Indiana University

subglobal1 link | subglobal1 link | subglobal1 link | subglobal1 link | subglobal1 link | subglobal1 link | subglobal1 link
subglobal2 link | subglobal2 link | subglobal2 link | subglobal2 link | subglobal2 link | subglobal2 link | subglobal2 link
subglobal3 link | subglobal3 link | subglobal3 link | subglobal3 link | subglobal3 link | subglobal3 link | subglobal3 link
subglobal4 link | subglobal4 link | subglobal4 link | subglobal4 link | subglobal4 link | subglobal4 link | subglobal4 link
subglobal5 link | subglobal5 link | subglobal5 link | subglobal5 link | subglobal5 link | subglobal5 link | subglobal5 link
subglobal6 link | subglobal6 link | subglobal6 link | subglobal6 link | subglobal6 link | subglobal6 link | subglobal6 link
subglobal7 link | subglobal7 link | subglobal7 link | subglobal7 link | subglobal7 link | subglobal7 link | subglobal7 link
subglobal8 link | subglobal8 link | subglobal8 link | subglobal8 link | subglobal8 link | subglobal8 link | subglobal8 link


Contents of this Section

Drosophila as a Model System

Why Model Systems are Necessary

When we first examined My Family, we found that it is difficult to unravel the rules of genetic inheritance when studying a species that has only a few offspring in each generation. If a human family has only one or two children, we just can't tell from this family how a trait is inherited. We may be able to examine the entire population to see that there are roughly as many males as females, so that we expect that any individual child has a 50:50 chance of being a boy or a girl, but we cannot test this expectation by observing a family that has two children, both of whom are boys. We need many more offspring to evaluate this type of expectation.

We also learned that it is difficult to follow genetic inheritance when individuals from other families marry into our study group. With human families, this type of "out-crossing" is the norm. We can get around this problem to some extent by studying other species for which many generations of intentional breeding have produced more homogeneous family lineages, such as specific breeds of dogs or sheep.

Unfortunately, large animals, and major agricultural crop plants such as corn and wheat, tend to breed seasonally. They may have no more than one generation per year. While we may be able to choose the parents for each cross, and establish homogeneous lineages for our studies, the studies are very slow.

What if we could find a small, rapidly-reproducing organism that would enable us to follow many generations in a year, instead of just one?

Back to top

Drosophila melanogaster

In 1906, Thomas Hunt Morgan began work with a small fly, Drosophila melanogaster, that meets this criterion. This fruit fly, as it is called, eats yeast that colonize fallen fruits. It was relatively easy to establish a laboratory population of fruit flies, growing them in half-pint milk bottles on a food of mashed bananas and yeast. In 1910, Morgan discovered a white-eyed variant in one of his bottles of flies. It was the result of a spontaneous mutation in a gene that Morgan named white. In the following century, hundreds of laboratories world-wide have made Drosophila one of the most intensively-studied species known.

When grown at room temperature (25C), it takes two weeks for D. melanogaster to develop from egg to adult. This makes it possible to analyze some 25 generations each year. A single female can lay as many as 2,000 eggs, and a single male can fertilize 20,000 eggs. Not only is it possible to set up any cross we want (even among siblings), we can examine very large numbers of offspring from each cross.

Therefore, where a human pedigree might look like (A) below, a similar pedigree of a laboratory stock of Drosophila might look like (B) below.

A. A human pedigree chart B. A pedigree chart for a laboratory stock of Drosophila

In general, laboratory stocks of Drosophila have been inbred for so many generations that there is no longer any significant genetic variation within any particular stock. Sibling matings, as illustrated in (B) above, completely eliminate the uncertainty in the genetic heritage of any particular individual fly. A stock of flies with red eyes "breeds true," with every individual having red eyes; a stock with brown eyes breeds true, and so does a stock with white eyes.

On the other hand, there are thousands of different laboratory stocks, each with its own particular alleles of particular genes. Most of these different stocks can be obtained from the Bloomington Drosophila Stock Center at Indiana University, or from other international stock centers in Umea, Kyoto, and Szeged. A smaller number of more-commonly used fly stocks can be obtained from science-teaching supply sources, such as Carolina Biologicals.

