Science in Society Archive

Safe Gene Therapy?

The first documented case of cancer from gene therapy has no sooner been announced than the hype of ‘safe’ alternatives begin. Prof. Joe Cummins gives his assessment.

Although gene therapy has been promoted optimistically for over a decade, the technical problems and safety concerns have proved formidable and insurmountable.

In its usual, accepted form, gene therapy is the delivery of genes to human somatic cells in order to remedy disease. Vectors are needed to deliver the genes successfully. The most popular vectors, adenovirus vectors, are not usually incorporated into the chromosome, and hence the transgenes don’t persist for long. The host immune response is another big problem particularly with vectors at high dosage. The first victim of gene therapy that the public heard about, teenager Gelsinger died of acute immune response to the adenovirus vector.

The technical problems with existing vectors has prompted the development of "improved" vectors clearly intended to restore the confidence of granting agencies and the public in gene therapy. Two newly described, "improved" vectors are described in the current issues of Nature Biotechnology and Nature medicine.

The first, a "gutless" adeno-transposon vector [1,2], is an adenovirus (used because it efficiently injects its DNA into mammalian cells and tissues) from which the viral chromosome has been removed leaving only the short terminal repeat sequences needed to package the chromosome in the virus. The adenovirus vector whose chromosome has been "gutted" is replaced with a transposon (a jumping gene or mobile genetic unit) from fish called "sleeping beauty" which has been modified to contain the genes for therapy along with the genes needed to integrate the vector into the mammalian chromosome. It has been found that the transposon integrates into the chromosome most efficiently as a DNA circle. So a gene, FLP, from a circular nuclear plasmid in yeast, is added to the vector, that creates DNA circles from the linear DNA released from the adenovirus vector after the latter is taken up into the mammalian cell. The gutless adeno-vector efficiently transfers the modified transposon to the mammalian cell, and the fish transposon integrates the therapy genes into the mammalian chromosome.

Unfortunately, the "sleeping beauty" transposon integrates into random sites in the genome, leading to insertion mutations. The transposon has, in fact, been used to create insertion mutations in numerous species. Such insertions are likely to cause cancer in the treated individual, as the recently confirmed case of leukemia in a child receiving gene therapy has shown all too clearly ("Predicted hazard of gene therapy a reality" ISIS Report, October 2002)

The recombinase genes and enzymes introduced into the cell by the vector may also cause both chromosome instability and a strong immune response. In conclusion, using adeno-vectors modified to transfer transposons carrying transgenes poses threats as great or greater than the original adenovirus (before it got gutted).

The second "improved" vector for gene therapy involves using a bacterial site- specific recombination system called a "bacteriophage" integrase [3]. The bacteriophage (bacterial virus) from which integrase was isolated infects Streptomyces bacteria. The phage system has the advantage that a relatively large piece of DNA can be integrated at specific sites (short DNA sequences), called attP. Mammalian cells contain a number of genes called pseudo attP sites - sites not identical to the attP site of bacteria, but similar - where genes may be inserted using the phage system. This alone should draw attention to the fact that most "site-specific" recombination systems are not specific, and insertion can happen at extraneous sites.

In practice, the attP site and genes to be inserted into the chromosome are inserted into a bacterial plasmid, and the plasmid, along with the integrase enzyme are used to transform human cells. Very large genes such collagen were used to transform human cells taken from patients with genetic disorder of collagen such as dystrophic epidermolysis bullosa.

The plasmid transfer is effective against cells in culture that may be used to restore diseased tissue, but the plasmid is relatively ineffective in the intact human or animals. Furthermore, the pseudo attP site on chromosome 8 is only a preferred site for integration but a number of other pseudo attP sites have been identified on other chromosomes. These sites must be most carefully studied for complications, such as cancer, arising from the insertion. Imprecise gene insertion in gene therapy can indeed scramble genomes [4] as well as cause cancer.

There are other problems of gene therapy that these "improved vectors" simply do not address, such as the difficulty of getting regulated transgene expression, and the tendency of the patient to regard the transgene-encoded protein as a foreign target for immune attack.

For an in-depth analysis of the problems of gene therapy see "Failures of gene therapy", Science in Society 2002, 16, now out.

Article first published 22/10/02


  1. Linden R. "therapy gets the Beauty treatment. Nature Biotech 2002, 20,l987-8
  2. Yant S, Ehardt A, Mikkelson J, Meuse L, Pham T and Kay M. Transposition from a gutless adeno-transposon vector stabilizes transgene expression in vivo. Nature Biotech 2002, 20,999-1005
  3. Ortiz-Urada S, Thyagarajan B, Keene D, Lin Q, Fang M, Calos M and Khavari P Stable nonviral genetic correction of inherited skin disease. Nature Medicine 2002, 8, 1166-70
  4. Cummins J and Ho M. Scrambled genomes in human gene therapy and transgenic plants" 2002

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a cat?