Download Rational selection of structurally diverse natural

Molecular Diversity (2005) 9: 131–139
c Springer 2005
Full-length paper
Rational selection of structurally diverse natural product scaffolds with favorable
ADME properties for drug discovery
D.S. Samiulla1,† , V.V. Vaidyanathan1,† , P.C. Arun1 , G. Balan1 , M. Blaze1 , S. Bondre1 , G. Chandrasekhar1 ,
A. Gadakh1 , R. Kumar1 , G. Kharvi1 , H.-O. Kim2 , S. Kumar1 , J.A. Malikayil1 , M. Moger1 , M.K. Mone1 ,
P. Nagarjuna1 , C. Ogbu2 , D. Pendhalkar1 , A.V.S. Raja Rao1 , G. Venkateshwar Rao1 , V.K. Sarma1 ,
S. Shaik1 , G.V.R. Sharma1 , S. Singh1 , C. Sreedhar1 , R. Sonawane1 , U. Timmanna1 & L.W. Hardy2,∗
1Aurigene
Discovery Technologies, Ltd., Electronic City Phase 2, Hosur Road, Bangalore-562158, India; 2 Aurigene
Discovery Technologies, Inc., 99 Hayden Avenue, Lexington, Massachusetts 02420, U.S.A.
(∗Author for correspondence, E-mail: [email protected], Tel: 508-347-5320, Fax: 978-685-3734)
Received 16 March 2004; Accepted 14 May 2004
Key words: computational design, early ADME, lead-like, metabolic stability, natural products, permeability, scaffold
Summary
Natural product analogs are significant sources for therapeutic agents. To capitalize efficiently on the effective features of
naturally occurring substances, a natural product-based library production platform has been devised at Aurigene for drug
lead discovery. This approach combines the attractive biological and physicochemical properties of natural product scaffolds,
provided by eons of natural selection, with the chemical diversity available from parallel synthetic methods. Virtual property
analysis, using computational methods described here, guides the selection of a set of natural product scaffolds that are both
structurally diverse and likely to have favorable pharmacokinetic properties. The experimental characterization of several in
vitro ADME properties of twenty of these scaffolds, and of a small set of designed congeners based upon one scaffold, is also
described. These data confirm that most of the scaffolds and the designed library members have properties favorable to their
utilization for creating libraries of lead-like molecules.
Abbreviations: ADME, absorption-distribution-metabolism-excretion; BCUT, Burden-Chemical Abstract Services – University
of Texas; DMSO, dimethylsulfoxide; e-ADME, early ADME; HPLC/UV, high performance liquid chromatography monitored by ultraviolet absorbance; LC-MS, liquid chromatography–mass spectrometry; MDCK, Madin-Darby canine kidney;
NCI, National Cancer Institute; PBS, phosphate buffered saline; PSA, polar surface area; TEER, transepithelial electrical
resistance
Introduction
Natural products have historically provided important and
effective therapeutic agents [1]. Aspirin, penicillin, and paclitaxel are just three of numerous well-known natural products used to improve human health. Moreover, analogs of
such natural products as benzylpenicillin and semisynthetic
drugs, such as taxotere, have proven to be enormously fruitful for new drug discovery. Recent studies have shown that
natural compounds and their analogs continue to be major
sources of new drugs [2, 3]. Collections of natural products
exhibit physicochemical property profiles that are favorable
compared to those of drugs and complementary to those
† These
authors have contributed equally to this work.
provided by synthetic compounds that derive from combinatorial chemistry [4, 5].
Despite these advantages, the classical processes to identify discrete new chemical entities from natural product
sources are too inefficient to have survived in many of the current discovery environments at pharmaceutical companies.
However, interest in natural products and their analogs as
sources of pharmaceutical agents has shown recent resilience
[6–8]. To capitalize more efficiently on the effective features
of naturally occurring substances, we have adopted a natural
product-based library production platform for lead discovery.
This approach combines the attractive biological and physicochemical properties of natural product scaffolds, provided
132
by eons of natural selection, with the chemical diversity available from parallel synthetic methods. The selection of natural
product scaffolds is guided by several criteria. Three major
criteria are: (1) computational analyses of virtual properties
calculated from the structures of the small molecule scaffolds
alone, (2) the density of patented entities in “chemical space”
around the each scaffold being considered, and (3) accessibility (by either total synthesis or extraction from a renewable source). The virtual property analysis considers both the
diversity for the scaffolds set as a whole (e.g., how well the
scaffolds complement each other in “chemical diversity”) and
the structural diversity accessible for each particular scaffold
(either by total synthesis of analogs or by derivatization of a
scaffold with multiple potential positions for modification).
