Laying the Groundwork: The Techniques and Applications of Recombinant DNA Technology

Leonard A. Post

Leonard A. Post, Ph.D., is Director of Molecular Biology Research at the Upjohn Company.

One of the problems in discussing "biotechnology" or "genetic engineering" is a definition of the subject matter. A U.S. Office of Technology Assessment definition of biotechnology was "any technique that uses living organisms (or parts of organisms) to make or modify products, to improve plants or animals, or to develop microorganisms for specific uses."¹ Genetic engineering could apply to any process that results in the derivation of living things which are genetically suited for some application. With these definitions, biotechnology and genetic engineering are not things that have come up in the last decade or even in the last century. These definitions include virtually all of agriculture, which is based on use of plants and animals for particular applications. Through selective breeding, modern agriculture has "engineered" plant and animal species to be very different from their naturally evolved ancestors. Biotechnology would have to include fermentation to produce alcoholic beverages. Genetic engineering would include derivation of microbial cultures that produce practical levels of antibiotics.

When someone speaks of biotechnology or genetic engineering in 1988, they are usually not referring to selective breeding of plants and animals, or the brewing industry. The terms are most often used for applications of recombinant DNA technology, and that would appear to be the major theme of this conference. Recombinant DNA is an idea that developed in the 1970s as tools for manipulation of DNA became more powerful than ever before. The result is that "genetic engineering," although in principle nothing new, became more rapid and more precise than it had been in the past.

The Central Dogma and Recombinant DNA

The 'Central Dogma of Molecular Biology' states that genetic information is stored in molecules called nucleic acids, and that the information stored in nucleic acid is decoded by translation into proteins. One nucleic acid which is the genetic information of nearly all living things is deoxyribonucleic acid, or DNA. At one level, DNA is an incredibly simple molecule. It is made up of repeats of only four chemicals, abbreviated A,G,C, and T. These four chemicals, called nucleotides or bases, are the alphabet that spells the genetic information, by the order of A's, G's, C's, and T's. To encode a bacterium like Escherichia coli takes a DNA molecule that is 5 million bases, i.e., 5 million A's, G's, C's, and T's in a specific order. To encode a human takes approximately 3 billion bases of DNA sequence! This large [19 ELR 10489] amount of DNA, all of which is chemically very similar, made specific manipulation of DNA an impossible task twenty years ago.

Recombinant DNA is the methodology by which specific pieces of DNA can be isolated from a massive genome, in short enough pieces to be manipulated conveniently. Enzymes called restriction enzymes were discovered to have the properties of cutting at specific DNA sequences (e.g., EcoRI cuts wherever the sequence GAATTC appears in a DNA molecule). This allows specific fragments of DNA to be isolated. The enzyme DNA ligase has the useful property of sticking fragments of DNA together. In combination, these reagents can make recombinant DNA. For example, one starts with a small DNA molecule, usually a small circle called a plasmid vector, that can replicate inside a bacterium like E. coli. That is cut by a restriction enzyme that converts the circle to a linear molecule. Then one mixes some other DNA fragments generated by restriction enzymes — from any source, plant, human, insect, or another bacterium — with the vector along with ligase. The ligase will stick some of the foreign DNA fragments to the vector. When the ligated vector is re-introduced into E. coli, it still has the ability to replicate and grow in the bacterium. If another DNA fragment has been stuck to the vector DNA, it will be replicated in the E. coli along with the vector DNA. This is a recombinant DNA molecule! If the recombinant DNA molecule consists of a vector plus a piece of human DNA, then the human DNA piece will be replicating in E. coli. The pieces transferred are much smaller than an intact human genome. The pieces of recombinant DNA transferred to E. coli are usually in the range of 500-50,000 bases long. This small fragment of DNA can be replicated free from the rest of the DNA of the organisms from which it was derived. Therefore, this fragment has been "cloned."

Why clone DNA fragments? One reason is simply to study individual parts of the DNA that have specific functions. Techniques have been developed that allow rapid determination of the order of the A's, G's, C's, and T's in a particular cloned DNA molecule. Although a 10,000 character string of A's, G's, C's, and T's may not look like exciting reading, it is often extraordinarily informative to a molecular biologist. This sequence of bases often tells what the DNA molecule does, including what kind of protein the DNA molecule encodes. Just by looking at sequences of cloned genes, advances have been made in the understanding of cancer and Alzheimer's disease. Recently, there has been much publicity about a national effort, either by the U.S. or by Japan, to determine the sequence of the 3 billion bases of the human genome. Although a monumental effort, having all of the information that encodes the components of a human organism would be an incredible boon to medical research.