In the previous pages, examining hair color in humans, dogs, and sheep, we we obtained enough information to build an explanatory model of genetic inheritance. Our study of Labrador Retrievers suggested that phenotypes such as hair color are based on the production of pigments; alleles that enable pigment production tend to be dominant to alleles that result in production of less pigment. Our study of sheep helped us see how to follow individual alleles in a single cross, and led us to Reginald Punnett's "square" diagram that is so useful for predicting the possible offspring from a particular cross. The "rules" we developed are helpful in understanding a number of different genotype/phenotype relationships, including those examined in peas by Gregor Mendel. Do the same rules hold true for Drosophila, and can we learn new and interesting things using this remarkable organism?

Back to top

Red Eyes and Brown Eyes

The normal eye color for Drosophila melanogaster is red, as shown in the photograph above (from the Wikipedia Commons collection). There are mutant varieties with brown eyes. What is the basis for the brown color, and how is this phenotype inherited?

In the absence of any information about red or brown eyes, we probably cannot even begin to phrase a hypothesis that states our current understanding. Nor can we make reasonable predictions based on such an hypothesis. We can, however, ask a question experimentally. We can ask whether the brown-eye phenotype is dominant or recessive (or somewhere in-between) to the normal red eye color. We can "ask this question experimentally" by crossing red-eyed flies to brown-eyed flies, and examining the offspring.

Starting with true-breeding stocks of flies (to ensure there are no unexpected genetic variations to confuse us), we see the following:

Illustrated as a Pedigree
Illustrated as Individual Flies

It doesn't matter whether the male has brown eyes (as shown here), or whether the female has brown eyes. The offspring all have red eyes. This tells us that this allele of the gene responsible for brown eyes is recessive to the normal allele found in wild flies.

This allele causes eyes to be brown; therefore, the gene has been named brown. There are many different alleles of this gene, including the original allele from flies captured in the wild. Here, we are working with the mutant allele of brown that was found first; using the convention of Drosophila genetics, the allele name (or number) is indicated as a superscript. Thus, brown-1 would be indicated as brown1. The normal allele from wild flies, referred to as the "wild type" allele, is indicated by a + symbol, as brown+. Using these conventions, we can write the cross illustrated above as:

or, abbreviating brown as bw,

Assuming that flies--like humans, sheep, dogs, and peas--inherit one set of genes from the mother and another set from the father, the parents should be homozygous for their particular brown alleles (likely, since the strains of flies "breed true"). The offspring should all be heterozygous for bw1 and bw+. If this is really true, we can make several clear predictions:

  1. The two alleles in the heterozygous flies should segregate during gamete formation, producing two classes of gametes. One class should carry bw1, and the other class should carry bw+. Therefore:
  2. If we back-cross the heterozygous offspring to the homozygous recessive (brown1 / brown1) parental strain of flies, we should find that half of the offspring have red eyes, and half have brown eyes. That is, half of the offspring should be bw1/bw1 homozgotes, and half should be bw1/bw+ heterozygotes. [Perhaps, if we use a Punnett Square to write down the types of gametes from each parent, it will be easier to see why this prediction makes sense.]
  3. If we cross the bw1/bw+ heterozygotes to each other, we should expect to obtain three classes of offspring--heterozygotes, and both types of homozygotes. For this, it will probably help significantly to write out a Punnett Square to keep track of the different possibilities.

Back to top

The Test-cross (or back-cross):

To test prediction #2, we must perform the following cross:

If we draw this out in a Punnett Square, we obtain this:

Only in the top row of the table do we find the bw+ allele, so only the flies represented in this row will have red eyes. The flies represented in the bottom row will have brown eyes. Two squares in the top row, two in the bottom row, and we get 50:50.

What do we find when we do the cross? Something like this:

This result is consistent with our expectation. It does not prove that our expectation is correct (there could be other mechanisms that could produce this result), but it lends confidence to our model of inheritance of brown1 as a single allele that is recessive to brown+.

