Western Blotting: A Guide to Current Methods

Transcription

A Sponsored Supplement to Science
Western Blotting:
A Guide to Current
Methods
Sponsored by
Produced by the
Science/AAAS Custom
Publishing Office
JOIN AAAS
Get
Getinstant
instantaccess
access to
to Science.
Science. Support
Support all of the sciences.
Since
1880
TheAmerican
AmericanAssociation
Associationfor
forthe
theAdvancement
Advancementof
ofScience
Science(AAAS)
(AAAS)isisaanon-profit
non-profit community
community that
that is
is open
open to
to everyone,
everyone, from
The
from
Nobel
laureates
to
high
school
students.
Ours
is
a
global
membership
of
over
120,000
people
who
believe
in the
the power
Nobel laureates to high school students. Ours is a global membership of over 120,000 people who believe in
power of
of
sciencetotomake
makethe
theworld
worldaabetter
betterplace.
place.
science
Fromthe
themoment
momentyou
youjoin,
join,you
youget
getimmediate
immediateaccess
access to
to everything
everything that
that AAAS’s
AAAS’s award-winning
award-winning journal
journal Science
Science has
From
has to
to
offer,
including:
offer, including:
weeksofofhome
homedelivery
deliveryofofScience;
Science;
§§5151weeks
§
Instant
online
retrieval
of
every
Sciencearticle
articleever
everpublished,
published, from
from today,
today, dating
dating back
back to
to 1880;
1880;
§ Instant online retrieval of every Science
§
Anytime,
anywhere
access
via
the
Science
mobile
site
and
apps
for
Android,
iPad,
and
iPhone
devices;
§ Anytime, anywhere access via the Science mobile site and apps for Android, iPad, and iPhone devices;
Members-onlynewsletters;
newsletters;and
andmore.
more.
§§Members-only
Asaamember,
member,you
youare
arealso
alsomaking
makingaacritical
criticalcontribution
contribution to
to AAAS’s
AAAS’s efforts
efforts to
to provide
provide aa public
public voice
voice for
for all
all of
of science.
As
science.
With
public
skepticism
about
science
increasing,
and
public
funding
for
research
more
uncertain
than
ever,
this
With public skepticism about science increasing, and public funding for research more uncertain than ever, this work
work has
has
neverbeen
beenmore
moreimportant.
important.
never
AAASisishard
hardatatwork
workpromoting
promotingscience
scienceiningovernment
government offices,
offices, in
in schools,
schools, and
and in
in the
the public
public commons
commons all
all around
around the
AAAS
the
world—with programs like AAAS Senior Scientists and Engineers, which brings volunteer scientists into public school
world—with programs like AAAS Senior Scientists and Engineers, which brings volunteer scientists into public school
classrooms, or our sweeping petition drives calling for the preservation of federal R&D funding.
classrooms, or our sweeping petition drives calling for the preservation of federal R&D funding.
Visit promo.aaas.org/joinaaas and join today. Together we can make a difference.
Visit promo.aaas.org/joinaaas and join today. Together we can make a difference.
TA BLE O F C O N T E N T S
Western
Blotting:
A Guide
to Current
Methods
Introductions
2
Western blotting: Surveying the past to the present
3
A new standard in reproducible quantitative
Western blotting
Tianna Hicklin, Ph.D.
Science/AAAS
Kim Cushing
Product Marketing Manager
GE Healthcare
Western blotting overview
4
7
10
A history of the old West(ern blot)
Jeffrey M. Perkel
Preparing for your Western blot
Jeffrey M. Perkel
Choosing your detection method
Jeffrey M. Perkel
Research articles
12 An analysis of critical factors for quantitative immunoblotting
Kevin A. Janes 22 A neuronal antigen in the brains of Alzheimer patients
Benjamin L. Wolozin, Alex Pruchnicki, Dennis W. Dickson et al.
25 Serological analysis of a subgroup of human About the cover: Western blotting images
provided courtesy of GE Healthcare.
This booklet was produced by the Science/
AAAS Custom Publishing Office and
sponsored by GE Healthcare.
Editor: Tianna Hicklin, Ph.D.
Proofreader/Copyeditor: Yuse Lajiminmuhip
Designer: Amy Hardcastle
Jörg Schϋpbach, Mikulas Popovic, Raymond V. Gilden et al.
28
BILL MORAN, GLOBAL DIRECTOR
Custom Publishing
[email protected]
+1-202-326-6438
ROGER GONCALVES, SALES MANAGER
Custom Publishing
Europe, Middle East, and India
[email protected]
+41 43 243-1358
T-lymphotropic retroviruses (HTLV-III) associated
with AIDS
Cell surface P-glycoprotein associated with multidrug
resistance in mammalian cell lines
Norbert Kartner, John R. Riordan, Victor Ling
Technical notes
31
Reproducibility using the Amersham™ WB system
Åsa Hagner McWhirter, Anita Larsson, Elisabeth Wallby et al.
© 2015 by The American Association for the Advancement
of Science. All rights reserved. 9 June 2015
1
W E S T E RN BLOT T ING : A G U I D E TO CU R R E N T M E TH O D S
W
Western
blotting:
Surveying
the past to
the present
estern blotting—easily recognizable by the distinct
images of laddered bands that it generates—has
remained a relatively popular protein detection technique over the past 36 years.
From immunology to neuroscience to drug discovery, the fields to
which Western blots have contributed are wide ranging. Science and
its family of journals have published hundreds of research articles that
have used the technique since it was first described, as it has been
a useful tool in many scientific discoveries. Shortly after the method
was first described in 1979, for example, Western blots began being
used to unravel questions about which proteins were contributing to
drug resistance (see page 28) and Alzheimer’s disease (see page 22)
as well as to characterize the human retrovirus that was causing a new
epidemic, AIDS (see page 25).
Though it has remained a staple for protein detection over the
years, the technique is not immune from the growing concerns of
the scientific community for increased reproducibility of data. Such
concerns have been recently described in an editorial by Science
Journals Editor-in-Chief Marcia McNutt in which she discussed the
importance of reproducibility, and transparency, when publishing
data (www.sciencemag.org/content/346/6210/679.full) as well as the
efforts a group of editors put forth to design guidelines for reporting
data in the hopes that they will be viewed “as part of the quality control that justifies the public trust in science.”
For Western blots in particular, the optimization of various steps
and a great deal of experience working with the technique are key to
gathering quality data that is reliable and reproducible. As a multistep process, the technique inherently houses many places in which
variability can be introduced during data collection. In a 2015 Science Signaling article, Kevin A. Janes details some of the critical factors for achieving accurate immunoblotting data, including sample
preparation, loading controls, and choice of reagents and buffers
(see page 12).
In this booklet, we invite you to explore some of the history of
Western blotting as we take a look back at some of the highly cited
advances to which the method has contributed and explore some of
the many ways the technique can be optimized to help create higher
quality and more reproducible data.
Tianna Hicklin, Ph.D.
Editor, Science/AAAS Custom Publishing Office
2
sciencemag.org SCIENCE
SE C T I O N O N E | I N T RO D U C T IO N S
W
A new
standard in
reproducible
quantitative
Western
blotting
estern blotting, also known as immunoblotting when
first described in 1979, is today a ubiquitous method
in the life science laboratory. From cell biology to
protein purification and characterization, Western
blotting remains an essential protein analytical technique that is fundamental to protein research.
One of the biggest challenges for Western blotting is data
reproducibility. Within the multistep immunoblotting process, there
are many potential ways to introduce error and variation. This inherent
risk often necessitates that experiments are repeated many times in
order to generate reproducible data. Every step of the process has
an impact on the result and must be controlled to ensure the quality
of the final data. This is particularly critical when data publication is
the ultimate goal and reproducible quantitative data is an absolute
prerequisite.
In 1990, Amersham International (now part of GE Healthcare) was
the first company to introduce enhanced chemiluminescent (ECL)
Western blotting detection reagents. Chemiluminescence is a highly
sensitive detection methodology and was the leading detection
technology used in Western blotting for many years; however, as the
critical horseradish peroxidase enzyme reaction is dynamic, the light
produced declines over time making it challenging to produce truly
quantitative data.
Advances in charge-coupled device (CCD) technology have
led to the use of digital imaging systems that can capture both
chemiluminescent and fluorescent signals. When using fluorescence
detection, the signal generated by dye excitation is stable for several
months, producing consistent and quantitative results, making it an
excellent choice for true protein quantitation.
Since the introduction of the first ECL detection reagent for
Western blotting, the portfolio of products offered by GE Healthcare
has been expanded, improved, and optimized to cover the breadth of
Western blotting requirements from sensitive protein identification to
reproducible protein quantification. The most recent addition to the
range is the new Amersham WB system, which integrates separation,
transfer, detection, and quantitative analysis of proteins. This fully
integrated system minimizes assay variability to provide consistent,
quantitative Western blot data for every sample, every time.
Western blotting continues to be an essential technique which,
while undergoing significant improvements remains largely faithful to
the tenet of the original protocol. There are opportunities to optimize
and refine the Western blotting protocol that can deliver some real
benefits
In addition to this publication we were delighted to sponsor two
webinars, which highlight some of the key considerations for optimal
Western blotting:
Quantitative Western blotting:
Improving your data quality and reproducibility
Western blot tips and tricks:
Filling the gap between art and science
Kim Cushing
Product Marketing Manager
GE Healthcare
SCIENCE sciencemag.org
3
A history of the old
West(ern blot)
By Jeffrey M. Perkel
W
hen Harry Towbin and colleagues at the Friedrich Miescher Institute in Basel, Switzerland first
detailed the procedure that would come to be
called the Western blot, almost nothing was easy.
Antibodies weren’t readily available in 1979, so they made their
own by injecting purified ribosomal proteins into mice or goats.
Their transfer procedure involved a gel and nitrocellulose membrane sandwiched between two Scotch-Brite scouring pads
and a “disposable micropipette tray” for structural support, and
bound together with rubber bands. The transfer itself was conducted in “an electrophoretic destaining chamber” (1).
Secondary antibodies were a little easier, at least for fluorescent and colorimetric detection since these could be purchased.
But for radioisotopic detection, the researchers once again were
on their own, labeling their own antibody preparations with
iodine-125 via “the chloramine T method”—and exposing the labeled blots to film for 6 days.
Today, the method is considerably easier: researchers can
purchase precast gels, transfer proteins to membranes using
commercial transfer apparatuses, and capture their data in
seconds using digital imagers. Towbin’s landmark paper has
been cited more than 51,000 times in the intervening 36 years,
according to Google Scholar. Yet for the most part, the method
remains largely the same. Modern Western blot sandwich
cassettes bear a strong family resemblance to the one Towbin
dreamed up. And researchers still use fluorophore and enzymeconjugated secondary antibodies to identify bands of interest,
though most researchers now favor chemiluminescence over
colorimetric detection.
That’s not to say methods developers have been content to
rest on their laurels. Researchers have extended and evolved
the method over the decades, making it easier, faster, and more
reliable.
Here, we travel back to the Old Western, to see where the venerable immunoblot has been and explore how far the technique
has come.
The Western blot
A Western blot is simply a way to identify proteins on a polyacrylamide gel. Proteins, separated by size on a polyacrylamide
4
gel, are transferred to a membrane such that their pattern in the
gel is retained, like making a photocopy, or what Towbin called
“a faithful replica” (1). That “blot” is then probed using “primary”
antibodies to a specific protein or proteins of interest, and developed by addition of secondary antibodies, which recognize
the primaries. These secondary antibodies are labeled with
either a radioisotope, fluorophore, or enzyme, all of which allow
the proteins’ positions to be identified.
The method is largely unchanged—at least in broad strokes—
since Towbin, with Theophil Staehelin and Julian Gordon, used
it to detect bacterial or chicken ribosomal proteins that had
been separated on urea-containing polyacrylamide gels. They
detailed their method in the Proceedings of the National Academy of Sciences in 1979, proposing such applications as screening for monoclonal antibody-producing hybridomas, autoimmune sera, enzymatic activities, and ligand binding (1).
Yet Towbin’s original article never actually names the method; it simply refers to an “electrophoretic blotting technique.”
The term “Western blot” was actually coined 2 years later by
W. Neal Burnette, of the Fred Hutchinson Cancer Research
Center in Seattle, Washington, who described several technical
improvements to the method, including applying it to the more
commonly used SDS-PAGE gels:
“With due respect to [Edwin] Southern, the established
tradition of ‘geographic’ naming of transfer techniques
(‘Southern,’ ‘Northern’) is continued; the method
described in this manuscript is referred to as ‘Western’
blotting” (2).
Unraveling multidrug resistance
One of the earliest applications of Towbin’s procedure used
the method to tease apart the mechanism of multidrug resistance in immortalized cells.
Researchers had observed that while many cultured cells
were sensitive to drugs like colchicine and actinomycin D, some
were resistant, and if they were resistant to one compound, they
often were resistant to many. Molecular analysis determined
that a 170 kD membrane protein seemed to correlate with multidrug resistance, but it was not yet clear if that protein was actually causative of the phenotype, or merely a passenger.
In 1982, Victor Ling of the Ontario Cancer Institute and
University of Toronto, and colleagues demonstrated that DNA
transfer from drug-resistant cells to sensitive cells induced
a resistant phenotype, including the appearance of that 170
kD protein on polyacrylamide gel separations of membrane
protein preparations. But the band was diffuse and partially
obscured by the complex membrane material, so, “To increase
the sensitivity and specificity of detecting the P-glycoprotein,”
they wrote, “we employed an antiserum raised against CHO
[Chinese hamster ovary] cell mutant membranes in a replica
Western blot procedure” (3).
PHOTO: © POLAKPHOTO/SHUTTERSTOCK.COM
SE C T I O N T W O | WESTERN BLOTTING OVERVIEW
Ling’s team didn’t
use that antiserum
directly, however;
they cleaned it up first
by preabsorbing it
against a preparation
from drug-sensitive
cells. In this way,
they were able to
clearly detect a band
consistent with the
P-glycoprotein in drugresistant but not sensitive lines.
But those results,
the researchers acknowledged, were still
only correlative—they
had no way of knowing, for instance, whether the P-glycoprotein gene was transferred with other sequences that actually conferred the drug
resistance itself. Much of that doubt, though, was eliminated
when the researchers demonstrated in 1983, again via Western
blotting, that drug resistant cells from Chinese hamster, Syrian
hamster, mouse, and human all contained the 170 kD P-glycoprotein, all of which could be detected using antibodies raised
against membrane preparations from both drug-resistant CHO
and human cell lines (4; see page 28).
These data, of course, suggested that whatever the
P-glycoprotein is, it is evolutionarily conserved. But the
implications were broader than that. “Our findings could
have important implications for cancer therapy,” Ling wrote in
Science. “It is possible that clinical resistance to combination
chemotherapy might result from an unchecked proliferation
of tumor cell subpopulations with a multidrug resistance
phenotype” (4).
Today, that prediction has been shown to be true. The
170 kD P-glycoprotein is an ATP-binding cassette (ABC)
transporter that actively pumps drugs and other compounds
out of cells, conferring drug resistance to cancer cells—a
clinically important observation that was made possible via the
Western blot.
Decoding AIDS
Another landmark in Western history occurred in 1984. At
the time, acquired immunodeficiency syndrome (AIDS) was
a disease of unknown etiology that was impacting mostly
intravenous drug users and homosexual men. The disease had
only been recognized in 1981, but within 2 years, researchers
had determined that the causative agent likely was a retrovirus,
a relative of the human T-lymphotropic viruses (HTLV). The
question was: which one?
To find out, Robert
Gallo and his team at
the National Cancer
Institute in Bethesda,
Maryland, performed
a battery of tests,
including electron
microscopy, reverse
transcriptase enzyme
assays, cell co-culture
experiments—and
Western blots. In one
series of experiments,
the team infected
human T cells with
the virus that they
called HTLV-III, then
separated cellular
proteins (including the
virus) on protein gels and blotted them to nitrocellulose. They
then cut those blots into strips and “tested [them] with samples
of human serum in a strip radioimmunoassay (RIA) based on the
Western blot technique” (5; see page 25). Those sera samples
had been taken “from patients with AIDS or pre-AIDS, from
contacts of such patients, and from homo- or heterosexual male
controls.”
The goal was to determine whether blood samples collected
from patients with AIDS contained antibodies capable of
recognizing HTLV-III proteins, and as it turns out, they did.
Sera from patients with AIDS or pre-AIDS lit up specific bands
in virus-infected cellular lysates, but not uninfected control
cell lines. Control patient sera did not bind HTLV-III proteins.
The analysis also demonstrated that HTLV-III is distinct from,
though related to, other HTLV family members, and began
the painstaking process of teasing apart the virus’ molecular
composition (5).
In total, the Gallo team published four studies in the 4 May
1984 issue of Science. In the final one, the team determined that
blood sera from 43 of 49 AIDS patients and 11 of 14 pre-AIDS
patients recognize antigens—particularly the 60 kDa env protein—
found in a viral extract from HTLV-III-producing cells, in this case
using ELISA and RIA based on the Western protocol (6).
“The data presented here and in the accompanying reports
suggest that HTLV-III is the primary cause of AIDS,” Gallo and
his team concluded (6). Within a few years, HTLV-III would be
rebranded with the name by which it is known today: human immunodeficiency virus (HIV).
That discovery, wrote Jean Marx in a news story accompanying the reports, opened the door to the development of reagents
for disease diagnostics and blood donor testing, as well as for research into potential vaccines (7). And Western blotting played
a key role. continued>
5
W E S T E RN BLOT T ING : A G U I D E TO CU
M ORDREERNNTM
MEETH
THOODDSS
Untangling Alzheimer’s disease
In 1986, researchers used Western blotting to address yet
another clinical puzzle, Alzheimer’s disease.
In an effort to begin to understand the biochemistry of disease, Benjamin Wolozin and colleagues at the Albert Einstein
College of Medicine used brain extracts from control patients
and patients with Alzheimer’s to identify an antibody that could
distinguish diseased and non-diseased brain homogenates.
The resulting reagent, called “Alz-50” bound an antigen that
was some 15 to 30 times more abundant in the temporal cortex,
hippocampus, and nucleus basalis of patients with Alzheimer’s
disease than non-disease individuals (8; see page 22). Alz-50
stained neurons in the brains of patients with Alzheimer’s but
not control individuals, and also lit up the neurofibrillary tangles
that characterize the disease.
Researchers have extended and evolved
the method over the decades, making it
easier, faster, and more reliable.
The question was, to what antigen was Alz-50 binding. Wolozin
and colleagues enriched the antigen from Alzheimer and control
brain homogenates, resolved them on an SDS-polyacryl amide gel, transferred to nitrocellulose, and probed the blot with
Alz-50. Detection, using an alkaline phosphatase-conjugated
secondary antibody, revealed a 68 kD protein in diseased brain
homogenates that was absent in normal tissue (8).
“Our results demonstrate that Alz-50 recognizes a protein
present in neuronal terminals in plaques and in most neurons
with tangles,” Wolozin et al. concluded. But they had no idea
what that protein might be, though they doubted it could
be tau protein, one of the few proteins known to be of the
right size, as it did not exhibit the differential abundance they
observed. Today, researchers know that presumption was incorrect: Alz-50 binds a phosphorylated form of tau, a major
component of neurofibrillary tangles in so-called tauopathies,
