User login

We offer our registered users tailored information, free online courses and exclusive content.

You have an old EXCEMED account ...

Our platform has been renewed. All users registered at any of the old websites are kindly requested to reset their password. Why is this?

... or you lost your password?

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.

Perspective interview - Søren Gersting

Perspective interview - Søren Gersting

PKU Academy Newsletter: Fourth Issue Feature interview with Dr Søren Gersting

Søren Gersting Talks About the Impact of Bioinformatics on PKU Research

Dr Søren Gersting trained as an MD but changed direction into basic research. Following the award of his PhD in 2003 he joined the Molecular Pediatrics group at the Dr von Hauner Children's Hospital, Ludwig-Maximilians-University, Munich (Germany). Here, his initial research focus was the mechanism of BH4 interaction with the PAH protein. He has published a number of papers on the mechanisms of action of cofactors that ameliorate the effects of missense mutations. Who better then, to comment on the ‘Zeitgeist’, bioinformatics?

New tools for old data

Dr Gersting told us ‘The potential in bioinformatics lies in the fact that it’s not so complicated to apply if you know what to do. Using these methods we can get more information out of the wealth of data that is already there.

‘We have so many studies on basic research in PKU over the last 10 years. These studies provide data from both basic research on the protein itself and clinical research on blood Phe levels, neurocognitive function, etc. All this can be brought together.

‘There are so many intertwined factors that influence patient outcome. Using bioinformatics, you can investigate the interplay between a number of these different factors, such as stress, metabolic state and BH4 supply, in different patients with different genotypes and mutations. The true relationship between multiple interacting factors that influence outcome can easily be obscured when only a single factor is studied.’

Blind dataset?

Dr Gersting continued ‘Bioinformatics is a marvelous tool for finding potentially meaningful correlations in vast datasets but, in order to function, such analyses need a lot of data. These data need to be accurate and accessible.’

Given the need for data, is enough material being made available? ‘There are some databases already available’ Dr Gersting said. ‘There’s the PAH database founded by Dr Charles Scrive1 , the BIOPKU  database curated by Prof Nenad Blau2 and RAMEDIS by Prof Fritz Trefz3, just to mention three of them. Here we find data on mutations, on the clinical phenotypes, on the BH4 supply and so on.

‘What is lacking at the moment is a database that brings together the basic research information on the mutations, so when variant protein is expressed you can find out, is it more unstable? Is it aggregation-prone? What is the affinity to the substrate or to the cofactor? We have tried to bring this together in our current study but a larger database is not yet available. Creating such a database, that could allow these data to be combined with the clinical databases, would be a step forward for the future.’

Towards individualized treatment

How, we asked, do advances in bioinformatics impact on patient confidentiality. ‘It’s true that individualized medicine requires us to treat the whole patient rather than specific mutations. With many other factors to consider, it is not enough to just know about the genotype, which could be non-personalized. We need more personal information about the patients. Capturing these data, gathered through newborn screening programs, for example, or adding in the available data on siblings or parents certainly raises issues about confidentiality. I think this is something the research community has to deal with because, although there is risk, there is so much we could potentially take out of it.’

Where next?

How is the greater understanding of how BH4 works as a pharmacological chaperone influencing research and the development of potential medicines for other IEM?

Dr Gersting said ‘Treatment with exogenous BH4 is now, I would say, proven to stabilize the PAH protein. This binding-induced stabilization has now been shown for other diseases. We have seen clearly that it is missense mutations that induce protein misfolding. If we keep in mind that about 60% of all mutations in IEM are missense, we can imagine how many patients could benefit from a pharmacological chaperone treatment. This clearly points the direction for further research.

‘One example would be the medium-chain acyl-Coenzyme A dehydrogenase (MCAD) deficiency, the deficiency in the metabolism of medium chain acyl-CoAs; another example is glutaryl-CoA dehydrogenase, which is related to the MCAD protein but behaves completely differently. In our group we are already looking at pharmacological chaperone therapy for glutaric aciduria type 1 patients.’

To what extent will bioinformatics play a role in the development of further treatments for PKU? ‘Historically, people tended to think of PKU as a simple Mendelian disorder but it is not’ said Dr Gersting. ‘It is a complex trait because every patient has his or her individual molecular phenotype. If you analyze a lot of parameters you will find that most mutations are unique in their behavior. Bioinformatics approaches will allow us to find out where there are similarities and where there are differences.

‘This approach can show us how patients can be grouped, and based on new sub-groups we might identify new biomarkers. This is an important step to creating the next generation of drugs beyond BH4.’

And what of the future?

‘When we think about basic research in PKU concerning protein misfolding and pharmacological chaperone treatment, one could have the misunderstanding that we went very far and that we know everything’ said Dr Gersting.

‘But, to be precise, there are more than 600 mutations known and only maybe 30 have been analyzed in depth – and to think that analyzing less than 10% is enough to provide an understanding of the whole picture might be misleading. I hope that the research community in PKU will continue to do this work as they have been for the last decade. This will give us expanded datasets and allow better bioinformatics.’


1 - Dr Charles R Scriver (Biochemical Genetics Unit, McGill University, Montreal, Canada)

2 - Prof Nenad Blau (Division of Inborn Metabolic Diseases, University Children’s Hospital, Department of General Pediatrics, Heidelberg, Germany)

3 - Prof Friedrich Trefz (Medical Centre Gammertingen, Kreiskliniken Reutlingen GmbH, Gammertingen, Germany)

Søren Gersting

Department of Molecural Pediatrics
Munich, Germany