Gene therapy is one of the many applications of genetic engineering. It involves introducing a new gene or modifying an existing gene (itself or its activity) in cells. It can be used to treat or prevent diseases.
There are two forms of gene therapy:
1. Germ cell therapy (unethical and not allowed in humans, easily performed in plants)
2. Somatic cell therapy (currently used to treat cystic fibrosis, severe combined immune deficiency [SCID], some tumors, etc). This kind of gene therapy can be applied to the whole body (in vivo therapy) or to the cells removed from a patient (ex vivo therapy). In the latter, the engineered cells are returned to the patient.
Gene delivery methods:
1. Recombinant retroviral or adenoviral vector-mediated: Retroviruses can only accept up to 7 kb of introduced DNA. They only infect dividing cells. Because of this, they are used in ex vivo therapies. They integrate into the host genome. Thus, their effect is long lasting, but insertional mutagenesis is a potential problem. Retroviruses are used in the treatment of SCID. Adenovirus can accommodate larger genes and does not integrate into the host genome. Their effect is transient. The main problem with adenovirus is the immune response they elicit (used in the treatment of cystic fibrosis and a1-antitrypsin deficiency).
2. Liposome-mediated transfer: DNA is encapsulated in liposomes (lipid micels). This method has no side effects but is less efficient in transferring DNA to target cells. It does not cause an immune reaction. Liposomes can be used in vivo and ex vivo and carry any size of DNA fragment.
3. Microinjection (germ cell therapy): DNA is injected directly into the nucleus of a fertilized ovum viewed under the microscope. Currently, this method is used routinely to produce transgenic animals.
4. Biolistics: This method is used in plants. It involves coating special metal spheres with DNA and firing them into the plant cell from a special gun. Biolistics is used as an alternative to Ti plasmid-mediated gene transfer. See Plant Genetics.
Gene therapy in humans
In humans, genetic engineering is only attempted in somatic cells in the form of gene therapy. Technically, it is possible to modify a fertilized egg at the four-cell stage (or even clone) but there is no ethical approval for such germ cell gene therapy at present. The first gene therapy for a human disease was successfully achieved for SCID by introducing the missing gene ADA into the peripheral lymphocytes of a 4-year-old girl and returning modified lymphocytes to her (in 1990). As of January 2000, more than 350 gene therapy protocols have been approved in USA.
Human gene therapy uses the same strategies to deliver the genes. This can be achieved ex vivo or in vivo (in vivo: viral vectors, liposomes; ex vivo: cells such as tumor-infiltrating lymphocytes (TNF insertion), bone marrow stem cells as in bone marrow transplantation, or peripheral lymphocytes -as in SCID (ADA insertion)- are taken out, modified and returned to the patient). Alternatively, antisense treatment can be tried to prevent the transcripts from being translated into unwanted proteins (as being tried to combat HIV infection). Even naked DNA (i.e., not inserted into a carrier virus or liposome) can be injected directly to the patients to substitute the defective gene (DNA vaccines). DNA vaccines are plasmids being inserted the desired gene. Plasmids may directly transfect (animal) cells in vivo. This strategy is usually used to elicit cytotoxic T cell type immune response using an antigenic product of a pathogen (e.g., HIV).
A popular gene therapy method is using the 'suicide gene'. The gene in question is thymidine kinase from a herpes simplex virus (HSV-tk) and is delivered to the target cells (usually cancer cells). When ganciclovir, an otherwise harmless antiviral agent, is given to patients, HSV-tk-bearing cells convert it to a toxic substance and the cells die. This is usually used in the treatment of (brain) tumors, but also used in the treatment of GvHD following BMT. Any other gene, which would render cancer cells highly sensitive to selected drugs, can be used as the suicide gene. A logical use of gene therapy in cancer would be either replacing the missing tumor suppressor gene or blocking the effects of an oncogene.
The tricks that can be used to treat HIV infections include using dominant negative mutations (generating inactive versions of proteins HIV needs to replicate), or delivering genes into CD4+ T lymphocytes (target cells for HIV infection) that would be transcribed to short mRNAs mimicking essential viral control mRNAs to interfere with the viral regulatory mechanisms.