Further, the virtual property assessment includes a consideration of the “lead-like” nature of the scaffolds, so that a
good starting point exists for lead generation and optimization. The goal of this process is to define a scaffold set that
is structurally diverse and contains members likely to have
favorable pharmacokinetic properties.
At present, fifty-one natural product scaffolds have been
selected, using the process broadly outlined above, for further exploration. Three examples are cryptoheptine, vasicinone, and Z-guggulsterone (Figure 1A). Z-Guggulsterone,
the active constituent of an ayurvedic medicine used in India for several thousand years, is especially exciting. ZGuggulsterone has clinically proven benefits in the treatment
of human hyperlipidemia, and is approved and marketed for
that purpose in India [9]. It was recently proven, using gene
knockout methods, that Z-guggulsterone targets the farnesoid
X receptor (also known as the bile acid receptor) in its effects
on cholesterol metabolism in the mouse [10].
The production of small molecule libraries based on
these natural product scaffolds is cost-effective, efficient,
expedient, and amenable to solid-phase and parallel solutionphase synthesis. For example, a small library of vasicinone
analogs was chosen for synthesis from a virtual library to provide a structurally diverse set with acceptable ADME parameters. Several members of the vasicinone library are shown
in Figure 1B 1 . The computational analysis of the vasicinone
library is described here. For target-driven drug discovery,
the designs of small molecule libraries based on the natural
product scaffolds are ideally guided not by diversity metrics primarily, but by the atomic structures of the targets.
Design of our targeted libraries is based both on in silico
estimates of the ADME properties of the virtual libraries
calculated solely from the small molecule structures and by
computational screening of virtual libraries against structural
models of specific proteins that are targeted for therapeutic
discovery.
Major reasons for the failure of early drug candidates to
reach the market include inappropriate ADME (absorption,
distribution, metabolism and excretion) properties of the candidates, and metabolism-based adverse interactions of the
candidates with existing drugs. By some estimates, 40–50%
of new chemical entities in the drug discovery pipeline fail
due to poor ADME properties [11, 12]. From a commercial
perspective, poorly behaved compounds need to be recognized and removed as early as possible in the discovery process, rather than during the more costly clinical testing and
development phase. To speed the provision of the data needed
for the design and discovery of well-behaved drug candidates, “early ADME” (e-ADME) assays have been developed
[13–15]. These assays use in vitro surrogates for in vivo
(A)
(B)
Figure 1. (A) Representative natural product scaffolds, (B) Representative analogs of vasicinone.
133
physiology, and allow information about ADME and
metabolism-based adverse drug-drug interactions to be incorporated early in drug discovery. The e-ADME parameters of twenty of our natural product scaffolds have been
characterized. The properties studied were aqueous solubility at pH 7.4, inhibition of several recombinant human cytochrome P450 isozymes, stability to exposure to pooled human hepatic microsomes, and permeability across the MDCK
(Madin-Darby Canine Kidney) cell monolayer. Incorporation
of early ADME assessments in our drug discovery projects
has helped us to identify natural product scaffolds and natural product analogs with high probability of good intestinal absorption and to flag potential problems arising from
hepatic metabolism. This approach allows potential problems with poor absorption or metabolic interactions to be
avoided by re-design of analogs at the discovery stage in a
cost-effective way.
Methods
the various structural fragments present were calculated [19].
The Tanimoto similarity index was defined as:
T ( f 1 , f 2 ) = N1&2 /N1|2
where, f1 = fingerprint 1, f2 = fingerprint 2, N1&2 = number of fields present in both f1 and f2, N1|2 = number of
fields present in either f1 or f2.
A variety of structure-based parameters, relevant to
ADME characteristics such as those described by Lipinski
[20] and by Veber and co-workers [21], were also evaluated
for the compounds. The parameters calculated were ClogP
(calculated using Sybyl from Tripos), molecular weight and
number of rotatable bonds (assessed manually), and number of hydrogen bond donors and acceptors (assessed using
ChemDraw for Excel). In the calculation of the number of
rotatable bonds, amide bonds are not counted, since the rotational barrier should prevent free rotation. The polar surface area (PSA) was calculated using the MolCad module of
Sybyl.
