Deciphering the origins of molecular toxicity of combretastatin A4 and its glycoconjugates: Interactions with major drug transporters and their safety profiles: In vitro and in vivo

Article abstract of DOI:10.1039/c7md00246g

Cellular uptake and transport mechanisms directly correlate with the drug-like profiles of lead compounds. To decipher the molecular origin of the toxicity of combretastatin A4 (CA4), an important microtubule targeting agent, we investigated the interacti

Full text of DOI:10.1039/c7md00246g

MedChemComm
View Article Online
View Journal | View Issue
RESEARCH ARTICLE
Deciphering the origins of molecular toxicity of
combretastatin A4 and its glycoconjugates:
interactions with major drug transporters and
their safety profiles in vitro and in vivo†
Cite this: Med. Chem. Commun.,
017, 8, 1542
2
a
a
a
c
Zhenhua Huang,‡ Gentao Li,‡ Xue Wang, Hu Xu,
a
ab
Youcai Zhang* and Qingzhi Gao
*
Cellular uptake and transport mechanisms directly correlate with the drug-like profiles of lead compounds.
To decipher the molecular origin of the toxicity of combretastatin A4 (CA4), an important microtubule
targeting agent, we investigated the interactions between CA4 and six key drug transporters, namely
hOAT1, hOAT3, hOCT1, hOCT2, hOATP1B3, and hOATP2B1. Three combretastatin-based glycoconjugates,
namely Glu-CA4, Man-CA4, and Gal-CA4 with glucose, mannose, and galactose respectively, were synthe-
sized and their in vitro and in vivo biological characteristics were evaluated. CA4 exhibited significant inhibi-
tion against hOAT3 and hOATP2B1, moderate inhibition of hOAT1 and hOCT2, and weak inhibitory effects
on hOCT1 and hOATP1B3. Compared to CA4, the inhibitory activities of Glu-CA4 on the six transporters
were minimal. The glycoconjugates were found to have a superior safety profile with their maximum toler-
ated dose (MTD) values exhibiting a 16–34-fold increase compared to CA4. Given the drawbacks of CA4,
the enhanced solubility and safety profiles of CA4 glycoconjugates augur well for further investigation into
these intriguing candidates' in vivo efficacy.
Received 11th May 2017,
Accepted 3rd June 2017
DOI: 10.1039/c7md00246g
rsc.li/medchemcomm
sized and evaluated for improved water solubility and in vivo
efficacy. Among these compounds, a few analogs with im-
1
. Introduction
6
Combretastatin A4 (CA4) is a natural vascular-disrupting
agent (VDA) that has shown potent inhibition of tubulin poly-
merization by binding to tubulin at the colchicine binding
proved water-solubility including CA4-disodium phosphate
(CA4P) and serine-conjugated CA4 (AVE8062) are currently un-
2
,7
der clinical investigation.
Nonetheless, some early-phase
1
site 5. Though CA4 exhibits strong cytotoxicity against vari-
clinical trials have revealed promising antineoplastic activity,
however, from the monotherapy studies, all tested CA4 ana-
logs have been found to cause systemic toxicity and severe
ous human cancer cell lines, studies revealed that it was not
promising as a therapeutic agent in clinical trials due in large
part to its poor water solubility and lack of tumor-specific-
8
side effects to the patients. The data observed from pre- and
2
,3
ity. Additionally, CA4 is only able to induce tumor regres-
sion at doses that are close to its maximum tolerated dose
clinical studies unequivocally confirm the shortcomings of
this natural compound as a monotherapy for cancer treat-
ment, and this has led us to investigate the cellular uptake
and drug transport mechanisms that may potentially be in-
volved in the toxic effect of CA4.
(
MTD), which has been hypothesized to be associated with its
4
,5
low tumor selectivity and narrow therapeutic margin. For
this reason, a large number of CA4 analogs have been synthe-
Drug uptake primarily relies on solute carrier (SLC) trans-
porters by facilitated diffusion or ion-coupled secondary ac-
tive transport. These transporters are localized in organs such
as the small intestine, liver, and kidney as well as the central
nervous system (CNS), and they play key roles in drug absorp-
tion, distribution, metabolism, excretion, as well as drug tox-
a
School of Pharmaceutical Science and Technology, Tianjin University, Tianjin
3
b
00072, P. R. China. E-mail: youcai.zhang@tju.edu.cn, qingzhi@tju.edu.cn
Tianjin Key Laboratory for Modern Drug Delivery & High-Efficiency,
Collaborative Innovation Center of Chemical Science and Engineering, School of
Pharmaceutical Science and Technology, Tianjin University, 92 Weijin Road,
Nankai District, Tianjin 300072, P. R. China
9
icity profiles (ADMET). Based on their mode of action, trans-
c
Department of Biochemistry, Gudui BioPharma Technology Inc., 5 Lanyuan
porters could be grouped into uptake transporters that
mediate the transport of drug molecules into cells and efflux
transporters that transport substances out of cells. Inhibition
or induction of transporters may also cause severe drug–drug
Road, Huayuan Industrial Park, Tianjin 300384, P. R. China
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
c7md00246g
‡
These authors contributed equally to this work.
