Proximity Ligation Assays: Detect protein-protein interactions and histone modifications

Proximity Ligation Assays: Detect protein-protein interactions and histone modifications


Welcome everyone to the final How-to
Talks by Postdocs sponsored by the VCU Tompkins McCaw Library the right center
for clinical and translational research and the VCU postdoctoral Association.
Today I’d like to introduce to you Dr. Salvador Sierra. Dr. Sierra obtained his MD
and his PhD from the University of Navarra. His PhD project is entitled
cannabinoid and adenosine heteromers and the output basal ganglia and nuclear and
the macaca model of Parkinson’s disease After obtaining his degree he joined the
Lakshmi W lab at Mount Sinai there he studied the interaction of cannabinoid 1
and Delta opioid receptors. Recently he joined VCU in the Gonzalez
Mesa lab as a postdoc where he’s exploring the epigenetic signature of
cannabinoid and opioid receptors. Today he’ll discuss the proximity ligation
assay that he’s been working with for over 10 years for this talk. Let’s
welcome him. [applause]. Thank you for this nice introduction. First I would like to thank VCU Libraries especially Karin Gau, Koenig, [total chameleon and dr. Denison]. Today I’m going to share with you my experience with proximity ligation assay. So we’re going to start by showing you my my presentation. So
first I will talk about colocalization and how we use PLA or drugs ligation
assay to detect it. Then I will move to talk of the auto fluorescence and how we
can reduce it, and finally I will talk a little bit about how to use PLA into
epigenetics. So first what do we mean by
colocalization. Colocalization means co-occurrence, the spatial overlap of two probes or the co-distribution of one protein into another in the sample of interest. So basically I show you here proteins-
those are G-Coupled Protein Receptors or GPCRs. Those are
proteins that have seven transmembrane domains with an N and C-terminal tail. So
those proteins, a in red and protein B in green are together, we can say that they are colacalizing. However, we cannot say that they are interacting with each other and it
will signal for trafficking basis. I’m going to show you one beautiful
example of that. GABA is the main inhibitory neurotransmitter to the
Central Nervous System and the target is the GABA(B) receptor and these are GPCRs as well. This receptor comprises two subunits. Subunit B1 (in red) and subunit B2 (in green). Only when the two subunits
interact with each other there is a migration of the receptor from the reticulum to
the plasma membrane and there is a signal upon the activation by the various
ligand GABA. So what happens when we delete the
subunit B2? The receptor will not reach the
plasma membrane and will be locked into the red reticulum. What happens when we
delete subunit B1? The receptor will reach the plasma membrane but we will
not find any signal response so basically it’s a nice example to show
what do we need to do when we want to show that there is a functional interaction
and experts in the field proposed three criteria to show the
functional interaction between two GPCRs. The first one is the
colocalization of two proteins in the same cell and same intracellular compartment
and for that we can apply a variety of techniques including the PLA I’m going to talk about today. The second
criteria refers to the properties of the interacting proteins. For that, we use
signaling ligand-binding and trafficking assays. The third one refers to the loss
of the protein-protein interacting signaling or binding properties. So for
that we use reagents that create interaction showing that there is a loss
of the properties we already saw in criteria 2. So among all of the
three criteria, the most important one is criterion 1 and this is the one
I’m going to talk about during this presentation. So the most widely used
technique to show colocalization is immuno- fluorescence. Here is indirect immunofluorescence where we use two primary antibodies against our targets those are raised in different species and basically energy
bodies are tagged with [fluorophobes] one for the red one for the green channel. We will say there is a colocalization when there is an overlap of the green
and red channel so we see our signal in yellow when they colocalize. Another
antibody based technique is the co-immuno precipitation or CoIP. Basically we use our antibody of interest against the
protein one and we pull down the lysate by using beads. After that we wash
and we elute and we run our sample then gel, then we obtain the different weight depending on the interaction of the
products and then there is another technique widely used in these days is
called FRET and BRET. Basically we use proteins that are that with luminous kind
or fluorescent proteins whenever those proteins are close enough, the
donor molecule here in blue will transfer the energy to the acceptor
molecule and we will measure the light and obtain this saturation
curve but the most important technique to show colocalization is immunoelectron
microscopy. Basically we use our primary antibodies and the secondary antibodies are
gonna be tagged with different sized particles so whenever we find two
different sized particles in our sample that are close enough, we will say that
there is a colocalization. Here I summarized all the techniques with the advantages
and drawbacks for all the techniques used should show colocalization. As I said,
the widely used one is immuno fluorescence but the resolution is very poor.
