Note: Descriptions are shown in the official language in which they were submitted.
_2159822
'r - .. ~ E ..---~
TEff TRA'j(SU~'t~;j
S015-231.62
Process for inhibiting the transcription of genes
The invention relates to influencing the transcription
of genes.
Inducible gene expression depends, amongst other things,
on the inducible activation of proteins, regulating
transcription by interacting with cis-regulatory DNA
elements. The activity of these proteins, referred to
as transcription factors, can be regulated by their de
novo-synthesis; this strategy requires additional
factors affecting the gene for the transcription factor.
On the other hand, posttranslational mechanisms for the
activation of transcription factors have the advantage
of being quicker than mechanisms at the transcription
level. For control of the activity of transcription
factors, inhibitory protein subunits play, inter alia,
an important role. An example for this is IP-1, a
Leucine zipper protein, inhibiting AP-1. In this case
the inhibitory subunit assumes the function of a trans-
dominant negative regulator based on a structural
homology with the activator. The NF-KB system is a
system in which the inhibitory subunits show no homology
with the DNA-binding subunits. The inhibitory subunits
of NF-KB and the related factors are referred to as IkB
proteins, reversibly inhibiting the binding of the
transcription factor to DNA (Baeuerle and Baltimore,
1988a, b) .
NF-KB, a heterodimeric factor consisting of a 50 kDa
(p50) and a 65 kDa (p65) DNA-binding subunit,
contributes to the so-called "immediate-early" activation
of defence genes if cells are exposed to primary or
secondary pathogenic stimuli (Baeuerle, 1991, Baeuerle
and Baltimore, 1991).
2159822
- 2 -
The transcription factor NF-KB (Grimm and Baeuerle,
1993; Blank et al., 1992; Nolan and Baltimore, 1992) is,
inter alia, activated by treating cells with
bacteriological stimuli (inter alia LPS), viruses (inter
alia HIV virus type 1), viral products, parasites,
inflammatory cytokines (inter alia TNF-(x, TNF-(3, IL-1,
IL-2), T-cell mitogens (inter alia lectines), protein
synthesis inhibitors (inter alia cycloheximide),
physical stress (UV-light, gamma radiation), oxidative
stress (inter alia hydrogen super oxide) and tumour
promoters (inter alia phorbol ester) (Baeuerle, 1991).
The activation of NF-KB in response to a large number of
pathogenic stimuli is carried out by a mechanism which,
as yet, has not been completely explained. Subsequent
to activation, NF-KB is transferred into the cell
nucleus and the target genes are activated by active NF-
KB. It is assumed that reactive oxygen compounds play a
role as messengers during the activation of NF-KB
(Schreck et al., 1991).
It has been proven that the activation of NF-KB is
connected with the release of the inhibitory subunit IkB
from a cytoplasmic complex including the DNA-binding
subunits p50 and Rel-A (formerly referred to as p65)
(Baeuerle and Baltimore 1988 a,b). As a result of
experiments with cell extracts it was assumed that the
release of the inhibitory subunit IkB-a and the
activation of NF-KB was due to the phosphorylation of
IkB-a by protein kinase C (PKC) and other kinases
(Shirakawa and Mizel, 1989; Ghosh and Baltimore, 1990;
Kerr et al., 1991).
The following gene classes are controlled by NF-KB; they
all contain a decameric DNA motif with the consensus
sequence 5'-GGGPuNNTPyCC-3' which is recognised by NF-
KB: viral genes (HIV-1-, cytomegalo-, SV 40-,
2~59822
3
adenovirus), immune receptors (inter alia light
immunoglobulin-k-chains, T-cell receptor B, adhesion
molecule 1), cytokine (IFN-(3, GM-CSF, IL-2, IL-6, TNF-(x,
TNF-(3), acute phase proteins (inter alia
angiotensinogen), transcription factors (inter alia
"interferon regulatory factor-i", NF-KB precursor p50),
vimentine. (Baeuerle, 1991).
In view of numerous pathological conditions in which the
activation of the genes by the transcription factor NF-
KB takes part, there is a need for NF-KB inhibitors to
inhibit the transcription of genes which could have a
harmful effect on the organism.
Prior art solutions suggested for inhibiting the
expression of genes under the transcriptional control of
NF-KB, are based on a control of the dissociation of the
NF-KB/IkB-a-complex (WO 89/08147 and WO 92/20795).
The present invention has the task of explaining the
mechanism of the NF-KB activation and to provide
inhibitors based on this which specifically intervene in
this mechanism.
Surprisingly it was found that IkB-a (previously
referred to as MAD-3; Haskill et al., 1991) disappears
within minutes after the stimulation of cells with
phorbol ester, IL-1, LPS or TNF-a, a process coinciding
with the appearance of active NF-KB. The treatment of
cells with protease inhibitors or an antioxidant
prevents the inducible depletion of IkB-a as well as the
activation of NF-KB. Experiments carried out as part of
the invention showed that the activation of NF-KB by PMA
or other stimuli were obviously due to a transiently
increased degradation of IkB-a by a chymotrypsin-like
protease. Furthermore it was shown, that certain
protease inhibitors cause the accumulation of a
2159822
- 4 -
phosphorylated form of IkB-a. The direct
phosphorylation and subsequent inactivation of IkB-a by
PKC or other kinases are, however, in contrast to
earlier assumptions, insufficient to activate NF-KB in
intact cells.
The present invention refers to a process for inhibiting
the transcription of genes in higher eukaryotic cells by
inhibiting the activation of NK-KB. The process is
characterised in that the cells are treated with a
substance, which specifically inhibits the proteolytic
degradation of IkB-a.
The inhibition of the proteolytic degradation can be
direct or indirect. It is insignificant, whether the
proteolytic degradation of IkB-a takes place after its
dissociation from the p50/p65 complex or if IkB-a is
degraded by the proteolytic degradation as part of the
heterotrimeric complex.
Substances directly inhibiting the proteolytic
degradation are protease inhibitors.
In the experiments an effect of serine protease
inhibitors on IkB-a and the NF-KB activation was
noticed. As the substances tested were toxic compounds,
these compounds can not be used therapeutically. For
therapeutic use substances effecting a specific
inhibition of the proteolytic degradation of IkB-a by
inhibiting the activation of the IkB-a protease, as
specifically as possible, can be considered.
Protease inhibitors can function on the basis of their
analogy with the substrate, i.e. IkB-a, by competing
with IkB-a as substrate for the protease.
A further mechanism by which an inhibitor can impair the
_2159822
- 5 -
proteolytic degradation of IkB-a is by blocking the
accessibility of IkB-a for the protease, e.g. by
accumulating or bringing about a conformational change
(allosterically effective inhibitor).
The results from the experiments carried out for the
present invention show that IkB-a only becomes
accessible to proteolytic degradation by being
posttranslationally modified by phosphorylation.
As part of the present invention experiments were
carried out with the protease inhibitor Z-Ile-Glu
(OtBu)-Ala-Leucinal (hereinafter also referred to as
NBIG), which is known to inhibit a component of the
proteasome, having a chymotrypsin-like specificity.
Experiments with regard to the influence of this
inhibitor on the activation of NF-KB and the
stabilisation of IkB-a have shown that it prevents the
activation of NF-KB in HeLa cells after TNF stimulation,
the IDso being of the same magnitude as the proteasome
inhibition previously shown by this inhibitor. The
inhibitor furthermore has the special feature that it
causes an accumulation of more highly phosphorylated
form of IkB-a. The occurrence and stabilisation of this
phosphorylated form do not occur simultaneously with the
activation of NF-KB, confirming the assumption that a
phosphorylation in intact cells is insufficient for the
activation of NF-KB. This indicates that the inducible
phosphorylated form of IkB-a is still contained in the
complex with NF-KB; the phosphorylation of IkB-a
consequently does not cause a separation of IkB-a from
the complex, as was initially assumed from in vitro
experiments. The results achieved with NBIG show on the
whole that the proteasome is responsible for the
selective degradation of the phosphorylated form of
IkB-a.
