Finding promiscuous old drugs for new uses
-
Upload
antony-williams-chemconnector-orcid-0000-0002-2668-4821 -
Category
Documents
-
view
283 -
download
4
Transcript of Finding promiscuous old drugs for new uses
1
Perspective
Finding Promiscuous Old Drugs For New Uses
Sean Ekins1, 2, 3, 4, Antony J. Williams5.
1 Collaborations in Chemistry, 601 Runnymede Avenue, Jenkintown, PA 19046, U.S.A.
2 Collaborative Drug Discovery, 1633 Bayshore Highway, Suite 342, Burlingame, CA
94010, U.S.A.
3 Department of Pharmaceutical Sciences, University of Maryland, MD 21201, U.S.A.
4 Department of Pharmacology, University of Medicine & Dentistry of New Jersey
(UMDNJ)-Robert Wood Johnson Medical School, 675 Hoes Lane, Piscataway, NJ
08854.
5 Royal Society of Chemistry, 904 Tamaras Circle, Wake Forest, NC-27587, U.S.A.
Running head: Repurposing old drugs
Corresponding Author: Sean Ekins, Collaborations in Chemistry, 601 Runnymede Ave,
Jenkintown, PA 19046, Email [email protected], Tel 215-687-1320.
2
From research published in the last 6 years we have identified 34 studies that have
screened various libraries of FDA approved drugs against various whole cell or target
assays. These studies have each identified one or more compounds with a suggested new
bioactivity that had not been described previously. We now show that thirteen of these
drugs were active against more than one additional disease, thereby suggesting a degree
of promiscuity. We also show that following compilation of all the studies, 109
molecules were identified by screening in vitro. These molecules appear to be statistically
more hydrophobic with a higher molecular weight and AlogP than orphan designated
products with at least one marketing approval for a common disease indication or one
marketing approval for a rare disease from the FDA’s rare disease research database.
Capturing these in vitro data on old drugs for new uses will be important for potential
reuse and analysis by others to repurpose or reposition these or other existing drugs. We
have created databases which can be searched by the public and envisage that these can
be updated as more studies are published.
Keywords: cheminformatics, Old drugs, repositioning, repurposing, HTS
3
Introduction
As productivity of the pharmaceutical industry continues to stagnate we call attention to
the merits of reconsidering new potential applications of drugs that are already approved,
whether they be old or new (1). This is commonly termed drug repositioning, drug
repurposing or finding “new uses for old drugs”, and has been reviewed extensively in
the context of finding uses for drugs applied to major diseases (2) but is also of value for
orphan or rare diseases. The benefits of repositioning include: the availability of chemical
materials and previously generated data that can be used and presented to regulatory
authorities and, as a result, the potential for a significantly more time- and cost-effective
research and development effort than typically experienced when bringing a new drug to
market.
To date multiple academic groups have screened 1,000-2,000 drugs against
different targets or cell types relevant to rare, neglected and common diseases and this
information has not been thoroughly compared or captured in a database for analysis until
now (Supplemental Table 1). We have identified at least 34 such studies published in the
last 6 years which have identified one or more drug molecule active in either whole cell
or target-based assays. Several of these studies attempt to find new molecules active
against diseases like malaria and tuberculosis for which there are several approved drugs,
yet there is still a need to find molecules with a better side effect profile or as a
replacement for drugs for which resistance has been shown. These issues alone justify the
continued search for drugs perhaps with novel mechanisms of action.
4
Several libraries of FDA-approved or foreign-approved drugs have been screened
but there is currently not one definitive source of all these molecules that researchers
could access at cost for themselves. For example, the John Hopkins Clinical Compound
Library (JHCCL) consists of plated compounds available for screening at a relatively
small charge and has been examined by more than 20 groups with more than a half dozen
publications to date (3-6). A number of new uses for FDA approved drugs have been
identified by screening these or other commercially available libraries of drugs or off-
patent molecules e.g. the NINDS/Microsource US drug collection and Prestwick
Chemical library (see Supplemental Table 1). In total a conservative estimate indicates at
least 109 previously approved drugs have shown activity in vitro against additional
diseases different than those for which the drugs were originally approved. For these
molecules to have any impact on their respective diseases they will obviously have to
show in vivo efficacy. Upon manual curation of this dataset we were able to create a
database of validated structures which is now publically available
(www.collaborativedrug.com). In addition we were able to generate molecular properties
for these molecules. We invite others to speculate as to which may show in vivo relevant
activity. We have performed several analyses of the dataset to understand how they
compare to drugs already repurposed for rare diseases.
Promiscuous in vitro repurposed drugs
Thirteen of these 109 drugs, (Figure 1), showed activity against more than one
additional disease, thereby suggesting a degree of promiscuity which we believe has not
been widely acknowledged elsewhere. We found through our meta-analysis that the class
5
III antiarrhythmic amiodarone was active in neurodegeneration assays and could also
selectively remove embryonic stem cells. The antidepressants amitriptyline and
clomipramine suppressed glial fibrially acidic protein (7) and inhibited mitochondrial
permeability transition (8). The anti-psychotic chlorprothixene showed antimalarial
activity (9) and suppressed glial fibrially acidic protein (7). The anti-cancer drug
daunorubicin was active against neuroblastoma (10) and was an NF-kB inhibitor (11).
The cardiac glycoside digoxin was active against retinoblastoma (12) and an inhibitor of
hypoxia inducible factor (13). The progestrogen hydroxyprogesterone has antimalarial (9)
and glucocorticoid receptor modulator activity. The antineoplastic mitoxantrone was
active against neuroblastoma and was a glucocorticoid receptor modulator (14). The
cardiac glycoside ouabain was an inhibitor of hypoxia inducible factor (13) and NF-kB
(11). The antipsychotic prochlorperazine was an inhibitor of mitochondrial permeability
transition (8) and myosin-II associated S100A4 (15). The antihelmintic Pyrvinium
pamoate has antituberculosis activity (6) and antiprotozoal activity against C. parvum
(16) and T. Brucei (17). The anti-psychotic thioridazine had antimalarial activity (9) and
was an inhibitor of mitochondrial permeability transition (8). Finally, the anti-psychotic
trifluoperazine was active in neurodegeneration assays (18), an inhibitor of mitochondrial
permeability transition (8) and myosin-II associated S100A4 (15).
