U.S. patent application number 12/124943 was filed with the patent office on 2010-01-14 for methods and compositions for cellular reprogramming.
Invention is credited to Larry J. Smith.
Application Number | 20100010065 12/124943 |
Document ID | / |
Family ID | 40525080 |
Filed Date | 2010-01-14 |
United States Patent
Application |
20100010065 |
Kind Code |
A1 |
Smith; Larry J. |
January 14, 2010 |
Methods and Compositions for Cellular Reprogramming
Abstract
Compositions and methods useful for the treatment of aberrant
programming diseases, particularly those associated with aberrant
apoptosis are disclosed
Inventors: |
Smith; Larry J.; (Omaha,
NE) |
Correspondence
Address: |
DANN, DORFMAN, HERRELL & SKILLMAN
1601 MARKET STREET, SUITE 2400
PHILADELPHIA
PA
19103-2307
US
|
Family ID: |
40525080 |
Appl. No.: |
12/124943 |
Filed: |
May 21, 2008 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
08472801 |
Jun 7, 1995 |
7517644 |
|
|
12124943 |
|
|
|
|
08426781 |
Apr 22, 1995 |
|
|
|
08472801 |
|
|
|
|
Current U.S.
Class: |
514/44A ;
435/375 |
Current CPC
Class: |
A61P 19/02 20180101;
C12N 2310/11 20130101; A61P 31/22 20180101; C12N 15/113 20130101;
A61P 31/16 20180101; C12Q 1/6886 20130101; A61P 37/00 20180101;
A61P 31/18 20180101; A61P 25/28 20180101; A61P 9/10 20180101; C12N
15/1135 20130101; C12Q 2600/106 20130101 |
Class at
Publication: |
514/44.A ;
435/375 |
International
Class: |
A61K 31/7088 20060101
A61K031/7088; C12N 5/06 20060101 C12N005/06 |
Claims
1. A method for inhibiting apoptosis in a cell or tissue
comprising: a) contacting said cells or tissue with an effective
amount of at least one agent which inhibits p53 dependent cell
death under conditions whereby said agent enters said cells and
reduces apoptosis relative to untreated cells.
2. The method of claim 1, wherein at least one said agent is a
nucleic acid molecule, said nucleic acid molecule optionally
comprising at least one modification which improves pharmacologic
function.
3. The method of claim 1 or 2, comprising co-administration of at
least one augmentation agent selected from the group consisting of
a free radical generator and a redox modifier.
4. The method of claim 1 or 2, comprising co-administration of at
least one augmentation agent selected from the group consisting of
an antioxidant, anti-inflammatory agent and a cytokine.
5. The method of claim 1 or 2, wherein said cells are normal cells
adversely affected by anti-cancer treatment for an existing
malignancy which treatment activates p53 dependent cell death.
6. The method of claim 5, wherein said treatment is radiation or
chemotherapy.
7. The method of claim 1 or 2 wherein said cells are damaged by a
process selected from the group consisting of Alzheimer's disease,
autoimmune disease, artherosclerosis, restenosis,
ischemia/reperfusion injury, rheumatoid arthritis, arthritis,
multiple sclerosis, lupus erythematosus, multiple organ dysfunction
syndrome, myocardial infarction, stroke, ionizing radiation, and
viral disease.
8. The method of claim 1 or 2, wherein said cell or tissue is
heart, skin, brain or a hematopoietic progenitor cell.
9. The method of claim 1 or 2, wherein said at least one agent is a
nucleic acid which is complementary to a portion of a p53 gene and
is selected from the group consisting of an
oligodeoxy-ribonucleotide, and an oligoribonucleotide.
10. The method of claim 9, wherein said nucleic acid is a double
stranded.
11. The method of claim 9, wherein said nucleic acid is present in
an expression vector.
12. The method of claim 9, wherein uptake of said nucleic acid into
said cell is facilitated by a carrier.
13. The method of claim 2, wherein said nucleic acid binds to a
transcriptional regulator binding site.
14. The method of claim 9, wherein said nucleic acid is single
stranded.
15. The method of claim 1 or 2, wherein said method further
comprises subjecting cells treated in a) to a Reprogramming
test.
16. The method of claim 1 or 2, wherein said nucleic acid is a
double stranded oligoribonucleotide comprising a sequence
complementary to a sequence encoding a transcriptional
regulator.
17. The method of claim 9, wherein said nucleic acid analog
comprises a sequence selected from the group consisting of SEQ ID
NOS: 1-4 and 863-872.
18. The method of claims 2 or 10, wherein said modification is
selected from the group consisting of a ethyl- or methylphosphonate
modified oligodeoxynucleotides, phosphorothioate modified
oligonucleotides, dithioates, oligonucleotide analogs,
oligonucleotides comprising ribozymes, oligoribonucleotides,
chimeric oligonucleotides that are composite RNA, DNA analogues,
oligonucleotides having a lipophilic backbone, methylphosphonate
analogs with ribozyme structures, and oligonucleotides with
2'-O-methyl modified ribose sugars.
19. The method of claims 1, 2 or 9, said method further comprising
administration of at least one agent or nucleic acid which inhibits
expression of a target selected from the group consisting of apo-1
(SEQ ID NOS: 3374-3443), bax alpha (SEQ ID NO: 1027-1042), bcl-beta
(SEQ ID NOS: 3610-3614), bcl-xl (SEQ ID NOS: 1000-1014), bcl-x (SEQ
ID NOS: 1015-1026), cyclin A (SEQ ID NOS: 2076-2099), cyclin B (SEQ
ID NO: 2066-2075), cyclin D exon 1 (SEQ ID NOS: 3589-3597), cyclin
D exon 2 (SEQ ID NOS: 3598-3603), cyclin D exon 3 (SEQ ID NO:
3604-3605, cyclin D exon 4 (SEQ ID NOS: 3606-3609), cyclin D3 exon
1 (SEQ ID NO: 3587-3588), cyclin D3 exon 3 (SEQ ID NOS: 3580-3586),
cyclin D3 exon 4 (SEQ ID NOS: 3575-3579) cyclin D1 (SEQ ID NOS:
727-738), CREBP-1 (SEQ ID NOS: 326-353), ICE (SEQ ID NOS:
2788-2802), ICH-1L (SEQ ID NOS: 1326-1343), TGF-beta (SEQ ID NOS:
3717-3741, TNF-alpha (SEQ ID NOS: 3145-3176), TNF-beta (SEQ ID NOS:
3112-3144, TR3 (SEQ ID NOS: 3216-3241), E2F-1 (SEQ ID NOS:
221-234), GADD153 (SEQ ID NOS: 1344-1353), IRF-1 (SEQ ID NOS:
271-289), ISGF-3 (SEQ ID NOS: 1051-1068), MSX2 (SEQ ID NO:
1481-1491), NF-IL-6 (SEQ ID NOS: 2803-2852), GADD45 (SEQ ID NOS:
1354-1404), c-jun (SEQ ID NOS:891-908), junB (SEQ ID NOS: 455-464),
junD (SEQ ID NOS: 5-17, 396, 399, 401, 403, 405, 407, 409, 412,
414, 416, and 418), c-fos (SEQ ID NOS: 465-482), Fra-1 (SEQ ID NOS:
449-454), Fra-2 (SEQ ID NOS: 627-638), PKC-alpha (SEQ ID NOS:
1869-1881), PKC-beta (SEQ ID NOS: 1882-1887), PKC-delta (SEQ ID NO:
1898-1917), CREBP-1 (SEQ ID NOS: 326-353), PKC-epsilon (SEQ ID NOS:
1918-1937), PKC-gamma (SEQ ID NOS: 1888-1897) PKC-iota (SEQ ID NOS:
1462-1480), PKC-mu (SEQ ID NOS: 1442-1461), PKC-theta (SEQ ID NOS:
3615-3626), PKC-zeta (SEQ ID NOS: 1417-1441) and c-myc (SEQ ID NOS:
657-676).
20. A pharmaceutical composition comprising the agent of claims 1,
2 or 9 in a pharmaceutically acceptable carrier.
21. The pharmaceutical composition of claim 20, further comprising
an effective amount of at least one of a free radical generator and
a redox modifier.
22. The pharmaceutical composition of claim 20 further comprising
at least one of an antioxidant, an anti-inflammatory and a
cytokine.
23. The pharmaceutical composition of claim 20, wherein said
nucleic acid is selected from the group consisting of SEQ ID NOS:
1-4 and 863-872.
24. The method of claims 1 or 2, wherein said cells are damaged as
a result of blood vessel occlusion followed by reperfusion
injury.
25. The method of claims 1 or 2, or 9 or 19 wherein said method is
for treatment of apoptosis associated with viral infection, wherein
said virus is selected from the group consisting of adenovirus,
cytomegalovirus, Epstein-Barr virus, hepatitis C virus, herpes
virus, hemorrhagic fever viruses, human immunodeficiency virus,
influenza virus, pox virus, vaccinia virus.
26. The method of claim 25, further comprising administration of at
least one augmentation agent selected from the group consisting of
an anti-oxidant, a redox modifier, a cytokine and an
interferon.
27. The method of claim 25, wherein said virus is HIV and said
nucleic acid inhibits virus expression or effects of virus on cells
and is selected from the group consisting of GADD-153 (SEQ ID NOS:
1344-1353), GADD-45 (SEQ ID NO: 1354-1404), MTS 1,2, (SEQ ID NOS:
2454-2472 and 2100-28), TGF beta (SEQ ID NOS: 3717-3741), USF (SEQ
ID NOS:1826-1852), Ap-2 (SEQ ID NOS: 1252-1289), Ap-4 (SEQ ID NOS:
2242-2264), Sp-1 (SEQ ID NOS: 989-999), Sp-3 (SEQ ID NOS: 985-988),
Sp-4 (SEQ ID NOS: 978-984), p53 (SEQ ID NOS:863-872), NF-kappa B
(SEQ ID NOS:1741-1774, 1720-1739, 2166-2205, 2007-2011), rel (SEQ
ID NOS:437-448), GATA-3 (SEQ ID NOS: 2992-3008), and Waf1 (SEQ ID
NOS: 2440-2453).
28. The method of claim 27, wherein said viral infection causes
AIDS.
Description
[0001] This application is a divisional application of U.S. patent
application Ser. No. 08/472,801 filed 7 Jun. 1995, now U.S. Pat.
No. 7,517,644, which is a Continuation-in-Part of co-pending
application Ser. No. 08/426,781, filed 22 Apr. 1995.
BACKGROUND OF THE INVENTION
[0002] 1. Field of the Invention
[0003] The present invention relates to methods and compositions
useful in treating disorders in which the direct cause of the
clinical disorder is the expression in the primary diseased cells
of a differentiation program that does not normally exist. Such
disorders are hereinafter referred to as Aberrant Programming (AP)
Diseases. The invention also relates to method and compositions
useful in therapeutically reprogramming normal cells.
[0004] As will be discussed more fully hereinafter, the AP diseases
of this invention constitute a new disease classification and there
is presented a novel molecular model of pathogenesis for these
diseases. According to the molecular model of this invention, the
basic disease causing entity in the AP diseases is a specific type
of relational alteration among certain cellular components involved
in program control. It is unlike any previously described molecular
pathogenic mechanism. This model defines the nature of the therapy
for these diseases, limits the potential set of therapeutically
useful targets to a relatively small number of genes and leads to
the unobvious conclusion that this includes the manipulation of
certain "normal" genes is an appropriate approach for the treatment
of AP diseases, thus, leading to a unique approach to therapy for
the AP diseases of this invention. This model makes the selection
of targets for proposed therapy straightforward and accessible to
anyone skilled in the art. Also provided herein is a novel approach
to diagnosing and developing prognostic criteria for the Aberrant
Programming Diseases.
[0005] A preferred embodiment relates to the reprogramming of cell
behavior through the manipulation of transcriptional regulators
(TRs). The invention includes systemic treatment and compositions
for such treatment, as well as in vitro manipulation of cells prior
to transplantation of such cells with the host under treatment.
[0006] As will be discussed more fully hereinafter, a method is
also disclosed for selecting target sites in the RNA transcripts of
particular genes that individually have a high likelihood of being
excellent target sites for the binding of antisense
oligonucleotides which are intended to inhibit the expression of
the corresponding gene. In a preferred embodiment, this target site
selection method can be used to select antisense oligonucleotides
for the treatment of Aberrant-Programming Diseases in accordance
with the molecular model of Aberrant Programming Diseases disclosed
herein. Taken together, this molecular model of Aberrant
Programming Diseases (AP Model), which delimits the preferred gene
targets for the development of the new therapies discussed herein
for said diseases, and the method for selecting target sites within
the transcripts of said genes, greatly simplifies the drug
discovery process for the development of new treatment modalities
and greatly increases the likelihood that clinically successful
compounds will, thereby, be generated.
[0007] In yet another embodiment, the novel method disclosed herein
for selecting antisense oligonucleotide target sites (and, thereby,
the sequence of the corresponding antisense oligonucleotides) can
be used with conventional rationale (independent of the AP model)
by one with ordinary skill in the art to select therapeutic
oligonucleotides for the treatment of a variety of diseases and
medical purposes. The conventional rationale is, in essence, a
rationale which links the known function(s) of particular
molecules, in terms of their direct effects on specific cellular
functions, with particular disease processes or therapeutic needs
(as opposed to molecules such as TR that--non-obviously--may act
indirectly as a result of being part of a combinatorial regulation
mechanism).
[0008] 2. Description of the Related Art
[0009] Very recent studies involving the use of antisense
oligonucleotides for treatment of cancer have been reviewed by
Stein and Cohen, Cancer Res. 48:2659 (1988). Several types of
antisense molecules have been screened for their ability to inhibit
the synthesis of particular proteins using both intact cells and in
vitro systems for protein synthesis (See Ld. and Paoletti,
Anti-Cancer Drug Design 2:325, 1988). For example, agents with
specificity for RNA transcribed from the myc gene have been
reported to inhibit the proliferation of the human AML line HL60
(Wickstrom, et al., Proc. Natl. Acad. Sci. USA 85:1028 (1988) and
normal T lymphocytes (Heikkila, et al., Nature 328:445 (1987), and
oligodeoxynucleotides complementary to cyclin mRNA have been
reported to suppress the division of 3T3 cells (Jaskulski, et al.
1988).
[0010] More recently, it has been found that in the treatment of
cancer with ODNs against myb, the proliferation of leukemic cells
was inhibited with an accompanying lower degree of inhibition
against normal cells. (Calabretta et al, PNAS, 88,2351,1991.) Also,
it has been shown that transient inhibition in a leukemia cell line
resulted with an ODN against myc; however, unfortunately, a
comparable inhibition against normal cells occurred (Zon et al
patent). This patent also discloses inhibition of HIV replication
using ODNs targeted to viral genes. Belenska et al (Science, 250,
997, 1990) have proposed the use of double stranded ODNs, binding
to TR ligands as potential therapeutic agents for disease causing
genes. They give blocking of NF-kB binding to HIV enhancer as an
example. The use of retroviral vectors carrying antisense oncogenes
for the treatment of cancer is known.
[0011] The fundamental problem with the foregoing part is that it
is based on the notion that the expression of specific molecular
abnormalities (altered regulation or mutation of endogenous genes
or expression of exogenous genes) in the disease cells of these
patients directly cause the clinical pathological features of the
AP disease. It follows from such thinking that the therapeutic
strategies should be directed to attacking these molecular
abnormalities.
[0012] In the case of cancer, contemplated therapy involving
antisense expression vector ODNs have been directed to oncogenes in
accordance with the oncogene/anti-oncogene cancer model, or to
growth factors expressed by cancer cells in accordance with the
autocrine model. In the case of AIDS therapeutic strategies
involving such agents being developed are directed toward blocking
HIV expression and/or infection. There are no counterpart causal
agents identified to the other AP diseases. Hence the therapeutic
approaches under development are more empirical.
[0013] The present inventor first described in detail, in writing,
in a confidential manuscript prepared for Gerald Zon, Ph.D. of
Applied Biosystems Incorporated (now LYNX Therapeutics, Inc.) and
completed in May of 1990, the concept of Aberrant Programming
disease and the derivative notion of using antisense ODNs to
modulate TR as a means of selectively reprogramming the cellular
programs of the diseases cells/tissue in question. According to the
AP disease model the fundamental pathology causing the clinical
pathological features of these disorders is both relational and
dynamic. In stark contrast to the prior art, the therapy of the
present invention involves manipulation of patterns of TR
expression. The invention provides an entirely new approach to the
treatment of said selected diseases and provides a rational,
empirical basis for the design of novel agents. The therapeutic
reprogramming of normal tissue involving ODNs is unprecedented.
[0014] The Aberrant Programming model indicates that
atherosclerosis is, for example, an Aberrant Programming Disease.
In this case, atherosclerosis is said to result from a change in
the pattern of expression of certain TR in the smooth muscle cells
(SMC) associated with blood vessels; the changed pattern of
expression of TR, then, is responsible for the particular
differentiated state that characterizes atherosclerotic SMC and
which, therefore, produces the disease. The conventional hypothesis
that attempts to provide a molecular explanation for the pathogenic
changes in atherosclerotic smooth muscle cells (SMC) is called the
"monoclonal hypothesis" (Senditt and Senditt, Proc. Natl. Acad.