Cloning DNA fragments to determine their structure is only the beginning, however, once cloned, DNA fragments can be manipulated further by the same recombinant DNA techniques described above. They can be cut at will, DNA sequences removed, and additional DNA sequences spliced on. The result is that the function of the cloned DNA can be exploited.

Protein Products from Cloned Genes

When the DNA encoding a given protein is moved from one species to another, it follows directly from the Central Dogma that all of the information necessary to make that protein has been transferred. For example, when DNA encoding human growth hormone is inserted into a plasmid vector that replicates in E. coli, it is possible to construct an E. coli that makes human growth hormone. Human growth hormone made in E. coli from recombinant DNA is now marketed by two companies. This removes the dependence on cadavers as a source of growth hormone for children with growth hormone deficiency. Thus, the source of growth hormone is much more dependable and safer than before recombinant DNA. Similarly, twenty years ago there was a genuine concern that the diabetic population would increase faster than the supply of swine pancreases from which insulin could be extracted. Extrapolation indicated that there would be a shortage of insulin. Now that insulin can be made by E. coli carrying DNA for human insulin, the supply of insulin is limited only by the size of the culture vessel.

In addition to making protein sources more reliable, the economics can be improved by recombinant DNA. Bovine growth hormone results in a significant increase in milk production efficiency. However, when the only source of bovine growth hormone was from bovine pituitary, it was never economical to produce enough of the protein for practical use. By inserting the DNA for bovine growth hormone into E. coli it is now possible to make economically large amounts for administration to dairy cattle.

Making some proteins by recombinant DNA can make other kinds of research safer, and thereby accelerate progress. A good example of this is in study of the human immunodeficiency virus (HIV) which causes AIDS. The only approved therapy for AIDs is the drug azidothymidine, which inhibits a viral enzyme called reverse transcriptase. Because azidothymidine has significant side-effects, an important direction in AIDS research is to find new reverse transcriptase inhibitors. By using recombinant DNA to move the piece of the HIV genome encoding reverse transcriptase into E. coli, reverse transcriptase can be made in E. coli. This means it is possible to do research on reverse transcriptase and its inhibitors without handling the HIV virus. This significantly increases the safety and speed of the research.

Once a recombinant DNA molecule is made, it can be introduced into types of cells other than E. coli. This is sometimes done because E. coli cannot make some complex animal proteins. For example, Genentech's tissue plasminogen activator is made by moving DNA encoding human tPA into Chinese hamster ovary cells. The Chinese hamster ovary cells can then be grown in large tanks to produce adequate amounts of tPA for use as a drug.

Discussion of tPA brings up another powerful application of recombinant DNA technology. Once a piece of DNA is cloned, it can be changed by replacing naturally occurring DNA sequences with chemically synthesized sequences. By the Central Dogma, changing the DNA sequence changes the encoded protein. When the altered DNA is inserted into E. coli, a Chinese hamster ovary cell, or whatever host cell is used for expression, the altered protein is made. The result is the capability to make proteins that do not occur in nature, that can be "designed" for specific purposes. One use of this technology is the attempt to make an improved tPA with greater efficacy and less potential for bleeding complications.

The application of recombinant DNA to produce protein products is not the area that usually generates the most environmental issues. As with any manufacturing process, there may be environmental implications from the kinds of waste generated. Most, if not all, processes involving recombinant [19 ELR 10490] DNA involve killing of the cells before disposal. Of course, the use of a protein product may have environmental implications. These considerations are product-specific and do not involve the process by which the product is made.

Viruses and Bacteria

Recombinant DNA can be used to introduce desired features into bacteria and viruses that make them particularly useful for some application other than to produce a protein product. In some cases, the recombinant organisms may be used in a manufacturing process such as production of an antibiotic. In other cases, for the organisms to be used for the intended purpose, they must be taken out of the manufacturing plant and used outside any containment. In this case, the environmental assessment involves the properties of the organism, since it will be introduced into the environment.

