The Potential Evolutionary Impact of Genetic Engineering

Lecture © John Brzustowski
BIOL 606 Session, University of Alberta, March 1, 2000

In the micro view of evolutionary patterns, individuals may survive to produce gametes, some of which combine into zygotes that themselves develop into reproductively viable individuals which can disperse to new habitats. The machinery underlying this picture involves heritable characters that are transmitted from parents to offspring through DNA in chromosomes and organelles. This information is translated via RNA into proteins. Gametes can include novel DNA sequences formed by recombination of parental chromosomes, and/or by mutation. Microevolution can be thought of as small scale motion through an adaptive landscape, with a tendency for populations to get "trapped" at local fitness maxima.

Genetic engineering is a term encompassing many new technologies, such as plasmid or viral transformation of bacteria, gene insertion (gene guns, vectors etc.) into crop plants, site-directed mutagenesis, and gene therapy. Potential applications of these include improvements on input and output sides of crops, treatment of polluted sites, and curing genetic diseases.

The challenge of curing genetic diseases is to get a copy of a functional gene into each (appropriate) cell in an individual with a defective gene. The first successful application in humans was for adenosine deaminase deficiency (ADAD), and involved ex vivo insertion of a working AD gene by a retroviral vector into blood cells from patients. This led to similar attempts to cure other diseases, but with little success. Trying to understand why leads us to explore how retroviruses function and evolve.

Retroviruses carry their genome in RNA. They have proteins which allow them to bind to and enter host cells, and to reverse transcribe their RNA into DNA which they insert into the host genome. The host transcribes this "provirus" into viral mRNA's which get translated into virus proteins. These self-assemble into viral particles, and leave the cell, often without rupturing it.

Retroviral vectors are made by replacing the viral genome with an RNA copy of the target gene. Such a vector can get itself into a host cell, and (RT)-insert the RNA into the host genome, but since that RNA no longer codes for viral proteins, the process ends with this genetically enhanced host cell. Humans are not original in doing this: because retroviral reverse-transcriptase (RT) requires several strand transfers when generating the proviral DNA, recombination sometimes occurs, leading to virus particles carrying RNA coding for non-viral products, and possibly having additional non-viral proteins in their coats. These particles are still able to enter host cells and RT-insert their RNA into the host genome, but will not reproduce. In an infected host, viral numbers are high enough that such Natural Viral Vecors (NRVs) will be replicated by entering host cells already (or soon to be) infected by normal virus. Some of the virus particles produced in that host cell will likely include vector RNA. NRVs might explain how viruses escape the dead end posed by their gradual incorporation into host genome (which eventually gives the host and its offspring immunity to reinfection). A retrovirus must eventually switch host species to survive, and if a reasonable amount of blood is transferred from one species to another, it can carry many NRVs, some potentially much more suited to the new host than the original virus.

Most human interventions in evolution have involved micro-evolutionary timescales and linear mechanisms. Genetic engineering uses some tools that resemble the parallel evolution seen, for example, in retroviruses, and so safety assessments of particular tools for particular uses should consider such less frequent but equally important evolutionary mechanisms.


Dornburg, R. 1997. From the Natural Evolution to the Genetic Manipulation of the Host-Range of Retroviruses. Biol. Chem. 378:457-468.


Rapporteur: Curtis Strobeck

Directly following the presentation, there was a discussion on the effect of keeping deleterious recessive alleles around by gene therapy. It was pointed out that deleterious alleles should only accumulate at the rate of mutations and since most alleles present in the population are in a heterozygous condition the amount of selection against these alleles is very small.

After a break, the general discussion began and was opened by asking of audience if they liked the paper and the speaker why he had chosen the paper. The audience as a whole felt that it was a worthwhile paper to discuss as it exposed them to new concepts and data. The rest of the discussion was covered such topics as "Should we be more concerned about the economic effects of transgenics (e.g. Monsanto) instead of gene therapy?" and "Should we really be concerned about gene therapy since the cure of a genetic disease may cause another disease? The later was thought improbable since it is not necessarily introduced into the germline and only about 10% of the genome codes for a gene product. It was also asked "If ancient DNA could be use to look at the phylogenetic relationships between retroviruses. It was felt that the discussion was somewhat unfocused because lack of specific knowledge on the part of staff and audience.