A Hip Approach to Gene Hunting


Horizons

COMPANY TO WATCH | IntegraGen makes human gene mapping easy 

By Michael A. Goldman

June 17, 2004 | IntegraGen, in Evry, France, is trying to get the laboriousness out of human genetics. Using a streamlined genetic mapping strategy and an ingenious hybridization procedure, the company says it can achieve with four staff members, two months, and 200 patients what would take most laboratories two years with large multigeneration families and up to 1,000 research subjects.

The company recently announced the discovery of novel genes involved in autism and obesity, both of which provide potential drug targets. So why isn't everyone using this technique? President and CEO Jan Mous says, "The enzyme cocktail required for the work just isn't so easy to perfect."

The IntegraGen method goes by the name GenomeHIP (Genome Hybrid Identity Profiling), and relies on the fact that "chromosomal regions with the disease gene will be 100 percent identical by descent between affected relatives," Mous says.

In essence, the GenomeHIP technique searches for all regions that are identical between the genomes of as few as two affected relatives. The sophisticated biochemistry uses a complex mixture of DNA repair enzymes to destroy mismatched regions of the genome. Then, the remaining candidate regions are hybridized to a bacterial artificial chromosome (BAC) DNA chip with about 3,000 markers, scanning the genome at a 900-kilobase resolution. (The BAC array has 3,000 human DNA fragments averaging 120 kilobases in length, printed in quadruplicate.)

IntegraGen
Founded: 2000
Location: Evry, France
Financing: $13 million to date
CEO: Jan Mous
No. employees: 27
As relatives typically share much of their genomes, a single affected-pair comparison gives numerous hits. But repeating the procedure on as few as 100 affected pairs and looking for common hits quickly narrows the overlap, identifying just a few 500- to 2,000-kilobase regions. With an average of 10 to 12 genes per megabase, that leaves just a handful of candidates for analysis. And IntegraGen's method works with both simple Mendelian disorders and complex, multifactorial traits.

One of the chief advantages of the GenomeHIP technology is that it does not require multigeneration families. In many cases, two affected individuals from the same family will suffice. There is also no need for extended pedigrees or for the use of homogeneous populations, such as the Icelandic population studied by deCODE. This broadens the array of genes that can be detected by allowing the exploration of the full range of variation and genetic heterogeneity. Moreover, GenomeHIP does not require PCR amplification.


Early Hits 
IntegraGen is confident it has identified an important candidate autism gene, which Mous categorizes as a "classical drug target." He estimates that this novel gene is one of three to eight major genes involved in autism, and that it will be useful in predictive testing or risk assessment. The gene, according to chief scientific officer Jörg Hager, "codes for a synaptic protein that is involved in modulating neuronal signals."

 

BETTER ODDS: Probability of successful disease gene detection as a function of disease allele frequency demonstrates up to a tenfold improvement in detection using IntegraGen's GenomeHIP protocol.

Remarkably, only 232 affected family members were needed for the study, and the gene was nabbed in less than six months. Full details should be published after the findings are confirmed on additional patient samples.

The company has also hauled in three obesity targets, all in the same pathway involved in regulation of food intake. Mous says the genes are "active in the periphery — intestinal and adipose tissue," which is good news for potential druggability. One mutation, in the coding region of a G-protein-coupled receptor (GPCR), occurs in more than 60 percent of affected individuals, increasing obesity risk about eleven-fold in the families studied.

IntegraGen was founded in 2000 by Hager, formerly at the National Genotyping Center in France; Peter Brooks, a biochemist expert in DNA repair enzymes; and Philippe Gesnouin, an information systems manager from Lexicon Genetics. Now headed by Mous, former CSO of LION Bioscience and former global head of genomics technologies for F. Hoffman-La Roche, its 7,500-square-foot facility is just a short drive from Paris.

The company obtained first-round venture financing at about $7 million in 2001, bridge financing of $6 million, and a second-round financing of $18 million planned for late summer 2005. The company's intellectual property portfolio includes four patents on GenomeHIP applications and the genes the company has identified. Its focus is the development and marketing of diagnostic tests. IntegraGen also markets its own BAC marker microarray chip, which can be used in mapping or in detecting chromosomal aberrations. In partnership with Aventis, IntegraGen is studying the elusive genetics of schizophrenia.



Michael Goldman is a professor of biology at San Francisco State University.











For reprints and/or copyright permission, please contact  Jay Mulhern, (781) 972-1359, jmulhern@healthtech.com.