The origins of recessive genes

[James_Frogan]

Hi everyone,

I'm studying physics and not biological sciences, but I've been wondering about recessive genes recently. Given my background, do be kind on the explanations. My question is: if a recessive gene tends to be overcome by the dominant gene, how do recessive traits still display today?

What I understand of genes: if there is a dominant gene in a pair, then the dominant gene trait will show (eg black hair in black-blonde combination). If 2 people with black hair have a child, it's possible to have a blond baby if they both have black-blonde genes.

From a mathematical viewpoint, the probability of black-black and blonde-blonde producing a black-blonde is 100%. The probability of a black-blonde and a black-black producing a black-blonde is 25% (and black-black 75%). And 2 black-blondes have 50% chance of black-blonde and 25% chance pure black or blonde.

If the early human population were mostly black haired, how did blonde haired people defy the probabilities and become numerous? They would mathematically start off and die shortly after, if I see my probabilities correctly. Obviously this is not the case, so what am I misunderstanding about genes?

Cheers for your responses :)

 

[atyy]

Gentlemen Prefer Blondes

But if they didn't, there would be Hardy-Weinberg equilibrium

That Hardy was, BTW, the famous pure mathematician.

 

[Ygggdrasil]

Two points:

1) Recessive traits will not die off on their own. Although the proportion of individuals with the recessive trait is small, there will still be some carriers of the trait (individuals who show the dominant trait but carry one copy of the recessive gene, in the example above the black-blonde individuals). If you do the math, you can find the proportion of people with the recessive trait will reach an equilibrium value, known as the Hardy-Weinberg Equilibrium.

2) Deriving the Hardy-Weinberg equilibrium requires assuming that every individual has an equal opportunity of reproducing and that reproduction is a random process. This is not true for many traits. Traits can be under natural selection such that individuals with a certain trait have a much higher probability of reproducing than individuals without that trait. Sometimes this can be due to a fitness advantage (individuals with a certain trait are able to outcompete others for resources) or due to sexual selection (individuals with a certain trait, due to biological or cultural reasons, are able to outcompete others for mates). If blonde hair were an adaptation that promoted survival as humans moved to more Northern climates, for example, then this could explain how the number of blonde individuals become more numerous over time.

Blonde individuals could also become more numerous for other reasons, such as luck or chance. For example, consider if only a small number of prehistoric humans migrated from Africa to Europe. These humans in Europe would have a large area with many resources, so their population could expand greatly. If this small group by chance had an unusually large proportion of individuals with blonde hair (or carriers of the blonde hair gene), this could have contributed to the increase in the proportion of the blonde hair trait. This effect is known as the bottleneck effect or the founder effect.

 

[bobze]

To add to the excellent post that Yggg has put up; Its actually a lot more complicated than just "dominant or recessive". Most genes aren't really an easy binary system like that and there is some degree of expression from both genes that contribute to the phenotype.

For the sake of learning genetics "pure" dominant/recessive traits are used, but that is only a very tiny fraction of the picture.

 

[James_Frogan]

Thanks atyy, Ygggdrasil and bobze.

I omitted the mutation that occurs to adapt to environments and the various other factors mentioned.

About the Hardy-Weinberg problem, doesn't the recessive trait population q^2 require a large enough sample size? For the first person that received a blonde-blonde or albino-albino gene, that would make his q^2 value a very very small number?

Cheers ;) learning so much from this

 

[epenguin]

Traits can be under natural selection such that individuals with a certain trait have a much higher probability of reproducing than individuals without that trait. Sometimes this can be due to a fitness advantage (individuals with a certain trait are able to outcompete others for resources) or due to sexual selection (individuals with a certain trait, due to biological or cultural reasons, are able to outcompete others for mates). If blonde hair were an adaptation that promoted survival as humans moved to more Northern climates, for example, then this could explain how the number of blonde individuals become more numerous over time.

This is the explanation (incomplete IMO but OK enough now) usually given to the blonde and light-skinned trait. That light skin allowed better vitamin D synthesis in the northern places which not much sun. But in Africa and Australia you have enough sun not to need it plus you need to wear less clothes - white then becomes a disadvantage. E.g. white farmers in Australia get a high frequency of skin cancers.