Coimmuno-precipitation tell us that two proteins are in the same immuno
complex but it doesn’t mean that there is a direct interaction. In the electron
staining as I said is the gold standard for colocalization however is
technically difficult and expensive. FRET/BRET give us a very close
resolution to immuno electron but the problem is that we usually do the experiments in
vitro and we are adding molecules that are not in the
native system, cells or tissue. Then I’m going to focus today about the
proximity ligation assay. We can obtain a very good resolution close to
FRET to detect protein-protein colocalization and most importantly we can apply this
technique in both cells and tissue. However as in your antibody based
technique, if our antibodies are not good it’s not going to work. Right? So a brief
introduction to PLA. This technique was developed by the early 2000 and it
was published in these two papers so in the last 10 years the numbers of papers
using PLA to show colocalization has doubled as you see here and I’m going to
show you just the basics of PLA and then I will go in detail through the
experiment, the pro/con in cells as well as in tissue. So basically we incubate our
sample of our protein 1 and protein 2 colocalizing with our primary
antibodies raised in different species Let’s say a mouse and a rabbit. Then we incubate
our sample with a secondary antibodies that are attached to a oligonucleotide and those are complementary to each other so they are called DNA probes and they are targeting
the species of the primary antibodies and you see here the anti mouse and the anti rabbit. Then there is
a ligation step where we basically use a ligase to make a circle template in
order to join those two whatever nucleotides and in the last step
polymerase will make several copies of this template and fluorescent label
oligonucleotides will hybridize to the amplification. There are two variations
of this technique. Let’s say we look up 2 antibodies that are raised in the same
species, rabbit and of course we cannot apply the secondary antibodies to target
because they are going to bind both ones so we can link the [reward] nucleotides
directly to the primary antibodies. The problem is that it is quite expensive
and then imagine we do have an antibody that give us a very high background, we
can use the single recognition approach to increase the signal to noise ratio. In
any case we are going to find a signal which is a fluorescent red dot like as
you see here of course we use another fluorophore, it might change. So I’m going to talk about the protocol usually route insulates it depends on the timing to
incubate your primary antibody. You can do cells
fixed + – frozen tissue and formalin-fixed paraffin-embedded or FFPE sections. So
basically we printed our sample we go through that later then we block one
hour 37 we might use the clock in [?] from the company or ours then
we incubate overnight at 4 degrees with our antibody. The next day we wash with a
buffer which contains sodium chloride [?] and we
incubate our sample with a PLA probes 1 hour and
37. Then we wash again and we incubate with ligands
for half to one hour at 37 and then we wash again and we perform the
amplification where we will obtain several copies of the template we obtained in the ligation step and then we just need to hydride our samples completely
and a coverslip by using a multi medium [?]. After that, we will need a confocal microscope to obtain our images. So I
want to give you some tips about pre-treatment in cells. I usually grow my cells in the 16 well-chambered slide or the chambered coverglass. I don’t
really recommend you use the cover to grow your cells for three reasons;
first they are really fragile and they can they break quite easily, second the
surface is bigger than the one we use for the chambered slide and we are
going to use more reagent and but you might know the PLA regions are not
very cheap and the third one, one of the critical things with PLA is that the sample
shouldn’t get dry during the [?] so we by using [?] they usually get dry.
