Optimizing solubility and permeability of a biopharmaceutics classification system (BCS) class 4 antibiotic drug using lipophilic fragments disturbing the crystal lattice

Article abstract of DOI:10.1021/jm301721e

Esterification was used to simultaneously increase solubility and permeability of ciprofloxacin, a biopharmaceutics classification system (BCS) class 4 drug (low solubility/low permeability) with solid-state limited solubility. Molecular flexibility was i

Full text of DOI:10.1021/jm301721e

Brief Article
pubs.acs.org/jmc
Optimizing Solubility and Permeability of a Biopharmaceutics
Classification System (BCS) Class 4 Antibiotic Drug Using Lipophilic
Fragments Disturbing the Crystal Lattice
Ulrika Tehler,†,§ Jonas H. Fagerberg, Richard Svensson, Mats Larhed, Per Artursson,
†
†
‡
†
,
†
̈
and Christel A. S. Bergstrom*
†
Uppsala University Drug Optimization and Pharmaceutical Profiling Platform, Department of Pharmacy, Uppsala Biomedical
Center, Uppsala University, P.O. Box 580, SE-751 23 Uppsala, Sweden
‡
Organic Pharmaceutical Chemistry, Department of Medicinal Chemistry, Uppsala Biomedical Center, Uppsala University, P.O. Box
74, SE-751 23 Uppsala, Sweden
5
*
S Supporting Information
ABSTRACT: Esterification was used to simultaneously increase solubility and permeability of ciprofloxacin, a biopharmaceutics
classification system (BCS) class 4 drug (low solubility/low permeability) with solid-state limited solubility. Molecular flexibility
was increased to disturb the crystal lattice, lower the melting point, and thereby improve the solubility, whereas lipophilicity was
increased to enhance the intestinal permeability. These structural changes resulted in BCS class 1 analogues (high solubility/high
permeability) emphasizing that simple medicinal chemistry may improve both these properties.
INTRODUCTION
permeability of this compound could be improved through
esterification producing derivatives with less strong crystal
lattice and improved lipophilicity and that these structural
features possibly position the ester derivatives in BCS class 1.
■
Poor aqueous solubility is a main hurdle to overcome in current
discovery programs with 40−70% of all new hits estimated to
have too low solubility to allow complete absorption from the
1
GI tract. We have investigated molecular features of
RESULTS
2
■
importance for poor solubility and identified structural
Compounds. Seven ciprofloxacin analogues (Figure 1, 5a−
g) were synthesized by esterification of ciprofloxacin. By
differences between compounds that are poorly soluble because
of strong crystal lattice and those that are poorly hydrated in
the aqueous environment. The molecules that are solid state
limited in their solubility, are smaller, rigid, and often flat
2
c
molecules. These structural features allow the molecules to
pack densely in the crystal and provide strong intermolecular
bonds through van der Waals interaction, π−π stacking, and
hydrogen bond formation. In comparison, the solvation limited
2a
compounds are larger, flexible, and lipophilic. In addition
there are intermediate compounds that have both a strong
crystal lattice and a poor solvation. Typically these are highly
lipophilic, small, and rigid molecules with a strong hydrogen
3
bond capacity within the crystal lattice. The underlying
molecular reason(s) for poor solubility can hence inform the
medicinal chemists upon strategies to improve the aqueous
solubility.
Solubility together with permeability form the basis for the
biopharmaceutics classification system (BCS). The BCS sorts
Figure 1. Stepwise scheme of the synthesis process with alkyl chains
used given as 5a−g.