Interestingly, the mean predicted molecular properties of these ‘promiscuous
compounds’ are AlogP 3.6 +/- and molecular weight 443 +/- (Table 1). These values are
not statistically significantly different when compared to the whole dataset of 109
molecules (mean AlogP of 3.1+/- and molecular weight of 428 +/-) and are closest to the
“natural product lead-like rules” (MW < 460, Log P< 4.2) described elsewhere (19). This
6
is suggestive that the 109 molecules are generally quite large compared to drugs in
general as for example, Vieth et al., 1193 oral drugs were shown to have a mean MWT of
343.7 and CLOGP of 2.3 (20). Another group has screened 3138 compounds against 79
assays, primarily GPCR, and showed that approximately 20-30 of the compounds were
promiscuous compounds and had a mean MWT (493) and AlogP (4.4) that was higher
than for selective compounds, 436 and 3.3, respectively (21). However, no statistical
testing was presented to show whether this was significant or not. It is possible that our
set of promiscuous compounds is too small to discern any meaningful difference.
Preventing rediscovery
From our analysis (see Supplemental Table 1) there are several examples in which
independent groups have screened drug libraries in whole cell assays or used different
assays to discover compounds with similar activity such as glial fibrially acidic protein
and mitochondrial permeability transition for neurodegeneration, and hypoxia inducible
factor and NF-kB for cancer. Additionally, several groups have screened FDA approved
drugs against malaria (9, 22). How do researchers now avoid repeating the same
discoveries that others have made? One way would be to capture all of the published uses
of these drugs in vitro and combine with information on uses that have already been
identified in the laboratory or clinic. This has not been done to date. The FDA has
recently provided a resource, the rare disease research database (RDRD), which lists
Orphan-designated products
(http://www.fda.gov/ForIndustry/DevelopingProductsforRareDiseasesConditions/Howtoa
7
pplyforOrphanProductDesignation/ucm216147.htm) with at least one marketing approval
for a common disease indication, for a rare disease indication, or for both common and
rare disease indications. In the last category there are less than 50 molecules (including
large biopharmaceutical drugs). These tables from the FDA do not capture the high
throughput screening (HTS) data generated to date from diverse laboratories involved in
screening libraries of drugs (Supplemental Table 1).
We have curated the molecular structures for these datasets and generated their
physicochemical properties. The mean predicted molecular properties of these
compounds in the RDRD databases with at least one marketing approval for a common
disease indication include AlogP 1.4 and molecular weight 353 (Table 2), while those
with at least one marketing approval for a rare disease indication have AlogP 0.9 and
molecular weight 344. Although these values have large standard deviations they are
close to the published “lead-like” rules (MW < 350, LogP< 3, Affinity ~0.1uM) (23, 24)
and closer to the properties of ‘oral drugs’ highlighted by Vieth et al., (20). When these
two datasets are compared with the 109 previously approved drugs shown to have
activity in vitro against additional diseases (Table 1) the differences in AlogP and MWT
are statistically significant. Also, the number of rings and aromatic rings are higher in the
in vitro dataset. It should be noted that these datasets are relatively small with several
showing skewed property distributions, hence the use of non-parametric testing and some
of the properties like LogP and MW correlate weakly (r2 = 0.07), while other properties
such as the number of rings and MW more strongly (r2 = 0.61). Such correlations
between physicochemical properties in large sets of FDA approved drugs have been
indicated by others (20). However, our analysis may suggest for the first time that
8
compounds with activity and approved for rare diseases have different LogP and MWT to
those compounds that have been shown to have in vitro activity for various diseases
(including rare and neglected).
The excel files provided by the FDA are not structure searchable or connected to
data in other NIH databases that may be of utility for assisting researchers. There are
other useful resources that are less well known. The Collaborative Drug Discovery
(CDD) database (25) has focused on collecting data for neglected diseases (26-28). Dr.
Chris Lipinski (Melior Discovery) provided a database of 1055 FDA approved drugs with
designated orphan indications, sponsor name and chemical structures. In addition, CDD
has collated and provided a database of 2815 FDA approved drugs from a list of all
approved drugs since 1938 (22). These data, can enable cheminformatics analysis of the
physicochemical properties of compounds (27, 29, 30) and are available for free access
and searchable by substructure, similarity or Boolean searches upon registration (e.g.,
see: http://www.collaborativedrug.com/register). We have therefore made the datasets
from this study, and those curated based on the content in RDRD, publically accessible in
the CDD database.
The curation of datasets of available drugs or orphan drugs with their uses could
be used for searching with pharmacophore models (31) or other machine-learning
methods to find new compounds for testing in vitro and to accelerate the repositioning
process or focusing of in vitro screening on select compounds (32, 33). A study using
similarity ensemble analysis, applying Bayesian models to predict off-target effects of
3665 FDA approved drugs and investigational compounds (34) and showed the
9
promiscuity of many compounds. While the in vitro validation of the computational
predictions focused on GPCRs, some of the collated data from the current study could
also provide a useful method for further validation of this or other future in silico
repositioning methods (35).
Making repositioning routine
As the availability, at a reasonable cost of FDA approved drugs in a format for
HTS is now commonplace, what remains necessary so that the burgeoning numbers of
academic screening centers or other groups can accelerate repositioning? An exhaustive
database that cross references the molecules, papers, and activities would certainly be a
valuable starting point and capturing the hit rates of such libraries versus other compound
library screening and clinical data would be valuable. It is not yet obvious whether a drug
has progressed straight from these in vitro screens to orphan drug status but the screening
of drug libraries may certainly accelerate this. Evidence of migration from in vitro
screens to orphan status would obviously be immensely valuable. Clearly very old drugs
like the tricyclic antidepressants, anti-psychotics and cardiac glycosides appear to be
promiscuous, having been found to possess many activities against additional diseases in
vitro. Whether these ‘new uses for old promiscuous drugs’ will translate into the clinic,
remains in question. The follow up of compounds from in vitro screening to appearance
in the clinic is limited as in the case of Ara-C (cytarabine) for Ewing’s sarcoma which
went to a Phase II clinical study and showed toxicity and minimal activity (36). To our
knowledge, in most cases clinical studies have not been described in over 6 years in
10
which this high throughput screening work has appeared. Perhaps focusing on screening
just these few classes of promiscuous compounds against any disease of interest would
yield additional activities and test this hypothesis.