Then there are some cells that are easily detach from the well so for
example red cells usually coat with Poly- -D lysine however maybe with [?] cells,
you won’t do it and then it is very important when you see your cells into the waste bag confluency is around forty to fifty
percent so you can see perfectly well the border of the cells. If we have
very low confluence with about morphology or mono layer is going to be
difficult for us to count the dots in the sample. Now I’m gonna go through
the pretreatment for the cells so here we do have our chambered slide I usually
don’t use the words that are in the side of the slide. Basically because if you
work with inverted confocal microscope these sites will get the
objective will get stuck and you will not be able to see this well. So we use
our medium with our [serum] and the antibiotics we incubate for 2 to 3 days
days depending on their morphology or the size and their confluence then we
gently wash with a pipette with videos then we fix by using a PFA of 15 minutes
room temperature and then we detach the plastic frame so we can wash our slide
in [compendium] with PBS but this step the cells would not detach because they
are fixed and we don’t want to pipette again because it will increase the
chance for the cells to get out of the one.Then we wash with a buffer which
contains glycine in order to quench the aldehyde group. We do that to get rid
of or reduce the auto fluorescence then we permeabilize by using triton x-100 and
finally we wash again with PBA x 2 without the detergent and we follow the
PLA protocol as I showed before. So this is the expected [whistles whistle cut] here
is the positive control wake up our cells with the two proteins of
interest and we can see the red dots in all of them. The negative control
we can make several; one is like avoiding the primary antibody. Its a [contrast] of the the background of the technique
or you can use cells where there is a lack of one of the proteins of interest
so [directly] background which is slightly higher. Here is another example
so here are cells that express the both proteins of interest in the second
column we avoid one of them and in the third we avoid the other and then we can
count the number of dots or cells I usually can’t say cells per condition,
I like to replicate my experiment at least three times to make
sure that the results are real. So we can also combine our PLA experiment with
markers especially when we want to explore the interaction that takes place.
For example we can use Phalloidin which is an F-actin marker to visualize the cytoskeleton of the cell. We can also use the BacMan 2.0 technology by adding the reagent into the living cells the night before so this [popular] virus will express GFP
into the plasma membrane and also you want to detect the lysosomes we can use an antibody Anti-LAMP1 and in this case as you see in the white arrows the
overlap of the PLA signal and the lysosome marker, you [pass] a yellow signal. I’ve used some markers which are used for living cells like [lysol cracker]. In my [case], it didn’t work however my [tracker] works with PLA and now I’m gonna move into the tissue samples. So I’m gonna start
with the FFPE samples we usually do a hydratation step where we immerse
our sample in Xylene and a series of ethanol to finally water then we do an
antigen retrieval permeabilization step where we heat our sample in sodium citrate with [twin 20] for
10 minutes at 95 to 100 degrees then we wash and I dry and we encircle our
sample by using a hydrophobic pen so so we will keep our
reagents the PLA in our sample. Now talk about the fixed and frozen samples [we so
our cryo value] then we wash [invidious with without] the protection solution
then we mount our sample by using a buffer with [debugging]so it makes
it easier to mount the [ ] and then we permeabilize by using x-100
the rest of the steps are very similar to the previous protocol and this is
[respected whistles] we set up this is a section of the brain lateral amygdala
and you can see how there are cells that are full of red dots whereas others do
not have so it means that this technique is showing specifically where
the protein-protein colocalization takes place and we are going to compare this
with the colocalization [we say no tweaking] with immunofluorescence. Here is the result so for this protein DPR 171
we used a green channel. For the DPR 83 we use the red Channel and they
overlap tell us the colocalization however as you see the signal is very blurry. Its difficult to say which cells are expressing both proteins. However PLA
gives us a better resolution in a more reliable
protein-protein colocalization that makes it easier to count the number of
these clocalizations by the number of dots. So we can combine with another
marker even though we also have the green and the far red channel. So as I
told you we can combine with other markers so this is a [instead] of a
section from the human and I use a marker for [European] receptor one
combined with PLA in the right figure I used [EF a field] antibody to to label the
astrocytes and you can see the PLA signal in the cytoplasm and close to the
nucleus. So sometimes I would like to comment with you is the choose a marker
that is not raised in the same species as the PLA probes. Let’s say you use your
proteins of interest and you use the antibodies raising mouse and rabbit for
the marker you will use a goat or guinea pig or a chicken right so incubate all together you don’t need to [extend] the experiments in great all together at
four degrees overnight then in the secondary antibody you can actually PLA probes in the step of one hour I recommend you that the secondary antibody is raised [] because the PLA probes are and also
linked with a fluorophore for the green channel if possible and make sure that
there’s a community body it’s not raised in a species targeted by the PLA
probes. Let’s say that the secondary antibody for this marker is
raised in goat but one of the PLA probes is targeting gold so we will get an
artificial [scene] and also we can do PLA combined with electron microscopy.
