Synthesis and characterization of bodipy-FL-cyclosporine a as a substrate for multidrug resistance-linked P-glycoprotein (ABCB1)

Article abstract of DOI:10.1124/dmd.119.087734

Fluorescent conjugates of drugs can be used to study cellular functions and pharmacology. These compounds interact with proteins as substrates or inhibitors, helping in the development of unique fluorescence-based methods to study in vivo localization and

Full text of DOI:10.1124/dmd.119.087734

Supplemental material to this article can be found at:
http://dmd.aspetjournals.org/content/suppl/2019/08/01/dmd.119.087734.DC1
1521-009X/47/10/1013–1023$35.00
DRUG METABOLISM AND DISPOSITION
https://doi.org/10.1124/dmd.119.087734
Drug Metab Dispos 47:1013–1023, October 2019
U.S. Government work not protected by U.S. copyright
Synthesis and Characterization of Bodipy-FL-Cyclosporine A as
a Substrate for Multidrug Resistance-Linked
P-Glycoprotein (ABCB1) s
Andaleeb Sajid,1 Natarajan Raju,1 Sabrina Lusvarghi, Shahrooz Vahedi, Rolf E. Swenson,
and Suresh V. Ambudkar
Laboratory of Cell Biology, Center for Cancer Research, National Cancer Institute (A.S., S.L., S.V., S.V.A.), and Imaging Probe
Development Center, National Heart, Lung and Blood Institute (N.R., R.E.S.), National Institutes of Health, Bethesda, Maryland
Received April 25, 2019; accepted July 24, 2019
ABSTRACT
Fluorescent conjugates of drugs can be used to study cellular
functions and pharmacology. These compounds interact with
and NBD-CsA to the human P-gp structure indicates that they both
bind in the drug-binding pocket with similar docking scores and
proteins as substrates or inhibitors, helping in the development of possibly interact with similar residues. Thus, we demonstrate that
unique fluorescence-based methods to study in vivo localization BD-CsA is a sensitive fluorescent substrate of P-gp that can be
and molecular mechanisms. P-glycoprotein (P-gp, ABCB1) is an used to efficiently study the transporter’s localization and function
ATP-binding cassette (ABC) transporter that effluxes most anti- in vitro and in vivo.
cancer drugs from cells, contributing to the development of drug
resistance. To study the transport function of P-gp, we synthe-
SIGNIFICANCE STATEMENT
sized a Bodipy-labeled fluorescent conjugate of cyclosporine A The goal of this study was to develop an effective probe to study
(BD-CsA). After synthesis and characterization of its chemical purity, drug transport by P-glycoprotein (P-gp). Fluorophore-conjugated
BD-CsA was compared with the commonly used 7-nitrobenz-2-oxa- substrates are useful to study the P-gp transport mechanism,
1,3-diazol-4-yl (NBD)-CsA probe. In flow cytometry assays, the
structural characteristics, and development of its inhibitors. Cy-
fluorescence intensity of BD-CsA was almost 10 times greater closporine A (CsA), a cyclic peptide comprising 11 amino acids, is a
than that of NBD-CsA, enabling us to use significantly lower concen- known substrate of P-gp. P-gp affects CsA pharmacokinetics and
trations of BD-CsA to achieve the same fluorescence levels. We found interactions with other coadministered drugs, especially during
that BD-CsA is recognized as a transport substrate by both human transplant surgeries and treatment of autoimmune disorders,
and mouse P-gp. The rate of efflux of BD-CsA by human P-gp is
comparable to that of NBD-CsA. The transport of BD-CsA was sized and characterized Bodipy-FL-CsA as an avid fluorescent
inhibited by tariquidar, with similar IC50 values to those for NBD- substrate that can be used to study the function of P-gp both
CsA. BD-CsA and NBD-CsA both partially inhibited the ATPase in vitro and in vivo. We demonstrate that Bodipy-FL-conjugation
when CsA is given as an immunosuppressive agent. We synthe-
activity of P-gp with similar IC50 values. In silico docking of BD-CsA
does not affect the properties of CsA as a P-gp substrate.
Introduction
and structure of P-gp but with limited success. P-gp is a highly
conserved protein, with homologs present in various species, from
mice and zebrafish to Caenorhabditis elegans. In humans, P-gp is
expressed on the epithelial cells of the liver, kidney, placenta, testes,
and adrenal gland and on endothelial cells of the blood-brain barrier
(Thiebaut et al., 1987; Cascorbi, 2011). The primary function of
this transporter is to efflux toxic metabolites and xenobiotics from
cells. Because of its functional requirement, P-gp has evolved to be
polyspecific, with the ability to transport a wide range of amphi-
pathic and hydrophobic compounds. As a consequence, P-gp can
also transport a variety of anticancer drugs out of cells, and its over-
expression leads to the development of drug resistance (Ambudkar et al.,
1999; Gutmann et al., 2010; Chufan et al., 2015).
The study of drug resistance mechanisms in cancer cells led to the
identification of P-glycoprotein (P-gp; ABCB1), an ABC-transporter
efflux pump (Juliano and Ling, 1976; Shen et al., 1986). In the last three
decades, numerous studies have elucidated the functional mechanism
This research was funded by the Intramural Research Program of the National
Institutes of Health, the National Cancer Institute, Center for Cancer Research and
the National Heart, Lung and Blood Institute.
1A.S. and N.R. contributed equally to this work.
https://doi.org/10.1124/dmd.119.087734.
s
This article has supplemental material available at dmd.aspetjournals.org.
ABBREVIATIONS: ABC, ATP-binding cassette; BD-CsA, Bodipy-FL conjugate of cyclosporine A; Boc, tert-butyloxycarbonyl; Bodipy NHS ester,
boron, [1-[3-[5-[(3,5-dimethyl-2H-pyrrol-2-ylidene-.N)methyl]-1H-pyrrol-2-yl-.k.N]-1-oxopropoxy]-2,5-pyrrolidinedionato]difluoro; CsA, cyclosporine
A; DCM, dichloromethane; DMAP, 4-N,N-dimethylaminopyridine; EDC.HCl, N-(3-dimethylaminopropyl)-N9-ethylcarbodiimide hydrochloride; EtOAc,
ethyl acetate; FBS, fetal bovine serum; Fmoc, fluorenylmethyloxycarbonyl; Fmoc-OSu, N-(9-fluorenylmethoxycarbonyloxy) succinimide; HEK293,
human embryonic kidney cell line 293; HPLC, high-performance liquid chromatography; IMEM, Iscove’s modified Dulbecco’s medium; LC-MS,
liquid chromatography-mass spectrometry; MeOH, methanol; NBD-CsA, 7-nitrobenz-2-oxa-1,3-diazol-4-yl conjugate of cyclosporine A; NHS,
N-hydroxysuccinimide; P-gp, P-glycoprotein; RT, room temperature; TFA, trifluoroacetic acid.
1013

1014
Sajid et al.
Structurally, human P-gp contains two transmembrane domains, each Chearwae et al., 2004; Shukla et al., 2011; Li et al., 2017; Vahedi
having six transmembrane helices and two nucleotide-binding domains, et al., 2018) and to assess P-gp’s function at the blood-brain barrier
connected by extracellular and intracellular loops. The nucleotide- using mouse or rat brain capillaries (Hartz et al., 2010). Thus, there has
binding domains bind and hydrolyze ATP, which is critical for been a need for a stable derivative of a substrate with a high yield
conformational changes associated with the translocation of the of fluorescence to assess the transport function of P-gp, especially in
substrates across the membrane (Ambudkar et al., 1999, 2006; in vivo studies using animal models. Given that CsA is a well character-
}
Szöllosi et al., 2018). Recently, the atomic structures of human P-gp ized substrate of P-gp, we synthesized and characterized BD-CsA as
bound to paclitaxel in the inward-open (Alam et al., 2019) and ATP-bound a substrate of P-gp and compared it with NBD-CsA. We showed that
E-Q mutant in inward-closed (Kim and Chen, 2018) conformations have BD-CsA can be transported by both human and mouse P-gp. The
been reported.
transport was inhibited by tariquidar, a known P-gp inhibitor. We also
To understand the mechanism of transport under in vitro and in vivo demonstrate that, owing to its stability and increased fluorescence
conditions, various fluorophore conjugates of drugs have been used that yield compared with that of NBD-CsA, BD-CsA is useful for labeling
act as substrates of P-gp (Gribar et al., 2000; Shukla et al., 2011; Strouse live cells at significantly lower concentrations.
et al., 2013; Vahedi et al., 2017; Sajid et al., 2018). These conjugated
substrates are useful for studying P-gp transport kinetics by flow
cytometry. Cyclosporine A (CsA), an immunosuppressive agent,
is a cyclic peptide comprising 11 amino acids (Tanaka et al., 1996;
Tedesco and Haragsim, 2012). It is often administered to patients
undergoing transplant surgeries or treatment of autoimmune disorders,
but the dosage must be carefully controlled to avoid side effects.