The largest lot of populations in Africa that you would call definitely 'black' may owe their predominance to this blackness. That is they are a successful offshoot which though predominant part of the population is minority part of the continent's genetic variation, they are relatively recent, originating around 60,000 years ago I heard. The closer-to-original trait is the colour of Bushmen and Pygmies - more of a reddish-brown. Considerations always subject to evolving evidence and I am not up to date. Put it past your teacher.

Remember a selection factor does not need to be very strong to predominate over time.

 

[thorium1010]

The largest lot of populations in Africa that you would call definitely 'black' may owe their predominance to this blackness. That is they are a successful offshoot which though predominant part of the population is minority part of the continent's genetic variation, they are relatively recent, originating around 60,000 years ago I heard. The closer-to-original trait is the colour of Bushmen and Pygmies - more of a reddish-brown. Considerations always subject to evolving evidence and I am not up to date. Put it past your teacher.

According to wikipedia first modern humans diverged from their common ancestor 200000 years ago.

Anatomically modern humans first appear in the fossil record in Africa about 195,000 years ago, and studies of molecular biology give evidence that the approximate time of divergence from the common ancestor of all modern human populations was 200,000 years ago.

http://en.wikipedia.org/wiki/Human

Skin color is older than modern human divergence. Again from wikipedia -

Rogers et al. (2004) performed an examination of the variation in MC1R nucleotide sequences for people of different ancestry and compared the sequences of chimpanzees and humans from various regions of the Earth. Rogers concluded that roughly five million years ago, at the time of the evolutionary separation of chimpanzees and humans, the common ancestors of all humans had light skin that was covered by dark hair.Over time human hair disappeared to allow better heat dissipation through sweating[3] and the skin tone grew darker to increase the epidermal permeability barrier[43] and protect from folate depletion due to the increased exposure to sunlight.[4] By 1.2 million years ago, around the time of homo ergaster and homo erectus, the ancestors of all people living today had exactly the same receptor protein as modern Africans

http://en.wikipedia.org/wiki/Human_skin_color

 

[Murdstone]

The details of dominant/recessive genes is missing from everything I have read. Sure - blue eye dominance or something is mentioned but the mechanics are not Is it that you have slightly different base sequence for a gene attributable to a particular trait - flavors if you like.

Then there is the whole thing of what makes a gene dominant?

 

[thorium1010]

The details of dominant/recessive genes is missing from everything I have read. Sure - blue eye dominance or something is mentioned but the mechanics are not Is it that you have slightly different base sequence for a gene attributable to a particular trait - flavors if you like.

Then there is the whole thing of what makes a gene dominant?

http://en.wikipedia.org/wiki/Dominance_%28genetics%29

Blue eye is not dominant, its a recessive trait.

 

[Moonbear]

Hair color isn't a good example. It isn't controlled by a single gene and you get a lot of ranges of color, and not even every hair of one individual is the same color. (Also a bad example because of the prevalence of hair dyes.)

I do want to refine the terminology being used a bit. When talking about dominant and recessive, we're referring to alleles for a gene, not the gene. The gene would be for something like hair pigments, but then there are different alleles that are specifying which pigment or if it's expressed.

The other thing to remember is that recessive doesn't mean bad. Sometimes, a dominant allele is the bad one, and you get a high prevalence of a recessive trait because both homozygotes and heterozygotes for the dominant allele die off faster than the homozygotes for the recessive trait.

 

[bobze]

I do want to refine the terminology being used a bit. When talking about dominant and recessive, we're referring to alleles for a gene, not the gene. The gene would be for something like hair pigments, but then there are different alleles that are specifying which pigment or if it's expressed.

The other thing to remember is that recessive doesn't mean bad. Sometimes, a dominant allele is the bad one, and you get a high prevalence of a recessive trait because both homozygotes and heterozygotes for the dominant allele die off faster than the homozygotes for the recessive trait.

Moon-- two great points.


I want to iterate again for others, the importance of the idea that simple "dominant and recessive" alleles tend to be the exception, rather than the rule for most genes we study. Not to mention complicating the fact that environment impact expression and degree of expression as well. Mendel was, quite possibly the luckiest biologist to have ever lived.

