Mutations: The Good, the Bad, and the Neutral (Part 5)

Lesson 5: A single mutation cannot create a new species

There was some debate during Dobzhansky’s time about whether new species could be formed through the birth of rare ‘monsters’.  Such monsters would be the product of a mutation, and would be effectively shut off from the rest of the population, forming their own species.  This theory was called saltationism.  Saltationists argued that a single mutation can have a very large effect.  A mutation somehow produces fundamental changes to the species, such that a new species is formed.

At the other end of the spectrum were those who thought that mutations could only produce superficial changes in an organism, and could not be responsible for the formation of new species.

Against the latter, argues Dobzhansky: a single set of wings and a set of balancers is a diagnostic character of dipterans (true flies).  An insect with four wings is, by definition, not a dipteran.  What, then, do we do with a certain fruit fly mutant, whose pair of balancers develop into a second pair of wings?  This is now a four-winged dipteran – do we call that a fundamental, or a superficial, change?  The implication is that, if such a mutation could occur within dipterans, then perhaps other characters that we use to distinguish other species could similarly evolve via simple mutations.

Against saltationists, Dobzhansky argues that this wing mutation does not alter anything else in the fruit fly – it is still, to all effects and purposes, a fruit fly.  Since most mutations are of small effect, many genes would need to produce many mutations in order for a new species to emerge.  But this could simply never happen at once – mutations would have to slowly build, from generation to generation, to produce a new species.

There is only a single way in which a new species can be born in a single generation: polyploidy, which we will discuss later.

Lesson 6 – Some mutations can be neutral

We haven’t really discussed this yet, so I thought  I would mention it briefly here.  Dobzhansky reported an experiment in which fruit flies which were missing an entire section of one of their chromosomes were bred with other fruit flies missing the same section.  These parents could survive because they still had one copy of the missing genes in their normal chromosome.  25% of their offspring, however, were homozygous for the missing section (that is, they had no copies at all of these missing genes).

In many cases, this gene deficiency was lethal.  But sometimes a fruit fly could be missing a gene entirely and it would have no effect on the fly at all.  Dobzhansky believed this had to be for one of two reasons: either certain genes have larger effects than others, or certain genes were multiplied throughout the genome, so that the loss of one gene is made up for by the presence of another.  In either case, one could imagine a less drastic scenario in which all of the genes are present but a mutation occurs in one of them; such a mutation would not be detectable, as it would produce no effect.  This mutation would be neutral.

Today neutral mutations in non-coding regions of DNA (regions that are not genes) are the cornerstone of molecular ecology.  These mutations hide from selection, and so can be used to determine relatedness between species and populations, and can be used to estimate the time at which a species diverged from another species.  People in my lab currently use these neutral markers to detect the movement of fish in western Canada after the last ice age.

Lesson 7: Mutations are random

In the 1800s it was believed that the environment caused internal changes that directly led to adaptation.  Genetics destroyed such a notion.  X-raying fruit flies induced mutations, but these mutations were in any direction, from lethal to beneficial, and were decidedly not adaptive to an X-ray environment.  So if an environmental change were to trigger mutations, there was no telling what the consequences would be.  Adaptation was only one possible scenario.  For proof, consider the experiments that have already been described, such as the one in which X-raying a bunch of males resulted in everything from lethal mutations to neutral mutations, with no evident pattern.  Says Dobzhansky, mutations ‘remain haphazard’.  We cannot experimentally induce a mutation in a target gene - the mutation will occur ‘wherever’ -  nor can we force a desired mutation to arise.

Lesson 8: Mutations are not completely random

When we say that a mutation is random, what we really mean is that we cannot determine when or where or of what nature a mutation will be.  There is nothing deterministic about mutations; many today argue that mutations are random in a quantum physics sense, in which future mutations are inherently unknowable, even if given perfect knowledge about the current state of the world and its past.

‘Random’ is almost an unfortunate word to use, however, because it could also imply something different.  Random could mean that all genes are equally likely to mutate, that all regions of a gene are equally likely to mutate, and that all organisms have the same mutation rates for each section of their genome.

But this definition of random does not hold.  Mutations are more likely to occur in some species than others; mutations are more likely to occur in some genes than others.  And mutations are more likely to occur in some directions than others.  We have, for instance, already noted that mutations are more likely to be harmful than beneficial in the environment to which an organism is already adapted.

Dobzhansky asks us to consider the rise of lethal mutations on the X chromosome of fruit flies.  If all mutations are equally likely, lethals should arise at an equal frequency in all fruit fly population, but this is not the case.  Some populations are more prone to developing lethal mutations than others.

And consider genes.  Some genes are more mutable than others.  In corn, genes responsible for colour would produce mutations in 492 out of one million gametes; genes responsible for producing shrivelled kernels would produce mutations in only 1.2 out of one million gametes. 

This seems fairly stunning.  Of course, it is possible that mutations are simply more likely to disrupt the effect of the colour gene, but be neutral within the shrivelled gene.  But evidence today supports Dobzhansky’s claim that certain genes are more likely to mutate than others.

Also consider that there can be directionality in mutations.  Fruit flies of different eye colours were exposed to the same level of X-rays.  Red, apricot and eosin eyes mutated more frequently to white eyes than to any other colour.  White eyes to red was never seen, but white to eosin to red was.  However, directionality is not a rule: for bristles, the mutation rate from wild to forked bristles was the same as the reverse.

Even this directionality differed from species to species or population to population:  Fruit flies from Russia were most likely to mutate from red eyes to white, while American fruit flies mutated from red to white or red to eosin or red to apricot equally frequently.  Furthermore, the Russian mutation rate was faster.

Why?  Why do species and genes differ in mutation rates?  On this, Dobzhansky was silent.  Today we know that mutations are often the result of copy errors when DNA is being produced, and the enzyme that copies DNA (called DNA polymerase) has a different structure from species to species.  Some structures are more prone to making mistakes than others, resulting in species-level differences in mutation rates.

‘The pace of evolution is not alike in all organisms.  Some groups seem to possess an unlimited store of variation and evolve rapidly, while others are conservative and undergo no change during geological epochs.’ 

We are now done chapter two of Genetics and the Origin of Species.  In the next chapter, Dobzhansky will argue that these principles of mutations are the basis for population and species-level differences among organisms.