8/16/01

Lecture #19 - Developmental Genetics&endash; Embryonic development of the fruit fly Drosophila melanogaster

Drosophila

Drosophila melanogaster, the fruit fly, has been an important experimental animal for approximately 100 years
From 1910-1925 Thomas Hunt Morgan used Drosophila to determine the basic principles of genetics
In the 1980s Drosophila was used to study development
Drosophila continues to be an important experimental animal today

Problem of development and pattern formation

Drosophila starts life as a single fertilized egg. It develops into an adult with highly differentiated cells.
The egg has polarity, with dorsal (top), ventral (bottom), anterior (left) and posterior (right) regions. There is positional information encoded in the egg and zygote.
We are interested in the molecular basis for positional information (pattern formation).

Drosophila life cycle

The Drosophila egg is polarized into the regions mentioned above.

Embryonic development is very rapid. Within 24 hours of fertilization the embryo hatches from the egg as the first instar larvae. In another 24 hours it develops into the second instar larvae, and by 72 hours it has developed into the third instar larvae. The instar larvae are overtly segmented; that is, they have an obvious segmentation pattern.

The pupal stage of Drosophila lasts four days, after which the adult is fully developed.

The adult animal is also overtly segmented, with a distinct thoracic (3 segments) and abdominal (8 segments) as well as several head segments.

Early embryonic development

Early embryonic development takes place in the first 6-8 hours after fertilization, especially in the first 3 hours following fertilization.
*See Figure 29.2
The phase from fertilization to first instar larvae involves nuclear divisions during the first 1.5 hours of life. At about 2.5 hours post-fertilization, nuclei migrate to the cell surface, forming the syncytial blastoderm.
At about 3.25 hours the nuclei become cellularized, forming the cellular blastoderm.
At this point the cells look identical although they express different gene products and they have differnent developmental fates.
*See Figure 29.21

Nüsslein-Volhard and Weischaus

Carried out a saturation genetic screen to isolate all genes which when mutated altered the pattern of the Drosophila embryo
Several experiments were done on early Drosophila development.
Beginning in 1980 Nüsslein-Volhard of the Max Planck Institute and Weischaus (presently at Princeton) carried out a saturation genetic screen that earned them the 1995 Nobel Prize in Medicine.
They wanted to isolate mutants with defects in developmental control. Their final goal was to isolate all genes in Drosophila which when mutated changed the pattern of the embryo (what hatched out of the egg as the first instar larvae). They wanted to isolate and identify the master developmental control genes.
They began by mutagenizing flies and looking for pattern alterations in first instar larvae (*see handout figure comparing wild type larvae to bcd^- mutant larvae).
They made several assumptions in conducting the experiment:
- 1. They predicted the mutants they isolated would survive until the end of embryonic development but would not hatch as first instar larvae; that is, those mutations would lead to a lethal phenotype after embryonic development. In making this assumption their experiment disregarded flies with mutations in DNA replication, RNA polymerase II, and splicing. These flies would all die very early, before they hatched from the egg.
- 2. They also assumed developmental defects would be evident in the cuticle patterns of dead embryos. They were looking for evidence of mutations in changes of the pattern of denticle bands (hairs). They thought they would isolate mutants with missing bands, transformed bands, and/or inverted bands.
They carried out this experiment on a large scale, and isolated more than 500 mutants that had altered developmental patterns.
They classified them into 50 complementation groups, approximately 10 alleles of each mutant.
From 1980-1995 all fifty genes were cloned and their protein products were identified. The expression patterns of RNA and protein for all 50 genes were determined.

Classes of pattern formation mutations

The developmental genes function as a temporal cascade.
Coordinate and segmentation genes are expressed during the early stages of development and function transiently (from about 0-3 hours). The homeotic genes are turned on at about 3 hours and continue to function throughout the fly's lifetime.

Coordinate

The coordinate genes function to lay down the overall polarity of the embryo on the anterior/posterior axis and the dorsal/ventral axis.
Another class of coordinate genes specifies the termini of the embryo.
Coordinate genes are maternal effect genes; that is, the passage of mutations in these genes is dependent on the genotype of the mother. For instance

Zyg

ote

Mom
Dad
genotype
phenotype

bcd/bcd
+/+
bcd/+
mutant

+/+
bcd/bcd
bcd/+
wild type

bcd/+
bcd/bcd
bcd/bcd* or bcd/+
wild type

Note that if the bcd/bcd* individuals in the bottom row of the chart are female, their offspring will be mutant.
The expression of phenotype is therefore dependent on the genotype of the mother.
In coordinate mutants the polarity of the embryo is altered. For example, bicoid (bcd) mutants are missing the anterior segments (head and first two thoracic segments) because the anterior-posterior axis is not patterned correctly.

Segmentation

Segmentation genes are not maternal effect genes. They are zygotic genes &endash; if the zygote has a mutant genotype it will have a mutant phenotype.

There are three subclasses of segmentation genes: gap, pair-rule, and segment polarity. These three classes of genes subdivide the embryo into segments and compartments (units within the segments)

Gap segmentation genes affect large blocks of segments. Mutations in these genes cause blocks of adjacent segments to be deleted. Examples of these types of mutations include Krüppel (Kr), hunchback (hb), and giant (gt)

Pair rule segmentation genes set up the overall segmentation of the embryo. Pair rule mutants have 1/2 the normal segments (every other segment is missing). An example of a pair rule mutation is even skipped (eve)

Segment polarity genes specify the polarity of individual segments. Mutants have polarity defects within every segment (for instance, the anterior and posterior portion of each segment may not be patterned correctly). Examples of this type of mutation include engrailed and wingless.