 

[Murdstone]

Is there a biochemical difference in the gene for a trait depending upon which allele of the trait it codes for?

I know you have emphasised that phenotype is not straight forward, however, I am trying to establish if whether the concept of biochemical dominance has merit.

 

[Ygggdrasil]

The biochemistry of the different alleles definitely affects whether the allele is dominant or recessive, but the connection is not always so straightforward. Many times a "broken" version of a gene is recessive. For example, cystic fibrosis is caused by a mutation in an ion channel called the cystic fibrosis transmembrane conductance regulator (CFTR) that prevents the ion channel for working properly. The disease allele is recessive because one functioning allele is sufficient to make enough CFTR for the body to behave normally. Only when both copies of the gene are non-functional does the disease occur.

Some "broken" version of genes, however, will inactivate the functioning copies of the genes and thus behave as dominant alleles (these are referred to as dominant negative alleles as they are dominant alleles that cause a loss of function). This is especially common when the proteins encoded by the genes function in a complex and one broken subunit in the complex is enough to render the entire complex non-functional. Another way for an allele to be dominant is though a gain-of-function mutation, in which the allele confers its trait by performing some new function (for example, an enzyme that cannot be turned off properly).

Here's a particularly good explanation of the topic with more examples:
http://www.thetech.org/genetics/ask.php?id=227

 

[Murdstone]

Thanks g - The article was very informative. Your willingness to help with focused information is appreciated.

 

[Murdstone]

In reflecting upon the information in the excellent article on the biochemical differences in gene type (dominant) (recessive), the distinction between a mutant gene and a recessive gene seem blurred to me.

Both mutant and recessive deviate from established. They can either code for a slightly different protein or no protein.

I would consider a recessive gene a subset of mutant gene.

 

[epenguin]

In reflecting upon the information in the excellent article on the biochemical differences in gene type (dominant) (recessive), the distinction between a mutant gene and a recessive gene seem blurred to me.

Both mutant and recessive deviate from established. They can either code for a slightly different protein or no protein.

I would consider a recessive gene a subset of mutant gene.

Just continue your reading and study without over-worrying to completely define words that you may find are sometimes used loosely without causing real misunderstanding. A recessive gene generally (with the exceptions mentioned by Ygggdrasil) is an unfunctional one as you say, and most often I guess because it codes for an unfunctional protein. A mutant gene means really an altered one, that is it is mutant with respect to its ancestor (Latin mutare - to change.) A mutation may result in either a functional or an unfunctional gene. All the genes you have are mutant though, with respect to some more or less distant ancestor.

If it's not confusing a mutation can sometimes restore functionality to an unfuctional gene. This can either be a 'back mutation' that restores the original gene sequence, or a mutation that restores not the polynucleotide sequence but the primary protein structure with a difference polynucleotide sequence, given the degeneracy of the genetic code, or some mutations can produce a new protein which have two mutations, two differences in protein primary structure compared with ancestor, but which compensate each other so as to give a functional protein.

 

[Ygggdrasil]

In reflecting upon the information in the excellent article on the biochemical differences in gene type (dominant) (recessive), the distinction between a mutant gene and a recessive gene seem blurred to me.

Both mutant and recessive deviate from established. They can either code for a slightly different protein or no protein.

I would consider a recessive gene a subset of mutant gene.

Mutations to a gene can also create dominant alleles, however. One good example here is the allele for lactase persistence (https://en.wikipedia.org/wiki/Lactose_intolerance#Lactase_persistence).

The ancestral state of this allele (in humans prior to their migration from Africa) encodes the enzyme responsible for breaking down lactose. This gene is regulated such that it gets produced in infancy and early childhood, but later gets turned off in adulthood, resulting in lactose intolerance as adults. In European populations, mutations arose in this gene that disabled the ability of the gene to be turned off in adulthood, creating alleles for lactase persistence. These alleles are dominant alleles associated with the trait of lactose tolerance in adulthood. Thus, in this case, the dominant allele is actually the mutant form of the gene.