Computational property and diversity analyses
General analytical methods
The BCUT parameter approach, pioneered by Pearlman [16],
was employed as provided in the software package from
Tripos [17]. To establish statistically valid orthogonal axes
of the chemistry vector space to use for the diversity analysis
of natural product scaffolds, BCUT parameters were generated for about 25,000 molecules from the National Cancer
Institute (NCI) database, as provided by Tripos. Correlation
between the axes was calculated and a chemistry space was
defined with minimal r 2 (<0.25) values between the axes.
The diversity of scaffold molecular structures was compared
in this chemistry space based on partitioning (cell based)
methods.
Evaluation of the vasicinone library for diversity was done
differently from the evaluation of the scaffolds. Relative diversity is important for a library with respect to the scaffold
[18]. Therefore, the BCUTs for the members of the vasicinone library were generated. Chemistry space definitions
were generated which maintained the r 2 (between the axes)
less than 0.25. As the number of library members was low,
it would not be statistically valid to assume the axis to be
orthogonal. Hence, the same BCUTs were generated for the
NCI set of molecules, and r 2 between the axes were evaluated. The chemistry space with r 2 (between the axes) less
than 0.25 was then accepted, and was partitioned into cells
such that the occupancy was 14.8%. Statistics of occupancy
for the compounds were calculated. The chemistry spaces in
both the cases are described as: two partial charges based
BCUTs, one H-donor based BCUT, one H- acceptor based
BCUT, and two polarizability based BCUTs.
A molecular fingerprint-based analysis was also performed for both the scaffolds and the vasicinone library members, using 992 bit Unity 2D fingerprint definitions. The
fingerprints were generated, and the correlations between
Fluorescence measurements were done in 96-well plates
using a Spectromax Gemini XS reader (Molecular Devices, USA). Transepithelial resistance (TEER) measurements were made using a World Precision Instrument
(Florida, USA) probe. HPLC analyses (performed on an
Agilent Technologies 1100 series system) were done using
a reverse phase column (Zorbax Eclipse XDB C18, 150 ×
4.6 mm, 3.5 micron), with quantitation based on UV absorbance at 254 nm. The LC-MS analyses were performed in
multiple reaction monitoring mode using an Applied Biosystems API 2000, fed by a in-line Agilent Technologies 1100
series HPLC using the same model of Zorbax reverse phase
column for sample chromatography.
Aqueous solubility assay
DMSO solutions of test compounds were added to phosphate
buffered saline pH 7.4 (PBS) in a 96 deep-well plate to generate theoretical concentrations of 200 µM in 2% DMSO.
The solutions were equilibrated by shaking (200 rpm, Ika
plate shaker) for 16 h at 25 ◦ C. Undissolved compound was
removed by centrifugation, and the supernatant was analyzed
by HPLC-UV or LC/MS/MS. Aqueous solubility was calculated using the equation:
Aqueous solubility = 200 µM × PAPBS /PADMSO
where PAPBS and PADMSO are the peak areas from the analyses of test compound in PBS with 2% DMSO and of test
compound in 100% DMSO, respectively.
Assays were performed in duplicate or triplicate.
134
MDCK monolayer permeability assay
Madin-Darby Canine Kidney (MDCK) cells (American Type
Culture Collection, Manassas, Virginia, USA) were used to
assess compound permeability [22]. The cells were grown in
Dulbecco’s modified Eagle medium supplemented with 10%
fetal bovine serum, 1 mM non-essential amino acids, 1 mM
sodium pyruvate and gentamicin sulphate (50 µg/mL) to 70–
80% confluency prior to seeding in 24-well plates loaded with
polycarbonate Millicell inserts (Millipore, 12 mm diameter,
0.4 µm, 50,000 cells/insert). Cells were fed with fresh
medium every other day and incubated at 37 ◦ C, with 5% CO2
for 3 days prior to drug transport experiment. Cell monolayer
integrity was assessed by measuring TEER values. Drugs
were applied at 50 µM in Hank’s buffered salt solution containing 0.5% DMSO to the apical chamber and the transport
assay was carried out for 2 h at 37 ◦ C. At the end of the
assay, samples from both apical and basal chambers were
collected for analysis by HPLC/UV or LC/MS/MS, and the
monolayer integrity was re-assessed by dye rejection using
Lucifer Yellow. Data from wells with dye rejection less than
98% were rejected. Apparent permeability (Papp ) values were
calculated using the equation:
BD Gentest): 7-methoxy-4-(trifluoromethyl) coumarin for
CYP2C9, 3[2-N,N-diethyl-N-methylammonium)ethyl]-7methoxy-4-methylcoumarin for CYP2D6, and 7-benzyloxy4-(trifluoromethyl)coumarin for CYP3A4. All assays were
performed in duplicate.