In performing our analysis of the literature it appears that many groups have taken
the ‘new uses for old drugs’ approach (37). At the same time it has not been recognized
that there appears to be a subset of ‘promiscuous’ old drugs (approximately 12% of the
compounds identified to date in vitro). We cannot however distinguish these molecules as
different from the complete dataset based on the simple molecular descriptors used in this
study. The 109 molecules identified by screening in vitro appear to be statistically more
hydrophobic and with a higher molecular weight and AlogP than orphan designated
products with at least one marketing approval for a common disease indication or one
marketing approval for a rare disease from the FDA RDRD. These may be useful
insights, suggesting that some compounds that may have different molecular properties to
those already orphan designated, may have many potential repositioning activities and
could be the focus of more aggressive screening against many more diseases. It will also
be important to rule out in vitro false positives due to aggregation (38) or other causes.
Capturing these in vitro data on promiscuous old drugs for new uses in a format that is
readily mined will be important for reuse and analysis by others and we welcome
suggestions as to who should be responsible for funding, developing and maintaining it.
Since this perspective was originally submitted for publication and passed through
the peer review process it has come to our attention that the NIH Chemical Genomics
Center has released a database described as a comprehensive resource of clinically
approved drugs to enable repurposing and chemical genomics (39). This will be used
11
along with the NCGC screening resources as a component of the NIH therapeutics for
rare and neglected diseases (TRND) program. The database has undergone a preliminary
evaluation by us and may indeed be a useful future resource for the community. However
we urge significant caution due to a large number of errors identified in the molecular
structure representations in the database (40) and hence this database will need further
curation and correction before the structures can be used for other applications such as
virtual screening. We believe there is scope for several efforts to provide databases of
validated compounds and data that may be useful for repurposing.
Conflicts of Interest
SE consults for Collaborative Drug Discovery, Inc on a Bill and Melinda Gates
Foundation Grant#49852 “Collaborative drug discovery for TB through a novel database
of SAR data optimized to promote data archiving and sharing”.
Acknowledgments
SE gratefully acknowledges David Sullivan (Johns Hopkins University) for
discussing and suggesting references for JHCCL. Accelrys are kindly thanked for
providing Discovery Studio.
12
References
1. C.R. Chong and D.J. Sullivan, Jr. New uses for old drugs. Nature. 448:645-646
(2007).
2. T.T. Ashburn and K.B. Thor. Drug repositioning: identifying and developing new
uses for existing drugs. Nat Rev Drug Discov. 3:673-683 (2004).
3. S.T. Byrne, P. Gu, J. Zhou, S.M. Denkin, C. Chong, D. Sullivan, J.O. Liu, and Y.
Zhang. Pyrrolidine dithiocarbamate and diethyldithiocarbamate are active against
growing and nongrowing persister Mycobacterium tuberculosis. Antimicrob
Agents Chemother. 51:4495-4497 (2007).
4. C. Gloeckner, A.L. Garner, F. Mersha, Y. Oksov, N. Tricoche, L.M. Eubanks, S.
Lustigman, G.F. Kaufmann, and K.D. Janda. Repositioning of an existing drug for
the neglected tropical disease Onchocerciasis. Proc Natl Acad Sci U S A.
107:3424-3429 (2010).
5. J.S. Shim, Y. Matsui, S. Bhat, B.A. Nacev, J. Xu, H.E. Bhang, S. Dhara, K.C.
Han, C.R. Chong, M.G. Pomper, A. So, and J.O. Liu. Effect of nitroxoline on
angiogenesis and growth of human bladder cancer. Journal of the National Cancer
Institute (2010).
6. K.E. Lougheed, D.L. Taylor, S.A. Osborne, J.S. Bryans, and R.S. Buxton. New
anti-tuberculosis agents amongst known drugs. Tuberculosis (Edinburgh,
Scotland). 89:364-370 (2009).
13
7. W. Cho, M. Brenner, N. Peters, and A. Messing. Drug screening to identify
suppressors of GFAP expression. Human molecular genetics. 19:3169-3178
(2010).
8. I.G. Stavrovskaya, M.V. Narayanan, W. Zhang, B.F. Krasnikov, J. Heemskerk,
S.S. Young, J.P. Blass, A.M. Brown, M.F. Beal, R.M. Friedlander, and B.S.
Kristal. Clinically approved heterocyclics act on a mitochondrial target and
reduce stroke-induced pathology. The Journal of experimental medicine. 200:211-
222 (2004).
9. J.L. Weisman, A.P. Liou, A.A. Shelat, F.E. Cohen, R.K. Guy, and J.L. DeRisi.
Searching for new antimalarial therapeutics amongst known drugs. Chemical
biology & drug design. 67:409-416 (2006).
10. J.S. Gheeya, Q.R. Chen, C.D. Benjamin, A.T. Cheuk, P. Tsang, J.Y. Chung, B.B.
Metaferia, T.C. Badgett, P. Johansson, J.S. Wei, S.M. Hewitt, and J. Khan.
Screening a panel of drugs with diverse mechanisms of action yields potential
therapeutic agents against neuroblastoma. Cancer biology & therapy. 8:2386-2395
(2009).
11. S.C. Miller, R. Huang, S. Sakamuru, S.J. Shukla, M.S. Attene-Ramos, P. Shinn,
D. Van Leer, W. Leister, C.P. Austin, and M. Xia. Identification of known drugs
that act as inhibitors of NF-kappaB signaling and their mechanism of action.
Biochem Pharmacol. 79:1272-1280 (2010).
14
12. C. Antczak, C. Kloepping, C. Radu, T. Genski, L. Muller-Kuhrt, K. Siems, E. de
Stanchina, D.H. Abramson, and H. Djaballah. Revisiting old drugs as novel
agents for retinoblastoma: in vitro and in vivo antitumor activity of cardenolides.
Invest Ophthalmol Vis Sci. 50:3065-3073 (2009).
13. H. Zhang, D.Z. Qian, Y.S. Tan, K. Lee, P. Gao, Y.R. Ren, S. Rey, H. Hammers,
D. Chang, R. Pili, C.V. Dang, J.O. Liu, and G.L. Semenza. Digoxin and other
cardiac glycosides inhibit HIF-1alpha synthesis and block tumor growth. Proc
Natl Acad Sci U S A. 105:19579-19586 (2008).
14. A.N. Gerber, K. Masuno, and M.I. Diamond. Discovery of selective
glucocorticoid receptor modulators by multiplexed reporter screening. Proc Natl
Acad Sci U S A. 106:4929-4934 (2009).
