In these questions I am providing a fairly realistic narrative of a type of investigation that has been
carried out for many years. The molecular biology approaches used have changed over time, so some
questions I pose refer to current approaches and others to older approaches.
The starting point is a genetic screen using a model genetic organism, Drosophila. The idea of a
classical genetic screen is to identify genes that normally contribute to a biological process of interest.
This is accomplished by generating mutations in genes at random in a large collection of individuals,
screening for any individuals that show an aberrant (“mutant”) phenotype of interest and then
identifying the gene affected by the functionally relevant mutation. The phenotype can be anything for
which you have a good screen (failure to produce the right number of segments, altered eye
morphology, male or female sterility…). Here, for the sake of interest, let us imagine a screen has
been conducted for flies that are not very good at learning and that the mutations identified are
recessive, i.e. flies with a relevant homozygous mutation learn poorly. Several such screens have, in
fact, been done, yielding a first crop of mutations named dunce, turnip, rutabaga etc.
This question is (mainly) about molecular biology rather than genetics, so we start from a point that
you have a homozygous mutation that produces learning-challenged flies and you want to identify the
gene that is affected.
Genetic screens can be conducted most efficiently with a chemical mutagen that generally induces
single bp changes (point mutations). An alternative mutagen is a transposable element- a piece of
DNA that can jump from one site in the genome to another. Here an advantage is that the transposon
acts by jumping into a gene and therefore molecularly tags the affected gene (making gene
identification easier). In our hypothetical experiment a 3kb transposon, known as a P-element, was
used. That means we had flies that contained a P-element in one location and a “transposase” that
catalyzed jumping to other (random) positions. Genetic derivatives of those flies were later (after
making chromosomes homozygous) tested for mutant phenotypes, so it is hoped (but not guaranteed)
that the mutations producing a phenotype are caused by a single P-element interrupting a single gene.
1. Southern blot analyses
(a) You want to know if there is just one P-element insertion site in your mutant strain or more than
one. You isolate genomic DNA from the mutant flies and digest with four different restriction
enzymes separately. The products are run on an agarose gel, blotted and hybridized to a probe that
copies the entire 3kb P-element DNA. You know the entire 3kb DNA sequence and hence the location
of all restriction sites. Imagine enzymes A, B and C cut at places shown below and that D has no sites
within the 3kb fragment.
If the P-element is at only one site in the genome what do you expect to see for the four lanes (cut with
A, B, C or D; four separate answers)? Explain carefully.
(b) Imagine that you had used a probe that contains biotinylated nucleotides and detected the signal
using horseradish peroxidase and chemiluminescence? What does that mean? Just draw or explain in
overview (actual chemical structures not needed) how a biotinylated probe is eventually visualized by
emission of light.
(c) Imagine that you could barely see any bands after completing the procedure and so you wanted to
enhance sensitivity. Which of the following might increase sensitivity - more cut genomic DNA per
lane, narrower lanes, more probe, smaller volume for hybridization, more time for hybridization?
(d) Imagine that for DNA cut with A you see bands of 1.2kb and 4.3kb; for B you see bands of 1.4kb,
3.5kb and 5.2kb. Draw what you can deduce from this about the locations of the nearest sites for A
and B in genomic DNA either side of the transposon insertion site.
(e) If you had also digested the DNA with both A and B together what results would you expect and
what further information would that give you? [Diagram probably essential].
2. Finding the exact transposon insertion site
If you had a cloned or PCR-amplified segment of DNA containing the transposon and neighboring
genomic DNA you would be able to use an oligonucleotide complementary to known P-element
sequences (and facing, 5’ to 3’, towards unknown genomic DNA) as a primer for dideoxy sequencing
to reveal exactly the junction between the transposon and genomic DNA (on either side, one at a time).
But how can you get that suitably pure and amplified template DNA?
(a) You consider cloning. Let us imagine that enzyme D produced a band of 8.4kb on a Southern blot.
Maybe you could use that knowledge.
(i) If you go back to your Southern blot (the membrane with the DNA on it) can you recover DNA of
the appropriate size from there? Explain.
(ii) You run another gel of DNA digested with enzyme D, cut a gel slice in the 8.4kb region, extract
the DNA (silica gel method), ligate to an ampicillin-resistant plasmid vector cut with D and treated
with alkaline phosphatase, transform competent E. coli and plate on ampicillin agar plates. You pick
one of the resulting colonies. Is it likely to contain the desired cloned DNA? Explain.
(iii) How could you expect to identify the desired colony from the experiment above or even from a
similar experiment where you ligate all fragments of genomic DNA cut with enzyme D to the vector
without any prior fractionation (i.e. what additional steps could you take after you have generated
colonies of transformed E. coli)?
(b) You consider PCR. Knowledge of transposon sequence allows you to design one primer for PCRamplification of neighboring genomic DNA but making an appropriate second primer to flank the
DNA you want to amplify appears to be a problem.
Might you be able to design a more elaborate protocol (not just straight PCR) where you use two
specific primers, both complementary to transposon sequences. You are given the clue that you should
consider using a restriction enzyme and DNA ligase in your procedure prior to the PCR reaction.