Back to top

The Cross Among Siblings (inter se cross):

If we cross the F1 offspring among one another, we expect to be crossing bw1/bw+ heterozygotes together. Drawing this out in a Punnett Square is the easiest way to see what's happening in this cross:

We expect 1 out of 4 offspring to be homozygous for brown1 (lower right corner of the table). Similarly, we expect 1 out of 4 to be homozygous for brown+ (upper left corner). We also expect 2/4 (i.e. 1/2) of the offspring to be heterozygous. If we are unable to distinguish by eye whether the brown+ homozgotes look significantly different from the heterozygotes, then we expect to find 3:4 of the offspring to have red eyes, and 1/4 to have brown eyes. What do we find when we do the cross?

We obtain the results we predict, within statistical variation:

The data we obtain from these crosses appear to support our hypothesis--our understanding of how the brown gene is inherited. This allele appears to be inherited as a simple recessive allele.

Despite the genetic analysis, we don't yet understand the link between genotype and phenotype here. There are several models that we can imagine--from brown1flies failing to produce a red pigment, to brown1 flies over-producing a brown pigment (due to some kind of biochemical problem). We need more information before we can understand how the brown1 mutation causes fly eyes to be brown.

Back to top

Red Eyes and White Eyes

What about the inheritance of white eyes, the first mutant Drosophila found by TH Morgan in 1909? TH Morgan named the gene white. Wild type flies (with red eyes) carry the white+ allele, and white-eyed flies carry the white1 allele. Again, we have no understanding of how this mutation might be inherited, so there is little point in guessing. Let's start by asking the simple question: is the white1 allele inherited as a simple recessive?

Again, we will ask this question experimentally, by crossing wild type, red-eyed flies with mutant, white-eyed flies. Again, just to be sure we don't overlook something interesting, we will perform the cross both ways--with white-eyed males, or with white-eyed females. The results are as follows:

(A) The cross using a white-eyed male
(B) The cross using a white-eyed female

The results of cross (A) are what we would expect if the white1 allele is inherited as a simple recessive. But the "reverse" cross, (B), is very different. Somehow, the sex of the fly matters.

To figure this out, it might be necessary to look into the mechanisms that determine sex.

Back to top

Sex Determination

In crocodiles and some species of turtles, sex is determined by the temperature at which the eggs develop. In some species, a warm nest produces all females, and a cold nest produces all males. In other species, it's the other way around. But in fruit flies and mammals, the temperature has no effect. Sex determination occurs genetically, in response to certain genes that control embryological development.

In species with genetic sex determination, there is often a visible difference in the chromosomes of males and females. This is true in humans and in Drosophila. For both of us, most of the chromosomes are identical in both males and females (though they may carry slightly different alleles). This is easiest to see in Drosophila, which have only 4 chromosome pairs (compared to humans, with 23 pairs). The image shown here (redrawn from J. Albert Vallunen's photographs in the Wikimedia Commons) shows chromosomes condensed at metaphase of mitosis. In Drosophila, mitotic chromosomes often associate in pairs, with the tiny chromosome #4 in the center of the cluster. It is evident from looking at the chromosomes that females have two X chromosomes, while males have only one. Instead of a pair of X's, males have an X chromosome paired with a Y chromosome.

The situation is similar in humans: XX individuals develop as females, and XY individuals develop as males.

Back to top

Sex-Linked Inheritance

The presence of X and Y chromosomes leads us to ask an interesting question: What happens with genes that are located on the X chromosome or on the Y chromosome? "Normal" inheritance (such as brown eye color, considered above) occurs for genes on the other chromosomes (called autosomes). But what if genes are on the sex chromosomes?

In humans, there is a gene on the Y chromosome that regulates the "hair-growth program" specifically in ears. Most alleles of this gene result in little or no ear hair. Some alleles, however, activate the hair growth program, resulting in hairy ears. (The image on the right is hosted at Images/gen17.gif). Because the Y chromosome also carries the gene that determines maleness, the hairy-ear phenotype rarely, if ever, occurs in women.

The white-eye phenotype in Drosophila does occur in female flies. This observation rules out the possibility that the white gene might be on the Y chromosome. Perhaps we can explain the data by the hypothesis that the white gene is on the X chromosome, with white+ being a functional allele, and white1 being a non-functional (and thus recessive) allele. Let's draw out a few Punnett Squares and see if this makes sense.