1542 | Med. Chem. Commun., 2017, 8, 1542–1552
This journal is © The Royal Society of Chemistry 2017

View Article Online
MedChemComm
Research Article
1
0
interactions (DDIs) in patients. Most DDIs via uptake trans-
porters involve the two gene superfamilies, namely the SLCO
superfamily, made up of the organic anion transporting poly-
peptides (OATPs), as well as the SLC22A superfamily,
containing the organic anion transporters (OATs) and organic
2.2. Synthesis of the glycoconjugates of CA4
As shown in Fig. 1, the preparation of glucose and galactose
derived glycoconjugates (Glu-CA4 and Gal-CA4) were accom-
plished by a two-step sequence from CA4. The mannose de-
rived glycoconjugate Man-CA4 was synthesized by a one-step
reaction from CA4. The final products were completely char-
1
1
cation transporters (OCTs). As key indicators of a potential
drug candidate, a variety of transporter assays for OATPs,
OATs, and OCTs are required by both the FDA and EMA prior
to testing of a New Chemical/Molecular Entity (NCE/NME).
Concerning R&D activities with CA4, although the over-
whelming majority of efforts have been expended in its phar-
macological and clinical evaluations, so far, to the best of our
knowledge, no study has investigated its uptake and trans-
port profiles, as well as its potential interactions with active
transporters.
In the current study, together with the evaluation of
the transport mechanism of CA4, we also investigated the
effects of glycoconjugation on CA4 in vitro and in vivo bio-
logical characteristics. As shown in Fig. 1, glucose, man-
nose and galactose derived CA4-glycosides were synthesized
and the differences in the transport properties between
CA4 and its glycoconjugates were compared, including
their potential interactions with uptake transporters,
namely OAT1, OAT3, OCT1, OCT2, OATP1B3 and OATP2B1.
The data generated from the current study provide funda-
mental information that is potentially relevant to the ori-
gin of the intrinsic toxicity of CA4, and our study also
highlights a promising new possibility for the development
of a safer CA4-based VDA.
1
13
acterized by H and C NMR spectroscopy. The stereo-
chemistry for both Glu-CA4 and Gal-CA4 was assigned to a
12,13
pure β-anomer,
whereas that of Man-CA4 was assigned to
a pure α-anomer (see the ESI† for the detailed procedure and
spectra data).
2.3. Solubility of CA4 and its sugar conjugates
Solubility testing of CA4, Glu-CA4, Man-CA4 and Gal-CA4 was
conducted by placing an excess of the compounds in deion-
ized water in a series of 10 mL stoppered volumetric test
tubes. The tubes were shaken in an incubating shaker at 25
°
C for 1 h. The test tubes containing equilibrated solutions
were then removed and the solutions were filtered immedi-
ately by passing through 0.2 μm filters. The filtered samples
(
0.5 mL) were diluted appropriately with deionized water and
concentration estimation was made by using HPLC-UV. All
experiments were performed in the dark. A reversed phase
column (Zorbax SB-C18, 5 μm, 4.6 × 150 mm, Agilent) was
used at room temperature for all analyses. The mobile phase
consisted of methanol and water (70 : 30, v/v), and the flow
−
1
rate was 1.0 mL min . The injection volume for CA4 was 20
μL and 2 μL for the conjugates. The average value of three tri-
als was taken. The standard curve obeyed Beer–Lambert's law
2
in the respective concentration range with >R = 0.999.
2
. Experimental
2.4. Cell culture
2
.1. Materials
Human colon cancer (HT29), human prostate cancer
(DU145), human breast cancer (MB231) and human ovarian
cancer (SKOV3) cell lines were obtained from the Central In-
stitute of Pharmaceutical Research, CSPC Pharmaceutical
All biological and chemical reagents were purchased from
commercial companies and directly used unless stated
otherwise.
Fig. 1 Combretastatin A4 and its three sugar-conjugates studied in this project.
This journal is © The Royal Society of Chemistry 2017
Med. Chem. Commun., 2017, 8, 1542–1552 | 1543

View Article Online
Research Article
MedChemComm
Group, China. SKOV3 was cultured in advanced Dulbecco's
modified Eagle medium (DMEM), and the other cell lines
were cultured as an adherent monolayer in RPMI-1640
supplemented with 1% (w/v) glutamine and 10% (v/v) fetal
bovine serum (Gibco) at 37 °C under a humidified atmo-
sphere containing 5% CO and 95% air. Human embryonic
2
kidney (HEK) 293 cells were purchased from ATCC, China
and cultured in DMEM supplemented with 10% FBS at 37 °C
(Sigma-Aldrich) was added to each well at a final concentra-
−
1
tion of 0.83 mg mL and incubated overnight. Cells were
lysed using MTT lysis buffer (15% SDS, 0.015 M HCl) and the
lysate was measured at 570 nm using a multi-well-reading
UV-vis spectrometer. For each drug, the cell survival rates
were expressed as the relative percentage of absorbance com-
pared to controls without drugs. The compounds were tested
for 72 h in HT29 and 24 h in DU145, MB231 and SKOV3
cells. Each experiment was performed in five replicates (5
wells of the 96-well plate per experimental condition).
under a humidified atmosphere containing 5% CO
air.