Sci. USA 70: 1753, 1973). In essence, this hypothesis argues that
atherosclerotic plaques are benign tumors that result from a
mutagenic event in some key regulatory molecule, in a manner
analogous to the conversion of a proto-oncogene to an oncogene in
the case of malignant tumor cell development. In support of this
"monoclonal hypothesis" it has been found that DNA isolated from
cells recovered from atherosclerotic plaques is capable of
transforming normal fibroblasts in a transfection-nude mouse assay,
whereas DNA extracted from normal control endothelial cells does
not induce such transformation. The gene(s) which are responsible
for encoding this transforming capacity have not been identified,
however. Ruled out so far are N-ras, K-ras, Ha-ras, erbA, erbS,
fes, src, mos, Abl, sis, c-fos, c-myb and c-myc which have been
shown to be expressed by SMC (Parkes et al., Am. J. Pathol. 138:
765, 1991). Thus, the "monoclonal hypothesis" requires the mutation
of some key regulatory molecule as the causal factor in the
development of atherosclerosis. It follows from the conventional
rationale that this key regulatory molecule which is altered is the
prime target for therapeutic intervention.
[0015] In contrast, the Aberrant Programming model argues that the
basic molecular pathology in the atherosclerotic SMC is to be found
in the pattern of TR expression, where the relevant TR are those
involved in cellular program control. The Aberrant Programming
model, therefore, identifies TR such as, for example, c-fos, c-myb
and c-myc as being appropriate targets for evaluating potential
therapeutic antisense ODNs for atherosclerosis in the Reprogramming
Test (as defined hereinafter), even though these TR have not been
found to be mutated in atherosclerotic SMC. If the molecular
mutations that have been detected in the transfection-nude mouse
assay contribute to the pathogenesis of atherosclerosis, they would
be considered by the AP model to be "risk factors". Risk factors in
this context are defined as determinants that increase the
probability that the afflicted cells/tissue will express an altered
pattern of TR, thereby facilitating the generation of an Aberrant
differentiation program. The AP model sets the foundation for a
novel therapeutic strategy. The model predicts, for example, that
there are antisense ODNs which, when targeted to certain TR, will
produce a therapeutic reprogramming of atherosclerotic SMC (such
as, for example, reversing the Aberrant cellular differentiation
program to a more normal state, or, inducing apoptosis in the
atherosclerotic SMC); the AP model also predicts that such effects
will not be seen when the antisense ODN is used to treat a wide
variety of other normal and diseased cell types that express
different differentiation programs, even though they express the
same TR target and the expression of said TR may be down regulated
by the antisense ODN treatment. The basis of this logic can best be
understood by making an analogy to "language", as is done in Table
I.
[0016] Rosenberg and his colleagues, using a rat model system,
explored the potential use of c-myb antisense ODNs for the
treatment of restenosis. Restenosis refers to the re-occlusion of
atherosclerotic blood vessels following a medical procedure to
reverse the obstruction to blood flow produced by atherosclerotic
plaques (Simons and Rosenberg, Circulation Res 70: 835, 1992; and,
Simons et al., Nature 359: 67, 1992) (U.S. Pat. No. 723,454, 28
Jun. 1991; U.S. Pat. No. 792,146, 8 Nov. 1991; U.S. Pat. No.
855,416, 18 Mar. 1992). The second of the published studies
demonstrates that the local delivery of phosphorothioate ODNs to
rat carotid arterial SMC in vivo results in a substantial uptake of
the ODNs by the SMC and a prolonged retention of these compounds by
the SMC. These investigators showed that c-myb antisense ODN, but
not the corresponding sense ODN, inhibited the proliferation of SMC
in the aorta of normal animals following regional damage to the
vessel wall resulting from balloon angioplasty. Balloon angioplasty
damages the endothelium underlying the region of treatment and
causes intimal migration and proliferation of the SMC over the
length of the damaged blood vessel. The result of treating the
damaged vessels with the c-myb antisense ODN was a substantial
improvement in the patency of the affected vessel after the induced
trauma, compared to control animals not treated with the c-myb
antisense ODN. The c-myb antisense ODN used to treat the normal
smooth muscle cells either in vitro or in vivo, however, had four
guanine bases in a row which could cause the formation of a
"G-quartet", while the control ODN did not. The suppression of SMC
growth, therefore, may not have been due to an antisense effect on
c-myb, but rather from a non-antisense effect, with the reduction
in c-myb expression in the SMC being a secondary event. This
possibility was not apparently explored by these investigators.
[0017] Three groups have examined the possibility that c-myc
antisense ODNs might be useful for the treatment of restenosis.
Zalewski and his colleagues were the first to carry out these
studies (Shi et al., Circulation 88: 1190, 1992; and Shi et al.,
Circulation 90: 944, 1994) (U.S. Pat. No. 4,799, published 7 Jan.
1993). The first of the Zalewski papers examined the usefulness of
c-myc antisense ODNs for inhibiting the proliferation of normal
human SMC. The c-myc antisense ODN was shown to inhibit the
proliferation of the SMC while the corresponding sense ODN and a
mismatched control ODN did not. The published in vivo work by Shi
et al. (1994) again demonstrated that phosphorothioate ODNs can be
readily delivered to SMC in the coronary vessels of animals where
the ODNs are taken up in sufficient quantities to produce
biological effects. Specifically, coronary blood vessels of pigs
were damaged by balloon angioplasty. The human c-myc antisense ODN
or the corresponding sense control ODN, were then applied to the
damaged vessels. It was not reported, however, whether or not the
human c-myc antisense ODN was sufficiently complementary to the
c-myc transcript of the pig to permit effective binding of the
human ODN sequence to the pig target transcripts. The c-myc, but
not the control ODN, substantially inhibited the proliferation of
SMC, resulting in improved blood flow through the affected vessels,
compared to control animals not treated with the c-myc antisense
ODN. Again, however, the antisense ODN used in the in vivo efficacy
studies had four guanine bases in a row, while the control ODN did
not. This four-guanine sequence could explain the capacity of the
"therapeutic" ODN to inhibit the proliferation of the pig SMC and
to inhibit the c-myc expression, while the "control" ODN did not.
Bennett et al. (J. Clin. Invest. 93: 820, 1994) and Biro et al.
(Proc. Natl. Acad. Sci. USA 90: 654, 1993), using the same c-myc
antisense ODNs and control ODNs, demonstrated an inhibition of rat
SMC proliferation in vitro.
[0018] Hence, the published studies of the use of c-myb or c-myc
antisense ODNs to block experimentally-induced restenosis in animal
models could be interpreted as showing that proliferation of cells
(in this case, normal smooth muscle cells) can be blocked simply by
exposure to compounds which have a non-specific capacity to inhibit
proliferation (such as, for example, by non-specific masking of
cell surface receptors, or by interference in essential metabolic
pathways). Hence, these findings do not constitute "cellular
reprogramming" for the purposes of achieving [; therapeutic effect
as defined herein. A "true reprogramming event" that involves
inhibition of proliferation in accordance with the rationale
provided herein would show a dependence on the differentiation
status of the target cells; i.e., the reprogramming event
(initiated by the antisense ODNs) must only work on cells that
exhibit a particular set of differentiation programs, and not work
on cells which exhibit a different set of differentiation programs.
For example, an antisense ODN capable of blocking SMC proliferation
by a reprogramming effect would not be able to block the
proliferation of human cells in general. The possibility remains,
however, that antisense ODNs directed to c-myc or c-myb could cause
a therapeutic reprogramming of atherosclerotic SMC, in accordance
with the present invention. This possibility remains because the
appropriate experiments have not yet been done.
[0019] The design of antisense oligonucleotides for the inhibition
of gene expression has been based primarily on one or the other (or
both) of two considerations. First, investigators have targeted
antisense oligonucleotides to regions of RNA transcripts known to
be involved in the control of pre-mRNA processing or mRNA
translation, such as splice sites or the start codon (AUG),
respectively. Second, investigators have used computer models of
the secondary structure of mRNA to "visualize" mRNA regions that
might be susceptible to ODN targeting; these structural modeling
procedures are not, however, highly predictive of the actual
secondary structure of the mRNA in situ. Design of antisense ODNs
according to novel methods disclosed in the present invention,
however, is not dependent on either of these approaches to
antisense ODN design. Rather, disclosed herein is a novel
computer-based method for selecting unique "hotspots" in RNA
transcripts that are particularly well suited for targeting
antisense oligonucleotides for the purpose of inhibiting the
expression of genes and thereby greatly enhancing the likelihood of
producing therapeutic effects. The method herein described appears
to be an unexpected and substantial improvement over the two
conventional approaches to selecting target sites for antisense
oligonucleotides. There are now many examples of the successful use
of antisense ODNs to selectively block the expression of any of a
wide variety of gene targets, both in in vivo and in in vitro
studies. For example, in in vivo model systems: inhibition of Human
Immunodeficiency Virus (HIV) gene expression (including tax gene)
in human cells grown as xenogeneic transplants in animal models
(Kitajima et al., J. Biol. Chem. 267: 25881, 1992); targeting genes
in xenotransplanted human cancer cells in animal models, including
targeting c-myc, c-Ha-ras, NF-KB, c-myb, c-kit and bcr-Abl (Agrawal
et al., Proc. Natl. Acad. Sci. USA 86: 7790, 1989; Agrawal et al.,
Proc. Natl. Acad. Sci. USA 88: 7595, 1991; Biro et al., Proc. Natl.
Acad. Sci. USA 90: 654, 1993; Gray et al., Cancer Res. 53: 577,
1993; Higgins et al., Proc. Natl. Acad. Sci. 90: 9901, 1993;
Ratajczak et al., Proc. Natl. J.l.cad. Sci. USA 89: 11823, 1992;
Wickstrom et al., Cancer Res. 52: 6741, 1992; Skorski et al., Proc.
Natl. Acad. Sci. USA 91: 4504, 1994). In each of these instances
involving the administration of antisense ODNs to treat animals
with xenogeneic human cancers, the transplanted malignant cells
were found to regress.
[0020] A number of difficulties, however, have also been reported
in the use of antisense ODNs in in vitro studies, none of which
have proven to be insurmountable, however, in view of the existing
art and technology. In general. problems in the in vitro use of
antisense ODNs (most commonly phosphorothioates) have centered
around what has been viewed as "poor uptake" and/or the production
of unintended biologic effects; i.e., non-antisense effects. These
unintended effects fall into two major categories: first, there are
biologic effects that are attributable to the backbone structure of
the oligonucleotides; and, second, there are sequence-specific
non-antisense (aptameric) effects that appear to be dependent upon
the three-dimensional conformation of the ODN in solution, and,
consequently, on the positioning of the molecular electrostatic
(ionic) charges associated with the ODN molecule. Like ODN
antisense effects, both of these non-antisense effects are dose
dependent.
[0021] There are large differences in the capacity of similar ODNs
directed to transcripts of a given gene to block the expression of
that gene in cells; the reasons appear to be related to variations
in the availability of the particular target site on the transcript
complementary to the antisense ODN. No method has previously been
described which permits antisense ODNs to be designed so that, with
a high probability, some will exhibit optimal activity in the
purpose intended. Hence, what has been referred to as the "poor
uptake" of ODNs by some cell types in vitro may in large part
reflect the use of antisense ODNs that are not properly designed
and are, therefore, not optimally potent. It is also possible that
the culturing of cell lines under atmospheric oxygen conditions
(which is the usual and common in vitro practice) produces a
situation in which antisense ODNs are made less active than they
may be at much reduced (and more physiologically-relevant) oxygen
tensions (Smith L J and Kay H O. Unpublished observations). The
basis of this latter phenomenon could be due, at least in part, to
the increased generation of reactive free oxygen radicals under
ambient (atmospheric) oxygen levels by cells following treatment
with any of several types of ODNs, such as phosphorothioates.
Highly-reactive free oxygen radicals have been shown to have the
capacity to alter the lipids in the surface membranes of cells, and
to activate certain second-messenger pathways. Such alterations
could lead to an inhibition of antisense ODN uptake and/or to other
non-antisense ODN-dependent biologic effects. A complete blockade
of the induction of free radical formation by cells in response to
exposure to ODNs at physiologic oxygen levels would require the
presence of potent antioxidants such as, for example, vitamin C or
vitamin E. Finally, in general, it appears that antisense ODNs are
more active when used on freshly-obtained patient specimens than
they are when used on established cell lines, either in vitro or in
vivo. Furthermore, at least some established cell lines appear to
be more responsive to antisense ODNs when studied in vivo in animal
models than when studied in vitro in cell cultures. Dean and McKay
(Proc Natl. Acad. Sci USA 91: 11762,1994), for example, found that
an antisense ODN directed to PKC.sub..alpha. could inhibit the
growth of the C127 murine mammary epithelial cell line both in
vitro and in vivo. To get enough ODN into the cells grown in vitro
to reduce PKC.sub..alpha. expression and inhibit growth, cationic
liposomes had to be utilized. The naked antisense ODN, however,
worked very well when injected into mice carrying the C127 cell
line as a transplant. Again, this greatly-superior in vivo response
is consistent with the concept that the ODNs cause a much lower
level of free radical production in the animal.
[0022] Similarly, the apparent unintended backbone-dependent
biologic effects of antisense phosphorothioate ODNs on treated
cells can be eliminated (or adequately reduced) by the use of more
appropriately designed (and, therefore, more efficacious) antisense
ODNs. These unintended biologic effects, of which the inhibition of
cell proliferation is most common, generally only occur at
phosphorothioate concentrations of 5-10 micromolar (.mu.M) or
greater in the final culture medium. The better designed and more
potent antisense ODNs, however, are biologically most active at at
least 10-fold lower concentrations, particularly when fresh tissue
is used, or when the antisense ODN is used in vivo.
[0023] Pronounced aptameric effects usually appear to be the
property of only a small proportion of ODN. Aptameric effects
result when an ODN binds tightly and specifically to a particular
biomolecule, and, as a result, modifies the biological
function/behavior of said biomolecule. Aptameric effects are
dependent on the nucleotide sequence in the ODN. Presumably, the
sequence dependence of these effects reflects the fact that ODNs
with different nucleotide sequences assume different spacial
conformations, dictated by neighboring nucleotide-nucleotide
interactions. The nature of the backbone chemistry, however, is
also relevant in aptameric effects since said chemistry (and
associated molecular electrostatic [ionic] charges) also influences
the overall spacial conformation which the ODN molecule can assume
in solution. Only a subset of the possible aptameric effects which
an ODN might produce, however, would be expected to be an absolute
counter-indication for therapeutic use as an antisense compound.
Any such undesirable effects can be overcome by simply choosing
another antisense ODN directed to the same target transcript, but
which contains a different nucleotide base sequence, or, in some
instances, by changing the ODN backbone. The former option may
involve selecting an entirely different "hotspot" on the
transcript, or simply making modest changes (length, position) in
the ODN in question. Changes in ODN secondary structure may also be
achieved by making a small number of base substitutions, such as
with inosine, that do not interfere significantly with binding of
the ODN to the target RNA transcript.
[0024] In contrast, some antisense ODNs may possess aptameric-like
effects that enhance their therapeutic efficacy. The present
inventor, for example, has found that some antisense ODNs (in
particular, phosphorothioate ODNs) which target MDR1 gene
transcripts (and thereby inhibit P-glycoprotein expression)
apparently can also reduce MDR1 mRNA levels by an aptameric-like
effect that presumably involves the inhibition of second messenger
pathways such as the protein kinase-C and/or protein kinase A
pathways.
[0025] Many of the in vitro successes in the application of
antisense ODNs for therapeutic purposes have been readily
extrapolated to in vivo use. This is evidenced by the numerous
publications showing the in vivo efficacy of antisense ODNs.
Furthermore, several ODNs have already been approved by the United
States Food and Drug Administration for clinical testing. It should
be noted that the ODN uptake problems sometimes encountered in in
vitro studies have not been reported to be problematic in in vivo
studies. Pharmacologic/toxicologic studies of phosphorothioate
antisense ODNs have shown that phosphorothioates are adequately
stable under in vivo conditions, and that they are readily taken up
by all the tissues in the body following systemic administration
(Iversen P I, Anticancer Drug Design 6: 531, 1991; Antisense Res
Develop 4:43, 1994; Crooke, Ann Rev Pharm. Toxicol. 32: 329, 1992;
Cornish et al., Pharmacol. Comm. 3: 239, 1993; Agrawal et al., Proc
Natl. Acad. Sci USA 88: 7595,1991; Cossum et al., J. Pharm. Exp.
Therapeutics 269: 89, 1994). In addition, these compounds readily
gain access to the tissue in the central nervous system following
injection into the cerebral spinal fluid (Osen-Sand et al., Nature
364: 445,1993; Suzuki et al., Amer. J. Physiol. 266: R1418, 1994;
Draguno et al., Neuroreport 5: 305, 1993; Sommer et al.,
Neuroreport 5: 277, 1993; Heilig et al., Eur. J. Pharm. 236: 339,
1993; Chiasson et al., Eur J. Pharm. 227: 451, 1992).
