Some inherited diseases are caused by simple changes ingenes. Sickle cell anemia is the classic example. In thisdisease, the malfunctioning that causes red blood cells tosickle results from a mutation that changes only oneamino acid. However, most genetically based diseasesand essentially all cancers result from far greatermodifications in the genome of affected individuals. Inthese diseases, chromosomal rearrangements are commonwith large deletions and duplications being the rule.

Genetic engineers can insert specific mutations into genesand often insert these modified genes into specific sites inchromosomes and then study the effects on cells. Butresearchers have been unable to efficiently andreproducibly engineer the important and dramaticchromosomal rearrangements that are more oftencharacteristic of inherited diseases and cancer. As aresult, the understanding of these diseases is limited by aninability to develop models that can identify the importantgenes impaired by these rearrangements.

But now Allan Bradley and his colleagues, in thedepartment of molecular and human genetics and theHoward Hughes Institute at Baylor College of Medicine,in Houston, have surmounted this difficulty. In an articletitled "Chromosome engineering in mice" that appears inthe Dec. 14, 1995, Nature, these molecular geneticistsdescribe a method by which they can systematicallycreate large deletions and duplications in the mousegenome. They also show that these mutations can bestably inherited in transgenic mice.

Embryonic Stem Cells With Engineered Deletions

These scientists used loxP DNA sequences to target theregions of mouse chromosome 11 that they wanted todelete or duplicate. Recombination can occur efficientlyin the presence of loxP sequences. Any genes presentbetween loxP sequences are deleted at a high frequencywhen a gene coding for Cre, an enzyme that causesrecombination, also is inserted into the same cell.

Their strategy was to transfect embryonic stem cells withloxP linked to a selectable marker gene and allow celldivision to occur in culture in the presence of theselectable marker. This allowed only the cells containingthe loxP sequences to divide and grow. This process wasperformed twice with loxP sites linked to two differentselectable markers so that these researchers were sure thatall of the selected cells would have two loxP sitesintegrated into their genomes.

Then the Cre gene was transiently transfected into theseselected cells, allowing it to delete the genes between thetwo integrated loxP sites. Since the two loxP sites wereboth embedded in a gene coding for yet another, differentselectable marker, cells in which recombination occurredcould be isolated because the recombination reassembledthe interrupted marker enzyme DNA sequences into afunctional gene that then encoded the third selectablemarker. By using this model system, Bradley andcolleagues have been able to generate deletions on mousechromosome 11 that range from 90 kilobases up to 7megabases.

Bradley told BioWorld Today, "With this technology, wecan apparently make deletions with little or no sizeconstraints." He went on to add that his lab is scaling upto tackle most of the mouse genome. Bradley reportedthat, "We have now generated deletions on a far largerportion of the mouse genome."

Bradley and colleagues are pursuing two avenues ofresearch that use this new ability to engineer chromosomerearrangements. He explained that, "We are identifyingand cloning tumor suppressor genes using this system.Our greatest interest is in suppressors that are related tobreast cancer." He indicated that, "We have very good invitro data showing that the regions we have selected forstudying tumor suppressors in the mouse genome areanalogous to those in humans."

The other active area of research at the Baylor labconcerns the generation of transgenic mice usingembryonic stem cells containing the engineered deletions."The transgenic mice that have been constructed showphenotypic differences. The deletions tell us where thegenes causing these differences are, but we now have toidentify these genes," explained Bradley. He noted thatthis latter work was being done with collaborators.

At present, this technology for chromosome engineeringhas not been commercialized. "Patents have been filed onthis technology, but we have not decided the best way tocommercialize it as yet. When we can show the ability toclone tumor suppressor genes, then commercializationwill be easier," Bradley said, adding that the real value inthis technology will be in identifying gene function.

Identification Of Tumor Suppressor Genes

Mario Capecchi, a well-known molecular geneticist at theUniversity of Utah, agreed, "The best use of Bradley'snewly developed technology is likely to be in identifyingtumor suppressor genes." Capecchi pointed out that, "Theendpoints of deletions vary from cancer to cancer. Sincechromosome engineering will delete a broad region withvariable endpoints, a biological assay will be used topinpoint genes of interest."

Capecchi told BioWorld Today that, "The power of theCre-loxP system is that it allows you to engineer any sizedeletion that you want."

He summed up the importance of this work by sayingthat, "Bradley has been able to combine site-specificrecombination and gene targeting to extend what ispossible beyond the deletion of small regions of thegenome. That is a significant jump forward intechnology." n

-- Chester A. Bisbee Special To BioWorld Today

(c) 1997 American Health Consultants. All rights reserved.