Results and discussion
To understand chemical diversity for sets of compounds
to be used for chemical biology, that diversity must be
quantified in a manner relevant to the biological role for
the compounds. The methods used to quantify diversity are
“fingerprint” based approaches and “chemistry space” based
approaches. In fingerprint-based approaches, bit strings
Papp = d Q/dt × 1/Co × 1/A
where, dQ/dt = permeability rate in µg/sec, Co = initial concentration in µg/mL, A = membrane surface area (0.6 cm2
for 12 mm inserts), Papp values were expressed in cm/sec.
All the assays were performed in triplicate.
Human hepatic microsomal stability assay
Figure 2. Distribution of the cell occupancy for the vasicinone library
members.
Pooled human liver microsomes were employed to assess the
potential instability of compounds to phase I metabolism [23,
24]. The microsomes (BD Gentest) were incubated with test
compounds (1–5 µM in 0.2% DMSO 1 buffer) at 37 ◦ C for
30 min. The reaction was stopped by the addition of acetonitrile containing haloperidol as internal standard. Precipitated
protein was removed by centrifugation and the supernatants
were analyzed by HPLC-UV or LC/MS/MS method. Stability was assessed by the disappearance of compound based on
the change in analyte to internal standard peak height ratio.
Metabolic stability was defined as the amount of test compound remaining after the incubation with human hepatic
microsomes, and expressed as a percentage of the initial concentration. All assays were performed in triplicate.
Human cytochrome P450 inhibition assay
Human recombinant Cytochrome P450 isozymes 2C9, 2D6
and 3A4 (BD Gentest) were incubated with 5 µM of test
compound in buffer containing 0.1% DMSO for 10 min at
37 ◦ C. The residual enzyme activity was measured [15] using
the following fluorogenic substrates (also obtained from
Figure 3. Scatter plot of molecular weight, ClogP, and PSA for natural
product scaffolds.
135
(molecular fingerprints) that indicate the presence of substructural features in each molecule are employed. Similarity
and dissimilarity between two molecules are measured as
the inner product of these bit string vectors, or a closely
related quantity called Tanimoto similarity/dissimilarity
indices [19]. This kind of approach is helpful in designing
combinatorial libraries as it checks for presence of different
kinds of fragments. However, with a collection of natural
product scaffolds, this method is not relevant for defining a
library, which is diverse in a general sense.
The chemistry space approach relies on the definition
of a set of quantifiable chemical descriptors, corresponding
to dimensions within chemistry space. As the position of
a particle in three dimensions is defined by its x-, y-, and
z-co-ordinates, the positions of molecules in N-dimensional
chemical space are defined by N chemical descriptors. A similar approach was exemplified by Pearlman and co-workers
in their analysis of acetylcholinesterase inhibitors [16]. The
axes of the chemistry space are BCUTs, which are eigenvalues of the connectivity matrices of the molecular graphs
with some properties as the diagonal elements. Pearlman and
co-workers found that three of the six BCUT metrics, they
had identified from an analysis of a database of 60,000 ‘druglike’ compounds, were receptor relevant, i.e. the actives were
clustered in these dimensions. We have employed this same
approach to identify scaffolds that are diverse in a biologically relevant chemical space, as defined by the diversity in
the compounds in the NCI database.
The natural product scaffolds are diverse in the chemical
space, since the different scaffolds are found in different cells.
Figure 4. Scatter plot of number of rotatable bonds, H bond acceptors, and
H bond donors for natural product scaffolds.
Figure 6. Scatter plot of number of rotatable bonds, H bond donors, and H
bond acceptors for the vasicinone library.