Phosphorothioates per se have been found to be relatively
non-toxic, although a few particular ODNs have produced unintended
toxic effects in animals. The latter instances of toxicity seem to
be attributable to an unexpected aptameric effect on the part of
the ODN in question.
[0026] In summary, it appears that antisense ODNs have the
essential properties which make them useful as therapeutic agents,
both in vivo and in vitro. In vitro antisense activities now can
reasonably be expected to be seen in vivo. Two major areas needed
for further development of antisense ODNs as therapeutic agents
involve (a) the choice of gene targets for diseases like cancer and
atherosclerosis, Alzheimer's and schizophrenia, and (b) methods for
the selection of optimally active antisense. ODNs directed to a
particular gene target. These needs are addressed by the novel
inventions herein described in the present invention.
SUMMARY OF THE INVENTION
[0027] In accordance with this invention, there is provided a
method for reprogramming cell behavior to achieve therapeutic
effects through manipulating patterns of TR expression.
[0028] Also provided is a method for treating an individual having
an AP disease comprising administering to said individual an
effective amount of a composition selected from the group
consisting of an expression vector, a double stranded ODN, and an
antisense ODN. Said composition must be capable of regulating
expression of a TR. Said TR is expressed by the AP cells and
further characterized by the fact that it exhibits a
therapeutically useful change in said cell behavior in the
Reprogramming Test of this invention (hereinafter more fully
described). It is noted that when the AP disease is AIDS, said TR
is not encoded by HIV. In the case of cancer, said TR is a Traitor
Gene of this invention (more fully discussed hereinafter) and,
preferably, excludes oncogenes, e.g. fos, myc, myb, rel, jun(in an
altered form).
[0029] Another embodiment of this invention is a method for
treating an individual having a clinical disorder comprising
administering to said individual an effective amount of a
composition selected from the group consisting of a double stranded
ODN and an antisense ODN. The composition is capable of regulating
expression of a TR. The TR is expressed by therapeutically relevant
cells and is further characterized by exhibition of a
therapeutically useful change in said cell behavior in the
Reprogramming Test of this invention.
[0030] The invention revealed here primarily embodies a new type of
therapy based on reprogramming cellular behavior. Collateral
inventions, however, also follow including: (1) the diagnosis
and/or staging of aberrant programming diseases by assaying for the
expression of particular transcriptional regulators and their
variants in diseased cells; and, (2) for any given aberrant program
disease, the use of test agents in vitro for determining the
optimum agent(s) for treating any particular patient.
[0031] Thus, there is provided a method for diagnosing or staging
an AP disease comprising identifying the relevant subset of TRs
expressed by AP cells from an AP patient. A method for selecting
the most efficacious treatment regimen for an AP disease forms
another embodiment. This embodiment comprises identifying the
relevant subset of TRs expressed by AP cells from an AP patient.
These embodiments are described more fully hereinafter.
[0032] In addition, the invention provides a method for treating
therapeutically relevant cells from an individual having a clinical
disorder prior to transplantation of the cells back into the
individual (autologous transplant) embodiment. This embodiment
comprises the steps of: [0033] a) obtaining therapeutically
relevant cells from the individual and [0034] b) exposing the
therapeutically relevant cells to a reprogramming amount of an ODN
having a sequence complementary to a sequence of RNA transcribed
from a TR regulated gene or double stranded ODN ligand of a
transcriptional regulator present in the TR cells. In a preferred
embodiment the cells are taken from prenatal tissue or from a
different donor than the individual under treatment (allogeneic
transplant).
[0035] Selection of the most efficacious treatment regimen for an
AP disease forms another embodiment of this invention. This method
involves removing and culturing AP disease cells from an AP disease
patient with an antisense ODN specific to a TR from the relevant
subset of TRs expressed by AP cells from an AP patient or a double
stranded ODN to the DNA binding domain of such TR to determine
optimal treatment.
[0036] In carrying out the methods of treating AP diseases of this
invention it is critical to select the proper targets. Hence, an
important embodiment of this invention is a method for the
selection of a target for the treatment of an AP disease comprising
(i) determining the subset of transcriptional regulators and their
direct modifiers expressed by the aberrantly programmed tissue, the
corresponding normal tissue, or the constitutively self-renewing
normal tissue or, alternatively, making a similar determination for
any other normal tissue that is to be therapeutically manipulated
in accordance with this invention; (ii) adding or subtracting
expression of transcriptional regulator(s), or their direct
modifiers, from cells to be therapeutically reprogrammed and the
appropriate control tissue; (iii) scoring effect on cellular
programming and selecting potential therapeutic agents according to
the Reprogramming Test; (iv) testing effect of addition or
subtraction of the function of particular transcriptional
regulators, using the agents selected, (in an animal model system
if the therapeutic agents are for systemic use), and (v) reducing
or eliminating any undesirable side effects that might be produced
by the potential therapeutic agents. This embodiment is described
in detail hereinafter.
[0037] Exploiting specific cell type differences in target RNA for
selecting differentially available sites for ODN binding forms
another embodiment of this invention. This embodiment comprises a
method for cell type dependant targeting of specific RNA
transcripts comprising selecting an ODN capable of binding to and
leading to the destruction of said RNA in the tissue to be
therapeutically manipulated, but not in tissue where side effects
are produced by destruction of said RNA. Exemplary is the use of an
antisense ODN directed to cyclooxygenase RNA that selectively binds
to and destroys said RNA in hematopoietic tissue while avoiding
said RNA in gastrointestinal tissue.
[0038] All of the foregoing embodiments involve reprogramming of
cell behavior to achieve therapeutic effects through manipulating
patterns of TR expression.
[0039] Yet another embodiment of this invention is a novel
computer-based method for selecting "hotspots" in the transcripts
of particular genes, which hotspots define the possible sequences
that the corresponding antisense ODNs can take. These hotspots
include the binding sites for the most highly active of the
possible antisense ODNs as judged by their capacity to suppress the
expression of the corresponding gene. This computer method, based
on the commercially-available "OLIGO" program created by Dr.
Wojciech Rychlik (Rychlik and Rhoads, Nucleic Acids Res. 17: 8543,
1989; copyrighted 1989) utilizes parameters intuitively chosen by
the inventor to direct the "OLIGO" computer program in the
selection of antisense ODNs that have a high probability of being
highly active. In this novel approach, the two criteria
conventionally used to select antisense ODNs (RNA secondary
structure and biologically functional domains within RNA
transcripts) are ignored. The criteria for ODN selection used in
the novel method described herein relate to the
base-sequence-dictated physical properties of the prototype or test
antisense ODNs, the nucleotide base sequence of which is determined
by the nucleotide sequence of the "hotspot" region of a particular
gene being evaluated.
[0040] When applied to the analysis of transcripts of TRs or their
direct modifiers, this novel computer-based method can be used to
select the most highly active therapeutic antisense ODNs for the
treatment of Aberrant Programming Diseases in accordance with the
Aberrant Programming Disease model disclosed herein, or for the
therapeutic reprogramming of normal cellular functions. In
addition, antisense ODNs designed by the novel computer-based
method disclosed can be used to block the expression of genes known
to be directly implicated in a variety of disease processes or
which are known to be directly involved in biologic functions of
therapeutic importance. ("Direct" involvement of a particular
molecule in this context is to be contrasted with the direct
involvement of patterns of expression of particular regulatory
molecules [as part of a combinatorial regulation system that
controls the cellular differentiation program and subprograms such
as proliferation and apoptosis], as considered by the Aberrant
Programming model in cellular reprogramming). These direct
cause-and-effect associations between particular molecules or
groups of molecules and particular disease or other biologic
processes make the choice of possible target gene for therapeutic
antisense ODN inhibition obvious to one with ordinary skill in the
art. As examples: (1) .beta.-amyloid precursor and apolipoprotein E
are implicated in the pathogenesis of Alzheimer's Disease. They
are, therefore, obvious antisense targets, as are (2) vascular
endothelial growth factor (VEGF) which is implicated in cancer and
in Rheumatoid Arthritis; (3) cyclooxygenase which is involved in
pathologic inflammatory conditions such as Arthritis; and (4) the
expression of molecules known to be directly involved in the
regulation of apoptosis, such as the variants of bax, bcl-2, and
bcl-x, which can be blocked by antisense ODNs for the purposes of
promoting or inhibiting apoptosis in accordance with the
therapeutic needs of the situation. It would be desirable to block
apoptosis, for example, following ischemic damage resulting from
the occlusion of blood vessels leading to an organ such as the
heart or brain. conversely, it would be desirable to promote
apoptosis in malignant or atherosclerotic-programmed cells.
Associations between particular target genes and disease processes
are shown in Table X.
[0041] Table XI lists some of the gene/proteins other than
transcriptional regulators that have been implicated in the control
of apoptosis or programmed cell death. Select antisense ODNs
directed to these targets should be therapeutically useful for the
treatment of the diseases/conditions listed in Table XII, which
lists some of the diseases/processes in which apoptosis (or
programmed cell death) appears to playa key role.
[0042] The antisense ODNs targeted to TR or their direct modulators
selected by the criteria disclosed herein can find uses beyond the
treatment of the Aberrant Programming Diseases in accordance with
the Aberrant Programming model. For example: [0043] (1) TR involved
in the control of cellular programming also can function to control
the expression of particular genes such as telomerase and
.beta.-amyloid precursor protein which are implicated in the
production of certain disease processes (Sp-1, Ap-1 and Ap-4, for
example, are among the TR known to regulate .beta.-amyloid
precursor protein expression; and certain Hox genes are likely to
be involved in the control of telomerase expression). Hence,
blocking the expression of TR required for the expression of these
medically important molecules can find therapeutic use. Other TR
not involved in the regulation of cellular programs are restricted
in their function to controlling the expression of particular genes
associated with a particular state of cellular differentiation, or
controlling the expression of housekeeping genes. These latter
genes can be of clinical importance, and antisense ODNs which
inhibit expression of TR involved in promoting or controlling the
disease process may produce a desirable therapeutic effect.
Examples of such TR include Ref-1, and, possibly, members of the
GADD family of TR. [0044] (2) TR encoded by the host cell are known
to be important for the expression and functioning of infecting
viruses. Indeed, blocking the action of NF-K.beta. in HIV-infected
cells by ODNs have been shown to reduce HIV expression. Examples of
virally-induced diseases that would benefit from such treatment
include, but are not to be limited to, those caused by HIV, HTL V,
CMV, Herpes simplex, measles viruses, the hepatitis virus variants,
rhinoviruses, influenza viruses and hemorrhagic fever viruses.
Host-encoded transcriptional regulators that are known to regulate
the following types of virus are given as examples: [0045] HIV:
USF, Elf-1, Ap-1, Ap-2, Ap-4, Sp-1, Sp-3, Sp-4, p53, NF-k.beta.,
rel, GATA-3, UBP-1, EBP-P; [0046] CMV: SRF, NF-k.beta., p53, Ap-1,
IE-2, C/EBP; [0047] Herpesviruses: USF, Spi-1, Spi-B, ATF, CREB and
C/EBP families, E2F, YY-1, Oct-1, Ap-1, Ap-2, c-myb, NF-k.beta.;
[0048] Hepatitis viruses: NF-1, Ap-1, Sp-1, RFX-1, RFX-2, RFX-3,
NF-k.beta., Ap-2, C/EBP. [0049] (3) Arguably, TR--particularly
those involved in the control of cellular programming--also
regulate higher-order functioning in the nervous system. Antisense
ODNs directed to c-fos, for example, have been shown to alter
neurological functioning in animal models (Dragunow et al.,
Neuroreport. 5: 305, 1993). According to the present invention,
altered patterns of TR expression in nerve cells can result in
Aberrant Programming of the nerve cells, resulting in certain
mental disorders such as schizophrenia. Hence, antisense ODNs,
selected using the method described herein are expected to be of
use for such diverse medical needs as the treatment of psychoses,
depression and epilepsy.
[0050] Table X provides a general overview of some of the diseases
in which the gene targets presented herein have been implicated.
All of the transcriptional regulators (and their direct modifiers)
known to be involved in the regulation of cellular differentiation,
proliferation and/or apoptosis cellular programs (or "Programmed
Cell Death") are candidates for the preferred embodiment which is
the treatment of Aberrant Programming Diseases in accordance with
the Aberrant Programming model. Particular subsets of these TR that
are appropriate targets for developing antisense therapeutics for a
particular disease are delimited by determining which of these TR
are expressed by the diseased cells in question.
[0051] Hotspots and prototype ODNs are also provided for certain
transcriptional regulators that have not yet been implicated in the
control of these cellular programs. This association, or lack
thereof, can be established by anyone with ordinary skill in the
art, using established methods. The appropriate disease
applications of antisense ODNs directed to the other gene targets
for which "hotspots" and prototype antisense ODNs have been
provided herein are related on the usual cause-and-effect basis
discussed elsewhere herein.
[0052] It follows from the AP model that there is not simply just
one possible "therapeutic solution" when one is confronted with
developing an antisense ODN therapeutic to treat AP diseases in
accordance with the present invention. That is, several different
antisense ODNs--directed against different members of a select set
of TR gene targets--may be active in treating the same disease.
This situation is a direct consequence of the facts that
(a) the TR involved in cellular programming are acting in an
interdependent way as part of a combinatorial regulation system,
and that (b) different TR combinations can direct the same change
in cellular programming. In the analogy given between the AP
cellular programming model and the grammatical rules of the English
language (see Table I), the latter phenomenon is referred to as
"synonyms".
[0053] Thus, as a consequence of the nature of the AP diseases and
as a consequence of the nature of the non-antisense ODN effects on
cells, it will not be possible a priori to determine on purely
theoretical grounds which particular antisense ODNs will be useful
for the treatment of any given AP disease. The major advantage of
the present invention over the prior art is that it makes the
fundamental nature of the AP diseases comprehensible, and, thereby,
allows the most rational approach possible to be applied to the
treatment of these diseases. This rational approach which is laid
out here is an enormous improvement over the prior art in that--in
combination with the methods described herein for selecting
antisense ODNs--it makes it possible to have a rapid advance in the
development of therapy for diseases that have been almost totally
intractable using the existing art.
BRIEF DESCRIPTION OF THE DRAWINGS
[0054] FIG. 1 is a graph showing the effects of antisense
oligonucleotides on incorporation of thymidine into DNA in G1
cells.
[0055] FIG. 2 is a graph showing the effects of different antisense
oligonucleotides on viability of AML blasts.
[0056] FIG. 3 is a graph showing the effects of different MDR
oligonucleotides on thymine incorporation into 8226/Dox4 cells.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
AP Disease Model and List
Definition of a "Cellular Program"
[0057] The coordinated appearance in cells of a cell type
restricted pattern of gene expression over time that provides for a
particular phenotype and as a result for the determination of the
range of possible cellular responses to exogenous stimuli.
[0058] The fundamental program can be thought of as a
differentiation program which in turn controls the subprogram
responses of the cell to environmental and other exogenous cues
where the subprograms include cellular viability (apoptosis) and
proliferation.
Definition of an "Aberrant Programming Disease"
[0059] One in which the direct cause of the clinical disorder is
the expression in the primary diseased cells of a differentiation
program that does not normally exist. That is, there is an
expression of normal genes that provide for particular
differentiated phenotype in abnormal combinations. The result is
that these diseased cells become capable of expressing pathogenic
behaviors involving cellular differentiation, viability and
proliferation. These attributes of the primary diseased cells can
also induce pathologic change, in their tissue environment.
[0060] The term "direct cause" with respect to pathogenesis is to
be distinguished from "risk factors." Typically an AP Disease will
be associated with numerous risk factors that in various
combinations appear to "cause" the appearance of the disease. In
fact, however, they cause the changes in the pattern of
transcription regulator (TR) expression and chromatin domain
availability which in turn causes the disease. This is important
because programs can evolve and can become independent of any risk
factors involved in their induction. Risk factors include mutagenic
events, viruses, chromosomal abnormalities, genetic inheritance,
and diet.
[0061] Aberrant programming disorders can be manifested as either a
hyperplastic or a hypoplastic (degenerative) disease or a
combination of both.
Examples of Diseases where the Aberrant Program Phenotype is
Expressed:
[0062] Cancer
[0063] Myeloproliferative Diseases [0064] polycythemia vera [0065]
agnogenic myeloid metaplasia [0066] essential thrombocytosis
[0067] Myelodysplasias [0068] refractory anemia [0069] refractory
anemia with ringed sideroblasts [0070] refractory anemia with
excess blasts [0071] refractory anemia with excess blasts in
transition
[0072] Atherosclerosis
[0073] AIDS-related complex
[0074] AIDS
[0075] Alzheimer's Disease
[0076] Autoimmune Diseases, including: [0077] inflammatory bowel
diseases [0078] multiple sclerosis [0079] Rheumatoid arthritis
[0080] Systemic lupus erythematosus
[0081] Schizophrenia
Molecular Model
[0082] According to the molecular model set forth herein, the basic
disease causing entity in the AP diseases is a specific type of
relational alteration among certain cellular components involved in
program control. It is unlike any previously described molecular
pathogenic mechanism. This model defines the nature of a novel
therapy for these diseases, limits the potential set of
therapeutically useful targets to a relatively small number of
genes and leads to the unobvious conclusion that the manipulation
of certain "normal" genes is an appropriate approach for the
treatment of AP diseases, in this way the model makes the reduction
to practice of the proposed therapy straightforward and accessible
to anyone skilled in the art.