Other applications of genetic engineering
Chakrabarty's bacterium was constructed by using classical genetic selection to combine genes originally located on four different plasmids onto one compound plasmid. It is used to clean up oil spills.
Proteins with industrial applications such as Rennin, a protein used in making cheese, can be produced by recombinant DNA technology. This technology deals with isolating a gene (or its cDNA) from its natural host and inserting it into a different (asexually reproducing) species' genome so that it can be copied every time the new host cell (usually a bacterium) replicates. Before the advent of genetic engineering it was extracted from the fourth stomach of cattle. The new technology is known as the 'cheese from microbes'.
Enzymes used in genetic engineering such as restriction endonuclease (biological scissors), DNA polymerase (for replication of DNA), reverse transcriptase (to make DNA from RNA) and ligase (to ligate two DNA fragments) are produced by recombinant DNA technology (by cloning in high copy number plasmids in bacteria). Other enzymes now routinely produced by recombinant DNA technology are rennin (an enzyme used in cheese making mentioned above), lipase (cheese making), a-amylase (beer making), bromelain (meat tenderizer, juice clarification), catalase (antioxidant in food), cellulase (alcohol and glucose production), and protease (detergents).
Proteins for medical applications such as insulin (previously extracted from pig pancreas, and since 1982 produced by recombinant DNA technology), clotting factor IX (which is lacking in hemophilia B patients and previously supplied as fresh plasma from volunteer donors), tissue plasminogen activator (t-PA, used in acute myocardial infarction), growth hormone (previously extracted from the pineal glands of cadavers), tumor necrosis factor (TNF), g-interferon (g-IFN) and interleukin-2 (IL-2) (TNF, g-IFN, IL-2 are used in immunotherapy of cancers), erythropoietin (EPO, to stimulate red blood cell production) and vaccines (such as HBV, rubella) can be produced by recombinant bacteria.
Secretion of valuable (modified) proteins in milk of transgenic animals such as factor IX and elastase inhibitor a1-antitrypsin (used in the treatment of emphysema) can be another source for certain protein. The gene for the desired product is usually combined with a gene coding for a milk protein that is expressed only in mammary glands. The combined gene is then inserted in fertilized eggs. These are implanted into recipient females. The desired protein can be harvested from a female's milk. Because the gene is now in all cells of the animal (including germ cells), its offspring will also have it. Transgenic and gene knockout animals are also used for research purposes. The first transgenic animal was a supermouse with a rat growth hormone gene (1982).
Plants: Plants most commonly used in genetic engineering are maize, tomato, potato, cotton and tobacco. The main aims are to induce tolerance to herbicides, resistance to insect pests or viral disease and to improve crop quality.
Herbicide tolerance: A gene from a soil bacterium (A. tumefaciens) codes for an enzyme (PAT) which inactivates the herbicide Basta. When plants are engineered to contain this gene they are not sensitive to the herbicide any more (already tried on sugar beet, tobacco and oilseed rape). This allows selective killing of weeds by herbicides. A plasmid from A. tumefaciens called Ti is used to integrate the gene into the plant genome (the PAT gene is inserted into the toxin gene of the plasmid).
Resistance to infection by viruses: Genes encoding antisense copies of viral genes were used with limited success to prevent viral infection. The transfer of a gene encoding a viral coat protein has been successful with an unknown mechanism.
Insect resistance: Potato plants have been engineered to contain a pea lectin gene. Lectin interferes with digestion of plants in insects but does not harm the plant. It is also possible to use a Bacillus thuringiensis toxin called protoxin as an insecticide. This has been tried in tomato but did not work very well because of low expression.