Figure 5. Scatter plot of molecular weight, ClogP, and PSA for the vasicinone library.
Figure 7. Aqueous solubility of standard drugs in solubility assay correlated
with published data.
136
As the distribution of partial charges, polarizability, and Hbonding properties of the structures are mapped by BCUTs
based upon them, these scaffolds are diverse in the distribution of these properties. The variety of distribution of these
properties in the natural product scaffolds gives them a better
chance to be selected by varied types of receptors.
The bits in the fingerprints indicate different structural
fragments present in the scaffolds. The fact that only 62
bits were highly correlated out of 992 indicates that these
scaffolds have different chemical cores in them. Thus, the
fingerprints based approach indicates a statistically significant difference in the types of chemical classes present in the
scaffolds. This conforms to the chemically intuitive definition
of the diversity of natural product scaffolds [4, 5]. Not surprisingly, the vasicinone library was considerably less diverse
than the Aurigene collection of natural product scaffolds, with
multiple compounds in many cells (cf., Figure 2).
As shown in Figures 3 and 4, the Lipinski parameters
for the natural product scaffolds satisfy “the rule of five”
limits [20]. The number of rotatable bonds is less than seven
for all the molecules, which is well within the maximum of
10 suggested for good bioavailability [21]. The maximum
PSA for these scaffolds is 310 Å2 . The acceptable maximum
fragment-based PSA is 140 Å2 , compounds with PSA
values above this limit are expected to have low oral
bioavailability [14]. In our experience, the PSA calculated
Figure 8. Permeability in MDCK model system correlates well with known human absorption data for orally dosed drugs.
Figure 9. Cytochrome P450 inhibition assay with standard inhibitors.
137
Table 1. ADME properties of natural product scaffolds
Solubility at pH
7.4, 10−6 M
Permeability through
MDCK monolayer (Papp ),
10−6 cm/s (±SD)
1 (Vasicinone)
>200
74 (±3)
2 (Cryptoheptine)
insoluble
nd
3
40
6.5 (±0.8)
4
89.7
60 (±2.4)
5
196.2
7.5 (±0.2)
6 (Z-Guggulsterone)
94
33.1 (±0.7)
7
195.9
68 (±4)
8
>200
83 (±3)
9
1.9
nd
10
191
19.1 (±0.4)
11
80
79 (±7)
12
>200
62 (±4)
13
197.5
113 (±8)
14
8.7
70.9 (±0.1)
15
>200
69 (±8)
16
6.1
nd
17
178.5
57 (±3)
18
>200
5.9 (±0.5)
19
>200
46 (±4)
20
13.5
10.5 (±0.8)
Scaffold
Values are mean (±SD) of triplicate assays. Green = Good, Blue = Acceptable, Red = May need to be modified by analog synthesis. nd = Not
determined.
using the MolCad method with the energy minimized
structures (as implemented by Tripos) is 2.5 times the value
calculated by the faster fragment-based approach [26].
Therefore, the acceptable limit for the PSA value computed
with MolCad is 350 Å2 . Thus, there are no computational
alerts for these molecules for solubility, permeability, and
oral bioavailability.
Distributions of the ADME relevant computed properties for the vasicinone library are shown in Figures 5 and 6.
Among the 172 molecules in the vasicinone library, there are
five molecules with molecular weight greater than 500, and
one molecule has a ClogP value greater than 5. The MolCad computed PSA has a maximum value of 311 Å2 for the
library. The maximum values observed for the numbers of
H-Bond donors and H-Bond acceptors are 4 and 6, respectively. The maximum number of rotatable bonds for the any
member of the library is 10. Therefore, six molecules have
computational alerts.
The experimental assays for the e-ADME parameters
were validated using drugs and other well characterized compounds with known values for solubility (Figure 7), permeability (Figure 8), inhibitory potency toward cytochrome
P450s (Figure 9), and metabolic stability (Figure 10). These
validated in vitro assays were then used to assess the eADME parameters (solubility, permeability, inhibition of cytochrome P450 isozymes, and hepatic microsomal stability)
for twenty of the natural product scaffolds, and for a subset of
compounds in the vasicinone library. Reference compounds
with low, medium, and high values for the assayed property are routinely included during the assessments of test
compounds to assure the acceptable performance of these
assays.
