Genetically engineered tomato

Patrick O'Neil patrick at corona
Thu Feb 16 01:43:20 EST 1995



On 13 Feb 1995, Brian Hachey wrote:

> > If, on the other hand, 
> > you grow that transformed bacteria in non-ampicillin media for a week or 
> > so - let's say, to be safe, three weeks - at the end of that time, you 
> > will most likely find no bacteria resistent to the antibiotic.  The gene 
> > will have mutated, and why not?  There is nothing selecting for a 
> > functional copy of the gene.  Moreover, the entire plasmid that harbored 
> > the gene in the first place is liable to be lost entirely unless it 
> > carries another gene that gives selective advantage in the circumstances 
> > within which the bacteria finds itself. 
>
> Do genes really mutate that quickly?
> What is the mutation rate and what is the average redundancy in a
> gene? 

The general mutation rate in bacteria and many other creatures, including 
us, is about 10-7 per base pair/per replication event.  This rather low 
number is the end result of base selection and proofreading by DNA 
polymerase added to the mismatch repair enzymes.  Even so, for E. coli, 
the doubling time (the time from one replication to the next) is about 20 
minutes under optimal conditions so over a relatively short period, they 
go through a great many replication events.  Their genome size is 
approximately 4 million base pairs.  
  For bacteria (and more for viruses), extraneous DNA is a burden.  
Bacteria that carry a plasmid with an ampicillin resistance gene (or any 
other antibiotic resistance gene), as well as a few other genes, do not 
compete favorably with wild-type bacteria under nonselective conditions 
that lack the extra DNA.  It takes time and energy to replicate and 
repair DNA and since the fastest reproducing bacteria wins, if a bacteria 
can dump extraneous, nonessential DNA, it is better off.  In addition, as 
with any gene in any creature, if a gene is not essential and doesn't 
confer a selective advantage given a certain set of environmental 
conditions, then it can sustain mutations without adversely affecting the 
host.  With an antibiotic resistance gene, if the bacteria carrying it 
are not being challenged by antibiotic, then the gene, under these 
conditions, is of no advantage and is just dead weight.  It can mutate 
beyond recognition and not hurt the bacteria at all.  That is not to say 
that it will have a higher REAL rate of mutation, which remains 10-7, its 
just that errors in extraneous DNA, being harmless, do not prevent 
bacterial propagation.  Until the bacteria eliminates the burdensome DNA, 
it will be mutated at no cost to the host.     

> I suppose my question really boils down to, given a medium
> which does not favour a specific gene, what is the average number of
> generations needed for that gene to be mutated out?

That is actually a very complicated question.  Some mutations within a 
gene will be silent (have no affect one way or another), most others will 
be deleterious, and a very few may improve function.  A mutation could 
affect the shape of the resulting protein, or alter the promoter that 
controls when and if the gene is expressed (it could be turned on 
uncontrollably, sucking up resources or otherwise hurting normal 
physiology), or it could be altered so as to never turn on.  The ultimate 
answer would require knowledge of what parts of the gene sequence are 
absolutely essential for normal function and the total length of the 
gene.  From this info you could then calculate the average number of 
replication cycles required for a deleterious mutation to occur within 
the gene. 
  The rate of mutation, of course, could be increased by environmental 
factors too, like concentrations of heavy metals, background radiation 
levels, etc.  I'm sorry, but I cannot give you a solid answer on this one.

> ...I suppose that most macroscopic
> organisms have lots and lots of redundant genes...  

Redundancy does exists but not for all genes.  It depends on the function 
of the gene product.  Ribosomes, used in translation of genes, are 
generally multicopy.  Any gene product that is needed in high 
concentration are often multicopy.

> Hmm...  I suppose
> I really should read an intro to genetics.  Any suggestions?

There are lots of possible books.  The Molecular Biology of the Gene is 
one (a textbook and the authors I cannot recall), Genes V is another, 
Recombinant DNA by Watson, Gilman, Witkowski, and Zoller, many 
biochemistry texts, etc.  The Recombinant DNA book is especially a nice 
source of basic information about various recombinant DNA techniques.  

Patrick



More information about the Plantbio mailing list