Basically instead of using fluorescent label or oligonucleotides, we use
peroxidase label oligonucleotides and then we increase our sample with
gold with an antibody start with gold particles and as you see in the image
every dot means the colocalization of two proteins in this case the one and the
two receptors and we can see that those are in the golgi. So the summary for this
part is that DNA is a powerful antibody basis technique for detecting protein-
protein colocalization involved in both in vitro and ex vivo. Also contrary to other
techniques you don’t need a lot training and it
allows quantification and third you can be detect where the protein-protein
organization takes place in a sub- cellular compartments. So,
question or for this part okay well to the next part. So this is an
image of PLA red channel combined yes this is PLA experiment
combined with the [new aim] in the green channel and [now before the new clip]
anyone when I say something about this okay
where are the red dots? There are none right? But there are yellow dots. There is not any yellow channel here. Anyone has any idea? Yellow is the red what do you mean? It combines with another signal. That’s what I thought So the PLA was
overlapping with a signal from the new end right so that’s why it is yellow. However, I showed you this image that there is no yellow dots you can see some
yellow but is mostly red okay so when something here is that didn’t work
PLA for some recently didn’t work at the [end of] primary antibodies or whatever
but it didn’t work so this is auto fluorescence and I’m going to introduce
this new guy or old guy because [well product up ] build with this
working you know fluorescence and so when we fix our cells or tissue with PFA
or any other fixative like glutar- aldehyde or some samples like Elastin or
red blood cells that contain hemosiderin have autofluorescence itself and as you see
here this sample has not been treated with any fluorophore or antibody
we can see dot-like signal in all three channels; green red and far red and they are in the
same location. Also it will use human and eight animals we can find this signal
which is called lipofuscin. Lipofuscin is a protein that is accumulated
in the lysosomes as we get older and you can see the signal is a
pseudo color image in all channels from blue green to red infrared. So how
do we how do we face things especially when our signal is not [mere talk]. So I’ve
been using several treatments in my hands I’m going to show you how I could
use it in order to reduce the fluorescence signal so here is a section of mice in the stratum which is a big ring and here I [end] it with any cure for a marker and you can see these dots
we which are present when we yes take our pictures with the four channels. If
we split our channels we don’t see that in the blue tell channel I must say that
here I use [19 million whipped] up in so we see the nucleus but we don’t see those dots
right but we can see them in the green [cretin] for red channel. If we use Sudan Black B, again no anti- bodies no markers we get rid of this
dot-like signal however we have some background mostly coming from the far
red channel so whenever we use Sudan Black B I recommend if possible to
work with green and red channel and leave the red channel for PLA. Now I’m going to combine this with a regular immunofluorescence for [big
blue tube] and secondary antibody with fluorophore for five six eight similar to the PLA
one and you can see how the signal is mostly in the red channel
[Maureen day for red even go] and what about the human sections the FFP, Lipofuscin. So here is a section of [the rod as the 4-h way] receptor in the
green channel we’ll do the same experiment with the same antibody in the
human [STI] we get this yellow brownish dot-like signal and when we split the
channels we can see the signal in the blue red green – far red that
I suppose will be there as well and we are going to do PLA by using Sudan Black B
so here is the positive control in which for PLA we even use to [them black P] if
we let this dot-thread like signal coming from the lipofuscin but also dot
signal which is very weak here compared to lipofuscin.