[0083] Specifically, the essential molecular pathology in the AP
diseases consists of changes in the interdependent patterns of TR
expression and/or chromatin domain availability for transcription.
In turn, these relational alterations provide for the expression of
abnormal cellular programs involving cellular differentiation which
are pathogenic. Particular TR or certain molecules involved with
the control of domain status may be structurally abnormal. However,
these are not necessarily useful targets for therapeutic
intervention.
[0084] Tenets of the model relevant to the development of specific
therapy:
A) Those True of both Normal and Aberrant Programming: [0085] 1)
The pattern of domain availability determines the possible range of
genes that can be expressed in the cell and, therefore, limits the
range of cellular programs that can be expressed. [0086] 2) The
pattern of TR expression is the molecular equivalent of a
programming code. By analogy with language particular combinations
of TR (letters) working as a unit (words) regulate the expression
of sets of genes in a coordinate fashion while the complete set of
TR combinations used in any given cell (sentence) determines which
of the possible phenotypes the cell will expressed, and therefore
the overall character of the cell's differentiation program (see
Table I for more details where cancer is used as an example).
[0087] 3) Only a subset of the total number of TRs involved in the
control of cellular differentiation for the total organism are
expressed in any given cell type and they are few in number. [0088]
4) Similar effects on particular patterns of gene expression
(programming) can be achieved by more than one specific combination
of TR (synonyms). [0089] 5) The specific functional consequences of
a particular TR's being expressed is context-dependent. That is,
its effects on cellular programming depends both on which other TR
it combines to regulate a particular set of genes (what words it
appears in) and on the total set of different TR combination
expressed by the cell (the sentence).
B) True of AP Cells but not Normal Cells:
[0089] [0090] 1) The combinations of TR seen in AP cells is
different from that seen in any normal cell (the sentence is not
expressed by any normal cell). [0091] 2) The specific functional
consequences of any given particular TR being expressed in an AP
cell, therefore, will be different from the consequences seen in a
normal cell. [0092] 3) AP cells, therefore, express a cellular
differentiation program that is different from any normal
differentiation program. As a result AP cells express pathogenic
behaviors resulting from their altered differentiation, viability
and proliferation characteristics. [0093] 4) Hence, equivalent
manipulations of the expression of a given TR in normal cells vs.
aberrantly programmed cells can produce differential effects on
cellular behavior. This can form the basis of therapeutic
intervention. [0094] 5) The subset of TRs expressed by any AP cell
is expected to include TR not expressed by the corresponding normal
cells and/or conversely. These TRs within the AP cells will be
normal TRs ectopically expressed or modified (alternate splicing
promoter use or post-translational modification) or mutated to a TR
with altered binding properties.
[0095] It follows from the rationale given herein that--for any
given AP disease made up of disease variants that contain AP cells
that have distinctive phenotypes of pathological significance--that
these distinct phenotypes will have correspondingly variant
patterns of TR expression. In cancer, for example, malignant cells
with a multidrug-resistant phenotype would be expected to have a
somewhat different pattern of TR expression than drug sensitive
malignant cells. However, these variant phenotypes, such as
multidrug resistance, should be considered as being an instance of
AP cells expressing a subprogram of the greater malignancy program
that produces cancer. It should be possible, therefore, to select
and target ODNs separately to the malignancy program or to the
multidrug resistance subprogram, for example, by attacking
different TRs; or to select ODNs which would simultaneously effect
both the cancer per se and the subprogram (which, in this example,
is multidrug resistance).
[0096] When a chronic AP disease undergoes clinical progression,
there is an associated progression in the pattern of some of the TR
expressed in the diseased cells, although expression of the
majority of TR involved will continue to be unchanged. This program
progression has several possible consequences, including the
consequence where the AP disease being expressed by the disease
cells may become independent of some disease risk factor such as,
for example, a mutated oncogene in the case of cancer. This is to
be contrasted with what could be called acute AP diseases in which
(a) at least one pathological risk factor must be currently acting
on the diseased cells to produce the disease; (b) the disease is
rapidly produced as a consequence; and (c) the disease involves a
re-programming of the diseased cells. Perhaps the best example of
this type of disorder are the result of an infection with certain
pathogenic viruses (Table XII). It follows from the definition of
these disorders as acute AP diseases that certain host cell-encoded
regulators of cellular programming are appropriate targets for
therapeutic intervention utilizing antisense ODNs disclosed herein,
even if said TR do not directly up-regulate the expression of the
virus in question. The host cell gene targets likely to be best
suited for this purpose are TR and other molecules involved in the
control of apoptosis and cellular responses to free radicals
(Tables X and XI).
[0097] The interpretation of these viral diseases as being acute AP
diseases leads to an important conclusion. That is, it will be
possible to inhibit the expression of a TR encoded by the host cell
where said TR is involved in controlling the cellular programming
of that cell, such as, for example, by exposing that host cell to
an antisense ODN directed to the particular TR in question, the
antisense ODN being used either alone or in combination with an
augmentation agent such as, for example, an anti-oxidant or an
interferon. The impact of this ODN treatment on the programming of
said cell will be different when comparing an uninfected with a
similar cell expressing the infecting virus. The choice of
antisense ODN for the purpose of this invention will be one that
exploits this difference to a therapeutic advantage, such as, for
example, blocking the ability of a virus to induce apoptosis. Thus,
as in the case of the chronic AP diseases, this situation provides
the basis for the novel approach to the development of therapeutic
agents as described herein.
Nature of Targets
[0098] TRs are the primary targets for therapeutic manipulations
based on the model. They may be manipulated directly or indirectly
through molecules such as tyrosine kinase, that can effectively
change a TR of one type to another through structural alterations
such as phosphorylation.
[0099] While all the TR that are involved in cellular programming
are, in accordance with the present invention, potential targets
for therapeutic antisense ODNs for the disease(s) in which they are
expressed, some TR appear to be more likely than others to be
highly useful. There is growing evidence, for example, that many
(if not all) of the chronic AP diseases frequently share a common
type of insult which contributes to the pathogenesis of the
disease. This insult consists of an attack on cells by oxygen
free-radicals. In AIDS, for example, there is evidence that
HIV-infected lymphocytes undergo apoptosis following exposure to
free-radicals (Sandstrom et al., AIDS Res Human Retroviruses 9:
1107, 1993). At least in part this induction of apoptosis is
mediated by lipid peroxidation (Sandstrom et al., J Biol. Chem.
269: 798, 1994). This general conclusion about AP Diseases suggests
that (1) those cellular program regulators (TRs, in particular)
that are involved with cellular responses to free-radicals should
be given priority for testing as targets for antisense ODN
inhibition in accordance with the concepts presented herein; and
(2) there is a rationale for adding a free-radical generator or an
antioxidant, whichever is appropriate, with one of the antisense
ODNs disclosed herein for the treatment of an AP disease.
Furthermore, these same pathways may also be activated by at least
some pathogenic viruses that produce acute disease.
[0100] The free-radical generator could be used as an "augmentation
agent" in combination with an antisense ODN designed in accordance
with the present invention in diseases where the objective is to
kill the AP cell (for example, atherosclerosis, or cancer) while
the antioxidant would be appropriate as an augmentation agent where
the objective is to block apoptosis (for example, AIDS). An
antioxidant would be appropriate, also, in the latter cause,
because many types of ODN themselves induce cells to generate free
radicals. Free radical generators include, but are not limited to,
certain polyunsaturated fatty acids (including gamma linolenic
acid, eicosapentaenoate and arachidonate), chemotherapeutic agents
and ionizing irradiation. Antioxidants include, but are not limited
to, certain vitamins, minerals, trace elements and flavinoids. A
complete listing of antioxidants would include those known to those
skilled in the art, and may be found in standard advanced
textbooks, such as, for example, Zubay G L: "Biochemistry" (3rd
edition), in 3 Volumes, Wm C Brown Communications, Inc., 1993; and
in: Rice-Evans C A and Burdon R H (eds): "Free Radical Damage and
Its Control", New York: Elsevier, 1994; and in: Yagi K et al (eds):
"5th International Congress on Oxygen Radicals and Antioxidants",
New York: Excerpta Medica Press, 1992 (International Congress
Series, No. 998). A partial listing of anti-oxidants that have been
used clinically include, but are not limited to: ascorbic acid
(vitamin C), allopurinol, alpha- and gamma-tocopherol (vitamin E),
beta-carotene, N-acetyl cysteine, Desferol, Emoxipin, glutathione,
histidine, lazaroids, Lycopene, mannitol, and
4-amino-5-imidazole-carboxamide-phosphate. The choice of which
anti-oxidant(s) to use (if any) in conjunction with a particular
antisense ODN can be determined on an ad hoc basis by one skilled
in the art.
[0101] The principal effects of free-radical generators or
anti-oxidants on cells from the perspective of the AP model is to
produce an alteration in the pattern of TR being expressed, or, in
the case of antioxidants, to prevent the effects on cells produced
by cellularly-generated free radicals subsequent to ODN binding. It
follows from the AP model that this pattern will be different
following treatment with these "augmentation agents" when normal
cells are compared with AP cells. Hence, it is possible to combine
this treatment with an antisense ODN selected according to the
criteria given herein (for example, in the Reprogramming Test) and
expect different results for normal versus AP cells.
[0102] The TR that are known to be involved in cellular responses
to free-radicals and apoptosis include, but are not limited to: the
AP-1 group, including junD; the Egr group; gadd group; Hox group;
IRF group; the MAO-, Max- and Mxi-groups; myc- and myb-groups;
NF-kB group; p53; Ref-1; Sp-1; TR-3 and TR-4; and USF (for a more
comprehensive list of TR, see Tables II and X). Other genes
involved include those directly involved in the regulation of
apoptosis, and are shown in Table XI.
[0103] It should be noted also that AP cells with major phenotypic
differences in accordance with the AP model would be expected to
have correspondingly somewhat different patterns of TR expression.
One example is the multidrug resistance phenotype that is expressed
by a substantial percentage of human cancer cells. Hence, it should
be possible in accordance with the present invention to select
antisense ODNs (preferably those directed to a TR involved in the
maintenance of the multidrug resistance phenotype) and
preferentially sensitize multidrug resistant cells to
chemotherapeutic agents without getting a correspondingly increase
in sensitivity to drug by drug sensitive cells.
Nature of Theraipeutic Intervention
[0104] The basis of the novel therapy is to differentially change
the pattern of gene expression in AP cells by altering the pattern
of TR expression. The model states that the specific functional
consequences of the expression of any given TR is
context-dependent. It therefore follows that the same TR present in
both normal and AP cells can be manipulated in the same way and a
different impact on cellular behavior obtained. A TR expressed only
by the AP cells, however, also may be targeted. The end result is
that the pattern of gene expression in the AP cells lose at least a
substantial portion of their disease-producing activity. This can
be manifested in numerous possible ways including death of the AP
cells, a change in their differentiation status with a concomitant
change in the production of disease-producing factors or to a loss
of proliferative potential.
[0105] The number of transcriptional regulators that will have to
be manipulated in any given cell type will be very small. There are
estimated to be 30,000 to 100,000 genes in the human genome
distributed over 3.times.10.sup.9 bp of DNA. In any given cell type
approximately 10,000 genes can be shown to be expressed. Greater
than 90% of these are expressed by many cell types and the large
majority of these are referred to as "housekeeping genes."
[0106] Typically, the number of genes that can be shown to be
differentially expressed in any given cell type account for only a
few hundred. It is these genes that make the difference between
liver cells and brain cells, for example. The large majority of
these are directly involved in carrying out the functions that
characterize the cell type. Liver cells, for example, express a
wide range of enzymes that are involved in ridding the body of many
types of chemicals. The genes of interest for the purposes of this
patent are the small subset of genes coding for molecules involved
in the differential regulation of cell type specific genes. In
particular, transcriptional regulators and their direct modulators.
The latter includes, for example, certain tyrosine kinases, that
can modify a particular transcriptional regulator and, in effect,
change it to a functionally different transcriptional regulator.
(Berk Biochem. Biophys. Acta. 1009, 103, 1989) For the purposes of
this invention transcriptional regulators are defined as molecules
that bind to specific DNA sequences variably expressed by different
genes and/or to other transcriptional regulators at least one of
which must bind to specific DNA sequences. As a result they control
the levels of gene expressions by means of modulating RNA
polymerase activity. The transcriptional regulators may be of
either endogenous or exogenous origin. They may either be normal or
be mutated.
[0107] The ability of transcriptional regulators to variably
interact with each other provides the basis for a combinatorial
regulatory system. This allows a very small number of
transcriptional regulators to control the expression of a large
number of genes in various patterns. Particular sets of genes being
controlled at any given time by a certain subset of the
transcriptional regulators being expressed by the cell. Each
transcriptional regulator subset, therefore, is a programming code
or an instruction or a "word" that directs the expression of a
particular gene set. The entire pattern of gene expression being
expressed by a given cell type can be thought of as a sentence,
since only certain words can appear together. Perhaps the first
demonstration of the involvement of combinatorial regulation in the
control of a cellular differentiation program was provided by
Chalfie and his colleagues as part of their studies of C. eleqans
(Mitani et al., Dev. 119: 773, 1993).
[0108] A general role for combinatorial regulation being involved
in eukaryotic gene expression has been previously postulated by
several investigators. (Scherrer, and Marcand J. Cell Phys 72, 181,
1968; Sherrer Adv. Esp. Med. Biol. 44, 169, 1924; Gierer Cold
Spring Harbor Symp Quant Biol 38; 951, 1973; Stubblefield J. Theor
Biol 118, 129, 1986, Bodnar J. Theor Biol. 132, 479, 1988) Lin and
Riggs (Cell 4, 107, 1975), demonstrated using biophysical arguments
the impossibility of having a separate regulator for every gene in
a eukaryotic cell. Combinatorial regulation models of eukaryotic
gene expression generally postulate multiple levels of regulation
in addition to transcription. In principle, these models show how
theoretically 100,000. genes could be selectively controlled by as
few as 50 regulatory molecules only a small subset of which would
operate at the level of what is defined here as transcriptional
regulators. Bodnar J. Theor. Biol. 132,479,1988. The actual number
of human transcriptional regulators are estimated to number on the
order of somewhere in excess of 100. (Table II lists those that
have been described in the literature.) Many, however, are known to
be expressed only in certain cell types. Since just a few hundred
genes determine the differences between particular differentiated
cell types and the large majority of these determine the particular
functional features of the cell, only a very small number of these
can be regulator gene products.
[0109] It follows, therefore, that the number of regulators that
must be manipulated to achieve the effects stipulated by this
invention for any given application is small and can be managed
with comparatively modest effort. It also follows from the notion
of combinatorial regulation that not all the transcriptional
regulators expressed by a given cell type need to be known before
this invention can be practiced.
[0110] The present inventor has found that antisense p53
oligonucleotides can inhibit the proliferation, including the
blocking of stem cell self-renewal, and ultimately kill primary
human leukemic blasts while not producing similar effects on fresh
normal bone marrow cells. This unobvious result indicates that the
interactive mechanisms for detecting, interpreting and responding
to environmental informational molecules involved in regulating
cell differentiation and proliferation and viability in AP cells
are so altered from normal in terms of their dynamic interactions
(involving signal transduction and interpretation) that the
inhibition of a single gene or set of genes coding for proteins
involved in this process by antisense oligonucleotides is
sufficient to change the impact of the informational molecules so a
change in cellular programming such as cellular death or growth
inhibition program can be selectively instituted in AP cells. The
term "traitor genes" is used herein to describe those genes in AP
cells that may be suitable for targeting for inhibition with
antisense molecules in accordance with the present invention
Suitable target or traitor genes may themselves either be
functionally abnormal or be normal but function to maintain the
pathological phenotype AP cells as part of an abnormal pattern of
gene expression. Such treatment results in differential programming
of AP cells, but not their normal counterparts over a selected dose
range. In the preferred embodiment the Traitor Genes to be targeted
are TRs.
[0111] The concentration of oligonucleotide to be used may vary,
depending upon a number of factors, including the type of cancerous
cells present in the marrow, the type, and the specificity of the
particular antisense oligonucleotide(s) selected, and the relative
toxicity of the oligonucleotide for normal cells. Although the
present inventor has observed significant AP cell programming at
oligonucleotide concentrations in extra-cellular fluid as low as 1
nanomolar, optimal inhibition was observed at concentrations of at
least 10 nanomolar in the model system described below. The upper
limit of the dosage range is dictated by toxicity and therapeutic
efficacy, and, generally will not exceed 5 micromolar. With the aid
of the techniques set forth in the present disclosure, those of
skill in the art should be able to determine the optimal
concentration to be used in a given case.
