| May
1, 2005 |
| By:
Michael
O'Grady, George
Hansen, Brian
Pollok, Debasish
Raha, Shelley
Hough, Kristin
Wiederholt, Michaeline
Bunting, Peter
Welch |
| Pharmaceutical
Discovery |
|
Alterations in cell signaling pathways are the biochemical basis
underlying a wide range of diseases, including inflammation, diabetes
and cancer. Cell survival, proliferation and differentiation are
regulated by a complex network of pathways in response to signals from
the environment. Protein kinases largely are responsible for transducing
these extracellular signals to the nucleus to modulate gene expression.
Sequencing of the human genome has revealed approximately 500 protein
kinases, demonstrating the complexity of kinase cascades in regulating
cellular decisions to grow, divide or die (1). Kinases are becoming
increasingly attractive targets for therapeutic development and
currently are the second most studied therapeutic target after
G-protein-coupled receptors (GPCRs). The recent successes, such as
Gleevec™ (Novartis Pharmaceuticals, East Hanover, New
Jersey, USA) and Tarceva™ (Genentech, South San Francisco,
California, USA) and the uncertain utility of approved drug entities
such as Iressa™ (AstraZeneca, Wilmington, Delaware, USA)
highlight the promises and potential pitfalls for developing
small-molecule kinase inhibitors as molecular-targeted therapies (2, 3).
Discovering new drugs has become increasingly expensive and time
consuming; current drug discovery efforts begin with gene profiling and
cellular genetics to identify potential drug targets, followed by
screening of small-molecule libraries to identify "hits" and
medicinal chemistry to convert hits into lead compounds. Despite a
seven-fold increase in research spending over the last two decades by
pharmaceutical companies, the number of new compounds approved by the
FDA each year has essentially remained static (4). While new
technologies may not necessarily shorten the time for drug discovery and
development (5), the use of enabling genomic tools and facile cellular
assays to qualify targets should help discovery groups make better
target choices and focus resources. We have used the combination of
Stealth™ RNAi (Invitrogen Corp., Carlsbad, California, USA)
and CellSensor™ (Invitrogen) technologies to create a
high-throughput platform for target discovery, which also can be
directly used for small-molecule screening.
Hitting the Target Target
discovery can be difficult for target classes like protein kinases due
to their significant structural similarity and functional redundancy. In
addition, changes in activity or regulation of a protein kinase in a
disease state may be a consequence rather than the cause of pathway
misregulation (e.g., a compensatory mechanism). There remains a need for
models that can demonstrate a clear correlation between inhibition of
the putative target and a functional consequence in a
physiologically-relevant cellular setting (6). Such information provides
greater confidence to a therapeutic team as they move downstream in the
discovery process.
RNA interference (RNAi) describes the sequence-specific degradation
of target mRNA that is mediated by double- stranded RNA (dsRNA) via the
RNAi pathway (7, 8). In this pathway, the dsRNA is recognized by a
series of proteins — referred to as the RNAi-induced silencing complex
(RISC) — that guides cleavage of homologous target mRNA (9, 10).
Introduction of dsRNA longer than approximately 30 bp in mammalian cells
induces the interferon response. Transfection of short stretches of
dsRNA, called short interfering RNAs (siRNAs), in mammalian cells causes
a specific silencing of target mRNAs without induction of the interferon
response (11). RNAi technology is broadly applicable for functional
genomics and target validation because oligonucleotides can be
rationally designed to specifically inhibit any gene target, provided
that the sequence is known.
While siRNAs can be rationally designed, some concerns about their
specificity have been raised. Recent reports in the literature suggest
that siRNAs can affect genes other than the intended target (off-target
effects) (12). Although it generally has been accepted that dsRNA less
than approximately 30 bp are not recognized as foreign by the immune
response, some siRNA duplexes have been found to stimulate an interferon
response (13). To address some of the limitations of traditional siRNAs,
Invitrogen has developed second-generation Stealth RNAi agents for
target discovery. The chemical modifications on Stealth RNAi are
designed to increase the specificity by allowing only the antisense
strand to efficiently enter the RNAi pathway and eliminating induction
of interferon-related pathways.
Kinases activate a cascade of phosphorylation events, resulting in
gene regulation at the transcriptional level. Biochemical screening
assays typically are employed as a starting point to identify
small-molecule inhibitors against purified protein kinase targets. There
are a number of different biochemical assays that can be used to screen
for kinase inhibitors, with many analyzing the phosphorylation of a
synthetic peptide substrate. These assays have the limitation of
measuring kinase activity under non-physiological conditions (e.g., low
concentrations of adenosine triphosphate [ATP]). Because the evaluation
of kinase inhibitor potency in biochemical assays often fails to
correlate with cellular systems, there is a need for high-throughput
functional cellular assays to be used in parallel.