4
compounds into four classes based on high/low permeability
and solubility, in which BCS class 1 represents highly
permeable and highly soluble compounds that will be well-
absorbed after oral administration. In contrast, class 4
compounds have both low solubility and low permeability. In
this study we investigated whether classical medicinal chemistry
using esterification could be used to simultaneously improve
solubility and permeability for ciprofloxacin, a rare model
compound reflecting BCS class 4 drugs with solid-state limited
solubility. Our hypothesis was that the solubility and
i
elongation of the alkyl chains in a stepwise manner (Me, Et, Pr,
Pr, Bu, Hex, and Bn as an example of a more rigid ester; see
Figure 1), a systematic study of the effect of esterification and
molecular flexibility on the lipophilicity, melting point,
solubility, and permeability could be undertaken. These
Received: November 22, 2012
Published: February 25, 2013
©
2013 American Chemical Society
2690
dx.doi.org/10.1021/jm301721e | J. Med. Chem. 2013, 56, 2690−2694

Journal of Medicinal Chemistry
Brief Article
structural modifications of ciprofloxacin produced derivatives
with increased flexibility and a lipophilicity range of almost
ciprofloxacin. For the more lipophilic derivatives (5f and 5g)
the solubility was significantly reduced compared to cipro-
floxacin, with the compounds having 10- to 30-fold lower
solubility in all media (Figure 3). A significant solubilization in
lipid aggregates naturally available in the intestine, as simulated
with FaSSIF, was only seen for the most lipophilic compound
(5f), i.e., the hexyl derivative with a log P of 3.3 (log DpH6.5 of
1.1; see Figure 2). This is in agreement with our previous
finding that for compounds with a log P of 3 or more,
significantly increased apparent solubility can be expected in
100 000-fold (Table S1). Further, the esterification eliminated
the zwitterionic nature of ciprofloxacin, resulting in basic
derivatives with a single pK of ∼8.6 and reduced hydrogen
a
bond capacity.
Solid State Characterization. Differential scanning
calorimetry (DSC) thermograms showed that only one
polymorph was obtained of each of the derivatives synthesized.
No indications of salt formations or solvent residues were
obtained, and the final product obtained was the free base. The
melting point of the derivatives ranged from 255 °C (5a, Me
ester) to 186 °C (5f, Hex ester) in comparison to the melting
point of 266 °C for ciprofloxacin. The melting temperatures of
2a
intestinal fluids compared to pure water or buffers.
The apparent solubility was used to calculate the dose
4
number (Do). Completely dissolved compounds have Do < 1.
Compound 5a resulted in complete dissolution of the
maximum dose in all media, and so did 5e when measured in
i
the unbranched alkyl chain esters, e.g., all derivatives except Pr
PhB . In comparison, ciprofloxacin obtained Do > 7 in all
and Bn esters, decreased with increased chain length (Figure 2).
6.5
three media. Hence, the esterification using methyl and butyl
alkyl chains transform ciprofloxacin to become soluble under
physiologically relevant conditions. The resulting different
apparent solubility measurements were not considered a result
of different particle size, since the particle size distribution of
derivatives was similar to that of ciprofloxacin.
Permeability. All derivatives showed higher permeability
than ciprofloxacin (Table 1) and were predicted to be >90%
absorbed (providing absorption is not solubility-limited) based
on our in-house correlation between Caco-2 permeability and
5
fraction absorbed. Several of the compounds showed
significantly higher permeability in the basolateral to apical
(
b−a) direction than in the apical to basolateral (a−b)
direction, e.g., derivative 5d displayed 146-fold higher
permeability in the b−a direction. This large ratio cannot
solely be explained by the ion trapping effect caused by the pH
Figure 2. Calculated lipophilicity (log DpH 6.5 from ADMET Predictor,
circles) and melting point (triangles) for derivatives. Standard
deviation of melting point is <2 °C for all derivatives and hidden by
the symbol.
6
gradient used in the experiment and hence suggests that active
efflux transporters are involved in the cellular permeation. This
ratio was significantly reduced when the derivatives were
coadministered with the P-gp inhibitor GF120918. Hence, it
was concluded that P-gp contributed to the large b−a/a−b
ratio observed.
BCS. BCS of the derivatives is shown in Figure 4. At the pH
of the intestinal fluid (pH 6.5) the 5a and 5e esters transform
ciprofloxacin from BCS class 4 to BCS class 1. All other
derivatives were BCS class 2 compounds for which appropriate
formulation may result in a BCS 1 performance in vivo.
Apparent Solubility. The chemical modifications led to
higher solubility for three of the derivatives of ciprofloxacin
(
Figure 3). Derivative 5a showed 48-, 66-, and 111-fold higher
solubility at pH 7.4 (PhB ), at pH 6.5 (PhB ), and in the
7
.4
6.5
fasted state simulated intestinal fluid (FaSSIF, pH 6.5). 5b
showed 2- to 3-fold higher solubility in the three media,
whereas 5e had 4-, 27-, and 10-fold higher solubility compared
to ciprofloxacin in PhB , PhB , and FaSSIF, respectively.