One such application is the construction of live vaccines. One particular example is a live vaccine for pseudorabies virus that the Upjohn Company field-tested in 1987 and was licensed by USDA in December of 1987. Pseudorabies virus (PRV) is a herpes virus of swine that is a significant economic problem. In this case, recombinant DNA was used not to add a new gene, but to remove two genes from the PRV genome. The recombinant DNA process — cutting with restriction enzymes and ligation of DNA back together in novel combinations — is nearly the same as described above, except in this case DNA is cut out of the parent DNA molecule and nothing added to replace it. One of the genes removed from Upjohn's PRV vaccine is thymidine kinase, or tk. Tk is essential for PRV to cause disease in pigs, and removal of the tk gene makes the vaccine unable to cause disease. This mutation is safer than most mutations that attenuate live vaccines. This is because most mutationsare "point mutations," in which a base in the DNA is changed from one to another (e.g., an A changed to a G). This can be a problem because if natural processes can convert an A to a G, similar processes can convert the G back to an A, making the virus pathogenic again. Such revertants cannot arise when a large sequence of the gene has been removed. In the PRV vaccine, 276 bases have been removed from the tk gene. The only way the vaccine could regain virulence is to regain something very nearly that precise sequence of 276 A's, G's, C's and T's, to recreate a functional tk gene. There is no spontaneous natural way for that to happen in the field. Even though a tk mutation could have been introduced into the vaccine by a process that does not involve recombinant DNA, the virus made by recombinant DNA is much safer. After extensive review for environmental implications by USDA, the field tests of this vaccine were conducted on hundreds of pigs without any safety problems.

There are many examples of using recombinant DNA to modify bacteria that may be useful in a field situation. The much-publicized "ice minus" bacteria was a case where recombinant DNA was used to remove a particular gene, such that the bacteria could possibly confer some frost resistance to crop plants. Recombinant DNA has been used to make microorganisms that normally live in a close relationship with plants into organisms that potentially kill plant pests. For example, the gene for the insecticidal Bacillus thuringenesis protein was moved into bacteria that normally live on plants. Rhizobium species that normally fix nitrogen to provide nutrients to plants have been manipulated with recombinant DNA in such a way that they may become more efficient at providing this "biological fertilizer." Other examples of environmental use of microorganisms with recombinant DNA that are further into the future include construction of organisms to degrade chemical and biological waste, and to enhance recovery of oil and mineral resources.

Field use of such bacteria and viruses requires evaluation of the likely environmental effects. This evaluation is certainly going to be the subject of discussion in later presentations at this conference. It is not possible to generalize that all microorganisms containing recombinant DNA are either safe or unsafe. In fact, it is not possible to generalize that all novel microorganisms that do not contain recombinant DNA are safe or unsafe for release into the environment. The facts of each particular novel microorganism must be evaluated. Whether the new organism was derived using recombinant DNA or some other method cannot be the basis of decision on its predicted safety.

Recombinant DNA Inplants

Just as it is possible to put recombinant DNA molecules into bacteria or animal cells, it is possible to introduce recombinant DNA molecules into plants. The most common way of effecting this is to use a natural genetic engineer, the bacterium Agrobacterium tumefaciens. Agrobacterium causes crown gall tumors on plants it infects, causing these tumors by transfer of a bacterial DNA sequence into the plant cells. Using recombinant DNA it is possible to remove the DNA that causes the tumor development, and replace it with DNA that encodes some desirable trait for the plant. Then the Agrobacterium can introduce the gene for the desired trait into the DNA of the plant. The actual manipulation of this system can be very complex, requiring regeneration of plants from tissue. The principle of the transfer or recombinant DNA is the same as discussed above for introduction of novel genes into bacteria.

It has been demonstrated that useful single genes can be transferred to important crop species using the Agrobacterium system. Transfer of a gene by this recombinant DNA technique can be used to make a plant resistant to some viruses; toxic to insects without being toxic to humans; and resistant to herbicides that may be more environmentally acceptable than others. Other contemplated applications of recombinant DNA to plants include resistance to fungal diseases, resistance to environmental extremes, and improved nutritional or processing properties.

Recombinant DNA provides a new technique to the plant breeding process. Individual genes can be added without a breeding program to eliminate other undesirable traits that might come in with a desired gene by a conventional cross. Also, it is possible to introduce genes from other sources than species of plants that can naturally cross. Of all of the properties to be introduced into plants via recombinant DNA in the foreseeable future, however, none is likely to introduce the magnitude of change from the natural species that centuries of traditional breeding have already made. Modern hybrid corn is vastly different from what the Indians grew, which was already so different from any natural species that the natural ancestor of corn was barely recognizable. Transfer of single genes is unlikely to make this magnitude of change in a plant species.

Of course, plants containing recombinant DNA must be grown in the field if they are to be of any practical use. Introduction of new plant species (with no recombinant DNA) has occasionally had deleterious environmental effects, and the USDA therefore has a regulatory mechanism [19 ELR 10491] in place. In the case of a plant with recombinant DNA, the problem is to assess the likely environmental consequence of a new variety of a species for which a large body of experience already exists. The new variety differs from the older well-known varieties by addition of one or a few discrete, characterized genes.