2159822
- 6 -
From the results it can be assumed, that the only
function of the phosphorylation is to mark IkB-a for a
subsequent rapid proteolytic degradation. The resulting
lack of activity of the phosphorylated form explains why
up to now no cell-free activation of NF-KB was observed.
In addition, therefore, substances indirectly inhibiting
the proteolytic degradation of IkB-a by preventing the
modification through which the IkB-a is changed to the
form accessible to the proteolytic degradation, that is
the phosphorylated form and especially substances
inhibiting the kinase responsible for this, can be
considered as inhibitors for the activation of NF-KB.
A further option for an indirect mechanism of protease
inhibition is an inhibitor which prevents the activation
of the protease, i.e. a hypothetical protease inhibitor
being prevented from being released from the IkB-a
protease.
The inhibitors used in the invention differ in their
operation from substances inhibiting an activation of
NF-KB, by inhibiting the dissociation of IkB- from
pSO/Rel-A- heterodimer.
The prior suggested methods for detecting substances
negatively or positively influencing the NF-KB
activation were based on testing chemical substances for
their ability to stabilise the complex or to promote its
dissociation (WO 92/20795, WO 89/08147).
Other screening procedures are based on testing
substances for their ability to modulate the
transcription of a specially interesting gene, the
modulation of transcription being carried out at the
level of DNA binding (WO 91/01379).
CA 02159822 2004-07-15
25771-608
- 7 -
The present invention has on the other hand a
further purpose of offering a screening process based on
knowledge gained about the mechanism of NF-KB activation,
enabling the detection of substances with a high specificity
for NF-KB activation, by recording substances inhibiting the
proteolytic degradation of IkB-a.
Thus, according to one aspect of the present
invention, there is provided an ex vivo method for
identifying a test substance which inhibits NF-KB activation
by inhibiting proteolytic degradation of IkB-a, comprising:
a) treating an IkB-(x containing substrate, in the presence
of the test substance, with a preparation having proteolytic
activity for IkB-a; and b) determining whether and to what
extent the test substance inhibits proteolytic degradation
of IkB-a.
A further aspect of the invention refers to a
procedure for identifying substances inhibiting NF-KB
activation. The procedure is characterised in that the
substances are tested for their ability to specifically
inhibit the proteolytic degradation of IkB-a, by
a) treating an IkB-a containing substrate, in the
presence of a test substance, with a preparation showing
proteolytic activity for IkB-a, and determining, whether and
to what extent the test substance specifically inhibits the
proteolytic degradation of IkB-a, and if necessary
b) and in the presence of the test substance,
inducing NF-KB activation in higher eukaryotic cells,
especially human cells that have been transformed with a
reporter gene construct responding to the NF-KB activation
and measuring the expression of the reporter gene.
CA 02159822 2004-07-15
25771-608
- 7a -
In a preferred embodiment of the process purified
IkB-a is used in a so-called "cell-free" assay as the
substrate of step a).
This can for instance be carried out by binding a
defined quantity of preferably recombinant IkB-a (Henkel
et al., 1992; Zabel et al. 1993), marked so as to be
measurable, to a fixed carrier.
A preparation showing proteolytic activity for
IkB-a is
2159822
- 8 -
applied to the immobilised IkB-a in presence of the test
substance. This preparation could either be a so-called
"activated" cell extract or IkB-a protease produced by a
recombinant method, as soon as it is available. If the
assumption is correct, that the IkB-a protease is a
component of the proteasome (Orlowski, 1990; Rivett,
1989), this can be used in a purified form or a part
thereof as the proteolytic activity.
An activated cell extract showing IkB-a degradation
activity is obtained from cells which were treated with
a known NF-KB inducer such as LPS, phorbol ester, TNF,
etc. The induction of NF-KB activity effects an
activation of the system responsible for the proteolytic
degradation of IBK-a. The cell extract preferably
contains no IkB-a degrading activity in its non-
activated state.
The activated cell extract can be used to detect direct
and indirect inhibitors of the proteolytic IkB-a
degradation. If using purified IkB-a protease, only
inhibitors acting directly on the protease are
registered.
As a preparation with proteolytic activity for IkB-a, a
preparation of induced cells may furthermore be used in
cell-free assays, being preferably a fraction of the
activated cell extract in which IkB-a proteolytic
activity was proven. The protein chemical methods for
fractionating the cell extracts are known to experts.
The fractions obtained for instance after ammonium
sulphate precipitation, gel filtration and/or ion
exchange chromatography, isoelectric focusing, etc., are
tested in pretrials for their ability to proteolytically
degrade IkB-a and the respective fraction is used for
the assay, the quantity being coordinated with the test
substrate IkB-a in such a way that the degradation can
2159822
- 9 -
be measured.
In the absence of a test substance (control), or with a
test substance not having an inhibiting effect on the
proteolytic activity, the immobilised IkB-a is
proteolytically degraded. This is shown by the fact,
that the marked form of IkB-a (e.g. a radioactive or a
fluorescent marker) passes completely or in part (during
complete or partial degradation of IkB-a) from the
carrier to the supernatant. The proportion of the
marker having changed from the carrier to the
supernatant is proportional to the proteolytic
degradation of IkB-a. In the presence of a test
substance, inhibiting the proteolytic degradation of
IkB-a, the immobilised IkB-(x is not degraded, the
marking remains on the carrier and none passes to the
supernatant.
Alternatively a stipulated quantity of marked IkB-a may
be present in soluble form instead of being immobilised
on a carrier during step a), whilst the other test
parameters are identical to the above test arrangement.
When applying a preparation with proteolytic activity
the IkB-a of the solution is broken down into smaller
species. The components of the solution are then
separated, e.g. by dialysis or gel filtration, the pore
size of the membrane or of the resin being chosen so
that the intact IkB-a is separated from the proteolytic
fragments. The marked fragments or the non-degraded
marked IkB-a substrate are now in the eluate, with the
fraction of the fragments to the total being
proportional to the proteolytic degradation of IkB-a.
In the presence of a test substance inhibiting the
proteolytic activity, the IkB-a contained in the
solution is not degraded and consequently no marked
fragments can be detected.
2159822
- 10 -
A further alternative for a cell-free assay in step a)
is the generation of a IkB-a complex with p50 and/or
p65, the components being preferably of recombinant
origin, as soon as this is completely available, and
IkB-a having a measurable marker to treat this complex
in the presence of the test substance with a preparation
having proteolytic activity. If proteolysis occurs,
which is the case if the test substance is not able to
inhibit this activity, the IkB-a is digested by the
complex and the marker is on the proteolytic fragments
which after filtration will, in this case too, be
detectable in the eluate. In case of an inhibitory
effect of the test substance on the proteolytic
activity, IkB-a remains associated with the complex and
no marker passes into the eluate. In principle the
proteolytic degradation of IkB-a can also be observed
with the change of the fluorescence of the complex,
depending on the structural change.
The findings of the present invention, that the protease
inhibitor NBIG effects the accumulation of a
phosphorylated form of IkB-a, enables a variant cell-
free assay in which the phosphorylated form of IkB-a can
be used as substrate for the protease. After having
initially tested by Western blotting with IkB-a
antibodies or ELISA with the addition of proteases, e.g.
the purified proteasome or protease containing cell
fractions, whether phosphorylated IkB-a is indeed the
form of the substrate inactivated by a protease
(constitutive) for the degradation, this substrate can
be generated as follows:
Suitable human cells, expressing NF-KB and IkB-a, i.e.