Although many investigators use CsA as an inhibitor or modulator of
P-gp, it is also transported by P-gp (Saeki et al., 1993; Chen et al., 1997;
Demeule et al., 1998). Interestingly, CsA treatment has been shown to
increase the expression of P-gp in rat tissues (Jette et al., 1996), and P-gp
is known to affect CsA pharmacokinetics and interactions with other
coadministered drugs (Kelly and Kahan, 2002; Dirks et al., 2004;
Pawarode et al., 2007; Barbarino et al., 2013; Yigitaslan et al., 2016). Of
all the known substrates of P-gp, CsA is unique, being a large cyclic
peptide (;1200 Da). Hence, the availability of fluorescent CsA would
Materials and Methods
Chemicals
Bodipy-FL-verapamil (BD-verapamil) was purchased from Setareh Biotech
(Eugene, OR). NBD-cyclosporine A was generously provided by Drs. Anika
Hartz and Björn Bauer, University of Kentucky (Lexington, KY). All remaining
chemicals were purchased from Sigma-Aldrich (St. Louis, MO) and Thermo
Fisher Scientific (Waltman, MA), unless otherwise specified.
Chemical Synthesis of Bodipy-FL-CsA
A schematic representation of the synthesis of BD-CsA is presented in Fig. 2.
General Procedures. All analytical high-performance liquid chromatography
(HPLC) was run on an Agilent 1200 analytical instrument (Agilent Technologies,
Inc., Santa, Clara, CA). Water (A) and acetonitrile (B) were used as mobile
phases with 0.1% trifluoroacetic acid (TFA) (v/v). Liquid chromatography-mass
spectrometry (LC-MS) data were obtained from an Agilent 1200 analytical HPLC
also help in its characterization, as it is the only representative of this and an Agilent Quadruple 6130 LC-MS. Preparative HPLC was performed using
a Shimadzu preparative instrument (Shimadzu Corp., Kyoto, Japan). Chromato-
graphic purifications were carried out using a Teledyne ISCO (Lincoln, NE) flash
chromatography system. Analytical HPLC conditions were the following: I:
Column, Agilent Zorbax-SB300 RP; 5.0 mm; 3.5 Â 50 mm; 1.0 ml/min; detection
at 220 nm; 50%–100% gradient B over 5.0 minutes and kept at 100% B for
5 additional minutes. II: Same column conditions as in I, 70%–100% gradient B
over 5 minutes. III: Same column conditions as in I, 50%–100% gradient B over
10 minutes. IV: Column: Waters X Bridge OBD protein BEH-C4 RP Waters
Corporation, Milford, MA), 3.5 mm, 300 Å, 4.6 Â 50 mm; solvent A, water;
solvent B, methanol (MeOH); elution rate: 1.5 ml/min: detection at 480 nm;
class of drugs.
In this study, we synthesized a Bodipy-conjugated derivative of CsA
(BD-CsA). We show that this derivative is a valuable tool to study P-gp
transport, as no fluorescent conjugate of CsA is commercially available.
Previously, the 7-nitrobenz-2-oxa-1,3-diazol-4-yl (NBD) conjugate of
CsA (NBD-CsA) was synthesized and used as a fluorescent substrate
(Wenger, 1986; Schramm et al., 1995; Hartz et al., 2010; Storck et al.,
2018) (Fig. 1). In the last few years, Bodipy-FL conjugates of sev-
eral substrates have been used for characterization of the transport
function of wild-type P-gp as well as its variants (Weiss et al., 2003; 50%–100% gradient B over 10 minutes.
Fig. 1. Chemical structures of cyclosporine A and its NBD- and Bodipy-fluorescent conjugates. Diagrams showing chemical structures, formula and molecular weights of
cyclosporine A, NBD-cyclosporine A, and BD-cyclosporine A. The fluorescent conjugate was linked to the D-Lys8 residue of a cyclosporine A cyclic peptide.

Development of ABCB1 Fluorescent Substrate Probe
1015
Fig. 2. Scheme for the synthesis of Bodipy-FL-cyclosporine A. (a) Pyridine, Ac2O, DMAP, 20 hours, RT; (b) P4S10-pyridine complex, CH3CN, reflux, 30 minutes;
(c) BnBr, 1,8-diazabicyclo(5,4,0)-undec-7-ene, DCM, 2 hours; (d) 6N HCl, CH3CN, RT, 1 hour; (e) PhNCS, CH3CN, RT, 2 hours; (f) 6N HCl, CH3CN, RT, 4 hours;
(g) Boc-D-Lys(Fmoc)-OH, EDC.HCl, DCM, RT, 20 hours; (h) 0.2 N NaOH, MeOH, RT, 72 hours; (i) Fmoc-OSu, THF, RT, 20 hours; (j) 50% TFA in DCM, 30 minutes;
(k) n-propylphosphonic anhydride, DMAP, DCM, 72 hours; (l) 50% diethylamine in CH3CN, RT, 30 minutes; (m) Bodipy-NHS ester, DMF, RT, 20 hours.
Thionation of Cyclosporine A Acetate; Product 2. Pyridine-P4S10 complex in anhydrous dichloromethane (DCM) (20.0 ml) and stirred for 20 hours at room
(Bergman et al., 2011) (0.642 g, 1.69 mmol) was added to a solution of the temperature (RT). The reaction mixture was concentrated under reduced pressure,
cyclosporine A acetate (1, 7.0 g, 5.63 mmol) in anhydrous acetonitrile (50.0 ml) and the residue was purified over 100.0 g of prepacked flash silica gel column
and refluxed for 30.0 minutes under argon. The solution was cooled and (25–50.0 mm). Elution with 0%–10% MeOH in DCM eluted the required peptide
concentrated to 15.0 ml and then diluted with 100.0 ml of water and extracted with at 8% MeOH concentration. The LC fractions with the pure product were pooled
5 Â 20.0 ml of ethyl acetate (EtOAc). The combined organic extracts were dried and concentrated to yield the fully protected linear peptide as a colorless foam.
(MgSO4), filtered, and concentrated, and the residue was chromatographed over Yield was 0.97 g (55%, 0.55 mmol); HPLC conditions were I, tR 5.3 minutes, and
330.0 g of prepacked silica gel column (50.0 mm). Elution with 1:1 EtOAc/ MS was [M 1 Na] 1771.1.
hexanes yielded 7-thioamide (checked by LC-MS), followed by unreacted
starting material on continued elution with 6:4 EtOAc/hexanes. Yield was a mixture of 0.2 M NaOH in water (22.5 ml) and MeOH (25.0 ml) and stirred
1.2 g (16.8%, 0.95 mmol, colorless gum); recovered starting material was at RT for 72 hours. The solution was concentrated under reduced pressure to
D-Lys8-Cyclosporine 5. Compound 3 (1.94 g, 1.5 mmol) was dissolved in
5.0 g (4.03 mmol, 71.2%). HPLC conditions I, tR 4.5 minutes; MS [M 1 H] 10.0 ml, and the pH was adjusted with 0.2 M of NaHSO4 to 8.0. Fmoc-OSu
was 1261.9. We observed the following: 1) the procedure using phosphorous (0.68 g, 2.0 mmol); THF (10 ml) was added and stirred at RT for 20 hours. The
pentasulfide in xylenes (Eberle and Nuninger, 1993) repeatedly yielded a mixture solution was concentrated under reduced pressure to 15.0 ml, and then the pH
of positional isomeric thioamides (Fig. 2, structure 1) 7 (required monothioamide), was adjusted with 0.2 M NaHSO4 to 3.0. The reaction mixture was diluted
4 (unwanted monothioamide), and 4,7-bis thioamide (major side product), which with water (50.0 ml) and extracted with 5 Â 30 ml of DCM. The combined
were difficult to separate, along with other products; and 2) the preceding procedure organic layer was washed with water, dried (Na2SO4), filtered, concentrated,
consistently yielded the required 7-isomer and occasionally the 4-isomer, which and purified by C18 RP flash silica gel column (120.0 g, solvent A, water;
could be removed at later stages with ease, and no bis isomer was detected by
LC-MS. The starting material was recovered (70%–75%) and recycled.