including Alzheimer’s.
Technical innovations
Western blotting has since participated in studies across
the biological spectrum, from identifying the substrates of the
aging-associated Sir2 deacetylase and probing the biology of
synthetic prions to modeling neurotransmitter synaptic vesicle
trafficking (10–12).
In the latter study, Silvio Rizzoli of the University of Göttingen
Medical Center, and colleagues, used “quantitative immunoblotting” to measure the abundance of each of 62 proteins in the
so-called synaptosome, a structure involved in synaptic vesicle
6
recycling. They then combined those figures with electron microscopy, mass spectrometry, and superresolution imaging data
to generate “a three-dimensional model of an ‘average’ synapse,
displaying 300,000 proteins in atomic detail” (12).
As its applications have evolved, so too has the Western blotting method itself. Shortly after the original Towbin paper, for
instance, researchers developed the so-called Southwestern
blot (13). Literally a cross between Southern and Western blotting, the technique probes protein blots with labeled DNA to
identify DNA-binding proteins.
Another variation is the “In-Cell Western,” combining the
features of Western blotting and ELISA. In a traditional ELISA,
an antibody to the target of interest is bound to a solid surface,
mixed with a sample, and then probed with a second, labeled
antibody. In an In-Cell Western, cells grown on the surfaces of
microtiter plates are lysed in situ, probed with primary antibodies to the antigen(s) of interest, and finally detected using fluorescently labeled secondary antibodies. The method avoids the
need for gel electrophoresis and blotting, and thus is faster than
a traditional Western. But it also assumes no background binding, as there is no electrophoretic separation step.
Methods developers have also found ways to accelerate and
optimize the Western procedure. As detailed in a Science magazine technology feature, for instance, EMD Millipore’s SNAP i.d.
2.0 system uses a vacuum manifold to reduce Western blotting
incubation and washing steps “from 4 hours to 30 minutes” (9).
Others have developed methods to reduce blotting time and
increase throughput, such as ProteinSimple’s Simple Western
system. Simple Westerns replace the traditional SDS-PAGE
gel and blotting steps with capillary electrophoresis, which
can separate proteins by size or charge. Up to 96 samples can
automatically be run in parallel and detected in the capillaries
directly via chemiluminescent detection. The resulting data are
then rendered as a traditional blot using the system software.
Another Western blotting optimization that has emerged
since Towbin is multiplexing. Researchers using fluorophoreconjugated secondary antibodies now can detect two or more
antigens simultaneously, which is particularly useful for quantitation and normalization of protein abundance (see Preparing for your Western blot, page 7). In one typical example, the
antigen of interest is detected using one fluorescent color and a
housekeeping protein with a second. Because both proteins are
detected on a single gel and simultaneously, the method saves
time. But it also improves reproducibility relative to the alternatives, which are either to probe, strip, and reprobe the blot with
a second primary antibody, or to run duplicate gels and probe
them with separate antibodies.
Perhaps the most significant technical advances in Western
blotting history, however, have been the most pedestrian. For
one thing, unless they are studying a new protein, researchers
rarely need, as Towbin did, to prepare their own antibodies;
with some 1.8 million antibodies listed in Antibodypedia, it’s a
SE C T I O N T W O | WESTERN
SE C T I OBLOTTING
N O N E | EDITORIAL
OVERVIEW
good bet somebody else has already done so. Meanwhile, using digital gel-documentation systems researchers can ditch
their manual cameras and lightrooms and capture digital data
instead. Such systems certainly are easier to use than X-ray film,
the traditional method of chemiluminescent detection, but they
also represent a tremendous advance in that they are more
sensitive, produce a linear signal over a wider dynamic range,
and enable more reproducible experiments. Equally significant
is the development of methods to make what effectively is a
qualitative technique, quantitative—a development that powered the synaptosome modeling study, for instance (12, 14; see
page 12).
The remarkable thing is, for all these advances, the original
Towbin protocol remains apparent. Countless labs around the
world still use plastic Western blotting cassettes and sponges
that look remarkably like the sketch in the seminal 1979 paper.
The process is more efficient and straightforward than ever,
yet the fundamental principle of creating and probing a
“faithful replica” of a protein gel remains. As the Western blot
closes in on four decades, it will be interesting to see where it
goes next.
Jeffrey M. Perkel is a freelance science writer based in Pocatello,
Idaho.
References
1. H. Towbin et al., Proc. Natl. Acad. Sci. U.S.A. 76, 4350 (1979).
2. W. N. Burnette, Anal Biochem. 112, 195 (1981).
3. P. G. Debenham et al., Mol. Cell Biol. 2, 881 (1982).
4. N. Kartner et al., Science 221, 1285 (1983).
5. J. Schüpbach et al., Science 224, 503 (1984).
6. M. G. Sarngadharan et al., Science 224, 506 (1984).
7. J. L. Marx, Science 224, 475 (1984).
8. B. L. Wolozin et al., Science 232, 648 (1986).
9. A. Harding, Science, (2013), doi:10.1126/science.opms. p1300073.
10. V. J. Starai et al., Science 298, 2390 (2002).
11. G. Legname et al., Science 305, 673 (2004).
12. B. G. Wilhelm et al., Science 344, 1023 (2014).
13. F. K. Y. Siu et al., Nat. Protocols 3, 51 (2008).
14. K. A. Janes, Science Signaling 8, rs2 (2015).
Preparing for your
Western blot
W
estern blotting is ubiquitous in modern molecular biology laboratories. It’s relatively simple to
perform, but good-looking data—and especially
quantitative and reproducible data—can be hard to
come by. A complete Western blotting protocol comprises dozens of steps, dozens of places in which something can go wrong,
from sample preparation all the way through to detection. Making a protocol that is reproducible from day to day and lab to lab
such that researchers can be confident of anything but the most
dramatic differences in abundance can be a challenge. Here we
review some key steps to consider.
Sample preparation
The first step in a Western blotting protocol is preparing a
cell or tissue lysate. The cell membranes must be broken open
to release the cellular contents, using a lysis or extraction buffer,
and the protein(s) of interest solubilized so they can be separated with gel electrophoresis. But there is no one best way to do
that. Some samples, for instance, require harsher lysis conditions
than others, and not all proteins are soluble or stable in the same
detergent.
Lysis buffers “differ in their ability to solubilize proteins, with
those containing sodium dodecyl sulfate [SDS] and other ionic
detergents considered to be the harshest and therefore most
likely to give the highest yield,” according to a guide published
by antibody-developer Abcam (1). Some common nonionic
detergents include Triton X-100, Nonidet P-40, and Tween; ionic
detergents include SDS, deoxycholate, and hexadecyltrimethylammonium bromide (CTAB).
In a recent report in Science Signaling, Kevin Janes of the University of Virginia documented several “critical factors for quantitative immunoblotting,” including sample preparation (2; see
page 12). He noted, for instance, that the commonly used RIPA
buffer (50 mM Tris-HCl pH 7.5, 150 mM NaCl, 1% Triton X-100,
0.5% sodium deoxycholate, 0.1% SDS, 5 mM EDTA, plus proteinase and phosphatase inhibitors)—efficiently solubilizes cellular
proteins, though some are left behind in an insoluble pellet.
In Janes’ own hands, analysis of the soluble and insoluble
fractions resulting from RIPA lysis of HT-29 human colon
adenocarcinoma cells revealed that though continued>
7
“many cytoplasmic proteins” were efficiently solubilized in
RIPA, some histones and transcription factors were confined
to the insoluble fraction. Cleaved caspase 8, an apoptotic
indicator, was soluble in Laemmli sample buffer (62.5 mM
Tris-HCl pH 6.8, 2% SDS, 10% glycerol, 100 mM dithiothreitol,
0.01% bromophenol blue), but not in RIPA or NP-40 buffer (50
mM Tris-HCl pH 8.0, 150 mM NaCl, 0.5% NP-40, 5 mM EDTA,
plus inhibitors). Other options include Tris-Triton buffer for
cytoskeletal proteins (10 mM Tris pH 7.4, 100 mM NaCl, 1 mM
EDTA, 1 mM EGTA, 1% Triton X-100, 10% glycerol, 0.1% SDS,
0.5% deoxycholate) and 20 mM Tris-HCl pH 7.5 for soluble
cytoplasmic proteins (1).
Researchers can cast their own gels
for separating proteins, but they
also can buy them precast. That offers
a speed advantage, but it’s also
safer and more reproducible.
In a separate experiment, Janes illustrated the importance of
phosphatase inhibitors during lysis in probing various members
of the “Akt-glycogen synthase kinase-3 (GSK3)-glycogen
synthase (GS) signaling axis.” In particular, he showed that
some post-translational modifications were more stable in the
absence of inhibitors in NP-40 lysis buffer than RIPA.
“These results collectively showed that lysis buffer
composition substantially affects the results of quantitative
immunoblotting,” Janes wrote (2).
Normalization
Western blotting typically is used to measure changes
in protein abundance across different conditions. For such
comparisons to be meaningful, each lane should contain the
same amount of protein, ideally representing equivalent cell
numbers.
First, researchers need to know how much protein they have
in each sample. The Abcam guide recommends quantifying
protein concentration using “a Bradford assay, a Lowry assay
or a BCA [bicinchoninic acid] assay,” usually with bovine serum
albumin (BSA) as a protein standard (1).
Then, because minor pipetting errors during loading the
gel or errors when measuring protein concentration can
dramatically over- or under-emphasize biological effects,
researchers should also take steps to ensure an equal amount
of protein is present in each lane.
In one common approach to minimize the effect of loading
variation, researchers compare the intensity of their protein of
8
interest to a control, such as a housekeeping protein.
But that assumes the housekeeping protein’s abundance
is constant across conditions—an assumption that must be
tested. Also, researchers must test to ensure that both the
housekeeping protein and protein of interest produce linear,
proportional responses, such that a twofold change in protein
abundance yields a twofold change in signal.
In one recent example, Alicia McDonough and colleagues
at the Keck School of Medicine at the University of Southern
California, loaded a gel with kidney homogenate samples.
Some lanes contained 3 mg of the lysate; others contained
twice as much. The team then probed with antibodies for
claudin-10 and actin. The claudin signal was twice as intense
in the high-abundance lanes, indicating that the protein was in
a linear detection range. But actin, a common control protein,
produced an equivalent signal in both cases, indicating that it
“is not an acceptable loading control for this application” (3).
One possible solution is normalizing against multiple
controls. For instance, Janes showed in his study that the
apparent abundance of phosphorylated Smad2 protein
changes dramatically depending on the choice and number
of normalization controls he used (he tested as many as five at
once). “As higher-order combinations of loading controls were
tested as normalizers, I found that the coefficient of variation
of p-Smad2 linker steadily improved toward 7% to 8%,” he
wrote (2).
An alternative approach normalizes each band’s intensity
to total protein in the lane using either a general protein stain,
such as Ponceau S, Memcode, or fluorescent dye. McDonough
recommends running a parallel “loading gel, which is loaded
and run identically to the gel(s) that will be immunoblotted” (3).
The gel is stained and imaged, and the intensities of several
discrete bands averaged for normalization. Alternatively,
the samples can be prelabeled prior to electrophoresis
and the total signal on the membrane after transfer can
be used as loading control. This is especially suitable for
fluorescent Western, since the target signals can be detected
simultaneously on the membrane by multiplexing.
Electrophoresis and blotting
Researchers can cast their own gels for separating proteins,
but they also can buy them precast. That offers a speed
advantage, but it’s also safer—unpolymerized polyacrylamide is
neurotoxic—and more reproducible. Precast gradient gels can
be especially useful, providing particularly sharp resolution over
specific size ranges; such gels are challenging to create from
scratch in a reproducible fashion. Precast gels also are available
for analysis of nondenatured (i.e., “native”) proteins. These are
useful especially when studying intact protein complexes or
using antibodies that recognize conformational epitopes.
Western blots typically use either nitrocellulose or
polyvinylidene fluoride (PVDF) membranes for protein blotting.
PHOTO: © PY PHOTO/SHUTTERSTOCK.COM
The latter tends
to be sturdier and
have a higher
protein-binding
capacity, but may
also yield higher
auto fluorescence
depending on
membrane type.
On the other hand,
nitrocellulose is
more compatible
with most protein
stains that enable
visual inspection of
transfer efficiency.
Once your
proteins are
transferred to a
membrane, blocking
reagents are used to
reduce antibodies
nonspecific binding
on the membrane. Common options include nonfat milk and
bovine serum albumin, but several commercial formulations exist
too. In many cases, any blocker will do, but it can be antibody
dependent. Milk is suboptimal for phosphorylation analysis,
for instance, because milk contains casein, a phosphoprotein
that can compete with the target for antibody binding and
reduce specific signal intensity. Milk also may contain biotin and
glycoproteins that will light up with streptavidin-based detection
schemes (4, 5). When in doubt, try several options to see which
works best for your system.
(TBST or PBST) (6),
but the number,
duration, and
stringency of washes
can vary, as can the
amount of detergent
used. Antibody
dilution also is key—
you need sufficient
antibody to detect
the target, but not
so much as to create
high background on
the blot, which can
obscure faint bands.
The secondary
antibody is likewise
critical, so be sure to
optimize your dilution and incubation
conditions there, as
well.
Like all methods,
Western blotting has both strengths and weaknesses. The
technique is easy enough that almost anyone can perform it.
The question is: how well does that signal reflect the cells it
represents? With careful experimental design and rigorous
control procedures, researchers can be sure it’s as close as
possible.
Antibodies
1. Abcam, “Sample preparation (WB guide),” [http://www.
abcam.com/index.html?pageconfig=resource&rid=11379
#A1].
2. K. A. Janes, Science Signaling 8:rs2 (2015).
3. A. A. McDonough et al., Am. J. Physiol. Cell Physiol. 308 C426 (2015).
4. LI-COR Biosciences, “Optimizing chemiluminescent Western blots,” 2011. [biosupport.licor.com/docs/Chemi_
Good_Westerns.pdf].
5. GE Healthcare Life Sciences, “Western Blotting Principles and Methods,” 2011. [www.gelifesciences.com/file_source/
GELS/Service%20and%20Support/Documents%20and%20
Downloads/Handbooks/pdfs/Western%20Blotting.pdf].
6. Cell Signaling Technology, “Western blotting protocol,” [http://
www.cellsignal.com/common/content/content.jsp?id=western].
The key reagent in a Western blot, of course, is the primary
antibody used to recognize the protein of interest. Thousands
of antibodies are commercially available, but not all are
equally good. Look for antibodies that have been validated as
compatible with Western blotting—typically, these recognize
linear peptides as opposed to conformational epitopes, as
most Westerns are run under denaturing conditions. (Native
Westerns, of course, retain protein conformation). You may need
to test several primaries to see which works best, but be sure
also to demonstrate that the antibody actually is specific for
your antigen in the first place—that is, that is produces a band of
the correct size in positive control samples and not in negative
controls (3).
Then, optimize how much and how long blots are incubated
with an antibody and the washing conditions used. Typical
wash conditions are three washes of 5 minutes each in Tris- or
phosphate-buffered saline (TBS or PBS) containing Tween 20
Idaho.
References
9
Choosing your
detection method
I
n their seminal Western blotting paper, Towbin et al. developed their blots using radioisotopic labeling, colorimetric
detection, or fluorescence (1). Thirty-six years later, the
method has changed remarkably little. But one thing that has
changed is how the experiment is read out: Modern researchers
most commonly develop their Western blots using chemiluminescence detection.
Fluorescence-based detection—which in Towbin’s study was
the least sensitive method tried—has also evolved into an increasingly sensitive and popular alternative.
Both chemiluminescent and fluorescent detection offer substantial advantages over the previously preferred options of colorimetry and radiolabeling, being more sensitive than the former
and less hazardous and troublesome than the latter. But they also
differ from each other in several important ways.
Chemiluminescence vs. fluorescence
In chemiluminescence, the secondary antibody—the antibody
that allows researchers to detect the antibody binding to the protein of interest—is conjugated to an enzyme [usually horseradish
peroxidase (HRP)], just as in Towbin’s original study (1). But where
Towbin and colleagues detected their HRP-tagged antibodies via
the deposition of a colored precipitate from a colorless substrate,
chemiluminescence uses hydrogen peroxide to oxidize luminol,
producing a transient burst of photons that must be detected using autoradiography or a digital imager.
Fluorescence detection uses secondary antibodies labeled
with fluorescent dyes, which also are detected on a digital imager. An alternative approach, chemifluorescence couples HRP
activity to formation of a fluorescent product.
Chemiluminescent Westerns generally are considered more
sensitive than their fluorescent counterparts, with detection down
to the low femtogram level in some cases. But as a practical matter, fluorescence Western blots are simpler to perform, as there’s
no chemiluminescent substrate to add (which saves time and
money) and the signal is generally stable over time (simplifying
data collection and enhancing reproducibility).
All things being equal, an exposure taken of a fluorescently
labeled blot today will be exactly the same as one taken a week
10
from now, assuming the blot is properly handled and stored.
Chemiluminescent signals, in contrast, are transient and variable.
Researchers performing fluorescence-based Westerns also
have the ability to combine two or more secondary antibodies
to detect multiple proteins at once, a process called multiplexing. Multiplexing—which requires only that the different primary
antibodies be raised in different organisms and that the corresponding secondaries be conjugated to spectrally distinct dyes—
reduces an experiment’s hands-on time and improves quantitative accuracy and reproducibility relative to chemiluminescence,
which is strictly a singleplex technique.
Quantitative vs. qualitative detection
By default, Western blotting is inherently qualitative—a protein
is either present, or it is not. But researchers frequently use the
technique to assess quantitative differences, such as how protein
abundance changes under different conditions.
Fundamentally, converting a band’s intensity into a reliable
and meaningful assessment of abundance requires that the overall amount of the protein does not vary from lane to lane, that the
signal any band generated be linearly related to that protein’s
abundance, and that the detector’s ability to capture that signal
also is linear. Film, for instance, is a poor choice for quantitative
Western blotting, writes Kevin Janes of the University of Virginia,
because its dynamic range “is so small that quantitative analysis
is virtually impossible. Film can make small differences in abundance appear as large differences in band intensity. When saturated, film exposures can also hide sample-to-sample variation
for high-abundance proteins such as loading controls” (2; see
page 12).
One way to ensure band intensity accurately reflects protein
abundance, and to account for sample-to-sample variation, is
multiplexing. Rather than assessing protein abundance from
the intensity of a single band in isolation, researchers typically
compare it to one or more invariant normalization controls, which
usually are housekeeping proteins (see Preparing for your Western blot, page 7).
In a fluorescent Western, this is easily accomplished. A researcher could use a goat anti-rabbit secondary coupled to Cy3
to detect their protein of interest, and a rabbit anti-mouse secondary conjugated to Cy5 (i.e., a dye with distinct excitation and
emission spectra) to detect, say, tubulin. Using the fluorescence
intensity of those two signals, they then can determine the abundance of their protein in their experimental condition relative to
the control—a ratio that is more reliably comparable from experiment to experiment than the single protein’s absolute intensity.
The problem is, chemiluminescence doesn’t offer multiple