 

[MrRagnarok]

It's not that a recessive allele is an undesired gene and causes a nonfunctional protein. A recessive allele will simply not be expressed in the presence of a dominant allele. It recedes expression to the dominant allele. Eye color is a good example.

To keep things overly simple, let's pretend everyone has a single gene that codes for the color of their eyes. Recall that each human has two copies of every gene, so you would have two copies of this eye color gene. Each copy is called an allele. Since we have two copies they can compete with each other for expression. Let's call the brown colored eye protein B and the blue colored eye protein b. With two possible proteins and two copies of the DNA we can have three different combinations. BB, Bb or bb.

When we have BB both instructions for eye color produce a brown colored eye.
When we have bb both instructions for eye color produce a blue colored eye.

What happens when we have Bb? One copy for brown and one for blue? Turns out the brown color wins out over the blue color. Because of this it is called 'dominant'. The blue pigment coded by b is not inferior to the brown pigment in B. But simply because the brown B will [STRIKE]silence[/STRIKE] overpower the expression of the blue b, brown (B) is dominant and blue (b) is recessive.




Changing gears here a little bit here is good because, like Ygggdrasil pointed out, dominant/recessive is not the only way genes influence protein production. In fact, if dominant/recessive was the only way genes were expressed, organisms would not change much after the original genetic programming during fertilization. Cells live in an always changing environment and must be able to dynamically change as well.

For example: Gene A (which is dormant until a stressful environment) inactivates Gene B. Normally, Gene B is highly expressed, but when the cell senses stress it rapidly produces Gene A, which in turn inactivates Gene B. Shown below:

Environmental stress off: [gene A] inactive GENE B active
Environmental stress on: GENE A active product inhibits [gene B]

When the environmental stressor is relieved Gene A is down-regulated and releases inhibition from Gene B which is expressed again. Thus the organism returns to normal baseline behavior. This is an alternative way of regulating genes that is responsive to the environment and can temporarily silence Gene B 'as if' it were recessive. It is through these 'epigenetic' mechanisms that cells can control their gene levels without mutation. I.e. The genetic code in Gene B was at no time altered during its silencing. So once Gene A is removed Gene B functions as usual. A mutation would alter the code of Gene B in a way that could permanently disable or alter the protein product.

There are also ways to permanently turn genes on and off, but I think I've already gone beyond the scope of this thread!

 

[Murdstone]

My understanding now is that alleles are slightly different coded genes for the same trait.

Most understand the BB, bb, Bb model. Where things become unclear is in explaining why B is dominant over b? There is a biochemical difference between B and b, what is it about this biochemical difference that makes B dominant over b?

While not objecting, I am of the understanding that Biological Evolution downplays the role of environment, stress, in initially determining which genetic combinations are offered for selection. The standard dogma is that these initial offerings are "random". Effects of environment - ex ante - no, ex post - yes.

 

[Ygggdrasil]

It's not that a recessive allele is an undesired gene and causes a nonfunctional protein. A recessive allele will simply not be expressed in the presence of a dominant allele. It recedes expression to the dominant allele. Eye color is a good example.

To keep things overly simple, let's pretend everyone has a single gene that codes for the color of their eyes. Recall that each human has two copies of every gene, so you would have two copies of this eye color gene. Each copy is called an allele. Since we have two copies they can compete with each other for expression. Let's call the brown colored eye protein B and the blue colored eye protein b. With two possible proteins and two copies of the DNA we can have three different combinations. BB, Bb or bb.

When we have BB both instructions for eye color produce a brown colored eye.
When we have bb both instructions for eye color produce a blue colored eye.

What happens when we have Bb? One copy for brown and one for blue? Turns out the brown color wins out over the blue color. Because of this it is called 'dominant'. The blue pigment coded by b is not inferior to the brown pigment in B. But simply because the brown B will silence the expression of the blue b, brown (B) is dominant and blue (b) is recessive.

This is incorrect. Dominant alleles are not silencing recessive alleles. In most cases, whether a particular allele is dominant or negative depends on the biochemistry of the gene and the effect of the mutation on the biochemistry. For a good example, see the explanation for why the gene for red hair is recessive in the link I posted earlier (http://www.thetech.org/genetics/ask.php?id=227 ).