15. S.C. Garrett, L. Hodgson, A. Rybin, A. Toutchkine, K.M. Hahn, D.S. Lawrence,
and A.R. Bresnick. A biosensor of S100A4 metastasis factor activation: inhibitor
screening and cellular activation dynamics. Biochemistry. 47:986-996 (2008).
16. A.S. Downey, C.R. Chong, T.K. Graczyk, and D.J. Sullivan. Efficacy of
pyrvinium pamoate against Cryptosporidium parvum infection in vitro and in a
neonatal mouse model. Antimicrob Agents Chemother. 52:3106-3112 (2008).
17. Z.B. Mackey, A.M. Baca, J.P. Mallari, B. Apsel, A. Shelat, E.J. Hansell, P.K.
Chiang, B. Wolff, K.R. Guy, J. Williams, and J.H. McKerrow. Discovery of
trypanocidal compounds by whole cell HTS of Trypanosoma brucei. Chemical
biology & drug design. 67:355-363 (2006).
15
18. L. Zhang, J. Yu, H. Pan, P. Hu, Y. Hao, W. Cai, H. Zhu, A.D. Yu, X. Xie, D. Ma,
and J. Yuan. Small molecule regulators of autophagy identified by an image-
based high-throughput screen. Proc Natl Acad Sci U S A. 104:19023-19028
(2007).
19. J. Rosen, J. Gottfries, S. Muresan, A. Backlund, and T.I. Oprea. Novel Chemical
Space Exploration via Natural Products. J Med Chem. 52:1953-1962 (2009).
20. M. Vieth, M.G. Siegel, R.E. Higgs, I.A. Watson, D.H. Robertson, K.A. Savin,
G.L. Durst, and P.A. Hipskind. Characteristic physical properties and structural
fragments of marketed oral drugs. J Med Chem. 47:224-232 (2004).
21. K. Azzaoui, J. Hamon, B. Faller, S. Whitebread, E. Jacoby, A. Bender, J.L.
Jenkins, and L. Urban. Modeling promiscuity based on in vitro safety
pharmacology profiling data. ChemMedChem. 2:874-880 (2007).
22. C.R. Chong, X. Chen, L. Shi, J.O. Liu, and D.J. Sullivan, Jr. A clinical drug
library screen identifies astemizole as an antimalarial agent. Nat Chem Biol.
2:415-416 (2006).
23. T.I. Oprea. Current trends in lead discovery: are we looking for the appropriate
properties? J Comput Aided Mol Des. 16:325-334 (2002).
24. T.I. Oprea, A.M. Davis, S.J. Teague, and P.D. Leeson. Is there a difference
between leads and drugs? A historical perspective. J Chem Inf Comput Sci.
41:1308-1315 (2001).
16
25. M. Hohman, K. Gregory, K. Chibale, P.J. Smith, S. Ekins, and B. Bunin. Novel
web-based tools combining chemistry informatics, biology and social networks
for drug discovery. Drug Disc Today. 14:261-270 (2009).
26. S. Ekins, J. Bradford, K. Dole, A. Spektor, K. Gregory, D. Blondeau, M.
Hohman, and B. Bunin. A Collaborative Database And Computational Models
For Tuberculosis Drug Discovery. Mol BioSystems. 6:840-851 (2010).
27. S. Ekins, T. Kaneko, C.A. Lipinksi, J. Bradford, K. Dole, A. Spektor, K. Gregory,
D. Blondeau, S. Ernst, J. Yang, N. Goncharoff, M. Hohman, and B. Bunin.
Analysis and hit filtering of a very large library of compounds screened against
Mycobacterium tuberculosis Molecular bioSystems. 6:2316-2324 (2010).
28. S. Ekins, M. Hohman, and B.A. Bunin. Pioneering use of the cloud for
development of the collaborative drug discovery (cdd) database In S. Ekins,
M.A.Z. Hupcey, and A.J. Williams (eds.), Collaborative Computational
Technologies for Biomedical Research, Vol. in press, Wiley and Sons, Hoboken,
2010.
29. S. Ekins and A.J. Williams. Meta-analysis of molecular property patterns and
filtering of public datasets of antimalarial “hits” and drugs. MedChemComm.
1:325-330 (2010).
30. S. Ekins and A.J. Williams. When Pharmaceutical Companies Publish Large
Datasets: An Abundance Of Riches Or Fool’s Gold? Drug Disc Today. 15:812-
815 (2010).
17
31. S. Kortagere, M.D. Krasowski, and S. Ekins. The importance of discerning shape
in molecular pharmacology. Trends Pharmacol Sci. 30:138-147 (2009).
32. X. Zheng, S. Ekins, J.-P. Rauffman, and J.E. Polli. Computational models for
drug inhibition of the Human Apical Sodium-dependent Bile Acid Transporter.
Mol Pharm 6:1591-1603 (2009).
33. L. Diao, S. Ekins, and J.E. Polli. Novel Inhibitors of Human Organic
Cation/Carnitine Transporter (hOCTN2) via Computational Modeling and In
Vitro Testing. Pharm Res. 26:1890-1900 (2009).
34. M.J. Keiser, V. Setola, J.J. Irwin, C. Laggner, A.I. Abbas, S.J. Hufeisen, N.H.
Jensen, M.B. Kuijer, R.C. Matos, T.B. Tran, R. Whaley, R.A. Glennon, J. Hert,
K.L. Thomas, D.D. Edwards, B.K. Shoichet, and B.L. Roth. Predicting new
molecular targets for known drugs. Nature. 462:175-181 (2009).
35. S. Ekins, A.J. Williams, M.D. Krasowski, and J.S. Freundlich. In silico
repositioning of approved drugs for rare and neglected diseases. Drug Disc
Today. In Press: (2011).
36. S.G. DuBois, M.D. Krailo, S.L. Lessnick, R. Smith, Z. Chen, N. Marina, H.E.
Grier, and K. Stegmaier. Phase II study of intermediate-dose cytarabine in
patients with relapsed or refractory Ewing sarcoma: a report from the Children's
Oncology Group. Pediatric blood & cancer. 52:324-327 (2009).
18
37. K.A. O'Connor and B.L. Roth. Finding new tricks for old drugs: an efficient route
for public-sector drug discovery. Nat Rev Drug Discov. 4:1005-1014 (2005).
