Application of protein microarray in biomarker discovery-I


Welcome to mooc interactomics course In today’s
lecture we will talk about applications of protein microarrays The protein microarray
technology has potential to one of the integral tools in proteomics field Due to the enormous
potential in diagnostic and drug discovery these high density protein arrays which are
printed on glass the arrays have been used for proteome wide screening Whether it is
human yeast or different type of bacterial species these wide variety of microarrays
have shown different applications and its potential for high throughput studies In today’s lecture first I will give you
an overview of protein microarray technology which we have already discussed in great detail
in the previous lecture I will then provide you an overview of how to perform protein
microarray based experiment and how can it be applied for biomarker screening studies
After discussing the overview and general strategies for protein microarray experiments
We will then talk about different applications in the form of case studies As you know the proteome is very dynamic and
it represents very high complexity whether we are studying it at the gnome level transcriptome
or proteome or metabolome level That is why proteomics on one hand aims to study the various
types of biological problems simultaneously on other hand researchers are trying to integrate
various type of homic technologies so that a robust high throughput platform can be generated
to study the dynamic proteome As you can see in the slide till dynamic proteome
are very much linked with physiological actions happening in the biological systems and that
is why we need to integrate many of these to obtain the compressive image of what is
happening inside eleven system We want to study the proteins but we cannot
do that in isolation We need to study them in high throughput manner by identifying to
whom they interact and work as a part of team player of a complex signal transaction cascade
To answer many of the questions protein microarray platform as well as some of the other techniques
have been generated and shown potential for high throughput applications In previous lecture I provided you an overview
of different type of protein microarray technology we discussed both abundance based and function
based protein microarray platforms We have talked about different type of labeling which
includes direct labeling sandwich assay type of technology we also talked about reverse
phase protein arrays and then discussed different ways of making functional based protein arrays It includes immobilizing the proteins which
is purified or peptide fusen with the tag as well as different type of cell free expression
based approaches With this now I hope you recall all the topics covered this far and
that there are many kind of protein microarray platform which could be used to address different
type of biological questions Let me show you one video to provide an overview
of this process of using protein microarrays Here I am giving you an overview of a protein
microarray experiment Regardless of what application one want to study the overall workflow of
the protein microarray experiments remain the same Here I am showing you the experiment
of protein interactions performed on the E coli proteome chip The first of all these
chips are stored at the minus 80 degree if they are purified proteins you want to store
them at the very freezing condition so that protein can remain functional Now these chips should be carefully removed
from frigid and that allow to thought briefly followed by the washing instance First of
all transfer these chips from the minus 80 freezer to a fresh slide holder
After these chips are thawed briefly at the room temperature then one can either directly
block it in the blocking solution or use the PBS twin for brief rinsing Blocking is usually
performed at room temperature for an hour or at four degrees overnight depending on
different type of applications PBS and milk or super block or BSA are commonly
used blocking solutions One can typically use even a pipette box or a small box at approximately
30ml or 50ml of the blocking solution dip the microarray slides inside the solution
And once the blocking is completed you can remove the slide and tap again the paper towel
to remove the excess milk By performing the blocking experiment or blocking
step make sure continuously makes the slide even if you are preparing the entire set If
slides are left uhh sitting on their side without mixing then they will dry and then
the slide will appear dark background after the scanning After thawing steps are completed
then depending on your application you can either apply primary antibody if you want
to do a quality control chip for example or one can use a query protein for example if
you want to study the protein protein interaction So in this study we are talking about protein
interactions So let us say a query protein of interest which for which you want to study
the protein interactions you can take that protein query protein and then apply that
on the protein microarray slide After addition of this query protein you need
to cover it with the cover slip and incubate it at the room temperature for an hour or
one can optimize this incubation condition depending upon their experimental requirement Once this step is done then you need to wash
the slide with PBS twin for three times usually five minutes wash for three times at the room
temperature with a gentle shaking on a rocker shaker is most commonly used In the microarray experiment one need to ensure
the proper washing and gentle shaking throughout various steps to ensure that the slides are
washed very neatly otherwise you will see very high background on the slides After this step one can add the secondary
antibody labeled with the either cy 3 or cy 5 conjugate labels or one can use the HRP
based detection system for detecting the signals but prior to the this step one need to dry
the chip by centrifugation So rinse the slides quickly and then centrifuge it to dry it or
one can also use the compressed air for drying these slides Once the slides are dried then they can be
scanned at the appropriate wavelength With this gives you a glimpse of an overview of
a microarray experiment which is shown in the 3d animation here but depending upon your