 

[MrRagnarok]

My understanding now is that alleles are slightly different coded genes for the same trait.

Exactly

Where things become unclear is in explaining why B is dominant over b? There is a biochemical difference between B and b, what is it about this biochemical difference that makes B dominant over b?

There are many ways that a dominant allele overpowers a recessive allele. The eye color gene encodes for melanin and B is a darker colored melanin while b is lighter. The darker color pigments in B reflects more than just blue wavelengths of light so it gives a darker appearance to the observer. Thus, in eye color both B and b are being expressed intracellularly but only B is being observed. Thus the dark brown phenotype. Link.

There is also an albino allele (a) for the eye-color gene. a is the absence of color (often due to impairments in melanin processing) and is recessive to b, so B>b>a. If our genotype was ba the we would have blue eyes even with one b simply because the blue pigment expressed by b would be more robust visually than the a pigment (which is actually a lack of pigmentation).

So there is not really a direct biochemical mechanism for eye color but rather the property of the B protein behaves differently (in terms of appearance) than b. Simply B is easier to see than b. But other dominant genes do exert competitive pressures on recessive genes. For example, let's say we have the Z dominant and z recessive protein. DNA for Z is written in an easier way to transcribe than z and this transcription efficiency makes it 20x easier to transcribe Z compared to z. This ease of transcription will result in baseline concentrations of the Z protein being 20x higher than z protein. If there are normally 1000 copies of this protein in the cell, Zz genotype will have 950 Zs and 50 zs. Even though they have the same number of alleles. In this way the Z is out-producing the protein levels vs z causing a Z trait to be expressed.

Another biochemical mechanism dominant alleles can use is RNA interference. Recall that the DNA codes for RNA and this RNA contains introns in the gene that are removed and do not directly influence the protein. These intronic pieces of RNA are usually degraded but some are designed to be complementary to the recessive allele's DNA and when this strand of RNA finds the recessive allele it attaches to the DNA in a way that inhibits this allele's transcription rendering it recessive.

This is not the end of the story and not all genes work this way, but simply some examples of what can happen.

Biological Evolution downplays the role of environment, stress, in initially determining which genetic combinations are offered for selection. The standard dogma is that these initial offerings are "random". Effects of environment - ex ante - no, ex post - yes.

I am unfamiliar with the field of bio evo but am familiar in epigenetics. You cannot directly change (without mutation which occurs so randomly it can be ignored) the sequence of the DNA you inherit from your mother and father. So in this regard, these initial genetic offerings are random in a sense that you get 50% mom 50% dads DNA.

But the cellular environment, even in the womb, greatly influences levels of genes that are expressed or silenced. For example a mother who is stressed while carrying a child can affect the levels of stress receptors (called glucocorticoid receptors) expressed by the fetus because the stress hormones (such as cortisol) present in the mother influence the levels of expression in the developing fetus. Outside of the womb during development, stressful times to the children (such as abandonment or reduced nurture) cause these stress receptors to be overly expressed in a way that low levels of stress cause a higher level of stress response, this is often observed as anxiety.

Other factors, such as cocaine intake by the mother, will also play a role in the expression levels of fetal (and eventually adult) proteins and anxiety levels, but science is still determining exactly what is happening there!

 

[MrRagnarok]

This is incorrect. Dominant alleles are not silencing recessive alleles. In most cases, whether a particular allele is dominant or negative depends on the biochemistry of the gene and the effect of the mutation on the biochemistry. For a good example, see the explanation for why the gene for red hair is recessive in the link I posted earlier (http://www.thetech.org/genetics/ask.php?id=227 ).

Well, the bulk of my post was correct. Just the usage of the word 'silence' in the last sentence was wrong so I fixed it. I agree with you in the case of eye color that b is not being silenced and have explained it more accurately in a more recent post.

But some alleles can silence other alleles. Just typically not alleles for the same gene. I posted a bit about this before I saw this post, but also in the case of genomic imprinting, maternal alleles can silence paternal alleles (or vice versa). But my understanding is that the imprinting field is still uncertain exactly how this is occurring (for instance how the cell differentiates between maternal and paternal autosomes) Link 1 Link 2

 

[Ygggdrasil]

It's imporant to distinguish between the interactions between alleles at the same location on the chromosome (locus) and interactions between alleles at different loci.