Coupling of Edman Degradation Product 2 with D-Lysine; Intermediate 3.
solvent B, MeOH; 0%–100% B over 60 minutes; elution rate, 60 ml/min;
detection at 220 and 280 nm). Fractions with the compound and purity of .90%
were (LC-MS) were pooled and concentrated to yield the acid 4 as a colorless gum.
EDC.HCl (0.77 g, 4.0 mmol) was added to a mixture of Boc-D-Lys (fluorenyl- Yield was 0.98 g, (40%, 0.61 mmol), MS [M 1 Na] 1622.7. This acid (150.0 mg,
methyloxycarbonyl, Fmoc)-OH (1.4 g, 3.0 mmol) and amine 2 (1.3 g, 1.0 mmol) 0.094 mmol) was dissolved in 50% TFA in DCM (20 ml) and kept at RT for

1016
Sajid et al.
30 minutes. All the volatiles were removed under reduced pressure, and the residue
was dried under high vacuum for 2 hours. The crude amino acid (0.113 g,
to determine the transport activity of human MRP1 (Müller et al., 2002), with
calcein-AM/NBD-CsA/BD-CsA (0.25 mM) and the inhibitor MK571 (25 mM).
0.07 mmol) in anhydrous DCM (300.0 ml), n-propyl phosphonic anhydride Flow cytometry was done using a FACS CANTO II instrument with BD
(44.5 mg, 50% in EtOAc, 90.0 ml, 0.21 mmol), and DMAP (68.0 mg, 0.56 mmol) FACSDiva software (BD Biosciences, BD Biosciences, San Jose, CA), and
were stirred under argon for 72 hours at RT. The solution was concentrated under spectra were collected in the fluorescein isothiocyanate region (Ex. 488 nm,
reduced pressure, and the residue was purified on a flash C18 RP column (275.0 g; Em. 525 nm). The data were analyzed using FlowJo software (Tree Star, Inc.,
25.0 mm; A, water; B, MeOH; elution rate: 80 ml/min; detection at 220 nm and Ashland, OR).
280 nm; 0%–100% B over 100 minutes). The fraction with the required mass and
Time Course of Efflux of BD-CsA and NBD-CsA
purity .90% were pooled and concentrated under reduced pressure to 5 as
a colorless foam. Yield was 51.6 mg (49%, 0.034 mmol). HPLC conditions: III;
HeLa cells were transduced with BacMam baculovirus as described earlier
(Sajid et al., 2018). After trypsinization, cells were resuspended in PBS containing
1 mM MgCl2 and 0.1 mM CaCl2. For depletion of ATP, 2-deoxyglucose (20 mM)
and sodium azide (5 mM) were added, and cells were incubated at 37°C for
10 minutes. Substrates (BD-CsA or NBD-CsA, 0.5 mM) were added to the
cells and incubated at 37°C for another 20 minutes to load the cells. After
incubation, cells were washed with cold PBS/Ca21/Mg21, and IMDM 1 5% FBS
(enriched medium) was added, followed by incubation at 37°C for different time
points (0, 1, 2, 5, 10, 20, and 30 minutes). Cells were washed with cold IMDM,
and PBS 1 1% BSA was added to be used for analysis by flow cytometry. The
time required for 50% (T1/2 min) efflux of BD-CsA or NBD-CsA was calculated
using GraphPad Prism version 7.0 (GraphPad Software, San Diego, CA).
t
R 5.2 minutes, and MS [M 1 H] was 1482.9.
Bodipy Fl-D-Lys8-Cyclosporine (BD-CsA) 6. [N«-Fmoc)-D-Lys8-cyclospor-
ine 5 (100.0 mg; 0.0625 mmol) was dissolved in 50% diethylamine in acetonitrile
(20.0 ml) and kept at RT for 30.0 minutes. All the volatiles were removed under
reduced pressure, and the residue was coevaporated with toluene (3 Â 10 ml).
The deprotected amine (above) in anhydrous DMF (3.0 ml) was stirred with
Bodipy N-hydroxysuccinimide (NHS) ester obtained from A1 Biochem
Laboratories (Wilmington, NC) (32.3 mg, 0.07 mmol), under argon protected
from light for 24 hours. DMF was evaporated under reduced pressure, and the
residue was purified by preparative HPLC [conditions: column: Waters
Corporation X-Bridge protein BEH-C4 RP; 5.0 mm; 300 Å; 19 Â 150 mm;
detection at 480 nm; solvent A, water; solvent B, MeOH; 20% B to 100% over
40.0 minutes; elution rate: 30 ml/min]. Fractions with .95% purity were
pooled and concentrated under reduced pressure at RT to yield the product
as a brown gum. Yield: 51.8 mg (52.8%, 0.033 mmol). Analytical HPLC
conditions, iv; tR 4.3 minutes; MS [M 1 H] 1534.3.
Fluorescence Microscopy-Based Transport Assay
Transduction was carried out as described in the earlier section using HeLa
cells seeded in a 24-well plate format (50,000 cells/well). After 20–22 hours, cells
were washed twice with PBS, and fluorescent substrates (BD-CsA or NBD-CsA)
were added at 0.5 mM concentration in 1 ml IMDM containing 5% FBS.
Tariquidar (200 nM) was added to the samples wherever indicated. Untrans-
duced cells were used as a control. The transport assay was carried out for
45 minutes at 37°C. Ten minutes before completion of the transport assay,
Nucblue Live cell stain (Invitrogen, Carlsbad, CA) was added to the cells. The
cells were washed with cold PBS three times, and 0.5 ml of PBS was added
to each well. Live cells were visualized in an Evos AMG microscope at
10Â magnification in transmitted light (to visualize the cells), green
fluorescent protein region (for fluorescence of BD-CsA and NBD-CsA) and
Cell Lines and Culture Conditions
HeLa cells were obtained from American Type Culture Collection (Manassas,
VA). Cells were maintained in Dulbecco’s modified Eagle’s medium (Difco
Laboratories, Detroit, MI) supplemented with 10% fetal bovine serum (FBS),
5 mM L-glutamine, 100 U/ml penicillin, and 100 mg/ml streptomycin at 37°C in
5% CO2.
Recombinant BacMam and Baculovirus Generation
The Bac-to-Bac Baculovirus Expression system (Life Technologies, Carlsbad, UV-blue region (to visualize stained nuclei with Nucblue).
CA) was used to generate recombinant baculovirus and BacMam baculovirus, as
described previously (Pluchino et al., 2016). Briefly, human MDR1 and mouse
ABCB1a and ABCG2 genes were cloned in pDonr-255 and used for Gateway
cloning into pDest-625 (mammalian cell expression) and pDest-008 (insect cell
expression). pDest clones were transformed in Escherichia coli DH10Bac
competent cells and used to prepare the recombinant Bacmids for generation
of BacMam and baculovirus according to the manufacturer’s protocol (Gibco:
ThermoFisher).
Preparation of Total Membranes from High Five Insect Cells
The ABCB1 gene cloned in pDEST-008 was transformed in E. coli DH10Bac
cells, and bacmids were prepared harboring recombinant P-gp with a 6X His-tag
and TEV-cleavage site at the C-terminal end, as described previously (Vahedi
et al., 2017). High Five insect cells (Invitrogen) were infected with recombinant
baculovirus carrying P-gp. Membrane vesicles were prepared by hypotonic lysis
of the insect cells, followed by ultracentrifugation to collect the membrane
vesicles (Ramachandra et al., 1998; Kerr et al., 2001).