To get rid of this ugly signal, we use Sudan Black B
after the PLA experiment we get these dots nicely here and most of the autofluorescence is [based on] but it doesn’t really interfere with our experiment I
think here is the negative control. So how do we [repair] the Sudan Black B, we basically dissolve our powder in ethanol we stir at room temperature overnight
we filter the next day and we incubate our slides after the PLA wash for 10
minutes at room temperature in an orbital shaker and then we just wash
with a buffer point 0.01x, 3 times and I recommend to use it yet into the sample to get rid of the any particle from the Sudan Black B. So the summary of this
part is that you should always check your auto fluoresence whenever you do experiment that
involves the use of fluorescent markers and I recommend to you use glycine in order
to reduce the autofluorescence for cells before experiment and Sudan Black B in order to get rid of the PFA or lipofuscin autofluorescence Any questions here? [] Any sort of cell [line] like because before, you were using a different type of dye. Why don’t you just use Sudan Black and not use any other dye at all? You mean for cells? Yes for cells. For cells, I never used Sudan Black B. [ ] I must say that when I used glycine, for me it works
so I never really used Sudan Black B Okay so you used glycine for everything but Sudan Black is just for those cells you don’t have some autofluorescence I just don’t know why Sudan Black you just can’t use it for everything even though it has all the fluorescence [ ]. Yeah so, I’m not the expert but people who have been working in histology they have tried many protocols
we use sodium borohydride, glycine we use sulfate [ ], Sudan Black B [mask block] is the commercial one and they show
their images and Sudan Black B is the best one. Why? I don’t know it’s very dirty so you need to use gloves and
I guess it also reduces your fluorescence from your marker but mostly it gets rid
of the fluorescence you don’t want for your experiment okay. So now I’m going to switch a
little bit so I’m going show you how we can [say] PLA for epigenetics. So briefly, PLA is the science that studies the the heritable changes in
gene expression either activation or repression. Itdoesn’t involve the changes
in the sequence so in the first level of organization of the DNA strand is around
the histones. Histone proteins control the access of the transcription
machinery into the DNA reading and for the histone to move, it meets some
modifications in [today] even as it grows The most common ones are acetylation
methylation, ubiquitination and the common technique we use is called chIP
or chromatin immunoprecipitation so we basically take our lysate and we closely by using PFA after that we isolate any shear our DNA we pull down our protein histone
bind to the DNA by using an antibody and [pits] and then we reverse cross link to
get just a DNA and do a real-time PCR and what we get is a graph where we get the histone modification. Here is a demethylated H3K4 for the Gene A whereas the negative control [ ] is very low. and this means just what I said However chIP does not permit the analysis of this histone modification in a given gene locus in individual cells especially when we use tissues that have a heterogenous
population of cells and there is a quite new paper show us how we can detect
this histone modification in a specific cells by using PLA. Basically we need to
design DNA probe labeling the gene promoter and it will be labeled with biotin then we use an antibody against the histone modification of interest and an antibody against biotin if we just do
the protocol I showed you before Here is the proof of concept for this technique
so basically we isolate a smooth muscle cells from the artery and endothelial
cells and by doing chIP we see that this histone modification is only present in
a smooth muscle cells so we check by chIP now we will now see if we can demonstrate this by doing PLA in the [ ] and we show here data for the nucleus PLA in
the red channel [I touch you ]is a marker for the smooth muscle cells and [CV 31 ] for the endothelial cells in the inner region and we can see how these dots are just in the outer
region of the artery so there are just in the smooth muscle cells here also in
the insect so this is telling us that PLA can detect a specific histone
modifications in certain or specific cells So as a summary of this part, chIP is a
powerful technique to detect histone modifications at the cells and tissues.
However it doesn’t allow us to characterize ex vivo tissues. In situ
PLA will be a useful tool for the investigation of the specific cellular
subtypes carrying this PLA modifications. The bad news contrary to
PLA in situ, PLA is a challenging technique where we need to combine in situ immuno
markers and PLA so every condition could or potentially interfere with one
another and I would like to thank the previous lab [ lab] at Mount Sinai where I did most of the experiments
I showed you here and also my current lab trying to develop this in situ PLA technique so if you have any questions if not feel free to contact me anytime. [applause] I asked you right at the end there will you try to target my understanding of
the at the end were you trying to target a specific gene by having a DNA
sequence and biotin? Yes. How close to that DNA sequence have to be to the promoter
region or you know what i mean yeah so basically you need to do in situ
right so if you see two works like histone the histone modification is
going to be detected if it’s there so I usually do first chiIP so you know
