theories of dominance

Toby Bradshaw toby at stein.u.washington.edu
Fri Mar 26 11:18:31 EST 1993


In article <robison1.733101702 at husc.harvard.edu> robison1 at husc10.harvard.edu (Keith Robison) writes:

[snip]

>Situation II
>	Dominant mutant allele
>	Recessive wild-type allele
>
>	Frequent molecular explanations:  
>	1) Dosage of this gene is critical
>	and is related to copy number.  Hence, one good copy will not
>	suffice (many developmental genes fall into this category)
>	2) E.g. retinoblastoma:
>	a non-functional allele has a dominant tumor phenotype because
>	the odds of a somatic mutation inactivating the wild-type
>	suppressor are high.  Because the "screen" involved (tumor
>	growth) is very sensitive, we detect these somatic mutations.

I thought of this example, but it is a straight case of tumor
formation recessive to tumor suppression.  Germ-line homozygotes
get bilateral retinoblastoma, and the heterozygotes can become
homozygous as you note above, usually only getting a tumor in
one eye.  Maybe the phenotypes of heterozygotes and homozygotes
would lead a classical geneticist to conclude that the alleles
are codominant, measured by the number of eyes with tumors :)
(hard to justify a smilie in cases of childhood tumors leading
to blindness, it's just there to reinforce Keith's point that
phenotypes can be a poor guide to the molecular details).

>	3) The mutant is a gain-of-function allele altering
>	the properties of the gene but does not destroy them.
>	Perhaps the new allele binds more tightly than before to another
>	molecule, or perhaps it binds something else.  Or, perhaps it
>	has lost a functional site required for regulation (some
>	oncogenes fall into this category -- mutations cause them
>	to be stuck on).
>	4) Mutant gene product "poisons" the system.  An example is
>	resistance to streptomycin in prokaryotes -- resistance is
>	generated by mutant ribosomal subunits which cannot bind
>	antibiotic.  Wild-type ribosomes stall on the mRNA in the
>	presence of antibiotic, and act as road-blocks for mutant
>	ribosomes.  Another frequent example is if the gene product
>	forms a homodimer -- wild-type/mutant dimers may be non-functional.
>
>
>Hence, knowing how things are dominant/recessive (or co-dominant) will
>not necessarily tell you much at the molecular level in itself.  More 
>likely, once you know the molecular picture you can predict the
>dominance of various mutations.  Dr. Bradshaw makes a good point
                                  ^^^^^^^^^^^^
Hey, even my mother doesn't call me "Dr." :)

>that dominant--recessive systems often appear to be co-dominant when
>viewed from a different viewpoint.  
>
>	
>toby at stein.u.washington.edu (Toby Bradshaw) writes:
>
>>In article <BRIANF.93Mar24160949 at dna.uvm.edu> brianf at dna.uvm.edu (Brain Foley) writes:
>
>>>Many students in molecular genetics are confused about what "dominant"
>>>and "recesive" mean.  Most often, recessive means that some gene
>>>product is lacking, so if a cell-cell hybrid is made, the cell with
>>>the dominant phenotype provides the lacking enzyme.  But this is not
>>>always the case.
>
>>OK.  It's late and my memory isn't the best.  I'm having a hard
>>time coming up with an exception to the above.  Care to help me
>>out?
>
>See Situation II, sub-cases 3 and 4.  

I don't really think that sub-case 3 is an exception.  The 
recessive "not neoplastic" phenotype is due to lack of a 
dominant allele for "neoplastically transformed".  Sub-case 3 
is different from the most common cases of dominance only in
that the mutant allele is dominant to the wild-type.  The same 
argument applies to sub-case 4, as far as I can tell, with 
streptomycin sensitivity dominant to resistance.  The recessive
phenotype, resistance, is due to lack of a gene product (the wild-type
ribosomal protein), and a cell fusion (or mating) experiment would give
you the answer as to which is dominant, as Brian has suggested.

>>>I think that "dominant" can be a fuzzy term in some cases now.
>
>Dominance is a very explicit term, but using it requires carefully defining 
>the system you are discussing and it's boundaries.  For example, suppose
>I am speaking of sickle-cell anemia.  Wild-type "S" is clearly dominant
>over mutant "s" at sea-level, but at high-altitudes they are co-dominant
>(and I suppose at really low oxygen partial pressures the dominance
>reverses).  At the molecular scale, however, you could argue that
>in terms of producing hemoglobin S and s are _always_ co-dominant --
>presence of an S allele does not _prevent_ generation of s-type chains.
>Put another way, do you define phenotype as the health condition or
>what you might see on a gel?  

Toby Bradshaw                       |
Department of Biochemistry          |  Will make genetic linkage maps
and College of Forest Resources     |            for food.
University of Washington, Seattle   |
toby at u.washington.edu               |



More information about the Mol-evol mailing list