BacMam Baculovirus Transduction of HeLa Cells and Transport of
Fluorescent Substrates
ATPase Assay
HeLa cells were transduced with human P-gp, mouse P-gp, or human ABCG2
BacMam baculovirus, as described previously (Shukla et al., 2012; Vahedi et al.,
2017; Sajid et al., 2018). Briefly, HeLa cells were incubated with the BacMam
baculovirus at a selected cell:virus ratio for 4 hours. Sodium butyrate (10 mM)
was added, and incubation was continued for an additional 12–16 hours at 37°C
in 5% CO2. For transport assays of P-gp, transduced HeLa cells were trypsinized
and resuspended in Iscove’s modified Dulbecco’s medium (IMDM) containing
5% FBS; 3 Â 105 cells were incubated with fluorescent substrates (BD-verapamil,
pheophorbide A, BD-CsA, or NBD-CsA) at selected concentrations for 45 minutes
at 37°C. After incubation, cells were washed with cold IMDM and resuspended in
cold PBS containing 1% BSA. The transport of substrates was measured by flow
cytometry using untransduced cells as a control. The mean fluorescence intensity
of P-gp-expressing cells after subtraction from that of untransduced cells (not
expressing P-gp) was taken as 100% efflux. For inhibition assays, tariquidar
(200 nM for steady-state assay and 0–100 nM for IC50 calculations) was
added wherever indicated. Both NBD-CsA and BD-CsA were tested with
ABCG2-expressing HeLa cells and MRP1-expressing human embryonic cell line
HEK293 cells as described below. To test the transport activity of ABCG2,
ATP hydrolysis was measured using membrane vesicles of High Five insect
cells expressing P-gp, as described previously (Sajid et al., 2018). Membrane
vesicles (10 mg of protein per 100 ml of reaction volume) were incubated in the
presence or absence of 0.3 mM sodium orthovanadate in ATPase assay buffer
(50 mM MES-Tris pH 6.8, 50 mM KCl, 10 mM MgCl2, 5 mM NaN3, 1 mM
EGTA, 1 mM ouabain, and 2 mM DTT). Basal ATPase activity was measured
in the presence of DMSO, and drug-modulated activity was measured in the
presence of selected drugs (drug stocks prepared at a 100Â concentration in
DMSO). The reaction was started at 37°C by the addition of 5 mM ATP and
stopped by the addition of 2.5% sodium dodecyl sulfate after a 20-minute
incubation. The level of generated inorganic phosphate was quantified with
a colorimetric method (Ramachandra et al., 1998; Vahedi et al., 2017). The
vanadate-sensitive ATPase activities were determined and plotted with
GraphPad Prism software (version 7).
In Silico Modeling
The recently published ligand (paclitaxel)-bound human P-gp structure (PDB ID:
pheophorbide A (a substrate) was used at 2 mM and Ko143 (an inhibitor) at 6QEX) (Alam et al., 2019) was used for docking of CsA, NBD-CsA, and BD-CsA
2.5 mM. HEK-MRP1 and HEK-PCDNA3.1 (vector control) cell lines were used with AudoDock Vina (Scripps Research Institute, La Jolla, CA). For this, the

Development of ABCB1 Fluorescent Substrate Probe
1017
transporter and ligands were prepared using the MGL tools software package
(Scripps Research Institute) (Sanner, 1999; Trott and Olson, 2010). Because
AudoDock Vina does not have a Forcefield for boron and given that the Bodipy
moiety in BD-CsA has boron in its structure, this atom was treated as carbon for
docking purposes. Close examination of published cryo-EM and X-ray structures
of mouse and human P-gp (Szewczyk et al., 2015; Nicklisch et al., 2016; Alam
et al., 2018, 2019), as well as our docking experiments, allowed us to identify 36
residues that interact with ligands in the drug-binding site. The side chains of these
36 residues, all located in the drug-binding pocket in the transmembrane region of
P-gp, were set as flexible. The residues were L65, M68, M69, F72, Q195, W232,
F303, I306, Y307, Y310, F314, F336, L339, I340, F343, Q347, N721, Q725,
F728, F732, F759, F770, F938, F942, Q946, M949, Y953, F957, L975, F978,
V982, F983, M986, Q990, F993, and F994. The receptor grid was centered at x 5
19, y 5 53 and z 5 3 and a box with inner box dimensions 40 Â 40 Â 44 Å was
used to search for all the possible binding poses within the transmembrane region
of the protein. The exhaustiveness level was set at 100 to ensure that the global
minimum of the scoring function would be found considering the large box size
and the number of flexible residues.
offers the flexibility of conjugation to the fluorescent group or to other
groups and simplifies the final purification.
Comparison of the Fluorescence Intensities of BD-CsA and
NBD-CsA
CsA has been widely used in studies of P-gp as an inhibitor or modulator
of activity. Even though cyclosporine A is transported by P-gp, because
of its relatively high affinity and slow efflux rate, it is often used as an
inhibitor of transport (Demeule et al., 1999; Muzi et al., 2009; Jouan
et al., 2016). In the past, several groups have characterized the transport
of NBD-CsA as a P-gp substrate (Schramm et al., 1995; Masereeuw
et al., 2000; Ott et al., 2010; Chufan et al., 2013; Miller, 2014; Vahedi
et al., 2017; Sajid et al., 2018). In the present study, we compared BD-
CsA and NBD-CsA transport by human P-gp expressed on the surface
of HeLa cells. HeLa cells were transduced with BacMam baculovirus
expressing human P-gp, and untransduced HeLa cells not expressing
detectable levels of P-gp were used as a control. First, the fluorescence
intensities of both BD-CsA and NBD-CsA were compared using
a concentration gradient of both the probes in the untransduced cells.
We found that under the same flow cytometry voltage settings, the
intensity of BD-CsA was almost 10 times greater than that of NBD-CsA,
with 0.5 mM of NBD-CsA showing the same fluorescence intensity as
that of 0.05 mM of BD-CsA (Fig. 3A). The P-gp-expressing cells efflux
these substrates, resulting in decreased fluorescence intensity. Next,
we calculated the efflux at different concentrations of both probes. As
shown in Fig. 3B, the efflux (the difference between the fluorescence
intensities of untransduced cells and cells expressing P-gp) was linear, in
the range of 0–0.5 mM BD-CsA and NBD-CsA. There was significant
efflux for BD-CsA but not NBD-CsA below 0.5 mM. Thus, in subsequent
assays, we used 0.5 mM of both the probes to compare their characteristics.
Our results showed that BD-CsA has a higher fluorescence yield that can
be used to develop more sensitive assays, which is critical in animal
studies in which lower doses of BD-CsA can be helpful to avoid toxicity.
Results
Chemical Synthesis of BD-CsA
To develop an efficient and unique probe as a P-gp substrate, we
decided to prepare a Bodipy-FL-labeled compound. Previously,
NBD-CsA was synthesized as a fluorescent probe (Wenger, 1988),
which was used as a reference in this study. The chemical structures
of CsA, NBD-CsA, and BD-CsA are shown in Fig. 1. Figure 2 shows
a schematic representation of the synthesis of BD-CsA. Commercially
available CsA was converted to intermediate 2 as detailed in the
literature (Eberle and Nuninger, 1993), except for step b. Thioamide
formation was accomplished using the pyridine-P4S10 complex (Bergman
et al., 2011), instead of phosphorous pentasulfide, yielding a much
cleaner reaction product (see experimental details). Intermediate 2 was
converted to 3 using standard peptide coupling conditions. Deprotection
of 3 and reprotection of «-amine of the lysine residue resulted in 4,
which was subjected to intramolecular cyclization to yield the required
D-Lysine8-cyclosporine protected as its Fmoc-derivative 5. Removal of the
protecting group from 5 followed by conjugation with Bodipy-NHS ester
yielded the labeled CsA derivative 6 after preparative HPLC (Supplemental
Figs. S1 and S2). The use of Fmoc protection of the «-amine of D-lysine
allows the introduction of a fluorescent group at the final step. This change
BD-CsA and NBD-CsA Are Transported by Both Human and
Mouse P-gp
Using the BacMam baculovirus transduction system, the efflux of
NBD-CsA and BD-CsA was compared using HeLa cells transduced with
human P-gp (ABCB1). We found that human P-gp was able to efflux both
Fig. 3. Kinetics of BD-CsA and NBD-CsA accumulation and efflux. HeLa cells were transduced with BacMam baculovirus to express P-gp, and untransduced cells
were used as control. Cells were incubated with different concentrations of BD-CsA and NBD-CsA (0–0.5 mM) (A) Comparison of fluorescence intensity of BD-CsA with
NBD-CsA in untransduced cells. The fluorescence intensity is proportional to the concentration of the substrates but is much higher for BD-CsA compared with NBD-CsA.
(B) Efflux of BD-CsA and NBD-CsA calculated by subtracting fluorescence of cells expressing P-gp from that of untransduced cells. The efflux was linear with increasing
concentrations of both the substrates but more significant at 0.5 mM. All the experiments were repeated three times; error bars show S.D.

1018
Sajid et al.
