Transcription Factors as Potential Targets for Therapeutic Drugs
David S. Latchman*
Institute of Child Health, University College London, 30 Guilford
Street, London WC1N 1EH, U.K.
*Address correspondence to this author at the Institute of Child
Health, University College London, 30 Guilford Street, London WC1N 1EH, U.K.;
Tel: 020-7905-2189; Fax: 020-7242-8437; Email: d.latchman@ich.ucl.ac.uk
Abstract: Although drugs which
target transcription are in wide therapeutic use, they were all identified on
the basis of their effect on a specific biological process such as inflammation
or hormone responses and were only subsequently shown to target transcription.
Our recent progress in understanding the mechanism of action of these drugs and
the mechanisms of transcriptional regulation in general offers hope for a new
generation of drugs isolated on the basis of their ability to modulate either
the synthesis of transcription factors, the regulation of their activity by
ligands or phosphorylation events, their protein-protein interactions or their
binding to DNA.
INTRODUCTION
Regulation of gene expression is central to
the normal development and proper functioning of all organisms since it
results, for example, in different proteins being made by different cell types
allowing these cells to perform different functions (for reviews see 1,2). Such
gene regulation is primarily achieved at the level of gene transcription
whereby the DNA is copied into an RNA transcript. Although some regulation
after transcription does occur (for review see 1), in general once
transcription occurs the other stages of gene expression such as RNA splicing
or translation into protein follow more or less automatically. Thus,
transcription of different genes in different cell types leads to the
production of their corresponding proteins whilst a particular stimulus such as
cyclic AMP or steroid hormones will produce new protein synthesis by activating
the transcription of genes which were not previously transcribed.
In view of this central role for
transcription in biological processes, it represents an obvious target for
therapeutic drugs which could act either by stimulating the transcription of
specific genes required for a desired beneficial effect or by inhibiting the
transcription of genes involved in an undesirable event. Indeed of the 50
FDA-approved best selling drugs, more than 10% target transcription and these
include such well known drugs as salicylate and tamoxifen [3]. Interestingly
however, these drugs were isolated in screens designed to produce specific
biological effects such as immunosuppression or inhibition of hormone action
rather than by screens for drugs which directly target transcription.
Subsequently, when the mechanism of action of these drugs was investigated,
they were found to affect transcription.
The existence of such drugs indicates that
transcription does represent a suitable target for therapeutic drugs. Moreover,
our increased understanding of the mechanisms of action of these drugs can be
used to offer insights into the types of transcriptional regulatory processes
which might be targeted by new drugs. Similarly, our increased understanding of
gene transcription in general provides indications of novel aspects of
transcription which might also be targeted.
It is the purpose of this review to
consider the manner in which such increased understanding of transcription in
general and of the actions of transcription modulating drugs in particular
could be used directly to identify new potential therapeutic agents modulating
transcription on the basis of their ability to do this. This will provide a new
level of cellular control processes which can be directly targeted for
therapeutic benefit.
Regulation of Promoter Activity
Transcription of a specific gene is dependent
upon an array of regulatory sequences known as the gene promoter which
determines both the basal transcription of the gene and its response to
specific stimuli. It is relatively simple to link this promoter to a reporter
gene encoding a protein whose expression can be measured. The construct can
then be introduced into cells and the effect of various agents on the activity
of the promoter measured simply by measuring the expression of the reporter
gene (by assaying the protein it encodes) in untreated cells and in cells
treated with the different agents. This can be done in mammalian cells using,
for example, the gene encoding the luciferase protein as a reporter, allowing a
high throughput automated screen with luciferase activity being read in a luminometer
[3]. This allows a wide range of compounds to be screened for their ability to
stimulate or inhibit the activity of a particular promoter. Evidently,
appropriate controls with a distinct promoter driving luciferase can be
included to confirm that any effects observed are not due to a particular
compound modulating luciferase mRNA stability, translation, etc. Such screening
systems can evidently be used for identifying compounds regulating viral as
well as cellular promoters allowing screening for anti-viral compounds [4].