2
and 95%
2
.5. Transporter overexpression and cell transfection
2.7. Transporter-mediated cellular uptake
4
Human embryonic kidney (HEK) 293 cells stably expressing
human OAT1, OAT3, OCT1, OCT2 and OATP1B3 were
established in the present study. Briefly, the open reading
frame of each transporter was subcloned into a pcDNA3.1/
Hygro (+) (Invitrogen) vector, which was then transfected into
HEK293 cells using Lipofectamine 2000 according to the
manufacturer's protocol (Invitrogen). Stable transporter-
expressing cells were obtained by hygromycin B selection.
The control (mock) HEK293 cells were obtained by trans-
fecting an empty vector into HEK293 cells followed by
hygromycin B selection. These transporter-expressing cells
were validated by both mRNA expression of transporters and
their uptake ability to their corresponding fluorescent sub-
strates (Table 1 and Fig. 2). The OATP2B1-expressing and
mock cells were kindly provided by Dr. Chunshan Gui from
Soochow University (Suzhou, China). All stable transporter-
expressing cell lines were maintained in DMEM
supplemented with 10% FBS, 1% L-glutamine, 1% penicillin/
Cells were seeded at a density of 7 × 10 cells per well in poly-
D-lysine-coated 96-well culture plates. Transport assays were
performed 16 h post seeding in preheated uptake buffer (135
mM NaCl, 5 mM KCl, 2.5 mM CaCl , 1.2 mM MgCl , 0.8 mM
4
, 28 mM glucose, and 13 mM HEPES, pH 7.2).
6-Carboxyfluorescein (6-CF) was used as the fluorescent probe
2
2
MgSO
1
5
for both hOAT1 and hOAT3 according to a previous study.
Fluorescein-methotrexate (FMTX) and dibromofluorescein
(DBF) were used as the fluorescent probe for OATP1B3 and
OATP2B1, respectively. The fluorescent probe alone, with dif-
ferent inhibitors (probenecid for OATs, quinine for OCTs,
and rifampicin for OATPs) or with different compounds was
incubated for 5 min in HEK293-OAT1/3 cells, 10 min in
HEK293-OCT1/2 cells, and 5 min in HEK293-OATP1B1/2B1
cells as well as in their corresponding mock cells. Uptake was
terminated by adding ice-cold phosphate-buffered saline
(PBS), and the mixture was quickly washed three times with
ice-cold PBS. The cells in each well were lysed with 100 μL of
20 mM Tris–HCl containing 0.2% TritonX-100. An aliquot of
50 μL lysate was used to determine fluorescence by using a
Tecan Infinite M200 (Austria, Swiss). The protein concentra-
tion of the cell lysate was determined using a BCA Protein As-
say Kit (Cwbio, Beijing, China). All the uptake values were
standardized against protein content, and were measured in
triplicate. The uptake experiment for each sample was in trip-
licate in the same assay, and repeated 2–3 times on other
plates.
−
1
streptomycin, and 75 μg mL hygromycin B at 37 °C with
% CO . The transporter function was evaluated according to
5
2
previous assays by using the corresponding fluorescent sub-
strates (OAT1/3, 6-carboxyfluorescein (6-CF); OCT1/2,
+
4
-(4-(dimethylamino)styryl)-N-methylpyridinium
(ASP );
OATP1B3, fluorescein-methotrexate (FMTX); OATP2B1, 4′,5′-
1
4–18
dibromofluorescein (DBF)).
