Til projektoversigt

Strengthening the Foundation for Treating Mental Illness

Semper Ardens forskningsprojekt | 16/06/2016

The serotonergic system in the brain originates from a small region in the hind brain denoted the raphe nucleus that makes projections and release serotonin (5-hydroxytryptamine; 5-HT) into almost all other brain regions. This system controls essential brain functions such as control of sleep, circadian rhythms, feeding, memory, and emotional state. Imbalances in serotonergic signalling have been implicated in several mental illnesses. Compared to most other neurotransmitter systems, serotonergic synapses are highly complex in terms of the number and variety of key serotonin signalling proteins (receptors and transporters) present across different types of serotonergic neurons. This complexity allows different serotonergic synapses to exert distinct effects on the nearby neurons. Serotonin itself was discovered in the 1930s; however, it was not until the late 1990s that the final human serotonin receptor and transporter proteins were identified. Considering the rather recent discovery of several of these key components, it is not surprising that there are many aspects of serotonergic signalling the scientific community do not understand yet. In the bio-modelling group at Aarhus University we apply computational tools to investigate the serotonin signalling process in collaboration with biochemists and neuro-pharmacologists at Copenhagen University. We model the key proteins involved in serotonergic signalling and assess their interactions with small molecules such as drugs in order to find out how they work and how this can be exploited in future drug design. 

Serotonin – an Important Messenger in the Human Brain

The serotonergic synapse comprises a complex set of proteins such as receptors and transporters. Traditionally, antidepressant drugs target only the transport proteins, but it is believed that drugs that also act on a subset of the receptors will decrease side effects and increase the antidepressant effect.

Serotonin, also called  5-HT (5-hydroxytryptamine), is a small molecule with a big impact. It is one of the major neurotransmitters in the human brain where it controls core functions such as appetite, sleep, memory, mood, and behaviour (1). Considering the importance of these functions, it is not surprising that imbalance in serotonergic signalling is linked to a wide range of disorders including depression, anxiety and migraines (Figure 1) (2). In principle, it is possible to treat such disorders using pharmaceutical drugs that restore the serotonin balance in the brain, but given the enormous complexity of signalling in the brain this has proven difficult. Despite of this several pharmaceutical drugs have been developed and are now successfully applied to patients worldwide. An example of such drugs are antidepressant drugs, the majority of which work by blocking a transport protein that would normally move serotonin into neurons, and the drugs hereby increase the level of serotonin outside the neuron. Despite the successes in treatment of depression there is still great room for improvement. Many patients are not responding to current treatment and others suffer side effects from their treatment. It is therefore of utmost importance that we continue to gain knowledge about the details of serotonergic signalling in order to continue to improve the lives of people suffering from mental illness.

“Signalling in the serotonergic synapse is fascinatingly complex and working with the elucidation of the underlying mechanisms is highly rewarding,” says Lucy Kate Ladefoged, PhD-student, Aarhus University.

For example, to avoid side effects, it is generally preferable to develop drugs that only bind to the intended target in the body. For example, a drug targeting serotonin binding to the transport protein important for the treatment of depression cannot target other serotonin binding proteins. For the serotonergic system, this is particular difficult because of the number of highly similar molecular components involved in serotonergic signalling and their widely different roles for normal function. However, designing such drugs requires a detailed knowledge of both the proteins that the drug should interact with and the proteins that it should not interfere with. Currently, we are lacking a complete understanding of what governs the binding of different drugs to the different proteins involved in serotonin signalling. We set out to provide this much sought-after knowledge for the central protein components of the serotonergic synapse, serotonin receptors and transporters (Figure 2).

Figure 1 The neurotransmitter serotonin is responsible for a wide array of normal physiological functions, but is also implicated in many diseases. Photo: Lars Kruse, Aarhus Universitet.

The Machinery Behind Serotonin Signalling

An estimated 350 million people suffer from depression worldwide and depression is considered the leading cause of disability in the world today according to the World Health Organisation (WHO, 3).

Communication between neurons occurs at the synapse through neurotransmitters which, when prompted, will be released into the synaptic space (Figure 2). While the neurotransmitter is in the synaptic space it is free to interact with different proteins in the cell membranes of nearby neurons as well as the neuron from where it was released. The principle group of proteins that serotonin interacts with which drives serotonergic signalling are serotonin receptors. These can be considered as tiny signalling machines that turn on once serotonin binds and relays a signal into the neuron. There are seven distinct types of serotonin receptors, named 5-HT1-7. Furthermore, each receptor type may have several subtypes. For example, the 5-HT2 family constitute the three unique subtypes 5-HT2A, 5-HT2B, and 5-HT2C. The different types and subtypes of receptors differ in their downstream effects in the cell environment, and it is this diversity that provides the ability for serotonin to control such a wide variety of physiological functions. 

Figure 2 The three principal components of the 5-HT system, the 5-HT3 Cys-loop receptors (red), 5-HT GPCRs (blue), and SERT (green), are localised both pre- and postsynaptically to the 5-HT neuron and on glial cells. Specific components of the 5-HT system is critically involved in a broad range of CNS disorders, such as anxiety and depression (SERT and 5-HT1A-C,2A-C,4,6,7), addiction (5-HT1A-B,2A,2C,3), migraine and nausea (5-HT1A,3), learning and memory (5-HT1A-B,2A,3,4,6,7), and obesity (5-HT1A,2A-C,4).     