Similar screens can also be set up in yeast using either a selectable marker or
b-galactosidase
to produce blue colonies when the promoter is active [5]. Obviously however,
any drug identified in this way will need to also be shown to produce a similar
effect in mammalian cells.
Although this approach has the advantage of
high throughput, it suffers from the disadvantage that it will not indicate in
any way the manner in which the particular compound which is identified
modulates promoter activity. Hence, further studies will be required to
determine the mechanism of action of the compound, both to indicate its
potential side effects and to allow rational design of more effective
derivatives. Moreover, this approach does not take advantage of any knowledge
which is available as to the processes which regulate the gene of interest.
Transcription Factors: Synthesis and Activation
Transcription is controlled by regulatory
proteins known as transcription factors which bind to specific DNA sequences in
the gene promoter and activate or inhibit transcription (for review see 2). The
critical importance of such factors is indicated by the fact that mutations in
the genes encoding them result in a wide range of human diseases ranging from
developmental disorders such as aniridia or Rubinstein-Taybi syndrome to
cancers such as leukaemia or retinoblastoma (for review see [6]. Most
importantly, transcription factors play a key role in regulating the expression
of specific genes in specific cell types or in response to specific stimuli.
This is achieved by regulating the transcription factor itself so that it is
either synthesized only in one particular situation or alternatively is
activated from a pre-existing inactive form by some post-translational
modification. Hence, the factor can only regulate the genes when it is present
in a particular cell and is in an active form [2].
Clearly, either the synthesis of a
transcription factor or its activity could be targeted for therapeutic benefit.
In terms of synthesis, the same techniques could be used as for any other
protein whose synthesis needs to be modulated. Thus, for example, considerable
progress has been made in the development of modified anti-sense
oligonucletides, which are complimentary to the mRNA encoding a specific
protein, and which can be used therapeutically to inhibit its synthesis [7].
Although most such anti-sense studies have been conducted in cultured cells,
this approach has been applied to target the synthesis of the transcription
factor NFkB in an in vivo
animal model of experimental colitis and was shown to be a more effective means
of treatment than treatment with glucocorticoid hormone which is normally used
to interfere with NFkB activity. [8]. Similarly, gene delivery
can be used to deliver the gene encoding a transcription factor in the same
manner as with any other gene in a gene therapy procedure and this method has
been used, for example, to deliver the gene encoding the anti-oncogenic
transcription factor p53 in a successful attempt to inhibit tumour growth [9].
Obviously, as in all such gene therapy approaches, the effectiveness of this
approach will depend on the development of safe and efficient gene delivery
systems for human use. Hence, transcription factor expression can be
manipulated pharmacologically or by gene therapy in exactly the same manner as
any other protein.
Evidently, however, such methods simply
mimic the methods used to control the synthesis of any protein rather than
taking advantage of the unique properties of transcription factors. This
approach is likely to be of use therefore only in a situation where it is
desirable to switch on or off an entire bank of genes involved for example, in
inflammation, which are modulated by a particular transcription factor. In such
cases, it would evidently be more effective to target the synthesis of the
factor itself rather than each of the proteins encoded by its target genes. In
many situations however, it will be preferable to develop ways of modulating
the activity of a transcription factor. Indeed, all the currently used
therapeutic drugs target this aspect and the processes which they target or
which could be targeted will now be discussed.
Ligand Binding
A number of transcription factors involved
in activating or repressing genes in response to a specific signalling molecule
are themselves activated from a pre-existing inactive form by direct binding of
the ligand. A classical example are members of the nuclear receptor family.
Individual members of this family are activated by binding of, for example,
glucocorticoid, oestrogen or thyroid hormone and they undergo a conformational
change which allows them to activate their target genes [10].
The important role of one of these
hormones, oestrogen, in the growth of breast cancer cells has led to particular
attention being given to the development of anti-oestrogens which could inhibit
its activity by, for example, binding to the receptor, thereby preventing the
binding of oestrogen, but not activating the receptor. One of these inhibitors,
tamoxifen, which is in clinical use in breast cancer as an anti-oestrogen can
also however, have an undesirable oestrogen-like activity in some situations
[11,12].