[0112] Phosphorothioates are currently the preferred chemistry for
antisense ODNs to be used clinically. A substantial body of
pharmacological/toxicological data for this class of compounds
shows that phosphorothioates generally behave in a manner similar
to most conventional drugs that are used systemically. As a result,
the basic pharmacologic principals that have been established over
the years apply here as well. For example, see the standard
textbook: "Principles of Drug Action: the Basis of Pharmacology",
3rd Edition (W. B. Pratt and P. Taylor, eds). New York: Churchill
Livingston, 1990. Thus, no novel pharmacologic principles or
procedures have had to be invented in order to adapt
phosphorothioates to in vivo use. There are some quantitative
differences between phosphorothioates and more conventional drugs
in terms of in vivo properties, but these differences can be
accommodated to particular needs by one with ordinary skill in the
art by an application of existing knowledge. In the future,
undoubtedly, clinically-superior ODN backbones will be identified
and/or developed. The "hotspots" for targeting specific genes and
the prototype ODNs revealed herein should work equally well with
any improvements in ODN backbone chemistry.
[0113] For the immediate purposes of this invention,
phosphorothioate antisense ODNs can be administered intravenously
(i.v.), intraperitoneally (i.p.), subcutaneously (s.c.) or
intramuscularly (i.m.), in order to treat systemic disease. Pendent
groups may be attached to the ODNs to aid tissue-specific
targeting, or the ODNs may be associated with carriers that
facilitate uptake, such as liposomes or charged lipids; in general,
however, such modifications will not be necessary. Antisense ODNs
can be delivered intrathecally or used in combination with agents
that interrupt the blood-brain barrier in order to treat conditions
involving the central nervous system. For local therapeutic
purposes, phosphorothioates can be applied topically in an
appropriate vehicle or delivered to particular organs or tissues by
catheters (or catheter-like devices) designed to direct the flow of
these compounds to particular sites in the body. Phosphorothioate
antisense ODNs may also be administered orally, associated with one
or more appropriate and acceptable carrier molecules or compounds,
processed for oral ingestion in the form of a tablet, capsule,
caplet or liquid. As described elsewhere herein, it may also be
desirable to administer the antisense ODN with other active agents.
For example, it may be desirable to add an antioxidant to an
antisense ODN preparation, or to administer both antioxidant and
antisense ODN simultaneously, to treat cancer patients whose tumors
express a multidrug resistance phenotype; or to treat AIDS patients
when the ODN backbone being employed is known to induce cells to
generate free-radicals. Free radicals can induce cells to express
higher levels of multidrug resistance, or boost the expression of
HIV, respectively.
[0114] The extracellular concentrations that must be generally
achieved with highly active phosphorothioate antisense ODNs is
believed to be in the 10-1000 nanomolar (nM) range. These levels
can readily be achieved in the plasma, for example, by infusing
phosphorothioates into patients at a rate of a few milligrams per
kilogram body weight per hour over a period of a few days. The
infusion rate will typically be higher for cancer patients because
they have higher levels of an as yet uncharacterized plasma protein
that tightly binds phosphorothioates. As for many drugs, dose
schedules for treating patients with phosphorothioates can be
readily extrapolated from animal studies.
[0115] For ex vivo applications, the concentration of the antisense
ODN(s) to be used is readily calculated based on the volume of
physiologic balanced-salt solution or other medium in which the
tissue to be treated is being bathed. In the large majority of
applications, the phosphorothioates can be assumed to be stable for
the duration of the treatment. With fresh tissue, 10-100 nM
represents the concentration extremes needed for an antisense
phosphorothioate ODN with a reasonably good to excellent activity.
Two hundred nanomolar (200 nM) is a generally serviceable level for
most applications. Incubation of the tissue with the ODN at 5%
rather than atmospheric (ambient) oxygen levels may improve the
results significantly, except for unusual situations (such as bone
marrow or peripheral stem cell purging with antisense ODNs directed
to p53) where generation of highly-reactive free radicals appears
to contribute to the desired therapeutic effects.
[0116] For some therapeutic applications, it may be preferable or
even necessary to administer the antisense ODN with another
augmentation agent. "Augmentation agents" as defined herein are
those agents that can alter patterns of transcriptional regulator
expression may include, but are not limited to: cytokines, cancer
chemotherapeutic agents, neuroleptics, anti-inflammatory agents,
anti-oxidants and free-radical generators such as certain
polyunsaturated fatty acids. The suitability of such potential
combination treatments generally can be determined initially in the
"Reprogramming Test" prior to studies in animal model systems. It
is also clearly desirable to add to the assay medium in the
Reprogramming Tests of said cells any cytokines which may be
associated in vivo with the diseased cells, in order to better
model in vivo conditions. The association of particular cytokines
with particular disease processes, or the usefulness of particular
cytokines in clinical procedures has been well described in such
standard reference works as: Thomson A W (Ed) "The Cytokine
Handbook" (2nd Edition), San Diego: Academic Press, 1994; Kunkel S
L and Remick D G: "Cytokines in Health and Disease", New York:
Decker, 1992; and Oppenheim J J et al (Eds), "Clinical Applications
of Cytokines", New York: Oxford Press, 1993. The choice of
particular cytokines as augmentation agents for a particular
disease application is to be limited to those cytokines to which
the disease and cells in question respond. Cancer chemotherapeutic
agents and free-radical generators could be used as augmentation
agents in applications where the objective is to destroy the
diseased cells, such as in, for example, cancer and
atherosclerosis. Anti-inflammatory agents and anti-oxidants could
be used as augmentation agents where the objective is to maintain
the viability and/or reduce the pathogenicity of the diseased
cells, such as in, for example, AIDS, Alzheimer's Disease, and
autoimmune diseases. Neuroleptics could be used as augmentation
agents in the treatment of diseases such as schizophrenia, where
they have already shown activity.
"Hardware" for Reduction to Practice
[0117] Using established techniques, assays and agents, the
following capabilities can be readily acquired. These can be used
by anyone skilled in the art to reduce the primary and collateral
inventions to practice.
[0118] 1) Assays for Transcriptional Regulators and their Direct
Modifiers. [0119] Preferred assays: RNA in situ hybridization (Lum
Biotech. 4, 32, 1986) or PCR (Block, Biochem 30, 2735, 1991) or
metabolic labelling (Ausubel et al (eds.) Current Protocols in
Molecular Biology, John Wiley NY, 1989 (updated semiannually) for
detecting expression at the protein level. [0120] Purposes: [0121]
To establish the subset of the known transcriptional regulators or
their direct modifiers that are expressed by a particular cell
type. This will serve the following functions: [0122] a) the
determination of the subset of transcriptional regulators, or their
direct modifiers, that are targets to be manipulated in the
reduction to practice; [0123] b) the evaluation of the
effectiveness of potential therapeutic agents in adding or
subtracting the expression of a particular transcriptional
regulator or its direct modifier cells; [0124] c) the diagnosis
and/or staging of a particular aberrant program disease; [0125] d)
the determination of the optimum therapeutic agent(s) in clinical
practice, when there are more than one option for a given
disease.
[0126] 2) Agents for adding or subtracting the expression of
particular transcription?` regulators or their direct modifiers in
cells to be therapeutically manipulate. [0127] a) Antisense
oligonucleotides (Zon, Pharmaceut. Res., 5, 539, 1988). [0128]
These agents can be used to subtract the expression of particular
genes from cells. Antisense oligonucleotides can produce an
induction of their target gene by inducing an induction of their
target gene by inducing an initial reduction in expression that is
then over compensated for as a consequence of a feedback loop in
the cells or by altering the half-life and/or translational rate as
a consequence of the antisense oligonucleotide producing a change
in the target RNA secondary structure and/or by altering the
binding of regulatory molecules to the target RNA. Such an acation
by antisense oligonucleotides is suitable for the pracatice of
theis invention. [0129] Design of "Test" Antisense Oligonucleotides
[0130] i) Using a computer program such as "Oligo" (Rychik and
Rhoads, Nucl. Acids Res., 17, 8543, 1989) select a set of antisense
oligonucleotides that bind to the RNA target of choice that have
the following characteristics: (1) length between 10 and 35 bases
with 20 being generally used; (2) negligible self-interaction
(self-dimers and hair pins) under physiologic conditions; (3)
melting temperature .gtoreq.40.degree. C. under physiological
conditions; and (4) no more than 40% of the oligonucleotide being a
run of guanines or cytosines); [0131] ii) Using a reference such as
Genbank ensure that the antisense oligonucleotide has .ltoreq.85%
homology with the RNA transcripts of other genes. An exception to
this is where an antisense oligonucleotide is selected on the basis
of its ability to bind to more than one member of a transcriptional
regulator family (such as the homeobox genes) on the basis of
sequence homology. [0132] b) Establishment of "prototype
therapeutic" antisense oligonucleotide from a set of test antisense
oligonucleotides. These prototype compounds will be used in the
reduction to practice. [0133] i) Synthesize test antisense
oligonucleotides using standard procedures, for example, those for
producing phosphorothioates (Vu et al, Tetrahedron Lett, 32, 3005,
1991). [0134] ii) Using assays for transcriptional regulators or
their direct modifiers select prototype therapeutic antisense
oligonucleotides out of the set of test compounds on the basis of
shutting down expression of the target gene in the cell types to be
therapeutically manipulated. In practice, the same set of prototype
agents capable of shutting down target gene expression in a variety
of cell types could be used in the Reduction to Practice, Step 2,
hereinafter, for multiple therapeutic objectives. [0135] c)
Synthetic double-stranded oligonucleotides that are ligands for the
DNA binding domain of one or more transcriptional regulators. (Wu
et al, Gene, 89,203, 1990).
[0136] Prototype therapeutic agents of this type for use in the
reduction to practice will correspond to actual gene sequences to
which the transcriptional regulator(s) will have been shown to bind
using standard techniques such as the gel mobility shift assay,
(Ausubel et al (eds,) Current Protocols in Molecular Biology, John
Wiley NY, 1989 (updated semiannually),) Alternatively, single
stranded oligonucleotides targeted to transcriptional regulator
binding sites and the adjacent sequences can also be used to block
the expression of particular genes in accordance with the present
invention. [0137] c) Expression Vectors
[0138] In the preferred embodiment a recombinant viral vector will
be used (Miller and Rosman, Biotech, 7, 980, 1989) that carries the
complete coding sequence of the transcriptional regulator or its
direct modifier. This will provide for expression of the regulator
or modifier in the cells of interest. It will be constructed and
tested using standard methods. (Ausubel et al, supra)
Alternatively, the viral vector will carry a sufficiently long
antisense sequence to such a regulator or modifier to provide for
the blocking of expression of the target gene in the cells of
interest.
[0139] 3) Preparation of Tissue
[0140] The preferred tissue is primary explant or early passaged.
It will be acquired using standard surgical procedures. Tissue
processing for culture and/or heterotransplant will be according to
established methods. Culture conditions for the disordered cells
from the various aberrant program diseases or their normal
counterparts are referenced in Table III. These references also
provide information on acquiring and processing the appropriate
cells.
[0141] Uses to provide the source material for:
[0142] a) determining the subset of the known transcriptional
regulators or their direct modifiers that are expressed by a
particular cell type.
[0143] b) practicing the collateral inventions; that is, diagnosis
and staging an aberrant program disease or for selecting optimal
treatment in clinical practice.
[0144] c) evaluating possible adverse effects of treatments for
aberrant program diseases on cultures of the three major
constitutively self-renewing tissues (bone marrow, gastrointestinal
epithelium, and skin). These cultures will also be used in some of
the reductions to practice involving therapeutic manipulations of
normal tissue. Culture conditions, Table IV.
[0145] d) The other cultures and heterotransplants to be used in
the reduction to practice.
[0146] 4) Discrimination of Normal vs Malignant Cells in a Mixed
Population.
[0147] Standard in situ hybridization procedures for detecting
chromosome and/or translocation specific changes will be utilized.
(Trask Trends in Genet. 7, 149,1991).
[0148] 5) Establish Assays for Scoring effects of Manipulating
Transcriptional Regulator Function or their Direct Modifiers on
Cellular Programming.
[0149] a) Aberrant Program Disease Tissue--
[0150] By definition the affected cells in these disorders express
abnormal patterns of gene expression that produce the
characteristic clinicopathologic features. Both of these can be
monitored using established molecular and cellular techniques. The
specific parameters to be assayed for each of the types of aberrant
program disease given as examples are shown in Table III.
[0151] b) Normal Tissue--
[0152] Reprogramming normal cell behavior where the relevant
programs are differentiation, proliferation and viability could
serve a variety of therapeutic uses. These would include but not be
limited to certain in vitro and systemic treatments: (1) expansion
of normal cell numbers in vitro prior to transplantation; (2)
promotion of the growth of gastrointestinal cells in the treatment
of peptic ulcers and inflammatory bowel disease; (3) liver
regeneration, for example, following partial destruction by a virus
or toxic chemicals; (4) expansion of one or more hematopoietic cell
lineages for a variety of clinical purposes including
reconstitution of immune function in immunodeficiencies,
counteracting the effects of agents toxic to bone marrow and in
fighting infection.
[0153] All of these changes in normal cellular programming can be
readily assessed using established techniques.
[0154] B) Reduction to Practice [0155] Step 1) Determine the subset
of transcriptional regulators, and their direct modifiers,
expressed by the aberrantly programmed tissue, the corresponding
normal tissue, and the constitutively self-renewing normal tissue.
Alternatively make a similar determination for any other normal
tissue that is to be therapeutically manipulated in accordance with
this invention. [0156] Step 2) Add or subtract expression of
transcriptional regulator(s) or their direct modifiers from cells
to be therapeutically reprogrammed and the appropriate control
tissue, as previously specified. [0157] a) Addition--Use expression
vector to insert expressible gene for a particular transcriptional
regulator or a direct modifier of a transcriptional regulator into
aberrantly programmed cells. The inserted gene will be one that is
expressed by the corresponding normal cells, but not by the
aberrantly programmed cells. [0158] b) Subtraction-- [0159] i) can
be achieved by the use of antisense oligonucleotides directed to
the RNA of a particular transcriptional regulator or direct
modulator or double-stranded oligonucleotide ligands for DNA
binding domain of one or more transcriptional regulators.
[0160] Using prototype antisense oligonucleotide(s) or
double-stranded oligonucleotides block function of specific
transcriptional regulator(s) in aberrantly programmed cells or
normal cells to be therapeutically manipulated through
reprogramming. Alternatively use an antisense oligonucleotide
directed to a direct modifier of a transcriptional regulator.
[0161] ii) Using expression vector carrying antisense DNA directed
to a particular transcriptional regulator or a direct modifier of a
transcriptional regulator, install the new gene in aberrantly
programmed cells. The therapeutic effect will be determined in
advance through the use of an antisense oligonucleotide. [0162]
Step 3) REPROGRAMMING TEST: [0163] Using the methods and procedures
described in the "Hardware for Reduction to Practice" and using the
information given in Tables III and IV, perform the following
functions. [0164] a) Utilize appropriate culture conditions for
normal cells to be therapeutically reprogrammed` or for AP disease,
the AP cells plus the corresponding normal cells and Constitutively
self-renewing normal tissues (gastrointestinal, bone marrow, skin);
[0165] b) For AP disease, assay one or more pathogenic features of
AP cells such as those shown in Table III and described in
references in Table XIII, according to established procedures;
[0166] c) Treat cultures with prototype agent with reprogramming
potential (as oligonucleotides to TR, as oligonucleotide ligands
for TR, or expression vectors). [0167] d) Score changes in
programming and choose those agents that are therapeutically
useful; for example: [0168] 1) cancer, myelodysplasia and
myeloproliferative syndrome and atherosclerosis--kill AP cells;
[0169] 2) AIDS, regenerate CD4.sup.+ lymphocytes; [0170] 3) Expand
normal hematopoietic stem cells for bone marrow transplant. [0171]
4) Alzheimer's cells: block apoptosis; [0172] 5) Rheumatoid
arthritis: block inflammatory responses, and block joint
destruction; [0173] 6) Schizophrenia: monitor in vitro predictors
of potential for altering nervous system functioning to improve
cognitive state of patients; [0174] Step 4) Test effect of addition
or subtraction of the function of particular transcriptional
regulators using the agents selected in an animal model system if
the therapeutic agents are for systemic use. (Table XIII)
[0175] Because of the need for a high degree of target homology
with the corresponding human transcriptional regulator or its
direct modulator the animals will of necessity nearly always be
non-human primates. Immunocompromised animals xenotransplanted with
human tissue are also of value for in vivo efficacy studies.
[0176] In the case of evaluating agents for the treatment of
aberrant program diseases the animal may either be afflicted with
the disease and both the efficacy of the treatment and the side
effect documented or the animal may be normal and only the side
effects tested. [0177] Step 5) Any undesirable side effects that
might be produced by the potential therapeutic agents can be
reduced or eliminated in several possible ways, all of which can be
implemented using existing technology.