Fig. 4. BD-CsA is a substrate of human P-gp. HeLa cells were transduced with BacMam baculovirus to express P-gp, and untransduced cells were used as a control. Cells
were incubated with BD-CsA or NBD-CsA (0.5 mM, each) for 45 minutes at 37°C, and fluorescence was measured using flow cytometry. Histogram traces show the
transport of substrates by human P-gp, (A) BD-verapamil, (B) BD-CsA, and (C) NBD-CsA. The efflux by P-gp was assayed by comparing the fluorescence intensity of cells
expressing P-gp (blue traces) with those that do not express P-gp (untransduced cells, red traces).
NBD-CsA and BD-CsA with similar efficiency. BD-verapamil was used Tariquidar Inhibits Transport of NBD-CsA and BD-CsA by P-gp
as positive control for transport, and the representative transport
Transport of substrates by P-gp is inhibited by tariquidar (originally
profiles of BD-verapamil, NBD-CsA, and BD-CsA are shown in
known as XR9576) (Martin et al., 1999). To compare the inhibition of
Fig. 4. As shown in the histograms, the substrates were transported by
efflux of BD-CsA and NBD-CsA by P-gp, tariquidar (200 nM) was
P-gp (blue traces), leading to a decreased fluorescence in the cell
added during the transport assay. Figure 6 shows that there is no efflux
of either of the compounds in the presence of tariquidar. Subsequently,
to test the kinetics of inhibition, tariquidar was used at different concen-
trations ranging from 0 to 100 nM. IC50 values were calculated from the
compared with untransduced cells (red traces).
The mouse homolog of human P-gp (mdr1a) has 87% sequence identity.
For comparison of transport profiles, mouse P-gp was expressed by using
BacMam baculovirus on the surface of HeLa cells, and a transport assay
curve generated using a variable slope nonlinear regression curve fit.
was carried out. As shown in Supplemental Fig. S3, mouse P-gp transports
The efflux of both the compounds was inhibited by tariquidar with
both BD-CsA and NBD-CsA at similar levels as human P-gp. Transport
comparable IC50 values of 21.6 6 0.6 nM for BD-CsA and 25.8 6 1.5 nM
for NBD-CsA (Fig. 6C). Complete inhibition of transport was observed
at .100 nM tariquidar.
of BD-verapamil as a control is also shown. Since BD-CsA is transported
by human and mouse P-gp to the same extent, subsequent experiments
were carried out with only human P-gp. We also tested the specificity of
these substrates for two other major ABC drug transporters, MRP1 and
ABCG2. As shown in Supplemental Fig. S4, ABCG2 does not efflux
either BD-CsA or NBD-CsA, although normal efflux of its substrate
To label the cells with BD-CsA and NBD-CsA and visualize the
pheophorbide A was observed. These data are consistent with an earlier
efflux, HeLa cells transduced with BacMam baculovirus expressing
report showing lack of transport of NBD-CsA by ABCG2 (Ejendal and
human P-gp were plated in 24-well plates. The monolayer of HeLa cells
Hrycyna, 2005). MRP1 shows marginal efflux of both BD-CsA and
Monitoring the BD-CsA Efflux from HeLa Cells in a Monolayer by
Fluorescence Microscopy
NBD-CsA, indicating that fluorescent conjugates of CsA are poor
substrates of MRP1 compared with calcein-AM, a known substrate.
Comparison of the Rate of Efflux of BD-CsA and NBD-CsA by P-gp
After confirming that BD-CsA and NBD-CsA are transported by P-gp
in a steady-state assay, we compared the rate of efflux for both the
compounds. HeLa cells were first depleted of ATP, as described in
Materials and Methods, followed by loading with fluorescent substrates
under ATP-depleted conditions. Because of the depletion of ATP, P-gp
was not active, and both the substrates diffused in the cells to reach an
equilibrium. A time-course efflux assay (0–30 minutes) was performed
after the addition of IMDM medium containing glucose. Fluorescence
intensity at time 0 (no efflux) was taken as 100%, and relative efflux
at indicated time points was calculated. As shown in Fig. 5, the percent
remaining BD-CsA and NBD-CsA (minimum remaining in cells) at
Fig. 5. The rate of efflux by human P-gp is similar for BD-CsA and NBD-CsA. Time-
dependent efflux of BD-CsA (blue circles) with NBD-CsA (red squares). Fluorescence
30 minutes are almost the same (BD-CsA 5 15.3 67.8% and NBD-CsA 5
17.3 6 4.9%; mean 6 S.D. from three independent experiments), and both
BD-CsA and NBD-CsA are effluxed at a similar rate by P-gp with T1/2 of
5.5 6 1.5 and 3.4 6 0.7 minutes, respectively. As expected, untransduced
cells did not show significant efflux of either substrate (blue and red dashed
intensity of untransduced cells and cells expressing human P-gp at the 0 time point was
taken as 100%, and relative fluorescence was calculated for 30 minutes. A similar
extent of efflux from P-gp-expressing cells was observed for both BD-CsA and NBD-
CsA with T1/2 of 5.5 6 1.5 and 3.4 6 0.7 minutes, respectively. Untransduced cells
did not show significant efflux of either substrate (blue and red dashed lines). The
lines in Fig. 5), showing that the efflux is specifically due to P-gp activity. mean values from three independent experiments are given, and error bars show S.D.

Development of ABCB1 Fluorescent Substrate Probe
1019
Fig. 6. Inhibition of BD-CsA and NBD-CsA transport by tariquidar. Histogram traces showing the steady-state efflux of BD-CsA (A) and NBD-CsA (B) by P-gp (blue
traces), and inhibition of efflux by tariquidar (200 nM) (gray traces). The fluorescence intensity was compared with the cells not expressing P-gp (untransduced, red traces).
(C) Graph plot showing the tariquidar concentration-dependent inhibition of efflux by P-gp and comparison of BD-CsA (blue circles) with NBD-CsA (red squares). The IC50
for tariquidar was 21.6 6 0.6 nM for BD-CsA and 25.8 6 1.5 nM for NBD-CsA. The experiment was repeated three times, and error bars show S.D.
was incubated with BD-CsA and NBD-CsA for 45 minutes, and nuclei were that CsA partially inhibits the ATPase activity of P-gp (Kerr et al., 2001;
stained with NucBlue. The fluorescence intensity of the cells owing to the Vahedi et al., 2017; Sajid et al., 2018). Here, we investigated whether
presence of NBD-CsA and BD-CsA was determined by using a fluorescence BD-CsA modulates ATPase activity in the same way as NBD-CsA. An
microscope. Untransduced cells showed a higher level of staining with both ATPase assay was carried out using increasing concentrations of the
BD-CsA (Fig. 7) and NBD-CsA (Fig. 8), whereas cells expressing P-gp substrates incubated with the protein at 37°C, and ATP hydrolysis was
showed negligible intracellular fluorescence attributable to their efflux. measured in a colorimetric assay. As shown in Supplemental Fig. S5,
To substantiate these results further, tariquidar was used to block the trans- both BD-CsA and NBD-CsA inhibit the activity of P-gp ATPase to
port function of P-gp. As evident, the addition of tariquidar increases the a similar extent, with maximum inhibition observed to be almost 60%
cellular fluorescence intensity of both BD-CsA and NBD-CsA, showing at 2.5 mM and comparable IC50 values of 30 6 2.9 nM for BD-CsA and
efflux of these compounds specifically by P-gp (Figs. 7 and 8).
17 6 2.6 nM for NBD-CsA. Thus, the BD-conjugated probe shows
properties similar to those of NBD-conjugated CsA.
Effect of BD-CsA and NBD-CsA on ATPase Activity
Transport of BD-CsA and NBD-CsA by P-gp Mutants
Substrates of P-gp usually affect the ability of the transporter to
hydrolyze ATP. Some substrates stimulate the ATPase activity, whereas
In our previous studies, we generated two mutants of P-gp, named
some inhibit it or have no effect. In our previous studies, we have shown 15Y and TMH1,7. The 15Y mutant has 15 conserved residues in the
Fig. 7. Monitoring transport of BD-CsA by fluorescence
microscopy. HeLa cells expressing P-gp were used for
labeling with BD-CsA and to assess its export by P-gp.
