Mendelian Genetics

Mendelian genetics is about heredity: the ways that genetic information is passed from parent to offspring. That’s slightly different from molecular genetics, which is about how nucleic acids and proteins function in cells, and it’s different from population genetics, which is about how genes spread or disappear in populations.

It’s an important set of ideas, and as a bio student you need to know how it works. However, I don’t particularly like the way this field is usually covered in the  textbooks, for reasons I’ll explain below.This page is intended to accompany the Mendelian genetics chapter in Campbell, not replace it.

Why do we study Mendel?

We’ve covered a lot of topics in Bio 6B, and we haven’t been looking at the original experiments and the people who carried them out. The reason is that there are just too many experiments. Our understanding of a topic like oxidative phosphorylation is based on hundreds of experiments, carried out over decades by a lot of different people. And fields of study aren’t usually named after the person who did the first experiments. Molecular genetics isn’t named after Jacob and Monod, who were the first to figure out how a gene works. 

So why do we call this field Mendelian genetics, and why is it almost always taught by explaining Mendel’s experiments with peas?

Mendel’s classic pea experiments both instruct and mislead

Gregor Mendel’s experiments, and his conclusions from those experiments, were brilliant. He was operating in complete darkness, in the sense that he had no idea of the physical reality of genes. DNA and protein structures hadn’t yet been figured out, so his idea of genes was simply as hereditary information that he surmised had to exist, even if its form was unknown. Amazingly, he got some important things right. So good job, Gregor!

On the other hand, by now you know far more molecular biology than he did. In fact, you know far more molecular biology than anyone in the world did for a hundred years after Mendel’s experiments. There’s no reason for us to pretend that we don’t know the molecular basis of genetics while  we talk about heredity. And yet that’s the way it’s almost always covered in the books. In Campbell, Mendelian Genetics is Chapter 14, at the beginning of the unit on genetics, and it doesn’t have a lot to say about DNA. 

The problem I have with this is that many bio students learn Mendelian genetics and molecular genetics, but don’t make a strong enough connection between the two. In fact, the simple pea-plant view of heredity sometimes seems to contradict molecular genetics. The view of genetics that comes from Mendel’s experiments isn’t always right. So let’s look at the famous experiments and consider how they relate to molecular biology.

Mendel’s pea flower color experiment

Mendel cross-fertilized different varieties of pea (and other) plants to figure out how traits were passed from parents to offspring, and he got a set of now-famous results.

You could do these simple experiments yourself, but you should keep in mind that Mendel chose very specific cases for his experiments.

First, he looked at binary traits. A trait is any observable characteristic of an organism. In this example, flower color is binary, because all the pea plans he looked at had either white or purple flowers, with nothing in between. Very few biological traits are so simple.

Second, he used true-breeding strains. In modern terminology, this would mean that his original parent plants were homozygous at the flower color locus, or whatever other locus he was investigating. One parent came from a variety (or strain) of pea plants that always produced purple flowers, generation after generation. The other came from a strain that always produced white flowers.

Mendel used common vegetable varieties for these experiments. People had been breeding peas and other vegetables for a long time (in fact, Mendel wasn’t the first person to do detailed experiments on heredity in pea plants. Similar experiments were done by Knight in 1799). Mendel could choose strains that had the characteristics he wanted, and know that he wouldn’t get any surprise characteristics in the next generation.

Also, it was easy for Mendel to control the cross-fertilization process. Like many plants, the peas produce flowers with both male and female reproductive parts. Peas often fertilize themselves (unlike most plants), but Mendel controlled this by removing the stamens and then transferring pollen from another plant himself.

Mendel’s results: Law of segregation

Here is one of the classic experiments:

Mendel started with two different parents: one with purple flowers and one with white flowers. These are called the P generation. This cross produces the F1 generation individuals, which all have purple flowers. When these F1 individuals are crossed with each other, or allowed to self-fertilize, most of the next generation (the F2 generation) have purple flowers, but some have white flowers. 

Here’s a summary, showing both the phenotypes (appearance) and the genotypes of the three generations:

  • P generation: All homozygous, for either purple flowers or white.
  • F1 generation: All heterozygous; one purple flower allele (P) and one white flower allele (p). All have purple flowers.
  • F2 generation: Mixed genotypes and phenotypes. 