Clearly, the improved understanding of the
structure of the ligand binding domains of the nuclear receptors [13], both
prior to and after binding of hormone and the manner in which this affects
their interactions with co-activators or co-repressors, which is currently
being obtained will facilitate the design of drugs which bind to the receptor
but do not mimic any of the effects of oestrogen so having a pure
anti-oestrogenic effect. Similarly, the recognition that one consequence of the
conformational changes which occurs upon oestrogen binding to its receptor, is
the binding to the receptor of co-activator proteins, which are necessary for
transcriptional activation [14,15], indicates that another approach would be to
develop drugs which block the protein-protein interaction between the receptor
and its co-activator(s) (see below).
Phosphorylation
Although transcription factors can be
regulated directly by ligands such as oestrogen which can enter the cell, many
other signalling molecules which cannot do so, set off cascades of kinase
and/or phosphatase enzymes which ultimately results in the phosphorylation or
dephos-phorylation of one or more transcription factors resulting in their
activation [16]. Several clinically-used drugs target this aspect by modulating
the phosphorylation state of one or more transcription factors. Thus,
cyclosporin and FK506 (tarcolimus) have an anti-inflammatory effect because
they prevent the enzyme calcineurin from dephosphorylating the NF-AT
transcription factor, such dephosphorylation being required for it to stimulate
the expression of several genes involved in the immune response [17,18,19].
Similarly, salicylate inhibits the activation of the NFkB transcription factor by
preventing the phosphorylation of the IkB transcription factor 20] which is
associated with the NFkB factor and inhibits its activity. Such
phosphorylation is required for the release of IkB from NFkB
and prevents it activating its target genes which are involved in immune and
inflammatory events [21].
Obviously, these drugs were identified on
the basis of their ability to modulate biological processes such as the immune
response rather than their effect on transcription factor phosp-horylation.
Therefore, our improved understanding of the effects of transcription factor
phosphorylation and the enzymes which regulate it should allow the setting up
of high throughput in vitro screens
aimed at identifying compounds which modulate the
phosphorylation/-dephosphorylation of a particular factor by a particular
enzyme. Such screens will be particularly valuable if they can identify
compounds which can specifically modulate the phosphorylation of a target
transcription factor by a specific enzyme without affecting the ability of the
enzyme to phosphorylate other transcription factors, which may produce
undesirable side effects. Thus, although NF-AT is expressed only in T cells, FK
506 and cyclosporin have toxic side effects due to their effects on other
tissues which are presumably due to their effects on other transcription
factors [22-24].
Protein-Protein Interactions
Interactions of transcription factors with
other proteins are central to the regulation of their activity and the
mechanism of their action. Thus, the NFkB factor discussed above, is an example of a
factor whose activity is regulated via interaction with the inhibiting IkB
factor [21] whilst the nuclear receptors need to interact with co-activator
proteins to activate transcription [14]. Similarly, in order to ultimately
stimulate transcription, activating factors or their co-activators need to
interact with the proteins of the basal transcriptional complex to stimulate
its activity [25].
Such protein-protein interactions represent
an obvious target for disruption for therapeutic purposes. Thus, once the site
of interaction on one or other of the interacting proteins has been mapped, a
short peptide prepared from this region can be prepared and used in attempts to
disrupt the interaction in vitro and
then in intact cells. For example, by using a peptide which disrupted the
interaction between the cellular transcription factor Oct-1 and the herpes
simplex virus (HSV) transactivator protein VP16, it was possible to inhibit the
HSV lytic cycle in intact cells [26]. Similarly, heterodimerization between the
E2F and DP-1 transcription factors could be inhibited using a peptide prepared
from the interacting region resulting in a failure of DNA binding which is
dependent upon heterodimerization, leading to apoptosis of tumour cells treated
in this manner [27]. Hence, peptides can be used to target protein-protein
interactions between transcription factors.