7
.4
6.5
Compounds 5c and 5d displayed solubilities of 0.5−1 of that of
DISCUSSION AND CONCLUSION
■
Ciprofloxacin is a compound with solid-state limited solubility,
and hence, we decided to improve the aqueous solubility by
introducing bulky, flexible side chains and reducing the
hydrogen bond capacity of the molecule to disrupt the crystal
lattice. Intermolecular interactions can be lowered by this
approach. The long, flexible alkyl chains will perturb the crystal
lattice, and the reduced hydrogen bond capacity, as a result of
esterification of the carboxylic acid function, will result in less
7
ordered intermolecular hydrogen bond patterns. A similar
approach in which the flat, planar molecular structure of
phosphodiesterase-4 inhibitors was disrupted through intro-
duction of bulky substituents based on cyclohexyl was recently
8
published. The authors speculated that the higher solubility of
the derivatives was an effect of a less ordered crystal lattice;
however, the solid state was not characterized to prove this. For
the ciprofloxacin derivatives herein melting point was used to
investigate how the crystal lattice was affected by the alkyl
chains. Indeed, melting point was shown to decrease with
Figure 3. Solubility ratio with standard error (solubility of derivative/
solubility of ciprofloxacin) in phosphate buffers pH 7.4 (PhB ) and
7.4
pH 6.5 (PhB ), and fasted state simulated intestinal fluid (FaSSIF).
6.5
2
691
dx.doi.org/10.1021/jm301721e | J. Med. Chem. 2013, 56, 2690−2694

Journal of Medicinal Chemistry
Brief Article
a
Table 1. Permeability, Solubility, and BCS Classification of Ciprofloxacin and Synthesized Derivatives
Papp(a−b)
Papp(b−a)
ratio
b/a
Papp(a−b + 1 μM GF)
Papp(b−a + 1 μM GF)
ratio b/a
+ 1 μM GF
BCS
Papp
BCS
SpH6.5
BCS
class
6
6
6
6
substance
(×10 cm/s)
(×10 cm/s)
(×10 cm/s)
(×10 cm/s)
ciprofloxacin
0.19 ± 0.06
1.01 ± 0.27
3.45 ± 0.53
4.69 ± 0.24
0.43 ± 0.04
4.95 ± 3.04
13.88 ± 3.40
6.42 ± 0.76
0.92 ± 0.03
5
nd
nd
nd
low
low
high
low
low
low
high
low
low
4
1
2
2
2
1
2
2
5
5
5
5
5
5
5
a
b
c
d
e
f
19.66 ± 0.60
24.31 ± 0.70
35.18 ± 4.51
62.75 ± 5.03
65.89 ± 0.65
263.55 ± 40.97
116.08 ± 9.24
19
7
22.01 ± 1.90
4.12 ± 0.21
4.56 ± 1.02
62.95 ± 1.93
21.56 ± 2.37
15.46 ± 1.59
21.54 ± 3.41
8.29 ± 1.32
3.69 ± 0.61
11.96 ± 0.76
26.93 ± 1.44
59.65 ± 2.42
7.76 ± 0.84
53.14 ± 3.50
0.4
0.9
2.8
0.4
2.6
0.5
2.5
high
high
high
high
high
high
high
7
146
13
19
18
g
a
nd: not determined. Values are presented as the mean ± standard deviation based on triplicates.
fragments so that improved permeability is obtained. Therefore,
little consideration was given to the biological effect and/or
toxicity of the structural modifications and the derivatives were
not tested for toxicity, potency, tissue distribution, or
elimination. In vitro metabolism experiments performed
showed that the derivatives were relatively stable with 0.5%
(
5a) to 2.4% (5f) ciprofloxacin formed after 1 h of incubation
in hepatocytes. The compounds were also stable during
transport and solubility studies.
To conclude, classical medicinal chemistry increased
molecular flexibility and lipophilicity, resulting in improved
solubility and permeability of a poorly absorbed model
compound. These findings can guide medicinal chemists in
optimizing poorly soluble, solid-state limited structures with
regard to absorption through synthesis of compounds with less
ordered crystal lattice using flexible, lipophilic side chains.