HeLa cells or 293 cells (ATCC CRL 1573), are treated
with a protease inhibitor, which caused in pretrials an
accumulation of a phosphorylated form of IkB-a, for
instance with NBIG at a concentration of 75 m. The
2159822
- 11 -
cells are then stimulated with an inducer of NF-KB
activation for a period required for this activation,
i.e. with TNF-a for 15 minutes, or with PMA or IL-1. The
NF-KB-IkB-a complex from cytoplasmatic extracts of
treated cells containing IkB-a in phosphorylated form,
is accumulated, if necessary, for example by a glycerine
gradient, gel filtration or ion exchange chromatography,
and is separated from surplus NBIG and from this
inactive endogenous protease. To transform IkB-a
completely into the phosphorylated form, preferably a
phosphatase inhibitor such as okadaic acid is added in a
suitable concentration, e.g. 100 nmol.
The phosphorylate IkB-a thus obtained, being the
substrate for the protease and leading to the activation
of NF-KB, may be used in the above assays for finding
inhibitors of the activation of NF-KB (alternatively
IkB-a may be provided in phosphorylated form, by
phosphorylating recombinant IkB-a in vitro. The
principle of the assay remains the same: without
protease inhibitors (without the addition of test
substances or in the presence of test substances having
no inhibitory effects) the phosphorylated substrate is
degraded. In the presence of the inhibitors the
phosphorylated form of the NF-KB-IkB-(x complex remains
detectable.
To narrow down the positive results of step a) with
regards to the specificity of the test substances for
the proteolytic degradation of IkB-a, additional
controls are required:
An inhibitor found for instance in step a) is tested to
determine whether it affects the activity of other
proteases. This is done in such a way that the effect
of the inhibitor on the activity of the protease having
a different substrate to the IkB-a is analysed.
2159822
- 12 -
To establish whether the inhibitor reacts indirectly,
that is on the IkB-oc substrate, the effect of the
inhibitor with a known protease is tested, e.g.
chymotrypsin (pretrials must have established that the
inhibitor does not directly influence the activities of
this control protease).
The results obtained from the present invention suggest
that the IkB-a protease is a chymotrypsin-like serine
protease. After confirmation of these findings in the
cell-free system by the inhibitor TPCK, the screening
procedure of the invention can be extended by testing
the main derivatives of serine protease inhibitors. As
a control, the effect of the established inhibitor on
the serine protease, e.g. chymotrypsin on one hand and
on the IkB-a protease on the other, is tested. (In case
that, contrary to expectations, these findings are not
confirmed, one can proceed analogously by testing the
main derivatives of inhibitors of the protease class to
which the IkB-(t protease is definitely assignable, and
implementing the respective controls.)
The selection of the most suitable assay variations is
made through pretrials, e.g. whether IkB-oc on its own,
or in a phosphorylated form or in association with its
complex partners p50 and/or p65 serves as test
substrates as well as the specific assay conditions.
The other assay conditions, such as the method and
concentration of the inducer, duration and condition of
the induction, cell disintegration, IkB-oc quantity or
that of possibly present complex partners, carriers and
binders of IkB-oc or of the complex partners to the
carrier (binding to microtiter plates using a bivalent
cross linker such as glutardialdehyde, to sepharose via
cyanogen bromide, etc.), markers (coupled enzyme,
radioisotope, fluorescent, chemiluminescent or
2159822
- 13 -
bioluminescent substances, etc.), period of treatment,
arrangement and number of controls, etc. are optimised
via pretrials.
The assay conditions in step a) are preferably chosen so
that the assay can be automated.
In the case that the specificity of the inhibitor within
the assay carried out using the cell extract was not
completely clear with regards to the inhibition of the
IkB-a protease, a so-called cellular assay (step b) is
carried out after the cell-free assay or parallel to it,
in which the effect of the test substance on the
activation of NF-KB in the intact cell is established.
Step b) is preferably carried out in any case to confirm
an inhibitory effect of the test substance in an intact
cell found in the cell-free assay.
The reporter gene construct with which the test cell
used in b) is transformed, is hereinafter referred to as
"sensor-DNA". This refers to a DNA construct containing
a reporter gene being controlled by regulatory sequences
and responding to the activation of NF-KB by containing
at least one binding sequence for NF-KB. During
activation of NF-KB, NF-KB binds to the recognition
sequence after which the expression of the reporter gene
is initiated, giving a measurable signal.
The sensor DNA is preferably located on a plasmid which
is highly reproducible in a suitable host organism,
preferably E. coli and facilitates the expression of a
reporter gene under the control of regulatory elements
after transfection into mammalian cells and integration
into the host genome. For this preferably a shuttle
vector is chosen containing an expression cassette for
the reporter gene (sensor DNA) and a selectable marker
for mammalian cells as well as at least one replication
215g$22
- 14 -
origin and a marker for the selection in E. coli.
To produce permanent cell lines containing stable sensor
DNA integrated into their genome, the vector contains a
dominant selection marker. The use of a specific
selection marker is non-critical, for which, for
instance, the gene for neomycin-phosphotransferase
(neo), offering resistance against antibiotic geneticin
(G-418) (Southern and Berg, 1982) the DHFR gene
(Dihydrofolatreductase) for DHF-deficient cells, the
gene for xanthine-guanine-phosphoribosyltransferase
(gpt), offering resistance against mycophenolic acid
(Mulligan and Berg, 1981) or the hygromycin-B-
phosphotransferase gene (hph; Gritz and Davies, 1983)
are suitable. Examples of promoters driving the
selection marker gene are the SV40 early promoter, the
cytomegalovirus promoter (CMV promoter), the promoter of
the thymidine kinase gene of the herpes simplex virus
(TK promoter), the Rous sarcoma virus (RSV) long
terminal repeat (LTR). The plasmids are preferably
constructed in such a way that individual important
elements such as the reporter gene, the promoter for the
reporter gene or the regulatory sequences for the
selection marker can be exchanged or changed to
correspond to possible changed requirements resulting
from the particular application, e.g. due to the use of
a different cell line. Such measures would be for
instance to install multi-cloning sites before the
promoter(s) or the reporter gene to facilitate the
cloning of regulatory sequences, modulating the
promoter, from various reporter genes.
When selecting a suitable reporter gene it is assumed
that preferably a non-radioactive, automatable assay of
high sensitivity is provided.
In principal all reporter genes fulfilling these
- 15 -
preconditions can be used for the present invention:
Alkaline phosphatase can be measured if using a
chemiluminescent substrate with high sensitivity,
although it has the disadvantage that this enzyme is,
relatively strongly expressed by many mammalian cells.
It is consequently only suited for cell lines which do
not express it or express it only to a limited extent.
The expression products of the (3-galactosidase and the
(3-glucuronidase gene can cleave the respective
methylumbeliferyl-galactoside or glucuronide whilst
forming fluorescent groups. These enzyme reactions are
observed by using established fluorescent assays
(Wieland et al., 1985; Kricka, 1988).
Expression of chloramphenicol actetyltransferase (CAT)
can be detected with relative sensitivity, the assay
has, however, inter alia the disadvantage that it is
radioactive and difficult to automate (Hartmann, 1991).
Within the context of the present invention the gene
encoding Photinus pyralis-luciferase (De Wet et al.,
1987) is used as a reporter gene. This enzyme has the
advantage that together with its substrate luciferone it
generates a high level of bioluminescence when ATP is
added, which can be measured with established,
automatable methods, and that this enzyme is not
produced endogenously by mammalian cells. Luciferase
also has a relative short in vivo half-life and is non-
toxic even in high concentration (Hartmann, 1991;
Brasier et al., 1989). The measurement of the activity
of firefly luciferase using bioluminescence is one of
the most sensitive methods for detecting enzymes.
Consequently, and due to the absence of luciferase
activity in normal mammalian cells, this enzyme is
especially suitable as a reporter gene (Subramani and
- 16 _2159822
DeLuca, 1987).
Alternatively the gene coding for the enzyme
apoaequorine of the jellyfish aequoria victoria
(Tanahashi et al., 1990) can be used as reporter gene.