Although such peptides are themselves
unlikely to be of use as therapeutic agents in
vivo, their structure can serve as a basis to design therapeutically useful
peptide mimetics which could be delivered in
vivo (for review see 28). Thus, this approach has been used, for example,
in the design of non-peptide inhibitors of the human immunodeficiency protease
based on structural analysis of peptide inhibitors which bind to its
substrate-binding site. Hence, this area holds considerable promise for the
future.
DNA Binding
Obviously, binding to a specific binding
site in DNA is necessary for the action of many activating or inhibitory
transcription factors, and thus represents an obvious target for therapeutic
drugs. Indeed, as noted above, the inhibition of E2F/DP-1 heterodimerization
actually achieved its effect by preventing DNA-binding of the factor which
required heterodimer formation. However, it is also possible to think in terms
of drugs which interact with the binding site of a transcription factor in the
DNA to prevent its binding. However, although a number of DNA-binding drugs
such as distamycin have been developed to inhibit DNA replication as a means of
cancer therapy, in general, these do not have sufficient sequence specificity
to specifically target the DNA binding site of a particular transcription
factor. However, numerous attempts are being made to produce derivatives with
greater specificity [29,30] utilizing improved structural information on the
binding of these drugs to DNA [31,32].
An alternative approach, however, comes
from the observation that specific sequences in double-stranded DNA can be
bound by a single-stranded oligonucleotide to form a triple helix structure
which is not recognized by a protein such as a transcription factor that would
normally bind to that site in double helical DNA. This approach has been used
for example, to inhibit the transcription of the c-fos proto-oncogene [33] or
that of the tumour necrosis factor gene [34] which in the latter case resulted
in inhibition of the growth of TNF-dependent tumour cells [34].
This approach, may thus, offer an effective
means of specifically inhibiting gene expression and may in future be combined
with the use of DNA-binding drugs particularly, since one such drug,
distamycin, has been shown to
differentially affect the stability of double helical versus triple helical DNA
[35].
CONCLUSIONS
Many millions of people daily take
therapeutic drugs such as tamoxifen, salicylate or FK506, which target
transcription. Yet, none of these drugs was identified on the basis of this
ability. Our increasing knowledge of the manner in which these drugs act at the
level of ligand binding or phosphorylation as well as our improved
understanding of protein-protein interactions and DNA binding by transcription
factors, offers hope of a new generation of drugs isolated specifically on the
basis of the ability to modulate transcription. These will arise both from high
throughput screens of compounds for their effect, for example, on promoter
activity or kinase activity and from designer approaches based on a detailed
structural understanding of an individual transcription factor and its
interaction with other proteins and with DNA.
REFERENCES
[1] Latchman
D.S. (1998) Gene Regulation: a
eukaryotic perspective. Third Edition. Stanley Thornes Publishers, Cheltenham.
[2] Latchman
D.S.: (1998) Eukaryotic transcription factors. Third Edition. Academic Press,
London, San Diego.
[3] Cai
W.; Hu L,; Foulkes J.G.; (1996) Current Opinion
in Biotechnology, 7, 608-615.
[4] Delrosaro
M,; Stephas J.C.; Zakel J.; Escobedo J.; Giese K. (1996). Nature Biotechnology, 14, 1592-1596.
[5] Shiau
S.P.; Glasebrook A.; Hardikar S.D.; Yang N.N.; Hershberger C.L. (1996) Gene, 179, 205-210.
[6] Latchman
D.S.: (1996). New England Journal of
Medicine, 334, 28-33.
[7] Agrawal
S. (1996). Trends in Biotechnology,
14, 376-387.
[8] Neurath
M.F.; Pettersson S.; Zumbuschenfelde K.H.M.; Strober W. (1996). Nature Medicine, 9, 998-1004.
[9] Roth
J.A.; Nguyen D.; Lawrence D.D.; Kemp B.L.; Carrasco C.H.; Ferson D.Z.; Hong
W.K.; Komaki R.; Lee J.J.; Nesbitt J.C.; Pisters K.M.W.; Putnam J.B.; Schea R.;
Shin D.M.; Walsh G.L.; Dolormente M.M.; Han C.I.; Martin F.D.; Yen N.; Xu K.;
Stepehsn L.C.; Mcdonnel T.J,; Jukhopadhyay T.; Cai D.; (1996) Nature Medicine, 2, 985-991.