Because we have two copies of each chromosome, we have two alleles of every locus in our genome. In the case where we have two different alleles at a particular locus, one allele can be dominant over the other. This dominance generally does not result from silencing or altering the protein produced by the recessive allele. Rather, how the dominance occurs is a result of the properties of the different alleles (e.g. whether the alleles are loss-of-function or gain-of-function).

As you mention, particular allele at one location on the chromosome can alter the expression another allele at a different locus. These interactions between different genes are often very important for determining the phenotype of an organism. However, having one gene alter or mask the effect of a gene at a different location is not an example of dominance, as dominance relates only to the relationship between alleles that reside at the same locus. Rather, the interactions between alleles that reside at different locations in the genome is referred to as epistasis.

I'm sorry if this is a bit pedantic and nitpicky, but students often get confused by the differences between genes and alleles, so I often find myself having to clarify this issue. Most of your posts here are correct, informative, and bring up important points, but the difference between dominance relationships vs epistasis is subtle and requires careful explanation. I'm sure all of this information is already clear to you, but I just want to make sure others reading the thread have a clear idea of what's going on.

 

[MrRagnarok]

Thanks for the precise clarification Ygggdrasil, it is important to not use terminology loosely. I thought that it was more common for the dominant allele to exert dominance over the recessive allele through some sort of direct or indirect inactivation mechanism (like Xist-mediated X chromosome inactivation). I learned this in school as the classic example, but further research now suggests, as you mention, that this appears to be an extreme example.

What happens in the case of less extreme mutations? For example, in the red hair example a mutated MC1R product impairs the natural degradation of red pigmentation. This mutation is pretty extreme since it causes a loss-of-function mutation for the MC1R protein. How is dominance exerted when two heterozygous alleles differ only in an exonic SNP (for others reading: one single base pair)? Let the SNP cause a missense (protein changing) and not nonsense (protein stopping) mutation. But the individual will still have two different proteins caused from two different (albeit slightly) alleles. Is dominance exerted here? Or does it just depend on how the missense mutation ultimately affects the final protein product?

 

[bobze]

What happens in the case of less extreme mutations? For example, in the red hair example a mutated MC1R product impairs the natural degradation of red pigmentation. This mutation is pretty extreme since it causes a loss-of-function mutation for the MC1R protein. How is dominance exerted when two heterozygous alleles differ only in an exonic SNP (for others reading: one single base pair)? Let the SNP cause a missense (protein changing) and not nonsense (protein stopping) mutation. But the individual will still have two different proteins caused from two different (albeit slightly) alleles. Is dominance exerted here? Or does it just depend on how the missense mutation ultimately affects the final protein product?

I pointed this out a few pages back, the answer to your question. Most protein coding genes (structural genes) aren't so simple as dominant or recessive. Most structural genes have some varying levels of codominance.

Lots of times we classify things as "dominant and recessive" based on the "big picture" (like clinical picture).

Think about sickle cell anemia. I'm sure you've learned ad nauseum that SCA is autosomal recessive in your classes. But when we call it this we are talking about how it looks clinically, not how it really works out at the molecular level.

Heterozygotes for SCA still produce the B-globin chains of sickle cell disease. If you run a western blot of a SCA carrier you can clearly see this. However, because of how hemoglobin loads oxygen (positive cooperativity) you normally don't get any SCA phenotype in the amounts the sickle cell B-globin are produced. But, under hypoxemic conditions (read; low Hb O₂ saturation) this can expose the amino acids responsible for polymerization of Hb and cause cell sickling. You see this clinically when an unknown carrier of SCA spends prolonged periods at high altitude and suddenly becomes symptomatic.

 

[Ygggdrasil]

What happens in the case of less extreme mutations? For example, in the red hair example a mutated MC1R product impairs the natural degradation of red pigmentation. This mutation is pretty extreme since it causes a loss-of-function mutation for the MC1R protein. How is dominance exerted when two heterozygous alleles differ only in an exonic SNP (for others reading: one single base pair)? Let the SNP cause a missense (protein changing) and not nonsense (protein stopping) mutation. But the individual will still have two different proteins caused from two different (albeit slightly) alleles. Is dominance exerted here? Or does it just depend on how the missense mutation ultimately affects the final protein product?