This enzyme has the advantage that it generates large
amounts of bioluminescence together with its co-factor
coelenterazine after binding calcium ions, which can be
measured with established automatable methods. Another
advantage is, that this enzyme is not endogenously
expressed by mammalian cells.
The reporter gene is driven by a promoter, which is
independent of other factors and makes the induction of
gene expression clearly visible, e.g. the thymidine
kinase promoter. The promoter must transcribe the
reporter gene it drives in a correct and measurable way.
Minimal promoters, for instance the minimal TK-promoter
and the minimal-(3-globin promoter, are available. For
the cellular assays carried out as part of this
invention, the -105 - +52 TK promoter was used, having a
relatively high basal activity and being easily
stimulated; in pre-trials the -30 and the -87 minimal-TK
promoters were also proven to be suitable.
In principle any of the consensus sequences 5'-
GGGPuNNTPyCC-3' are suitable as binding motifs for NF-
KB, e.g. the well characterised motif of the light
immunglobulin K-chain, having the sequence 5'-
GGGACTTTCC-3' may be used. The NF-KB binding motif
appears for this purpose at least twice within a short
distance (approx. 10 nucleotides). For the experiments
six KB motifs were used.
In principle NF-KB binding sequences can be used from
any gene which is controlled by NF-KB. (Baeuerle,
1991). If the genes also contain binding domains for
2159822
- 17 -
other transcription factors, apart from the NF-KB
binding sequence, these are preferably removed.
Control cells are preferably cells containing no DNA
binding motif for NF-KB in their reporter gene. The
control cells facilitate the detection of a non-specific
effect of the substance on gene expression; for
instance, a signal contained in the control cell would
originate from baseline transcription and should
therefore be deducted from the signal contained in the
test cell.
The test cells must fulfill the condition that
endogenous NF-KB can easily be activated inside them.
NF-KB should not be constitutive in these cells.
The cells should also be induciable with substances
activating NF-KB, such as PMA, LPS, H202, UV.
Cells can be tested for their suitability as test cells
by transforming them with the sensor DNA; and the
kinetics of induction of reporter gene expression
determining the concentration dependence of the cell
inductor.
For an automatic process the cells should be as adherent
as possible.
In the procedure according to the invention, substances
directly or indirectly inhibiting the proteolytic
degradation of IkB-a and thus the activation of NF-KB
are detected, as well as substances inhibiting the
phosphorylation of IkB-a. It is insignificant for the
inhibitory effect, whether the protease itself is
specific to IkB-a, essential is the specificity with
regard to the activation.
18 2159822
- - '
In a further aspect, the invention relates to substances
which specifically inhibit the proteolytic degradation
of IkB-oc, for the treatment of pathological conditions,
to which expression of genes controlled by the
transcription factor NF-KB contribute.
Pathological conditions caused by an adverse effect of
gene expression by NF-KB activation include, inter alia,
inflammatory illnesses resulting from an activation of T
cells, macrophages or B cells, toxic shock, illness
after infection by a virus containing the KB motif, UV-
damage (sunburn), radiation damage, burns, transplant
rejection, reperfusion damage.
For therapeutic use, substances identified according to
the procedure of the invention are formulated, which are
then characterised in more detail for the development of
medication in the usual way with regard to their
pharmacological characteristics, e.g. in secondary
screening and animal tests, depending on the delivery
with suitable pharmaceutically acceptable carrier and
auxiliary substances, guaranteeing the bioavailability
of therapeutically effective substances and not having
damaging effects on the organism. Methods for the
formulation of pharmaceutic preparations can be found in
standard text books, e.g. Remington's Pharmaceutical
Sciences, 1980.
As part of the present invention, the effect of protease
inhibitors on the proteolytic degradation of IkB-a was
determined. The inhibition of NF-KB activation in
induced test cells containing an NF-KB-responsive
luciferase gene construct was confirmed, as well as
inhibition of the NF-KB regulated expression of IL-6 and
IL-8. As part of the present invention the fate of the
NF-KB specific inhibitory subunit IkB-a was observed
after the treatment of cells with NF-KB activating
19 2159822
- -
stimuli. For this purpose, highly purified recombinant
human IkB-a was used to generate a rabbit antiserum;
specific IgG was affinity-purified by immobilised IkB-a.
The IkB-a specific polyclonal IgG recognised a single
38 K band at the correct size for IkB-a (Fig. 1A, lane
1) in Western blots in a total cell extract of mouse
70Z/3 pre-B cells. This band was not recognised by a
control antibody (anti-rabbit IgG antibody) (Fig. 1A,
lane 2). Between 2 and 5 minutes after the addition of
phorbol-l2-myristyl-13-acetate (PMA) to the 70Z/3 cell
cultures the IkB-a band disappeared almost completely
from the cells (Fig. 1B, compare lanes 3 and 4). The
cells then showed no IkB-a immunoreactivity until 40
minutes after stimulation (Fig. 1B, lane 7). Aliquots
of the same cell extract were analysed using the
Electropheric Mobility Shift Assay (EMSA) for KB-
specific DNA binding activities. As apparent from Fig.
1C (compare lanes 2 and 3), the disappearance of IkB-a
coincided exactly with the appearance of the NF-KB-DNA
binding activity, suggesting a causal relationship
between these two events.
NF-KB may also be activated in 70Z/3 cells by treatment
with IL-1(3 and LPS. The activation of NF-KB by TNF-.a
can be studied in HeLa cells. As shown in Fig. 2, IL-
10, LPS and TNF-a all induced a decay of IkB-a in 70Z/3
or HeLa cells. There were, however, no kinetic
differences with regard to the start of the
disappearance of IkB-a. Most of the IkB-a had already
decayed 5 minutes after stimulation of the 70Z/3 cells
with PMA (Fig. 1B) or of the HeLa cells with TNF-a (Fig.
2). When stimulating the 70Z/3 cells with IL-1B, the
decay started somewhat later; after 5 minutes of
stimulation more IkB-a remained than had been observed
in the cells treated with PMA or TNF-a (Fig. 2B). In
70Z/3 cells stimulated by LPS most of the IkB-a was not
degraded before 30 minutes from the induction (Fig. 2).
20 - 2159822
-
Again and for all three inducers, the disappearance of
IkB-a coincided with the appearance of NF-KB-DNA binding
activities in the cells. These observations show that
four different inducers of NF-KB use a mutual mechanism
of NF-KB inactivation, contributing to a decay of IkB-a.
The varying kinetics of the degradation suggests that
the inducers use different signal transduction paths
before NF-KB activation. Based on the fact that a
polyclonal antibody was used, it is unlikely that only
an epitope of IkB-a was lost or modified after the
stimulation. It is more likely that the loss of IkB-a
immunoreactivity is due to a rapid and complete decay of
the protein leading to no immunologically detectable
degradation product. After stimulation with PMA, IL-1B,
LPS or TNF-a, no change of the electrophoretic mobility
of IkB-a in SDS gels was visible, indicating a
posttranslational modification of IkB-a before this
degradation.
It is known that the protein synthesis inhibitors
cycloheximide and anisomycin, activate NF-KB in 70Z/3
cells (Sen and Baltimore, 1986). The one-hour treatment
of 70Z/3 cells with cycloheximide carried out within the
context of this invention induced only slightly the
binding of NF-KB to DNA (Fig. 3A, lane 1, top panel).
In these cells large amounts of IkB-a could still be
detected by Western blotting (Fig. 3A, lane 1, lower
panel). This leads to the conclusion that the
inhibition of the normal IkB-a transformation is
insufficient for an efficient and rapid activation of
NF-KB. If cells pre-treated with cycloheximide were
induced with PMA a rapid decay of IkB-a and consequently
an induction of the binding of NF-KB to DNA may be
observed (Fig. 3A); the early kinetics of the
established IkB-a reduction could not be differentiated
from those observed with PMA alone (compare Fig. 3A with
Fig. 1B). The protein synthesise inhibitor cyclo-
21 - 2159822
-
heximide prevents, however, the reoccurrence of IkB-a,
which was observed 40 minutes after the treatment with
only PMA (compare Fig. 3A with Fig. 1B), suggesting that
this IkB-a was newly synthesised, possibly under
transcriptional control of NF-KB itself.