detection channels. The only way to quantify multiple proteins in
a chemiluminescent Western is to strip off the primary antibody
(using, for instance, 6M guanidine-HCl), and reprobe the blot with
another primary antibody against the second protein of interest—
a process that adds time and impacts reproducibility, since
PHOTO: © EXTENDER_01/SHUTTERSTOCK.COM
stripping can also remove
target proteins. Alternatively,
researchers can run and probe
replicate blots in parallel, but
this, too, is problematic, as the
blots may not be treated identically.
Complicating quantification
of chemiluminescent Westerns
is the detection mechanism
itself. Chemiluminescence
stems from an enzymatic reaction that consumes a finite
supply of substrate. Thus, the
signal intensity of a chemiluminescent blot will vary with
time and substrate concentration. A 30 second exposure taken
immediately after substrate addition will look very different from
an exposure of the same duration collected 10 minutes later. It
may also vary if the substrate reagent is a little older, or doesn’t
cover the blot uniformly. As a result, it can be difficult to acquire
an “ideal” exposure, as it isn’t necessarily obvious how different
exposures will appear.
The situation can be particularly tricky in cases involving multiple bands of dramatically different intensities. At the very least,
researchers may end up wasting considerable time fine-tuning
their exposures to capture both bands in the detector’s linear
range, such that their intensities accurately reflect their abundance. Experimental timing becomes key in such situations, and
replicating such conditions from day to day requires technical
skill. Even then, it isn’t easy to compare one “ideal” image with
another captured from a different blot, on a different day, so researchers often are advised against direct comparisons between
experiments.
In contrast, fluorescence signals are largely quantitative, so a
band with twice the intensity of another band can generally be
assumed to contain twice as much material.
In his recent study on the factors influencing the reliability
of quantitative Western blotting, Janes wrote that his lab favors
fluorescence over chemiluminescence Westerns “whenever possible” (2).
“I compared the linearity of [fluorescent dye]-conjugated
secondary antibodies to horseradish peroxidase conjugates that
were incubated with an enhanced chemiluminescence (ECL)
cocktail or a commercial substrate marketed for high-sensitivity
applications,” Janes wrote. “Under the same immunoblotting conditions, chemiluminescent exposures consistently yielded stronger band densities. However, the linear dynamic range was very
limited, and signals often decreased at high protein inputs” (2).
That’s not to say quantitative chemiluminescence is impossible. In one 2013 report, Sanjai Kumar of the U.S. Food and
Drug Administration’s Laboratory of Emerging Pathogens,
and colleagues described
a quantitative assay for the
detection of Plasmodium
falciparum circumsporozoite
protein (PfCSP), a potential
vaccine antigen.
The assay, the authors report, was linear between about
3 and 12 pg of protein, making
it “the most sensitive assay for
immunoassay-based detection of PfCSP or any malarial
protein.” (ELISA assays, by comparison, required at least 100
ng of protein for detection.)
Furthermore, the authors found
the assay to be highly reproducible, with an inter-assay coefficient
of variation (CV) of 10.31% and a mean intra-assay CV of 3.16%
(3). But to achieve that level of reproducibility, the authors had to
go beyond the typical Western protocol, using serial dilutions of
protein preparations, standard curves, and multiple replicates.
Equipment and reagents
Other differences between chemiluminescence- and fluorescence-based Westerns include the equipment and reagents
required to run the experiments. Nitrocellulose membranes, for
instance, are compatible with both approaches, but are relatively
fragile. Polyvinylidene fluoride (PVDF) membranes are sturdier,
but exhibit lower sensitivity and higher autofluorescence. Dedicated “low-fluorescence” membranes are commercially available
to address these shortcomings.
Chemiluminescence can be detected in most standard gel
documentation systems, and even on X-ray film (though that is
not advised, given film’s poor sensitivity and narrow dynamic
range). Fluorescence, however, requires a dedicated instrument,
such as a laser scanner or CCD camera equipped with suitable
light sources and emission filters.
Given all these variables, which option should you choose?
In many cases, it really doesn’t matter. Researchers interested in
a qualitative assessment of their samples can often use either
chemiluminescence or fluorescence detection, depending on
the antibodies and equipment they have on hand. But for a deep
dive into protein quantitation, they may want to give fluorescence
a try.
Idaho.
References
1. H. Towbin et al., Proc. Natl. Acad. Sci. U.S.A. 76, 4350 (1979).
2. K. A. Janes, Science Signaling 8, rs2 (2015).
3. S. Kumar et al., J. Immunol. Methods 390, 99 (2013).
11
An analysis of critical factors for
quantitative immunoblotting
Kevin A. Janes
Immunoblotting (also known as Western blotting) combined with digital image analysis
can be a reliable method for analyzing the abundance of proteins and protein modifications, but not every immunoblot-analysis combination produces an accurate result. I
illustrate how sample preparation, protocol implementation, detection scheme, and
normalization approach profoundly affect the quantitative performance of immunoblotting. This study implemented diagnostic experiments that assess an immunoblot-analysis
workflow for accuracy and precision. The results showed that ignoring such diagnostics
can lead to pseudoquantitative immunoblot data that markedly overestimate or underestimate true differences in protein abundance.
INTRODUCTION
mong the most indispensible tools in cell
signaling research is the immunoblot.
The premise of immunoblotting is simple, but execution is tricky, and there are
many variations in the method that can
affect the outcome (1). Add quantitation to the
end of an immunoblot and the complexity of
implementations increases even further. Surprisingly, there are few objective studies on
quantitative immunoblotting in the primary
literature (2, 3). Lacking a systematic assessment of key factors, researchers are prone to
repeat or reinforce mistakes that others have
made before them.
Here, I analyze how various methodological
choices affect the ability to perform quantitative immunoblotting accurately and precisely.
The analysis revealed how seemingly minor
variations affect immunoblot linearity and
reproducibility, yielding pseudoquantitative
numbers that are not directly proportional
to the input material. After background subtraction, quantitative immunoblots should
strive for zero-intercept linearity: y = bx,
where y is the quantified band intensity, x is
the abundance of the protein or modification
state in the sample, and b is a proportionality coefficient. The value of b is flexible, but
lines with nonzero intercepts indicate errors
in background subtraction, and nonlinear relationships suggest problems with detection
sensitivity or saturation. Either scenario will
yield fold change estimates that are skewed
relative to the true differences among samples.
Throughout this work, I systematically
altered several experimental parameters
that are often neglected or overlooked when
immunoblotting. Many other parameters
were kept fixed: All gels were run as 15well, 1.5-mm-thick, tris-glycine minigels on
a Bio-Rad Protean III platform; all wet electrophoretic transfers were done onto lowautofluorescence, 0.45-μm polyvinylidene
A
Department of Biomedical Engineering, University of Virginia,
Charlottesville, VA 22908, USA
E-mail: [email protected]
12
difluoride (PVDF) under modified conditions of Towbin et al. (4) (25 mM tris, 192
mM glycine, 0.0375% SDS, and 10% methanol
unless otherwise indicated); detection was
performed on either a LI-COR Odyssey instrument (for fluorescence detection) or a Bio-Rad
ChemiDoc MP gel imager (for chemiluminescence detection); and quantitation of raw 16bit digital images was implemented with the
ImageJ gel analysis plug-in (5). Using film to
perform quantitative immunoblotting was
avoided entirely, because the dynamic range
of film is so small that quantitative analysis is
virtually impossible (3). Film can make small
differences in abundance appear as large differences in band intensity. When saturated,
film exposures can also hide sample-to-sample variations in high-abundance proteins
such as loading controls. Therefore, throughout this study, all data were acquired as digital images. The diagnostic experiments shown
here can be easily adapted for other hardware
and reagent configurations.
RESULTS
Sample preparation: A critical factor for
quantitative immunoblotting
The conditions of cell lysis have a profound
impact on the proteins that are extracted and
the condition in which they are preserved. For
example, lysis of cells or tissues with purely
nonionic detergents (Triton X-100 or NP-40)
causes some proteins to partition into the
soluble and insoluble (pellet) fractions after
centrifugation. Radioimmunoprecipitation
assay (RIPA) buffer—containing dilute SDS
(a denaturing detergent) and deoxycholate (a
disruptor of protein-protein interactions)—is
widely used as a lysis buffer for whole-cell extraction. Nonetheless, RIPA buffer lysis still
generates an insoluble fraction with major
protein constituents from the cytoskeleton
and extracellular matrix (6).
To test how RIPA lysis conditions affected
immunoblotting results, I lysed HT-29 human colon adenocarcinoma cells in RIPA
buffer, boiled the RIPA-insoluble pellet in an
Originally published 7 April 2015 in SCIENCE SIGNALING
equal volume of dithiothreitol-containing
Laemmli sample buffer (7), and then immunoblotted for 20 different protein targets. As
expected, RIPA lysis buffer efficiently solubilized many cytoplasmic proteins [glyceraldehyde-3-phosphate dehydrogenase (GAPDH)
and heat shock protein 90 (Hsp90)] and signaling proteins [inhibitor of nuclear factor
κB α (IκBα) and various kinases] (Fig. 1, A
and B). RIPA buffer also extracted the cytoskeletal and cytoskeleton-associated proteins, actin and focal adhesion kinase (FAK).
However, tubulin and intermediate filament
proteins (lamin A and KRT5) showed substantial losses into the RIPA-insoluble fraction (Fig. 1C). Remarkably, RIPA insolubility
was not limited to cytoskeletal proteins: the
transcription factor GATA2 and the cell-cell
adhesion protein β-catenin were also present in the insoluble fraction. In contrast, lysis with Laemmli sample buffer, followed by
shearing of the viscous genomic DNA with a
high-gauge needle, solubilized proteins that
are tightly associated with DNA, such as histones (Fig. 1D). Despite rules of thumb for
protein solubility in various lysis buffers (6),
these results show that it is best to confirm
proper solubilization of proteins of interest
before embarking on an immunoblot study.
If 10 to 30% of a protein were consistently lost in the insoluble fraction, then
the choice of lysis conditions would not be
critical. However, specific proteins can shift
between soluble and insoluble fractions in
a stimulus-dependent manner. As an example, I activated the FAS death receptor
in MCF10A-5E human breast epithelial cells
(8) and lysed the cells in NP-40 buffer (lacking SDS and deoxycholate), RIPA buffer, or
Laemmli sample buffer. Although the cleavage of caspase-3 was readily detected in all
three preparations (Fig. 2A), cleaved forms
of caspase-8 were only detected in Laemmli
sample buffer (Fig. 2B). Thus, quantitative
measures of caspase-8 processing would require the Laemmli preparation for accurate
results (9–11). My laboratory has found that
similar precautions are required for monitoring regulated changes in intermediatefilament proteins (12), such as KRT5 (Fig. 1C).
Regulated oligomeric or polymeric protein
assemblies may be particularly susceptible to
differential partitioning between soluble and
insoluble fractions.
The stability and posttranslational modifications of lysate proteins are also affected by
the activity of co-mingling cellular enzymes,
such as proteases and phosphatases. These
enzymes are usually blocked with inhibitors
that are supplemented into nondenaturing
lysis buffers, but the SDS and deoxycholate
in RIPA are sometimes assumed to inactivate most cellular enzymes. I tested how effectively RIPA and NP-40 buffers inhibited
protein phosphatases by omitting from both
buffers the Ser-Thr phosphatase inhibitor
microcystin-LR (13) and the Tyr phosphatase
SE C T I O N T H R E E | RESEARCH ARTICLES
Fig. 1. RIPA buffer solubilizes many, but not all, cellular proteins. (A)
Examples of proteins that are entirely solubilized [100% in the supernatant (Sup)]. (B) Examples of proteins that are mostly solubilized (>90%
Sup). (C) Examples of proteins that are partially solubilized (≤90% Sup).
(D) Dimethyl-lysine 4 histone H3 (H3K4me2) resides almost entirely in the
RIPA-insoluble pellet (Pel). Band intensities were quantified from the 16-bit
digital image by densitometry in ImageJ and are shown normalized to the
Sup lane for each target. n.d., not detected. Data are representative of two
to four experiments.
inhibitor orthovanadate (14). I lysed parallel
cultures of AC16 ventricular cardiomyocytes
(15) and immunoblotted for multiple phosphorylation sites along the Akt–glycogen
synthase kinase-3 (GSK3)–glycogen synthase
(GS) signaling axis (Fig. 3, A to D). For these
phosphoproteins, the addition of phosphatase inhibitors was more critical in RIPA buffer than in the nondenaturing NP-40 buffer.
The extent of sensitivity depended strongly
on the phosphorylation site, with Thr308
of Akt, Ser21 of GSK3α, and Ser9 of GSK3β
showing greater lability than Ser473 of Akt
and Ser641 of GS. These results collectively
showed that lysis buffer composition substantially affects the results of quantitative
immunoblotting.
Assessing immunoblot protocols by
serial dilution
For immunoblotting, a single protocol that is
optimal for all electrophoresis transfer setups and
detection methods does not exist. However, the
quantitative accuracy and dynamic range of any
protocol can be assessed using a serial dilution
of cell extract and a panel of primary antibodies. As an example, I sought to determine
wet-transfer conditions (specifically, methanol concentration) that enabled quantitative
detection of most target proteins. Here, the
transfer buffer always included 10% methanol
(see Materials and Methods), but other protocols use 20% methanol according to the original conditions of Towbin et al. (4). Although
immunoblot bands are brighter and crisper
with the higher methanol concentration, how
Fig. 2. Posttranslational modifications can move protein into the
insoluble fraction of common lysis buffers. MCF10A-5E cells were exposed to the Fas cross-linking agent anti-APO-1 (1 μg/ml) (48) for 24 hours,
and then floating and adherent cells were lysed in NP-40 lysis buffer, RIPA
buffer, or dithiothreitol-containing Laemmli sample buffer (SB). (A) Effect
of solubilization conditions on the detection of cleavage (Clv.) products of
caspase-3. (B) Effect of solubilization conditions on the detection of cleavage products of caspase-8. Vinculin, tubulin, GAPDH, Hsp90, and p38 were
used as loading controls where indicated. Data are representative of three
experiments.
methanol percentage
affects quantitative
accuracy and dynamic
range of immunoblot
band intensities is
unknown.
Using HT-29 cell
extracts, I performed
an extended twofold serial dilution from a
grossly overloaded sample (200 μg of extract)
to one below the limit of detection (100 ng of
extract). Two replicate gels were transferred
in buffer containing 10 or 20% methanol, and
then membranes were probed for seven different targets (file S1). For actin and p38, I found
that detection was linear up to 50 μg of total
protein, irrespective of the methanol concentration (Fig. 4, A and B). This zero-intercept
linearity is ideal, because band density is directly proportional to input material without
the need to refer to a calibration curve. By
contrast, Hsp90 and tubulin showed a hyperbolic saturation under both transfer conditions with less than 25 μg of total protein
(Fig. 4, C and D). Saturation can result from
steric crowding of antibody epitopes, quenching of fluorescently labeled secondary antibodies, or oxidation of enzyme-conjugated
secondary antibodies. Regardless of the
source, band densities in this regime no longer provide a linear estimate of sample abundance, and calibration is required to obtain
accurate measurements. The improved transfer of proteins in 20% methanol appeared to
shift the detection of multiple targets from a
linear regime to one of hyperbolic saturation
(Fig. 4, E to G). The results from this diagnostic study indicated that the lower methanol
concentration was preferred for my immunoblot protocol.
Similar comparisons have also caused my
laboratory to favor fluorescence-based detection over chemiluminescence whenever
possible. I compared the linearity of IRDye-
conjugated secondary antibodies to horseradish peroxidase conjugates that were incubated
with an enhanced chemiluminescence (ECL)
cocktail (16) or a commercial substrate marketed for high-sensitivity applications (file
S2). Under the same immunoblotting conditions, chemiluminescent exposures consistently yielded stronger band densities (Fig.
5, A to C). However, the linear dynamic range
was very limited, and signals often decreased
at high protein inputs (Fig. 5C). This can occur when side products of the peroxidase-catalyzed reaction are oxidized and precipitated,
causing the membrane to “brown out” and
absorb the emitted photons. Titrating down
the amount of protein or primary antibody
can avoid the problem, provided that the researcher is aware of it.
Optimizing loading controls
Arguably, the biggest source of confusion in
quantitative immunoblotting is the role of
protein loading and loading controls (17). Normalizing for cell numbers across samples is
challenging, because it is difficult to estimate
changes in cell proliferation and death under
different conditions. Such estimates also do
not account for variations in initial seeding
density and final lysate volumes, which will
affect the observed protein abundance. Consequently, immunoblot samples are often prepared according to total cellular protein (18,
19), assuming that the average protein content per cell is constant across the different
conditions.
To complement total protein estimates, immunoblots typically include loading control
proteins, which provide a secondary check
that roughly equal amounts of cellular material have been added. Two key assumptions of
the loading control are that (i) its abundance
is roughly constant across the different conditions, and (ii) its immunoblot band intensity
is linearly reflective of its abundance. How13
ever, a single loading control may not fulfill
both of these assumptions (2).
If total cellular protein can be quantified accurately, then total protein is loaded
equivalently across samples, and one or two
loading controls suffice as a qualitative confirmation of overall protein abundance (20–22).
Conversely, if the total cellular protein is not
known or cannot be determined accurately,
then the input must be normalized to some
estimate of protein loading. A common approach found in the literature is to normalize
by only one loading control, but this scaling
is highly problematic. Taking one unknown
quantity (the protein of interest) and dividing it by another unknown quantity (a single
loading control) creates a number with very
poor statistical properties, including an undefined mean. The dangers of single-variable
normalization have long been recognized in
data from microarrays (23) and quantitative
polymerase chain reaction (PCR) (24), but not
in data from immunoblotting. A solution is to
aggregate the band intensities from multiple
loading controls, calculating a mean estimate
of total cellular content that is less sensitive
to the technical or biological fluctuations of a
single loading control (12, 25).
To demonstrate the utility of multiprotein
normalization, I immunoblotted for linker
phosphorylation of Smad2 on Ser245/250/255 (pSmad2 linker) in MCF10A-5E cells that had
been stimulated with transforming growth
factor–β (TGFβ) (file S3). Under these conditions, Smad2 linker phosphorylation is mediated by cyclin-dependent kinases (CDKs) (26),
so cells were additionally pretreated with or
without the pan-CDK inhibitor flavopiridol
(27). The TGFβ stimulation ± flavopiridol
inhibition experiment was performed in biological quadruplicate to assess reproducibility
of p-Smad2 linker quantification. Lacking any
total protein normalization, p-Smad2 linker
densitometry was variable, with flavopiridol
producing only a marginally significant decrease in phosphorylation (Fig. 6, A and B).
To improve reproducibility, I blotted the same
membrane for five potential loading controls:
total Smad2, tubulin, Hsp90, GAPDH, and
p38. The antibodies for these proteins are
from various hosts and yield single immunoreactive bands under the blotting conditions
used here. The properties of the antibodies
and the detected control proteins enabled
multiplex detection of loading controls together with p-Smad2 linker or after a single
round of membrane stripping (see Materials
and Methods).
Upon quantifying band densities in
ImageJ, I compared the reproducibility
of p-Smad2 linker after normalization to all
possible combinations of loading controls: 25