38. B.Y. Feng, A. Simeonov, A. Jadhav, K. Babaoglu, J. Inglese, B.K. Shoichet, and
C.P. Austin. A high-throughput screen for aggregation-based inhibition in a large
compound library. J Med Chem. 50:2385-2390 (2007).
39. R. Huang, N. Southall, Y. Wang, A. Yasgar, P. Shinn, A. Jadhav, D.T. Nguyen,
and C.P. Austin. The NCGC Pharmaceutical Collection: A Comprehensive
Resource of Clinically Approved Drugs Enabling Repurposing and Chemical
Genomics. Science translational medicine. 3:80ps16 (2011).
40. A.J. Williams. Reviewing Data Quality in the NCGC Pharmaceutical Collection
Browser. http://www.chemconnector.com/2011/04/28/reviewing-data-quality-in-
the-ncgc-pharmaceutical-collection-browser/.
41. C.R. Chong, J. Xu, J. Lu, S. Bhat, D.J. Sullivan, Jr., and J.O. Liu. Inhibition of
angiogenesis by the antifungal drug itraconazole. ACS chemical biology. 2:263-
270 (2007).
42. C.R. Chong, D.Z. Qian, F. Pan, Y. Wei, R. Pili, D.J. Sullivan, Jr., and J.O. Liu.
Identification of type 1 inosine monophosphate dehydrogenase as an
antiangiogenic drug target. J Med Chem. 49:2677-2680 (2006).
19
43. L.P. de Carvalho, G. Lin, X. Jiang, and C. Nathan. Nitazoxanide kills replicating
and nonreplicating Mycobacterium tuberculosis and evades resistance. J Med
Chem. 52:5789-5792 (2009).
44. D. Shahinas, M. Liang, A. Datti, and D.R. Pillai. A repurposing strategy identifies
novel synergistic inhibitors of Plasmodium falciparum heat shock protein 90. J
Med Chem. 53:3552-3557 (2010).
45. S. Chopra, M. Torres-Ortiz, L. Hokama, P. Madrid, M. Tanga, K. Mortelmans, K.
Kodukula, and A.K. Galande. Repurposing FDA-approved drugs to combat drug-
resistant Acinetobacter baumannii. The Journal of antimicrobial chemotherapy.
65:2598-2601 (2010).
46. T.L. Biechele, N.D. Camp, D.M. Fass, R.M. Kulikauskas, N.C. Robin, B.D.
White, C.M. Taraska, E.C. Moore, J. Muster, R. Karmacharya, S.J. Haggarty, A.J.
Chien, and R.T. Moon. Chemical-Genetic Screen Identifies Riluzole as an
Enhancer of Wnt/beta-catenin Signaling in Melanoma. Chem Biol. 17:1177-1182
(2010).
47. Y. Zhang, Y. Byun, Y.R. Ren, J.O. Liu, J. Laterra, and M.G. Pomper.
Identification of inhibitors of ABCG2 by a bioluminescence imaging-based high-
throughput assay. Cancer Res. 69:5867-5875 (2009).
48. N. Masuda, Q. Peng, Q. Li, M. Jiang, Y. Liang, X. Wang, M. Zhao, W. Wang,
C.A. Ross, and W. Duan. Tiagabine is neuroprotective in the N171-82Q and R6/2
20
mouse models of Huntington's disease. Neurobiology of disease. 30:293-302
(2008).
49. J.D. Rothstein, S. Patel, M.R. Regan, C. Haenggeli, Y.H. Huang, D.E. Bergles, L.
Jin, M. Dykes Hoberg, S. Vidensky, D.S. Chung, S.V. Toan, L.I. Bruijn, Z.Z. Su,
P. Gupta, and P.B. Fisher. Beta-lactam antibiotics offer neuroprotection by
increasing glutamate transporter expression. Nature. 433:73-77 (2005).
50. S.A. Peterson, T. Klabunde, H.A. Lashuel, H. Purkey, J.C. Sacchettini, and J.W.
Kelly. Inhibiting transthyretin conformational changes that lead to amyloid fibril
formation. Proc Natl Acad Sci U S A. 95:12956-12960 (1998).
51. X. Wang, S. Zhu, Z. Pei, M. Drozda, I.G. Stavrovskaya, S.J. Del Signore, K.
Cormier, E.M. Shimony, H. Wang, R.J. Ferrante, B.S. Kristal, and R.M.
Friedlander. Inhibitors of cytochrome c release with therapeutic potential for
Huntington's disease. J Neurosci. 28:9473-9485 (2008).
52. H. Wang, Y. Guan, X. Wang, K. Smith, K. Cormier, S. Zhu, I.G. Stavrovskaya,
C. Huo, R.J. Ferrante, B.S. Kristal, and R.M. Friedlander. Nortriptyline delays
disease onset in models of chronic neurodegeneration. The European journal of
neuroscience. 26:633-641 (2007).
53. U.A. Desai, J. Pallos, A.A. Ma, B.R. Stockwell, L.M. Thompson, J.L. Marsh, and
M.I. Diamond. Biologically active molecules that reduce polyglutamine
aggregation and toxicity. Human molecular genetics. 15:2114-2124 (2006).
21
54. K. Stegmaier, J.S. Wong, K.N. Ross, K.T. Chow, D. Peck, R.D. Wright, S.L.
Lessnick, A.L. Kung, and T.R. Golub. Signature-based small molecule screening
identifies cytosine arabinoside as an EWS/FLI modulator in Ewing sarcoma.
PLoS medicine. 4:e122 (2007).
55. Y. Han, A. Miller, J. Mangada, Y. Liu, A. Swistowski, M. Zhan, M.S. Rao, and
X. Zeng. Identification by automated screening of a small molecule that
selectively eliminates neural stem cells derived from hESCs but not dopamine
neurons. PloS one. 4:e7155 (2009).
56. H.C. Ou, L.L. Cunningham, S.P. Francis, C.S. Brandon, J.A. Simon, D.W. Raible,
and E.W. Rubel. Identification of FDA-approved drugs and bioactives that protect
hair cells in the zebrafish (Danio rerio) lateral line and mouse (Mus musculus)
utricle. J Assoc Res Otolaryngol. 10:191-203 (2009).
57. S.M. Pollard, K. Yoshikawa, I.D. Clarke, D. Danovi, S. Stricker, R. Russell, J.
Bayani, R. Head, M. Lee, M. Bernstein, J.A. Squire, A. Smith, and P. Dirks.