Quality improvement: Genetic engineering has also been used to modify plants to create genetically-modified (GM) foods (also called genetically engineered organisms ‘GEO’ perhaps more appropriately). As tomatoes age, they soften due to the effects of an enzyme called polygalactorunase which breaks down cell walls. Its production can be blocked by activating the antisense gene to inactivate mRNA for its gene. Thus, it is still transcribed but no translation occurs. These tomatoes can ripen on the plant and are still suitable (hard enough) for mechanical handling and transport (long lasting tomatoes = FlavrSavr tomato). The FlavrSavr tomato was the first GM food approved by the FDA to go on the market in 1994 (now discontinued). Tomato paste from genetically engineered tomatoes (Zeneca Tomato Paste; also discontinued) and oil from genetically engineered oilseed rape were the first two whole foods declared safe in the UK (in 1995) (see the link for all Transgenic Crops). Transgenic plants such as soybean and rice can be engineered to have the essential amino acids they normally lack. Golden Rice is for example genetically enhanced with beta carotene. Genetic engineering in plants has also been used to alter pigmentation in flowers, to improve nutritional quality of seeds and to obtain seedless fruits. See also Plant Improvements: Biotechnology, Transgenic Plants and Genetic Engineering in Plants (National Geographic, May 2002).
Drawbacks and potential dangers of genetic engineering
1. Gene integration into the right place: Position effect or tissue/cell-specific expression of genes in gene therapy depend on the effects of enhancers and promoters.
2. Immune response against the vector used for gene delivery
3. Spread of the gene conferring antibiotic resistance to the vector
4. Escape of the introduced virus or bacteria to the environment (to avoid this a modified strain of E.coli is used in recombinant DNA applications which cannot survive in nature but is useful in the laboratory. The laboratories using this technology are also under stringent control to avoid any contamination of the outside world)
5. Acceleration of the evolution of resistance to the toxins or antibiotics introduced
Modern biotechnology is achieving what classical biotechnology could not have in such a short time and across a species barrier. There is no need for generations of classical breeding schemes any more and hybridization between different species can be achieved (like inserting a human gene into a mouse germ cell).
1. Nuclear transfer in frogs resulting in viable juveniles up to the tadpole stage in the 1950s (Elsdale et al. J Embryol Exp Morph 1960;8:437)
2. Nuclear transfer in sheep using donor nucleus from early morula stage (8-16 cell) embryos (Willadsen SM. Nature 1986;320:63)
3. Nuclear transfer in sheep using a cultured embryonic (totipotent) cell's nucleus (March 7, 1996 - Nature)
4. Cloning of Dolly the sheep by nuclear transfer from an adult's udder cell (Febr 27, 1997 - Nature) with a success rate of 0.3% (1 in 277 attempts)
5. Cloning of Polly the Sheep by transfer of a transgenic nucleus of a fetal fibroblast cell - (Dec 19, 1997 - Science) [an alternative to pronuclear injection in creation of transgenic animals]
6. Successful use of nuclear transfer in mice (Cumulina, success rate 2-2.8%) and cows in 1998 followed by cloning of many other mammals (cattle, goats, rabbits, cats, pigs, mules, horses). The most recent example is the cloned horse who was delivered by her dam twin in 2003 (see below).
Cloning is the ultimate phase in creating an animal with a desired trait. It follows the same path as selective breeding, artificial insemination, egg transplantation and in vitro fertilization. In genetics, cloning means creation of genetically identical animals (mammals) by means of nuclear transfer. In nuclear transfer, the chromosomes of an unfertilized oocyte (egg) are removed (enucleation) and a nucleus from a mature (diploid) cell is placed into the egg. Note that no fertilization takes place and it is the presence of a diploid nucleus inside an egg, which initiates cleavage divisions. It has been established that an unfertilized oocyte is a better recipient than a zygote. This is asexual reproduction of an animal, which would normally reproduce sexually. A clone resembles its (single) parent, the animal from which the donor cell was taken. This technology can be used either to create genetically identical animals or to introduce a genetic modification to the mammal. Another application is the production of undifferentiated cells which can then be induced to differentiate to any organ or cell type. The success rate is invariably very low. It may take hundreds of tries to succeed in nuclear transfer. It is, however, highest in sheep. This is probably because of the fact that transcription of the embryonic genome does not begin until the 8-16 cell stage in sheep, whereas it is the late 2-cell stage in mice.