= 32 combinations. For single loading control normalization, the effect on replicate-toreplicate reproducibility heavily depended on
the choice of loading control. Normalizing to
GAPDH decreased the p-Smad2 linker coef14
Fig. 3. Phosphatase inhibitors are
critical to preserve
certain phosphorylated residues
under certain lysis
conditions. (A and
B) Effect of lysis buffer and presence or
absence of phosphatase inhibitors (PPIs)
on the detection of
phosphorylated Akt
(p-Akt) on Thr308
(T308) and Ser473
(S473). (C) Effect of
lysis buffer and presence or absence of
PPIs on detection of
GSK3α/β phosphorylated on Ser21 and
Ser9 (p-GSK3α/β).
(D) Effect of lysis
buffer and presence
or absence of PPIs
on detection of GS
phosphorylated on Ser641 (p-GS). AC16 cells were lysed in RIPA or NP-40 lysis buffer with or without
PPIs. Vinculin, tubulin, GAPDH, and actin were used as loading controls where indicated. Total Akt,
GSK3α/β, and GS were used to monitor specific changes in protein abundance. Band intensities were
quantified from the 16-bit digital image by densitometry in ImageJ and are shown normalized to the
average +PPI conditions for each target across both lysis conditions. Data are representative of two
experiments.
ficient of variation by more than twofold (21
to 9%), but normalizing to tubulin had virtually no effect (Fig. 6C). This does not imply
that GAPDH is always a good loading control
or that tubulin is always a bad one; rather, it
emphasizes the danger of relying on a single
measured variable to estimate total protein
content. As higher-order combinations of
loading controls were tested as normalizers,
I found that the coefficient of variation of pSmad2 linker steadily improved toward 7 to
8%, consistent with values reported previously (10, 11). This reproducibility became
less dependent on the specific combination
of loading controls (note the shrinking error
bars in Fig. 6C), indicating that I had converged on a measure of cellular content per
lane that was truly representative. With all
five loading controls, both the TGFβ-induced
stimulation of p-Smad2 linker and its inhibition by flavopiridol were highly significant
(Fig. 6D).
Others have noted that many common
loading controls are abundant proteins (for
example, tubulin) that result in saturated
band intensities under the conditions needed
to detect proteins or modification states of interest (2). On the basis of the results with the
serial dilutions (Fig. 4D), saturation is probably the reason why tubulin worked poorly as
a loading control for p-Smad2 linker in this
setting (Fig. 6B). However, in the context of
multiple loading controls that are averaged,
a saturated loading control negligibly affects
normalization because of its reduced sampleto-sample variation. An alternative strategy
is to prepare a separate set of immunoblot
lanes with decreased protein content for
loading controls (2). However, normalizing
from different lanes will miss lane-specific
irregularities in sample preparation or electrophoretic transfer, which can be minimized
with good technique but not eliminated.
A possible alternative to multiple loading controls is to use reversible total protein
stains that are compatible with PVDF membranes (such as MemCode or Ponceau S).
For a PVDF-compatible stain to be useful, it
must be quantitative for total cellular protein
(directly proportional or, at least, hyperbolically saturating), and it must not interfere
with the subsequent immunoblot. I tested
two total protein stains that can be reversibly applied to PVDF membranes: MemCode
Reversible Protein Stain (commercially available from Thermo Scientific) and Ponceau S
(28). MemCode yielded a strong blue banding
pattern that was readily detected by whitelight epi-illumination and a charge-coupled
device (CCD) camera (fig. S1A). Using a blank
region of the PVDF membrane to define a
background for subtraction, I found that total MemCode band intensity increased hyperbolically with zero intercept over a relevant
range of lysate amounts (fig. S1B). However,
the staining procedure markedly increased
the 700-channel background fluorescence of
the membrane, and this background was not
Fig. 4. Linearity and hyperbolic saturation
of immunoblots determined by serial dilution. (A and B) Immunoblots for actin and p38
are linear under both transfer conditions. (C
and D) Immunoblots for Hsp90 and tubulin are
hyperbolically saturated under both transfer
conditions. (E to G) Linear detection of immunoblots for E-cadherin, ERK1/2 (extracellular
signal–regulated kinases 1 and 2), and GAPDH
occurred with tank transfer conditions containing 10% methanol. HT-29 cells were lysed in RIPA
buffer, immunoblotted for the indicated targets,
and imaged. Left: Immunoblots. Middle: Log-log
plots of the quantified band intensities from the
blots on the left. Right: Linear plots of the same
data. Linear fits are gray when the hyperbolic
model is no better than the linear model for that
transfer condition. Linear fits are red when the
linear fit of the associated transfer condition is
better than the linear fit of the other transfer
condition. Hyperbolic fits are green when the
hyperbolic model is better than the linear model
for that transfer condition. Data are in blue when
neither the linear nor the hyperbolic model provides a better fit. Model comparisons were done
by the F test [false discovery rate (FDR) = 5%;
n = 5 to 8 dilutions]. See file S1 for raw images
and calculations.
removed by the recommended stain removal
(erasing) procedure (fig. S1C). Although potentially useful for immunoblots detected by
chemiluminescence, MemCode is not suitable
for two-color fluorescence detection.
I uncovered a different set of problems with
Ponceau S. Compared to MemCode, the red
Ponceau S stain was not as efficiently detected
by the CCD camera (fig. S1, A and D). However, its image densitometry was linear as a
function of lysate amount, and there was no
background fluorescence caused by Ponceau
S staining or erasure (fig. S1, E to G). Ponceau S appeared to fill all the requirements
for a total protein stain, except for one major
drawback—the zero intercept of its densitometry could not be accurately estimated from a
blank region of the PVDF membrane (fig. S1D,
lane B), causing a negative bias of ~10,000 intensity units (fig. S1E). Without a lysate calibration curve to estimate this bias on a PVDF
membrane, I concluded that Ponceau S cannot be used for relative protein quantification.
When immunoblotting for phosphoproteins,
there are additional complications with erasing Ponceau S from PVDF membranes, because the alkaline conditions for erasure will
chemically modify phosphorylated Ser and
Thr residues (29).
Given these data with the reversible total
protein stains, accurate quantitation of immunoblot data should adopt the best practices of quantitative PCR (24) and use an
assortment of three or more loading controls,
spanning a range of abundances, when direct total protein measures are lacking. To
minimize cost and effort, my laboratory detects constitutively produced proteins with
15
high-affinity antibodies that work reliably
at low concentrations (25 to 100 ng/ml),
yield single immunoreactive bands, and
thus are ideal for multiplexing (see Materials and Methods; Table 1).
Stripping, reprobing, and the total
protein control
Aside from loading controls for total cellular content, immunoblots that quantify protein modification states should
contain an additional control: an immu- Fig. 5. Quantitative immunoblotting is challenging when imaging by chemiluminescence. (A to C)
noblot for the total protein. This control HT-29 lysates were prepared as in Fig. 4, immunoblotted for the indicated proteins, and imaged by IRDye
serves to gauge how much of the observed fluorescence, ECL, or SuperSignal West Femto chemiluminescence as described (12, 16, 45, 46). Linear
change in protein modification can be fits are shown in gray when the hyperbolic model is not significantly better than the linear model for that
explained by differences in target abun- imaging condition. Linear fits are shown in red when the linear fit of the associated imaging condition is
dance. For rapid experiments that are significantly better than that of the other imaging conditions. Hyperbolic fits are shown in green when the
expected to avoid protein turnover and hyperbolic model is significantly better than the linear model for that imaging condition. Data are interposynthesis (such as the one shown in Fig. lated in blue when neither the linear nor the hyperbolic model provides a better fit. All model comparisons
6), the total protein control can addition- were done by the F test at an FDR of 5% (n = 4 to 8 dilutions). See file S2 for raw images and calculations.
ally contribute to a panel of loading controls if changes in band intensity are clearly
sidual GAPDH antibody staining. By contrast,
Recombinant ERK2 and p38 proteins were
cofluctuating with other loading controls.
the guanidinium strip removed most of the
cloned, purified, and quantified by protein
However, for comparisons on long time scales,
GAPDH antibody but left a clear p-ERK1/2
assay. Using a known mass of bovine serum
across different cell types, or with rapid proartifact in the total ERK1/2 reprobe, with inalbumin (obtained as a calibrated standard
tein turnover, the total protein control is specreased staining in the +EGF lane. The SDS
from a protein assay kit), a standard curve
cifically important on its own and may not be
plus β-mercaptoethanol strip completely rewas constructed by running serially diluted
a reliable indicator of loading.
moved the GAPDH antibody and yielded the
bovine serum albumin on a polyacrylamide
A common way to estimate target protein
closest approximation of total ERK1/2 abungel alongside recombinant glutathione Sabundance from a modification-specific imdance, although with substantial loss of total
transferase (GST)–tagged purified ERK2 and
munoblot is to strip the membrane of antiprotein from the membrane based on the imp38 (22) (Fig. 8A). The gel was then stained
body and then reprobe with an antibody that
munoreactivity of the three additional loadwith Coomassie blue and scanned for its nearrecognizes the total protein (or at least the uning controls (compare the glycine-stripped
infrared fluorescence (34), yielding a digital
modified form). Essential to this approach is
membrane to the membrane stripped with
image for densitometry in ImageJ (see Maconfirming that the stripping conditions have
SDS plus β-mercaptoethanol). Moreover, after
terials and Methods). To accommodate some
fully removed the original modification-speSDS plus β-mercaptoethanol stripping, the todegree of band saturation and improve the
cific antibody from the membrane of choice
tal ERK1/2 bands still showed an artifactual
dynamic range of detection, I fit the protein
(PVDF or nitrocellulose). If not, the reprobed
increase in the EGF-treated sample. To avoid
band intensities to a simple hyperbolic curve:
blot will show artifacts from the residual first
the need to strip and reprobe the same blot,
antibody. As an example, I acutely stimulated
an alternative is to use two-color fluorescence
a × Protein
Intensity =
AC16 cells with epidermal growth factor (EGF)
detection with phosphorylation-specific and
b + Protein
for 5 min and immunoblotted for phosphorytotal antibodies that are raised in different
lated ERK1/2 (p-ERK1/2) with an antibody
hosts and against different epitopes (Fig. 7C).
with two free parameters (a and b) that were
that is difficult to remove. Technical replicates
Another option is to measure total target proestimated by least-squares regression. The
of the same two lysates showed good reprotein by immunoblotting a replicate set of sammodeled fit captured all of the albumin standucibility in the observed EGF-stimulated
ples and confirming that the matched loading
dards and enabled mapping the measured
induction of p-ERK1/2, and loading controls
controls are comparable to those in the modiGST-ERK2 and GST-p38 band densities to
were consistent across the membrane (Fig.
fication-specific immunoblot (Fig. 7D).
total protein amount (Fig. 8B). Dividing
7A). After cutting the membrane into thirds, I
by the volume of sample loaded into the
Advanced quantitative immunoblotting:
tested three stripping conditions: (i) a gentle
gel resulted in an estimated concentration
Absolute quantification
low-pH glycine buffer solution [1.5% glycine
of full-length protein in the recombinant
(pH 2.2), 0.1% SDS, and 1% Tween], (ii) a more
For some computational models of biochemipreparation.
stringent guanidinium solution (6 M guanical networks, the absolute abundances of
Size separation of the recombinant protein
dine-HCl) (30), and (iii) a high-stringency SDS
specific cellular proteins are needed (32, 33).
before quantification is critically important,
plus β-mercaptoethanol solution with heat
Such applications require explicit calibrabecause purified preparations often contain
[62.5 mM tris (pH 6.8), 2% SDS, and 100 mM
tion using proper absolute standards on the
cleavage products that add to total protein
β-mercaptoethanol at 50°C] (31). The stripped
same immunoblot. For the cellular proteins
but are not immunoreactive. Commercial
membranes were blocked identically and repof interest, appropriate standards are purivendors of recombinant protein may simply
robed for total ERK1/2 along with three addified recombinant proteins, which have been
quantify total protein (full length plus fragtional loading controls. Reprobed membranes
calibrated against a purified protein of known
ments that are not useful for calibration),
were imaged and displayed identically to emmass.
ultimately resulting in an overestimation of
phasize differences in the immunoblot signal
To illustrate this process, I quantified the
cellular protein. For calibration, one should
detected (Fig. 7B).
absolute masses of ERK2 and p38 (per 25 μg
also remember that unknown samples must
The experiment revealed that the low-pH
of cellular extract) in HT-29 and AC16 cells
fall between calibration samples that are well
glycine strip was ineffective at removing the
(file S4). Absolute quantification enabled
fit by the linear or hyperbolic curve. Samples
antibodies. Indeed, “total” ERK1/2 looked
ERK2-p38 comparisons within each cell line
that fall outside the calibration range will
identical to p-ERK1/2, and there was even reas well as between cell lines.
cause the linear or hyperbolic equation to
16
make extrapolations that could
persistent concern is saturation
be highly inaccurate.
(Fig. 4). Saturated immunoblots
Before making comparisons
do not overestimate a change
between cell lines, I verified
in protein; instead, they can
that loading the same total mass
substantially underestimate it.
of HT-29 and AC16 extract gave
To demonstrate, I plotted diluapproximately equivalent band
tions of an unstimulated and
intensities for several loadstimulated extract for a theoing controls (Fig. 8C). Hsp90
retical immunoblot band that
abundance was higher (per
is hyperbolically saturated (Fig.
total mass) in HT-29 cells,
9). There is a clear threefold
whereas vinculin and tubulin
change in abundance when the
were higher in AC16 cells, and
target is immunoblotted under
GAPDH and actin were approxiconditions where the linear apmately equal. The uncorrelated
proximation is accurate (<10
abundance of these “housekeepμg of extract in this example).
ing proteins” suggests that the
However, overloading the gel
cell extracts can be fairly comwith 50 μg of extract reduces
pared on a total protein basis.
the difference to 1.4-fold. This
The results further reinforced
type of dampening has led some
the importance of using multo conclude that fluorescence
tiple loading controls.
detection is not sensitive comTo quantify the protein of
pared to chemiluminescence
interest in cellular extracts, I
(especially on film). However,
ran the calibrated standards
the comparison is not fair if the
and the extracts alongside one
1.4-fold difference in antibody
another and blotted with an
binding has been exaggerated
antibody that detects both the
by a nonlinear chemiluminesrecombinant protein and the
Fig. 6. Reproducibility of quantitative immunoblots across biologicent reaction (Fig. 5, B and C).
protein of interest (Fig. 8, D to
cal replicates is improved after normalization to multiple loading
The analysis here also ilG). Note that the ERK2 calibracontrols. (A) Representative immunoblot for phosphorylated Smad2 on
lustrated
that
quantifying
tion cannot be used to quantify
Ser245/250/255 (p-Smad2 linker) in MCF10A-5E cells stimulated with TGFβ (50
phosphorylation-specific immuthe ERK1 band at 44 kD because
ng/ml) for 30 min with or without 1 hour of preincubation with 300 nM flanoblots as a “phosphorylated-toof differences in the immunovopiridol. Tubulin, Hsp90, GAPDH, and p38 were used as loading controls.
total ratio” is fraught with both
reactive epitope. In addition,
Total Smad2 was used to monitor overall changes in protein abundance
numerical complications (Fig. 6)
a mass correction is needed
and served as a fifth candidate loading control for this analysis. (B) Raw
and potential experimental artito account for the size differp-Smad2 linker densitometry quantified in ImageJ. (C) Decrease in the cofacts (Fig. 7). The ratio is further
ence between the GST-tagged
efficient of variation among p-Smad2 biological replicates with increasing
prone to be misinterpreted as a
recombinant protein and the
numbers of loading controls. The best (GAPDH) and worst (tubulin) single
phosphorylation stoichiometry,
endogenous protein, because
loading control normalizations are highlighted. (D) p-Smad2 linker densiwhich cannot be calculated in
immunoblotting quantifies antometry after normalization to the mean band intensity of tubulin, Hsp90,
experiments that use differtigen independently of its mass
GAPDH, p38, and total Smad2 for each biological replicate. For (B) and (D),
ent antibodies to detect the
(the correction is 42/69 for
data are shown as means ± SE of n = 4 biological replicates, with differences
phosphorylated and total proERK2 and 38/65 for p38; Fig. 8,
in means assessed by Welch’s two-sided t test. For (C), data are shown as
tein because of differences in
E and G). This analysis showed
mean coefficients of variation ± SD of n = 1 to 10 possible normalization
antibody affinity. Calculating
that AC16 cells had greater than
combinations for the indicated number of loading controls. See file S3 for
phosphorylation stoichiometry
sevenfold more ERK2 protein
raw images and calculations.
by immunoblotting requires
compared to HT-29 cells, despite
electrophoretic conditions that
only about twofold difference in
separate the phosphorylated
band intensity (Fig. 8E). The limited differto be meaningful (8, 36–40). What are the
and total forms by mobility (41). Modificaence in band intensity was due to the saturaimplications of pseudoquantitative immution stoichiometry can be critical under
tion of the ERK2 immunoblot at high protein
noblotting? When chemiluminescence is
certain conditions (42), but there are also
concentrations, which was captured by the
used cavalierly, there is a danger of wildly
examples where increases in total protein
calibration curve. ERK2 abundance in AC16
exaggerated claims. For example, a twofold
have affected signal flow (43). Considering
cells was also high compared to the abunchange in Hsp90 abundance might appear
these caveats, my conclusion is that phosdance of p38, which had a concentration that
as a fivefold change in the intensity of the
phorylated-to-total ratios based on stripped
was an order of magnitude lower (Fig. 8G).
band on an immunoblot (compare the ECL
and reprobed membranes should be avoided
Although laborious, this type of absolute
values for ~6 and ~3 μg in Fig. 5B). A twowith immunoblot data.
quantification can be done systematically for
fold change in GAPDH abundance could be
A major challenge in evaluating the immultiple cellular proteins and provide new
“quantified” as a >20-fold change on the
munoblot data of others is that publications
insight into signaling function (35).
basis of band intensity (compare the ECL
and manuscripts will often omit details on
values for 25 and ~13 μg in Fig. 5C). This
acquisition that are considered “routine.”
DISCUSSION
makes numerical results look very impresThose details matter, because the evaluJust because we can put numbers on an imsive, but the truth is still a twofold change
ation criteria of an immunoblot detected
age does not imply that we should—a quanin abundance.
by film are different than those of an imtified biomolecule should relate directly to
With fluorescence detection, there is not
munoblot detected by fluorescence. When
the true quantity of that biomolecule if it is
the danger of a runaway reaction, but a
such information is missing, readers and
17
Fig. 7. Membrane stripping and reprobing is a quantitative trade-off between antibody removal and total protein loss. (A) Replicate immunoblots
for ERK1/2 phosphorylated on Thr202 and Tyr204 of ERK1 and Thr185 and Tyr187 of ERK2 (p-ERK1/2) in AC16 cells stimulated with EGF (100 ng/ml) for 5 min.
GAPDH and tubulin were used as loading controls in the first immunoblot. (B) Reprobe of the membrane in (A) for total ERK1/2 after stripping with glycine
buffer, guanidinium, or β-mercaptoethanol (βME) stripping buffer. Vinculin, Hsp90, and actin were used as loading controls for the reprobed blots. (C)
Two-color fluorescence immunoblot for p-ERK1/2 (green) and total ERK1/2 (magenta) of the same lysates as in (A). Vinculin and Hsp90 were used as