Most of the scaffolds and vasicinone library members had
acceptable or good aqueous solubility and permeability parameters (Tables 1 and 2), which supports the accuracy of
the favorable computational estimates of their physicochemical properties. The majority of the natural product scaffolds
have excellent microsomal stability, with more than 50% of
the parent compound surviving the 30-min exposure for 18
of the 20 scaffolds examined (Figure 11). Most of the natural
product scaffolds and vasicinone analogs also have acceptably weak or moderate inhibitory potencies towards the three
cytochrome P450 isozymes that were assayed (Table 2 and
Figure 12), which are the prevalent hepatic P450 isozymes in
Table 2. ADME properties of Vasicinone analogs
Solubility, 10−6 M
Papp, 10−6 cm/s
3A4 % inhibition
2C9 % inhibition
2D6 % inhibition
1
>200
69 (±8)
34
43
14
2
>200
87 (±4)
34
26
33
3
>200
67 (±2)
32
25
16
4
>200
75 (±4)
30
27
19
5
>200
46 (±2)
31
22
24
6
>200
65 (±2)
31
32
6
7
>200
64 (±17)
31
29
19
8
>200
73 (±5)
31
36
23
9
>200
35 (±2)
49
66
16
10
>200
2.2 (±0.6)
31
44
7
11
11.6
nd
32
32
17
12
>200
97 (±1)
26
36
24
Vasicinone analogs
Green = Good, Blue = Acceptable, Red = requires modification by analog synthesis. nd = Not determined.
138
Figure 10. Microsomal stability assay with standard compounds.
Figure 11. Microsomal stability of natural products scaffolds.
Figure 12. Inhibition of CYP450 isozymes by natural product scaffolds.
139
most humans [27]. Generally, the natural product scaffolds
have good e-ADME properties. This strongly supports their
suitability for library generation for drug discovery. Preservation of the favorable e-ADME properties during library
buildouts starting with the scaffolds is suggested by the good
e-ADME properties exhibited by the vasicinone library members. A few scaffolds have marginal e-ADME properties, indicating that if they are pursued, these properties will likely
require modification during library build-outs.
In summary, the computational and experimental assessments of the properties of the natural product scaffolds are
in agreement. These findings support the utility of the rational process that we have established to choose scaffolds for
hit and lead generation libraries for drug discovery. Computational methods clearly have great utility to combine good
physicochemical properties (and thereby increase the likelihood of favorable bioavailability) and structural diversity in
libraries of natural product analogs. We are pursuing a similar approach in the design, synthesis, and characterization
of small, target-focused libraries of natural product analogs
with good physicochemical properties, guided by computational methods and the structural templates of therapeutic
target proteins.
Note
1. The chemical structures of most of the scaffolds and vasicinone library
members cannot be disclosed for proprietary reasons.
References
1. Samuelsson, G., Drugs of natural origin, Swedish Pharmaceutical Society, Stockholm, Sweden, 1991.
2. Buss, A.D., Cox, B. and Waigh, R.D., Natural products as leads for new
pharmaceuticals, In Abraham, D.J. (Ed), Burger’s Medicinal Chemistry and Drug Discovery, Sixth Edition, Vol. 1, Wiley Interscience,
New York, 2003, pp. 847–900.
3. Newman, D.J., Cragg, G.M. and Snader, K.M., Natural products as
sources of new drugs over the period 1981–2002, J. Nat. Prod., 66
(2003) 1022–1037.
4. Feher, M. and Schmidt, J.M., Property distributions: Differences between drugs, natural products, and molecules from combinatorial
chemistry, J. Chem. Inf. Comput. Sci., 43 (2003) 218–227.
5. Henkel, T., Brunne, R.M., Muller, H. and Reichel, F., Statistical investigation into the structural complementarity of natural products and
synthetic compounds, Angew. Chem. Int., 38 (1999) 643–647.
6. Rouhi, A.M., Rediscovering natural products, Chem. Eng. News, 81
(2003) 77–91.
7. Rouhi, A.M., Betting on natural products for cures, Chem. Eng. News,
81 (2003) 93–103.
8. Rouhi, A.M., Moving beyond natural products, Chem. Eng. News, 81
(2003) 104–107.
