Lecture 4: Transcriptional Regulation 2
- Last lecture: taken a look at the interplay b/w DNA & its organization & its transcription to RNA & the
regulation of those processes. When we did that, started to talk about the components of genetic switches &
talked about the role of sequence specific transcription factors/gene regulatory proteins & their interaction
with short segments or motifs of DNA. Took a look how one identified the motifs that those gene
regulatory proteins are going to bind to – did that through 3 different kinds of experiments: loss of function
experiment demonstrating necessity & a gain of function experiment demonstrating sufficiency of a
particular segment of DNA to support regulation of transcription.
- The loss of function experiment: deleting portions of the gene regulatory sequences (promoter) – call this
upstream or 5’ regulatory region – did promoter deletion analysis, removing chunks of it & asking which
chunks are necessary for the regulation of the corresponding gene & in this case a reporter gene that is
reporting on the activity of these endogenous gene – has been artificially constructed & inserted into the
genome & asked when it turns on/off & identified motif/sequence segment that is necessary to drive
expression in a particular tissue, at a particular point in time.
- Imagine that that promoter is actually chalk-block with different motifs, one that is important for, for
example, driving expression in guard cells, another chunk of that promoter may be important for driving
expression, for example, xylem cells. Identify therefore this necessary motif & asked is it also sufficient to
drive expression from a minimal promoter (by analogy: thought of like being a car that is missing its keys –
got everything you need for car to drive somewhere but without that key, it’s not going anywhere) – those
sequence specific motifs/necessary sequence that we put in front of the minimal promoter turns out to be
the useful car to drive the expression in guard cells, for example – shows that that specific chunk/motif is
sufficient for promoter activity – driving expression at a particular place at a particular time.
- Then we talked about sort of motifs that one might identify through such a process – these less than 20 in
length nucleotides motifs where we can have slight sequence variation in them & nucleotide changing here
or there but overall, they are minimally going to be 3 invariant nucleotides.
Slide 1 - How do we know that specific proteins interact with these DNA motifs
that we identified in the experiments described previously (gain
Slide 2 - Way we can look at the interplay b/w a protein & a DNA segment is
done very nicely in vitro (outside of the cell) using electrophoresis.
- Have indentified target sequence, now we want to ask can a specific
protein interact with those sequences.
Slide 3 - Just imagine that we’ve already identified a protein that can interact
with this target motif.
- Going to run motif through electrophoresis gram using PAGE gel
electrophoresis – run it from the cathode to the anode – as DNA is
negatively charged, it’s going to migrate down the length of the gel &
we set the whole thing up so this fragment is going to migrate right
down to the bottom.
- At the bottom, have the unbound (free-running) motif called the target.
- Way in which we’re going to eventually visualize where the migration
has occurred in the gel is by making use of radioactivity – the target
DNA has been labeled with radioisotope in one of the nucleotide bases
that has been added to the double stranded probe target. At the end of the
process, if we take our gel & expose x-ray film to it & develop x-ray
film, we will see particular bands that correspond to the location of the
- We’ve hypothesized that a particular protein can interact with this
motif – how can we test that the interaction actually occurs? We either
purify the protein through some purification process from the cells in
which it’s normally found or we make it recombinantally & purify it from E. coli. In the end you’ve got one protein & you want to ask if it
can bind to this DNA motif.
Slide 4 - Note what happens is that in addition to the free target running down to
the bottom of this particular lane, we have another band that we see on
the x-ray film that is higher up the gel, indicating that something has
retarded/impeded the progress of the probe of the target through the gel.
What could have possibly done that? Well of course it’s got to be the
protein impeding the progress – made it for much larger MW complex &
through non-denaturing polyacrylamide gel electrophoresis, it’s going to
migrate at a slower rate & create another band – that’s our protein bound
to the target DNA.
- Nice simple method to test the hypothesis that a particular protein acts
with a specific DNA motif.
- Example of how one actually visualizes this method at the end of
looking at the x-ray film. This then is what is the output of such an
experiment where what we have on the lane on the far left hand side is
the labeled target alone – can see that it runs right down to the bottom.
Then what we have is our labeled target (our DNA sequence motif) +
recombinant protein that’s in the next lane & what we observe is a
shifted or retarded band in the gel indicative then of binding of the motif
by the protein.