2
.6. Cytotoxicity assay
Cells were seeded in a 96-well flat-bottomed microplate at
000–7500 cells per well in 100 μL of growth medium solu-
2
.8. Quantification of CA4 and Glu-CA4 by HPLC-UV
2
tion on day 0. On day 1, the cells were treated either with a
vehicle or the drugs that were dissolved in DMSO and diluted
under cell culture conditions for an indicated time. The
amount of DMSO used for dissolving the samples was kept
below 0.1% (v/v) as the final concentration for each test. MTT
Analyses of CA4 and Glu-CA4 concentrations were carried out
using a Waters HPLC system with a UV detector at 294 nm. A
reversed phase column (Zorbax SB-C18, 5 μm, 4.6 × 150 mm,
Agilent) was used at room temperature for all analyses. The
mobile phase consisted of methanol and water (70 : 30, v/v),
Table 1 The RT-PCR primer sequences. Primers were designed with Primer3 software (version 4)
Gene name
Forward primer (5′ → 3′)
Reverse primer (5′ → 3′)
Product size (bp)
OAT1
OAT3
OCT1
OCT2
GGAGCCAAATTGAGTATGGAGG
CCCACAGTCATCAGGCAAACA
TGTCACCGAAAAGCTGAGCC
GGCTCTATGAGTATCGGCTACA
TGGAGCAACAGTACGGTCAG
GTATGCAAAGCTAGTGGCAAAC
AGGGCGGTGATCCCGTAGA
TCCGTGAACCACAGGTACATC
TCCACGTATAGGTTGGGGAAAT
TGCTTTCGCAGATTAGAGGGAA
159
137
96
121
210
OATP1B3
1544 | Med. Chem. Commun., 2017, 8, 1542–1552
This journal is © The Royal Society of Chemistry 2017

View Article Online
MedChemComm
Research Article
Fig. 2 Characterization of transporter protein transfected cells established in the present study. (A) The mRNA expression of transporters in stable
over-expressing cells compared to control (mock cells). Total RNA was isolated with an RNAiso Plus reagent (Takara Biotechnology, Dalian, China)
according to the manufacturer's protocol. RNA was then reverse transcribed to cDNA using a SuperScript II RT kit (Invitrogen). Quantitative reverse
transcriptase PCR analysis was carried out using a SYBR green PCR mastermix (ABI Inc.). Values were normalized to glyceraldehyde-3-phosphate
dehydrogenase (GADPH), and expressed as the relative fold change to control (mock) group. (B) The fluorescence uptakes in mock and
transporter-expressing cells. Values were expressed as the relative fold change to the control (mock) group.
−
1
and the flow rate was 1.0 mL min . The injection volume 3. Results and discussion
was 20 μL. Glu-CA4 was used as the internal standard for
CA4 quantification, and CA4 was used as the internal stan-
3.1. Synthesis of the three sugar conjugates of CA4
dard for Glu-CA4 quantification.
As shown in Scheme 1, the coupling reactions of CA4 with
tetra-O-acetyl-D-glucopyranosyl and D-galactopyranosyl bromide
were carried out in a bilayer system using CHCl3 and H O
2
2
.9. Determination of maximum-tolerated dose (MTD)
(
3 : 1, v/v) in the presence of sodium hydroxide. After
Four-to-five-week old male DBA/2 mice from Beijing Vital
River Laboratory Animal Technology Co. Ltd. were adminis-
tered with CA4 and the sugar-conjugates by i.p. injection at
deprotection of the acetyl groups using NaOH in methanol,
the desired products were afforded in 79% and 76% yields
for Glu-CA4 and Gal-CA4, respectively. To directly afford man-
nose conjugated Man-CA4, the coupling reaction of CA4 with
tetra-O-acetyl-D-mannopyranosyl bromide was carried out
using lithium hydroxide in methanol as a homogeneous reac-
tion system. The stereochemistry of the conjugates was deter-
−
1
daily intervals for 3 days. CA4 (10–100 mg kg ) and the sugar
−
1
conjugates (200–600 mg kg for Glu-CA4, Man-CA4 and Gal-
−
1
CA4) were administered via i.p. injection (10 mL kg ) to the
randomly grouped mice. A total of five animals were used per
treatment group, and eight animals for the control group.
The control group was given an equivalent volume of sterile
saline. The MTD was defined as the allowance of a median
body weight loss of 15% of the weight before pharmacologic
treatments causing neither death due to toxic effects nor re-
markable changes in the general signs within 3 days after ad-
ministration. The study was approved by the Committee for
the Protection of Animal Care at Tianjin University. All exper-
imental protocols were in accordance with the Guidelines for
Experimental Animal Administration published by the State
Committee of Science and Technology of the People's Repub-
lic of China.
1
mined by H NMR analysis with the proton on the 1-position
of the sugar moiety. Glu-CA4 and Gal-CA4 were assigned to a
1
2,13
pure β-anomer,
and the mannose derived glycoside Man-
1
9
CA4 was assigned to a pure α-anomer.
The stereochemical outcomes for glycosylation of CA4
with different sugar-bromides can be explained by the role of
“Neighbouring Group Participation”. Thus, the protecting
group, typically one with a carboxyl group at the 2-position of
the glycosyl donor, will predominantly result in the formation
of glycosides as an “anti-form” of the protecting group at po-
sition 2. In the current study, when sugar-bromide is derived
from acetyl protected glucose and galactose (Scheme 1), both
the acetyl protecting groups at position 2 will allow for the
formation of an “acetoxonium ion” intermediate that blocks
the attack of CA4 from the same side of the protecting group
(α-side), therefore predominantly allowing for the formation
of the β-glycosides. Similarly, when the glycosyl donor of the
2
.10. Statistical analysis
Data were analyzed by two-tailed unpaired Student's t-test.
The p value for statistical significance was set to <0.05.