[10] Mangelsdorf
D.J.; Thummel C.; Beato M.; Herrlich P.; Schütz G.; Umesono K.; Blumberg B.;
Kastner P.; Mark M.; Chambon, P, Evans R.M. (1995). Cell, 83, 835-839.
[11] Fan
J.D.; Wagner B.L.; McDonnell D.P. (1996). Molecular
Endocrinology, 10, 1605-16.
[12] Gallo
M.A.; Kaufman D. (1997). Seminars in
Oncology, 24, S1-71-S1-80.
[13] Wurtz
J-M.; Bourguet W.; Renaud J-P.; Vivat V.; Chambon P.; Moras D.; Gronemeyer H.
(1996) Nature Structural Biology, 3,
87-94.
[14] Glass
C.K.; Rose D.W.; Rosenfeld M.G.; (1997). Current
Opinion in Cell Biology, 9,
222-232.
[15] L’Horset
F.; Dauvois S.; Heery D.M.; Cavaillès V.; Parker M.G.; (1996) Molecular and Cellular Biology, 16,
6029-6036.
[16] Treisman
R. (1996). Current Opinion in Cell
Biology, 8, 205-215.
[17] Shibasaki
F. Price E.R.; Milan D.; McKeon F. (1996). Nature,
382, 370-373.
[18] Beals
C.R.; Clipstone N.A.; Ho S.N.; Crabtree G.R.; (1997). Genes and Development, 11, 824-834.
[19] James
D.G. (1996) Postgraduate Medical Journal,
72, 852-856.
[20] Pierce J.W.; Read
Ma.; Ding H.; Luscinskas F.W.; Collins T. (1996) Journal of Immunology, 156, 3961-3969.
[21] Baeurale
P.A.; Baltimore D. (1996) Cell 1996,
87, 13-20.
[22] Mosieniak
G.; Figiel I.; Kaminska B. (1997) Journal
of Neurochemistry, 68, 1142-1149.
[23] Su
Q.; Eugster H.P.; Ryffel B.; Dumont F.J.; (1996) Biochemical Biophysical Research Communications, 229, 249-256.
[24] Muraoka
K.; Fujimoto K.; Sun X.; Yoshioka K.; Shimizu K.; Yagi M.; Bose H Jr.; Miyazaki
I.; Yamamoto K.; (1996) Journal of
Clinical Investigation, 97, 2433-2439.
[25] Greenblatt
J. (1997) Current Opinion in Cell Biology,
9, 310-319.
[26] Haigh
A.; Greaves R.; O’Hare P. (1990) Nature, 344, 257-259.
[27] Bandara
L.R.; Girling R.; La Thangue N.G.; (1997) Nature
Biotechnology, 15, 896-901.
[28] Kauvar
L.M. (1996) Nature Biotechnology, 14,
709.
[29] Bailly
C.; Brana M.; Waring M.J. (1996) European
Journal of Biochemistry, 240, 195-208.
[30] Gao
Y.G.; Priebe W.; Wang A.H.J. (1996) European
Journal of Biochemisry, 240,
331-335.
[31] Kumar
R.A.; Ikemoto N.; Patel D.J. (1997) Journal
of Molecular Biology, 265, 173-186.
[32] Kumar
R.A.; Ikemoto N.; Ptel D.J. (1997) Journal
of Molecular Biology, 265, 187-201.
[33] Lavrosky
Y.; Stoltz R.A.; Vlassov V.V.; Abraham, N.G. (1996) European Journal of Biochemistry 1996, 238, 582-590.
[34] Aggarwal
B.B.; Schwarz L.; Hogan M.E.; Rando R.F. (1996) Cancer Research, 56,
5156-5164.
[35] Durang
M.; Maurizot J.C. (1996) Biochemistry,
35, 9133-9139.