Furthermore, the half-life of IkB-a in protein synthesis
arrested cells was determined without or with subsequent
PMA stimulation of the cells. The half-life of IkB-a in
cycloheximide-treated 70Z/3 cells was approx. 138
minutes, as analysed by quantitative Western blotting
(Fig. 3B). The half-life of IkB-a was reduced to only
1.5 minutes during the period of its most rapid
degradation, 2 to 5 minutes after the PMA stimulation;
this shows that PMA induces an approx. 90-fold
degradation of IkB-a.
To determine whether the inducible degradation of IkB-a
is a necessary step for the activation of NF-KB, cell
cultures were treated with protease inhibitors. Of the
various substances found to be active, the chymotrypsin
inhibitor p-tosyl-L-phenylalaninechloromethylketone
(TPCK) was the most potent inhibitor of NF-KB activation
in the various cell types. A one-hour treatment of
70Z/3 cells with 25 M TPCK was sufficient to completely
suppress the NF-KB-DNA binding activity in a subsequent
PMA treatment (Fig. 4, top panel). The ID50 of TPCK was
approx. 8 M. The constitutive DNA binding activity,
including nuclear-factor oct-1, was not significantly
influenced (Fig. 4A). As shown by the Western blot, the
protease inhibitor was able to completely prevent the
degradation of IkB-a in PMA treated cells (Fig. 4A,
lower panel). TPCK had an inhibitory effect even when
added simultaneously with the PMA and was at the same
time effective in preventing the activation of NF-KB by
IL-IB and LPS in 70Z/3 cells (Fig. 4A, lanes 6 and 7)
and with TNF-a in various other cell lines. The
22 - 2159822
-
treatment of cells with the protease inhibitor after
stimulation with PMA, IL-i or LPS hardly impaired the
NF-KB activation, but should, however, have stopped
further activation (Fig. 4A, lanes 10-12). This shows
that TPCK did not just simply impair the DNA binding of
NF-KB or lead to the degradation of NF-KB, processes
which had occurred during the cell lysis. It is equally
unlikely that TPCK inhibited PKC, as IL-1B and TNF-ot
were shown to activate NF-KB independently from PMA-
induciable PKC isoenzymes (Bomsztyk et al., 1991;
Meichle et al., 1990).
p-tosyl-L-lysinchloromethylketone (TLCK) is an inhibitor
of trypsin-like serine proteases and is quite similar to
TPCK in structure and chemical activity. TLCK was
unable to prevent the activation of NF-KB in 70Z/3 cells
at a concentration of 25 M (Fig. 4B, lane 3). At a
concentration of 100 M TLCK, however, a partial
inhibition of the NF-KB activation was observed.
Various other protease inhibitors were effective, albeit
at higher concentrations. The selective and strong
inhibitory effect of TPCK suggests that a chymotrypsin-
like serine protease is part of the NF-KB activation.
This protease is hereinafter referred to as IkB-a
protease.
Recently, it was reported that reactive oxygen compounds
could play a role as messenger substances in the
activation of NF-KB through many inducible factors
(Schreck et al., 1991; Schreck et al., 1992 a,b). This
assumption was partially based on the fact that
antioxidants such as thiol compounds, dithiocarbamate
and chelating agents for free iron, suppress the
activities of NF-KB in intact cells. A very strong
antioxidant inhibitor of the NF-KB activation was
pyrrolidindithiocarbamate (PDTC; Schreck et al., 1992b).
If the degradation of IkB-a plays a significant part in
~1598~2
- 23 -
the activation of NF-KB, PDTC should, similarly to TPCK,
prevent the disappearance of IkB-a, although this
inhibition could operate due to a mechanism operating
prior to the IkB-a protease. 100 M of PDTC effectively
suppressed the activation of NF-KB binding activities in
PMA treated 70Z/3 cells (Fig. 5). As expected no
significant degradation of IkB-a protein was shown by
the Western blot. Recently it was shown that PDTC does
not interfere with the PMA induced membrane association
and the kinase activity of PKC, arguing against a direct
effect of PKC on IkB-a in intact cells.
From the results of the present invention it can be
concluded that the proteolytic degradation of IkB-a in
response to PMA, IL-1R, LPS and TNF-a is a necessary
step in the process of NF-KB activation. This is not
only apparent from the near simultaneous coupling of
IkB-a degradation with the NF-KB activation, but also
and more importantly from the selective inhibitory
effect observed at low concentrations of the well
characterised protease inhibitor TPCK. The inducible
degradation was caused by a dramatic reduction in
stability of the protein and not by blocking de novo
synthesis. The normal half-life of the inhibitory
subunit could explain the weak activation of NF-KB by
protein synthesis inhibitors, but this is definitely
insufficient to induce a rapid and complete depletion of
IkB-a and to potently activate NF-KB as shown by PMA and
other stimuli.
An increased degradation of IkB could be controlled by
various mechanisms: Firstly, a new modification of the
IkB protein could make the inhibitor more susceptible to
a degradation by a constitutive protease. Secondly a
protease selectively breaking down IkB could be
activated. An interesting aspect of this possibility is
that the IkB-a protease or a protease inhibitor can be a
24 - 2159822
-
direct upstream target for messenger substances such as
protein kinases. Thirdly a modification of IkB or an
induction of protease activity could be required. The
results available to date cannot decisively
differentiate between these possibilities.
The present invention also raises the question, to what
extent direct phosphorylation of IkE contributes to the
release of IkB and the activation of NF-KB in intact
cells. Based on the investigations carried out so far,
no phosphorylation data from intact cells is available,
suggesting an induciable phosphorylation of IkB-a,
although a constitutive phosphorylation of IkB-a was
observed. The present results only agree with the in
vitro phosphorylation data (Shirakawa and Mizel, 1989;
Ghosh and Baltimore, 1990; Kerr et al., 1991) under the
assumption that degradation is required to quickly
eliminate the IkB released by the phosphorylation.
Degradation could be required to prevent IkB, once
dephosphorylated, from inhibiting new NF-KB. In view of
the strong inhibitory effect of the protease inhibitor
TPCK, however, phosphorylation of IkB-a should be too
transient to activate NF-KB and to permit its transport
into the cell nucleus or to be detected by the usual
methods.
A controlled proteolytic degradation of IkB would be an
excellent mechanism to make the activation of NF-KB
irreversible. The only option to inhibit activated NF-
KB would in this case be newly synthesised IkB. An
important question is whether the IkB-a protease attacks
IkB-a in the cytoplasmatic complex with NF-KB, or
whether a prior release of IkB from NF-KB is required.
Experiments have shown that overexpressed IkB-a is not
significantly degraded by a TNF-a stimulation, which,
however, is the case with the endogenous version. This
suggests that only the IkB associated with NF-KB is a
25 - ,-2159822
-
substrate for the IkB-a protease making the requirement
of identification, but not modification of IkB-a by a
direct phosphorylation, obsolete. A specific structure
for a subunit of a transcription regulator has already
been established for the a2 repressor from yeast
(Hochstrasser and Varshavsky, 1990).
Specific inhibitors of the IkB-a protease are suitable
as pharmaceutical agents for preventing the activation
of NF-KB, which is responsible for various pathological
conditions. Examples of the numerous biomedically
important conditions to which NF-KB contributes
significantly as signal transferrer and activator of
immediate-early genes, are the progression of AIDS, the
activation of T-cells, B-cells and macrophages during
the immune response, the so-called acute phase response,
toxic shock, transplant rejection and the response of
the cell to gamma radiation and W light. Specific
inhibitors of the IkB-a protease are effective as anti-
inflammatory and immune suppressive drugs.