loading controls. (D) Direct immunoblot for total ERK1/2 of the same lysates as in (A). A lower percentage polyacrylamide gel was used in (C) and (D) to
emphasize the total ERK1/2 upshift after stimulation with EGF. GAPDH, vinculin, Hsp90, and tubulin were used as loading controls. Data are representative
of two experiments.
Fig. 8. Workflow for absolute protein quantification. (A) Serial dilution of an albumin standard to calibrate recombinant purifications of GST-ERK2
and GST-p38 by Coomassie staining. (B) Albumin band intensity (black) plotted as a function of protein and fit to a hyperbolic model (gray) that infers
the amounts of GST-ERK2 (green) and GST-p38 (purple) proteins. (C) HT-29 and AC16 cells have roughly equal protein constituents by mass based on
the amount of Hsp90, vinculin, tubulin, GAPDH, and actin detected in 25 μg of each sample. (D) Serial dilution of the GST-ERK2 standard to calibrate endogenous abundances of ERK2 in HT-29 and AC16 cells. (E) GST-ERK2 band intensity (black) plotted as a function of protein input and fit to a hyperbolic
model (gray) that infers the amount of ERK2 in HT-29 cells (blue) and AC16 cells (red). (F) Serial dilution of the GST-p38 standard to calibrate endogenous
abundances of p38 in HT-29 and AC16 cells. (G) GST-p38 band intensity (black) plotted as a function of protein input and fit to a hyperbolic model (gray).
The model was used to infer the amount of ERK2 in HT-29 cells (blue) and AC16 cells (red). Data are representative of two experiments. See file S4 for raw
images and calculations.
reviewers can distinguish a film exposure
by the hazy gray of the background and the
blurred borders of bands resulting from the
flatbed optical scan of a film at arbitrarily
high resolution (44). Digitally acquired immunoblots will often have a whiter background with crisper bands that may appear
pixelated because of binning on a CCD camera or the step size of a fluorescence scanner.
Although some may find them less aesthetically appealing, digitally acquired images
provide the more accurate representation of
band intensity and its relationship to sample
abundance.
The message of this research resource
is not that chemiluminescence cannot be
quantitative or that film exposures are al18
ways inappropriate. Rather, I want to convey
that with numbers comes great responsibility. There are straightforward ways to diagnose immunoblot accuracy (Figs. 4, 5, and
8) and precision (Fig. 6). We should all be
encouraged to complete these diagnostics
on our own targets and immunoblot setups
before diving in to generate “real data.” The
stakes are simply too high to do otherwise.
MATERIALS AND METHODS
Cell lines, stimulation, and lysis
HT-29 cells were obtained from the American
Type Culture Collection and maintained as
recommended. The 5E clone of MCF10A cells
was isolated and maintained as previously
described (8). AC16 cells (15) were purchased
from M. Davidson (Columbia University) and
maintained in Dulbecco’s modified Eagle’s
medium/F-12 medium (Life Technologies)
plus 12.5% tetracycline-free fetal bovine serum (Clontech) and penicillin-streptomycin
(Gibco).
Cells were stimulated with the indicated
concentrations of anti-APO-1-3 cross-linking
antibody (Axxora), TGFβ (PeproTech), or EGF
(PeproTech) for the indicated times, washed
with ice-cold PBS (phosphate-buffered saline),
and then lysed in RIPA buffer [50 mM tris-HCl
(pH 7.5), 150 mM NaCl, 1% Triton X-100, 0.5%
sodium deoxycholate, 0.1% SDS, 5 mM EDTA
supplemented with aprotinin (10 μg/ml), leusciencemag.org SCIENCE
peptin (10 μg/ml), pepstatin (1 μg/ml), 1 mM
phenylmethylsulfonyl fluoride (PMSF), microcystin-LR (1 μg/ml), and 200 μM sodium orthovanadate], NP-40 buffer [50 mM tris-HCl
(pH 8.0), 150 mM NaCl, 0.5% NP-40 substitute,
5 mM EDTA supplemented with aprotinin (10
μg/ml), leupeptin (10 μg/ml), pepstatin (1 μg/
ml), 1 mM PMSF, microcystin-LR (1 μg/ml),
and 200 μM sodium orthovanadate], or dithiothreitol-containing Laemmli sample buffer [62.5 mM tris-HCl (pH 6.8), 2% SDS, 10%
glycerol, 100 mM dithiothreitol, and 0.01%
bromophenol blue]. Laemmli sample buffer
lyses were performed at room temperature
to avoid precipitating the SDS, and viscosity of these lysates was reduced by passing
them vigorously through a 25-gauge needle 5
to 10 times. For RIPA and NP-40 lysates, protein concentrations were determined with
the bicinchoninic acid assay (Thermo Fisher
Scientific).
For solubilization comparisons (Fig. 1),
HT-29 cells were lysed in RIPA buffer supplemented with protease and phosphatase inhibitors as described above. After incubation
on ice and centrifugation, the RIPA-insoluble
pellet was boiled for 5 min in an equivalent
volume of dithiothreitol-containing Laemmli
sample buffer.
Immunoblotting: Polyacrylamide gel
electrophoresis
Immunoblotting was performed as described
(12, 45, 46), but the details of the implementation will be elaborated upon here. Samples
were prepared in dithiothreitol-containing
Laemmli sample buffer to a total volume of
20 or 40 μl. Polyacrylamide gels (8, 10, 12, or
15%) of 1.5 mm thickness were cast according
to (1), and samples were electrophoresed in
tris-glycine running buffer (25 mM tris base,
250 mM glycine, and 0.1% SDS) at 130 V until
the dye front reached the end of the gel.
Immunoblotting: Electrophoretic
transfer
Proteins from the polyacrylamide gel were
transferred to a PVDF membrane (Millipore; Immobilon-FL, 0.45-μm thickness) in
a Mini Trans-Blot Electrophoretic Transfer
Cell (Bio-Rad) under modified conditions of
Towbin et al. (4) (25 mM tris, 192 mM glycine, 0.0375% SDS, and 10% methanol unless
otherwise indicated). Transfers were electrophoresed at 100 V for 1 hour under ambient
conditions with an ice block in the transfer
tank and the transfer tank surrounded by
ice.
Immunoblotting: Membrane blocking
After transfer, the molecular weight markers
on the membrane were overwritten with a
lead pencil (to provide 700-channel fluorescence), and the membrane was blocked with
0.5× blocking buffer: Odyssey blocking buffer (LI-COR; #927-40000) diluted in an equal
volume of PBS. Although not observed for the
immunoblots here, some phosphorylation
site–specific antibodies can be competed
away from their target epitopes with PBS
buffers. In this circumstance, the 0.5× blocking buffer should be prepared with Odyssey
blocking buffer (LI-COR; # 927-50000) and
diluted in an equal volume of tris-buffered
saline (TBS) throughout the procedure. TBS
buffers were also used for the fluorescencechemiluminescence comparison (Fig. 5).
All blocking steps used a surface-to-volume
ratio of 5-ml 0.5× blocking solution per 2
¾-inch × 3 ¼-inch membrane from a 15-well
minigel. The membrane and 0.5× blocking
solution were sealed in a plastic bag and incubated on a rotating platform for 1 hour at
room temperature.
Immunoblotting: Antibody probing
After blocking, membranes were incubated
with 0.5× blocking solution + 0.1% Tween
20 containing primary antibodies recognizing the proteins or epitopes listed in Table 1.
All primary antibody steps used a surfaceto-volume ratio of 5-ml primary antibody
solution per 2 ¾-inch × 3 ¼-inch membrane
from a 15-well minigel. The membrane and
primary antibody solution were sealed in a
plastic bag and incubated on a rotating platform overnight at 4°C.
Antibody pairs raised in different species
were routinely multiplexed when using twocolor fluorescence detection. In addition,
primary antibodies with negligible off-target bands could be multiplexed in the same
detection channel if the molecular weights
of their protein targets could be clearly resolved from one another. This single-color
multiplexing enabled concurrent detection
of multiple loading controls. Common primary single-channel combinations included
antibodies recognizing tubulin (50 kD) +
GAPDH (36 kD), Hsp90 (90 kD) + p38 (38
kD), vinculin (120 kD) + actin (42 kD), and
vinculin (120 kD) + tubulin (50 kD) + GAPDH
(36 kD).
Immunoblotting: Fluorescence
detection
Membranes were removed from primary antibody solution and washed on a rocking platform for 4 × 5 min in ~25 ml of PBS + 0.1%
Tween 20 (PBS-T). After washing, membranes
were incubated with 0.5× blocking solution +
0.1% Tween 20 + 0.01% SDS containing one
or more of the following secondary antibodies: IRDye 800CW–conjugated goat antirabbit (LI-COR; #926-32211, 1:20,000), IRDye
800CW–conjugated goat anti-mouse (LI-COR;
#926-32210, 1:20,000), IRDye 680–conjugated goat anti-rabbit (LI-COR; #926-32221,
1:20,000), IRDye 680LT–conjugated goat
anti-mouse (LI-COR; #926-68020, 1:20,000),
and IRDye 680LT–conjugated donkey antichicken (LI-COR; #926-68028, 1:20,000). All
secondary antibody steps used a surface-tovolume ratio of 5-ml primary antibody solu-
Fig. 9. Quantifying partially saturated immunoblots can markedly underestimate differences between samples. In this theoretical
example, a serial dilution is performed with unstimulated and stimulated extracts. The relative
change in the linear range (b) of the immunoblot
is [99 (blue)/33 (red)] ~ threefold, whereas the
relative change at fivefold higher loading is only
1.4-fold (36%).
tion per 2 ¾-inch × 3 ¼-inch membrane from
a 15-well minigel. The membrane and secondary antibody solution were sealed in a plastic
bag and incubated on a rotating platform for
1 hour at room temperature.
Membranes were removed from secondary
antibody solution and washed on a rocking
platform for 4 × 5 min in ~25 ml of PBS-T.
To remove residual Tween 20, which is highly
autofluorescent upon 700-nm excitation, the
membrane was washed for 5 min in ~25 ml
of PBS before scanning. Fluorescence images
were obtained on an Odyssey infrared scanner (LI-COR) at 169-μm resolution and 0-mm
focus offset. Fluorescence channel intensities ranged from 5.0 to 8.5 depending on the
immunoblot.
Immunoblotting: Chemiluminescence
detection
Membranes were removed from primary antibody solution and washed on a rocking platform for 4 × 5 min in ~25 ml of TBS + 0.1%
Tween 20 (TBS-T). After washing, membranes
were incubated with 0.5× blocking solution
+ 0.1% Tween 20 containing horseradish peroxidase–conjugated goat anti-rabbit (Jackson
ImmunoResearch; #111-035-144, 1:10,000)
or anti-mouse (Jackson ImmunoResearch;
#115-035-146, 1:10,000). All secondary antibody steps used a surface-to-volume ratio of
5 ml primary antibody solution per 2 ¾-inch
× 3 ¼-inch membrane from a 15-well minigel. The membrane and secondary antibody
solution were sealed in a plastic bag and incubated on a rotating platform for 1 hour at
room temperature.
Membranes were removed from secondary
antibody solution and washed on a rocking
platform for 4 × 5 min in ~25 ml of TBS-T.
To enable a fair comparison with fluorescence detection, the membrane was washed
for 5 min in ~25 ml of TBS before exposing.
19
Table 1. List of antibodies used and proteins or epitopes detected. BD,
BD Biosciences; CST, Cell Signaling Technology; SCBT, Santa Cruz Biotechnology; Thermo, Thermo Fisher Scientific; KLF4, Kruppel-like factor 4;
MCL1, myeloid cell leukemia 1; PDI, protein disulfide isomerase.
acetic acid, and PEG-400 (polyethylene glycol,
molecular weight 400). For erasing, Ponceau
S–stained membranes were treated with 0.1
N NaOH for 30 s and washed with running
deionized water for 2 min.
For both total protein stains, digital images
were captured on a ChemiDoc MP gel imager
(Bio-Rad) with “Colorimetric” settings (2 × 2
camera binning). Fluorescence images were
obtained on an Odyssey infrared scanner (LICOR) at 169-μm resolution and 0-mm focus
offset, with a 700-channel intensity of 5.0 and
an 800-channel intensity of 8.0.
Coomassie staining and digital image
acquisition
Polyacrylamide gels were stained with 0.1%
(w/v) Coomassie blue R-250 in 40% methanol
and 10% glacial acetic acid on a rocking platform for 1 hour at room temperature and then
destained for several hours with 30% methanol + 10% glacial acetic acid until the background was acceptable. The stained gel was
scanned on an Odyssey infrared scanner (LICOR) at 169-μm resolution and 0.5-mm focus
offset for 700-channel fluorescence.
Image densitometry
Membranes were covered with an ECL solution composed of 1.25 mM luminol, 2 mM
4-iodophenylboronic acid, and 0.0162% H2O2
(16). Alternatively, membranes were covered
with SuperSignal West Femto reagent according to the manufacturer’s instructions
(Thermo Fisher Scientific). Chemiluminescent exposures were captured on a ChemiDoc MP gel imager (Bio-Rad) with “Chemi Hi
Resolution” settings (2 × 2 camera binning).
Exposure times were set manually to fill the
bit depth of the CCD camera without saturating any binned pixels.
Immunoblotting: Stripping and
reprobing
For the low-pH glycine strip, membranes
were incubated with low-pH glycine buffer
solution [1.5% glycine (pH 2.2), 0.1% SDS,
and 1% Tween] for 2 × 10 min at room
temperature on a rocking platform. The
stripped membranes were washed 2 × 10
min in ~25 ml of PBS before blocking and
immunodetection as described above. For
the guanidinium strip, membranes were
incubated with 6 M guanidine-HCl for 10 min
at room temperature on a rocking platform,
20
followed by a 5-min wash with PBS before
blocking and immunodetection as described
above. For the SDS plus β-mercaptoethanol
strip, membranes were incubated with
SDS plus β-mercaptoethanol solution [62.5
mM tris (pH 6.8), 2% SDS, and 100 mM
β-mercaptoethanol] and incubated in a dryair oven at 50°C for 30 min with occasional
agitation by hand. Stripped membranes were
washed 3 × 5 min in ~25 ml of PBS before
blocking and immunodetection as described
above.
Total protein staining after
electrophoretic transfer
PVDF membranes were stained for total
protein with MemCode Reversible Protein
Stain (Thermo Scientific) according to the
manufacturer’s recommendations. Erasure
of the MemCode stain was performed with
the Eraser/Methanol solution for 20 min.
For total protein staining with Ponceau S,
membranes were incubated with 0.1% (w/v)
Ponceau S in 5% acetic acid for 5 min and
washed 2 × 5 min in 10% acetic acid, followed
by washes of 5 min in 100% methanol and 5
min in a 70:30:4 volume ratio of methanol,
Raw 16-bit TIFs (tagged image files) were
opened in ImageJ (5) and rotated to align
immunoblot bands horizontally in the window. The rectangle tool was then used to
select lanes containing the band of interest.
The width of the lane rectangle was drawn
as wide as possible without causing overlap
with bands from adjacent lanes. The height
of the lane rectangle was drawn long enough
to get a sample of the local background surrounding the band of interest. Lane profiles
were plotted with the gel analysis plug-in,
and background was subtracted by connecting the background intensity profiles to the
left (top) and the right (bottom) of the band
of interest by using the line tool. Last, the
magic-wand tool was used to calculate the
integrated area within the band profile of interest and obtain the final raw densitometry
value.
Recombinant protein purification
Recombinant GST-ERK2 and GST-p38 were
prepared by glutathione affinity chromatography in RIPL cells (Stratagene) as described
(22).
Statistical analysis
Serial dilutions were fit to linear or hyperbolic
models by least-squares regression in IGOR
Pro (WaveMetrics). The χ2 statistic from each
model fit was used together with the number
of data points and the number of fitted
parameters (one for the linear model and two
for the hypergeometric model) to calculate
an F statistic that compares goodness of fit
between models. To correct for multiplehypothesis testing, FDRs were calculated
according to Benjamini and Hochberg (47).
REFERENCES AND NOTES
1. M. R. Green, J. Sambrook, Molecular Cloning:
A Laboratory Manual (Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, NY, ed.
4, 2012), p. 1890.
2. S. C. Taylor, A. Posch, Biomed. Res. Int. 2014,
361590 (2014).
3. S. C. Taylor, T. Berkelman, G. Yadav, M.
Hammond, Mol. Biotechnol. 55, 217–226
(2013).
4. H. Towbin, T. Staehelin, J. Gordon, Proc. Natl.
Acad. Sci. U.S.A. 76, 4350–4354 (1979).
5. C. A. Schneider, W. S. Rasband, K. W. Eliceiri,
Methods 9, 671–675 (2012).
6. L. C. Ngoka, Proteome Sci. 6, 30 (2008).
7. U. K. Laemmli, Nature 227, 680–685 (1970).
8. K. A. Janes, C. C. Wang, K. J. Holmberg, K.
Cabral, J. S. Brugge, Nat. Methods 7, 311–317
(2010).
9. K. A. Janes, J. G. Albeck, S. Gaudet, P. K.
Sorger, D. A. Lauffenburger, M. B. Yaffe, Science
310, 1646–1653 (2005).
10. S. Gaudet, K. A. Janes, J. G. Albeck, E. A. Pace,
D. A. Lauffenburger, P. K. Sorger, Mol. Cell.
Proteomics 4, 1569–1590 (2005).
11. K. A. Janes, S. Gaudet, J. G. Albeck, U. B.
Nielsen, D. A. Lauffenburger, P. K. Sorger, Cell
124, 1225–1239 (2006).
12. C. C. Wang,S. S. Bajikar, L. Jamal, K. A. Atkins,
K. A. Janes, Nat. Cell Biol. 16, 345–356 (2014).
13. C. MacKintosh, K. A. Beattie, S. Klumpp, P.
Cohen, G. A. Codd, FEBS Lett. 264, 187–192
(1990).
14. G. Huyer, S. Liu, J. Kelly, J. Moffat, P. Payette,
B. Kennedy, G. Tsaprailis, M. J. Gresser, C.
Ramachandran, J. Biol. Chem. 272, 843–851
(1997).
15. M. M. Davidson, C. Nesti, L. Palenzuela,W. F.
Walker, E.Hernandez, L. Protas, M. Hirano,
N. D. Isaac J. Mol. Cell. Cardiol. 39, 133–147
(2005).
16. C. Haan, I. Behrmann, J. Immunol. Methods
318, 11–19 (2007).
17. E. Marcus, Cell 159, 965–966 (2014).
18. B. J. Olson, J. Markwell, Curr. Protoc. Protein
Sci. Chapter 3, Unit 3.4 (2007).
19. P. K. Smith, R. I. Krohn, G. T. Hermanson, A. K.
Mallia, F. H. Gartner, M. D. Provenzano, E. K.
Fujimoto, N. M. Goeke, B. J. Olson, D. C. Klenk,
Anal. Biochem. 150, 76–85 (1985).
20. K. J. Jensen, F. S. Garmaroudi, J. Zhang, J. Lin,
S. Boroomand, M. Zhang, Z. Luo, D. Yang, H.
Luo, B. M. McManus, K. A. Janes, Cell Host
Microbe 13, 67–76 (2013).
21. B. H. Kang, K. J. Jensen, J. A. Hatch, K. A. Janes
Sci. Signal. 6, rs13 (2013).
22. A. K. Bose, K. A. Janes, Mol. Cell. Proteomics
12, 797–806 (2013).
23. J. P. Brody, B. A. Williams, B. J. Wold, S. R.
Quake, Proc. Natl. Acad. Sci. U.S.A. 99, 12975–
12978 (2002).
24. C. Tricarico, P. Pinzani, S. Bianchi, M.
Paglierani, V. Distante, M. Pazzagli, S. A. Bustin,
C. Orlando, Anal. Biochem. 309, 293–300
(2002).
25. J. Vandesompele, K. De Preter, F. Pattyn,
B. Poppe, N. Van Roy, A. De Paepe, F.
Speleman, Genome Biol. 3, research0034–
research0034.11 (2002).
26. C. Alarcon, A. I. Zaromytidou, Q. Xi, S. Gao,
J. Yu, S. Fujisawa, A. Barlas, A. N. Miller, K.
Manova-Todorova, M. J. Macias, G. Sapkota, D.
Pan, Cell 139, 757–769 (2009).
27. G. I. Shapiro, Clin. Cancer Res. 10, 4270s–
4275s (2004).
28. O. Salinovich, R. C. Montelaro, Anal. Biochem.
156, 341–347 (1986).
29. G. Taborsky, Adv. Protein Chem. 28, 1–210
(1974).
30. S. Gallagher, S. E. Winston, S. A. Fuller, J. G.
Hurrell, Curr. Protoc. Immunol. Chapter 8, Unit
8.10 (2001).
31. S. H. Kaufmann, C. M. Ewing, J. H. Shaper,
Anal. Biochem. 161, 89–95 (1987).
32. O. E. Sturm, R. Orton, J. Grindlay, M. Birtwistle,
V. Vyshemirsky, D. Gilbert, M. Calder, A. Pitt,
B. Kholodenko, W. Kolch, Sci. Signal. 3, ra90
(2010).
33. B. Schoeberl, C. Eichler-Jonsson, E. D. Gilles, G.
Müller, Nat. Biotechnol. 20, 370–375 (2002).
34. S. Luo, N. B. Wehr, R. L. Levine, Anal. Biochem.
350, 233–238 (2006).
35. C. Kiel, E. Verschueren, J. S. Yang, L. Serrano,
Sci. Signal. 6, ra109 (2013).
36. K. A. Janes, D. A. Lauffenburger, Curr. Opin.
Chem. Biol. 10, 73–80 (2006).
37. K. A. Janes, D. A. Lauffenburger, J. Cell Sci. 126,
1913–1921 (2013).
38. K. A. Janes, J. G. Albeck, L. X. Peng, P. K. Sorger,
D. A. Lauffenburger, M. B. Yaffe, Mol. Cell.
Proteomics 2, 463–473 (2003).
39. M. D. Shults, K. A. Janes, D. A. Lauffenburger, B.
Imperiali, Nat. Methods 2, 277–284 (2005).
40. L. Wang, K. A. Janes, Nat. Protoc. 8, 282–301
(2013).
41. J. G. English, J. P. Shellhammer, M. Malahe, P.
C. McCarter, T. C. Elston, H. G. Dohlman, Sci.
Signal. 8, ra5 (2015).
42. O. Stemmann, H. Zou, S. A. Gerber, S. P. Gygi,
M. W. Kirschner, Cell 107, 715–726 (2001).
43. Y. H. Hsu, J. Yao, L. C. Chan, T. J. Wu, J. L. Hsu,
Y. F. Fang, Y. Wei, Y. Wu, W. C. Huang, C. L. Liu,
Y. C. Chang, M. Y. Wang, C. W. Li, J. Shen, M. K.
Chen, A. A. Sahin, A. Sood, G. B. Mills, D. Yu,
44. 45. 46. 47. 48. G. N. Hortobagyi, M. C. Hung, Cancer Res. 74,
4822–4835 (2014).
M. Gassmann, B. Grenacher, B. Rohde, J. Vogel,
Electrophoresis 30, 1845–1855 (2009).
S. S. Bajikar, C. Fuchs, A. Roller, F. J. Theis, K. A.
Janes, Proc. Natl. Acad. Sci. U.S.A. 111, E626–
E635 (2014).
L. Wang, J. S. Brugge, K. A. Janes, Proc. Natl.
Acad. Sci. U.S.A. 108, E803–E812 (2011).
Y. Benjamini, Y. Hochberg, J. R. Statist. Soc. B
57, 289–300 (1995).
B. C. Trauth, C. Klas, A. M. Peters, S. Matzku, P.
Möller, W. Falk, K. M. Debatin, P. H. Krammer,
Science 245, 301–305 (1989).
ACKNOWLEDGMENTS
I thank D. Gioeli, M. Weber, and D. Brautigan for
comments on this manuscript; Z. Chitforoushzadeh
and S. Bajikar for lysate preparations; M. Shah for
recombinant proteins; T. Allen and C. Smolko for
MemCode Reversible Protein Stain; S. Gaudet for
immunoblotting anecdotes; and C. Borgman for
copyediting the manuscript. Funding: This work
was supported by the NIH (#1-R21-AI105970),
the American Cancer Society (#120668-RSG11-047-01-DMC), the Pew Charitable Trusts
(#2008-000410-006), and The David and Lucile
Packard Foundation (#2009-34710). Author
contributions: K.A.J. conceived the work,
performed all immunoblots and quantitative
analyses, and wrote the paper. Competing
interests: K.A.J. is a paid consultant for Cell
Signaling Technologies.
Submitted 29 September 2014
Accepted 16 March 2015
Final Publication 7 April 2015
10.1126/scisignal.2005966
Citation: K. A. Janes, An analysis of critical factors
for quantitative immunoblotting. Sci. Signal. 8, rs2
(2015).
SUPPLEMENTARY MATERIALS
www.sciencesignaling.org/cgi/content/full/8/371/
rs2/DC1
Fig. S1. Challenges with using total protein stains
for normalization of quantitative immunoblots.