In doing so, let's represent the Y chromosome with the symbol, Y. We hypothesize that the Y chromosome has no white gene on it. We also hypothesize that the X and Y chromosomes behave at meiosis like any other pair of chromosomes, and segregate from each other. That is, males should produce some sperm cells carrying the Y chromosome, and some carrying the X chromosome--and only the X carries an allele of white.

(A) The cross using a white-eyed male
(B) The cross using a white-eyed female
all offspring have red eyes because all carry w+
only the female offspring have red eyes, because only they carry w+

These Punnett Squares predict that cross A should produce entirely red-eyed offspring, and that cross B should produce red-eyed females and white-eyed males. This is exactly the result we obtained. Thus, it looks like we can explain the observations with the model that the white gene is on the X chromosome, and that the normal,white+allele enables flies to produce the eye pigment, while white1 does not enable pigment production (and is thus recessive to white+). Let's test this model with one more cross: let's allow mating between the male and female offspring from cross A. That is, let's perform inter se crosses among the F1 flies from cross A, and examine the F2 generation. First, what do we predict?

Using the handy tool of the Punnett Square, we predict that all of the females from this cross will be heterozygous for white+ and white1. They should, therefore, have red eyes. The males will inherit the Y chromosome from their fathers, and therefore should show the phenotype that is determined by the only white allele they have -- which they inherit from their mothers. Therefore, half of the males should have red eyes, and half should have white eyes. When we perform this cross, we observe the following:

The results of the cross are as we predicted, based on our tentative model of inheritance of the white gene.

Additional work, ranging from genetic mapping to DNA sequencing, has shown unambiguously that the white gene is, indeed, on the X chromosome. Therefore, this analysis of white provides a good model of sex-linked inheritance.

  • Genes that are on the X chromosome are said to be X-linked.
  • Genes on the X chromosome show inheritance patterns in which the traits of interest are co-inherited with the sex of the individual; such traits are therefore said to be sex-linked.
  • Very few genes are on the Y chromosome. Therefore, most sex-linked inheritance involves genes on the X.

Back to top

Dominance (Again)

There are hundreds of alleles of the white gene, several of which are described here (from "Genetic Variations of Drosophila melanogaster" by D. L. Lindsley and E. H. Grell, Carnegie Institution of Washington Publication No. 627, 1968). Most of them produce various shades of the reddish-brown eye color of normal flies, as illustrated in the image on the right. Some produce mottled or variegated patterns of color.

Because there are so many different alleles, we cannot refer to the wild-type allele with a capital letter (W) and the mutant alleles with a lower-case letter (w). For one thing, W is the symbol for the gene, Wrinkled. Instead, we use superscript abbreviations to keep track of the specific alleles, such as w+, we, wch, wa, and w1 for the series shown on the right. As with our own hair color (different shades of brown), these different alleles show dominance relationships that reflect how much pigment each allele enables the eyes to accumulate. We illustrate this below with whiteapricot.



The w+ allele is dominant to both w1 and wa. This makes sense, since w+ is the darkest of the white alleles.

Does this mean that wa is "a recessive allele"? It is recessive to w+, but it is dominant to w1. Again we see that "dominance" is a relationship between alleles within an individual. A particular allele may be dominant, or may be recessive, depending on the other allele with which it interacts.

Again, we can understand this by thinking about the mechanism by which the genotype gives rise to the phenotype. The wa allele enables eyes to accumulate only a small amount of pigment. This small amount of pigment is more than the amount accumulated by w1, so wa appears dominant to w1. This small amount of pigment is much less than the wild-type amount of pigment, so wa appears recessive to w+.

Again, we do not see 100%, strict dominance. wa/wa flies have slightly darker eyes than do wa/w1 flies. Again, this makes sense, based on the amount of pigment that each allele can contribute to the overall total that we see as the phenotype of the fly.

Back to top

Other Model Systems Used in Genetics

Unicellular Organisms:

E. coli

Back to top

Caenorhabditis elegans -- the transparent soil nematode

Back to top

Arabidopsis thaliana -- a tiny mustard plant

Back to top

Danio rerio -- the Zebrafish

Back to top

Mus musculus -- the mouse


Back to top

Working With More Than One Gene

Creating New Alleles by Induced Mutation