biological question various type of modifications can be made After giving you an overview of protein microarray
experiment let us now talk about protein microarray based applications Protein microarray chips
have been used to assay for wide range of applications such as protein protein interactions
protein lipid protein nucleic acid which includes protein DNA and protein small molecules or
protein substrate identification Protein chips have also been used for drug
and drug target identification and kinase substrate identification I will try to summarize
few more studies in detail to familiarize you with different type of applications which
can be performed using protein microarrays Protein microarrays have been widely used
for biomarker identification and protein protein interactions as well as different type of
protein modifications As you can see in the slide the protein interaction network transcription
regulation DNA damage repair antiviral response drug target identification are various type
of applications which researchers have tried using protein chips Similarly studying the kinase networks the
dynamic function regulation as well as studying protein turnover and protein modification
have been used on protein chips Let us further discuss biomarker detection
this is one of the most commonly used applications which researchers have tried using protein
arrays Because the clinical samples are very precious very limited and usually do not have
access to the large volume of clinical samples So on one hand you would like to study as
many as protein possible but at the same time we would like to use very small volume So protein chips provide this high throughput
platform to screen patient samples in a very volume Therefore protein microarrays offer
a very appreciable platform because you can use few microliters of sample and that can
be used to prove thousands of proteins On one hand we are using very small amount of
sample that you can actually probe for large number of proteins Often some of these samples
are very challenging but by using protein microarray applications one could get some
very useful information for biomarker discovery Let us start with case study first identification
of differentially expressed proteins .. in cancer using high density protein microarrays
a study performed by Hudson Nitol In this study authors took serum from 30 individuals
suffering from ovarian cancer and 30 healthy individuals For each of this type of samples
they used 300 microliter of serum sample diluted in correct buffer and then applied that on
human proteome arrays These arrays are obtained commercially so
they wanted to compare antibody response of various type of proteins in ovarian cancer
serum as compared to the healthy controls Identification of tumor associated antibodies
and targeted protein antigens were performed on protein microarrays These protein microarrays
contain more than 5000 GST fusion proteins which were first probed with antigens tmt
body As you can see in the slide first of all quality
control experiments were performed by using entire GST antibody because all the clones
contain GST tag So first of all there was a need to ensure that all the clones express
protein uniformly and whole of the protein array can be used for uniform screening Now once the quality control experiment is
done then you can start actual screening of large number of samples on these arrays for
discovering biomarkers In this study the authors used sample from ovarian cancer and healthy
individuals and then applied it using human proteome chips The different boxes which are
shown on the bottom panel on the right hand side indicates there are several positive
and negative controls which were printed on the chips to unsure that assay is working
fine and there are certain proteins shown in the centers which are potential biomarkers Based on this study authors werel able to
identify several differentially expressed proteins which included lamin A which is one
of the nuclear membrane organization protein Structure specific recognition protein SSRP1
which is involved in the regulation of transcription RAL binding protein which is important in
the transportation IRF6 which is crucial for transcription regulation MAGEB4 is reported
as cancer biomarker COIL which is found in the nuclear body although its function is
unknown NOB1P which is adenocarcinoma antigen the function of this protein is unknown CBLB
which is involved in signal transaction By using high throughput approach high throughput
protein microarrays and screening for more than 5000 proteins authors will able to identify
several differentially expressed proteins These are some of the proteins which they
conceded quite interesting and selected them for further validation experiments Once protein targets were identified authors
used immunoblot and western blots to validate their findings The protein sample of cancer
and normal cells were analyzed by using immunoblot assays and they used antibody the specific
for Lamin A or C and SSRP1 proteins These proteins were considered interesting based
on their proteomic finding The Lamin A and C were greatly elevated in cancer samples
as compared to the healthy individuals and appeared quite interesting although the western
blot analysis was performed by using healthy and ovarian patients samples and then they
used antibodies specific for Lamin proteins as well as p53 protein As you can see on the right hand side of this
slide the Lamin A and C were probed by using western blot Similarly p53 protein was probed
by using anti p53 antibody so these immunoblot assays they have performed The western blot
analysis confirmed and validated that the proteomic findings based on protein microarrays
were quite relevant and elevation of this proteins were validated by independent techniques Now further to the immunoblot assays authors
also used tissue microarray based analysis So the tissue microarray analysis of different
stage of ovarian cancer was performed These microarrays containing the representative
tissues from various stages of ovarian cancer such as stage 2 3 and 4 tumors were probed
for Lamin A or C SSRP1 and cancer antigen CA 125 These results are shown in the slide
which is redrawn from the results presented in the manuscript Lamin A and C shown in the top panel SSRP1