File S1. Raw 16-bit images and densitometry
calculations from Fig. 4.
21
A neuronal antigen in the brains of
Alzheimer patients
Benjamin L. Wolozin, Alex Pruchnicki, Dennis W. Dickson, Peter Davies
A monoclonal antibody was prepared against pooled homogenates of brain tissue from
patients with Alzheimer’s disease. This antibody recognizes an antigen present in much
higher concentration in certain brain regions of Alzheimer patients than in normal brain.
The antigen appears to be a protein present in neurons involved in the formation of
neuritic plaques and neurofibrillary tangles, and in some morphologically normal neurons
in sections from Alzheimer brains. Partial purification and Western blot analysis revealed
the antigen from Alzheimer brain to be a single protein with a molecular weight of 68,000.
Application of the same purification procedure to normal brain tissue results in the detection of small amounts of a protein of lower molecular weight.
A
lzheimer’s disease is a neurodegenerative disorder characterized clinically
by progressive loss of intellectual function. This impairment of function appears to be correlated with numbers of
neuritic plaques in the neocortex and with
loss of presynaptic markers of cholinergic
neurons (1). Neuritic plaques are composed
of degenerating axons and nerve terminals,
often surrounding an amyloid core and usually containing reactive glial elements (2).
Another characteristic pathologic feature
of Alzheimer’s disease, the neurofibrillary
tangle, is an intraneuronal mass composed
of normal intermediate filaments and paired
helical filaments (PHF) with unusual properties (3).
In studying the topographic distribution
of plaques and tangles in the brains of Alzheimer patients, we noted that the lesions
occur with high frequency in regions receiving cholinergic innervation from the ventral
forebrain (4). This cholinergic cell group
appears to be extremely vulnerable to the
disease process, and evidence that cholinergic nerve terminals participate in plaque
formation has been presented (5). To better
define the relation between ventral forebrain
cholinergic neurons and the lesions of the
Alzheimer brain, we have prepared monodonal antibodies to homogenates of ventral
forebrain tissue taken at autopsy from four
patients with Alzheimer’s disease. The resulting antibodies were screened on the basis
of their ability to differentiate brain tissue
from patients with Alzheimer’s disease and
from normal subjects in both immunochemical and immunocytochemical procedures.
Antibodies were initially assayed according to their ability to bind to brain homogenate that had been immobilized onto
polyvinyl plates (1, µg per 50-mm diameter
well) by drying at 37°C for 1 hour. Antibody
binding was detected with peroxidase-conjuDepartments of Pathology and Neuroscience, Albert Einstein
College of Medicine, 1300 Morris Park Avenue,
Bronx, NY 10461
22
gated goat antibody to mouse immunoglobulins. Those antibodies that showed greater
than a 50% increase or decrease in binding
to homogenates of Alzheimer brain relative
to normal tissue were studied further. One of
these antibodies, Alz-50, is described below.
Initial assays showed that the binding of
Alz-50 was highly selective for brain tissue
from Alzheimer patients. Figure 1 shows that
0.33 µg of temporal cortex homogenate from
Alzheimer patients gave an optical density
only slightly lower than 10 µg of temporal
cortex homogenate from normal patients.
From these data we conclude that the antigen is elevated 15 to 30 times in the temporal
cortices of the Alzheimer patients. Alz-50 reactivity was similarly elevated in the nucleus
basalis and hippocampus. These areas, cortex, nucleus basalis, and hippocampus, are
all known to contain neuritic plaques and
neurofibrillary tangles in brains of patients
with Alzheimer’s disease. Brain areas less affected by the disease, such as caudate, thalamus, or cerebellum demonstrated little or no
reactivity.
The immunocytochemistry of Alz-50 on
Formalin-fixed tissue was dramatic and was
consistent with the hypothesis that Alz-50
was highly selective for neuronal components in Alzheimer brain (Fig. 2A). Many
neurons in the pyramidal layer of the hippocampus were stained. The antigen was present in cell bodies and in neurites. Figure 2C
shows that plaques were strongly stained by
the antibody. The staining was confined to
the neuritic meshwork present in plaques.
Darkly stained neurons and plaques were
found throughout Alzheimer hippocampus
and cortex. By contrast, there was virtually
no staining of normal brain (Fig. 2B). This
pattern of specificity was observed in a total
of eight brains from Alzheimer patients and
five brains from normal subjects.
To determine the relation between Alz-50
staining of neurons and the presence of neurofibrillary tangles we used a double staining
technique. Vibratome sections of Formalinfixed Alzheimer tissue were reacted with AlzOriginally published 2 May 1986 in SCIENCE
Fig. 1. Quantitation of Alz-50 reactivity in
temporal cortex of patients that had died of
Alzheimer’s disease and normal individuals.
Alzheimer’s disease cases were typical in both
clinical and neuropathologic features (3). Brains
were obtained from normal individuals dying in
hospital from lung or heart disease. These patients were not demented prior to death and had
no history of neurologic or psychiatric disease;
neuropathologic studies failed to reveal any significant pathology. The reactivity in the brains of
the Alzheimer patients is estimated to be 15 to
30 times greater than in the brains of the normal
subjects. Various amounts of antigen (x axis)
was dried onto 90-well polyvinyl plates (NUNC,
Germany). Nonspecific binding of protein to the
antigen was blocked by incubating the plates
with 0.01M TBS, pH 7.4, plus 5% dried milk for 1
hour. Alz-50 was diluted 1:5 in blocking solution
and incubated overnight at 4°C. Unbound antibody was removed by washing five times with
0.02% Tween-TBS. Peroxidase-coupled goat antibody to mouse immunoglobulins (Kirkegaard
& Perry) was diluted 1:100 in blocking solution,
added to the plates, and incubated for 1 hour
at room temperature. After five washes with
0.02% Tween-TBS, reactivity was visualized with
2,2’-azino-di’-3-ethyl-benzthiazoline solution
(ABTS) (Kirkegaard & Perry). Results are expressed as means, with standard deviations indicated. Numbers of cases were ten for Alzheimer
and six for normal.
50, and reactivity was visualized with the use
of peroxidase conjugated goat antibody to
mouse immunoglobulins. 4-Chloronaphthol
was used to visualize the peroxidase reaction; this compound, a peroxidase substrate,
yields a product that precipitates in aqueous
solution but is soluble in organic solvents.
The tissue section was photographed and the
4-chloronaphthol was removed by dehydration and xylene treatment. Finally, plaques
and tangles were stained with thioliavine S,
a sensitive histologic reagent for the demonstration of these lesions, and the section was
photographed again.
Comparison of the staining patterns (Fig.
3) revealed that many neurons were stained
both by the antibody and by thioflavine
S. However, several neurons were darkly
stained by the antibody and did not appear to
contain neurofibrillary tangles. A small fraction of neurons that contained tangles and
were thioflavine-positive were
in native form on a Sepharose
not positive for Alz-50 (Fig. 3, A
6B column. Temporal cortex
and B). The staining of plaques
homogenate from Alzheimer
by Alz-50 was also studied by
patients or normal subjects was
this method. All plaques bound
centrifuged at 27,200g for 20
both Alz-50 and thioflavine.
minutes, and the supernatant
In addition, these results conwas run through the Sepharose
firmed the neuritic nature of
6B column. An immunoreacthe antibody staining; staining
tivity profile was obtained by
was present in the neuritic pediluting each fraction 1:100 in
riphery of the plaques but abwater, drying 50 µl of diluted
sent in the amyloid core.
fraction onto polyvinyl plates,
Simple biochemical experiand assaying by ELISA. The
ments suggest that the Alz-50
profile from Alzheimer brain
Fig. 2. (A) Immunocytochemistry of Alz-50 staining in the pyramidal
antigen is distinct from PHF,
revealed a single immunoreaclayer of hippocampus from a brain affected by Alzheimer's disease.
the major tangle component.
tive peak at the void volume.
Vibratome sections (40 µm) were cut from Formalin-fixed brain. The tisUnlike PHF, the Alz-50 antiBy contrast, no immunoreacsue was washed twice in TBS, incubated for 30 minutes in 0.25% Triton Xgen is largely soluble in 0.01M
tivity was seen in the column
100-TBS, washed once in TBS, and incubated for 30 minutes in dried milk
tris-buffered saline (TBS) and
fractionation of normal brain
(blocking solution) for 1 hour at room temperature to prevent nonspecific
completely soluble in TBS conat this dilution. At a lesser dibinding of antibody. Alz-50 was diluted 1:5 in blocking solution and incutaining 5% sodium dodecyl
lution, normal brain immunobated with the tissue sections overnight at 4°C. Unbound antibody was
sulfate (TBS-SDS) (6). Solubilreactivity was seen at the void
removed by washing in TBS. Peroxidase coupled goat antibody to mouse
ity was tested by vortexing
volume. This result supports
immunoglobulin G was diluted 1:100 in blocking solution and incubated with
Alzheimer cortex homogenate
the quantitative data shown in
the tissue sections for 1 hour. Antibody was visualized by incubating the secfor 2 minutes in TBS or TBSFig. 1 demonstrating 15 to 30
tions for 8 minutes in a 0.1M tris solution, pH 7.4, containing 0.45 mg/ml
SDS. The homogenate was then
times as much inimunoreactivof diaminobenzidine and 0.44 mM hydrogen peroxide. The tissue was then
centrifuged at 10,000g for 10
ity in the Alzheimer brain as in
washed in TBS, dehydrated, and mounted. Magnification x 12.5. (B) Alz-50
minutes. Supernatant and pelthe normal brain.
staining in hippocampus from a normal brain. (C) A plaque from frontal corlet were separated, and the
Samples of supernatant from
tex of a brain affected with Alzheimer’s disease. Magnification x25.
pellet was washed twice by
Alzheimer and normal corhomogenizing and centrifugtex were then studied by the
ing as above. The supernatant
Western blot technique (Fig. 4).
and pellet were homogenized
The major band in Fig. 4, lane
in water and various amounts
A, has an apparent molecular
of each sample (10, 3, 1 µg per
weight of 68,000 (68K). In Fig.
50, µl) were dried onto poly4, lane B, the reactivity from
vinyl plates. The presence of
the normal void volume fracantigen was determined by an
tion was brought out by loadenzyme-linked immunosorbent
ing onto the polyacrylamide gel
assay (ELISA). PHF reactivity
three times as much protein as
was monitored by means of
was used for the Western blot
an antibody to PHF, antibody
of the Alzheimer void volume
704.1 (10). Alz-50 reactivity was
Fig. 3. (A) Immunocytochemistry of Alz-50 in hippocampus from a brain
fraction. The major band has
found in the TBS supernatant
affected by Alzheimer’s disease. Arrows point to neurons that react with
an apparent molecular weight
and was quantitatively recovAlz-50 (A) but do not stain with thioflavine S (B). For details on the immunoof 59K. A doublet at 245K is
ered in the TBS-SDS supernacytochemistry see legend to Fig. 1. The peroxidase reaction was developed
also present.
tant, whereas PHF reactivity
by using 0.2 mg/ml of 4-chloronaphthol instead of diaminobenzidine (as in
These results show that in
remained in the pellet after
Fig. 2). Microscopy was performed by using phosphate buffered saline (PBS)TBS the antigen recognized
TBS-SDS extraction. Thus, Alzglycerol instead of dehydration. Magnification x 12.5. (B) Thioflavine S histoby Alz-50 is either aggregated
50 immunoreactivity is soluble
chemistry of the same section from (A). Note the absence of thioflavine S
or is part of a large complex.
and segregates away from PHF
staining of neurons (arrows). To remove the 4-chloronaphthol the section was
When dissociated, the epitope
immunoreactivity.
occurred as a single 68K prodehydrated, whereupon the 4-chloronaphthol dissolved in the xylene, and
Enzyme experiments contein that was distinctly differthen subsequently rehydrated. The section was then incubated in a 0.01%
firmed this result. Before we
ent from the 59K antigen from
thioflavine S solution in Formalin and rinsed by dipping three times in 80%
dried the Alzheimer brain
normal brain. Several relevant
ethanol (fluorescence microscopy; magnification x 12.5.)
homogenates onto polyvinyl
proteins have subunit molecuplates, we treated them for 0,
lar weights in the 68K range: in
20, or 60 minutes with trypsin or alkaline
lated, although it is of interest because many
the cytoskeletal family there are neurofilaphosphatase. An ELISA was used to measure
antibodies to neurofilament epitopes that
ment and tau proteins (7), in the cholinergic
the sensitivity of the Alz-50 epitopes to the
identify neurofibrillary tangles fail to react
family there is choline acetyltransferase (8).
treatment. Unlike the PHF, the Alz-50 epitafter phosphatase treatment. In addition,
The Alz-50 antigen is therefore unlikely to be
ope was highly sensitive to trypsin. This supthis result suggests that phosphorylation of
neurofilament, tau protein, or choline acetylports the hypothesis that the Alz-50 antigen
the epitope does not account for the ability
transferase, because none of these proteins
is not PHF, and also demonstrates that the
of the antibody to distinguish between Alseems to be elevated in concentrations 15 to
antigen is a protein. The Alz-50 epitope is not
zheimer and normal brain.
30 times in the brains of Alzheimer patients.
phosphatase sensitive; this does not prove
A Western blot of the Alz-50 antigen was
Further, if the 68K Alzheimer type prothat the Alz-50 antigen is not phosphoryobtained by first purifying the Alz-50 antigen
tein is related to the 59K protein, then the
23
Fig. 4. Western blot of (A) Alzheimer and
(B) normal temporal cortex fractions from
the void volume of a Sepharose 6B column;
(C) prestained molecular weight standards
(Bethesda Research Laboratories). Temporal cortex (2.5 g) was homogenized into 10 ml of
PBS and centrifuged at 20,000g for 20 minutes.
The supernatant was run through a Sepharose
6B column and the void volume was collected.
Protein determinations were performed and
the samples were run on a 10% SDS-PAGE gel.
Fifteen micrograms of protein was in the Alzheimer sample (A) and 45 µg of protein was in
the normal sample (B). The protein was transferred to nitrocellulose for 3 hours at 125 mA;
the buffer contained 19.2 mM glycine, 2.5 mM
trizma base, and 20% methanol at pH 8.3. The
antibody reaction was developed as in Fig. 2, except that phosphatase-coupled antibody was
used instead of peroxidase-coupled antibody.
Color development was achieved by means of
BCIP/NBT (Kirkegaard & Perry). Omission of
the incubation with Alz-50 completely abolished staining.
Alz-50 antigen is unlikely to be neurofilament, tau protein, or choline acetyltrarsferase. The relation among the proteins
detected is questionable: the relatively large
amounts of samples of normal brain tissue
needed to detect any reactivity raises doubts
about the specificity of the antibody binding.
Some monoclonal antibodies to tangles react
with proteins in the 59K to 68K range (9), but
none of these show quantitative differences
between normal and Alzheimer brain. The
points raised above suggest that the epitope
recognized by Alz-50 is a novel antigen.
Our results demonstrate that Alz-50 recognizes a protein present in neuronal ter-
24
1. D. M. Bowen, C. B. Smith, P. White, A. N.
Davison, Brain 99, 459 (1976); P. Davies,
Brain Res. 171, 319 (1979); P. Davies and A. J.
F. Maloney, Lancet 1976-II, 1403 (1976); E. K.
Perry, P. H. Gibson, G. Blessed, R. H. Perry,
B. G. Tomlinson, J. Neurol. Sci. 34, 247
(1977); G. Blessed, B. G. Tomlinson, M. Roth,
Br. J. Psychiatry 114, 797 (1968).
2. R. D. Terry, N. K. Gonatas, M. Weiss, Am. J.
Pathol. 44, 269 (1964).
3. H. M. Wiseniewski, R. D. Terry, A. Hirano, J.
Neuropathol. Exp. Neurol. 29, 163 (1970).
4. M. M. Mesulam, E. J. Mufson, A. L. Levey, B.
H. Wainer, J. Comp. Neurol. 214, 170 (1983).
5. R. G. Struble, L. C. Cork, P. Whitehouse, D. L.
Price, Science 216, 413 (1982); D. L. Price et
al., Neurosci. Comment. 1, 84 (1982).
6. D. J. Selkoe, Y. Ihara, F. J. Salazar, Science
215, 1243 (1982).
7. F. C. Chiu, W. T. Norton, K. L. Fields, J.
Neurochem. 37, 147 (1981).
8. A. J. Levey, D. B. Rye, B. H. Wainer, ibid. 39,
1652 (1982).
9. D. W. Dickson, Y. Kress, A. Crowe, S.-H. Yen,
Am. J. Pathol. 120, 292 (1985).
minals in plaques and in most neurons with
tangles. The surprising finding is that Alz50 immunoreactivity appears to precede the
deposition of neurofibrils and PHFs to form
tangles. Thus Alz-50 may recognize a precursor to tangle formation. Alternatively, Alz-50
immunoreactive neurons may be affected
by Alzheimer pathophysiology but may not
form classical tangles. The biochemical data
showing that Alz-50 is not PHF antigen is
consistent with this discordance between
the presence of Alz-50 and the presence of
neurofibrillary tangle. The function and
identity of the proteins that are recognized
by Alz-50 remains to be elucidated.
ACKNOWLEDGMENTS
Supported by NIH training grant T32 GM7288
from the National Institute of General Medical
Sciences, The Mcknight Foundation, The
Joyce Mertz-Gilmore Foundation, and the
Commonwealth Fund. We thank S.-H. Yen for
antibody 704.1 to PHF and M. Scharff and E.
Fischberg for advice and assistance. We also
acknowledge use of the hybridoma facility of the
cancer center (CA 13330) at the Albert Einstein
College of Medicine.
30 October 1985; accepted 18 February 1986
Serological analysis of a subgroup of
human T-lymphotropic retroviruses
(HTLV-III) associated with AIDS
Jörg Schüpbach,1 Mikulas Popovic,1 Raymond V. Gilden,2 Matthew A. Gonda,2
M. G. Sarngadharan,3 and Robert C. Gallo1
The two main subgroups of the family of human T-lymphotropic retroviruses (HTLV) that
have previously been characterized are known as HTLV-I and HTLV-II. Both are associated with certain human leukemias and lymphomas. Cell surface antigens (p61 and p65)
encoded by HTLV-I are frequently recognized, at low titers, by antibodies in the serum of
patients with acquired immunodeficiency syndrome (AIDS) or with signs or symptoms
that precede AIDS (pre-AIDS). This suggests an involvement of HTLV in these disorders. Another subgroup of HTLV, designated HTLV-III, has now been isolated from many
patients with AIDS and pre-AIDS. In the studies described in this report, virus-associated
antigens in T-cell clones permanently producing HTLV-III were subjected to biochemical
and immunological analyses. Antigens of HTLV-III, specifically detected by antibodies
in serum from AIDS or pre-AIDS patients and revealed by the Western blot technique,
are similar in size to those found in other subgroups of HTLV. They include at least three
serologically unrelated antigenic groups, one of which is associated with group-specific
antigens (p55 and p24) and another with envelope-related (p65) proteins, while the antigens in the third group are of unknown affiliation. The data show that HTLV-III is clearly
distinguishable from HTLV-I and HTLV-II but is also significantly related to both viruses.
HTLV-III is thus a true member of the HTLV family.
M
embers of the family of human lymphotropic retroviruses (HTLV) have
the following features in common: a
pronounced tropism for OKT4+ lymphocytes (1), a reverse transcriptase
(RT) with a high molecular weight (100,000)
and a preference for Mg2+ as the divalent cation for optimal enzymatic activity (2, 3), and
the capacity to inhibit T cell function (4) or,
in some cases, kill T cells (5). Many HTLV
also have the capacity to transform infected
T cells (1). The two major subgroups that have
been characterized (6) are HTLV-I, which is
causatively linked to certain adult T-cell malignancies (7), and HTLV-II, which was first
identified in a patient with hairy cell leukemia (8).
Viruses of the HTLV family have been
detected in some patients with the acquired
immunodeficiency syndrome (AIDS) (9)
or with pre-AIDS, a condition frequently
progressing to AIDS (10). A high proportion
of patients with AIDS or pre-AIDS, as well as
a significant number of hemophiliacs, have
antibodies in their serum that recognize
a cell surface glycoprotein (gp6l) that is
present on certain human T cells infected
with HTLV-I (11). Gp6l and p65, a slightly
larger protein that is a homolog of gp6l and
occurs in another cell line producing HTLV-I,
were subsequently shown to be related to
the HTLV viral glycoprotein (12, 13). Studies
1
Laboratory of Tumor Cell Biology, National Cancer Institute,
Bethesda, Maryland 20205.
2
Program Resources, Inc. NCI-Frederick Cancer Research Facility,
Frederick, Maryland 21701.
3
Department of Cell Biology Litton Bionetics, Inc., Kensington,
Maryland 20895.
of blood transfusion recipients who later
developed AIDS and of their blood donors
have revealed the presence, in the blood of
the donors, of antibodies to a retrovirus of
the HTLV family (14). These findings suggest
an involvement of viruses of the HTLV
family in the cause of AIDS and pre-AIDS.
An involvement of HTLV-I alone appeared
doubtful, however, because antibody titers
to gp6l of HTLV-I in these patients are
generally very low and antibodies to the
structural proteins of HTLV, notably p24
and p19 (15), are not detectable in most AIDS
patients (16). Instead, it seemed likely that
another member of the HTLV family might
be involved in the etiology of AIDS. Here we