EXPERIMENTAL SECTION
■
Figure 4. BCS classification of ciprofloxacin and synthesized
derivatives.
Synthesis of Derivatives. All reactions were stirred under
ambient conditions and monitored by TLC if not otherwise is stated.
For detailed synthesis and analysis of intermediate compounds see the
Supporting Information. Purity was determined with analytical
RPHPLC−MS and elemental analysis. The HPLC system (Gilson)
was connected to a Finnigan AQA quadrupole mass spectrometer.
Onyxmonolithic C18 column (50 mm × 4.6 mm) and a gradient of
CH CN/H O (0.05% HCOOH) with a flow rate of 4 mL/min were
increased chain length, and the reduction was 80 °C for the
derivative with the most flexible substituent (5f) compared to
the parent compound. 5a reduced the melting point with 11
°
C, although this substituent does not significantly increase the
number of conformations formed. However, the methyl ester of
a blocks the intermolecular hydrogen bond formed between
3
2
used, and detection was performed with UV (DAD, 214 and 254 nm)
and MS (ESI). Elemental analysis was performed at Mikrokemi AB,
Uppsala, Sweden. All products were >95% pure according to LC−MS
and elemental analysis. Microscopy (Nikon Eclipse TS100, 40×
magnification) was used to visually inspect particle size of obtained
material. DSC (DSC6200, Seiko, Japan) using 0.5−1.0 mg of each
5
the hydroxyl function in the carboxylic acid and the carbonyl of
the quinolone which is likely to affect the molecular packing
and the hydrogen bond structure within the crystal lattice. The
9
less ordered crystal lattice resulted in equal or increased
aqueous solubility for 5a−e, whereas 5f and 5g as a
consequence of increased lipophilicity showed 10- to 30-fold
lower water solubility compared to ciprofloxacin. The results
from the solubility and permeability studies were combined to
sort the compounds into the BCS. All analogues were high
permeability compounds, hence pushing the BCS 4 compound
into class 1 or 2. 5a showed high solubility in all media studied
and hence was sorted as a BCS 1. 5e was sorted as BCS 2 at pH
derivative was performed to determine the melting temperatures (T )
m
and to analyze possible occurrence of polymorph material or residual
traces of organic solvents. Each sample was weighed into an aluminum
pan and covered with a pierced lid before being heated to 50 °C above
the expected melting temperature at a rate of 10 °C/min while being
purged with liquid nitrogen. All experiments were performed in
triplicate.
Free Bases of 1-Cyclopropyl-6-fluoro-4-oxo-7-(piperazin-1-
yl)-1,4-dihydroquinoline-3-carboxylates 5a−g. An amount of
50−100 mg of 4a−g (see Supporting Information) was dissolved in
7
.4 and as BCS 1 at pH 6.5. However, all other derivatives were
4−10 mL of MeOH and applied to a PL-HCO MP StratoSphere SPE
3
BCS class 2 analogues and hence may be possible to transform
to BCS 1 behaving compounds through formulations. Of the
BCS 2 derivatives produced in this work, 5b would be the most
interesting to undertake such efforts for because of its low Do
column (Varian). Solvent was removed under vacuum, and cold
diethyl ether was added to precipitate the final products which were
obtained after filtration. The white crystals obtained were dried under
1
19
vacuum and analyzed by H and F NMR, LC−MS, DSC, XRPD and
1
by elemental analyses. 5a was prepared from 4a. H NMR (400 MHz,
(
3 and 5 at pH 7.4 and 6.5, respectively) and comparably low
D O): δ 1.11−1.20 (m, 2H), 1.34−1.43 (m, 2H), 3.50−3.68 (m, 9H),
2
efflux ratio.
3
.89 (s, 3H), 7.52 (d, J = 7.6 Hz, 1H), 7.71 (d, J = 13.1 Hz, 1H), 8.67
This work was undertaken to prove the impact of small
molecular changes on the crystal lattice and hence the aqueous
solubility and to simultaneously design these molecular
+
(s, 1H). C H FN O MW: 345. MS (ESI): m/z 346 [M + H ]. Anal.