shown in the middle panel and cancer antigen 125 shown in the bottom panel Overall this
was very interesting study which revealed that these three can add a tissue markers
which were immunostained can produce a very robust signature of ovarian cancer in tissue
specimen Although authors have used serum and their
motivation was to develop the serum or blood based assays they thought that at least the
robust signature in the tissue indicates that these proteins are overly presented in serum
sample and probably this marker could be used for blood based or serum based assays Let us now move on to case study 2 identification
of novel serological biomarkers for inflammatory bowel disease using E coli proteome chip crone
disease and ulcerative colitis These are chronic idiopathic and clinically heteropathic intestinal
disorders which are In this study authors have used E coli proteome
chip to screen and identify novel biomarkers associated with inflammatory bowel disease
study was performed by Chan Etal in 2009 In this study first of all authors have used
E coli proteome chip they used more than 4200 proteins obtained from E coli and then they
collected serum from healthy controls and clinically well characterized patients with
inflammatory bowel disease .. examples from chrome disease and 29 from ulcerative colitis
were collected and compared with 39 healthy individuals The protein spots were recognized by serum
antibodies and visualized by using cy 3 labeled goat antihuman antibody After looking at the
overview of the experiment let us now discuss some of the findings This E coli proteome chips were probed with
serum from CD patients which is shown on the left side which is crohn’s disease and healthy
control is shown on the right panel in the slide The cy 3 labeled antihuman immunoglobulin
antibodies were probed on the chips which allowed the visualization of immuno reactive
protein spots Some of this spots are shown in the center
and it shows the comparison of crohn’s disease versus healthy individuals After performing this screening experiment
on chip author generated a heat map of 273 differentially expressed proteins which were
identified using healthy controls and crohn’s disease samples They also performed a comparison
of UC HC and CD which are shown on the slide in Venn diagram on right side It shows that
differentially expressed proteins should very limited overlap even the HC versus CD and
CD versus UC After studying these differentially expressed
proteins authors used these proteins to understand there functional role and how they are distributed
in the silver compartments HC CD and UC were then further used to define the silver components
and functional role which is show in this slide Authors tried to categorize the proteins
in all the three groups as per that silver location as membrane cell wall macromolecular
complex intracellular and periplasmic space and the cell projections and looked at the
differential response By using the supervised learning algorithms
ktop scoring peers authors identified two sets of serum antibodies that were novel biomarkers
for specifically distinguishing crohn’s disease from healthy controls As you can see
in the slide that healthy controls crohn’s disease and ulcerative colitis all these samples
can be distinguished and especially the healthy controls versus crohn’s disease cross by
using supervised learning algorithms After studying the application one based on
the biomarkers and another based on E coli proteome chips let us now look at few more
examples of biomarker discovery in brief to give you an overview of different type of
applications and data analysis which can be performed using protein microarrays In this study we will talk about identification
of potential diagnostic markers for infection from neisseria meningitides Neisseria is most
common cause of meningital disease and also causes epidemic outbreaks To investigate the
immune responses to the phase variable express proteins .. apply protein microarrays to screen
the meningitidis patient serum This is the first study which aims to investigate
the genetic phase variations in the pathogens Authors first amplified all the 102 known
phase variable genes obtained from the neisseria meningitides and expressed and purified these
proteins in hydrologal system using E coli They were able to purify 67 recombinant proteins
because not all the proteins are able to produce and therefore they were limited with 67 purified
proteins for further investigation using protein microarrays This protein microarray platform was used
to screen 20 patients serum and healthy individuals After screening authors identified 47 immunogenic
proteins out of which 9 proteins were quite reproducible including its phase variable
opacity protein OPAV which was very reproducible in many patients Let me provide you the overview
of this study in following animation This study was performed by Stellar et al
Bacterial protein microarrays for the identification of new potential diagnostic markers for neisseria
meningitides infections Authors simplified and sub cloned 102 genes from the neisseria
species for expression in the E coli These clones were grown for overnight at 37 degrees
centigrade in the antibiotic containing medium after which the protein expression was induced
by addition of IP2G The cells were harvested 4 hours after induction and then protein was
purified The proteins were purified based on the specific
nickel NTA binding after various elutions were collected these
fractions were further separated on the SDS page gel to check the purity of these probes
This shows the SDS page separation of these proteins based on the molecular weight Now authors were successful in purifying 67
proteins and then these purified proteins were further printed on the nitrocellulose
coated glass slide by using the robotic printer Once these protein arrays were generated it
was used to probe the serum from 20 convalescent patients by incubating it overnight at 4 degree
centigrade After overnight step array was washed with PBS and detection was carried
out by using cy 5 labeled secondary antibody The excess detection antibody was washed off
array was dried and then scanned by using a microarray scanner Authors detected 47 immunogenic