Glioma stem cell lines expanded in adherent culture have tumor-specific
phenotypes and are suitable for chemical and genetic screens. Cell stem cell.
4:568-580 (2009).
22
Table 1. Calculated mean molecular properties (±SD) of orphan designated products and compounds identified with
additional potential therapeutic uses through in vitro high throughput screening of approved drug libraries. Properties
calculated using Discovery Studio 2.5.5 (Accelrys, San Diego, CA). The datasets of approved drugs repositioned for common or rare
diseases from the FDA’s rare disease research database were compared with the in vitro dataset (N = 109) curated in this study using a
Non-parametric Wilcoxon / Kruskal-Wallis 2 sample test, a p < 0.05, b p < 0.0001. Comparison of the mean molecular properties for
the subset of thirteen in vitro inhibitors with the larger dataset (n = 109) did not show a statistically significant difference. Range is in
parenthesis. All datasets are available at www.collaborativedrug.com.
Dataset ALogP Molecular
Weight
Number
of
Rotatable
Bonds
Number
of Rings
Number of
Aromatic
Rings
Number of
Hydrogen
bond
Acceptors
Number of
Hydrogen
bond Donors
Molecular
Polar
Surface
Area
Compounds identified in vitro with
new activities (N = 109) *
3.1 ± 2.6
(-4.3 –
13.93)
428.4 ± 202.8
(167-2 –
1255.42)
5.4 ± 3.8
(0 – 20)
3.8 ± 1.9
(0 – 12)
2.0 ± 1.4
(0 – 12)
5.6 ± 4.2
(1 – 27)
2.0 ± 1.9
(0 – 9)
89.6 ± 69.3
(3.2 –
379.6)
23
Compounds identified in vitro with
multiple new activities (N = 13)
3.6 ± 2.7
(-2.2 –
7.2)
442.8 ± 150.0
(277.4 – 780.9)
5.1 ± 3.1
(1 – 12)
4.2 ± 1.5
(3 – 8)
1.8 ± 1.2
(0 – 4)
5.5 ± 4.6
(1 – 14)
2.2 ± 3.3
(0 – 8)
79.5 ± 78.8
(3.2 –
206.6)
Orphan designated products with at
least one marketing approval for a
common disease indication (N = 79) #
1.4 ± 3.0 b
(-12.6 –
6.4)
353.2 ± 218.8 a
(78.1 –
1462.71)
5.3 ± 6.4
(0 – 37)
2.8 ± 1.7 a
(0 – 8)
1.2 ± 1.3 b
(0 – 6)
5.3 ± 6.0
(1 – 51)
2.5 ± 3.0
(0 – 18)
99.2 ± 110.7
(12.5 –
839.2)
Orphan designated products with at
least one marketing approval for a rare
disease indication (N = 52) #
0.9 ± 3.3 b
(-13.1 –
8.3)
344.4 ± 233.5 a
(30.0 – 1394.6)
5.3 ± 5.3
(0 – 34)
2.4 ± 1.9 b
(0- 10)
1.3 ± 1.4 b
(0 – 6)
6.2 ± 4.2
(2 – 25)
2.7 ± 2.8
(0 – 17)
114.2 ± 85.3
a
(37.3 –
544.8)
*disulfiram excluded from this analysis.
# Compounds from the FDA rare disease research database (RDRD), which lists Orphan-designated products
(http://www.fda.gov/ForIndustry/DevelopingProductsforRareDiseasesConditions/HowtoapplyforOrphanProductDesignation/ucm2161
47.htm)
24
Figure 1. Structures of FDA approved drugs found to have multiple activities beyond
what they were approved for when screened in vitro. Structures downloaded from
www.chemspider.com.
Amiodarone Amitriptyline
Clomipramine Chlorprothixene
Daunorubicin Digoxin
25
Hydroxyprogesterone Mitoxantrone
Ouabain Prochlorperazine
Pyrvinium Thioridazine
Trifluoperazine
26
Supplemental data for:
Perspective
Finding Promiscuous and Non-promiscuous Old Drugs For New Uses
Sean Ekins1, 2, 3, 4, Antony J. Williams5.
1 Collaborations in Chemistry, 601 Runnymede Avenue, Jenkintown, PA 19046, U.S.A.
2 Collaborative Drug Discovery, 1633 Bayshore Highway, Suite 342, Burlingame, CA
94010, U.S.A.
3 Department of Pharmaceutical Sciences, University of Maryland, MD 21201, U.S.A.
4 Department of Pharmacology, University of Medicine & Dentistry of New Jersey
(UMDNJ)-Robert Wood Johnson Medical School, 675 Hoes Lane, Piscataway, NJ
08854.
5 Royal Society of Chemistry, 904 Tamaras Circle, Wake Forest, NC-27587, U.S.A.
Running head: Repurposing old drugs
Corresponding Author: Sean Ekins, Collaborations in Chemistry, 601 Runnymede Ave,
Jenkintown, PA 19046, Email [email protected], Tel 215-687-1320.
27
Supplemental Table 1. Drugs identified with new uses using HTS methods. This
table greatly extends a previously published version (35).
CCR5, Chemokine receptor 5; DHFR, Dihydrofolate reductase; DOA, Drugs of abuse,
FDA, Food and Drug Administration; GLT1, Glutamate transporter 1; HSP-90, Heat
shock protein 90; JHCCL, John Hopkins Clinical Compound Library; Mtb,
Mycobacterium tuberculosis; NK-1, neurokinin- 1 receptor; OCTN2.
Molecule Original use (Target
or drug class)
New use and activity How discovered Ref
Itraconazole Antifungal –
lanosterol 14-
demethylase inhibitor
Inhibition of angiogenesis by
inhibiting human lanosterol 14-
demethylase IC50 160nM.
In vitro HUVEC
proliferation screen
against FDA approved
drugs (JHCCL)
(41)
Astemizole Non-sedating
antihistamine
(removed from USA
market by FDA in
1999)
Antimalarial IC50 227nM against
P. falciparum 3D7.
In vitro screen for P.
Falciparum growth of
1937 FDA approved
drugs (JHCCL)
(22)
Mycophenolic
acid
Immunosuppresive
drug inhibits guanine
nucleotide
biosynthesis
Inhibition of angiogenesis by
targeting Type 1 inosine
monophosphate dehydrogenase
IC50 99.2nM.