This journal is © The Royal Society of Chemistry 2017
Med. Chem. Commun., 2017, 8, 1542–1552 | 1545

View Article Online
Research Article
MedChemComm
Scheme 1 The synthetic scheme of CA4 glycoconjugates.
2
0–23
sugar-bromide is derived from mannose, the reaction will
predominantly result in the formation of the α-glycoside. For
a detailed mechanistic explanation, see Fig. S1.† Due to the
steric hindrance imposed by the two neighbouring β-acetoxyl
groups in acetobromo-D-mannose, the O-glycosylation of man-
nose with CA4 was found to be very slow in the bilayer sys-
tem (Scheme 1, condition a) but occurs smoothly under the
homogeneous conditions with lithium hydroxide in
methanol.
technologies.
As summarized in Table 2, all three sugar
conjugates had markedly increased water solubilities com-
pared with CA4. Conjugation with glucose, mannose, and ga-
lactose resulted in a 610-, 1963-, and 853-fold increase in the
water solubility of CA4, respectively.
3
.3. In vitro cytotoxicity of CA4 and the sugar conjugates
CA4 has been found to have potent cytotoxic activity in colo-
2
4
rectal cancer cell lines, such as HT-29 cells. In this study,
the cytotoxicity of CA4 and its sugar conjugates was first eval-
uated in HT-29 cells. As illustrated in Fig. 3A, the sugar con-
jugates of CA4 showed less cytotoxicity in HT-29 cells than
CA4. The IC50 values of Glu-CA4 (IC50 = 6.91 μM), Man-CA4
(IC50 = 11.44 μM), and Gal-CA4 (IC50 = 39.32 μM) were 9.59-,
3
.2. Water solubility of CA4 and its sugar conjugates
One major purpose of the present study was to improve
the solubility and the toxic profiles of CA4 by leveraging
our expertise in drug-glycoconjugation and related
Table 2 Solubility and safety improvement of glucose, mannose and galactose conjugated CA4 analogs
−1
−1
−1
Compound
M.W. (g mol
)
Solubility (mg mL
)
MTD weight loss (mg kg
)
Ratio (solubility)
Ratio (MTD)
HT29 cell IC50 (μM)
CA4
316.4
478.5
478.5
478.5
0.003
1.83
5.89
2.56
25
400
400
550
1
610
1963
853
1
16
16
34
0.72 ± 0.10
6.91 ± 0.51
11.44 ± 1.70
39.32 ± 3.13
Glu-CA4
Man-CA4
Gal-CA4
1546 | Med. Chem. Commun., 2017, 8, 1542–1552
This journal is © The Royal Society of Chemistry 2017

View Article Online
MedChemComm
Research Article
Fig. 3 Comparison of the cytotoxicity of CA4 with the glycoconjugates in different cell lines.
1
5.89-, and 54.61-fold higher than that of CA4 (IC50 = 0.72
possibly be one of the reasons for Glu-CA4 exhibiting a rela-
tively stronger cytotoxicity in HT-29 cells. To explain the de-
creased cytotoxicities of the sugar conjugates compared to
μM) (Table 2). Since Glu-CA4 has a slightly lower water solu-
bility compared to Man-CA4 and Gal-CA4 (Table 2), this could
Fig. 4 Cytotoxicity comparison study of CA4 and glycoconjugates in the HEK293FT and GLUT1 overexpressing cell line: HEK293FT-GLUT1.
This journal is © The Royal Society of Chemistry 2017
Med. Chem. Commun., 2017, 8, 1542–1552 | 1547

View Article Online
Research Article
MedChemComm
CA4, in section 3.5, we surveyed the cellular uptake proper-
ties of CA4 and Glu-CA4 in different cell systems and demon-
strated that Glu-CA4 exhibited a more than ten-fold decrease
in cellular accumulation. These results can be indicative of
the lower IC50 values of the sugar conjugates, and the re-
duced cytotoxic effect of the conjugates can be also indicative
of the potential for alleviating drug toxicity via
glycoconjugation.
Among the three CA4 sugar conjugates, Glu-CA4 was
the most cytotoxic in HT-29 cells. Thus, Glu-CA4 was cho-
sen for further evaluation. The 24 h cytotoxicity studies of
CA4 and Glu-CA4 were conducted in DU145 (human pros-
tate cancer), MB231 (human breast cancer), and SKOV3
above, theoretically, more hydrophilic molecules have a
lower ability to cross the cell membrane without the aid of
transport proteins. However, in the 24 h cytotoxicity study,
Glu-CA4 did not show a marked difference in cytotoxicity
with these cancer cell lines as compared to CA4. Some
monosaccharides, such as D-glucose, D-mannose, and D-
galactose, have been identified as substrates for glucose
transporters (GLUTs). Human GLUT1-overexpressing cells
were used to evaluate the potential role of GLUTs in the
cytotoxicity of CA4 and its sugar conjugates. As a result
(Fig. 4), the sugar conjugates of CA4 showed comparable
cytotoxicity in mock and GLUT1-overexpressing cells,
suggesting that GLUT1 had little effect on the cytotoxicity
of the CA4 sugar conjugates.