After transduction with BacMam baculovirus in 24-well
plates, cells in monolayer were incubated with BD-CsA
and NucBlue stain (nuclear staining) and visualized using
a phase-contrast microscope. Untransduced HeLa cells not
expressing P-gp are on the left, HeLa cells expressing P-gp
in the center, and HeLa cells expressing P-gp in the presence
of 200 nM tariquidar (P-gp inhibitor) are on the right. All
three sets were analyzed in transmitted light (top row).
NucBlue-stained UV region for visualization of the nucleus
(middle row), BD-CsA (0.5 mM)-stained GFP region (bottom
row). All images were taken at original magnification, 10Â;
scale bar, 200 mm.

1020
Sajid et al.
Fig. 8. Monitoring transport of NBD-CsA by fluorescence
microscopy. HeLa cells transduced with P-gp were used for
labeling with NBD-CsA and to measure its export by P-gp.
After transduction in 24-well plates, cells in monolayer
were incubated with 0.5 mM NBD-CsA and NucBlue stains
(nuclear staining) and visualized using a phase-contrast
microscope. Untransduced HeLa cells not expressing
P-gp are on the left, HeLa cells expressing P-gp are in
the center, and HeLa cells expressing P-gp with 200 nM
tariquidar (P-gp inhibitor) are on the right. All three sets
were analyzed in transmitted light (top row); NucBlue-
stained UV region for visualization of the nucleus (middle
row) and NBD-CsA (0.5 mM)-stained GFP region (bottom
row). All images were taken at original magnification 10Â;
scale bars, 200 mm.
drug-binding pocket substituted with tyrosine. These mutations affect of P-gp, owing to the flexible nature of the transmembrane region (Alam
the transport of large (.1000 Da) substrates only, including NBD-CsA, et al., 2019). This could also be the case for CsA and other molecules.
by 15Y mutant P-gp (Vahedi et al., 2017). The TMH1,7 mutant harbors We then focused on the pose with the lowest energy score for each of the
12 mutations, six in both transmembrane helices 1 and 7, which leads to molecules. As expected by the large size of the ligand, several amino
loss of polyspecificity of the transporter, as it can transport only 3 out
of 25 substrates (Sajid et al., 2018), including NBD-CsA. We tested
whether these mutants recognize BD-CsA in the same way as NBD-CsA.
As shown in Fig. 9A, the 15Y mutant does not efflux either BD-CsA or
NBD-CsA (same fluorescence as untransduced cells), whereas TMH1,7
efficiently effluxes both of the probes (.75% efficiency as compared
with wild-type P-gp) (Fig. 9B). Thus, BD conjugation does not affect the
recognition of CsA by P-gp mutants.
Docking of CsA, NBD-CsA, and BD-CsA in the Drug-Binding
Pocket of P-gp
To compare the interaction of NBD-CsA and BD-CsA with P-gp, we
docked these two molecules and CsA (parent unconjugated compound)
in the drug-binding pocket. These studies were carried out with the recently
published structure of ligand (Taxol)-bound P-gp (PDB ID:6QEX) as
a template. For the purpose of docking, we defined a box that included
all the residues in the drug-binding pocket within the transmembrane
region of P-gp. AutoDock Vina (Scripps Research Institute) generated
a total of nine different poses (Fig. 10). By examining the group of
docking poses, we found that for both BD- and NBD-CsA, often the
fluorophore was found within the inner leaflet of the membrane,
toward the intracellular region. Supplemental Table S1 shows the
docking scores for all three ligands. We found that all the poses (for CsA
and fluorophore-conjugated CsA) had comparable docking scores for
the lowest energy pose (214.5 kcal/mol). Particularly for CsA, the two
lowest energy poses had identical docking scores; however, these poses
differed significantly in the relative orientation of CsA, even though they
Fig. 9. Transport of BD-CsA and NBD-CsA by P-gp mutants. Transport of NBD-
shared many of the interacting residues (data not shown). The high-
resolution atomic structure of human P-gp in the presence of paclitaxel
(Taxol) revealed a structure in which the ligand could be fitted to
the density in more than one orientation, supporting the possibility
of different binding modes for the same molecule within the binding site
CsA (A and B, left) and BD-CsA (A and B, right) by 15Y and TMH 1,7 P-gp
mutants. In previous studies, we showed that NBD-CsA is not transported by 15Y
(Vahedi et al., 2017), but it is transported by TMH 1,7 mutant P-gp (Sajid et al.,
2018). Compared with untransduced cells (not expressing P-gp, orange traces), the
fluorescence intensity of HeLa cells expressing P-gp (blue traces) was taken as
100%, and relative transport by mutant P-gp (red traces) was calculated.

Development of ABCB1 Fluorescent Substrate Probe
1021
measure multiple types of readouts using techniques such as flow
cytometry, microscopy, and spectroscopy. Fluorescently labeled drugs
and chemicals are useful for in vitro assays to study the functions of
proteins, cellular signaling pathways, cell division, labeling of organ-
elles, metabolic functions, nucleotide labeling, etc. These probes are
used for in vivo research applications as well, such as live cell labeling;
intravital imaging; labeling of organs in animal models; localization of
tumors, stones, or infections; and labeling of transporters at the blood-brain
barrier.
As P-gp is associated with the development of multidrug resistance
in cancer, understanding its function is essential. Additionally, P-gp is
a highly conserved protein across several species, with homologs present
in popular animal models, such as the mouse (87% identity with Mdr1a)
and zebrafish (64% identity with Abcb4). Several in vitro and in vivo
techniques can be used to study different aspects of P-gp, such as
expression, topology, transport mechanism, and ATP hydrolysis. As
these methods can be applied to a broad range of species to study
P-gp, the synthesis of relevant fluorescent substrate probes plays an
important part in elucidating the function of this transporter.
CsA, being a cyclic peptide of ;1200 Da, is a unique substrate of
P-gp. Therefore, synthesizing its fluorescent conjugate is valuable.
Additionally, as P-gp substrates, fluorescent compounds can be useful
to study the drug pharmacokinetics, biodistribution, cytotoxicity, and
characterization of drug-resistant cancer cell lines. CsA conjugated to an
NBD fluorophore has been used as a P-gp substrate in only a limited
number of studies. We developed a simplified approach for the synthesis
of Bodipy-FL-CsA. Our approach has the advantage that the fluorophore
is conjugated during the last synthesis step, allowing the option of intro-
ducing other functionalities at the D8-lysine, if needed, by a simple
amide formation. During synthesis of the CsA conjugates, we made
two significant modifications to the reported synthesis of D-Lys-cyclosporine
(Eberle and Nuninger, 1993). Thionation, using the pyridine-phosphorous
pentasulfide complex (Fig. 2, step B), resulted in a much cleaner thioamide
product compared with the reported procedure. During the cyclization step k,
the «-amine of D-lysine was protected, as its Fmoc derivative provided
a UV-positive chromophore, thereby making the separation of the
required product easier by using column chromatography.
In this study, we used Bodipy-FL, a well characterized fluorophore,
and compared it with NBD, both covalently linked to CsA, a P-gp
substrate. Both fluorescent moieties can be excited and detected with
similar filter settings (NBDex 5 468 nm and NBDem 5 538 nm; BDex 5
470 nm and BDem 5 520 nm), but BD-CsA has sharper emission spectra
compared with NBD-CsA (Supplemental Fig. S6). The Bodipy fluores-
cent group has several advantageous optical properties over NBD. Its
long excited-state half-life is useful for fluorescence polarization studies
(Ulrich et al., 2008). It has a higher extinction coefficient and quantum
yield (extinction coefficient .80,000 cm21 M21, w ; 1.0, independent
of the solvent); it is more stable, has a smaller Stokes shift, and is
relatively easier to synthesize as a conjugated probe (Johnson et al.,
1991). In comparison, the NBD moiety has a relatively low extinction
coefficient and quantum yield (.22,000 cm21 M21, w ; 0.018, in water
but increases to 0.3–0.4 in organic solvents). Thus, BD-conjugated
chemicals or drugs are brighter and show less to no background noise. In
addition, the BD conjugate does not affect the permeability of associated
molecules across biologic membranes. These properties are particularly
useful for in vivo experiments. In past studies, NBD-CsA has been used
to study P-gp function in renal tubules and brain capillaries (blood-brain
barrier) (Ott et al., 2010; Miller, 2014). It would be beneficial to compare
such studies with BD-CsA, with better fluorescence that allows the use
Fig. 10. Docking of CsA, BD-CsA, and NBD-CsA in the drug-binding pocket of
human P-gp. Cartoon representation of the transmembrane region of human P-gp
(PDB ID: 6QEX) in the inward-open conformation and docked ligands (A) CsA, (B)
BD-CsA, and (C) NBD-CsA. Transmembrane helices 9–12 were removed for
clarity. The fluorophore is highlighted in blue (Bodipy) and red (NBD); the CsA
backbone is black. Docking of ligands using the Autodock Vina program, and the
cryo-EM structure of human P-gp was performed as described in the Materials and
Methods section. The image was prepared using Pymol.
acids of P-gp are close enough to interact with the ligands (Supplemental
Table S2). Although the relative orientation of the ligands is not the same,
they all interact comparably well with residues in the drug-binding pocket
of P-gp. We also replaced boron with carbon in the bodipy group for
docking of BD-CsA. Although the substitution of boron with carbon has
been used by others (Naaresh Reddy and Giri, 2016; Verwilst et al.,
2017; Zhao et al., 2017; Bonacorso et al., 2018) and by us (Fig. 10), the
interaction of boron compared with carbon in the BD-CsA with residues in
the drug-binding pocket of P-gp might be different. Resolution of the
atomic structure of P-gp bound to BD-CsA would help to resolve this issue.
