Snipping away at the causes of disease

Biotech '03, held recently in Helsinki, Finland, highlighted investigations into how SNP variations affect human health, and how this research has been helped by the Human Genome project.

Biotechnology will generate at least half of all new medicines by the end of the decade. So said Dr Kalevi Kurkijaervi of BioFund Management at the recent Biotech '03 event held in Helsinki, Finland, where there is a burgeoning biotech industry. 'There are 1,900 biotech companies in Europe, compared with 1,500 in the US,' he said. 'But maybe 60% of these have business issues and could do with being consolidated.'

And indeed, there is no large biotech company in Europe along the lines of Amgen in the US. 'There are too many companies in Europe, and not enough world-class managers for them,' Kurkijaervi said. 'Governments continue to be catalysts to the emerging markets with funding, and substantial support from the capital markets is still needed for success. Darwinism is currently rationalising the market, and consolidation should be a strategy, not a destiny.'

Despite the current difficulties facing the biotech sector on the business side, the rapid strides made recently in the science of biotechnology are having a dramatic effect on drug discovery. The unravelling of the human genome, in particular, is providing a massive amount of information about the causes of diseases that just a few years ago could only have been dreamed of.

Perhaps the human genome project's biggest surprise so far is that the number of genes is much lower than had been predicted, as Professor Leena Peltonen-Palotie of the department of medical genetics and molecular medicine at the University of Helsinki explained at the conference. 'Even one year before the announcement, the human genome was expected to contain about 100,000 genes. The actual number is only about a third of that. And the rule "one gene, one protein" is wrong - some sequenced regions could give thousands of proteins! There is also a number of large duplications.'

useful information

Until now, only single or small groups of genes or proteins have been analysed for their involvement in the disease process. It is now becoming possible, thanks to high throughput methods such as gene chips, for information on all genes to be found using one microscale test. Bioinformatics is how science is trying to elicit useful knowledge out of the mass of information provided by genome mapping. It is used to predict transcript and protein structure, and also protein function. And it is being used to identify novel metabolic pathways and protein interactions.

Several of the speakers addressed the issue of genealogical gene databases. Perhaps the most famous - and most extensive - of these is that being created by deCODE in Iceland. As the company's president and ceo Dr Kari Stefansson explained, in Iceland there is a whole population database stretching back over 1,100 years, which has provided a powerful basis for the genetic investigation of disease.

gene implications

'Mendelian diseases are simple,' he said. 'They are accidents of evolution. However, common diseases are much more complex, often involving several genes. Another problem is that many common diseases occur at the interface between genetics and the environment. When we're looking for the genetic causes of lung disease or COPD, for example, we must make sure we are not studying the genetics of nicotine addiction!' Isolated populations have big advantages, as they have a higher degree of homogeneity, with fewer mutations in the disease genes, hence making them easier to identify.

deCODE ran incidences of cancer within Iceland through the genealogical database, and found that some did indeed run in the family. It has now created a number of disease collections, including Alzheimer's, stroke, rheumatoid arthritis, osteoarthritis, non-insulin dependent diabetes, obesity and osteoporosis, and some more surprising ones such as anxiety. And some responsible genes have also been pinpointed, such as one implicated in schizophrenia, where 31% of patients compared with just 14% of the control group carried the at-risk halotype. Similarly, a gene implicated in stroke has also been found. Another interesting finding was that a gene involved in myocardial infarction coded for a protein that is targeted by an existing, marketed drug for a totally different indication, and this is now being evaluated as a potential treatment for myocardial infarction.

Iceland is not the only country building up a genome database. Estonia has recently begun a similar project, with the goal of collecting data on at least a million of the country's 1.4m population and creating a health and genetic database. 'The aims are to apply DNA-based diagnosis and personalised treatment methods to achieve better healthcare service at lower cost,' explained Professor Andres Metspalu from Estonia's University of Tartu. 'Estonia is ideal for such a project. The population is large enough to provide sampling for common diseases. There is a developed infrastructure and nationwide primary healthcare system, which will be the main data collector, as well as an efficient IT and data communications infrastructure, a transparent legal and ethical environment, quality staff and a nascent biotechnology industry. And it's financially possible!'

genetics research

Participation in the gene research studies is voluntary, with complete confidentiality for gene donors, who also have the right to know or not know their genetic data, as well as to apply for their data to be destroyed at any time. And non-discrimination by employers and insurers is guaranteed. Estonian academic institutions will have free access to the database, and commercial companies will have to pay a fee for access. 'Some 43% of informed people are willing to be gene donors already,' Metspalu said, 'and 36% would like to know more before they decide. Just 6% do not support it. By 2007, we hope that 75% of the country's population will be included.'

Finland, too, is an excellent candidate for a genetic library. 'Finland is a good place to do genetics research,' said Kari Paukkeri, ceo of Jurilab, which is located in Kuopio and uses its DNA database for drug target discovery and medical DNA microchips. 'It is 200 times cheaper to carry out genetic research in eastern Finland than in the general population.'

'Finland has good population records that have been kept by the church, with information on families dating back to about 1600,' said Peltonen-Palotie. 'Many studies have been carried out in Finland, and most loci are replicated in other populations.' One example is a study on familial combined hyperlipidaemia - the most common form of genetic hyperlipidaemia, estimated to cause 10-20% of all premature coronary heart disease. It has a prevalence of 1-2%, and both genetic and environmental effects are implicated. A linkage to a gene has been found in Finland, with the gene believed to be associated also with non-insulin dependent diabetes and plasma free fatty acids. By looking for genes on the web, said Peltonen-Palotie, it is possible to identify the effect SNPs have. 'It's a slow and expensive process,' she said.

lactose intolerance

Another genetic condition that has been studied in Finland is lactose intolerance. It is common in Finns, and two SNPs have been found, which perhaps surprisingly are a long way from the lactase gene. 'This means they must be very old variants,' she said. Based on genetic data, it appears that the lactose intolerance allele represents the ancient, original form of the human allele - those who are lactose tolerant are actually the mutants. 'Hundreds of different allele types have been found globally, so common disease mutations have been beneficial during evolution, and it is our current lifestyle that has made them disease-forming.'

Genetics also has a bearing on how our bodies respond to drugs. As Prof Magnus Ingelman-Sundberg of the Institute of Environmental Medicine at the Karolinska Institutet in Sweden explained, drug absorption, excretion, metabolism and receptor interactions all have a genetic basis. 'Only 30-60% of patients respond to the common drugs,' he said. 'And serious adverse drug reactions (ADRs) are responsible for 5-7% of all hospital admissions. This costs US$100bn a year in the US alone.' This is where pharmacogenetics can help - it can take a patient's genetic constitution into account when a drug is prescribed to increase the number of responders and decrease the incidence of ADRs.

Populations have been established for adverse reactions to warfarin and anticancer agents. 'The lesson for drug design is to avoid a design which makes the candidate a high affinity substrate.'

Although big strides have been made in understanding how our genes cause illness and interact with medicines, we are still a very long way from a comprehensive understanding of what each gene does. As Peltonen-Palotie said, 'We've finished the genome map. We just don't know how to fold it.'