PLEASE START A NEW PAGE FOR THE REMAINING QUESTIONS
3. Making simple DNA constructs using cloning or PCR
If you succeeded in obtaining the sequence of the genomic DNA immediately adjacent to the
transposon you would know the exact insertion site because the (almost) complete Drosophila genome
sequence was deciphered more than 15 years ago. Over the last few decades there have been
increasingly comprehensive and sensitive approaches to also map the sequence of RNA transcripts
(mRNAs and other types of RNA) derived from the Drosophila genome. Imagine here that we don’t
have that RNA mapping information and hence that we know nothing about any RNAs that normally
derive from the gene in this question or how they may be affected by the transposon insertion.
(a) If you had cloned the 8.4kb “D-fragment” you would have 5.4kb of genomic DNA neighboring the
insertion site. You could make a probe from the entire clone or from a smaller segment cut from that
recombinant DNA and use it to probe a Northern (RNA) blot. There are several choices for the RNA
that could be run on the gel. Your objective is to figure out clearly, but also economically (i.e. more
work is justified only if it is likely to be necessary to give a clear answer) if an RNA normally derives
from the region of DNA where the transposon is inserted. Explain what would guide your choices for
using the following types of RNA sample (and/or means you should consider choosing one, the other
(i) total RNA and/or polyA+ RNA?
(ii) RNA from whole flies and/or from fly heads?
I am looking for fair discussion of issues rather than the final decisions.
(b) Imagine you see a single RNA band of about 2.5kb in one or more of the tests above using a probe
derived from the 2.0 kb EcoRI fragment shown below (left). You want to know whether that RNA
runs in the direction from EcoRI to XbaI (within the 1.8kb EcoRI-XbaI fragment) or vice versa. One
way to determine that is to repeat the Northern blotting experiment with a single-stranded RNA probe.
To generate such a probe you first need to connect relevant cloned genomic DNA to a short segment of
DNA (less than 30bp) that is a binding site for SP6, T3 or T7 RNA polymerase (different sequence for
each). This could be done by cloning a relevant fragment of DNA into the vector shown below on the
right, which has binding sites for T3 and T7 RNA polymerase either side of convenient potential
cloning sites. (In all diagrams assume that all occurrences of named sites are shown).
(i) Which genomic fragment will you try to clone into the vector? Explain your choice.
(ii) Will you purify the fragment you want to clone after cutting and prior to ligation? Explain.
(iii) After transformation will you assume than any colony you pick will contain the desired
recombinant clone? Explain (AND say what else you might do).
(iv) Imagine you have the right clone and you purify the recombinant plasmid DNA. You then cut with
KpnI to linearize the template (not essential but better), add T3 RNA polymerase and suitable other
reagents to synthesize a biotinylated RNA probe. You use the probe on a Northern blot and see a band
of about 2.5kb. In which direction do you deduce the 2.5kb RNA runs 5’ to 3’? Explain. 
(v) You recall that almost anything can be done with PCR. How could you use PCR rather than
cloning to make a suitable template for making an RNA probe starting from the cloned 8.4kb
(c) Even though you now know that some part of the 2kb EcoRI fragment region of DNA gives rise to
an RNA of known polarity you do not know the exact structure of the RNA. As a small step in that
direction you consider using RT-PCR on Drosophila RNA samples using primers complementary to
sequences within the 2kb EcoRI fragment.
(i) if you pick any pair of primers of opposing polarity within this region would you expect to see an
(ii) If you saw a product with a particular pair of primers how could you generate an RT-PCR product
that extends all the way to the 3’ end of the RNA (assuming it is a mRNA)?
4. Back to the project and genetics (or common sense).
Imagine you had obtained full molecular details about the P-element insertion site and were convinced
it was the only site of insertion and you also found an RNA transcribed from the region that is altered
in some way by the P-element insertion. Would you be sure that the learning defect of the flies was
due to disruption of the gene in question? Explain (and include any additional test that would make
you more certain).
These solutions may offer step-by-step problem-solving explanations or good writing examples that include modern styles of formatting and construction
of bibliographies out of text citations and references. Students may use these solutions for personal skill-building and practice.
Unethical use is strictly forbidden.
Answer for 1c)
If we are unable to get proper signal in our southern blot then we should find ways to enhance the sensitivity. To increase the sensitivity means to increase the signal or eventually the number of probes (labels) in specified areas. Hence, we can digest more DNA and then load it per lane. Probe is generally taken in excess and I think taking more probe would not affect the sensitivity, but if we think that we had taken a less concentration of probe, then we should increase it. Narrower lane would eventually means reducing the total volume of DNA that can be loaded. If we think that we are taking DNA that gets “diffused” then we can reduce the lane diameter, else, I do not think it is a crucial parameter. Smaller volume of hybridization would not affect the sensitivity unless there is an enhanced amount of probe used. Hybridization time can be increased that would provide more chance for the probe molecules to bind with their target. However, we should note that the number of molecules of probe bound would only be up to the total number of DNA molecules available on the membrane; hence once saturated, we cannot enhance the signal by increasing the time of incubation....