Cover image for Highlights in Chemical Biology

Highlights in Chemical Biology

Chemical biology news from across RSC Publishing.



Interview: Quick on the uptake


14 October 2008

 

Douglas Kell tells Elinor Richards about his findings on drug uptake and the implications for drug discovery and development                

Douglas Kell

Douglas Kell is director of the Manchester Centre for Integrative Systems Biology at the University of Manchester. His work focuses on systems biology, analytical science and computational biology. He is the recipient of numerous awards, including the RSC's Interdisciplinary Award, and is the current holder of the RSC Chemical Biology Award and the SAC/RSC Gold Medal. As of 1 October, he is the Chief Executive of the UK BBSRC.

 

Can you explain about your role and the aims of the Manchester Centre for Integrative Systems Biology? 

We have a team of eight principal investigators who help manage the work of the centre, and I am the director. We are developing a pipeline that covers all areas of systems biology, including purifying (usually recombinant) biomolecules, assessing both qualitatively and quantitatively what they interact with and in effect doing quantitative enzyme kinetics with them. 

"We are developing a pipeline that covers all areas of systems biology, including purifying biomolecules, assessing both qualitatively and quantitatively what they interact with and in effect doing quantitative enzyme kinetics"
This gives us parameters such as Km and kcat for each component of the network. We can then use modelling software to turn the individual reaction kinetics into a system of ordinary differential equations that describe the time-course of evolution of the system variables, usually metabolite fluxes and concentrations. We then compare this with measurements made experimentally, including those of protein concentrations, and adjust the model to try to make it fit the reality. When the model begins to look sound, we can analyse the model for its properties, and make predictions that can be further tested. All the data and models are stored in online databases according to recognised standards that allow integration and interoperability. 

Our 'set-piece' activity is to be the production of a genome-wide kinetic model of the yeast metabolic pathways that accounts for the bulk of the flux of substrates. If we can show that our methods work in yeast, we can then apply them to other systems, including humans.

In addition to the MCISB project, I have a personal research group of 25 people working on problems with systems biology, analytical science and computational biology. 

What are you currently working on? 

About 80% of my group is 'dry' and works on computational modelling and data analysis, with a lot of the 'wet' (experimental) work focused on the development of novel analytical methods for biomolecules. For instance, one major project, in collaboration with GlaxoSmithKline and AstraZeneca, is a mixed wet-and-dry project that concerns the analysis of human serum samples to understand their complement of metabolites. We have developed novel methods that can detect, and in many cases quantify, thousands of these metabolites, and these can be useful in diagnosing various diseases. We have recently published work on molecules diagnostic of heart failure and on pre-eclampsia, a major disease of pregnancy. The challenge is to integrate the measurements obtained with systems biology models.

"We have developed novel methods that can detect, and in many cases quantify, thousands of these metabolites, and these can be useful in diagnosing various diseases"

What fascinated you about this area of research?

When I finished my DPhil in 1978, DNA sequencing had just been established, albeit at a rate of just three bases per day, and this was evidently going to usher in a new phase of the molecular biology agenda. I never felt that this would directly tell me how cells worked, and I went into other areas such as metabolic control analysis that subsequently morphed into systems biology. This happened in particular following the systematic genome sequencing programmes, where it was found that classical methods had missed the existence, let alone the function, of about half of the genes in even well-worked organisms such as baker's yeast and Escherichia coli. It was and is clear that to understand a system, one needed to understand, quantitatively, its internal interactions more than just the qualitative list of its components. This is a kind of engineering-type approach.

Similarly, metabolism is the part of biology that is closest to the phenotype, and all the present concerns about diet and health are playing out on a background in which the genotype has not changed at all. So, just as many were attracted to chemistry through cookery, I always felt that small molecule metabolism, an obvious branch of physical organic chemistry, was where I would wish to focus. 
 
Can you tell us about your work on drug uptake? How do your findings change the current understanding about drug uptake into cells?

We have a taste for iconoclastic areas of research, and this is one such area. According to a widespread assumption, the ease with which drugs diffuse ('leak') into cells across the cell membrane is a function of their lipophilicity, measured as the octanol : water partition coefficient log P. It is also recognised as an 'exception' that drugs may occasionally hitchhike on proteinaceous transporter molecules that normally move endogenous metabolites about. Actually, the evidence for the lipophilicity story is at best circumstantial, and we have adduced evidence that drug uptake into cells is almost certainly the rule and not the exception. The challenge then is to find out which drugs use which transporters; that changes this from a problem of biophysics (lipophilicity) to one of mechanistic systems biology. 

We have also been engaged in an experimental program, with baker's yeast, that has developed methods for detecting which carriers use particular drugs.

What are the implications for drug discovery and development, and chemical genetics?

"By contrast, the systems biology approach offers the opportunity to understand, and to model precisely, the expected distributions of drugs based on the distributions of the transporter molecules for which they are the substrates"
The implications are profound, because much of the design of libraries of candidate substances that become drugs has been predicated on estimates or measurements of log P. This is despite the fact that only about half the marketed drugs fulfil the stated criteria. By contrast, the systems biology approach offers the opportunity to understand, and to model precisely, the expected distributions of drugs based on the distributions of the transporter molecules for which they are the substrates. This will help to account for both the lack of efficacy and the toxicity that are the main reasons for drugs failing before (and in some cases after) they reach the market. The same holds true for the methods of chemical genetics.

What benefits do collaborations bring to your research?

We have numerous collaborations, and in fact I can think of no research grant that I hold alone. The drug uptake project was done in collaboration with Professor Steve Oliver, now of the University of Cambridge, and was funded jointly by the Biotechnology and Biological Sciences Research Council and by GlaxoSmithKline. As well as a financial contribution from the latter, they provided tremendous expertise in pharmacokinetics that we simply did not have. The research laboratories of many industries are little different from academic laboratories in terms of their scope and focus, and these kinds of partnerships allow complementary expertise to flourish for the greater good.

You've received numerous awards for your work. What do these awards mean to you and to your research?
 
I think they represent a recognition that some things we have done have been of value, and thereby help promote some of our approaches among others in the community.

 

Related Links

Link icon Bioanalytical Sciences Group
Professor Douglas Kell's homepage


External links will open in a new browser window



Also of interest

Analysis of aptamer sequence activity relationships
Mark Platt, William Rowe, Joshua Knowles, Philip J. Day and Douglas B. Kell, Integr. Biol., 2009, 1, 116
DOI: 10.1039/b814892a

Proximate parameter tuning for biochemical networks with uncertain kinetic parameters
Stephen J. Wilkinson, Neil Benson and Douglas B. Kell, Mol. BioSyst., 2008, 4, 74
DOI: 10.1039/b707506e

Insights into the behaviour of systems biology models from dynamic sensitivity and identifiability analysis: a case study of an NF-B signalling pathway
Hong Yue, Martin Brown, Joshua Knowles, Hong Wang, David S. Broomhead and Douglas B. Kell, Mol. BioSyst., 2006, 2, 640
DOI: 10.1039/b609442b

Fast automatic registration of images using the phase of a complex wavelet transform: application to proteome gels
Andrew M. Woodward, Jem J. Rowland and Douglas B. Kell, Analyst, 2004, 129, 542
DOI: 10.1039/b403134b