[0178] a) Antisense Oligonucleotides
[0179] FIG. I demonstrates that there are cell type specific
differences in effects of particular antisense oligonucleotides
targeted to different sites on specific RNA transcripts on cell
behavior. Such differences can be used to select antisense
oligonucleotides that produce the desired therapeutic effects with
minimal undesirable side effects. These differences could be due to
several factors, including differences in the availability of
particular "hotspots" for antisense ODN binding between different
cell types/specimens, apta.meric or backbone effects.
[0180] b) Double-Stranded Oligonucleotide Ligands
[0181] Typically more than one transcriptional regulator can bind
to the same double-stranded DNA sequence, but with variable
affinities. It is, therefore, possible to change the competitive
inhibitor effect of such an agent relative to the potential set of
target transcriptional regulators by introducing base changes.
These can include mismatches. The melting temperature of the two
resulting strands, however, must be >40.degree. C. under
physiologic conditions. The effect of such changes, therefore, can
produce a more favorable therapeutic index.
[0182] c) Expression Vectors
[0183] The levels of expression and efficiency of gene transfer can
be readily adjusted on a tissue specific basis by changes in the
viral envelope and/or the promoter/enhancer combination used to
achieve gene expression.
[0184] The forgoing, then, can be reduced to a novel "Method of
Rational Discovery" for antisense oligonucleotides for the
treatment of Aberrant Programming disease comprising the following
steps: [0185] (i) select one or more transcriptional regulator gene
targets implicated in the regulation of cellular programming in the
Aberrant Programming Disease, these targets also being expressed by
the diseased cells to be reprogrammed; [0186] (ii) select one or
more prototype antisense oligonucleotides to target transcripts of
the selected transcriptional regulators, where the prototype
antisense oligonucleotides are chosen from among those targeting
gene "hotspots" defined by the Tertiary Selection Method, as
described hereinafter; [0187] (iii) evaluate the prototype
antisense oligonucleotides in the Reprogramming Test for their
capacity to therapeutically reprogram the Aberrant Programmed
disease cells while at the same time not adversely reprogramming
normal cells; [0188] (iv) combine the selected prototype antisense
oligonucleotide with another augmentation agent which is capable of
altering the pattern of transcriptional regulator expression
through a change in the expression of or physical state of the TR,
where the TR is expressed in the Aberrant Programmed cells, to
discover if the agent improves the ability of the selected
antisense oligonucleotide to therapeutically reprogram the Aberrant
Programmed cells; [0189] (v) design sequence- and length-variants
of the selected prototype antisense oligonucleotides which scored
well in the Reprogramming Test (i.e., which exhibited capacity to
reprogram Aberrant Programming cells); the variants should be
selected so that they bind to sites overlapping or contiguous to
the binding site of the corresponding prototype antisense
oligonucleotide, and have negligible capacity to form
self-complementary dimers or hairpin structures; [0190] (vi)
evaluate these variant ODNs alone, or with the agent selected in
step (v), in the Reprogramming Test for capacity to effectively
reprogram the Aberrant Programmed disease cells in the Aberrant
Programming Disease while not adversely reprogramming normal cells;
and then [0191] (vii) test the most active of the evaluated
antisense oligonucleotides, either alone or with the agent selected
in step (v), for therapeutic efficacy in an animal model system if
the therapeutic agents are for systemic use; [0192] (viii) evaluate
the data, and select the antisense ODN which exhibits the best
therapeutic index with the least toxicity, for potential use as a
therapeutic agent in clinical trials.
[0193] Demonstration of the Reduction to Practice with a P53
Target
[0194] Step 1
[0195] It is known that p53 is expressed by primary human leukemia
blast cells using the metabolic labeling technique (Smith, et al.,
J. Exp. Med. 164, 751, 1986.)
[0196] Step 2
[0197] A set of four different phosphorothioate antisense
oligonucleotides directed to p53 RNA were prepared using an Applied
Biosystems, Inc. (AB!) DNA synthesizer (Model 380B) according to
the manufacturer's protocols. An antisense oligonucleotide against
the HIV rev gene was used as a negative control. The sequences are
set forth in the Sequence Listing hereinafter as SEQ ID NOS: 1-4.
These were used to treat primary human leukemic blasts, normal
human bone marrow, normal human circulating T-lymphocytes, normal
adult human gastrointestinal epithelium, normal human fetal
gastrointestinal epithelium and Rhesus monkey T-lymphocytes.
Destruction of p53 RNA by the antisense p53 oligonucleotides was
documented using PCR and/or dot blotting.
[0198] The four p53 sequences are as follows:
TABLE-US-00001 SEQ ID NO. 1: 5'-AGTCTTGAGC ACATGGGAGG-3' C(1)p53
SEQ ID NO. 2: 5'.quadrature.ATCTGACTGC GGCTCCTCCA-3' A(1)p53 SEQ ID
NO. 3: 5'-GACAGCATCA AATCATCCAT-3' A(3)p53 SEQ ID NO. 4:
5'-CCCTGCTCCC CCCTGGCTCC-3' OL(1)p53
[0199] Step 3
[0200] The following effects of the antisense p53 oligonucleotides
on cellular programming were evident from the results found. [0201]
1) They can irreversibly block the proliferation of, block stem
cell self-renewal, or kill human cancer cells. This coupled with
the lack of toxic effects on normal tissue indicates these agents
can have a role in the treatment of cancer. (See Tables V-VIII).
[0202] 2) They promote the proliferation of gastrointestinal
epithelium, indicating a role in the treatment of peptic ulcer and
inflammatory bowel disease (FIG. I). The suppressive effect of
these agents on mature lymphocyte (Table IX) proliferative also
supports their role in diseases such as inflammatory bowel disease
that have an autoimmune component. [0203] 3) The data also
demonstrates that there are cell type specific differences in
responses to antisense oligonucleotides targeted to different sites
on RNA transcripts of the same gene (FIG. I). This provides a basis
for optimizing therapeutic effects and for minimizing undesirable
side effects. [0204] 4) These results support the general principle
that antisense oligonucleotides directed to a transcriptional
regulator can be used to expand particular normal adult or fetal
tissues in vitro that could then be used for various medical
purposes including transplantation (FIG. I). [0205] 5) The cell
type dependency of the effects of particular antisense
oligonucleotides directed to a transcriptional regulator support
the cellular program model in general and the aberrant program
model in particular.
[0206] Step 4
[0207] The ability of the antisense p53 oligonucleotides to
recognize the p53 RNA of Rhesus monkeys was demonstrated by showing
a similar inhibitory effect on mature T-cell proliferation for both
Rhesus and human cells (Table IX).
[0208] Two Rhesus monkeys weighing 8.9 kg and 6.8 kg were infused
with 52.5 mg and 75.8 mg of the OL(1)p53 antisense oligonucleotide
(SEQ ID NO:4) which was radiolabeled over four hours. In keeping
with rodent data, tissue distribution analysis showed substantial
oligonucleotide uptake compared to the levels needed to block p53
expression. Excretion studies demonstrated retention of the infused
agent for more than two weeks. During this time and subsequently,
the animals were extensively monitored for signs of toxicity and
none were seen. These and additional results were reported in:
Cornish K G, P Iversen, L Smith, M Arneson and E Bayever:
Cardiovascular effects of a phosphorothioate oligonucleotide with
sequence antisense to p53 in the conscious Rhesus monkey. Pharm.
Comm. 3(3): 239-247 (1993), which is expressly incorporated herein
by reference.
[0209] It was not possible to do in vivo efficacy studies with the
p53 antisense ODN in an animal model for two reasons: (1) there
were no readily-available animal model systems that utilized
freshly-obtained human leukemia cells as xenotransplants, including
no readily available primate leukemia animal models; (2) there is
at best only a modest effect of the p53 antisense ODN on
established in-vitro-adapted human leukemia cell lines, in contrast
to the significant p53 effects observed on primary (fresh) human
leukemia cells in vitro.
[0210] These latter data were reported in: Bayever E, I Smith et
al: Selective Cytotoxicity to human leukemic myeloblasts produced
by oligodeoxyribonucleotide phosphorothioates complementary to p53
nucleotide sequences. Leukemia and Lymphoma 12: 223-231 (1994),
which is expressly incorporated herein by reference.
[0211] Step 5
[0212] Since no unacceptable side effects, were produced in the
monkeys, it has not been necessary to modify the antisense
oligonucleotides.
[0213] Step 6
[0214] The non-toxicity of the antisense p53 phosphorothioate
oligonucleotide OL(1)p53 was confirmed after systemic
administration into humans in a Phase I clinical trial designed
with the review and approval of the U.S. Food and Drug
Administration. The initiation of this world's-first clinical trial
of systemically-administered antisense ODN was reported by: Bayever
E, I Smith et al: Systemic human antisense therapy begins.
Antisense Res. Develop. 2: 109-110 (1992), which is expressly
incorporated herein by reference.
[0215] In addition to the treatment of patients with cancer, the
p53 antisense ODNs may be used to treat certain other diseases or
medical conditions that include, but are not limited to, the
following: AIDS; Atherosclerosis/restenosis; Alzheimer's disease;
autoimmune diseases including, for example, multiple sclerosis,
rheumatoid arthritis, and systemic lupus erythematosus, in
accordance with the AP model. The results may be substantially
improved by adding either a free-radical generator or an
anti-oxidant as augmentation agents, as previously described. In
addition, free-radicals may cause an alteration in the conformation
of p53, with the result that p53, so altered, can function like an
essentially different TR (Maxwell and Roth: Crit. Rev. Oncog. 5:
23, 1994; Hainaut and Milner: Cancer Res. 53: 4469, 1993). In this
instance, a free-radical generator or redox modifier used to
produce this effect would be a direct modifier of p53. These p53
antisense ODNs, used alone or with an augmentation agent, may also
be used to block apoptosis in damaged normal tissue, for example,
as a result of blood vessel occlusion and/or reperfusion injury, or
ionizing radiation damage to the skin.
Novel Computer-Based Method for Selecting Target Sites for Highlv
Active Antisense ODNs
[0216] The four p53 antisense ODNs (SEQ ID NOS.1-4) mentioned above
were all designed using the method described above under the title
"Design of `Test` Antisense Oligonucleotides", along with certain
other considerations. The general area to which A(1)p53 (SEQ ID
NO.2) and A(3)p53 (SEQ ID NO.3) were targeted were, or were thought
to be, the translational start site for regular or for
alternatively spliced p53, respectively. The practice of targeting
antisense ODNs to areas of RNA transcripts thought to have a
regulatory role (beyond coding a protein) is well established in
the literature. C(1)p53 (SEQ ID NO.1) was targeted to a general
area of approximately 150 base pairs (bp) in length that was
selected by a computer program designed to pick out potential
functional areas in nucleotide sequences. The basis of this latter
program is Chaos Theory. This C(1)p53 oligonucleotide binds to one
strand of the DNA duplex in the p53 gene that includes a
transcriptional regultor binding site.
[0217] In contrast, the target area for OL(1)p53 (SEQ ID NO. 4) was
selected from a group of possible sites within the p53 transcripts,
without consideration of whether or not a biologically significant
function was associated with any of these regions of the p53
transcript. Instead, the inventor simply further restricted the
criteria presented under "Design of `Test` Antisense
Oligonucleotides". The restriction was to obtain an ODN with the
highest possible Tm value (equivalent to lowest .DELTA.G.degree.
value) without compromising the other selection criteria. In
subsequent discussions which follow, the originally-described
selection method and criteria under the title "Design of `Test`
Antisense Oligonucleotides" will be termed "the primary selection
method", while the method with the more restricted criteria to be
described next will be termed "the secondary selection method."
Again, it was this secondary selection method that was utilized to
design OL(1)p53.
[0218] The "OLIGO" computer program (Version 3.4), created by Dr.
Wojciech Rychlik (Rychlik and Rhoads, Nucleic Acids Res. 17: 8543,
1989; copyright 1989), was utilized in all of the hotspot
selections included herein. This is the same "OLIGO" computer
program originally used by the present inventor to examine the p53
cDNA sequence. The p53 sequences were examined using
computer-generated probe of all possible antisense ODNs containing
20 nucleotide bases (i.e, a "20-mer") that could bind with the p53
cDNA. The inventor instructed the computer program to calculate the
binding affinities for these antisense ODNs under the conditions of
138 mM salt and an ODN concentration of 250 nM. Next, the inventor
asked the computer to list all of the possible 20-mer antisense
ODNs that would have negligible self complementarity, which is
defined as not having hairpin structures within the 20-mer (neither
negative nor positive .DELTA.G.degree. values), and no self
dimerization with more than two continuous base matches. Using the
"Option `Z`" function in the "OLIGO" program, the program was then
asked to print out a list of the 5-prime terminal base positions of
the sites on the p53 sense strand to which the ODNs would bind, the
extreme 5'-nucleotide of the entire gene/cDNA sequence as it came
from Genbank or the literature being designated nucleotide position
number 1. These sites were entered into the primary array of the
program's memory. Next, using the memory for the secondary array,
all the 20-mer ODNs with a higher binding affinity for their target
than a certain value (again, selected by the inventor) were
entered. In standard physical chemistry terms, the more negative
the value of the .DELTA.G.degree. indicated for a particular 20-mer
ODN, the higher the binding affinity to its corresponding target
sequence. Therefore, beginning with a .DELTA.G.degree. value of
-43.0 kcal per mole (kcal/mol), and then making the
.DELTA.G.degree. value more negative in steps of -2 kcal/mol (i.e.,
for example: -45 kcal/mol, -47 kcal/mol, -49 kcal/mol), the
inventor directed the computer to compare primary and secondary
arrays and to print out the start sites for all the 20-mers meeting
the criteria used in both arrays. The inventor initially decided
intuitively that he would assume that the more negative the
kcal/mol value for a given 20-mer ODN (i.e., the more tightly it
bound to its corresponding target sequence), the higher would be
the biological activity of the corresponding antisense ODN. This
assumption could then be tested.
[0219] Next, the 20-mers meeting the criteria in both arrays were
examined to further determine the suitability of these antisense
ODNs. Any ODN having four (4) or more guanine bases in a row were
eliminated. Originally, this was done because ODNs having four or
more guanine bases in a row were thought to be more difficult to
synthesize. More recently, however, it has been determined that
four guanine bases in a row may cause the formation of a
"G-quartet" which, through formation of intermolecular complexes,
may cause production of significant adverse, sequence-specific,
non-antisense effects on cells. Further, any ODN having .gtoreq.85%
match (i.e., 17 matches out of 20 bases) with any other known human
gene sequence whose blockade by ODN would negatively influence the
therapeutic utility of the antisense ODN in question were also
eliminated. OL(1)p53 constituted the single 20-mer antisense ODN
that met all these individual criterion and which had the most
negative .DELTA.G.degree. value (-47.5 kcal/mol) when these
criteria were applied to the p53 gene/cDNA sequence (Lamb and
Crawford, Mol. Cell. Biol. 6: 1379, 1986).
[0220] Finally, the method for selecting antisense ODNs underwent a
final stage of development to yield the "tertiary selection
method." It is this tertiary selection method that was used to
select the many additional "hotspots" and prototype antisense ODNs
that are disclosed herein. The tertiary selection method consists
of the following steps: (1) the secondary selection method is
utilized to find the binding sites for the ODNs that have a high
probability of being the most active, with one modification: i.e.,
22-mer lengths of ODN rather than 20-mer was used for probing the
target gene sequences. This was accompanied by a decrease in the
required -kcal/mol value for binding affinity to -45.0 kcal/mol for
primary hotspots. Identification of a continuous and contiguous run
of 5'-end starting positions on the computer print out was
considered to reveal the core of a "hotspot" for antisense ODN
targeting. Using the OLIGO program in the manual mode, examining
contiguous overlapping binding sites on either side of the "core
hotspot" for self-complementarity, potential antisense ODNs varying
in size from 16-27 nucleotides in length were identified. Of these,
ODNs contiguous with the core hotspot were rejected if they were
shown to contain sequences which would generate a stable self
hairpin structure with a negative .DELTA.G.degree. value
(.ltoreq.1.0 kcal/mol) as calculated by the OLIGO program, and/or
if the ODN was estimated to have significant self dimerization at
37.degree. C. (Self dimerization is most readily evaluated by using
the upgraded OLIGO program in Version 4.0). Once all of the
acceptable contiguous ODNs are determined by this procedure, then
the "hotspot" has been defined. As a rule of thumb, the present
inventor has determined that if 10 such hotspots can be identified
for a particular gene target, then one should find that at least
three (3) of these sites will yield very highly active antisense
ODNs, and this probability is inversely correlated with the
calculated -kcal/mol binding affinity for the antisense ODN in
question. Most of the prototype antisense ODNs designed by the
inventor for the evaluation of these hotspots are 22-mers that
optimize minimal self interaction with a large -kcal/mol value. In
some instances, these parameters can best be optimized by going to
a longer or a shorter antisense ODN. Multiple active antisense ODNs
within any given "hotspot" are needed for several reasons,
including the possibility that some ODNs will have unexpected but
biologically important aptameric (or aptameric-like) effects which
may either increase, decrease, or have no effect on the therapeutic
endpoint to which the antisense ODN in question is directed.