describe our studies of a group of cytopathic
yiruses (collectively designated HTLV-III)
isolated from patients with AIDS or preAIDS. Isolation of these viruses was achieved
by means of a novel system permitting the
continuous growth of T-cell clones infected
with the cytopathic types of HTLV found in
these disorders (17). We show that antigens
associated with human cells infected by
HTLV-IlI are specifically recognized by
antibodies in serum from AIDS and preAIDS patients, and present a preliminary
biochemical and immunological analysis of
these antigens.
Lysates of two immortalized and infected
human T-cell clones, H4/HTLV-III and H17/
HTLV-III (17), were tested with samples of
human serum in a strip radioimmunoassay
(RIA) based on the Western blot techniq1ue
(18). The sera were from patients with
AIDS or pre-AIDS, from contacts of such
patients, and from homo- or heterosexual
Originally published 4 May 1984 in SCIENCE
male controls. Sera from the same patients
were also tested by the enzyme-linked
immunosorbent assay (ELISA) with purified
HTLV-III as part of a larger, systematic
serologic study of the prevalence of
antibodies to HTLV-III in AIDS and preAIDS patients (19).
Representative results are shown in Fig.
1. Sera from patients with AIDS or preAIDS, and from some homosexuals and
heroin-addicts, recognized a number of
specific antigens not detected by sera from
heterosexual subjects. The most prominent
reactions were with antigens of the following
molecular weights: 65,000, 60,000, 55,000,
41,000, and 24,000. Antigens with molecular
weights of approximately 88,000, 80,000,
39,000, 32,000, 28,000, and 21,000 gave less
prominent reactions. The reaction with the
antigen of 55,000 (p55) only occurred in
sera that also recognized p24, suggesting a
relationship between the two antigens.
The specificity of these reactions was
studied by comparing lysates of H4/HTLVIII and HI7/HTLV-III with lysates of the
same cell clones, H4 and H17, before viral
infection (Fig. 2A). No antigen from the
uninfected clones reacted with the sera,
with the exception of a protein with a
molecular weight of 80,000 in HI7 which
bound antibodies from all of the human
serum samples tested (see Fig. 1B) but not
from rabbit or goat serum. Antigens newly
expressed after viral infection and recognized
by the human serum used for this analysis
included p65, p55, p41, p39, p32, and p24.
A large protein with a molecular weight
of approximately 130,000 and a protein of
48,000 were also detected. With this serum,
p55 consistently appeared as a doublet of
bands of similar intensity. With normal
human serum, none of the antigens was
detected (not shown). These results show
clearly that the antigens detected after virus
infection are either virus-coded proteins or
cellular antigens specifically induced by the
infection.
The antigens of H4/HTLV-III were also
compared with antigens from virus purified
from the culture fluids of H4/HTLV-III (Fig.
2B). Extensive accumulation of p24 and p41
[see (20)] occurred in the virus preparation
(Fig. 2B, panels I and II). Protein stains
showed that these molecules are the major
components of the virus preparation (19).
P24 and p41 may therefore be considered
viral structural proteins. Furthermore,
an antigen with a molecular weight of
approximately 110,000 was detected in the
virus preparation but was below limit of
detection in the cells. Also, p39 [see (20)]
was present in the virus preparation. It is
interesting that p24 in the virus preparation
consistently appeared as a doublet (p24/
p23), whereas in the cells it appeared as
p24 alone. The significance of this is under
investigation. P55 was not detected in the
25
Fig. 1. Serologic detection of antigens in HTLV-III producer cell
clones. Strip RIA were performed with human serum as described
elsewhere in detail (21). Briefly, lysates of HTLV-IlI producer cell
clones were subjected to electrophoresis under reducing conditions
on preparative sodium dodecyl sulfate (SDS)-polyacrylamide slab
gels, and electroblotted to nitrocellulose sheets (18). The sheets were
cut into strips. These were incubated with human serum diluted 1:100.
After three thorough washings, bound antibodies of immunoglobulin
G (lgG) and immunoglobulin M (lgM) classes were made visible with
radiolabeled, affinity-purified goat antiserum to human IgG and IgM
(H-chain specific) and autoradiography. (A) Analysis with H4/HTLVIII cells. (Lanes a, d, and g) U.S. patients with AIDS; (lane b) a French
heterosexual male who developed AIDS after receiving a blood
transfusion in Haiti (24); (lane c) an AIDS patient from Switzerland;
(lane e) a normal heterosexual control; (lane f) a French pre-AIDS
patient (24); (lane h) a Swiss heterosexual drug addict; (lane i) a
normal homosexual control. (B) Analysis with H17/HTLV-IlI cells.
(Lane a) An infant with AIDS whose mother is a prostitute; sera from
both are highly positive for antibodies to the HTLV membrane antigen
(11, 25) and in our ELISA with disrupted HTLV-III (19); (lane b) same
serum as in (A), lane d; (lane c) normal heterosexual control; (lane
d) another Swiss AIDS patient; (lane e) a Swiss heterosexual male
intravenous drug abuser with generalized lymphadenopathy and
thrombocytopenic purpura (pre-AIDS).
Fig. 3. Relation between HTLVIII and HTLV-II. Serum of an
AIDS patient at a dilution of 1:
500 was tested in a competition
RIA on strips (20) prepared with
H4/HTLV-III cells. (Lane a) The
human serum was added directly
to the strip (uncompeted control);
(lanes b to e) the serum was first
absorbed for 3 hours at 37°C with
1 mg of cellular extract. In (b) the
absorption was with uninfected
H4 cells (not producing virus); in
(c) the absorption was with H4/
HTLV-IlI cells producing HTLVIII (positive control); in (d) the
absorption was with C3/44 cells
(26) producing HTLV-II; in (e) the
absorption was with HUT 102
cells producing HTLV-I (2).
26
Fig. 2. (A) Specificity of the antigens recognized. Lysates of
cloned cells before and after infection with HTLV-III were analyzed
by the Western blot technique (18) with a 1:500 dilution of the
serum shown in Fig. 1B, lane e. (Lane a) The HI7 clone before and
(lane b) the same clone after infection (HI7/HTLV-IlI); (lane c) the
H4 clone before and (lane d) the same clone after infection (H4/
HTLV-III). All reactive antigens are virus-related with the exception
of that with a molecular weight of 80,000 in HI7 cells; this antigen
binds antibodies from all human sera investigated. Normal human
serum did not bind to any of the virus-related bands (not shown).
(B) Comparison of antigens in (lanes a) cells and (lanes b) virus.
Lysates of H4/HTLV-III (250, µg per lane) or virus purified from
the cell culture fluids (19) (5 µg per lane) were analyzed with
1:500 dilutions of human sera. (Panel I) Same serum as in Fig.
2A; (panel II) serum of a Swiss male homosexual with fatigue and
generalized lymphadenopathy (pre-AIDS); (panel III) serum from
same AIDS patient as in Fig. IB, lane d. An antigen with a molecular
weight of 110,000 and p41, p39, and p24 are enriched in the virus
preparation [see (20)]. The serum in panel III recognized a subset
of the antigens recognized by the sera used in panels I and II.
Fig. 4. Electron microscopy of thin sections of cells producing HTLV-I, -II, and
-III. (Top) HUT 102 cells producing HTLV-I (2). (Middle) Cells from an AIDS patient
(J.P.) producing HTLV-II (24). (Bottom) Cells from a patient [described in (27)]
with pre-AIDS, producing HTLV-III. (Panels a) Virus particles budding from the cell
membrane. (Panels b) Free particles have separated from the membrane. (Panels c)
Free particles sectioned in a different plane, Note the dense, cylindrical core region
of HTLV-III.
virus; however, the intensity of the p55 band
in the cells (Fig. 2B, lanes a) appeared to
correlate with the intensity of p24/p23 in
the virus preparation (Fig. 2B, lanes b), thus
again suggesting a relation between these
antigens. The p55 is probably a precursor of
p24, since a group-specific antigen of similar
size (Pr 54gag ) in HTLV-I-infected cells is the
precursor of p24 and the other gag-coded
proteins (21). Occasionally an additional
set of antigens was recognized by a serum
(Fig. 2B, panel III) but their relation to the
antigens described above is unclear.
Thus we have shown that viral or virusinduced antigens in cloned human T cells
infected with HTLV-III are specifically
recognized by antibodies in the serum
of patients with AIDS or pre-AIDS. The
detection of p65 by many of the serum
samples is of special interest. We have tested
these sera on strips prepared from lysates
of cells producing HTLV-I or -II. Some of
these cells produce a p65 that has been
shown (13) to be coded for by the env gene
of HTLV-I and to be the homolog of the gp61
described by others (11, 12). Many of the sera
recognizing p65 in HTLV-III–infected cells
also recognized, though somewhat faintly,
p65 in cells producing HTLV-I or -II, and
some of them also recognized gag-related
antigens (data not shown).
In addition, the reaction of some human
sera with virus-related antigens of HTLV-III–
infected cells could be partially inhibited by
large amounts of extracts of cells producing
HTLV-II (Fig. 3). When a human serum not
recognizing p65 was used, the antigens for
which there was competition included p55,
p48, p41, p39, and p24. These results were
confirmed by the demonstration that a rabbit
antiserum raised against purified HTLV-III
showed some reactivity with antigens of
HTLV-II and, to a lesser extent, with HTLV-I.
In contrast, antiserum to HTLV-II recognized
both HTLV-I and -III antigens, and an
antiserum to HTLV-I reacted well with
HTLV-II, but only faintly with HTLV-III (22).
Moreover, nucleotide sequences of HTLV-III
have been found to be related to HTLV-I and
-II (23). Although the morphology of HTLVIII particles appears to be somewhat different
from the morphology of HTLV-I and -II (Fig.
4), and although some differences are also
found in the protein patterns of purified virus
preparations (19), these immunological and
nucleic acid data clearly indicate that HTLVIII is a true member of the HTLV family and
that it is more closely related to HTLV-II
than to HTLV-I.
1. M. Popovic, P. S. Sarin. M. Roben-Guroff, V.
S. Kalyanaraman, D. Mann, J. Minowada, R. C.
Gallo, Science 219, 856 (1983); P. D. Markham,
S. Z. Salahuddin, V. S. Kalyanaraman, M.
Popovic, P. Sarin, R. C. Gallo, Int. J. Cancer 31,
413 (1983); S. Z. Salahuddin. P. D. Markham, F.
Wong-Staal, G. Franchini, V. S. Kalyanaraman,
R. C. Gallo, Virology 129, 51 (1983).
2. B. J. Poiesz, F. W. Ruscetti, A. F. Gazdar, P.
A. Bunn, J. D. Minna. R. C. Gallo, Proc. Natl.
Acad. Sci. U.S.A. 77, 7415 (1980).
3. H. M. Rho, B. Poiesz, F. W. Ruscetti. R. C.
Gallo, Virology 112, 355 (1981); M. Seiki, S.
Hattori, Y. Hirayama, M. Yoshida, Proc. Natl.
Acad. Sci. U.S.A. 80, 3618 (1983).
4. M. Popovic et al.. in preparation.
5. H. Mitsuya, H. G. Guo, M. Megson, C. O.
Trainor, M. S. Reitz, S. Broder, Science 223,
1293 (1984).
6. For a brief review, see M. G. Sarngadharan et
al., in Human Carcinogenesis, C. C. Harris and
H. H. Autrup, Eds. (Academic Press, New
York, 1983), p. 679.
7. V. S. Kalyanaraman, M. G. Sarngadharan. Y.
Nakao, Y. Ito. T. Aoki, R. C. Gallo, Proc. Natl.
Acad. Sci. U.S.A. 79, 1653 (1982); M. Robert
Guroff, Y. Nakao, K. Notake, Y. Ito, A. H.
Sliski, R. C. Gallo, Science 215, 925 (1982); Y.
Hinuma et al., Int. J. Cancer 29, 631 (1982);
W. A. Blattner et al., ibid. 30, 257 (1982);
J. Schüpbach, V. S. Kalyanaraman, M. G.
Sarngadharan, Y. Nakao, R. C. Gallo, ibid. 32,
583 (1983); J. Schüpbach, V. S. Kalyanaraman,
M. G. Sarngadharan, W. A. Blattner, R. C. Gallo,
Cancer Res. 43, 886 (1983).
8. V. S. Kalyanaraman et aI., Science 218, 571
(1982); E. P. Gelmann et al., Proc. Natl. Acad.
Sci. U.S.A. 81, 993 (1984).
9. R. C. Gallo et al., Science 220, 865 (1983); E. P.
Gelmann et al., ibid., p. 862.
10. F. Barré-Sinoussi et al., ibid., p. 868.
11. M. Essex et al., ibid., p. 859; M. Essex et al.,
ibid. 221, 1061 (1983).
12. T. H. Lee, J. E. Coligan, T. Homma, M. F.
McLane, N. Tachibana, M. Essex, Proc. Natl.
Acad. Sci. U.S.A., in press.
13. J. Schüpbach, M. G. Sarngadharan, R. C. Gallo,
Science, in press.
14. H. W. Jaffe et al., Science 223, 1309 (1984).
15. V. S. Kalyanaraman, M. G. Sarngadharan, P.
A. Bunn, J. D. Minna, R. C. Gallo, Nature
(London) 294, 271 (1981); V. S. Kalyanaraman,
M. Jarvis-Morar, M. G. Sarngadharan, R. C.
Gallo, Virology 132, 61 (1984).
16. M. Robert-Gurotr et al.; in Cancer Cells, vol. 3,
Human T-Cell Leukemia Viruses, R. C. Gallo
and M. Essex, Eds. (Cold Spring Harbor
Laboratory, Cold Spring Harbor, N.Y., in press),
17. M. Popovic, M. G, Sarngadharan. E. Read, R.
C. Gallo, Science 224, 497 (1984).
18. H. Towbin, T. Staehelin, J. Gordon, Proc. Natl.
Acad. Sci. U.S.A. 76, 4350 (1979).
19. M. G. Sarngadharan, M. Popovic, L. Bruch, J.
Schüpbach, R. C. Gallo, Science 224, 506
(1984).
20. Although in Fig. 2B the p41 in the virus
preparation appears to be larger than the p41
in cells, the two molecules are of the same
size. During application of the lysates to the
gel in another experiment, a small amount of
cellular lysate was spilled into the neighboring
lanes and the cellular p41 moved with the
same velocity as the viral p41. A connecting
band was thus formed between the p41 in
the lane containing the cells and the p41 in
the lane with the virus. The same situation
occurred with p39 in cells and virus.
21. J. Schüpbach, V. S. Kalyanaraman, M. G.
Sarngadharan, R. C. Gallo, in preparation.
22. M. G. Sarngadharan et al., in preparation.
23. S. Arya et al., in preparation.
24. J. B. Brunet et al., Lancet 1983-I, 700 (1983).
25. M. Essex, personal communication.
26. M. Popovic, V. S. Kalyanaraman, D. L. Mann,
E. Richardson, P. S. Sarin, R. C. Gallo, in
Cancer Cells, vol. 3, Human T-Cell Leukemia
Viruses, R. C. Gallo and M. Essex, Eds. (Cold
Spring Harbor Laboratory, Cold Spring Harbor,
N.Y., in press).
27. R. C. Gallo et al., Science 224. 500 (1984).
ACKNOWLEDGMENTS
We thank J. Ahmad for technical assistance and
R. Lüthy and M. Vogt, Division of Infectious
Diseases, Department of Medicine, University
Hospital, Zurich, and O. Haller, Institute for
Immunology and Virology, University of Zurich, Zurich, Switzerland, for making some sera
from AIDS and pre-AIDS patients available and
for providing clinical information. J .S. is a Fogarty
International Fellow of the National Cancer Institute.
30 March 1984; accepted 19 April 1984
27
Cell surface P-glycoprotein associated
with multidrug resistance in
mammalian cell lines
Norbert Kartner,1 John R. Riordan,2 and Victor Ling3
The plasma membranes of hamster, mouse, and human tumor cell lines that display
multiple resistance to drugs were examined by gel electrophoresis and immunoblotting.
In every case, increased expression of a 170,000-dalton surface antigen was found to be
correlated with multidrug resistance. This membrane component is of identical molecular size and shares some immunogenic homology with the previously characterized
P-glycoprotein of colchicine-resistant Chinese hamster ovary cells. This finding may have
application to cancer therapy.
S
election of variants in mammalian cells
that are resistant to specific drugs, such
as Vinca alkaloids, maytansine, colchicine, anthracyclines, actinomycin D,
or bleomycin, is often accompanied by
expression of a complex phenotype of cross
resistance to various unrelated drugs (1–14).
This characteristic is referred to as the multidrug resistance phenotype. The generation
of such variants in tumor cells may be an important mechanism by which neoplasms become resistant to treatment by combination
chemotherapy.
Ontario Cancer Institute, Princess Margaret Hospital, and
Department of Medical Biophysics, University of Toronto, Toronto,
Ontario, M4X 1K9, Canada.
2
Research Institute, Hospital for Sick Children, and Departments
of Biochemistry and Clinical Biochemistry, University of Toronto,
Toronto, Ontario, M5G 1X8.
3
Ontario Cancer Institute, Princess Margaret Hospital, and
Department of Medical Biophysics, University of Toronto.
1
Studies in model systems indicate that
multidrug resistance results from a reduced
cellular accumulation of the drugs involved
(5–19), and changes in the plasma membrane
have been observed (17–23). In the well-characterized colchicine-resistant (CHR) Chinese
hamster ovary (CHO) system, for example,
genetic analyses involving cell-cell hybrids,
drug-sensitive revertants, and DNA-mediated transformants of the CHR phenogenesis
type indicate that multidrug resistance, colchicine resistance, and reduced drug accumulation are the result of the same genetic
alteration (21, 24, 25). Moreover, the expression of a 170,000-dalton plasma membrane
glycoprotein (P-glycoprotein) is invariably
associated with this pleiotropic phenotype
(20–22, 24). The degree of drug resistance is
correlated approximately with the amount of
P-glycoprotein present (20, 22). The objective
of the present study is to determine whether
or not P-glycoprotein expression is also asso-
ciated with the multidrug resistance phenotype observed in other cell systems.
Each of the different mammalian cell lines
examined in this study (Table 1) was originally
selected for resistance to a specific drug, and
in each case a multidrug resistance phenotype typified by cross resistance to unrelated
compounds was observed. Such a phenotype
appears to reflect a membrane-associated
alteration (6, 24, 26). We therefore prepared
membranes from these cell lines component
could not be detected by for analysis of their
components by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)
(22). For CHR CHO cells (lanes b to d in Fig.
1A), a protein having a molecular size of about
170,000 daltons appeared in the plasma membrane with a staining intensity proportional to
the degree of resistance to the drug. As previously characterized, this band is referred to as
the P-glycoprotein (20, 22). Similarly stained
bands of approximately the same molecular
size are seen in the other drug resistant cell
lines that we examined (lanes e, h, j, and l in
Fig. 1A). Such a component could not be detected by protein staining in the corresponding drug-sensitive parental cells (lanes a, g, i,
and k) or drug-sensitive revertant (lane f ).
As a means of further characterizing the
relatedness of the 170,000-dalton membrane
components observed in these drug-resistant
lines, an antiserum to plasma membrane vesicles of the highly colchicine-resistant CHO
line CHRC5 was prepared. The specificity of
this antiserum for the P-glycoprotein was improved by cross absorption with immobilized,
detergent-solubilized plasma membrane
proteins of the parental drug-sensitive line
(21, 27). The absorbed antiserum was then
used to examine the membrane components
of the lines shown in Fig. lA. Each of the
Table 1. Description of cell lines. Cell line CHRA3 was selected from AUXB1; CHRB3 was selected from CHRA3 in a second step; and CHRC5 was selected
from CHRB3 in a third step (6). DNRR51 was selected from AUXBI in two steps (27). 110 is a revertant cell line, selected in a single step from CHRC5 (24).
Abbreviations for drugs are: ACR, acriflavine; ADR, adriamycin; AMD, actinomycin D; CCH, colchicine; CMD, colcemid; CYB, cytochalasin B; DNR, daunorubicin; EME, emetine; ERY erythromycin; ETB, ethidium bromide; GRD, gramicidin; PRO, proflavin; PUR, puromycin; VCR, vincristine; and VLB, vinblastine.
Relative resistance was calculated as the ratio of drug concentration tolerated by the resistant cells to that tolerated equally by the sensitive parent cell.
Tolerance was assessed by relative growth rates except for cell lines 110 and ECHR, whose colony-forming abilities in the presence of drug were compared
with the appropriate parental lines. Since cross resistance is shown only for drugs that have been tested and the results reported, absence of a drug from
the column on cross resistance does not imply lack of cross resistance to that drug. The drugs are listed in order of decreasing relative cross resistance.
28
Originally published 23 September 1983 in SCIENCE
Fig. 1. Electrophoretic and immunochemical
analysis of plasma membrane components.
Cell culture, plasma membrane isolation, and
electrophoresis were carried out as described
previously (21, 22, 27). (A) Gels were loaded
with 10 µg of membrane protein per lane. They
were overstained by the silver staining method
of Switzer et al. (34) and were then reduced to
acceptable intensity by soaking in a solution of
0.03M ferric ammonium sulfate in 0. 18M sulfuric acid diluted 1:10, followed by a final rinse in
0.3M sodium carbonate and thorough washing
in water. Molecular size standards (Bio-Rad)
are shown in the first lane. These consist of 0.1
µg each of myosin, (200,000), β-galactosidase
(116,250), phosphorylase b (92,500), bovine
serum albumin (66,200), ovalbumin (45,500),
carbonic anhydrase (31,000), soybean trypsin inhibitor (21,500), and lysozyme (14,400).
(Lane a) Drug-sensitive parent CHO cell,
AUXB1; (lane b) colchicine (0.10 µg/ml)-resistant
CHO cell CHRA3; (lane c) colchicine (3.0 µg/ml)resistant CHO cell CHRB3; (lane d) colchicine
(10 µg/ml)-resistant CHO cell CHRC5; (lane e)
daunorubicin-resistant CHO cell DNRR51; (lane
f) colchicine-revertant CHO cell I10 (derived
from CHRC5); (lane g) sensitive parent SV40transformed Syrian hamster cell Cl2TSV5S; (lane
h) actinomycin D-resistant SV40-transformed
Syrian hamster cell Cl2TSV5R2; (lane i) sensitive
mouse L cell LMTK–; (lane j) colchicine (0.50 µg/
ml)-resistant mouse L cell ECHR; (lane k) sensitive parent human lymphoid cell CCRF-CEM;
(lane l) vinblastine-resistant human lymphoid