The MC1R gene encodes a transmembrane receptor protein (the melanocortin 1 receptor) that is important for regulating the expression of many genes in melanocytes, the cells responsible for producing pigmentation. When MC1R signaling is low, melanocytes produce primarily a red/yellow pigment (phaeomelanin), and when MC1R signaling is high, the melanocytes instead produce a brown/black pigment (eumelanin).

The alleles responsible for red hair in humans encode defective versions of the MC1R receptor. Thus, if both copies of the MC1R gene are defective, MC1R signaling will be low, and the melanocytes will produce much more phaeomelanin than eumelanin, resulting in red hair. As it turns out, if one copy of the MC1R gene is normal and the other copy is defective, the normal copy is able to signal well enough to still allow the cell to produce eumelanin. The case where one working copy of the gene is capable of performing the function of the gene is called haplosufficiency. Many genes exhibit haplosufficiency, so many loss-of-function alleles end up being recessive.

Of course, not all genes display haplosufficiency. Sometimes having one working copy of a gene does not produce enough protein for the cell to function normally, and we get a case of haploinsufficiency. Interestingly, individuals who are heterozygous for the red hair allele, while they may not have red hair, do show increased sensitivity to UV radiation. Therefore while the defective MC1R allele is recessive with respect to red hair color, it displays partial dominance with respect to sensitivity to UV radiation. As you can see from this case, whether a particular allele acts as a dominant, recessive or other type of allele depends a lot on the particular trait you are asking about, and the same allele can display different types of dominance for different traits.

While above we have examples of loss-of-function alleles acting as recessive and partial dominant alleles, loss-of-function alleles can also act as dominant alleles. These dominant negative alleles often come up when the protein needs to assemble into a complex in order to function (for example, when four of the same proteins form a tetramer). If all four subunits of the complex need to be active in order to work properly, the defective subunits will "poison" most of the complexes in the cell, preventing the normal proteins from working properly.

Gain-of-function alleles (such as those that result in hyperactive proteins) generally act as dominant alleles.

In fact, in a classic genetics paper, researchers examined four different alleles of the MC1R gene in mice. One was a recessive allele for yellow hair, while the other three were dominant alleles for dark hair. When they examined the mechanistic basis for the traits, they found that the yellow hair allele encoded a defective MC1R, while the three dark hair alleles encoded hyperactive MC1Rs, reflecting the generalization that loss-of-function alleles are generally recessive while gain-of-function alleles are generally dominant. The citation for the paper is below for any who are interested in the topic:

Robbins et al. 1993. Pigmentation phenotypes of variant extension locus alleles result from point mutations that alter MSH receptor function. Cell, 72: 827. http://dx.doi.org/10.1016/0092-8674(93)90572-8 .

 

[MrRagnarok]

Thanks for the great information. Sorry to hijack this thread, but in light of these last posts I still have some questions I feel compelled to ask!

MC1R is a G-coupled protein receptor that binds pituitary hormones called melanocortins, including ACTH. MC1R Wiki. ACTH is upstream of the stress molecule cortisol. I may be looking too much into this, but does this imply that abnormal levels of stress may influence hair color? I'm sure MC1R binds more than just ACTH but the idea becomes more interesting (to me at least) in light of anecdotal evidence such as gray hair during stressful times.

On that note one final question, what happens to cause the melanocytes to produce gray (or white) pigmentation in old age?

 

[Ygggdrasil]

Those are all good questions that, unfortunately, I don't know the answer to. A quick look at Wikipedia suggests that the death of melanocytes in hair follicles (perhaps via apoptosis since Bcl2 and Bcl-w are involved) leads to no pigments being supplied to the hair, resulting in the grey coloration (https://en.wikipedia.org/wiki/Grey_hair#Conditions_affecting_hair_color). It doesn't seem clear whether we know how (or even if) stress affects the greying of hair (http://www.scientificamerican.com/article.cfm?id=fact-or-fiction-stress-causes-gray-hair).