[0221] Not all gene targets will yield 10 or more hotspots
according to the criteria just presented. In these cases,
additional ("alternative" or "secondary") hotspots can be found by
manually examining each of the areas of the gene that yield
antisense ODNs with the highest binding affinity for their targets
as measured by the calculation of the -kcal/mol value (this, too,
is to be considered part of the tertiary selection procedure). Each
of the potential antisense ODNs in these areas are examined for
self complementarity using the aforementioned relaxed standards of
no hairpin structures with self-binding affinities of .ltoreq.1
kcal/mol or significant self dimerization at 37.degree. C. as
judged by the OLIGO program, version 4.0. The set of contiguous
overlapping antisense ODNs uncovered in this way then define a
"hotspot" within the transcript (sense) sequence. Again, the
prototype antisense ODNs selected for each of these hotspots are
optimized for two (2) parameters which are: minimal self
interaction, and a more highly negative kcal/mol value for binding
to the target site in the gene transcript.
[0222] Using this tertiary selection method, the most highly active
antisense ODNs for blocking the expression of particular genes can
be discovered. The steps are: (1) design antisense ODNs according
to the tertiary selection method; (2) synthesize and test the
prototype antisense ODNs for each hotspot; (3) pick the most active
prototype antisense ODNs; and (4) design and test new; variants
derived from these prototypes. The variants should be designed to
bind further upstream or downstream of the prototype, in short
steps of two or three bases, and variants of shorter or longer
lengths varying by two bases at a time should be tested. In an
initial test to determine if shortening the ODN improves its
antisense activity, or to determine if shifting the ODN two bases
downstream improves antisense activity, for example, the subsequent
variants that continue in the same direction of change should be
evaluated. All of these variants, however, will fit into the range
of possible antisense ODNs defined by the corresponding
hotspot.
[0223] Following standard drug development procedures, these
antisense ODNs will be first tested in vitro for activity whenever
possible, and for toxicity. Next, in vivo studies of potential
efficacy (where there is an appropriate animal model) (Table XIII)
and pharmacology/toxicology analyses will be carried out. Antisense
ODNs successfully making it to this stage of development will be
subject also to any additional studies that might be required by
the U.S. Food and Drug Administration for achieving approval of an
Investigational New Drug (I.N.D.) application.
Testing of the Novel Computer-Based Method for Selecting Target
Sites for Highly Active Antisense ODNs.
Example 1
[0224] In vitro comparisons of the relative activity levels of the
four p53 antisense ODNs (SEQ ID NOS. 1-4) showed that OL(1)p53 (SEQ
ID NO.4) was consistently the most active in terms of producing
anti-tumor effects. Freshly obtained acute myeloid leukemia (AML)
blast cells, as well as ovarian, colon, lung and brain cancer
specimens, were tested. In nearly every experiment in which the
four p53 antisense ODNs were evaluated simultaneously on such tumor
specimens, the OL(1)p53 antisense ODN exhibited the most potent
activity. Activity was determined as a reduction in viable cell
counts and by reduced capacity of treated cells to grow in methyl
cellulose as colonies in colony forming assays, compared to cells
treated with control ODNs. The results with fresh AML blast cells
have been published (Bayever et al., Leukemia and Lymphoma 12:
223-231,1994). These data were consistent with the notion that the
secondary selection method would yield ODN sequences with a greater
probability of being the most highly active than selecting ODNs by
applying the primary selection method to regions of transcripts
thought to have an important regulatory function beyond encoding
for a particular protein (such as the 5'-cap site, the AUG start
site, or splice junctions).
Example 2
[0225] Having discovered a method for selecting highly active
antisense ODNs, the present inventor designed a series of antisense
ODNs against two other gene targets in order to further test and
confirm the value of the selection method herein disclosed. The
MDR1 gene that encodes P-glycoprotein, and the gene encoding the
multidrug resistance-associated protein (MRP) were selected for
these studies. The protein products of these genes constitute
molecular pumps that have been implicated in the production of the
multidrug resistance phenotype in cancer cells. Hence, antisense
ODNs with the capacity to block the expression of these genes would
be expected to increase the sensitivity of treated multidrug
resistant tumor cells to chemotherapeutic agents such as
vincristine, doxorubicin and VP-16. The entire MDR1 gene has been
cloned and sequenced, while only the MRP cDNA has been cloned and
sequenced. consequently, there is a much larger number of
functional sites that can be targeted on the MDR1 transcript.
[0226] The primary selection method was used to design antisense
ODNs to target selected regions of the MDR1 transcript; the
functional significance of the selected regions had been described
(Chin et al., Mol. Cell. Biol. 9: 3808, 1989; Chen et al., J. Biol.
Chem. 265: 506, 1990; Kohno et al., J. Biol. Chem. 265: 19690,
1990). These regions included the 5'-cap site, the 5'-untranslated
leader, splice junctions and the translational start site
(containing the AUG codon). These types of functional sites have
frequently been identified as suitable targets by other
investigators. In addition, other antisense ODNs were selected
using the secondary selection method described herein, and also by
the method for finding alternative or secondary hotspots described
herein; these ODNs will be discussed hereinafter.
[0227] To determine relative activity among the various MDR1
antisense ODNs, they were incubated at several different ODN
concentrations in cell culture medium with the human multiple
myeloma cell lines 8226/Dox4 or 8226/Dox6 (gifts of Dr. William
Dalton, University of Arizona Cancer Center, Tucson) (Dalton et
al., Cancer Res. 46: 5125, 1986), Specifically, the human myeloma
cells were incubated with the antisense ODNs for four (4) days, in
a humidified, 5% CO.sub.2 atmosphere containing a reduced oxygen
level of 5%, maintained by purging the culture incubator with
nitrogen gas in place of ambient room air. Addition of an
antioxidant to the cultures during exposure of the cells to the ODN
further reduces the effects of the free radicals induced by the
ODNs on cells (this effect can include increased expression of
MDR1). The cell cultures were then incubated with a
chemotherapeutic anti-cancer agent overnight (15-18 hrs), after
which the cell cultures were washed repeatedly and then seeded into
96-well microtest plates to permit quantitative analysis of their
growth. Tritiated-thymidine was added to these cultures for 18-hrs
prior to termination of the assays on day 4 following exposure to
the anticancer agents. Each test combination of cells, antisense
ODN and anti-cancer agent was run in quadruplicate replicates.
[0228] The data for MDR1 antisense ODNs are shown in Table XIV, and
show that the most active ODNs were those which had been selected
using the secondary selection method. In this Table, the most
active MDR.sup.1 antisense ODN is OL(1C)mdr. Re-iterating, the
criteria of the secondary selection method (1) permit no self
interactions in the ODN (such as hairpin structures) and no more
than two contiguous base pair matches in selfdimers; and (2)
maximize the negativity of the calculated -kcal/mol value to ensure
the most avid binding of the antisense ODN to its corresponding
target sequence, without compromising the other criteria provided.
These are the same parameters that were used to select the OL(1)p53
antisense ODN (SEQ ID NO. 4).
[0229] Next, variants of OL(1)mdr and OL(12)mdr--the most active of
the antisense ODNs targeting MDR1 transcripts--were tested. The
results of one such experiment are shown in Table XV and in FIG. 2.
Among the several OL(1)mdr variants, OL(1C)mdr and OL(1Q)mdr were
found to be two of the most active in terms of producing the
greatest "fold-increase" in drug sensitivity, when calculated using
as the "treatment control" cells treated only with culture media
("Method 2", last two columns, Table XV). If, on the other hand,
the "treatment control" cells used in the calculation were cells
treated with the corresponding ODN ("Method 1", middle two columns,
Table XV), then the OL(1C)mdr variant is calculated to be less
active than certain other MDR-ODNs, including OL(12)mdr, OL(12A)mdr
and OL(1B)mdr. The OL(1)mdr ODN and its variants were selected
using the secondary selection method, while the OL(12)mdr ODN and
its variants were selected using the method for finding
alternative, or secondary, "hotspots".
[0230] The very high potency of the OL(1C)mdr prototype MDR-ODN on
8226/Dox4 cells, however, appears possibly to be due to a combined
antisense effect on MDR1 transcripts and to an aptameric-like
effect that also can cause suppression of MDR1 expression in these
highly drug resistant cells. This aptameric-like effect is also
associated with a partial inhibition of proliferation of this
highly P-glycoprotein expressing cell line (FIG. 2). It is this
effect that explains the differences seen with the "Media control"
versus "Corresponding-ODN Control" in calculations of
fold-increases in drug sensitivity of ODN-treated cells (Table
XVI). A likely explanation for this aptameric-like effect is that
the ODNs showing this effect are binding to cellular proteins
involved in the regulation of MDR.sup.1 expression, such as, for
example, protein kinase C and/or protein kinase A.
[0231] The enhanced reduction in P-glycoprotein expression due to
the combined antisense and aptameric-like effects likely causes the
associated reduction in cell proliferation. It has been previously
shown that treating P-glycoprotein-expressing cells with monoclonal
antibodies capable of binding to and blocking P-glycoprotein
function leads to a substantial reduction in the proliferation of
the treated cells. The fact that OL(1C)mdr is more active than
OL(12)mdr and it variants on cell lines that express less
P-glycoprotein (such as 8226/Dox6 and CEM/VLB cells) is consistent
with this hypothesis. It is also possible that the aptameric-like
effect that leads to a reduction in P-glycoprotein expression also
inhibits a cellular pathway involved in proliferation, such as the
protein kinase C pathway. The lack of the aptameric-like effect on
less drug-resistant cells, or on drug sensitive cells, could
reflect the fact that such cells do not have the up-regulated PKC
pathway commonly seen in highly drug-resistant cell lines (Table
XVII). The relative ranking of the most potent MDR-ODNs was
confirmed using less drug-resistant P-glycoprotein-expressing CEM
human leukemia cells. The MDR-ODNs showing the aptameric-like
effect on 8226/Dox4 cells did not produce this effect on these
P-glycoprotein-expressing CEM cells. The term "aptameric-like" is
used because the effect seen is seen with some sequences but not
others (aptameric); aptameric-like effects are substantially more
common among ODN sequences than are the usual aptameric effects.
OL(1C)mdr did not produce drug sensitization of drug sensitive
lines, and has no significant effect on the proliferation or
differentiation of normal human bone marrow progenitor cells in
standard in vitro assays. In general, this situation in which the
antisense and an aptameric-like effect of an ODN both act to reduce
expression of the same gene (e.g., MDR.sup.1) can be expected to be
a very uncommon finding among antisense ODN therapeutics directed
to other gene targets. Hence, despite the calculated (Method 2)
superiority of OL(12) and its variants over OL(1C)mdr, it appears
reasonable to conclude that OL(1C)mdr is the more active antisense
ODN.
Example 3
[0232] As was done for the MDR1 gene transcripts in Example 2, the
primary selection method was also used to design antisense ODNs to
target selected regions within the MRP cDNA gene transcript, based
on what is known in the literature about the functions of the
various regions of the transcript (Cole et al., Science 258: 1650,
1992). Functional sites targeted included the 5'-cap site, the
5'-untranslateu leader and the translational start site. In
addition, other antisense ODNs Io were selected using the secondary
selection method described above.
[0233] Relative antisense ODN activity among the various MRP
antisense ODNs was determined in exactly the same manner as was
described in Example 2 above for analysis of the MDR1 antisense
ODNs, except that the MRP antisense ODNs were tested on the
MRP-expressing, multidrug-resistant A427 human lung cancer cell
line (Giard et al., J. Natl. Cancer Inst. 51: 1417, 1973), Table
XVI summarizes the prototype MRP antisense ODNs tested in these
studies, and shows their relative ranking in terms of in vitro
drug-sensitizing activity. OL(8)MRP, which was selected using the
secondary selection method, was found to be the most effective
among the various MRP antisense ODNs tested.
[0234] Based on the results of our studies of variants of the most
active antisense ODNs selected by the Secondary Selection Method,
the latter selection method was modified to become the Tertiary
Selection Method already described.
[0235] These data support the notion that the method of selecting
"hotspots" for targeting antisense ODNs and for determining the
best antisense ODNs for any given hotspot will yield highly active
antisense ODNs for any gene of choice. These, in turn, can be used
for the therapeutic purposes disclosed herein.
Evaluation of the Usefulness of the Aberrant Programming Model with
the Novel Method Presented herein for Selecting Highly Active
Antisense ODNs for the Discovery of Antisense ODNs for Therapeutic
Use in the Aberrant Programming Diseases.
[0236] The most important feature of the invention disclosed herein
is the fact that it makes possible a straight-forward and
comparatively simple approach to the discovery of novel
therapeutics for the treatment of several major diseases that
historically have been intractable with respect to the development
of new therapies, such as, for example, cancer, atherosclerosis,
AIDS and Alzheimer's Disease. Specifically, the Aberrant
Programming disease model combined with the method disclosed herein
for selecting highly active antisense ODNS makes the discovery of
new therapeutic agents for these diseases a comparatively small
scale screening process. In contrast, for example, conventional
drug development for cancer has yielded no major new breakthrough
drugs for more than a decade, despite the huge expense and the
screening of hundreds of thousands of compounds.
[0237] To demonstrate a reduction of this invention to practice,
the inventor decided to select a very small number of gene targets
from those predicted by the Aberrant Programming model to be of
therapeutic use and to design and test antisense ODNs directed to
these targets on normal and cancer cells. The first gene target
selected was junD (Hirai et al., EMBO J. 8: 1433, 1989; Ryder at
al., Proc. Natl. Acad. Sci. USA 86: 1500, 1989), The reasons were
the following: (1) junD is a TR involved in the control of cellular
programming; (2) it is expressed by many types of both normal and
malignant cells; and (3) the expression levels of junD and its
physical state appear to be the same in both normal and malignant
cells. Therefore, junD appears to be neither an oncogene nor an
anti-oncogene in actual tumors seen in patients (certain junD
mutants may, however, function as an oncogene in some experimental
situations). Hence, if treatment of normal and malignant cells with
antisense ODNs to junD leads to adverse effects on malignant but
not on normal cells, then such data would constitute prima facia
evidence in support of the Aberrant Programming model and its value
for the development of novel therapeutics.
Example 4
[0238] Two junD antisense ODNs were tested in vitro: H(1)junD (SEQ
ID NO.5) and H(2)junD (SEQ ID NO.6):
TABLE-US-00002 SEQ ID NO. 5: H(1)junD GTCGGCGTGG TGGTGA SEQ ID NO.
6: H(2)junD GCTCGTCGGC GTGGTGGTGA
[0239] Control ODNs included those which had the same nucleotide
base sequence but in the reverse order, as well as ODNs which were
directed to irrelevant genes. These ODNs were designed using the
criteria listed for the primary selection method, with the added
restriction that (a) only hotspots capable of giving rise to
antisense ODNs with no hairpins and no more than two contiguous
base matches in a row for self-dimers were allowed; and (b) the
melting temperature (T.sub.m) had to be .gtoreq.40.degree. C. for a
20-mer (corresponding to a negative kcal/mol value of .ltoreq.-33.8
kcal/mol or even more negative. Under these conditions, 26
acceptable hotspots were identified. The inventor decided to test
initially antisense ODNs with binding affinities in the mid-range
between the maximal value of -33.8 and the preferred values of
.ltoreq.-46.0 kcal/mol. The hotspot with the 5'-terminal nucleotide
of 515 was chosen. The corresponding 20-mer prototype antisense ODN
[=H(2)junD] had a binding affinity of -43.0 kcal/mol. There were 11
acceptable hot spots in the junD sequence that would yield
prototype antisense ODNs with this or even greater binding
affinity. Based on subsequent work by the inventor, the selection
criteria used for H(1)junD and H(2)junD ODN are expected to yield
good but not necessarily exceptional antisense ODNs.
[0240] In the initial tests, the H(1)junD oligo or a control ODN
were incubated with AML blast cells for seven days. The blast cells
were freshly obtained from patients with AML and were cultured and
evaluated for ODN effects as described by Bayever et al. (Leukemia
and Lymphoma 12: 223-231, 1994). In brief, after the incubation
with ODNs, the blast cells were then washed, and adjusted to 105
cells per ml and continued in culture for several more days. Viable
cell counts were performed every 2-3 days on suspension cultures
and oligo-treated blast cells were plated in the "blast colony
assay" to determine the effects of the ODNs on the leukemic stem
cells. A representative example of the suspension culture data is
shown in FIG. 3. H(1)junD is consistently found, as in the example
shown, to have as great or a greater antileukemic effect than
OL(1)p53. Similarly, H(1)junD has been found to have anticancer
effects on solid tumors such as, for example, ovarian
carcinoma.
[0241] To test the effects of the H(1)junD ODN on normal cells,
normal human bone marrow cells were incubated with from 10 nM to 10
.mu.M of the H(1)junD ODN for 7 days. The bone marrow cells were
cultured and evaluated as described in Bayever et al. (op.cit.). In
brief, viable cell counts were performed every two days following
ODN treatment and then the cells were plated in mixed colony assays
to determine what effects (if any) the ODNs would have on the
proliferation and differentiation of various types of hematopoietic
colony forming units. H(1)junD was not found to have any impact on
normal cell growth, viability or differentiation.