chIP if that works I will probably use this antibody to do in situ PLA but first you need to make sure that your in situ is working and that’s
I think the most challenging part for this. Let’s say the chiIP probably you use
you can use that to make the DNA for the biotin. Is that right? So for
chIP we use the antibody yeah if there isn’t one. What I usually do is I
test if the antibody works also for immuno right so I check if yes in a
sample if my histone modification and antibody targets the nuclei
of the sample and then the in situ So by using this PLA tool can we identify as new I mean new protein [it
writing partner of our] interest you I mean can you identify the new protein
interacting for test by using this PLA tool? You mean regular PLA or in situ PLA? No the regular PLA [I’m now protein] yes well you need your antibody right? To target your protein yes no you know that
anybody but we don’t know that the complex is still associate with the
protein of interest or not yeah I mean so I mean I will probably do do first
taste in vitro and if that works I will go through PLA. Because they are trying to pull an experiment [looks like how this job you know the complex is a stable
with that] our interest of protein so we couldn’t bring it down without regular
full-dome protocol. It means you pull down your proteins [that yes ]maybe it’s cheaper
and if its works I will do in a section of tissue with PLA if it makes more sense to do that. There are many variations of these techniques so some people are doing [ ] PLA so you can also detect by [ear I start to identify because who
needs the anytibody very few I guess the idea that in some
nonspecific that kid] yeah of course I mean this is basically so if your
antibodies are not good enough or not specific enough they are not gonna work
even though it’s a background is high but it’s a specific. With PLA you can
increase this [signal to noise the] by doing the PLA signal recognition but you want to see the
interactions either [suffocate] or yeah one more question so what’s the PLA probe are
you making [artwork] oh yeah so I didn’t say that PLA was first developed
by [X values company holding] and then it was bought by [multiple] Sigma so you buy
all the regions the PLA probes the blocking reagent [on the mountain] medium
from [Millie power Sigma]. The buffers I don’t usually buy from them. I use my own. Its cheap. How expensive is the PLA probe?
The PLA probe is around $300 to $400 each In case if you want to test one
particular protein, you need to buy one PLA probe?
yeah yeah so I see. Let’s say that you are using your antibody to tighten your
protein of interest and is raising product and the other one is [raising] gold. You need two PLA probes [one until right one and] [ ]
Of course I mean you can do lot of reactions the amount of reagent I have
is very low so I cover one spinal cord section of the mice with just twenty
micrometers and the dilution for the PLA probes is 185 so and
if you get one buyer of the PLA probe you you have 1 millimeter so yeah as
long as we use the hydrophobic pen you keep your reagent there you don’t need to
[ ] something. So how can we quantitate [oh yeah ] let me show you here well yes right so
in this protocol we shall how to do it with imaging so you can count the dots
in one cell then you do your statistics I [salute] the convolution with image [ ] so yeah I need
to pay for yeah but does it significantly differ from the
[Colo Playstation like immunofluorescence] experiments?
Well that’s a good question I haven’t done the I haven’t checked the colocalization so I can measure the colocalization by doing regular immunofluorescence and doing PLA I haven’t done but for me [you said you can dots] instead of intensities; overlap intensities and also the
overlap doesn’t mean that they are closing yes as I said the colocalization
could be very unreliable nowadays with the super super resolution microscopy
you can go like lower than 250 nanometers but you cannot get 40 or 17
nanometers or proximity if you want to know a single protein I want to know the localization
instead of using I say you just using the PLA. Can you use two different antibodies
targeted to different yes at all yes then do PLA. What are the bodies of using
that will localize your protein you can do it yes absolutely [so yes did you yeah
so yeah]. So you can use one antibody so you want to detect with an antibody
against one epitope and the other antibody targeting the other epitope so we’ll do something like that so one is raised in rabbit and one is raised in mouse putting gold right basically it is the same as this because
here we just need one primary antibody taking one epitope but for your
experiment it will do that. okay so that single [return is kind of grub you just have
well] antibody but the second [algorithm] we had two different PLA probes yes it works so both are [indeed] the species but one is the PLA probe
is plus and the other is minus so you need to put probes that are complimentary
[competition right yeah] yeah there is competition and is that a problem [ ] but yes there is a competition. You are not going to get a full, yes you’re right but also for the
suspension cell to do [PLA] using cytospin to a spin [class] what
question yes you can do it I I haven’t [studied] that but it has been published so
I cannot give you the information but yes. It has been done Well if there aren’t any questions, we would like to thank [ ] for his presentation today. [Applause]

One Reply to “Proximity Ligation Assays: Detect protein-protein interactions and histone modifications”

Leave a Reply

Your email address will not be published. Required fields are marked *