(human ovarian cancer) cell lines (Fig. 3B). As discussed
Fig. 5 Dose-dependent inhibition of CA4 and Glu-CA4 on six key drug transporters.
1548 | Med. Chem. Commun., 2017, 8, 1542–1552
This journal is © The Royal Society of Chemistry 2017

View Article Online
MedChemComm
Research Article
3
.4. Concentration-dependent inhibition of CA4 and Glu-CA4
the six transporters, CA4 showed very strong inhibition of OAT3
and OATP2B1 with IC50 values being 16.35 and 12.27 μM, respec-
tively. Additionally, CA4 also showed moderate inhibition of
OAT1 and OCT2, whereas its inhibitory effects on OCT1 and
OATP1B3 were very weak. Compared to CA4, the inhibitory activi-
ties of Glu-CA4 with the six uptake transporters were minimal,
with IC50 values being larger than 200 μM. It is well known that
both OATs and OCTs are ubiquitously expressed in tissues and
organs of murines or humans, including kidney, liver, choroid
plexus, olfactory mucosa, brain, retina, and placenta. Further-
more, biological studies using knockout mice demonstrated that
both OAT and OCT systems are important for the transport of an
extraordinarily broad range of molecules, including a number of
endogenous hormones, nutrients and toxicants, as well as many
on cell uptake transporters
As shown in Fig. 5, using transfected cell lines, we evaluated the
inhibitory effect of CA4 and Glu-CA4 against six key drug trans-
porters (hOAT1, hOAT3, hOCT1, hOCT2, hOATP1B3 and
hOATP2B1). Since the toxicity of the drugs could cause potential
false-positive results in the cellular uptake assays, HEK293-mock
cells were incubated with CA4 or Glu-CA4 at two high concentra-
tions (100 and 200 μM) for 10 min, which was the maximal time
for later cellular uptake assays, and the cytotoxicity was evaluated
by both the MTT and total protein assays. As shown in Fig. 6,
both high concentrations of CA4 and Glu-CA4 had little effect on
the cell viability and total protein amounts in HEK293-mock cells.
The concentration-dependent inhibition of CA4 and Glu-CA4 on
the six key drug transporters was then examined to compare the
inhibitory potencies of the two compounds. Dramatic differences
between CA4 and Glu-CA4 were observed for their inhibitory ef-
25,26
clinically important drugs.
Besides the fact that DDIs might
originate from the inhibition of both OAT and OCT transporters
with CA4 as such a potent inhibitor of the most important or-
ganic anion and cation transporters for OAT1, OAT3, OATP2B1
and OCT2, there is a definite possibility that CA4 might trigger
severe adverse effects and may cause systemic toxicities.
+
fects on OAT1/3-mediated 6-CF uptake, OCT2-mediated Asp up-
take, as well as OATP2B1-mediated DBF uptake (Fig. 5). Among
Fig. 6 Validation of the toxic effect that might occur under inhibitory assay conditions: high concentration (up to 200 μM) of CA4 and Glu-CA4
do not induce toxicity in HEK293 cells for 10 min. A: Microscopic images of HEK293 cells treated with different concentrations of CA4 and Glu-
CA4. B: Influence of different drug concentrations on cell viability.
This journal is © The Royal Society of Chemistry 2017
Med. Chem. Commun., 2017, 8, 1542–1552 | 1549

View Article Online
Research Article
MedChemComm
3
.5. Accumulation of CA4 and Glu-CA4 in transporter-
the uptake of Glu-CA4 increased by about 60% in hOATP2B1-
expressing cells compared to that in the mock cells. However, co-
incubation of rifampicin, a known inhibitor of OATP2B1, in-
creased the uptake of Glu-CA4 in both mock and hOATP2B1-
expressing cells. Thus, Glu-CA4 was not likely a substrate of
hOATP2B1. Upon careful analysis of the cellular uptake data, we
found that the cell concentrations of CA4 and Glu-CA4 were sig-
nificantly different. Except for the hOATP1B3 overexpressed cells,
CA4 in all tested cell lines showed more than a 10-fold higher cel-
lular accumulation than Glu-CA4, which parallels with the cyto-
toxicity of this compound in HT29 cells and reflects the effect of
increased hydrophilicity through glycoconjugation.
overexpressing cells
To determine whether CA4 and Glu-CA4 could be substrates of
hOAT1/3, hOCT1/2, hOATP1B3, or hOATP2B1, the cellular uptake
of CA4 and Glu-CA4 in transporter-overexpressing cells and in
their mock cells was investigated in the presence and absence of
the well-known corresponding inhibitor for each transporter. As
shown in Fig. 7A, uptake of CA4 in these transporters expressing
cells and their corresponding mock cells showed no significant
differences. Overexpression of OAT1/3, OCT1/2, and OATP1B3 did
not increase the cellular uptake of Glu-CA4 (Fig. 7B). Interestingly,
Fig. 7 Cellular uptake of CA4 (A) and Glu-CA4 (B) in transporter-overexpressing cells and their mock cells.