Transgenic Animals

It is also possible to transfer recombinant DNA into animals. This technique is well established only for mice. The most common method is that a mouse embryo is surgically collected shortly after fertilization, and recombinant DNA is microinjected into the embryo with a very fine needle. The embryo is then re-implanted into a surrogate mother mouse to develop. Some of the mice born after such a procedure contain the recombinant DNA inserted into their own DNA, and sometimes express the inserted gene. The most famous examples of transgenic mice are the mice which contain extra growth hormone genes such that they overproduce growth hormone and grow abnormally large. More useful transgenic mice have made significant contributions to understanding of the development of cancer.

Transgenic animal technology is now being applied to livestock species. There has been successful production of transgenic rabbits, sheep, chickens, and pigs. If useful improvements in livestock can be achieved by transgenic technology, the changes will probably be very much like what was discussed above for plants. Single genes will be transferred, in a way to complement the centuries of experience in traditional breeding techniques of livestock. As an example of the power of the traditional breeding techniques, one can compare lean modern domestic swine with the pigs bred for lard production at the beginning of this century, and compare both with the wild ancestor of the pig. Parenthetically, the transgenic livestock technology is very different from the widely publicized proposals for "gene therapy" for genetic disease of humans. The proposed experiments for humans involve providing a gene to the somatic cells of the patient, and not in the germ cells. This means that the recombinant DNA would be in, for example, only the blood cells of the treated patient, and would not be transmitted to the offspring of the patient. In livestock breeding the goal is introduction of a gene into the germ line to establish a desirable trait in a breeding line of livestock.

Summary

All of the above applications of recombinant DNA are variations on a theme: particular sequences of DNA can be isolated, manipulated, and re-introduced into many different kinds of living things. The desired endpoints are improvement of microorganisms, plants, and animals, for a particular purpose. These goals are not new, and have been achieved by techniques other than recombinant DNA. Recombinant DNA provides a new level of control to the genetic manipulation, which allows some new kinds of things to be done and certainly greater precision in the manipulations. In evaluation of the environmental implications of new microorganisms, plants, or animals, however, one cannot rely on the fact that recombinant DNA was used in their construction to indicate that they are necessarily safe or unsafe.

DISCUSSION

JAMES FAUSONE: Are there any limitations on recombinant DNA research in terms of techniques or handling methods that still have to be developed?

POST: There are many limitations; I have oversimplified the technique quite a lot, and every process has its own set of technical problems. First, there are limitations on the size of the DNA molecule that can be manipulated — it is hard to work with anything more than 20,000 bases long. There are also species limitations. It is very easy to work with E. coli and Chinese hamster ovary cells, but other bacteria and cell lines are much harder to work with. It is easy to work with tobacco, but not with soybeans and corn. It has become relatively easy to make transgenic mice, but no one has ever made a transgenic cow.

PARTICIPANT: Do these changes transmit to the next generation?

POST: For the kinds of things I talked about today, the goal is to transmit the changes to the next generation. One of the limitations is that, depending on the stability of the engineered trait, the changes are not always transmitted. But in order to make a usable vaccine or plant or animal, the trait must be amenable to transmittal to the next generation.

The one exception is human gene therapy, where the human genome is modified by recombinant DNA techniques to correct a genetic defect. Human gene therapy is a type of somatic therapy, so the changes are not inherited; no one proposes manipulating inheritance of the human genome. For example, one could provide sickle cell anemia patients with a normal hemoglobin gene in their blood cells — the trait would not need to be inherited. But for everything else that I have talked about, a goal of the research is to ensure that the trait is relatively stable.

PARTICIPANT: Why did the biotechnology industry develop sorapidly in the 1970s? Was the computer the catalyst?

POST: No, other factors fell in place to make biotechnology develop so rapidly. One was the discovery of restriction enzymes, which allowed scientists to manipulate and isolate discrete pieces of DNA. That discovery was probably the single biggest factor that made people realize that genetic engineering would be practicable. The DNA sequencing method was the other discovery that accelerated the biotechnology industry. Once we found what the sequences were, the sequencing went rapidly and DNA manipulations were much easier.

PARTICIPANT: When you have done one of these mixing procedures, how formidable a task is it to screen the results to find out which changes have taken and which have not?

POST: That is a good point, because most of the time involved in this process is devoted to screening. It is not hard to clone human genes, but you usually end up with what are called libraries of clones. It can take hundreds of thousands or millions of clones to produce the clone that you want. With plants, you usually end up making on the order of hundreds or thousands of plants to find a stable clone that effectively expresses the trait you want. So, it takes a lot of time to screen the results to find out if what you have behaves exactly the way you want it to.

1. U.S. CONGRESS OFFICE OF TECHNOLOGY ASSESSMENT, COMMERCIAL BIOTECHNOLOGY: AN INTERNATIONAL ANALYSIS, OTA-BA-218, Jan. 1984, at 3.