Mendel’s insight was that a gene (allele) for white flowers must have somehow been hidden in the purple F1 generation individuals, but the white flower trait only showed up when some F2 individuals inherited two copies of the white flower allele.

This is Mendel’s law of segregation: the P and p alleles are segregated, or separated, and heterozygous individuals will pass on either P or p, but not some mix of the two. We now know that segregation happens in meiosis: when chromosome number is cut in half, at each locus the resulting haploid cells carry nucleotide sequences inherited from one parent, but not both.

 

Phenotype and genotype ratios

Mendel’s most important contribution may be that he brought quantitative thinking to the study of heredity. Look closely at the Punnett square showing the distribution of possible phenotypes and genotypes. You should be able to solve problems like this, including those with different parental genotypes than those shown above under the law of segregation.

Locus and allele

In more modern terms, we can describe these results in terms of locus and allele.

Locus” means place (plural: loci), and a locus is literally a place on a chromosome. Originally, the idea was that the locus is the place where there is a gene that codes for a protein that generates a particular phenotypic trait. The word locus is now used to refer to any location in a chromosome (whether it’s a gene or not), and we know that traits are likely to be controlled by more than one locus.

Allele” means "different." Mendel didn’t know anything about DNA; his alleles were simply different versions of hereditary information. Now we know that alleles are specific nucleotide sequences at a given locus. If we compare the genomes of two different pea plants, they would both normally have all the same loci, but at any given locus, they might contain different alleles.

So what is a gene? The word gene can be used to refer to a locus or an allele; if you want to be precise, you’ll have to use more specific terms.

Dominant and recessive

According to these results, the purple flower allele can be called dominant: as long as there is one copy of the P allele, the flowers are purple. The white flower allele p is recessive, because white flowers only appear on individuals that are homozygous for this allele. This dominant/recessive relationship seems simple in this case, but in many cases it isn’t.

Mendel’s gene model

Based on both his experimental results and some assumptions, Mendel created a conceptual model to explain how heredity works. It goes like this (rephrased to fit some of the biological concepts you’ve already learned): 

  1. Differences in phenotypic characters are caused by different alleles. 
  2. Diploid individuals inherit one copy of each locus from each parent.
  3. If there are two different alleles at a locus, one is dominant and one is recessive. 
  4. Alleles at a locus get separated from each other during meiosis, so only one allele is passed on at each locus (the law of segregation).

Mendel’s model was, for its time, a brilliant insight that helped to unveil the mechanisms of heredity. However, from a modern perspective, it’s far too limiting, and it would be very misleading to apply this model to all genetic situations, for several reasons:

  • Traits are influenced by more than one locus. An extreme example of this would be the genetics of height in humans. Researchers have searched the human genome in detail to find loci and alleles that influence height. They found that height is influenced by thousands of loci, and the 200 most important loci only explain about 20% of the variability in height. None of these loci is "the height gene;" all of them affect other aspects of biology, such as signaling pathways, energy use, and cell architecture.
  • Most loci affect more than one trait, as was exemplified by the study on human height.
  • Most traits are continuously variable, like height, rather than falling into discrete categories like white or purple flowers.
  • Traits are also affected by environmental conditions and epigenetics, which often interact. For example, childhood nutrition and health can influence height.
  • Alleles and traits aren't necessarily dominant or recessive.

Mendel had to start with the simplest cases if he was going to be able to explain anything, and his model provides a good explanation of his pea plant experiments. However, now we know that most genetic situations don’t fit into this simple model. Mendel pictured genes as units of hereditary material: one gene, one trait. Now we know that genes are more like networks than units, and traits aren't clearly tied to single genes. We begin to see the limitations of this model if we look more closely at some specific genes.

The molecular genetics of pea flower color

The diagrams in Campbell show that the P/p genotype determined the color of the pea flowers in Mendel's experiment. It's tempting to say that P/p is the flower color locus, but that isn’t quite right. The purple color of pea flowers comes from a pigment called anthocyanin, which is the product of a multi-step biochemical pathway (called the flavonoid biosynthesis pathway). There are multiple genes involved in this pathway, and mutations in any of them could alter flower color.

It wasn’t until 150 years after Mendel’s publications that molecular biologists identified the specific alleles behind the purple and white pea flowers. The purple/white flower locus from Mendel’s pea plants apparently encodes a transcription factor protein that belongs to a protein family called the basic helix-loop-helix or bHLH transcription factors. Another member of this large protein family is MyoD, which controls muscle cell differentiation. (It‘s featured as a gene expression example in Chapter 18 of Campbell.) Myc, which is involved in cell cycle control, is also a bHLH protein. All eukaryotes have multiple bHLH genes.