Figure overview
Fig. 1: Fate of IkB-a in stimulated 70Z/3 pre-B cells
Fig. 2: Effect of IL-1(3, LPS and TNF-a on the
activation of NF-KB and the stability of IkB-a
Fig. 3: Effect of a protein synthesis inhibitor on the
stability of IkB-a and the activation of NF-KB
in 70Z/3 cells
Fig. 4: Effect of the protease inhibitors TPCK and TLCK
on the activation of NF-KB and the stability of
IkB-a in 70Z/3 cells
Fig. 5: Effect of PDTC on the activation of NF-KB and
the stability of IkB-a in 70Z/3 cells
Fig. 6: Effect of the protease inhibitor TPCK on the
activation of a reporter gene by NF-KB in HeLa
2159822
- 26 -
cells
Fig. 7: Effect of the protease inhibitor Z-Ile-Glu (Ot-
Bu)-Ala-leucinal and Z-Leu-leucinal on the
activation of NF-KB and the stability of IkB-a
in HeLa cells.
Fig. 8: Testing of the protease inhibitors Z-Ile-
Glu(Ot-Bu)-Ala-leucinal and Z-Leu-leucinal;
inhibiting effect of the proteolytic
degradation of IkB-a induced by TNF-a
Fig. 9: The context of the induction of the NF-KB
activation using TNF-a and the appearance of
phosphorylated IkB-a
Example 1
Fate of IkB-a in stimulated 70Z/3 pre-B cells
70Z/3 cells (ATCC No. TIB 158) were cultivated in RPMI-
1640 medium (Gibco BRL), supplemented by 10% FCS (Gibco
BRL) and 50 M 2-mercaptoethanol. 2 ml aliquots with
approx. 2 - 3 x 106 of suspended cells was treated with
50 ng/ml PMA (Sigma) for various periods. The treatment
was interrupted by immediate centrifugation (5 s in an
Eppendorf Microfuge) and cooling on ice. The cell
pellets were lysed with 60 l of a high-salt extraction
buffer, containing the non-ionic detergent nonidet P-40
(Baeuerle and Baltimore, 1988b). The supernatant of a
15 minute centrifugation at 13.000 rpm in an Eppendorf
Microfuge was analysed by Western blotting and EMSA
(electrophoretic mobility shift assay). For SDS-PAGE
and Western blotting 30 l aliquots of the extracts were
mixed with 15 l SDS sample buffer (Laemmli, 1970),
boiled and subjected to a SDS-PAGE on 12.5%
polyacrylamide minigels (Biorad). The proteins were
transferred by using a so-called semi-dry blotting
device (Biorad; 1 h at 15V/2.5mA/cm2) from the gels onto
Immobilon-P-filter (0.45 m; Millipore). The efficiency
CA 02159822 2004-07-15
25771-608
- 27 -
of the blot was checked by protein staining of the
filters with Ponceau S (Serva). The filters were
blocked overnight in a Tris-buffered saline solution,
containing 0,1% (v/v) Tween-20~(TBST), 5g low-fat milk
powder (Nestle) and 1% BSA. The filter was then
incubated for one hour at ambient temperature with anti
IkB-a IgG in a dialysed eluate of the IkB-a affinity
column and diluted by 1:100 in the blocking buffer.
After washing for 30 minutes in TBST the filter was
incubated in a 1:4000 solution of goat-anti-rabbit
IgG/horseradish-peroxidase conjugate (Biorad) in
blocking buffer. After washing for 30 minutes in TBST
the filter was treated with ECL detecting reagent
(Amersham) and exposed (Kodak XR film, less than 1
minute). For EMSA 2 l aliquots of the extracts were
added to a binding mixture, resulting in a final
concentration of 100 nM KC1, 20 nM HEPES, pH 7.9, 2.5 nM
dithiothreitol, 0.5 nM phenylmethylsulfonylfluoride
(PMSF), 0.2a Nonidet P-40, 5 % Ficoll; 20 g BSA, 3 g
Poly (dl-dC) and 10.000 cpm (Cerenkov counting method)
of a 32P marked double-stranded KB-oligonucleotide probe
(Gibco BRL). The specificity of the protein DNA complex
was confirmed by competition experiments and
supershifting with a polyclonal antibody, recognising
the C-terminal 100 amino acids of Rel-A (p65). After 20
minutes of incubation on ice the samples were subjected
to electrophoresis on native 411 polyacrylamide gels, run
in 0.5 x TBE. The dried gels were exposed (Kodak XR
film, -70 C, overnight).
Human IkB-a was produced in E. coli and purified, as
described by Zabel et al., 1993. An antiserum against
IkB-a was produced in rabbits, the specific IgG in 2 ml
serum, affinity purified to IkB-a sepharose (Henkel et
al., 1992). 2 mg of 6 x His-marked, affinity-purified
human IkB-a were coupled to 0.5 ml cyanogen bromide
activated sepharose 4CL-B (Pharmacia) according to the
*Trade-mark
- - --------- - -----
2159~22
- 28 -
manufacturer's instructions. After thorough washing
with PBS and 2 column volumes 0.1 M glycine-HC1, pH 2.7,
specific antibodies were eluated with 2 volumes 4 M
guanidinhydrochloride. The eluate was dialysed
thoroughly with TBST.
The results of the experiments are listed in Fig. 1. A:
Specificity of an affinity-purified polyclonal rabbit-
anti-IkB-a antibody in Western blots. As stated above,
the proteins were separated by non-stimulated 70Z/3
cells in a high-salt total cell extract by sodium
dodecylsulphate-polyacrylamide gel electrophoresis (SDS-
PAGE) and transferred to a membrane filter. A strip of
the filter was incubated with anti-IkB-a and a
peroxidase labelled anti-rabbit IgG (lane 1). A
duplicate filter was incubated on its own with the
second antibody (lane 2). The figure shows the
fluorograms of the chemiluminescence marked filter. The
molecular weights are stated in kDa. The arrow shows
the position of a single 38 K-band. Phosphorylase (97),
bovine serum albumin (BSA, 67), ovalbumin (45) and
carboanhydrase (30) were used as molecular weight
standards. B: Effect of a treatment of cells with PMA
on IkB-a. The cells were treated with PMA for the
period stated in the figure (in minutes) (lanes 2-9).
The zero value (lane 1) stems from untreated cells.
After treatment the cell extracts were subjected as
stated to a SDS-PAGE and were Western blot tested for
IkB-a by using anti-IkB-a IgG. The arrow indicates the
position of IkB-a. The figure shows an extract of a
fluorogram. C: Effect of a treatment of cells with PMA
on the DNA binding activity of NF-KB. Aliquots of
extracts from control cells (lane 1) and PMA treated
cells (lanes 2-6; the treatment period is stated at the
top in minutes; the number underneath the figure shows
the lane) were incubated, as stated, with the 32P-marked
DNA-probe containing the NF-KB binding motif of the
2159822
- 29 -
enhancer of the gene "mouse k light chain enhancer";
(Sen and Baltimore, 1986) after which an analysis of the
DNA binding activity was carried out by using gel
electrophoresis. The figure shows a fluorogram of a
native gel. The filled arrowhead shows the portion of
the NF-KB-DNA complex, the open arrowhead the position
of the non-complexed DNA probe.
Example 2
Effect of IL-1(3, LPS and TNF-a on the activation of NF-
KB and the stability of IkB-a
The methods used in this example for the culture of the
70Z/3 cells and the production of the cell extracts were
implemented as stated in example 1. The 70Z/3 cells
were treated with 50 E/ml IL-1(3 (Boehringer Mannheim) or
with 15 g/ml LPS (Sigma). Human HeLa cells were
cultivated in DMEM, supplemented with 10oFCS and 1o L-
glutamine and treated with 200 E/ml human recombinant
TFN-a (genzyme).