18 20 3 3
(C H FN O ·0.2H O) C, H, N. Shows diffraction peaks in X-ray
18
20
3
3
2
powder diffraction (XRPD) studies, indicating crystalline material. Tm:
2
692
dx.doi.org/10.1021/jm301721e | J. Med. Chem. 2013, 56, 2690−2694

Journal of Medicinal Chemistry
Brief Article
1
5
2
54.5 ± 0.4 °C. 5b was prepared from 4b. H NMR (400 MHz, D O):
monolayers formed from 3.3 × 10 cells seeded on polycarbonate filter
inserts (12 mm diameter, pore size 0.4 μm; Costar) were used. The
transport experiments were performed in a pH gradient of 6.5/7.4 in
the apical to basolateral direction at 37 °C and were started by the
application of the drug solution (10 μM) to the donor side. The filter
inserts (n = 3) were stirred at 500 rpm, and the receiver chambers
were sampled continuously for 60 min. The transport experiment was
2
δ 1.09−1.21 (m, 2H), 1.33−1.45 (m, 5H), 3.44−3.66 (m, 9H), 4.37
m, 2H), 7.52 (m, 1H), 7.71 (d, J = 13.1 Hz, 1H), 8.67 (s, 1H).
C H FN O MW: 359. MS (ESI): m/z 360 [M + H] . Anal.
C H FN O ) C, H, N. Shows diffraction peaks in XRPD studies,
(
+
19
22
3
3
(
19 22 3 3
indicating crystalline material. T : 222.2 ± 1.9 °C. 5c was prepared
m
1
from 4c. H NMR (400 MHz, D O): δ 1.12−1.21 (m, 2H), 1.34−1.46
2
1
4
(
7
(
m, 8H), 3.49−3.71 (m, 9H), 5.22 (m, 1H), 7.55 (d, J = 7.5 Hz, 1H),
repeated with 1 μM GF120918, to specifically inhibit P-gp. All
samples were analyzed with LC−MS/MS directly after termination of
the experiment. The derivatives were stable in the transport assay (no
detection of ciprofloxacin) and used at nontoxic concentrations.
Apparent permeability (Papp) was calculated from
.76 (d, J = 13.3 Hz, 1H), 8.68 (s, 1H). C H FN O MW: 373. MS
20 24 3 3
+
ESI): m/z 374 [M + H] . Anal. (C H FN O ·0.1H O) C, H, N.
20 24 3 3 2
Shows broad unresolved diffraction peaks in XRPD studies, indicating
low crystalline material or amorphous material. T : 228.7 ± 0.7 °C. 5d
was prepared from 4d. H NMR (400 MHz, D O): δ 1.03 (t, J = 7.3
m
1
2
Hz, 3H), 1.11−1.19 (m, 2H), 1.34−1.41 (m, 2H), 1.80 (m, 2H),
ΔQ
1
P
app
=
3
.49−3.67 (m, 9H), 4.17 (t, J = 6.7 Hz, 2H), 7.51 (d, J = 7.3 Hz, 1H),
.69 (d, J = 13.1 Hz, 1H), 8.63 (s, 1H). C H FN O MW: 373. MS
Δt AC0
(3)
7
2
0
24
3
3
+
(
ESI): m/z 374 [M + H] . Anal. (C H FN O ) C, H, N. Shows
20 24 3 3
where ΔQ/Δt is steady-state flux (mol/s), C0 is the initial
concentration in the donor chamber at each time interval (mol/
mL), and A is the surface area of the filter (cm ).
Metabolism of the derivatives was investigated by incubating the
derivatives (1 μM) for 60 min in 0.5 mg/mL pooled human liver
microsomes (Invitrogen) and in 0.5 × 10 cells/mL freshly prepared
human hepatocytes at 37 °C isolated according to a previously
published protocol. After termination of the experiment with MeCN
50% v/v) the samples were analyzed for concentration of
ciprofloxacin and parent derivative with LC−MS/MS. Positive
controls for esterase, CYP450, and UGT activity were p-nitrophenol
acetate, dextromethorphan, midazolam, diclofenac, and 7OH-coumar-
in.