In vitro HUVEC
proliferation screen 2450
FDA and foreign
approved drugs (JHCCL)
(42)
28
Disulfiram Alcohol deterrent Anti-tuberculosis - MIC 8 to 16
g/ml – screen also identified
sodium diethyldithiocarbamate
(metabolite of disulfiram) and
pyrrolidine diethocarbamate
which were more active.
Mechanism may be due to metal
chelation
Screened 3360
compounds (JHCCL)
against Mtb H37Ra
(3)
Nitazoxanide Infections caused by
Giarda and
cryptosproridium
Anti-tuberculosis – multiple
potential targets.
Screens against
replicating and non
replicating M.
tuberculosis
(43)
(±)-2-amino-3-
phosphonopropio
nic acid,
Acrisorcin,
Harmine
Human metabolite,
mGLUR agonist,
Antifungal,
anticancer
Antimalarial – Inhibits HSP90
against P. falciparum 3D7.
HTS screening 4000
compounds
(44)
Levofloxacin,
Gatifloxacin,
Sarafloxacin,
Moxifloxacin,
Gemifloxacin
DNA gyrase Active against ATCC17978
inactive against BAA-1605 MIC
<0.03 – 0.04 (mg/L)
Screened Microsource
drugs library of 1040
drugs versus A.
baumannii
(45)
Bithionol, Various NF-B inhibitors IC50 0.02- Screened NCGC (11)
29
Bortezomib,
Cantharidin,
Chromomycin
A3,
Daunorubicin,
Digitoxin,
Ectinascidin 743,
Emetine,
Fluorosalan,
Manidipine HCl,
Narasin,
Lestaurtinib,
Ouabain,
Sorafenib
tosylate,
Sunitinib malate,
Tioconazole,
Tribromsalan,
Triclabendazole,
Zafirlukast
39.8M. pharmaceutical collection
of 2816 small molecules
in vitro
Pyrvinium
pamoate
Antihelmintic Anti-tuberculosis – Alamar blue
assay MIC 0.31 M.
In vitro screen against
1514 known drugs –
many other previously
(6)
30
unidentified hits found.
Pyrvinium
pamoate
Antihelmintic Anti-protozoal –
Cryptosporidium parvum IC50
354nM.
In vitro screen for P.
Falciparum growth of
1937 FDA approved
drugs hypothesized to be
active due to confined to
intestinal epithelium.
(16)
Pyrvinium
pamoate
Antihelmintic Anti-protozoal – against T
Brucei IC50 3 M.
Screened 2160 FDA
approved drugs and
natural products from
Microsource. 15 other
drugs active IC50 0.2 – 3
M
(17)
Riluzole Amyotrophic Lateral
Sclerosis - inhibits
glutamate release and
reuptake
Riluzole enhanced Wnt/-
catenin signaling in both the
primary screen in HT22 neuronal
cells and in adult hippocampal
progenitor cells. Metabotropic
glutamate receptor GRM1
regulates Wnt/-catenin
signaling.
Screened 1857
compounds (1500
unique) in vitro treating
melanoma cells with
riluzole in vitro enhances
the ability of WNT3A to
regulate gene expression.
(46)
Closantel A veterinary Onchocerciasis, or river Screened 1514 FDA (4)
31
antihelmintic with
known proton
ionophore activities
blindness IC50 1.6 M
competitive inhibition constant
(Ki) of 468 nM.
approved drugs (JHCCL)
against the chitinase
OvCHT1 from O.
volvulus.
Nitroxoline Antibiotic used
outside USA for
urinary tract
infections.
Anti-angiogenic agent inhibits
Type 2 methionine
aminopeptidase (MetAP2) IC50
54.8nM and HUVEC
proliferation. Also inhibits
Sirtuin 1 IC50 20.2 M and
Sirtuin 2 IC50 15.5 M.
Screened 2687 FDA
approved drugs (JHCCL)
for inhibition of HUVEC
cells. Also found the
same compound in HTS
of 175,000 compounds
screened against
MetAP2. Also active in
mouse and human tumor
growth models.
(5)
Glafenine Analgesic Inhibits ABCG2 IC50 3.2 M
could be used with
chemotherapeutic agents to
counteract tumor resistance.
Screened FDA approved
drugs (JHCCL) with
bioluminescence imaging
HTS assay. Discovered
37 previously unknown
ABCG2 inhibitors.
(47)
Tiagabine Antiepileptic
(enhances gamma –
aminobutyric acid
Neuroprotective in N171-82Q
and R6/2 mouse models of
Huntington’s disease (HD).
Initial screen of NINDS
Microsource database of
drugs (1040 molecules)
(48)
32
activity). against PC12 cell model
of HD found nepecotic
acid which is related to
tiagabine.
Digoxin,
Ouabain,
Proscillardin A
Cardiac glycosides
used to treat
congestive heart
failure and arrthymia.
Anticancer – Inhibition of
hypoxia-inducible factor 1 IC50 <
400 nM
Screened 3120 FDA
approved drugs (JHCCL)
screened against reporter
cell line Hep3B-c1.
Digoxin also tested in
vivo xenograft models.
(13)
Ceftriaxone -lactam antibiotic Neuroprotection – Amyotrophic
lateral sclerosis (ALS) -
increasing glutamate transporter
(GLT1) expression
EC50 3.5 M. Other-lactams
also active.
Screen of NINDS
Microsource database of
drugs (1040 molecules)
against rat spinal cord
cultures followed by
immunoblot for GLT1
protein expression. Also
tested in ALS mouse
model, delaying neuron
loss, increased survival.
(49)
Flufenamic acid Non steroidal anti-
inflammatory drug
Familial amyloid
polyneuropathy – Inhibits
transthyretin.
Screening library not
described.
(50)
33
Chlorprothixene,
Dihydroergotami
-ne, Hycanthone,
Hydroxyprogeste
-rone,
Perhexiline,
Propafenone,
Thioridazine
Anti-psychotics,
vasoconstrictor,
vasodilator,
anhelmintic,
progestrogen,
antiarrhythmic
Antimalarial - EC50 1-3 M
against P. falciparum 3D7.
Used the Microsource
screening and killer
collections (2160
compounds). Many other
compounds identified
e.g. topical and IV drugs.