Figure 1. The EGFR pathway and the
ME-180 AP-1 CellSensor. A) The EGFR pathway is shown with an EGF
agonist-stimulated signal cascade, leading to AP-1 activation.
B) The ME-180 AP-1 CellSensor was used to generate a
dose-response curve with increasing concentrations of rhEGF.
ME180 AP-1 cells were plated in 96-well plates. The following
day the cells were treated with increasing concentrations of
rhEGF. Cells were incubated for 5 hr at 37 °C with CO2
and then assayed for β-Lactamase activity. Each point
represents an n = 8.
|
As a compliment to our biochemical
screening assay platforms — Z'-Lyte™, PolarScreen™
and LanthaScreen™ — we have developed the CellSensor
technology, a cell-based functional reporter assay that can be used for
both target discovery work and high-throughput screening of small
molecules. Reporter assays have become the standard for cell-based
analysis studies due to their ease of use and widespread acceptance.
Reporter assays such as β-galactosidase, CAT and luciferase
commonly are used, but all require another reporter to normalize the
results. The GeneBLAzer® technology (Invitrogen) overcomes
this limitation by using a fluorescence resonance energy transfer (FRET)
read-out to obtain blue (460 nm) and green (530 nm) signal from a single
enzyme (β-lactamase) and a single substrate, allowing for a
ratiometric read-out (14). This technology eliminates variability
attributable to transfection efficiencies and/or plating errors, which
can occur with other reporter read-outs, and also allows for cell
imaging concurrent with the assay itself. We have established several
CellSensor™ assays to study individual kinase targets or
entire signaling pathways.
Figure 2. Knockdown of EGFr
expression using Stealth RNAi. A) Transfection of Stealth RNAi
duplexes in ME-180 AP-1 CellSensor cells resulted in >80%
knockdown of the EGFR component of the AP-1 pathway. ME180 AP-1
cells were transfected with EGFR and medium GC control Stealth
RNAi and Block iT™ Fluorescent Control using
Lipofectamine 2000. Transfection efficiency was confirmed using
the Block iT™ Fluorescent Oligo (panel B). Gene
expression was assessed using the SuperScript™ III
Platinum Two-Step qRT-PCR kit 48 h post-transfection. The
expression level of EGFR was normalized against cyclophilin, and
percent inhibition of EGFR transcript by EGFR Stealth RNAi was
calculated relative to the medium GC negative control Stealth
RNAi. B) Fluorescent uptake with the Block iT™
Fluorescent Oligo is seen in ME-180 AP-1 CellSensor cells.
|
Experimental Conditions and Results
CellSensors have been developed for the study of numerous cell signaling
pathways, including the epidermal growth factor receptor (EGFR) pathway.
Stimulation of the EGFR by an epidermal growth factor (EGF) ligand
results in activation of the RAS/MAPK signaling pathway and
transcription factors such as AP-1 (Figure 1A). We have generated an
ME180 cell line with stable expression of β-lactamase under control
of the AP-1 transcriptional response element. EGF was used as the ligand
in dose-dependent response experiments to confirm that the EGFR pathway
was functional in the AP-1 CellSensor (Figure 1B).
Figure 3. Validation of EGFR
involvement in the ME-180 AP-1 CellSensor. Traditional assay
methods for effects of RNAi knockdown use lytic methods for
analysis of a target. The CellSensor™ and
GeneBLAzer® technologies allow for the functional
cellular assay of a physiologically-relevant effect, shown here
at the level of AP-1 transcription. ME180 AP-1 cells were plated
in 96-well plates. The next day, cells were transfected with the
Stealth™ RNAi duplexes indicated. Following a 40-
or 60-h incubation at 37 °C with CO2, cells were
treated with 1 ng/ml rhEGF for 5 h. The cells were assayed for
β-Lactamase activity. All results were normalized to medium
GC control, with each bar representing an n = 8.
|
The EGFR pathway was interrogated in
the AP-1 CellSensor line using Stealth™ RNAi. Stealth RNAi
duplexes were designed to target EGFR and also β-lactamase as a
positive control. Inhibition of EGFR was confirmed by assessing
knockdown of the target RNA using reverse-transcriptase polymerase chain
reaction (RT-PCR) in parallel assays (Figure 2). The targeted and
negative control Stealth RNAi duplexes then were transfected into the
AP-1 CellSensor line using Lipofectamine™ 2000 (Invitrogen).
At 40 and 60 hours post-transfection, β-lactamase expression was
assessed using the GeneBLAzer® reporter assay. The
ratiometric readout at 60 hours shows 75% and 90% inhibition of β-lactamase
expression in cells treated with EGFR and β-lactamase Stealth RNAi,
respectively, relative to treatment with the negative control (Med GC
Stealth™ RNAi). These results are confirmed by the
corresponding images of cellular fluorescence shown in Figure 3.