Figure 2 shows the treatment of 70Z/3 cells in the upper
and middle panel with IL-1(3 (lanes 2-4) and LPS (lanes
6-8) for the individually stated, different periods (in
minutes), followed by a test of the total cell extract
with regards to the DNA binding activity of NF-KB and
the IkB-a immune reactivity, using EMSA or Western
blotting. The lower panel shows the results of the
treatment of HeLa cells with TNF-a (lanes 10-13). This
figure shows sections of Western blots and sections of
fluorograms of native gels. The position of the NF-KB
DNA complex in the EMSAs is shown by the filled
arrowheads. The position of the IkB-a band on the
Western blot is shown by an arrow.
215 9822
- 30 -
Example 3
Effect of a protein synthesis inhibitor on the stability
of IkB-a and the activation of NF-KB in 70Z/3 cells
70Z/3 cells were treated with 25 g/ml cycloheximide
(Sigma) (as determined by Wall et al., 1986, already at
a concentration of 10 )ug/ml the protein synthesis is
inhibited in 70Z/3 cells by 900). Cell culture, extract
production, SDS-PAGE and Western blotting was carried
out according to example 1. The densitometric test was
then carried out by a Howtek Scanmaster 3; the data was
evaluated with Quantity One Version 2.2 software.
The result of the experiments is shown in figure 3: A:
Effect of cycloheximide on the DNA binding of NF-KB and
the cellular concentration of IkB-a. The cells were
pre-treated with cycloheximide for one hour, followed by
a stimulation with PMA for the stipulated period (in
minutes). The top panel shows the test of the cell
extracts for the DNA binding activity of NF-KB using
EMSA, showing a section of the fluorogram of a native
gel; the filled arrowhead shows the position of the NF-
KB-DNA complex. The lower panel shows the test of the
cell extract aliquot on IkB-a using Western blotting;
the position of the 38 K IkB-a band is shown by an
arrow. B: Stability of IkB-a without and in the
presence of PMA. The left panel shows the treatment of
cells with only cycloheximide; the right panel shows the
treatment of cells with PMA (treatment duration is
stipulated in the figure) after a one-hour cycloheximide
pre-treatment. (The total cell extracts in high-salt
concentrations were produced from cell culture aliquots
and identical protein amounts were subjected to SDS-
PAGE. The IkB-a concentration was determined by Western
blotting and densitometric quantification of the 38 K
band in the fluorogram.) The left panel shows the
2159822
- 31 -
proven IkB-a (in o) at the start of the cycloheximide
treatment, the right at the start of the PMA treatment.
Example 4
Effect of protease inhibitors on the activation of NF-KB
and the stability of IkB-a in 70Z/3 cells
Cell cultures, extract production, SDS-PAGE and Western
blotting were carried out according to example 1. The
cells were dissolved in 25 m TPCK or TLCK (both from
Sigma), and pre-treated in dimethylsulfoxide (DMSO).
The control cultures received an equivalent amount of
DMSO.
The results of the experiments are shown in Fig. 4: A:
Effect of TPCK on the induction by PMA. The cells were
treated without (left panel) or in presence of TPCK with
PMA for the stipulated periods. The cell extracts were
tested for the DNA binding activity of NF-KB using EMSA
(top panel) and for the IkB-a protein quantity using
Western blotting (bottom panel). The position of the
NF-KB-DNA complex in a section of a fluorogram of a
native gel is shown by a filled arrowhead. An arrow
shows the position of the 38 K IkB-a band in the Western
blot. The weaker lower band is non-specific; its
quantity was highly reduced after an affinity-
purification of the antibody. B: Effect of TPCK on the
activation of NF-KB by IL-1R (I), LPS (L) and PMA (P) in
70Z/3 cells; in control cells (Co). The cells were
treated with TPCK for 10 minutes before stimulation
(lanes 6-8) or for an additional 10 minutes after
stimulation (lanes 10-12; 30 minutes with IL-1(3 and PMA
or 60 minutes with LPS). The total cell extracts of
control cells (lanes 1-4) and cells treated with TPCK
(lanes 5-12) were tested for the DNA binding activity of
NF-KB using EMSA; a section of a fluorogram of a native
2159822
- 32 -
gel is shown. The filled arrowhead shows the position
of the NF-KB-DNA complex. C: Effect of TLCK on the
activation of NF-KB by PMA. 70Z/3 cells were left
untreated (lane 1) or were treated for 10 minutes with
either 25 m TPCK (lane 2) or 25 m TLCK (lane 3),
followed by the addition of PMA. The total cell
extracts were tested for DNA binding activity of NF-KB
using EMSA. The figure shows a section of the
fluorogram; the position of the NF-KB-DNA complex is
again shown by a filled arrowhead.
Example 5
Effect of PDTC on the activation of NF-KB and the
stability of IkB-a in 70Z/3 cells
The cell culture, extract production, EMSA, SDS-PAGE and
Western blotting were carried out according to the
previous examples. The treatment of the cell cultures
with 100 M ammonium salt of PDTC was implemented as
described by Schreck et al. 1992.
The results of the experiments are shown in fig. 5. The
cells were treated during the stipulated times (in
minutes) with PMA without (left top panel) or one-hour
prior to incubation in the presence of PDTC (right
panel). The total cell extracts were tested for DNA
binding activity of NF-KB (top panel) using EMSA and
IkB-a protein content (bottom panel) using Western
blotting, in which, as in the above figures, sections of
fluorograms of native gels are shown and the position of
the NF-KB-DNA complex is shown by filled arrow heads. A
section of a Western blot is shown, in which the
position of the 38 K IkB-a band is indicated by an
arrow.
2159892
- 33 -
Example 6
Effect of the protease inhibitor TPCK on the activation
of a reporter gene by NF-KB in HeLa cells
The cell culture was carried out as in example 1. The
cells were transfected by the calcium phosphate method
(Wigler et al., 1978). 105 cells per tray were either
transfected with 1.5 g control plasmid, 1.5 .g pUC-TK-
Luc or 1.5 g pUC-KB-TK-Luc. The plasmid pUC19 (ATCC
no. 37254, New England Biolabs) was used as a control
plasmid. The plasmid pUC TK-Luc contains the sequence
coded for luciferase (De Wet et al., 1987) under control
of the -105 -+52 region of the Herpes simplex thymidine
kinase promoter. The plasmid pUC-KB-TK-Luc contains
three HIV-LTR fragments of 28 base pairs each, before
the TK promoter region, with two KB binding positions
each (Nabel et al., 1988). The cells were treated with
TPCK (Sigma) 20 minutes before stimulation. The lysis
of the cells and the luciferase assay were carried out
according to Brasier et al., 1989. The results are
shown in relative light units (RLU).
The results of the experiments carried out are shown in
figure 6. Formulations 1, 2 and 3 show the basic
expression of luciferase in non-stimulated cells.
Formulations 4, 5 and 6 show the expression of
luciferase after stimulation of the cells with 200 U/ml
of human TNF-a for 3 hours. Formulations 7, 8 and 9
show the effect of 25 M TPCK (in DMSO) on the
luciferase expression after TBF-a stimulation.
Formulation 10 shows the in vitro effect of TPCK on the
luciferase activity. For this purpose 250 M TPCK
(compare formulations 9 and 10) was added to the cell
extract.
2159822
- 34 -
Example 7
Effect of the protease inhibitors Z-Ile-Glu(Ot-Bu)-Ala-
leucinal (NBIG) and Z-Leu-leucinal on the activation of
NF-KB and the stability of IkB-a in HeLa cells
The test arrangement in this example corresponded to
that of example 4; HeLa cells were cultivated as stated
in example 2 and TNF-a was used as inductor for the NF-
KB activation. As protease inhibitors Z-Ile-Glu(Ot-Bu)-
Ala-leucinal (NBIG) and the chemically related substance
Z-Leu-leucinal (Cbz-L-L) were used (Z standing for
benzyloxycarbonyl).
a) The electrophoretic mobility shift assays were
carried out by using a 32P labelled oligonucleotide with
a NF-KB binding site on whole cell extracts from HeLa
cells. Fig. 7a shows the prevention of the activation
of NF-KB after TNF-a stimulation: the comparison after 5
and 10 minutes clearly show that after stimulation with
TNF-a in the presence of 60 gM NBIG, considerably less
NF-KB-DNA complex is formed than in the control cells
(compare lanes 3 and 4 with lanes 8 and 9; "n.s." in Fig.