Homeotic

Homeotic genes are involved in specifying segmental identity. These genes are zygotic genes.
Homeotic genes begin to function about 3 hours after fertilization and remain functional throughout the lifetime.
Homeotic mutants have segment identity transformations (eg, T3 ý A1). Examples of these mutants include antennapedia (some dominant alleles have phenotypes where the antennae are changed to legs) and ultrabithorax.
*See Figures 29.21 and 29.20

Temporal hierarchy of gene action

After mutants were isolated, the genes were cloned and people looked at when in development and where in the embryo RNA protein was expressed.
Generally the domain of expression of protein or mRNA corresponds to the region of the animal that is defective in the mutants.
This tells us that genes are expressed in spatially and temporally regulated patterns.
*See Figure 29.22

even skipped stripe 2 experiments

even skipped stripe 2 becomes established by action of the coordinate and gap genes
*See anterior-posterior polarity figure in handout for genes important in even skipped stripe 2 expression
The 7 stripe expression pattern happens within two hours after fertilization. It is the result of the activation of maternal effect and gap genes.
Experiments studying even skipped stripe 2 involved the construction of promoter-reporter gene fusions. These fusions were inserted into the fly genome so transgenic flies were produced in which every cell had the gene fusion in it.
*See the "Are effects direct?" figure in the handout
Referring to the "Are effects direct?" figure, each stripe has a separate element; that is, there is a region of the promoter for each stripe. Initially the experiments had a large region of the promoter fused to the reporter gene. The embryo showed stripes 2, 3 and 7. They determined that the promoter region was too big to study only stripe 2 in isolation, so they fused a smaller promoter piece to the reporter gene. The embryo showed only stripe 2. Using this piece of promoter that controlled stripe 2, a series of experiments were conducted. Crossing the transgenic fly (with the promoter-reporter fusion) with bcd^- or hb^- mutants yielded progeny with no stripe 2.
Crossing the transgenic fly with gt^- mutants yielded progeny with a stripe 2 that was expanded towards the anterior end of the fly.
Crossing the transgenic fly with Kr^- mutants yielded progeny with a stripe 2 that was expanded towards the posterior end of the fly.
From these results we can conclude that bcd and hb proteins are needed for the production of stripe 2 &endash; they are positive regulators of eve stripe 2. Also, gt defines the anterior boundary of eve stripe 2 expression and Kr defines the posterior boundary of eve stripe 2 expression. Kr and gt are negative regulators that function spatially.

Footprints

Scientists then wanted to determine if the positive and negative regulatory effects are direct. They wanted to know if the proteins bound directly to the promoter region that controls the stripe 2 expression. They did a series of footprints to find this out. Kr, gt, hb, and eve proteins all bind to the eve stripe 2 element in a footprint experiment.

Co-transfections

To determine if the four proteins discussed above functioned as activators or repressors of transcription, co-transfection experiments were conducted. Tissue cultures from Drosophila were used. Reporter plasmids were added to the cells, and expression of the plasmid was measured to determine the activity of the protein. The reporter plasmid contained the eve stripe 2 element and the CAT (chloramphenicol acetyl transferase) reporter gene.
When just the reporter gene on the plasmid was added to the cells, a 1X level of expression of CAT was measured.
When the plasmid with bcd and the reporter gene was added to cells, the CAT expression was 18X.
When the plasmid with hb and the reporter gene was added to cells, the CAT expression was 4X.
When the plasmid with hb, bcd, and reporter genes was added to cells, expression was 44X. This phenomenon is called synergistic effect &endash; the effect is greater than the sum of its parts.
From these results it was concluded that bcd and hb proteins functioned as transcriptional activators.
Negative regulators were then looked at. The same experimental setup with bcd + hb + Kr and the reporter gene on the plasmid gave cells with 2X expression.
The same experimental setup with bcd + hb + gt on the plasmid gave cells with 1X expression.
It was therefore concluded that gt and Kr proteins can function as repressors of transcription.

Site-specific mutagenesis of binding sites

Next scientists wanted to determine if the effects observed above were from site-specific binding of the DNA binding proteins on eve stripe 2 DNA.
To determine this they conducted experiments in which binding sites were mutated in the context of an eve stripe 2/reporter gene fusion. The general idea in this experimental setup is if the regulators function by binding to specific sites then the effect should be the same as crossing a wild-type promoter-reporter gene fusion with the various kinds of mutants.
First the bcd binding site was mutated (2 of 5 sites were affected). The result was a decrease in the levels of eve stripe 2 expression. When all 5 sites were mutated, there was no stripe in the flies.
Mutation of hb binding sites yielded decrease eve stripe 2 expression.
Mutation of 3 gt binding sites yielded anterior expansion of eve stripe 2 (note this is the same effect as the wild type X gt mutant cross)
Mutation of 2 Kr binding sites did not give posterior expansion (as was expected). This was a puzzling result. One possible reason for this result is that Kr might not function via the Kr binding sites in the stripe 2 element. The accepted explanation, though, is that Kr has more than one role in regulating transcription.
Kr protein is a negative regulator of eve stripe 2 in that it sets the proper posterior boundary. Kr protein is also a negative regulator of hb (which is a positive regulator of eve stripe 2). There is thus no posterior expansion when Kr binding sites are mutated, because Kr protein keeps the posterior edge of the stripe in check by regulating hb, and not by direct DNA binding.

*See Figure 29.25