- Now what we do to show the strength of the interaction is to add
increasing quantities of unlabelled competitor & as one adds increasing
quantities of unlabelled motif, it should outcompete the labeled motif &
that we will see is a decrease in the amount of radioactive in the band as
we go across. Increase concentration of unlabelled target DNA that
functions as a competitor & again, instilled with the protein present –
note how the amount of radioactivity in that band decreases – that’s b/c
we’ve just displaced the radioactively labeled with non radioactively
labeled competitor – this shift is still there – the protein is still binding to
the DNA but b/c we’ve added cold unlabelled non-radioactive
competitor, it’s bound to that preferentially over the radio-labeled &
therefore when we visualize using x-ray film, the band diminishes.
- By taking a look at how much the concentration of the unlabelled
target is required to get rid of the band, we have a measure of the affinity
b/w the protein & its target.
Slide 5 - This method has tested the hypothesis that a specific protein has bound
to the DNA target motif. What if we don’t know anything about what is
important to bind to this sequence that we’ve identified as being
necessary & sufficient. What if we’re just interested in identifying the
proteins that can bind to it in a mixture of all proteins that are in a given
- Have again our target sequence that has been radioactively labeled &
it’s going to run to the bottom of the gel but in contrast to what was
shown previously, now what we have is a mixture of proteins – some of
them are going to bind to the motif, others will not, there may be
multiple proteins that can bind to the motif & in the example provided,
that is precisely what occurs – multiple proteins can bind to this DNA
- B/c the proteins themselves all vary in either MW or iso-electric points,
they’re going to migrate differentially in the gel so as opposed to seeing
just one shifted band (one shift corresponding to one protein), we’ll see
multiple that correspond to the different proteins binding.
- There’s the unlabelled probe at the bottom & when you run it together with the mixture, you end up with 4 bands – the unbound that is down at
the bottom, then the 3 shifted bands representing the sequence being
bound by the different proteins.
- What if the red & green bound simultaneously, would I end up with a
shift that is higher or lower than that? Higher up the gel – that can occur.
Slide 6 - Now we’re interested in identifying what those proteins are.
- Start with a column that has a matrix (very small gelatinous beads)
with beads that have DNA bound to them of many different kinds of
sequences. Now we flow through that column/matrix total cellular
proteins. The only things that are going to bind to the column are going
to be those that are able to bind to DNA & so at low salt what will
happen is that those proteins that have very poor affinity for DNA will
flow right through the column & those that are able to bind will remain
bound. Then add medium salt wash (increase salt concentration) & that
will displace through ionic competition the interaction b/w the proteins
that had specifically bound & the DNA that is on the column – wash all
the DNA-binding proteins.
- Recall we’re after a protein that binds to a specific motif so now we
take our general DNA binding proteins & we ask which ones bind to the
motif that we’re interested in so we run them through a column using the
same method again but in this instance we’ve replaced the generic
sequences with our specific sequence. Now bound to the matrix we have
this specific sequence GGGCCC & its complementary strand. Then we
take our DNA binding proteins, allow them to flow through the column
& now we start with the medium salt wash & anything that just binds
DNA generically will come off with the medium salt wash.
- In order to get the proteins that specifically bind, they’ll remain bound
to the column/DNA sequences there. Now we need to use a high salt
wash & that will remove the proteins that are specifically bound to the
Slide 7 - Now what we can do is go forward & characterize those particular
proteins & their interactions with their DNA binding partners.
**Don’t need to know anything besides the specifics in this lecture**
Slide 8 - Binding site selection assay is a specialized electrophoretic mobility
shift assay conducted in multiple steps. So far we’ve identified a protein
that we know can bind to a motif but we don’t know if that’s the motif
that it preferentially binds to.
- In this instance we take a pool of all possible radioactive
oligonucleotides & we mix it & conduct an electrophoretic mobility shift
assay that’s going to run right down to the bottom – this is a mixture of
all possible nucleotides in order. What we want to do is ask which one of
that mixture is preferentially bound by our protein of interest. What will
happen is those preferentially bound will end up being shifted (retarded)
in their mobility in the gel.
- Use scalpel to cut the band out & purify it – designed oligonucleotide
that the random sequence that is in the middle is flanked by 2 primer
binding sites that we know the identity of that so we can use as primer
binding sites in a PCR reaction.
- Take purified band, PCR amplify it – whatever those sequences were
in the middle that our protein was able to bind to & conduct the assay
where you radiolabel it & conduct the assay again – end up with a
shifted band & now it’s made it through 2 rounds of selection.
- Do the whole process again, repeat it multiple times (usually 5 times)