The white-flower mutant that Mendel worked with is a loss-of-function mutation in a bHLH transcription factor. Here's how the mutation works:

Conjugation detail 4

The gene has multiple exons, and the RNA must be spliced correctly to produce the correct protein product. The mutant version contains a single-nucleotide substitution at a splice site. With this mutation, the RNA cannot get spliced correctly. One of the introns contains extra nucleotides, which disrupt the reading frame. Eventually a stop codon turns up in the wrong place. When this mutant mRNA is translated, a shorter-than-normal, nonfunctional protein is produced.

We can describe this mutation several ways:

  • It’s a substitution, because one nucleotide is changed to another.
  • It causes mis-splicing, because it changes a splice site and causes different nucleotides to be included in the finished mRNA.
  • It’s a nonsense mutation, because it causes a premature stop codon.
  • It’s a loss-of-function mutation, because the mutant protein product doesn’t do its normal job.

This particular bHLH transcription factor is necessary for turning on the genes for the enzymes that synthesize purple pigment. The protein encoded by the P/p locus doesn't directly make the flowers purple, but it turns on the genes for other proteins that make the purple pigment.. One gene doesn’t control one trait, even in this simple example.

Dominant and recessive alleles, revisited

In Mendel’s experiments, there were always two alleles for each locus; that’s because he crossed two strains that were mostly genetically identical, but differed in one trait. The examples he used, such as flower color, pea color, and pea shape, all showed simple dominance: the heterozygous phenotype is the same as the homozygous dominant phenotype; only homozygous recessive individuals have the recessive phenotype, such as white flowers. All these examples have two things in common:

Loss-of-function mutations: 
The recessive allele has a loss-of-function mutation, so it doesn’t encode a functional protein; the recessive phenotype arises from the lack of a particular functioning protein. (There are many other examples in which mutant alleles are neither loss-of-function nor gain-of function, but simply encode slightly different proteins.)
Haplosufficiency: 
Only one copy of the functional allele is necessary to generate the dominant phenotype. This is true because the pathway is apparently regulated by feedback mechanisms that allow the anthocyanin-producing flavonoid pathway to keep functioning until a specific amount of pigment is produced, resulting in uniformly purple flowers, regardless of whether there are one or two functioning copies of the bHLH transcription factor allele. 
There are many examples that don’t fit this pattern. For example, the snapdragon flower color example described in Campbell, in which heterozygotes have an intermediate phenotype, is an example of haploinsufficiency; one copy of the red flower allele isn’t sufficient to generate completely red flowers. This is called incomplete dominance in the textbook.

Even in cases of simple dominance, the dominant allele doesn’t prevent the recessive allele from being expressed; this is a common misconception. The dominant allele is functional, the recessive allele is non-functional, but the two don’t directly influence each other. In these examples, the loss-of-function allele is recessive, but this isn’t always the case.

Dominant-negative alleles

Consider the tumor suppressor p53. As you know this gene, along with its protein product, is essential for cell cycle control and protecting genome integrity. The p53 pathway stops the cell cycle and may initiate either repair or apoptosis when DNA is damaged. A cell that lacks a functioning p53 pathway is very likely to become a cancer cell. However, loss-of-function mutations in this locus aren’t always recessive. 

The p53 protein is a transcription factor. When activated, p53 proteins form tetramers: four copies of the same protein, bound together to form a functional transcription factor that can bind to particular regions of DNA and up-regulate other genes. Some mutant versions of p53 are able to bind to normal p53 to form tetramers that can’t bind to DNA. In this case, the mutant version of p53 not only fails to carry out its function, it also prevents normal p53 proteins from functioning. It’s called a dominant-negative mutation, because a single copy of the mutant allele is enough to alter the normal phenotype and increase the risk of cancer. Some loss-of-function mutant versions of p53 are recessive, while others are dominant-negative.