Discussion
With the availability of a vast number of fluorescence-based technol- of lower concentrations.
ogies, the use of fluorescent probes in the development of sensitive, as well
We used HeLa cells expressing P-gp after transduction with the
as quantitative, assays has become common. These assays can be used to BacMam baculovirus to study the transport of BD-CsA. Flow cytometry

1022
Sajid et al.
Bergman J, Pettersson B, Hasimbegovic V, and Svensson PH (2011) Thionations using
a P4S10-pyridine complex in solvents such as acetonitrile and dimethyl sulfone. J Org
Chem 76:1546–1553.
Bonacorso HG, Calheiro TP, Iglesias BA, Acunha TV, Franceschini SZ, Ketzer A, Meyer AR,
Rodrigues LV, Nogara PA, Rocha JBT, et al. (2018) 1,1-Difluoro-3-aryl(heteroaryl)-1H-pyrido
[1,2-c][1,3,5,2]oxadiazaborinin-9-ium-1-uides: synthesis; structure; and photophysical, electro-
chemical, and BSA-binding studies. New J Chem 42:1913–1920.
Cascorbi I (2011) P-glycoprotein: tissue distribution, substrates, and functional consequences of
genetic variations. Handb Exp Pharmacol 201:261–283.
Chearwae W, Anuchapreeda S, Nandigama K, Ambudkar SV, and Limtrakul P (2004) Biochemical
mechanism of modulation of human P-glycoprotein (ABCB1) by curcumin I, II, and III purified
from Turmeric powder. Biochem Pharmacol 68:2043–2052.
Chen G, Durán GE, Steger KA, Lacayo NJ, Jaffrézou JP, Dumontet C, and Sikic BI (1997)
Multidrug-resistant human sarcoma cells with a mutant P-glycoprotein, altered phenotype, and
resistance to cyclosporins. J Biol Chem 272:5974–5982.
Chufan EE, Kapoor K, Sim HM, Singh S, Talele TT, Durell SR, and Ambudkar SV (2013)
Multiple transport-active binding sites are available for a single substrate on human P-glycoprotein
(ABCB1) (Abstract). PLoS One 8:e82463.
Chufan EE, Sim HM, and Ambudkar SV (2015) Molecular basis of the polyspecificity of P-glycoprotein
(ABCB1): recent biochemical and structural studies. Adv Cancer Res 125:71–96.
Demeule M, Laplante A, Murphy GF, Wenger RM, and Béliveau R (1998) Identification of the
cyclosporin-binding site in P-glycoprotein. Biochemistry 37:18110–18118.
Demeule M, Laplante A, Sepehr-Araé A, Beaulieu E, Averill-Bates D, Wenger RM, and Béliveau
R (1999) Inhibition of P-glycoprotein by cyclosporin A analogues and metabolites. Biochem Cell
Biol 77:47–58.
Dirks NL, Huth B, Yates CR, and Meibohm B (2004) Pharmacokinetics of immunosuppressants:
a perspective on ethnic differences. Int J Clin Pharmacol Ther 42:701–718.
Eberle MK and Nuninger F (1993) Preparation of [D-cysteine]8-cyclosporin via intramolecular
sulfur transfer reaction. J Org Chem 58:673–677.
Ejendal KF and Hrycyna CA (2005) Differential sensitivities of the human ATP-binding cassette
transporters ABCG2 and P-glycoprotein to cyclosporin A. Mol Pharmacol 67:902–911.
Gribar JJ, Ramachandra M, Hrycyna CA, Dey S, and Ambudkar SV (2000) Functional charac-
terization of glycosylation-deficient human P-glycoprotein using a vaccinia virus expression
system. J Membr Biol 173:203–214.
Gutmann DA, Ward A, Urbatsch IL, Chang G, and van Veen HW (2010) Understanding poly-
specificity of multidrug ABC transporters: closing in on the gaps in ABCB1. Trends Biochem Sci
35:36–42.
Hartz AM, Miller DS, and Bauer B (2010) Restoring blood-brain barrier P-glycoprotein reduces
brain amyloid-beta in a mouse model of Alzheimer’s disease. Mol Pharmacol 77:715–723.
Jetté L, Beaulieu E, Leclerc JM, and Béliveau R (1996) Cyclosporin A treatment induces over-
expression of P-glycoprotein in the kidney and other tissues. Am J Physiol 270:F756–F765.
Johnson ID, Kang HC, and Haugland RP (1991) Fluorescent membrane probes incorporating
dipyrrometheneboron difluoride fluorophores. Anal Biochem 198:228–237.
Jouan E, Le Vée M, Mayati A, Denizot C, Parmentier Y, and Fardel O (2016) Evaluation of
P-glycoprotein inhibitory potential using a rhodamine 123 accumulation assay (Abstract).
Pharmaceutics 8:E12.
and microscopy-based assays were used to compare the kinetics of
transport of BD-CsA with NBD-CsA and its inhibition by tariquidar. We
show that BD-CsA is a much better probe in terms of its fluorescence
intensity, requiring a lower concentration of the compound. Thus,
BD-CsA can be used to label live cells, as shown in microscopy-based
assays, using concentrations an order of magnitude lower than those of
NBD-CsA. The kinetics of transport show that BD-CsA is efficiently
exported out of the cells by P-gp with a T1/2 of 5.5 minutes, which is
comparable to the NBD-CsA T1/2 of 3.4 minutes.
In our recent studies, we characterized the transport profile of P-gp
mutants with altered properties. We found that 15Y mutant P-gp could
not transport large substrates, such as NBD-CsA, whereas TMH1,7
mutant P-gp could transport only three of the substrates tested, including
NBD-CsA (Vahedi et al., 2017; Sajid et al., 2018). In this study, we
observed that the 15Y mutant failed to transport BD-CsA, whereas the
TMH1,7 mutant transported it to same extent as NBD-CsA (Fig. 9).
Thus, BD-CsA can be used for characterization of P-gp mutants with
substitutions in the drug-binding pocket to understand the transport
mechanism.
In silico docking experiments show that both BD- and NBD-CsA bind
to the substrate-binding pocket in the transmembrane region of P-gp,
interacting with numerous residues (Supplemental Table S2). We found
that the docking scores are comparable for CsA, BD-CsA, and
NBD-CsA, and conjugation of either fluorophore does not interfere
with the binding and transport of CsA by P-gp (Fig. 10; Supplemental
Table S1). We found that for residues within 5 Å of the ligand (Fig. 10;
Supplemental Tables S1 and S2), interaction with the ligand increased
with the increase in the molecular weight of the molecule. This
included 34 residues for CsA, 42 residues for NBD-CsA, and 46
residues for BD-CsA. In addition, most of the residues that interacted
with the parent CsA moiety were shared by all three molecules, whereas
the residues that interact with the NBD or BD fluorophores were
different.
Juliano RL and Ling V (1976) A surface glycoprotein modulating drug permeability in Chinese
hamster ovary cell mutants. Biochim Biophys Acta 455:152–162.
Kelly P and Kahan BD (2002) Review: metabolism of immunosuppressant drugs. Curr Drug
Metab 3:275–287.
Kerr KM, Sauna ZE, and Ambudkar SV (2001) Correlation between steady-state ATP hydrolysis
and vanadate-induced ADP trapping in human P-glycoprotein. Evidence for ADP release
as the rate-limiting step in the catalytic cycle and its modulation by substrates. J Biol Chem 276:
8657–8664.