9. Urizar, N.L. and Moore, D.D., Gugulipid: A natural cholesterollowering agent, Annu. Rev. Nutr., 23 (2003) 303–313.
10. Urizar, N.L., Liverman, A.B., Dodds, D.T., Silva, F.V., Ordentlich, P.,
Yan, Y., Gonzalez, F.J., Heyman, R.A., Mangelsdorf, D.J. and Moore,
D.D., A natural product that lowers cholesterol as an antagonist ligand
for FXR, Science, 296 (2002) 1703–1706.
11. van de Waterbeemd, H. and Gifford, E. ADMET in silico modelling: Towards prediction paradise? Nat. Rev. Drug. Discov., 2 (2003) 192–204.
12. Kennedy, T., Managing the drug discovery/development interface,
Drug Discov. Today, 2 (1997) 436–444.
13. Stoner, C.L., Cleton, A., Johnson, K., Oh, D.M., Hallak, H.,
Brodfuehrer, J., Surendran, N. and Han, H.K., Integrated oral
bioavailability projection using in vitro screening data as a selection tool in drug discovery, Int. J. Pharm., 268 (2004) 241–
249.
14. Yu, H., and Adedoyin, A., ADME-Tox in drug discovery: Integration
of experimental and computational technologies, Drug Discov. Today,
8 (2003) 852–861.
15. Crespi, C.L., Miller, V.P. and Stresser, D.M., Design and application of
fluorometric assays for human cytochrome P450 inhibition, Methods
Enzymol., 357 (2002) 276–284.
16. Pearlman, R.S. and Smith, K.M., Metric validation and the receptorrelevant subspace concept, J. Chem. Inf. Comput. Sci., 39 (1999)
28–35.
17. Diverse Solutions Users Manual version 3.0.1: Laboratory of
Molecular Graphics and Theoretical Modeling, University of Texas,
Austin,Texas copyrighted 1996, 1997, Distributed by Tripos Inc.,
1699 S. Hanley Road, St. Louis, MO 63144 (1996, 1997).
18. Mason, J.S., Absolute versus relative similarity and diversity, In
Dean, P.M., and Lewis, R.A. (Eds ) Molecular Diversity in Drug
Design, Kluwer Academic Publishers, Dordrecht (1999), pp.88–
90.
19. Holliday, J.D., Ranade, S.S. and Willett, P., A fast algorithm for
selecting sets of dissimilar molecules from large chemical databases,
Quant. Struct.-Act. Relat., 14 (1995) 501–506.
20. Lipinski, C.A., Drug-like properties and the causes of poor solubility
and poor permeability, J. Pharmacol. Toxicol. Methods, 44 (2000)
235–249.
21. Veber, D.F., Johnson, S.R., Cheng, H.Y., Smith, B.R., Ward, K.W. and
Kopple, D.K., Molecular properties that influence oral bioavailability
of drug candidates, J. Med. Chem., 45 (2002) 2615–2623.
22. Irvine, J.D., Takahashi, L., Lockhart, K., Cheong, J., Tolan, J.W.,
Selick, H.E. and Grove, J.R., MDCK (Madin-Darby canine kidney)
cells: A tool for membrane permeability screening, J. Pharm. Sci., 88
(1999) 28–33.
23. Iwatsubo, T., Hirota, N., Ooie, T., Suzuki, H., Shimada, N., Chiba, K.,
Ishizaki, T., Green, C. E., Tyson, C.A. and Sugiyama, Y., Prediction of
in vivo drug metabolism in the human liver from in vitro metabolism
data. Pharmacol. Ther., 73 (1997) 147–171.
24. Mandagere, A.K., Thompson, T.N. and Hwang, K.K., Graphical
model for estimating oral bioavailability of drugs in humans and
other species from their Caco-2 permeability and in vitro liver
enzyme metabolic stability rates, J. Med. Chem., 45 (2002) 304–
311.
25. Palm, K., Stenberg, P., Luthman, K. and Artursson, P., Polar molecular surface properties predict the intestinal absorption of drugs in
humans, Pharm. Res., 14 (1997) 568–571.
26. Clark, D.E., Rapid calculation of polar molecular surface area
and its application to the prediction of transport phenomena. 1.
Prediction of intestinal absorption, J. Pharm. Sci., 88 (1999) 807–
814.
27. Yan, Z. and Caldwell, G.W., Metabolism profiling, and cytochrome
P450 inhibition and induction in drug discovery, Curr. Top. Med.
Chem., 1 (2001) 403–425.