BCS. If the maximum oral dose (2.3 mmol, corresponding to 750
mg of ciprofloxacin) was soluble in 250 mL of PhB6.5 and PhB7.4, the
derivatives were defined as highly soluble. Permeability values were
corrected for the pH gradient and then superimposed onto our in-
house correlation of Papp and fraction absorbed to determine high or
diffraction peaks in XRPD studies, indicating crystalline material. T :
m
1
2
17.4 ± 1.2 °C. 5e was prepared from 4e. H NMR (400 MHz, D O):
2
2
δ 1.03 (t, J = 7.4 Hz, 3H), 1.09−1.19 (m, 2H), 1.33−1.53 (m, 4H),
1
7
.77 (m, 2H), 3.48−3.68 (m, 9H), 4.32 (t, J = 6.7 Hz, 2H), 7.50 (d, J =
.3 Hz, 1H), 7.66 (d, J = 13.1 Hz, 1H), 8.62 (s, 1H). C H FN O :
21
26
3
3
6
+
MW 387. MS (ESI): m/z 388 [M + H] . Anal. (C H FN O ) C, H,
21
26
3
3
N. Shows diffraction peaks in XRPD studies, indicating crystalline
1
5
1
material. T : 197.0 ± 0.7 °C. 5f was prepared from 4f. H NMR (400
m
(
MHz, D O): δ 1.03 (t, J = 7.4 Hz, 3H), 1.09−1.19 (m, 2H), 1.33−1.70
2
(
m, 8H), 1.77 (m, 2H), 3.48−3.68 (m, 9H), 4.32 (t, J = 6.7 Hz, 2H),
7
.50 (d, J = 7.3 Hz, 1H), 7.66 (d, J = 13.1 Hz, 1H), 8.62 (s, 1H).
+
C H FN O MW: 415. MS (ESI): m/z 416 [M + H] . Anal.
23
30
3
3
(
C H FN O ·0.2H O) C, H, N. Shows broad unresolved diffraction
23 30 3 3 2
peaks in XRPD studies, indicating low crystalline material or
amorphous material. T : 185.9 ± 0.5 °C. 5g was prepared from 4g.
m
1
H NMR (400 MHz, CDCl ): δ 1.07−1.16 (m, 2H), 1.26−1.34 (m,
3
2
7
H), 3.11 (m, 4H), 3.26 (m, 4H), 3.41 (m, 1H), 5.39 (s, 2H), 7.23−
.55 (m, 6H), 8.05 (d, J = 13.3 Hz, 1H), 8.53 (s, 1H). C H FN O
2
4
24
3
3
+
5
MW: 421. MS (ESI): m/z 422 [M + H] . Anal. (C H FN O ) C, H,
low permeability.
24
24
3
3
N. Shows diffraction peaks in XRPD studies, indicating crystalline
Analytical Methods. A ThermoFinnigan TSQ Quantum Discov-
ery triple−quadrupole (electrospray ionization) coupled to a Waters
Acquity UPLC instrument was used for concentration determination.
A gradient was used (1% mobile phase B to 90% over 2 min total run)
on a Waters HSS T3 1.8 μm column, 2 mm × 50 mm with mobile
phase A consisting of 5% MeCN in water (0.1% formic acid) and
mobile phase B 100% MeCN with 0.1% formic acid. The flow rate was
0.5 mL/min. An amount of 5 μL of sample was injected and detected
in MRM mode. The limit of quantification (LoQ) was estimated based
on the concentration that yielded a signal-to-noise ratio of >10. For
the compound with lowest sensitivity (ciprofloxacin) the LoQ was 2
nM.
material. T : 200.0 ± 1.4 °C.
m
Solubility Measurements. The apparent solubility were meas-
ured in PhB , PhB , and FaSSIF using a previously described small
7
.4
6.5
1
0
11
shake flask method. PhB7.4 was prepared according to USP 25.
PhB6.5 and FaSSIF were prepared according to the protocol developed
by Galia and co-workers. The compounds were weighed in excess
12
into vials (n = 3) with 500 μL of solubility medium and shaken (37 °C,
3
00 rpm), and after 4 h (selected to reflect the time allowed for
dissolution in the small intestine in vivo) the suspensions were
centrifuged twice at 10 000g for 10 min to separate solid materials
from the solution. The supernatant was determined for concentration
with LC−MS/MS analysis. The derivatives were stable during the 4 h
in all solvents. Standard error of solubility ratio (SE ) was calculated
SR
ASSOCIATED CONTENT
from
■
*
S
Supporting Information
2
2
σSA
SA
σSB
SB
Detailed experimental, analytical, and spectral data for 2, 3a−g,
and 4a−g and physicochemical properties of derivatives. This
material is available free of charge via the Internet at http://
pubs.acs.org.