1550 | Med. Chem. Commun., 2017, 8, 1542–1552
This journal is © The Royal Society of Chemistry 2017

View Article Online
MedChemComm
Research Article
Since OAT1 and OAT3 are almost exclusively expressed in Conflict of interest
the kidney and responsible for the renal excretion of a broad
The authors declare no competing interests.
range of drugs, including anticancer drugs, such as metho-
trexate, 6-mercaptopurine, azathioprine, cisplatin, imatinib,
2
7
Acknowledgements
cytarabine and vinblastine, in addition, both OCT2 and
OATP2B1 as renal and hepatic transporters play key roles in
disposition and clearance of endogenous molecules and drug
compounds, therefore, CA4 has great potential for causing
DDIs through inhibitions of OATs and OATPs. In contrast,
the inhibitory effect of Glu-CA4 on the six transporters was
minimal and thus the DDI potency of Glu-CA4 is negligible.
The authors would like to thank Prof. Mark Olson from the
Health Science Platform of Tianjin University for critical re-
view of the manuscript. This research was supported by
Grants from the Tianjin Municipal Applied Basic and Key Re-
search Scheme of China (10ZCKFSH00300, 11JCYBJC14400,
12ZCDZSY11500), and by the Project of the National Basic Re-
search (973) Program of China (2015CB856500).
3
.6. In vivo toxicity of CA4 and its sugar conjugates
Notes and references
Since toxicity is a major concern for CA4 due to its narrow
therapeutic margin, maximal tolerated dose (MTD) studies
were conducted in order to evaluate the in vivo toxicity of the
synthesized conjugates. Adult DBA/2 male mice were treated
with either CA4 or its sugar conjugates by i.p. injection at
daily intervals for 3 days, and the MTD was defined as the al-
lowance of a median body weight loss of 15% of the body
weight before treatment. As summarized in Table 2, the MTD
values for CA4, Glu-CA4, Man-CA4, and Gal-CA4 were 25, 400,
1
G. R. Pettit, C. J. Temple, V. L. Narayanan, R. Varma, M. J.
Simpson, M. R. Boyd, G. A. Rener and N. Bansal, Anti-Cancer
Drug Des., 1995, 10, 299–309.
2
3
4
S. L. Young and D. J. Chaplin, Expert Opin. Invest. Drugs,
2
004, 13, 1171–1182.
J. Griggs, J. C. Metcalfe and R. Hesketh, Lancet Oncol.,
001, 2, 82–87.
2
J. Griggs, R. Hesketh, G. A. Smith, K. M. Brindle, J. C.
−
1
4
00, and 550 mg kg , respectively. Thus, the present study
Metcalfe, G. A. Thomas and E. D. Williams, Br. J. Cancer,
confirmed that sugar conjugates of CA4 have a superior
safety profile with their MTD values exhibiting a 16–34-fold
increase compared to CA4.
2
001, 84, 832–835.
5
6
G. Kremmidiotis, A. F. Leske, T. C. Lavranos, D. Beaumont,
J. Gasic, A. Hall, M. O'Callaghan, C. A. Matthews and B.
Flynn, Mol. Cancer Ther., 2010, 9, 1562–1573.
R. Mikstacka, T. Stefanski and J. Rozanski, Cell. Mol. Biol.
Lett., 2013, 18, 368–397.
4
. Conclusion
With regard to drug efficacy, the pharmaceutical industry has
great interest in drug transporters, particularly those with
7 T. J. Kim, M. Ravoori, C. N. Landen, A. A. Kamat, L. Y. Han,
C. Lu, Y. G. Lin, W. M. Merritt, N. Jennings, W. A. Spannuth,
R. Langley, D. M. Gershenson, R. L. Coleman, V. Kundra and
A. K. Sood, Cancer Res., 2007, 67, 9337–9345.
8 I. M. Subbiah, D. J. Lenihan and A. M. Tsimberidou,
Oncologist, 2011, 16, 1120–1130.
9 Y. Zhang and D. W. Miller, Pharmaceutical Sciences
Encyclopedia, 2010, vol. 12, pp. 1–22.
10 S. Klatt, M. F. Fromm and J. Konig, Pharmaceutics, 2011, 3,
680–705.
11 K. M. Giacomini, S. M. Huang, D. J. Tweedie, L. Z. Benet,
K. L. Brouwer, X. Chu, A. Dahlin, R. Evers, V. Fischer, K. M.
Hillgren, K. A. Hoffmaster, T. Ishikawa, D. Keppler, R. B.
Kim, C. A. Lee, M. Niemi, J. W. Polli, Y. Sugiyama, P. W.
Swaan, J. A. Ware, S. H. Wright, S. W. Yee, M. J. Zamek-
Gliszczynski and L. Zhang, Nat. Rev. Drug Discovery, 2010, 9,
215–236.
