study has side effects. This compound could be used in the clinical trial if the patients consented, however, it may not be so positive to use this directly in daily clinic. By using mouse models we may be able to screen new drugs to find another compound with the same mechanism for the intervention but with much lower side effects.
This study indicates that adolescent isolation stress for somebody who may have a genetic risk is not so ideal. So we may be able to do some environmental adjustments. This may be the concrete way to utilise the study in the short run, but in the long run we need a clinical study to move all the findings to the patient care setting.
Does your work shed any light onto the “nature versus nurture” debate with regards to mental illnesses?
Cancer, diabetes etc. are all a combination of multiple genetic factors and multiple environmental factors – mental health is exactly the same. Usually people have a stigma to mental illness and people really emphasise the involvement of genetic factors in mental illness, but no single gene can cause mental illness, it is always a combination of multiple genetic and environmental factors. Even if somebody may have a genetic risk, there are many ways to prevent the onset of the disease.
How do you think the future of mental illness treatments will develop?
I am classified as a biological psychiatrist, however, I also respect non-biological assessment of patients with mental illness. Psychiatric disorders are purely brain disorders; however, the function impaired by such brain defects may have many indications in not only biological aspects but also psychological and social aspects.
We wish to understand mental illnesses from a biological viewpoint, however, from a treatment viewpoint I would like to emphasise multiple approaches. Not only a biological approach – mainly medication - but I believe that social and psychological intervention is very important in the treatment of patients with mental illness.
Are there any plans in place to look at the role of stress on other illnesses?
As I mentioned to you cancer, diabetes, heart disease are all caused by a combination of genetic factors and environmental factors. However, there has been no clear direct evidence in mouse models that show that genetic factors and environmental factors can synergistically cause a phenotype that is relevant to mental illness. This is the focus of our paper.
Where can readers find more information?
They can read our paper here: http://www.ncbi.nlm.nih.gov/pubmed/23329051
They can visit our Schizophrenia centre website:
About Professor Akira Sawa
Based on the training in both clinical psychiatry and basic molecular neuroscience, Dr. Sawa currently serves the director of Johns Hopkins Schizophrenia Center in which efforts on clinic, research, professional education, and public outreach towards better care and cure for patients with schizophrenia are taken place.
Dr. Sawa also leads P50 Silvo O. Conte center grant, which takes multifaceted translational approach on DISC1 interactome and schizophrenia.
In addition to these administrative responsibilities, Dr. Sawa's own research program aims translational study for major mental illnesses, such as schizophrenia and bipolar disorder, with a particular emphasis on molecular pathways of the diseases.