SESR = SR
+
2
2
(1)
in which SR is the solubility ratio between mean derivative solubility
S ) over mean ciprofloxcacin solubility (S ). σ and σSB are the
(
A
B
S
A
standard errors for the solubility mean values, respectively. The Do
was calculated from
AUTHOR INFORMATION
■
M0
V0CS
Corresponding Author
*Phone: +46 18 4714118. Fax: +46 18 4714223. E-mail:
christel.bergstrom@farmaci.uu.se.
Do =
(2)
where M0 is dose, V0 the volume (set to 250 mL), and C the
S
4
Present Address
measured apparent solubility. The maximum oral dose given
§
(
according to the Swedish Physician Desk Reference) was 2.3 mmol,
For U.T.: AstraZeneca R&D Mo
31 83 Molndal, Sweden.
Notes
̈
lndal, Pepparedsleden 1, SE-
which was used for the BCS.
4
̈
Cell Culture and Metabolism. Caco-2 cells (American Tissue
Collection, Rockville, MD) were maintained in an atmosphere of 90%
13
air and 10% CO as described previously. For transport experiments,
The authors declare no competing financial interest.
2
2
693
dx.doi.org/10.1021/jm301721e | J. Med. Chem. 2013, 56, 2690−2694

Journal of Medicinal Chemistry
Brief Article
(
9) Mahaptra, S.; Venugopala, K. N.; Guru, T. N. A device to
ACKNOWLEDGMENTS
■
crystallize organic solids: structure of ciprofloxacin, midazolam, and
ofloxacin as targets. Cryst. Growth Des. 2010, 10 (4), 1866−1870.
(10) Bergstrom, C. A. S.; Norinder, U.; Luthman, K.; Artursson, P.
We thank Dr. Seth Bjo
̈
rk, AstraZeneca R&D So
̈
dertal
̈
je for
conducting XRPD analyses and SimulationsPlus (Lancaster,
CA) for providing a reference site license of ADMET Predictor.
The authors gratefully thank the Swedish Research Council
̈
Experimental and computational screening models for prediction of
aqueous drug solubility. Pharm. Res. 2002, 19 (2), 182−188.
(11) United States Pharmacopeia and National Formulary (USP 32-NF
(
Grants 621-2008-3777 and 521-2009-2651), The Swedish
2
7); United States Pharmacopeia Convention: Rockville, MD, 2009;
Agency for Innovation Systems (Grant 2010-00966), the Knut
and Alice Wallenberg Foundation, AstraZeneca, and The
Disciplinary Domain of Medicine and Pharmacy at Uppsala
University for their financial support.
Vol. 1.
(12) Galia, E.; Nicolaides, E.; Horter, D.; Lobenberg, R.; Reppas, C.;
Dressman, J. B. Evaluation of various dissolution media for predicting
in vivo performance of class I and II drugs. Pharm. Res. 1998, 15 (5),
6
(
98−705.
13) Hubatsch, I.; Ragnarsson, E. G.; Artursson, P. Determination of
ABBREVIATIONS USED
■
drug permeability and prediction of drug absorption in Caco-2
monolayers. Nat. Protoc. 2007, 2 (9), 2111−2119.
FaSSIF, fasted state simulated intestinal fluid; PhB , phosphate
6
.5
buffer pH 6.5; PhB , phosphate buffer pH 7.4; BCS,
(14) (a) de Bruin, M.; Miyake, K.; Litman, T.; Robey, R.; Bates, S. E.
Reversal of resistance by GF120918 in cell lines expressing the ABC
half-transporter, MXR. Cancer Lett. 1999, 146 (2), 117−126.
7
.4
biopharmaceutics classification system; Do, dose number;
Papp, apparent permeability; LoQ, limit of quantification;
DSC, differential scanning calorimetry; T , melting point;
(b) Taipalensuu, J.; Tavelin, S.; Lazorova, L.; Svensson, A. C.;
m
Artursson, P. Exploring the quantitative relationship between the level
of MDR1 transcript, protein and function using digoxin as a marker of
MDR1-dependent drug efflux activity. Eur. J. Pharm. Sci. 2004, 21 (1),
69−75.
log P, octanol−water partition coefficient
REFERENCES
■
(
15) LeCluyse, E. L..; Alexandre, E. Isolation and culture of primary
hepatocytes from resected human liver tissue. Methods Mol. Biol. 2010,
40, 57−82.