[0242] When the H(1)junD and H(2)junD ODNs were tested on malignant
cell lines, they were found to have less of a cytotoxic or
anticancer growth-inhibitory effect than they had on
freshly-obtained cancer specimens. However, the inventor discovered
that these antisense ODNs could be used to dramatically sensitize
various types of multi drug-resistant cancer cells to anti-cancer
chemotherapeutic agents. Remarkably, these sensitizing effects were
operative on cancer cells that have differing mechanisms for their
multidrug resistance. Table XVII shows that H(1)junD or H(2)junD
can be used to sensitize P-glycoprotein-expressing drug-resistant
8226/Dox cell to vincristine, while H(1)junD also can sensitize
DU-145 prostate cancer cells that express MRP and not
P-glycoprotein. These finding support the conclusion that
suppressing the expression of junD, such as by treatment with
antisense ODNs, can be used to reverse multidrug resistance
resulting from any of a variety of mechanisms, including those that
require expression of molecular transporters such as P-glycoprotein
and MRP, and perhaps also the less-well characterized drug
resistance mechanisms such as those producing resistance to
cisplatin and related compounds. It has also been found that jund
is up-regulated in cancer cells following treatment with the
chemotherapeutic agent Ara-C (Kharbanda et al., Biochem. Pharm. 45:
2055, 1993). Hence, increased junD expression could be part of a
cellular response mechanism that tends to block the toxic effects
of chemotherapeutic agents on cancer cells.
[0243] In contrast to the effects on multidrug resistant cancer
cell lines, the H(1)junD ODN had minimal sensitizing potential when
used to treat drug-sensitive (parent) cancer cell lines (Table
XVII). This difference in the responsiveness of the drug-resistant
versus drug-sensitive cancer cells to H(1)junD is consistent with
the prediction made by the Aberrant Programming model that the
pattern of transcriptional regulator expression associated with
different states of the Aberrant Programming diseases will vary in
such a way that the effects of modulating these patterns with any
particular modulator will produce somewhat different effects,
depending on the specific character of the Aberrant Programming
disease cells. Hence, the need for the type of novel diagnostic and
prognostic testing disclosed herein. The junD antisense ODN data
taken as a whole also illustrate that junD is involved in an
abnormal combinatorial regulation system that maintains both (a)
the viability of cancer cells, and (b) the drug resistance
phenotype.
[0244] In conclusion, the Aberrant Programming model coupled with
the antisense ODN selection method disclosed, can readily be used
to identify novel antisense ODNs with therapeutic potential for
treating Aberrant Programming Diseases.
Selection of Hotspots and Prototype Antisense ODNs for Transcripts
of Genes of Therapeutic Importance
[0245] The tertiary selection method was applied to more than 200
genes and the resulting hotspots and prototype antisense ODNs are
disclosed herein, in Tables XVIII and following. The selection of
this large number of genes is based on the Aberrant Programming
model and on the scientific literature concerning the
probable/possible involvement of particular molecules or types of
molecules in particular disease processes. The details of this
extensive literature is readily available to anyone through
computer searching of a variety of scientific and medical
databases, using "keyword" searches. The list of genes for which
hotspots have been identified herein should be expanded in the
future. Certain genes (human) that are clearly of potentially great
therapeutic importance have been identified, but either the gene
sequences have not been made available or only partial sequences
are available. The genes in the former category include telomerase
which is implicated in the promotion of cancer, and several of the
homeobox genes (including Hox A1, Hox A10, Hox B6, Hox C6, Hox 2.4,
Hox 2.6 and Hox 5.4) which fall into both sequence categories. The
homeobox genes are apt to be important for more than one of the
Aberrant Programming diseases, and some are likely to be regulators
of telomerase expression, for example. TR that are required for
telomerase expression are potential targets for antisense
therapeutics for the treatment of cancer. Once the MRP gene is
completely sequenced, it is likely that the new sequence data will
allow for more useful antisense ODNs to be designed for therapeutic
inhibition of the gene. Other genes that are to be included in the
list of possible therapeutic targets, for which sequences are not
yet available, are mdm-2, LBP-1 a,b,c and d, Rp-8, fos-B, ATBF1,
hu-cut, maf, ERP, ELAM-1, ERM, AP-12, SAP-1, OxyR (human
counterpart), HES-1, NM23-H1, LR1, and E2F1. Application of the
tertiary selection method disclosed herein can and will be readily
applied to these gene sequences as they become available. The
resultant hotspots and the associated antisense ODNs will be
utilized in the practice of this invention.
Treatment of Atherosclerosis using Antisense ODNs Directed to
Transcriptional Regulators or their Direct Modulators
[0246] The present inventor, to the best of his knowledge, was the
first to describe the c-myc gene (and perhaps the c-myb gene, as
well) as being potentially useful gene targets for the treatment of
atherosclerosis, since these gene targets encode transcriptional
regulators known to be involved in the control of cellular
programming. As previously reviewed herein, antisense ODNs directed
to c-myc or c-myb have now been shown to have in vivo activity for
inhibiting the proliferation of smooth muscle cells (SMC) following
balloon angioplasty in atherosclerosis. In these publications, only
a single c-myc antisense ODN, however, directed to the transcript
of the human c-myc gene, has been tested. Analysis of this 15-mer
c-myc antisense ODN by the criteria disclosed in the present
invention indicate that this 15-mer is a very sub-optimal antisense
ODN for the purpose of reducing c-myc expression. Specifically, it
has four guanines in a row which may cause the formation of a
"G-quartet" (as does the c-myb antisense ODN used to treat SMC)
which is apt to inhibit the proliferation of some normal cell types
independently of any direct effect on c-myc expression. In fact, it
has been shown in the inventor's laboratory that an oligonucleotide
comprising the reverse sequence of this 15-mer c-myc antisense ODN
inhibited the proliferation of human lung cancer cell lines nearly
as well as did the c-myc antisense ODN (unpublished data).
Furthermore, the published 15-mer c-myc antisense ODN has a greater
degree of self dimerization than is acceptable according to the
criteria herein described in the present application, since it has
six (6) contiguous base matches; similarly, it has a much lower
binding affinity for its target than is desirable even in a 15-mer
(only being -30.9 kcal/mol). The application of the tertiary
selection method described herein will provide much improved c-myc
antisense ODNs, as well as highly active c-myb antisense ODNs
useful for modulating expression of the human c-myb gene, although
the c-myb sequence has not proven to be a particularly good
target.
[0247] It also is likely that the Reprogramming Test disclosed
herein could be used to find genes that would be more suitable
targets for antisense ODNs for the treatment of
atherosclerosis/restenosis than are the c-myc and c-myb genes. This
assumes, however, that the SMC involved in restenosis .are
atherosclerotic and not normal. If this is the case, then
restenosis is just a particular form of atherosclerosis. Indeed, it
could be argued that the published c-myc and c-myb antisense ODNs
that have been used in the restenosis system work simply by
inhibiting cell proliferation, and that many other antisense ODNs
directed to other gene targets involved in the control of cell
proliferation would work equally well. Such a broad effect based
simply on inhibiting a cellular program (proliferation) from
proceeding is very different from the cellular reprogramming effect
which is the basis of the AP model approach. With respect to the
strategy of simply inhibiting the proliferation of SMC following
angioplasty, it would be preferable to use gene targets for
antisense ODN inhibition that are more restricted to SMC than are
c-myb or c-myc; such as, for example, NF-IL6. Antisense ODNs that
can be effectively used to block cell proliferation have been
disclosed herein.
[0248] According to the Aberrant Programming model, atherosclerotic
SMC will express a different pattern of TR than is to be found in
normal SMC. Therefore, finding the most appropriate TR target
gene(s) requires a comparison of the effects of potential
therapeutically-useful antisense ODNs on atherosclerotic SMC, along
with normal cells (normal SMC, and normal bone marrow if possible).
It would be preferable for the treatment of atherosclerosis
generally to identify an antisense ODN capable of selectively
inducing programmed cell death (or "apoptosis") in the
atherosclerotic SMC while sparing normal cells. This could be
achieved by the use of antisense ODNs of the type disclosed herein,
used either alone or in combination with another augmentation agent
also capable of modulating TR expression in atherosclerotic SMC,
such as, for example, a growth factor. Miano et al.(Amer. J.
Pathol. 142: 715, 1993) have observed that the TR c-fos, fos-B,
fra-1, c-jun, jun-B and junD are expressed by SMC. Each of these,
along with c-myc and c-myb, should be evaluated in the
Reprogramming Test applied to atherosclerosis. The p53 gene is also
expressed in atherosclerotic SMC and should be tested as a target
gene along with genes regulated by p53, as well as those which, in
turn, regulate p53 expression. These genes, along with the genes
previously mentioned that are implicated in controlling changes in
cellular programming in response to oxidative damage, should be
evaluated as potential therapeutic targets for inhibition by
antisense ODN. Atherosclerosis has been postulated by some to be an
inflammatory disease linked to an abnormality in oxidationmediated
signals in the vasculature (Offermann et al., Heart Dis. Strokes 3:
52, 1994). Any other TR (or the direct modulators of the TR)
involved in the control of cellular reprogramming that can be shown
to be expressed in atherosclerotic SMC should also be evaluated in
the Reprogramming Test. The preferred methods for screening for the
expression of TR are provided herein. In addition, antisense ODNs
directed to genes involved in programmed cell death (Le.,
apoptosis) should be evaluated in the Reprogramming Test using
atherosclerotic SMC. Genes such as E2F-1 and cyclin-A or cyclin-D 1
that have been implicated in the expression of c-myc also should be
given priority testing (Oswald et al., Oncogene 9: 2029, 1994). A
listing of some of the in vitro and in vivo assays that should be
of use for developing antisense ODNs for the treatment of AP
diseases (including atherosclerosis) are provided in Table
XIII.
[0249] The antisense oligonucleotide selected for practice of the
invention may be any of the types described by Stein and Cohen,
Cancer Research 48:2569-2668 (1988), and including without
limitation, unmodified oligodeoxynucleotides, ethyl- or
methylphosphonate modified oligodeoxynucleotides, phosphorothioate
modified oligonucleotides, dithioates, as well as other
oligonucleotide analogs, including those incorporating ribozyme
structures, and oligoribonucleotides such as those described by
Inove et al., Nucleic Acids Res. 15:6131 (1987); and Chimeric
oligonucleotides that are composite RNA, DNA analogues (Inove, et
al, FEBS Lett. 2115:327 (1987). Oligonucleotides having a
lipophilic backbone, for example, methylphosphonate analogs with
ribozyme structures, may prove advantageous in certain
circumstances; these molecules may have a longer half-life in vivo
since the lipophilic structure may reduce the rate of renal
clearance while the ribozyme structure promotes cleavage of the
target RNA. Gerlach, Nature 334:585 (1988). Oligonucleotides with
2'-0-methyl modified ribose sugars are also suitable for the
practice of this invention. This ribose sugar modification can be
used with a variety of backbone types such as phosphorothioates and
may be limited to only some of the sugar residues such as those at
the ends of the oligonucleotide.
[0250] The oligonucleotides may be formulated into pharmaceutical
compositions and administered using a therapeutic regimen
compatible with the particular formulation. As described further
below, with the aid of present disclosure, those of skill in the
chemotherapeutic arts should be able to derive suitable dosages and
schedules of administration for any of a number of suitable
compositions that contain the compounds. Thus, pharmaceutical
compositions within the scope of the present invention include
compositions where the active ingredient is contained in an
effective amount to kill the cells of the cancer without causing
unacceptable toxicity for the patient. However, a preferred dosage
comprises that which is sufficient to achieve an effective blood
concentration of between about 1 and about 5 micro molar. Although
a preferred range has been described above, determination of the
effective amounts for treatment of each type of tumor may be
determined by those of skill in the art of chemotherapeutic
administration.
[0251] In addition to the antisense oligonucleotide compounds, the
pharmaceutical compositions of the invention may contain any of a
number of suitable excipients and auxiliaries which facilitate
processing of the active compounds into preparations that can be
used pharmaceutically. Preferably, the preparations will be
designed for parental administration. However, compositions
designed for oral or rectal administration are also considered to
fall within the scope of the present invention. Preferred
compositions will comprise from about 0.1 to about 1% by weight of
the active ingredients.
[0252] Suitable formulations for parental administration include
aqueous solutions of the active compounds in water-soluble or
water-dispersible form. Alternatively, suspensions of the active
compounds may be administered in suitable lipophilic carriers. The
formulations may contain substances which increase viscosity, for
example, sodium carboxymethyl cellulose, sorbitol, and/or dextran.
Optionally, the formulation may also contain stabilizers.
Additionally, the compounds of the present invention may also be
administered encapsulated in liposomes. The oligonucleotide,
depending upon its solubility, may be present both in the aqueous
layer and in the lipidic layer, or in what is generally termed a
liposomic suspension. The hydrophobic layer, generally but not
exclusively, comprises phospholipids such as lecithin and
sphingomyelin, steroids such as cholesterol, more or less ionic
surfactants such a diacetylphosphate, stearylamine, or phosphatidic
acid, and/or other materials of a hydrophobic nature.
TABLE-US-00003 Lengthy table referenced here
US20100010065A1-20100114-T00001 Please refer to the end of the
specification for access instructions.
TABLE-US-00004 Lengthy table referenced here
US20100010065A1-20100114-T00002 Please refer to the end of the
specification for access instructions.
TABLE-US-00005 Lengthy table referenced here
US20100010065A1-20100114-T00003 Please refer to the end of the
specification for access instructions.
TABLE-US-00006 Lengthy table referenced here
US20100010065A1-20100114-T00004 Please refer to the end of the
specification for access instructions.
TABLE-US-00007 Lengthy table referenced here
US20100010065A1-20100114-T00005 Please refer to the end of the
specification for access instructions.
TABLE-US-00008 Lengthy table referenced here
US20100010065A1-20100114-T00006 Please refer to the end of the
specification for access instructions.
TABLE-US-00009 Lengthy table referenced here
US20100010065A1-20100114-T00007 Please refer to the end of the
specification for access instructions.
TABLE-US-00010 Lengthy table referenced here
US20100010065A1-20100114-T00008 Please refer to the end of the
specification for access instructions.
TABLE-US-00011 Lengthy table referenced here
US20100010065A1-20100114-T00009 Please refer to the end of the
specification for access instructions.
TABLE-US-00012 Lengthy table referenced here
US20100010065A1-20100114-T00010 Please refer to the end of the
specification for access instructions.
TABLE-US-00013 Lengthy table referenced here
US20100010065A1-20100114-T00011 Please refer to the end of the
specification for access instructions.
TABLE-US-00014 Lengthy table referenced here
US20100010065A1-20100114-T00012 Please refer to the end of the
specification for access instructions.
TABLE-US-00015 Lengthy table referenced here
US20100010065A1-20100114-T00013 Please refer to the end of the
specification for access instructions.
TABLE-US-00016 Lengthy table referenced here
US20100010065A1-20100114-T00014 Please refer to the end of the
specification for access instructions.
TABLE-US-00017 Lengthy table referenced here
US20100010065A1-20100114-T00015 Please refer to the end of the
specification for access instructions.
TABLE-US-00018 Lengthy table referenced here
US20100010065A1-20100114-T00016 Please refer to the end of the
specification for access instructions.
TABLE-US-00019 Lengthy table referenced here
US20100010065A1-20100114-T00017 Please refer to the end of the
specification for access instructions.
TABLE-US-00020 Lengthy table referenced here
US20100010065A1-20100114-T00018 Please refer to the end of the
specification for access instructions.
TABLE-US-00021 Lengthy table referenced here
US20100010065A1-20100114-T00019 Please refer to the end of the
specification for access instructions.
TABLE-US-LTS-00001 LENGTHY TABLES The patent application contains a
lengthy table section. A copy of the table is available in
electronic form from the USPTO web site
(http://seqdata.uspto.gov/?pageRequest=docDetail&DocID=US20100010065A1).
An electronic copy of the table will also be available from the
USPTO upon request and payment of the fee set forth in 37 CFR
1.19(b)(3).
Sequence CWU 0 SQTB SEQUENCE LISTING The patent application
contains a lengthy "Sequence Listing" section. A copy of the
"Sequence Listing" is available in electronic form from the USPTO
web site
(http://seqdata.uspto.gov/?pageRequest=docDetail&DocID=US20100010065A1).
An electronic copy of the "Sequence Listing" will also be available
from the USPTO upon request and payment of the fee set forth in 37
CFR 1.19(b)(3).
0 SQTB SEQUENCE LISTING The patent application contains a lengthy
"Sequence Listing" section. A copy of the "Sequence Listing" is
available in electronic form from the USPTO web site
(http://seqdata.uspto.gov/?pageRequest=docDetail&DocID=US20100010065A1).
An electronic copy of the "Sequence Listing" will also be available
from the USPTO upon request and payment of the fee set forth in 37
CFR 1.19(b)(3).
* * * * *
References