All of the serotonin receptors, except for the 5-HT3 type, belong to the family of so-called G-protein coupled receptors (GPCRs) which communicate directly with other proteins within the cell. These types of receptors all share a similar 3D structure and have a single serotonin binding site. On the other hand, the 5-HT3 receptor, which is able to form ion channels in the membrane, is part of the so-called Cys-loop receptor family, and has a completely different structure compared to the GPCRs. This receptor consists of five proteins that have joined together and collectively form a channel. Another important protein in the serotonergic synapse is the serotonin transporter. This protein is responsible for the re-uptake of serotonin back into the presynaptic neuron thereby controlling the duration of serotonin signalling. The overall structure of this protein is different from the structure of both types of receptors (Figure 3). Despite the widely different overall structure and function, these proteins all have the ability to recognise and bind serotonin in common. Understanding the molecular similarities and differences of these serotonin binding pockets that can be found across the synapse is central to designing drugs that selectively target only specific components of the synapse.

“Making models of complex chemical systems is attractive since we can really predict how complicated processes occur in life, hereby providing links from atomic to anatomic levels of understanding,” says Birgit Schiøtt, Professor, Aarhus University.

Figure 3 Cartoon models of the three protein classes found in the serotonergic synapse; Cys-loop ion channel, G-protein coupled receptor and transport protein. Serotonin is shown in yellow in all proteins.

Combining Virtual and Real-Life Experiments

In our research lab, we use computers to make 3D models of proteins. Such models can be used to make very detailed predictions on how the proteins function as well as how drugs and other small molecules interact with them. All models have shortcomings and there are limitations to what a model can describe and the reliability of the predictions must thus always be checked against all available data. Therefore, we typically use our models to form hypotheses and then go into the lab and perform experiments on the protein in question that can confirm or reject the hypothesis. In other cases, the experiments themselves lead to hypotheses, which we can then test in our models (Figure 4). This way, we get the combined benefit of the detailed information from the virtual models and the reliability of the cell-based experiments. In this project, supported by the Carlsberg Foundation, we are constructing models of all of the serotonin receptor subtypes that are currently pursued as drug targets in depression, as well as a model of the serotonin transporter. The models are challenged by our collaborators at Copenhagen University. Using our models, we can determine the similarities and differences between the serotonin- and drug binding sites in these different targets, as well as how different types of molecules bind to the protein binding sites. Thus, we obtain a fingerprint for each substrate binding pocket across the serotonergic synapse that uniquely describes its properties. Knowing the layout of the binding sites makes it possible for researchers to rationally develop new molecules that can bind selectively to a single or a few subtypes of serotonin receptors and thus make highly effective drugs with very few side effects.

Figure 4 A cartoon representation of the never-ending workflow employed in our collaborative research. Illustration made by Johan Jarnestad, The Royal Swedish Academy of Sciences.

Societal Impact and Scientific Social Responsibility

According to the Worlds Health Organisation (WHO), 350 million people are estimated to suffer from depression and depression is the leading cause for disability in the world (3). It is estimated that the global cost of treating mental illness will reach 6 trillion US dollars by 2030 (4) and it is thus crucial that better treatment options emerge. Commonly, depression and some forms of anxiety are treated by the administration of drugs that interact with proteins in the serotonergic system with the main target being the serotonin transporter. Due to a lack of knowledge about several of the proteins in this system many of the drugs used today are also interacting with other non-target proteins and use of the drugs is thus accompanied by side effects. On the contrary, the recently introduced multimodal antidepressants vilazodone and vortioxetine rely on interactions with multiple serotonergic targets in a specific manner. Clearly, there is a great need for expanding our knowledge about the serotonergic nervous system if we as a society want to be able to help and improve the lives of the millions of people who suffer from mental illness as well as decrease the economic burden of treatment. It is therefore our aim to help in closing these gaps in our knowledge and form the basis for the rational design of next generation antidepressant drugs. Our continued research in this area has been made possible by a Semper Ardens Research Grant granted by the Carlsberg Foundation and has resulted in several new scientific articles (5-7).

“Collaborating with brilliant academic groups on fundamental questions related to drug discovery is fun and important for generating new ideas and eventually projects,” says Benny Bang-Andersen, Senior Director at H. Lundbeck A/S, who also participates in the project with access to data.


(1) Pytliak, M. et al, Physiol Res, 2010, 60, 15 [link]

(2) Mohammad-Zadeh, L. F. et al, J Vet Pharmacol Therap, 2008, 31, 187 [link]

(3) WHO fact sheet 2016 [link]

(4) The Global Economic Burden of Non-communicable Diseases (page 27) [link]

(5) Grouleff, J. et al, Front Pharmacol, 2015, 6, 235 [link]

(6) Koldsø, H. et al, Front Pharmacol, 2015, 6, 208 [link]

(7) Andersen, J. et al, ACS Chem Neuroscience, 2015, 6, 1892 [link]

The researchers from Aarhus University and University of Copenhagen meeting with researchers from H. Lundbeck A/S in August 2014 to discuss the project.
(Left to right: Lucy Kate Ladefoged (PhD-student, AU), Anne Laustsen (PhD-student AU), Lachlan Monro (PhD-student UC), Birgit Schiøtt (Professor, AU, project leader), Julie Grouleff (postdoc, AU) Lena Tagmose (Lundbeck), Ana Negri (Lundbeck), Jacob Andersen (postdoc UC), Benny Bang-Andersen (Lundbeck), Anders Skov Kristensen (Associate Professor, UC, co-PI)