(
1) Rane, S. S.; Anderson, B. D. What determines drug solubility in
lipid vehicles: Is it predictable? Adv. Drug Delivery Rev. 2008, 60 (6),
6
6
(
38−656.
2) (a) Bergstrom, C. A. S.; Wassvik, C. M.; Johansson, K.; Hubatsch,
̈
I. Poorly soluble marketed drugs display solvation limited solubility. J.
Med. Chem. 2007, 50 (23), 5858−5862. (b) Fagerberg, J. H.; Tsinman,
̈
O.; Sun, N.; Tsinman, K.; Avdeef, A.; Bergstrom, C. A. S. Dissolution
rate and apparent solubility of poorly soluble drugs in biorelevant
dissolution media. Mol. Pharmaceutics 2010, 7 (5), 1419−1430.
(
c) Wassvik, C. M.; Holmen, A. G.; Draheim, R.; Artursson, P.;
Bergstro
solubility. J. Med. Chem. 2008, 51 (10), 3035−3039.
3) Zaki, N. M.; Artursson, P.; Bergstrom, C. A. S. A modified
̈
m, C. A. S. Molecular characteristics for solid-state limited
(
̈
physiological BCS for prediction of intestinal absorption in drug
discovery. Mol. Pharmaceutics 2010, 7 (5), 1478−1487.
(
4) Amidon, G. L.; Lennernas, H.; Shah, V. P.; Crison, J. R. A
theoretical basis for a biopharmaceutic drug classification: the
correlation of in vitro drug product dissolution and in vivo
bioavailability. Pharm. Res. 1995, 12 (3), 413−420.
(
̈
5) Bergstrom, C. A. S.; Strafford, M.; Lazorova, L.; Avdeef, A.;
Luthman, K.; Artursson, P. Absorption classification of oral drugs
based on molecular surface properties. J. Med. Chem. 2003, 46 (4),
5
(
58−570.
6) Neuhoff, S.; Ungell, A. L.; Zamora, I.; Artursson, P. pH-
dependent bidirectional transport of weakly basic drugs across Caco-2
monolayers: implications for drug−drug interactions. Pharm. Res.
2
(
003, 20 (8), 1141−1148.
7) (a) Stella, V. J.; Martodihardjo, S.; Terada, K.; Rao, V. M. Some
relationships between the physical properties of various 3-acylox-
ymethyl prodrugs of phenytoin to structure: potential in vivo
performance implications. J. Pharm. Sci. 1998, 87 (10), 1235−1241.
(
b) Yamaoka, Y.; Roberts, R. D.; Stella, V. J. Low-melting phenytoin
prodrugs as alternative oral delivery modes for phenytoin: a model for
other high-melting sparingly water-soluble drugs. J. Pharm. Sci. 1983,
7
2 (4), 400−405. (c) Varia, S. A.; Schuller, S.; Sloan, K. B.; Stella, V. J.
Phenytoin prodrugs III: water-soluble prodrugs for oral and/or
parenteral use. J. Pharm. Sci. 1984, 73 (8), 1068−1073.
(
8) Press, N. J.; Taylor, R. J.; Fullerton, J. D.; Tranter, P.; McCarthy,
C.; Keller, T. H.; Arnold, N.; Beer, D.; Brown, L.; Cheung, R.;
Christie, J.; Denholm, A.; Haberthuer, S.; Hatto, J. D.; Keenan, M.;
Mercer, M. K.; Oakman, H.; Sahri, H.; Tuffnell, A. R.; Tweed, M.;
Tyler, J. W.; Wagner, T.; Fozard, J. R.; Trifilieff, A. Solubility-driven
optimization of phosphodiesterase-4 inhibitors leading to a clinical
candidate. J. Med. Chem. 2012, 55 (17), 7472−7479.
2
694
dx.doi.org/10.1021/jm301721e | J. Med. Chem. 2013, 56, 2690−2694