12 H. K. Chenault and L. F. Chafin, J. Org. Chem., 1994, 59,
6167–6169.
13 T. Doura, K. Takahashi, Y. Ogra and N. Suzuki, ACS Med.
Chem. Lett., 2017, 8, 211–214.
1
1
broad substrate specificities. In the present study, we dem-
onstrated for the first time the relationship between CA4 and
six key drug transporters, namely, hOAT1, hOAT3, hOCT1,
hOCT2, hOATP1B3 and hOATP2B1, which as an origin, might
directly contribute to the safety profiles and affect the poten-
tial DDI properties of CA4 and its derivatives. CA4 was found
to be a highly potent inhibitor of hOAT3 and hOATP2B1 and
showed moderate inhibition of hOAT1 and hOCT2, whereas
its inhibitory effects on hOCT1 and hOATP1B3 were very
weak. With the synthesis of CA4 sugar conjugates using glu-
cose, mannose, and galactose as sugar motifs, we systemati-
cally investigated the effects of sugar conjugation on the wa-
ter solubility, in vitro cytotoxicity and in vivo safety, as well as
their transport mechanisms. The results revealed that sugar
conjugation greatly improved the water solubility and in vivo
safety profile of CA4. Additionally, we demonstrated that the
sugar conjugated CA4 analogs have no interactions with the
six key transporters that were tested and thus may circum-
vent the interaction risk associated with the most important
drug transporters, therefore significantly reducing the DDI
potency of CA4. Given the drawbacks of CA4, the enhanced
solubility and safety profiles of CA4 sugar conjugates augur
well for further investigation into these intriguing candidates'
in vivo efficacy.
14 A. Bahn, M. Ljubojevic, H. Lorenz, C. Schultz, E.
Ghebremedhin, B. Ugele, I. Sabolic, G. Burckhardt and Y.
Hagos, Am. J. Physiol., 2005, 289, C1075–C1084.
15 P. Duan, S. Li, N. Ai, L. Hu, W. J. Welsh and G. You, Mol.
Pharmaceutics, 2012, 9, 3340–3346.
This journal is © The Royal Society of Chemistry 2017
Med. Chem. Commun., 2017, 8, 1542–1552 | 1551

View Article Online
Research Article
MedChemComm
1
6 E. Schlatter, P. Klassen, V. Massmann, S. K. Holle, D.
Guckel, B. Edemir, H. Pavenstadt and G. Ciarimboli, Pflugers
Arch., 2014, 466, 1581–1589.
22 M. Wu, H. Li, R. Liu, X. Gao, M. Zhang, P. Liu, Z. Fu, J.
Yang, D. Zhang-Negrerie and Q. Gao, Eur. J. Med. Chem.,
2016, 110, 32–42.
1
1
7 C. Gui, A. Obaidat, R. Chaguturu and B. Hagenbuch, Curr.
Chem. Genomics, 2010, 4, 1–8.
8 S. Izumi, Y. Nozaki, T. Komori, O. Takenaka, K. Maeda, H.
Kusuhara and Y. Sugiyama, Mol. Pharmaceutics, 2016, 13,
23 R. Liu, H. Li, X. Gao, Q. Mi, H. Zhao and Q. Gao, Biochem.
Biophys. Res. Commun., 2017, 487, 34–40.
24 D. Tarade, D. Ma, C. Pignanelli, F. Mansour, D. Simard, S.
van den Berg, J. Gauld, J. McNulty and S. Pandey, PLoS One,
2017, 12, e0171806.
4
38–448.
1
2
2
9 B. G. Davis, M. A. T. Maughan, M. P. Green, A. Ullman
25 S. K. Nigam, K. T. Bush, G. Martovetsky, S. Y. Ahn, H. C. Liu,
E. Richard, V. Bhatnagar and W. Wu, Physiol. Rev., 2015, 95,
83–123.
26 A. T. Nies, H. Koepsell, K. Damme and M. Schwab, Handb.
Exp. Pharmacol., 2011, 201, 105–167.
27 K. Mori, Y. Ogawa, K. Ebihara, T. Aoki, N. Tamura, A.
Sugawara, T. Kuwahara, S. Ozaki, M. Mukoyama, K. Tashiro,
I. Tanaka and K. Nakao, FEBS Lett., 1997, 417, 371–374.
and J. B. Jones, Tetrahedron: Asymmetry, 2000, 11,
2
45–262.
0 P. Liu, Y. Lu, X. Gao, R. Liu, D. Zhang-Negrerie, Y. Shi, Y.
Wang, S. Wang and Q. Gao, Chem. Commun., 2013, 49,
2
421–2423.
1 X. Gao, S. Liu, Y. Shi, Z. Huang, Y. Mi, Q. Mi, J. Yang and Q.
Gao, Eur. J. Med. Chem., 2017, 125, 372–384.
1552 | Med. Chem. Commun., 2017, 8, 1542–1552
This journal is © The Royal Society of Chemistry 2017