Figure 4. Kinase inhibitors
analyzed against a validated EGFR target in an AP-1 CellSensor
correlate with results from a SelectScreen biochemical assay. A)
ME180 AP-1 cells were plated in 96-well plates. The next day,
cells were treated with 500 nM of the indicated kinase
inhibitor. Cells were incubated for 30 min. at 37 °C with CO2
and treated with 1 ng/ml rhEGF for 5 h. The cells then were
assayed for β-lactamase activity. Specificity and potency
can be determined through the parallel analysis of the
complementary methods. Percent inhibition of AP-1
transcriptional activation normalized to a DMSO control is shown
for each compound and represents an n = 8. B) PD153035 and AG183
were further analyzed in a dose-response manner using the
ME180-AP-1 CellSensor. ME180 AP-1 cells were plated in 96-well
plates. The next day, cells were treated with increasing
concentrations of the indicated compound for 30 min. at 37 °C;
with CO2. The cells then were treated with 1 ng/ml
rhEGF for 5 h and assayed for β-lactamase activity. IC50
concentrations were calculated. C) The ability to analyze the
specificity of the small-molecule compound in a purified setting
enables the researcher to compare biochemical specificity with
the effect seen in a functional cellular assay, as shown in
Figure 4A. All compounds were tested at a concentration of 1 _M
and an ATP concentration of Km app.
|
The validated EGFR CellSensor then was
used to analyze kinase inhibitors targeting EGFR in the AP-1 pathway. A
panel of inhibitors specific for EGFR and other specific inhibitors of
non-EGFR/AP-1 pathway kinases were analyzed at a 500 nM concentration
for their efficacy in a functional cellular assay (Figure 4A). A
dose-response curve was generated to determine the potency of the
compound hits, and an IC50 concentration was calculated using
GraphPad Prism™ (GraphPad Software, San Diego, California,
USA) software. The PD153035 compound (Calbiochem, La Jolla, California,
USA) showed strong inhibition of EGFR (IC50=15 nM), while AG183 (Calbiochem),
which was not a hit in the inhibitor panel, showed no effect across the
dose-response curve (Figure 4B). Profiling services, such as the
SelectScreen™ Kinase Profiling Service (Invitrogen),
provide access to a large panel of kinases for determining inhibitor
specificity against a particular target. SelectScreen results for the
inhibitory activity of these compounds against a panel of protein
kinases (Figure 4C, EGFR results only) showed excellent correlation
between the biochemical and functional cellular assays, a key hallmark
of any valuable suite of assay technologies.
Conclusions Drug
discovery groups have a continuing need for improved assay technologies
that can enable improved decision making. Using "linked"
technologies that provide ready transitions from target discovery to
biochemical screening, as well as functional cellular secondary assay
studies, provides a platform that should make discovery more efficient.
The iterative process of advancing a compound series through the
discovery gates would clearly benefit from matched assays that are
functionally cross-validated.
References
1. L.B. Ray, N.R. Gough, Science's STKE 162, eg13, 1
(2002).
2. P. Cohen, Nat.Rev. Drug Disc.
1, 309-315 (2002).
3. M. Muhsin, J. Graham and P.
Kirkpatrick, Nat. Rev. Drug Disc. 2, 515-516 (2003).
4. S. Bernard, Pharmaceutical
Executive 6-16, (2002).
5. W. Bains, Drug Discovery
World 5, 9-18 (2005).
6. D. Samarsky and P. Welch, Preclinica
2, 396-396, (2004).
7. A. Fire et al., Nature 391,
806-811 (1998).
8. S.M. Elbashir, W. Lendeckel and
T. Tuschl, Genes Dev. 15, 188-200 (2001).
9. G.J. Hannon, Nature418,
244-251 (2002).
10. M.T. McManus and P.A. Sharp, Nat.
Rev. Genet. 3, 737-747 (2002).
11. S.M. Elbashir et al., Nature
411, 494-498, (2001).
12. A.L. Jackson and P.S. Linsley, TRENDS
in Genetics 20, 521-524, (2004).
13. V. Hornung et al., Nature
Medicine 11, 263-270, (2005).
14. G. Zlokarnik et al., Science
279, 84-88, (1998).
Michael O'Grady is a
research associate, George Hansen is senior research scientist Brian
Pollok is chief science officer at Invitrogen Drug Discovery
Solutions in Madison, Wisconsin, USA. Debasish Raha is senior
scientist, Shelley Hough is a scientist, Kristin Wiederholt
is technical area manager of RNAi, Michaeline Bunting is R&D
area manager of RNAi and Peter Welch is associate director of
R&D at Invitrogen Corporation in Carlsbad, California, USA. Michael
O'Grady can be reached at 501 Charmany Drive, Madison, Wisconsin 53719
USA; e-mail michael.ogrady@invitrogen.com
.
|