7 and 8 stands for "non-specific"). With Cbz-L-L this
effect did not occur (lanes 11 to 15) . The ID50 value for
NBIG was approx. 50 M (Fig. 7B).
b) The Western blot analysis showed that TNF-a induces
the proteolytic degradation of IkB-a (Fig. 8, lanes 3 to
5). In the presence of 60 M NBIG, this degradation is
greatly reduced. (Fig. 8, compare lanes 4 and 9, 10
minute value). The inhibition of the NF-KB activation
after stimulation with TNF-a consequently also creates
an inhibition of the degradation of the IkB-a (Fig. 8A).
In contrast to the protease inhibitor TPCK (see example
4) the NBIG at a concentration of 60 m had the special
feature of causing an accumulation of a modified form of
2159822
- 35 -
IkB-a running slower in the SDS gel (Fig. 8a, compare
lanes 4 and 9). The inhibitor Cbz-L-L (50 m) does not
show these effects (Fig. 8B). In the figure the slower
migrating form of IkB-a is identified with an asterisk.
The incubation of cell fractions with acidic phosphatase
from potatoes results in the slower migrating form of
IkB-a being transformed back into the faster migrating
form. This finding shows that the slower migrating form
of IkB-a, which is accumulated after stimulation by
TNF-a and under the effect of NBIG, is a phosphorylated
form in IkB-a.
In a further experiment NBIG was used together with the
phosphate inhibitor okadaic acid (100 nM). The presence
of okadaic acid caused the quicker migrating non-
phosphorylated form of IkB-a to disappear completely.
The phosphorylation of IkB-a is not induced by treatment
with NBIG (Fig. 9, lanes 8 to 14), not even at a
concentration of 125 M (lane 14). The phosphorylated
form of IkB-a is thus only created when the cells are
induced with TNF-a (Fig. 9, lanes 2 to 7), the quantity
of modified IkB-a rising with the concentration of NBIG.
The observation that the occurrence and the
stabilisation of the phosphorylated form in IkB-a is not
simultaneous with the activation of NF-KB (compare Fig.
8A, lanes 6 to 9 with Fig. 7A, lanes 6 to 9), reinforces
the assumption, that phosphorylation of IkB-a in intact
cells is insufficient for an activation of NF-KB.
Example 8
Inhibition of the expression of NF-KB regulated cytokine
IL-6 and IL-8 in HeLa cells by the protease inhibitor
NBIG
2159822
- 36 -
The principle of the test arrangement corresponded to
that of the previous examples.
The cells were stimulated - with or without 30 minutes
of pre-treatment with 100 mol NBIG - with TNF (200
E/ml) or with PMA (50 ng/ml) and the concentration of
IL-6 or IL-8 was measured in the cell supernatant at
various intervals using ELISA (British Biotechnology
Products Ltd.). The expression values are listed in the
table. The values show that after 120 minutes of
stimulation in the presence of the protease inhibitors,
the production of IL-6 is inhibited by 790 or 730 (TNF
stimulation) and the production of IL-8 is inhibited by
900 or 980 (PMA stimulation).
Table
IL-6 (pg/ml) IL-8 (pg/ml)
0' 15' 120' 0' 15' 120'
TNF - 875 6750 - - 52
TNF+NBIG - 875 1438 - - 5
pMA - 750 12813 - - 410
PMA+NBIG - 1125 3438 - - 10
2159822
- 37 -
References
Baeuerle, P.A. and Baltimore, D., 1988a, Cell 53,
211-217.
Baeuerle, P.A. and Baltimore, D., 1988b, Science 242,
540-546.
Baeuerle, P.A., 1991, Biochim. Biophys. Acta 1072,
63-80.
Baeuerle, P.A. and Baltimore, D., 1991, In: Molecular
Aspects of Cellular Regulation Vol. 6, Hormonal
Control Regulation of Gene Transcription, Cohen, P.
and Foulkes, J.G. (eds.), Elsevier/North Holland
Biomedical Press, Amsterdam, pp. 409-432.
Blank, V., Kourilsky, P. and Israel, A., 1992, Trends
Biochem. Sci. 17, 135-140.
Bomsztyk, K. et al., 1991, Cell. Regul. 2, 329-337.
Brasier, A.R., Tate, J.E. and Habener, J.F., 1989,
BioTechniques 7, 1116-1122.
De Wet, J.R., Wood, K., DeLuca. M. Helsinki, 13. and
Subramani, S., 1987, Mol. Cell. Biol. 7, 725-737.
Gosh, S. and Baltimore, D,, 1990, Nature 344, 678-682.
Grimm, S. and Baeuerle, P.A., 1993, Biochem. J. 290,
297-308.
Gritz, L. and Davies, J., 1983, Gene 25, 179-188.
Hartmann, A., 1991, BioTec 5, 40-45.
Haskill et al., 1991, Cell 65, 1281-1289.
Henkel et al., 1992, Cell 68, 1121-1133.
Hochstrasser, M. and Varshavsky, A., 1990, Cell 61,
697-708.
Kerr et al., 1991, Genes Dev. 5, 1464-1476.
Kricka, L.Ji, 1988, Analyt. Biochem. 175, 14-21.
Laemmli, U.K., 1970, Nature 227, 680-685.
Meichle, A., Schiztze, S., Hensel, G., Brunsing, D. and
Kronke, M., 1990, J. Biol. Chem. 265,8339-8347.
Mulligan, R. and Berg, P., 1981, Proc.Natl.Acad.Sci.
2159822
- 38 -
USA 78, 2072-2076.
Nabel et al., 1988, Science 239, 1299-1301.
Nolan, G.P. and Baltimore, D., 1992, Curr. Opin. Genet.
Dev. 2, 211-220.
Orlowski, M., 1990, Biochemistry 29, 10289-10297.
Remington's Pharmaceutical Sciences, 1980, Mack Publ.
Co., Easton, PA, Osol (ed.).
Rivett, A.J., 1989, Arch. Biochem. Biophys. 268, 1-8.
Schreck, R., Rieber, P. and Baeuerle, P.A., 1991,
EMBO J. 10, 2247-2258.
Schreck, R., Albermann, K. and Baeuerle, P.A., 1992a,
Free Rad. Res. Comms. 17, 221-237.
Schreck, R., Meier, B., Mannel, D., Dr6ge, W. and
Baeuerle, P.A., 1992b, J. Exp. Med. 175, 1181-1194.
Shirakawa, F. and Mizel, S., 1989, Mol. Cell. Biol. 9,
2424-2430.
Sen, R. and Baltimore, D., 1986, Cell 47, 921-928
Southern, P. and Berg, P., 1982, J. Mol. Appl. Gen, 1,
327.
Subramani, S. and DeLuca, M., 1987, Genetic
Engineering, Principles and Methods, J.K. Sedlow
ed., Plenum Press, New York, Band 10, 75-89.
Tanahashi, H., Ito, T., Inouye, S., Tsuji, F.I. and
Sakaki, Y., 1990, Gene 96, 249-255.
Wall, R. et al., 1986, Proc.Natl.Acad.Sci. USA 83,
295-298.
Wieland, E. et al., 1985, Arztl. Lab. 31, 203-214.
Wigler et al., 1978, Cell 14, 725-731.
Zabel, U., Henkel, T., dos Santos Silva, M. and
Baeuerle, P.A., 1993, EMBO J. 12, 201-211.