Summary: what Mendel got right, and what he got wrong

Many biology students seem to learn Mendelian genetics and quickly forget it. Frankly, that’s not a bad idea, because assuming that Mendel’s ideas apply to everything in genetics can lead to more confusion than understanding. However, the fundamental ideas of Mendelian genetics are still useful if we look at them critically. Here’s my assessment:

  • Wrong: One gene controls one trait. This is wrong because very few traits are binary (like white or purple flowers), and because every characteristic is affected by more than one gene.
  • Right: In a diploid organism, there are two copies of each locus, and only one copy is passed to offspring in eggs or sperm (segregation of alleles). 
  • Misleading: there aren’t always two alleles for each locus. In a diploid individual, there could be one or two different alleles; in a population, there could be thousands.
  • Often wrong: One allele is dominant, the other recessive. This dichotomy applies only in specific cases; it doesn’t fit most loci.
  • Sometimes right: Separate loci are often passed on independently from one another (independent assortment). This is wrong in cases of gene linkage, and those cases are useful and biologically important.

Review

Terms & concepts

  • Allele
  • Dominant and recessive
  • Dominant-negative alleles
  • Gene linkage
  • Genotype and phenotype ratios
  • Haplosufficiency (and its opposite, haploinsufficiency)
  • Haplotype
  • Independent assortment
  • Locus
  • Loss-of-function mutations
  • Mendelian genetics
  • Segregation
  • Trait

About the codes for alleles 

The pea flower color alleles are abbreviated in Campbell as P and p. In Mendel’s view, there is always a dominant allele and a recessive one, so this code makes sense. In the larger world of biology, it’s not a good system, because in a population there could be more than two alleles at a locus, and they’re not always dominant or recessive. This type of abbreviation is normally used only in introductory textbooks.

In Morgan’s experiments with fruit flies, featured in Campbell, a different code was used. They started with wild fruit flies, so they called the typical alleles from the wild population “wild type,” which they indicated with “+.” They looked for mutants that affected traits that could easily be observed, and named the loci after the mutant version of the trait. For example, b for black body. The b+ allele is the wild type, which doesn’t produce a black body; the recessive loss-of-function allele is b, and it results in a black body only in homozygous individuals.

For modern biology, neither of these abbreviations is sufficient. The range of possible mutations is practically unlimited. Even in a single locus, such as the coronavirus spike protein, dozens of different mutant alleles may arise. Often, mutations are described in terms of the effect they have on the amino acid sequence of a protein. For example, one common mutation of the coronavirus spike protein is called D614G; it changes amino acid #614 of the spike protein from D (aspartic acid) to G (glycine). A similar code can be used for nucleotide sequences.

Review questions

  1. Look at the four statements of Mendel’s gene model. In what situations is each statement true, and in what situations might it be false?
  2. Explain the law of segregation in terms of the events of meiosis.
  3. Why are loss-of-function mutations often recessive? Why are they sometimes not recessive?
  4. How does a mutation at one locus cause flower color to change from purple to white for pea plants? Are there other mutations that could produce the same phenotypic effect?

References & further reading

General genetics reference 

Talking Glossary of Genetic Terms. National Human Genome Research Institute. 

Molecular genetics of Mendel’s experiments

Identification of Mendel's White Flower Character. Hellens, 2010. PLOS. "The A gene encodes a bHLH transcription factor. The white flowered mutant allele most likely used by Mendel is a simple G to A transition in a splice donor site that leads to a mis-spliced mRNA with a premature stop codon." Surprisingly, the molecular basis of Mendel’s flower color alleles wasn’t figured out until 2010.

p53 mutants

Mutant p53 exerts a dominant negative effect by preventing wild-type p53 from binding to the promoter of its target genes. Willis et al, 2004. Nature.

Re-engineered p53 activates apoptosis in vivo and causes primary tumor regression in a dominant negative breast cancer xenograft model. Okal et al., 2014. Gene Therapy. Researchers engineered a new version of p53 that continues to function even in the presence of dominant-negative mutant proteins. Graphic abstract: 

figure1

Ancient history

An Account of Some Experiments on the Fecundation of Vegetables. Knight, 1799. That’s right, the year 1799! Read this if you want to know what science sounded like before Mendel was born. Knight actually did some experiments that were very similar to Mendel’s, but he lacked Mendel’s quantitative framework. Mendel probably read this paper and built on it.

Experiments in Plant Hybridization. Mendel, 1865. A 1901 translation of  Mendel’s original paper, which was in German. Unfortunately, this translation wasn’t available in time for Darwin to read it; no doubt, he would have made good use of the concepts of genetics.

A- A A+