proteins one of which showed response in 11 of patients This protein microarray platform
was successfully used for detection of several other disease biomarkers and this is one of
the application which is shown in this animation Let us now move on to case study 4 the human
prostate cancer screening for identification of potential biomarkers study by Milet et
al In this study researchers used antibody microarrays containing 184 unique antibodies
which were printed on microarray surface and they used two different type of substrates
containing polyacrylamide as well as polylysine coated glass slides and further used to screen
the prostate cancer patients for potential biomarker identification In this study authors used 33 cancer patients
and 20 controls and obtained serum samples from these subjects and employed to study
the abundance of various proteins present on microarray slide They optimized different
parameters for the measurement and once these conditions were optimized they used the microarray
system for identification of various potential biomarkers Investigators used robotically spotted 184
unique antibodies on polyacrylamide based hydrogels and polylysine coated glasses slides
which they probed with sera from prostate cancer patients and healthy individuals Let
me use this interactive animation and show you the results obtain for the hydrogel slides Authors used robotically spotted 184 unique
antibodies on polyacrylamide based hydrogels and polylysine coated glass slides These slides
were probed with the sera obtained from the prostate cancer patients and healthy controls
Now let me use this interactive animation and show you the results obtained for the
hydrogel slides So left slide is probed with the prostate
cancer patient sera and the right one is used with the control group sera From this study
five proteins were shown to have significantly differential expression in the prostate cancer
patients as compared to the control group .. PWF protein was found to be elevated and
the remaining four proteins were down regulated as compared to the control group We will now move on to the final case study
for today which focuses on autoantibody screening in glyomous serum samples which is published
by Sayed et al Glioma is a very aggressive brain cancer with great heterogeneity It has
been classified by WHO into four grades Grade 1 being the least benignant to grade 4 being
highly aggressive and associated with poor prognosis The focus of this work was to devise some
minimally invasive biomarkers which could help differentiate between each of the grades
as well as identification of early diagnostic markers so that clinicians could identify
the disease using blood at early stage as well as predict the prognosis and identify
abrasions which could help them target therapeutics In this work investigators used serum from
17 subjects with grade 2 18 with grade 3 and 34 with grade 4 of glioma and they screened
using human proteome arrays or uprot arrays harboring over 17000 unique proteins The work flow of this study is shown in the
slide which is very similar to the one which we have discussed in case study 1 However
investigators here used different type of data analysis scheme In this study stringent quality control checks
were performed after data acquisition quintile normalization was performed to reduce the
technical variability This normalized data was further subjected to dimensionality reduction
techniques like correspondence analysis or CA CA provides a shorter list of differentially
expressed proteins that are statistically more significant This data was then subjected to recursive
feature elimination models such as support vector machine SVM to deduce a list of significant
classified proteins from this list The study revealed panel of interesting proteins
like IJG1 SNX1 They were up and PQBP1 I1 proteins as down regulated The sorting nexin 1 or SNX1
is known to interact with EGA and is highly relevant in gliomas EYA1 protein has putative
role in in net immunity DNA damage repair angiogenesis and cancer metastasis IJG1 or immunoglobulin heavy constant gamma
1 protein may play a role in immune system evasion mechanisms The polyglutamine binding
protein 1 or PQBP1 binds to be add into an POU3 class of neural transcription factors
which inhabits the transcription activation of BRN2 BRN2 is known to be expressed in gbms
and associated with development of neural and clear cells As you can see in these MDS plots that the
disease cohorts show a significant segregation from healthy and disease cohorts when this
specifier proteins are applied Apart from differentiation between grade and healthy
controls the authors have also used a same platform to identify the productive markers
for better prognosis It has been observed by the clinicians that
if glioblastoma multiforme tumors occur in .. zone of the brain known as SVZ positive
the prognosis of such of patients are poor as compared to the grade 4 tumor occurring
outside the SVZ region or SVZ negative Using similar data analysis strategies they
investigators here identified protein net nine a protein which is involved in cell migration
as one of the differentiating protein for SVZ positive with SVZ negative Mutation in a gene called isocitrate dehydrogenase
IDH is also associated with better prognosis in glioma patients So patients with IDH1 mutation
or IDH positive show better prognosis than those with IDH1 wild type gene Similar comparisons using protein array data
led to the identification of a protein YWHAH and STUB1 which were two of the twenty two
distrivet proteins investigators have identified which are known to be involved in proliferation
of glioma cells They were down regulated in IDH1 patients which allowed the authors to
correlate the pathological prognosis with their findings So today we have discussed several examples
how protein arrays can be used for verity of clinical research questions especially
pertaining to the biomarker identification Considering the vast diversity of protein
array types each array can be used for multiple set of applications In the next lecture we
will talk about some of the example as to how to understand the scope of this powerful
technology especially to understand protein protein interactions studies using protein
microarrays Thank you

Leave a Reply

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