cell CEM/VLB100 (see Table 1 for further details). The P-glycoprotein region is indicated (P)
at molecular size 150,000 to 170,000 daltons.
Lanes a to f and g to I represent two separate
gels that were run simultaneously and treated
identically. (B) SDS-PAGE was performed as in
(A), with the exception that 50 µg of membrane
protein was loaded per lane. Western blots were
overlaid with previously absorbed rabbit antiserum (diluted 1:100) against isolated plasma
membranes from the colchicine-resistant CHO
cells CHRC5 and processed as described (21,
27). Lanes a to f and g to l are two separate blots
prepared and treated identically, but lanes k and
l were exposed to film four times longer than
other lanes. All lanes are as described in (A). (C)
SDS-PAGE was performed as in (B), but blots
were overlaid with unabsorbed serum prepared
against CEM/VLB100 plasma membranes (diluted 1:100). Lanes k and I were exposed to film
for one-tenth the exposure time of other lanes.
drug resistant lines expressed a 170,000-dalton component that was stained by this antiserum (Fig. 1B). It is clear from the size and
cross-reactivity of this component that it is
similar to the P-glycoprotein of the CHR CHO
cells. Staining of other components of about
50,000 and 200,000 daltons is also observed
with this antiserum. We believe that these
components are not related to the P-glycoprotein or multidrug resistance because they vary
with different preparations of membranes.
The relatively faint staining of the presumptive human P-glycoprotein by the antiserum
is likely due to reduced cross-reactivity of this
component with the antiserum against the
CHO cell P-glycoprotein.
To further corroborate the above conclusions, we stained the membrane components
with an antiserum to plasma membrane of
the vinblastine-resistant, human cell line,
CEM/VLB100 (Fig. 1C). In this case, the serum
was not previously absorbed with membrane
proteins from drug-sensitive cells, and many
antigens common to both drug-resistant and
drug-sensitive human cells were strongly
stained (lanes k and l; note the exposure
time). This crude antiserum stained only the
drug resistance-associated P-glycoprotein in
the rodent cell membranes (Fig. 1C). Thus, of
the dozen or so different cell surface antigens
29
that can be detected with the two antiserums,
only the P-glycoprotein is consistently stained
in all drug-resistant mammalian cell lines
tested. These observations, and the fact that
a side-by-side comparison of the different Pglycoprotein bands reveals no significant difference in molecular size, as shown in Fig. 1,
strongly indicate that the P-glycoprotein is
conserved relative to other mammalian membrane antigens that are detectable by Western
blotting.
Our observations that P-glycoprotein is
present in increased amounts in various drugresistant lines, and that in the CHR CHO system the amount of P-glycoprotein expressed is
correlated with the degree of resistance, are
consistent with a mechanism of resistance involving gene amplification. The appearance of
double minute chromosomes has been correlated with unstable drug resistance in known
gene-amplified systems (28, 29). Double minute chromosomes have also been observed
in association with multidrug resistance in
several mouse cell lines including colchicineresistant lines for which the degree of resistance and P-glycoprotein expression correlate
with the number of double minutes contained
within the cells (14, 30). Double minutes have
also been reported in multidrug-resistant
hamster cells that were originally selected for
colchicine resistance (30, 31). This speculation
on the origin of the P-glycoprotein in drug-resistant cells suggests its preexistence, in much
smaller amounts, in the drug-sensitive parent
cell. In this context, we observed a barely detectable, antigenically cross-reactive band of
the same molecular size as the P-glycoprotein
in all the drug-sensitive cells examined in
Fig. lB when the film was exposed for a much
longer period (data not shown). What specific role the P-glycoprotein might play in the
maintenance of the multidrug resistance phenotype, or in the drug-sensitive cells, is unknown. It is also not yet known whether the
P-glycoprotein is expressed in normal tissues.
Our findings could have important implications for cancer therapy. It is possible that clinical resistance to combination chemotherapy
might result from an unchecked proliferation
of tumor cell subpopulations with a multidrug resistance phenotype (33). The present
data indicate that there is a strong correlation
between the expression of multidrug resis-
30
tance and surface P-glycoprotein in different
species of drug-resistant cells established in
vitro. It is reasonable to suppose that such
a relationship might also exist in vivo. If Pglycoprotein is commonly present in tumor
cells from patient biopsies, and if this antigen
is expressed in increased amounts in tumors
nonresponsive to treatment by combination
chemotherapy, immunochemical screening
for the antigen could provide a rapid diagnostic basis for planning treatment of cancer
patients. Moreover, hitherto unresponsive
tumors may become amenable to treatment
with P-glycoprotein–targeted antibodies conjugated with toxins.
1. N. T. Bech-Hansen, J. E. Till, V. Ling, J. Cell.
Physiol. 88, 23 (1976).
2. S. Brabbs and J. R. Warr, Genet. Res. 34, 269
(1979).
3. L. J. Wilkoff and E. A. Dulmadge, J. Natl.
Cancer Inst. 61, 1521 (1978).
4. B. Salles, J.-Y. Charcosset, A. Jacquemin-SabIon, Cancer Treat. Rep. 66, 327 (1982).
5. T. Skovsgaard, Cancer Res. 38, 4722 (1978).
6. V. Ling and L. H. Thompson, J. Cell. Physiol.
83, 103 (1974).
7. K. DanØ, Acta Pathol. Microbiol. Scand. Suppl.
256, 11 (1978).
8. M. P. Chitnis and R. K. Johnson, J. Natl.
Cancer Inst. 60, 1049 (1978).
9. J. L. Biedler and R. H. F. Peterson, in Molecular
Action and Targets for Cancer Chemotherapeutic Agents, A. C. Sartorelli et al., Eds.
(Academic Press, New York, 1981), p. 453.
10. Y. Langelier, R. Simard, C. Brailovsky,
Differentiation 2, 261 (1974).
11. P. D. Minor and D. H. Roscoe, J. Cell Sci. 17,
381 (1975).
12. V. Crichley, D. Mager, A. Bernstein, J. Cell.
Physiol. 102, 63 (1980).
13. C. D. Aldrich, J. Natl. Cancer Inst. 63, 751
(1979).
14. F. Baskin, R. N. Rosenberg, V. Dev. Proc.
Natl. Acad. Sci. U.S.A. 78, 3654 (1981).
15. S. A. Carlsen, J. E. Till, V. Ling, Biochim.
Biophys. Acta 455, 900 (1976).
16. M. E. Lalande, V. Ling, R. G. Miller, Proc.
Natl. Acad. Sci. U.S.A. 78, 363 (1981).
17. D. Kessel and H. B. Bosmann, Cancer Res. 30,
2695 (1970).
18. D. Garman and M. S. Center, Biochem.
Biophys. Res. Commun. 105, 157 (1982).
19. W. T. Beck and M. C. Cirtain, Cancer Res. 42,
184 (1982).
20. R. L. Juliano and V. Ling, Biochim. Biophys.
Acta 455, 152 (1976).
21. P. G. Debenham, N. Kartner, L. Siminovitch, J.
R. Riordan, V. Ling, Mol. Cell. Biol. 2, 881
(1982).
22. J. R. Riordan and V. Ling, J. Biol. Chem. 254,
12701 (1979).
23. R. H. F. Peterson and J. L. Biedler, J. Supramol.
Struct. 9, 289 (1978).
24. V. Ling, Can. J. Genet. Cytol. 17, 503 (1975).
25. __ and R. M. Baker, Somat. Cell Genet. 4,
193 (1978).
26. R. M. Baker and V. Ling, in Methods in
Membrane Biology., E. D. Korn, Ed. (Plenum,
New York, 1978), vol. 9, p. 337.
27. N. Kartner, M. Shales, J. R. Riordan, V. Ling,
Cancer Res., in press.
28. P. C. Brown, S. M. Beverley, R. T. Schimke,
Mol. Cell. Biol. 1, 1077 (1981).
29. R. J. Kaufman, P. C. Brown, R. T. Schimke,
ibid., p. 1084.
30. S. M. Robertson, V. Ling, C. P. Stanners, in
preparation.
31. B. P. Kopnin, Cytogenet. Cell Genet. 30, 11
(1981).
32. We have recently observed, using the
methods described, highly elevated
P-glycoprotein expression in vincristineresistant Chinese hamster cells, which bear a
chromosomal homogeneous staining region
[see T. Kuo et al. in Gene Amplification, R. T.
Schimke, Ed. (Cold Spring Harbor Laboratory,
Cold Spring Harbor, N.Y., 1982), pp. 53-57].
33. N. T. Bech-Hansen et al., J. Natl. Cancer Inst.
59, 21 (1977).
34. R. C. Switzer, C. R. Merril, S. Shifrin, Anal.
Biochem. 98, 231 (1979).
ACKNOWLEDGMENTS
Supported by grants from the National Cancer
Institute of Canada and the Medical Research
Council of Canada. We thank W. T. Beck, P. G.
Debenham, T. Kuo, and R. Simard for making
available their drug-resistant cell lines; and N.
Alon, S. Fahim, and M. Naik for technical
assistance.
13 May 1983
GE Healthcare
GE Healthcare
Reproducibility using the
Amersham™ WB system
High- and sup
DeltaV
Åsa Hagner McWhirter, Anita Larsson, Elisabeth Wallby, Anna Edman-Örlefors, Ylva Laurin and Ola Rönn
DeltaVision E
high-resoluti
Introduction
DeltaVision Elite is a
imaging system sp
for imaging challen
Amersham WB system is a fully integrated system for separation,
transfer, detection, and quantitative analysis of proteins that gives
consistent, quantitative data for every sample, every time. The system
achieves this high quality of data by combining three elements:
a SMART system design that standardizes and monitors every step
of the process; dedicated consumables and optimized protocols
that minimize assay variability; and built-in data normalization. The
system delivers consistent, quantitative data with fewer technical
replicates, very little hands-on intervention and less overall time
required to generate results.
Whereas traditional manual Western blotting can take up to 24 h/run
or experiment, the Amersham WB system delivers Western blotting
results, based on proven methods, and automatic evaluation of
membrane image in just 4 to 5 h. The system uses fluorescence
detection to enable a wide dynamic range, excellent sensitivity,
and color multiplexing for reliable normalization and applications
such as the simultaneous detection of phosphorylated/nonphosphorylated proteins. The primary antibodies that work with
traditional Western blotting protocols will work on the Amersham
WB system, preserving a link to valuable historical data. The
Amersham WB system enables researchers, staff, and students
new to Western blotting to generate high-quality, quantitative
data from day one, since the ease of use is high and there is error
proofing in a majority of the steps.
Here we present results for SDS PAGE and Western blotting
showing high reproducibility within or between gels/membranes
for different users, both experienced and first time users.
Methods
Protein mixes or Chinese hamster ovary (CHO) cell lysates (prepared
at GE Healthcare) were pre-labeled with Amersham WB Cy™5
according to SDS PAGE and Western pre-labeling protocols.
The samples were applied to Amersham WB gel cards. The
electrophoresis, transfer to Amersham WB PVDF cards and probing
was performed using the default settings of the Amersham WB system.
Tubulin and ERK 1/2 was detected using primary antibodies from
Sigma-Aldrich (mouse anti-tubulin, diluted 1:1000 or 1:2500 and
rabbit anti-ERK1/2, diluted 1:5000) and anti-mouse and anti-rabbit
• Improved contr
secondary Amersham WB antibodies labeled with Cy3 (diluted
1:2500). The gels and membranes were scanned with optimal
• More efficient li
settings in the system and the images evaluated using Amersham
exposure condi
WB evaluation software. The data was exported to Microsoft® Excel®
for coefficient of variation (CV%) calculations and linearity
• graphs.
Fully integrated
Un-labeled CHO cell lysate or HeLa and NIH 3T3 cell lysates
(Santa platfor
• Modular
Cruz) dilution series were used to compare reproducibility of target
expand capabil
signals. For chemiluminscence detection, Amersham ECL Gel 12%
and ECL™ Box electrophoresis unit (both from GE Healthcare)
were flexib
Maximum
used for the SDS-PAGE. The proteins were transferred using wet transfer
(TE22, GE Healthcare) onto Amersham Hybond™ P PVDFDeltaVision
membrane Elite ca
including
(GE Healthcare). ERK1/2 was targeted using primary rabbit
anti- widefield
ERK1/2 (Sigma) diluted 1:20 000 (incubated overnight at multipoint
+4°C) and cell trac
secondary HRP conjugated anti-rabbit IgG (GE Healthcare)
diluted
(TIRF),
fluorescence
1:100 000 and finally the signals were detected using Amersham
photokinetics, and
ECL Select detection reagent (GE Healthcare) and a CCD camera
(ImageQuant™ LAS 500 Imager, GE Healthcare). The signal
intensity
Deconvolution
s
for each sample amount was determined using ImageQuantTL
Deconvolution imp
image analysis software (GE Healthcare).
contrast without sa
For fluorescence detection, the Amersham WB system with default
exclusive deconvol
settings was used as described above. Tubulin was targeted using
in discovery resear
primary rabbit anti-Tubulin (Sigma) diluted 1:1000 and secondary
part
of the system.
Amersham WB Cy3 labeled anti-rabbit IgG (GE Healthcare)
diluted
studies at
1:2500. The signal intensity for each sample amount was enabling
determined
using Amersham WB evaluation software (GE Healthcare). For more
Outstanding foc
details, see figure legends.
imaging experim
UltimateFocus auto
regardless of mech
Conclusion
impact your experi
Amersham WB system is very reproducible and user
of UltimateFocus d
independent with built-in total protein normalization which
objective
can efficiently correct for uneven sample loading. This
results and the c
objective into the a
in CVs between 5 and 10% compared with a traditional
a camera.
manual workflow where CVs can be between 15 andor
35%.
Technical note
31
Reproducible Cy5 pre-labeling of proteins
MW
Markers
(A) Concurrent
ECL detection
225 kDa
100
80
97 kDa
R2 = 0.98833
Conalbumin CV: 4.3 %
66 kDa
50 kDa
60
35 kDa
40
25 kDa
20
20 kDa
0
14 kDa
0
5
10
15
20
Ratio
Conalbumin/Aldolase
CV: 5.7 %
10 kDa
25 µg
Comparing variation of non-normalized and normalized target signals
Cy3/Cy5
Image overlay
Cy3 Non-normalized
Target signal (CV%)
15 000
8.0
500
16%
300
250
200
22%
150
100
16%
50
Fig 2. 14 Individual samples containing 2 µl
with a mix of Conalbumin and Aldolase
(0.1 µg/µl) were pre-labeled with Cy5 in 20 µl
Amersham WB labeling buffer for 30 minutes
at room temperature. The reactions were
stopped by adding 20 µl sample loading buffer,
heated to 95°C for 5 min, and applied to an
Amersham WB gel card, 8-18%. CV% of the
signal intensity for each reaction as well as
the ratio between the two protein signals
were calculated.
Fig 1. Samples containing 10, 20, 30 or 40 µg of
CHO cell lysate in 80 µl Mammalian Protein Extraction
Buffer (GE Healthcare) containing protease inhibitor
mix (GE Healthcare) were pre-labeled for 30 min
in room temperature. The reactions were stopped
by adding 80 µl sample loading buffer followed by
boiling at 95°C for 5 min. Samples containing 5,
10, 15 or 20 µg of CHO cell lysate were applied in
triplicate to an Amersham WB gel card, 8-18%. The
Cy5 total signal within each lane on the membrane
was plotted against the protein amount loaded.
400
350
Aldolas CV: 7 %
Cy3/Cy5
Normalized ratio (CV%)
0.20
5.4
(B) Non-concurrent
ECL detection
Signal intensity (× 105)
120
Traditional manual workflow results in high
variation of target signals
Cy5 total protein signal (volume × 106)
Linearity of Cy5 total protein pre-labeling
0
27%
400
300
34%
200
35%
100
35%
20%
0
2.5
1.3
0.6
0.3
NIH/3T3 cell lysate (µg)
2.5
1.3
0.6
0.3
NIH/3T3 cell lysate (µg)
Fig 4. A dilution series of NIH/3T3 cell lysate was
applied a SDS-PAGE gel (Amersham ECL gel/
ECL gel box, GE Healthcare) and the proteins were
transferred to a PVDF membrane. The membranes
were blocked, probed and washed manually using
trays and a rocker. ERK1/2 was targeted using
primary and secondary HRP conjugated anti-rabbit
IgG and the signals were detected using ECL Select
detection reagent and a CCD camera. The signal
intensity for each sample amount was determined
and the average and CV% was calculated using
Microsoft Excel.
Six membranes were analysed concurrently (A) or
on three separate occasions (B) but otherwise using
identical conditions. Sample dilution series, antibodies,
probing protocol, ECL reagent and detection exposure
time were identical for the signals to be compared.
0.10
Amersham WB standardized, fluorescence based
workflow results in low variation of target signals
5000
0.05
0
15 000
1 2 3 4 5 6 7 8 9 10 11 12
6.7
0
0.12
1 2 3 4 5 6 7 8 9 10 11 12
175
User 2
0.08
0.06
0.04
5000
200
5.0
0.10
10 000
(A) Concurrent AWB
(B) Non-concurrent AWB
Fluorescence detection
Fluorescence detection
30 000
1 2 3 4 5 6 7 8 9 10 11 12
11.2
0
0.30
1 2 3 4 5 6 7 8 9 10 11 12
6.6
User 3
0.25
20 000
0.20
0.15
10 000
0.10
0.05
0
1 2 3 4 5 6 7 8 9 10 11 12
0
5.4%
125
100
7.5%
75
6.4%
50
0
0
1 2 3 4 5 6 7 8 9 10 11 12
Fig 3. Three different first time users of the Amersham WB system performed Cy5 pre-labeling of CHO cell
lysate and 7.5 µg total protein per well were loaded to an Amersham WB gel card, 13.5%. Western blotting
was performed targeting ERK1/2 with primary antibody and Amerham WB Cy3 labeled secondary
antibody. The target signals were normalized using the Cy5 total protein signal. The CVs were all reduced
compared to non-normalized target signals and below 7% for the normalized ratio (Cy3 target/Cy5 total
protein) indicating correction of the uneven loading.
7.1%
3000
150
25
0.02
3500
5.9%
User 1
0.15
10 000
6.2%
5.0
2.0 1.0 0.5 0.1
HeLa cell lysate (µg)
2500
7.5%
2000
7.2%
1500
1000
6.6%
500
0
10.0
5.0
2.5
0.6
CHO cell lysate (µg)
Fig 5. A dilution series of HeLa cell lysate was applied
to an Amersham WB gel card, 8-18%, and the
proteins were transferred to a Amersham WB PVDF
card. The membranes were blocked, probed and
washed using standardized automated protocol
in the Amerhsam WB system. Tubulin was targeted
using primary antibody and secondary Amersham
WB Cy3 labeled anti-rabbit IgG. The signal intensity
for each sample amount was determined using
Amersham WB evaluation software and the average
and CV% was calculated using Microsoft Excel.
Four sample replicates on two membranes were
analysed concurrently in the same Amersham WB
run (A) or on three separate occasions (B) but otherwise
using same conditions. Sample dilution series,
antibodies, and the probing protocol were identical for
the signals to be compared.
www.gelifesciences.com
GE and GE monogram are trademarks of General Electric Company. Amersham, ECL, Cy, CyDye, Hybond and ImageQuant are trademarks of General Electric Company or one of its subsidiaries. Excel is a registered trademark of Microsoft
Corporation. All other third-party trademarks are the property of their respective owners.
CyDye: This product is manufactured under an exclusive license from Carnegie Mellon University and is covered by US patent numbers 5,569,587 and 5,627,027.
The purchase of CyDye products includes a limited license to use the CyDye products for internal research and development but not for any commercial purposes. A license to use the CyDye products for commercial purposes is subject to a
separate license agreement with GE Healthcare. Commercial use shall include: 1. Sale, lease, license or other transfer of the material or any material derived or produced from it. 2. Sale, lease, license or other grant of rights to use this material
or any material derived or produced from it. 3. Use of this material to perform services for a fee for third parties, including contract research and drug screening. If you require a commercial license to use this material and do not have one, return
this material unopened to GE Healthcare Bio-Sciences AB, Bjorkgatan 30, SE-751 84 Uppsala, Sweden and any money paid for the material will be refunded. © 2015 General Electric Company — All rights reserved. First published May 2015.
GE Healthcare UK Limited Amersham Place, Little Chalfont, Buckinghamshire, HP7 9NA, UK
29133719AA 05/2015
32
What if Western blotting was
consistent and quantitative?
Introducing the fully integrated Amersham™ WB system designed
to deliver consistent quantitative data every sample, every time.
With reduced repetition and fewer control experiments, you will feel
confident that you have the results to move forward.
Go to www.gelifesciences.com/artofwesternblotting to find out more.
Amersham | Biacore | ÄKTA | Whatman | Cytell | Xuri*
GE and GE monogram are trademarks of General Electric Company.
* Amersham, Biacore, ÄKTA, Whatman, Cytell and Xuri are trademarks of General
Electric Company or one of its subsidiaries.
© 2015 General Electric Company—All rights reserved. First published May 2015.
GE Healthcare UK Limited, Amersham Place, Little Chalfont, Buckinghamshire, HP7 9NA, UK
Amersham™ Western blotting:
expertise for results without
compromise
The Amersham Western blotting portfolio brings together Rainbow™ markers,
Hybond™ membranes, blockers, CyDye™ conjugated secondary antibodies,
Amersham ECL™ detection reagents, imaging systems and Amersham WB system.
Whether you need to confirm the presence of your protein or quantify multiple
targets with fluorescence detection, Amersham Western blotting has a
solution for your needs.
Go to www.gelifesciences.com/amershamwb to find out more.
Amersham | Biacore | ÄKTA | Whatman | Cytell | Xuri*
GE and GE monogram are trademarks of General Electric Company.
* Amersham, Biacore, ÄKTA, Whatman, CyDye, Cytell, ECL, Hybond, Rainbow
and Xuri aretrademarks of General Electric Company or one of its subsidiaries.
© 2015 General Electric Company—All rights reserved. First published May 2015.
GE Healthcare UK Limited, Amersham Place, Little Chalfont, Buckinghamshire, HP7 9NA, UK

Western Blotting: A Guide to Current Methods

Transcription

Similar documents

How to Tie a Woggle

lnstallation - ConservationMart.com

Turkey - EOPOWER

INMUNOLOGIA MOLECULAR

Key - Scioly.org

RÉSoNANSS: a regional contribution to

Real Time Control of Hand Prosthesis Using Surface EMG

Kerntech Diagnose-System Loose Part Monitoring System

CoUNTerMEASURES CoUNTerMEASURES

PrimeFilm 1800 AFL Low-Cost 35mm Slide/Film Scanning Through