Perhexiline, propafenone
and thioridazine may be
the most useful
(9)
Methazolamide Carbonic anhydrase
inhibitor, diuretic
Glaucoma
Inhibitors of cyctochrome c
release with therapeutic potential
for Huntington’s disease (HD).
NINDS database of drugs
(1040 molecules),
Methazolamide used in
transgenic muse model of
neurodegeneration
resembling HD. 20 other
compounds (including
antibiotics) identified that
inhibit cytochrome c and
cross the blood-brain
barrier.
(51)
Aclacinomycin,
Mitoxantrone,
Antineoplastics,
Antifungal, steroidal
Selective glucocorticoid receptor
(GR) modulators.
NINDS database of drugs
(1040 molecules)
(14)
34
Ciclopirox
olamine, Rosolic
acid,
Pararosaniline,
Hydroxyprogeste
-rone caproate
Anthracyclines were inhibitors
of GR.
screened simultaneously
against 4 promoters,
followed by luciferase
assays.
Trifluoperazine,
Promethazine,
Clomipramine,
Fluphenazine,
Nortriptyline,
Thioridazine,
Mefloquine,
Desipramine,
Chlorpromazine,
Prochlorperazine
, Perphenazine,
Amitriptyline,
Amoxapine,
Maprotiline,
Mianserin,
Antidepressants,
antipychotics,
antihistamine,
antimalarial,
antiemetics, muscle
relaxant,
anticholinergic, anti
ulcer.
Inhibitors of mitochondrial
permeability transition (mPT),
for stroke.
NINDS database of drugs
(1040 molecules)
screened to find those
that delay mPT in
isolated rat liver
mitochondria. 23 out of
32 hits are approved for
human use and 4
molecules were approved
but no longer in use.
Promethazine protected
mice in vivo from
occlusion/ repurfusion.
Nortriptyline delayed
disease onset and
(8)
35
Cyclobenzaprine,
Imipramine,
Clozapine,
Doxepin,
Loratidine,
Thiothixene,
Propantheline,
Pirenzepine
mortality in ALS mice
and R6/2 mice (amodel
for HD) (52)
Fosfosal,
Levonordefrin,
Molsidomine,
Nadolol,
Gefitinib
NSAID, -adrenergic
agonist, nitric oxide
releasing prodrug, -
adrenergic receptor
antagonist
Neurodegeneration, reduce
polyglutamine aggregation and
toxicity for Huntington’s disease
and X-linked spinolbulbar
muscular atrophy
NINDS database of
drugs, annotated
compound library and
Kinase library (>4000
molecules) screened in
FRET assay of androgen
receptor aggregation,
follow up testing in
Drosophila. 5 other
compounds identified.
(53)
Fluspirilene,
Trifluoperazine,
Pimozide,
Nicardipine,
Niguldipine,
Antipsychotics,
Calcium channel
antagonists
(cardiovascular),
opiod receptor
Neurodegeneration – regulate
autophagy a target for
Huntington’s disease Alzheimers
disease
Biomol known bioactive
library (480 molecules)
screened against a human
glioblastoma cell line
using an image based
(18)
36
Loperamide,
Amiodarone
antagonist method and a long-lived
protein degradation
assay. Penitrem A was
also identified (non-FDA
approved).
Cytosine-
arabinoside
(ARA-C)
Nucleoside analog
that inhibits DNA
synthesis.
Ewing sarcoma targeting
EWS/FLI oncoprotein
NINDS database of drugs
(1040 molecules)
screened with a ligation-
mediated amplification
assay with a bead-based
detection. The study used
a gene expression
signature (14 genes) of
EWS/FLI off state. Also
showed efficacy in
mouse Ewing’s sarcoma
model.
(54)
5-Azacytidine,
Colchicine,
Dactinomycin,
Daunorubicin,
Mitoxantrone,
Paclitaxel,
Anticancer drugs
with different
mechanisms
Neuroblastoma (NB) Screened 96 drugs
against the SK-N-AS cell
line derived from a stage
4 neuroblastoma tumor.
Secondary screening in a
second NB cell line. 30
(10)
37
Teniposide,
Thioguanine,
Valrubicin
compounds active, 15
FDA approved (5
currently used for NB)
Digoxin,
Pyrithione zinc
Cardiac glycoside
used for treating heart
failure, antimicrobial
Retinoblastoma Microsource and
Prestwick libraries (2640
molecules) screened
against retinoblastoma
cell lines followed by
xenograft model of
retinoblastoma. 9 other
compounds identified
(12)
Amiodarone Class III
antiarrhythmic
Selective removal of
undifferentiated embryonic stem
cells (ESC) for cell replacement
therapy.
720 FDA approved drugs
from the NINDS
collection were tested in
ESC and neural stem
cells (NSC). 8 other
compounds identified as
hits that were selectively
toxic to NSCs by
reducing ATP levels.
Follow up in postmitotic
neurons.
(55)
Carvedilol, 2-adrenergic Prevention of hearing loss – NINDS database of drugs (56)
38
Phenoxybenzami
ne, Tacrine
blocker, diuretic, 1-
blocker,
anticholinergic
lowest dose tested that shows
protection is 10 M
(1040 molecules) tested
in zebrafish larvae to find
those that modulate
neomycin induced hair
cell toxicity. 4 other non
FDA approved molecules
were also active. Tacrine
was also protective in
mouse utricle explants.
Rimcazole,
Sertraline,
Tegaserod,
Roxatidine,
Paroxetine,
Indatraline
Monoamine signaling
modulators
Malignant glioma NIH clinical collection
(480 molecules) screened
using live cell imaging in
glioma neural stem cells.
32 other hits identified
including anticancer
compounds.
(57)
Trifluoperazine,
Prochlorperazine
, Fluphenazine
Antipsychotics Inhibitors of myosin-II
associated S100A4 - benign
tumors
400 FDA approved drugs
screened using a
fluorescent biosensor to
report on the calcium
bound. 9 other diverse
compounds had activity.
(15)
Clomipramine, Antidepressants, Suppression of glial fibrillary Prestwick and spectrum (7)
39
Amitriptyline,
Chlorprothixene,
Tamoxifen
citrate,
Amlodipine
anticancer, calcium
channel antagonist
acidic protein – CNS disorders
e.g. Alexander Disease
libraries (2880
molecules) screened for
suppression of GFAP. 5
other compounds active –
not FDA approved.
Reduction in GFAP
protein, also see in mice
in vivo using
clomipramine.
40