In the previous cloning experiments, embryonic cells were used as the source of nucleus and the nucleus was transferred to a fertilized egg whose own nucleus was removed (in mice in 1980s). In the creation of Dolly, however, the nucleus of an adult -differentiated- cell was used. The key to the new procedure was synchronization of the cell cycle of the adult cell with that of the egg. The Dolly experiment showed that for successful cloning, an -undifferentiated- embryonic cell nucleus was not the only choice. Dolly was created from a mammary gland (udder) cell of a six-year old ewe. This showed that any adult cell still has the potential to start from the undifferentiated state and give rise to new differentiated cells (reversibility of differentiation). To achieve synchronization, the (donor) mammary cell is deprived of nutrients to stop its cell cycle (Go phase). The nucleus is then implanted into the egg, and an electrical current is applied to simulate fertilization which initiates egg activation which normally occurs at the time of sperm penetration (in parthenogenesis, egg activation is achieved by physical or chemical stimuli). The egg recombined with a new nucleus begins dividing and proceeds to become an early embryo (blastocyst). This is then implanted into the uterus of an ewe (surrogate mother). The lamb that is born is a clone of the donor. At present, dedifferentiation can only be achieved by forming an embryo from the donor cell and culturing the embryo to the stage when it has a few hundred still undifferentiated cells. Then, the cells would be separated and grown in culture. Cloning, when it becomes more practical to do, has the potential to make copies of a bred line of cattle, sheep or other economically important animals without having to do artificial selection. Another application is to clone genetically modified pigs to use their organs for xenotransplantation. The modification is required to prevent immune response.
It has to be remembered that the cloned animal may not be fully identical to the animal who donated the nucleus. This is because of some epigenetic phenomena (methylation patterns) and cytoplasmic (extra-nuclear) inheritance. The low success rate in nuclear transfer may be due to a random loss of correct imprinting. Differentiated female cells, for example, contain one active X chromosome, but in early female embryos both X chromosomes are active. It is also a concern that imprinting that results in functional differences in the male and female genome may complicate the reprogramming of the genome after nuclear transfer. This may be why out of the 277 mammary-gland-cell nuclei that were fused with enucleated eggs, only 29 developed to the blastocyst stage and only one of those resulted in a live birth, Dolly, in surrogate mothers. Similarly, it took scientists 87 tries to successfully clone the first cat ‘CopyCat’ in 2001. Cloning of a horse in 2003 surprised many scientists because the surrogate mother was the donor of the egg. Therefore she gave birth to her own clone (Nature, New Scientist, Science News, BBC). This time Italian researchers had to try more than 800 fusions. Most failed to develop from the cleavage stage to blastocysts and at early implantation. Interestingly, the failure rate was higher in male embryos (8 of 467 males vs 14 of 286 females, meaning three times high failure in male embryos; P = 0.01), which is probably the first concrete documentation of the same phenomenon known to occur in human pregnancies (see Dorak MT et al, 2002 for references).
Because of the totipotency of the first four cleavage cells in the early embryo, a woman who is suffering from a mitochondrial DNA disease can still have a healthy baby by having the nucleus of her embryo implanted into a donor oocyte whose own nucleus has been removed. This is nuclear transfer (from an embryonic cell) but not cloning. Another development in recent years in reproductive physiology is the feasibility of ICSI (intracytoplasmic sperm injection) that enables sterile man to have a child. The bioethical implications of these applications are discussed elsewhere (see Bioethics). One important and usually neglected piece in discussion of the ethics of cloning is what will happen in the possible outcome of having a disabled / malformed cloned offspring. Are the scientists who have done the cloning going to be held responsible? It appears that cloned mammals usually have a collection of disorders called large offspring syndrome (LOS). Most cloned mammals so far have also developed deformities and arthritis (like Dolly). There is still a long way to go to optimize the conditions of cloning if it will ever be a common practice.
Suggestion for further reading
Wilmut I: Cloning for medicine. Scientific American 1998 (Dec):30-35
A Student’s Guide to Biotechnology - Debatable Issues. Greenwood Press, 2002
Rhind et al. Human Cloning: Can it be made safe? Nature Reviews Genetics 2003 (PDF)
M.Tevfik Dorak, B.A. (Hons), M.D., Ph.D.
Last updated on Dec 7, 2004