In conclusion, we synthesized a Bodipy-FL conjugate of CsA and
demonstrated that it is a stable probe with high fluorescence yield for
monitoring the transport function of P-gp. It has the potential to become
the probe of choice for in vivo experiments in animal models, including
the mouse and zebrafish.
Kim Y and Chen J (2018) Molecular structure of human P-glycoprotein in the ATP-bound,
outward-facing conformation. Science 359:915–919.
Li MJ, Nath A, and Atkins WM (2017) Differential coupling of binding, ATP hydrolysis, and
transport of fluorescent probes with P-glycoprotein in lipid nanodiscs. Biochemistry 56:
2506–2517.
Martin C, Berridge G, Mistry P, Higgins C, Charlton P, and Callaghan R (1999) The molecular
interaction of the high affinity reversal agent XR9576 with P-glycoprotein. Br J Pharmacol 128:
403–411.
Masereeuw R, Terlouw SA, van Aubel RA, Russel FG, and Miller DS (2000) Endothelin B
receptor-mediated regulation of ATP-driven drug secretion in renal proximal tubule.
Mol Pharmacol 57:59–67.
Acknowledgments
We thank Kristina Folta and Vivian Lin for technical help and George Leiman
for editing the manuscript. The high-performance computational capabilities of
the Helix and Biowulf Systems at the National Institutes of Health, Bethesda, were
used for docking studies.
Miller DS (2014) Sphingolipid signaling reduces basal P-glycoprotein activity in renal proximal
tubule. J Pharmacol Exp Ther 348:459–464.
Authorship Contributions
Participated in research design: Sajid, Raju, Swenson, Ambudkar.
Conducted experiments: Sajid, Raju, Lusvarghi, Vahedi.
Contributed new reagents or analytic tools: Sajid, Raju, Swenson, Ambudkar.
Performed data analysis: Sajid, Raju, Lusvarghi, Vahedi, Swenson, Ambudkar.
Wrote or contributed to the writing of the manuscript: Sajid, Raju, Lusvarghi,
Swenson, Ambudkar.
Müller M, Yong M, Peng XH, Petre B, Arora S, and Ambudkar SV (2002) Evidence for the role of
glycosylation in accessibility of the extracellular domains of human MRP1 (ABCC1). Bio-
chemistry 41:10123–10132.
Muzi M, Mankoff DA, Link JM, Shoner S, Collier AC, Sasongko L, and Unadkat JD (2009)
Imaging of cyclosporine inhibition of P-glycoprotein activity using 11C-verapamil in the brain:
studies of healthy humans. J Nucl Med 50:1267–1275.
Naaresh Reddy G and Giri S (2016) Super/hyperhalogen aromatic heterocyclic compounds. RSC
Advances 6:47145–47150.
Nicklisch SC, Rees SD, McGrath AP, Gökirmak T, Bonito LT, Vermeer LM, Cregger C, Loewen
G, Sandin S, Chang G, et al. (2016) Global marine pollutants inhibit P-glycoprotein: environ-
mental levels, inhibitory effects, and cocrystal structure (Abstract). Sci Adv 2:e1600001.
Ott M, Huls M, Cornelius MG, and Fricker G (2010) St. John’s Wort constituents modulate
P-glycoprotein transport activity at the blood-brain barrier. Pharm Res 27:811–822.
Pawarode A, Shukla S, Minderman H, Fricke SM, Pinder EM, O’Loughlin KL, Ambudkar SV,
and Baer MR (2007) Differential effects of the immunosuppressive agents cyclosporin A,
tacrolimus and sirolimus on drug transport by multidrug resistance proteins. Cancer Chemother
Pharmacol 60:179–188.
Pluchino KM, Hall MD, Moen JK, Chufan EE, Fetsch PA, Shukla S, Gill DR, Hyde SC, Xia D,
Ambudkar SV, et al. (2016) Human-mouse chimeras with normal expression and function reveal
that major domain swapping is tolerated by P-glycoprotein (ABCB1). Biochemistry 55:1010–1023.
Ramachandra M, Ambudkar SV, Chen D, Hrycyna CA, Dey S, Gottesman MM, and Pastan I
(1998) Human P-glycoprotein exhibits reduced affinity for substrates during a catalytic transition
state. Biochemistry 37:5010–5019.
References
Alam A, Kowal J, Broude E, Roninson I, and Locher KP (2019) Structural insight into substrate
and inhibitor discrimination by human P-glycoprotein. Science 363:753–756.
Alam A, Küng R, Kowal J, McLeod RA, Tremp N, Broude EV, Roninson IB, Stahlberg H,
and Locher KP (2018) Structure of a zosuquidar and UIC2-bound human-mouse chimeric
ABCB1. Proc Natl Acad Sci USA 115:E1973–E1982.
Ambudkar SV, Dey S, Hrycyna CA, Ramachandra M, Pastan I, and Gottesman MM (1999)
Biochemical, cellular, and pharmacological aspects of the multidrug transporter. Annu Rev
Pharmacol Toxicol 39:361–398.
Ambudkar SV, Kim IW, and Sauna ZE (2006) The power of the pump: mechanisms of action of
P-glycoprotein (ABCB1). Eur J Pharm Sci 27:392–400.
Barbarino JM, Staatz CE, Venkataramanan R, Klein TE, and Altman RB (2013) PharmGKB
summary: cyclosporine and tacrolimus pathways. Pharmacogenet Genomics 23:563–585.

Development of ABCB1 Fluorescent Substrate Probe
1023
Saeki T, Ueda K, Tanigawara Y, Hori R, and Komano T (1993) Human P-glycoprotein transports
cyclosporin A and FK506. J Biol Chem 268:6077–6080.
Sajid A, Lusvarghi S, Chufan EE, and Ambudkar SV (2018) Evidence for the critical role of
transmembrane helices 1 and 7 in substrate transport by human P-glycoprotein (ABCB1). PLoS
One 13:e0204693.
Thiebaut F, Tsuruo T, Hamada H, Gottesman MM, Pastan I, and Willingham MC (1987) Cellular
localization of the multidrug-resistance gene product P-glycoprotein in normal human tissues.
Proc Natl Acad Sci USA 84:7735–7738.
Trott O and Olson AJ (2010) AutoDock Vina: improving the speed and accuracy of docking with
a new scoring function, efficient optimization, and multithreading. J Comput Chem 31:455–461.
Ulrich G, Ziessel R, and Harriman A (2008) The chemistry of fluorescent bodipy dyes: versatility
unsurpassed. Angew Chem Int Ed Engl 47:1184–1201.
Sanner MF (1999) Python: a programming language for software integration and development.
J Mol Graph Model 17:57–61.
Schramm U, Fricker G, Wenger R, and Miller DS (1995) P-glycoprotein-mediated secretion of
a fluorescent cyclosporin analogue by teleost renal proximal tubules. Am J Physiol 268:F46–F52.
Shen DW, Fojo A, Chin JE, Roninson IB, Richert N, Pastan I, and Gottesman MM (1986) Human
multidrug-resistant cell lines: increased mdr1 expression can precede gene amplification. Science
232:643–645.
Shukla S, Schwartz C, Kapoor K, Kouanda A, and Ambudkar SV (2012) Use of baculovirus
BacMam vectors for expression of ABC drug transporters in mammalian cells. Drug Metab
Dispos 40:304–312.
Shukla S, Skoumbourdis AP, Walsh MJ, Hartz AM, Fung KL, Wu CP, Gottesman MM, Bauer B,
Thomas CJ, and Ambudkar SV (2011) Synthesis and characterization of a BODIPY conjugate of
the BCR-ABL kinase inhibitor Tasigna (nilotinib): evidence for transport of Tasigna and its
fluorescent derivative by ABC drug transporters. Mol Pharm 8:1292–1302.
Storck SE, Hartz AMS, Bernard J, Wolf A, Kachlmeier A, Mahringer A, Weggen S, Pahnke J,
and Pietrzik CU (2018) The concerted amyloid-beta clearance of LRP1 and ABCB1/P-gp across
the blood-brain barrier is linked by PICALM. Brain Behav Immun 73:21–33.
Strouse JJ, Ivnitski-Steele I, Waller A, Young SM, Perez D, Evangelisti AM, Ursu O, Bologa CG,
Carter MB, Salas VM, et al. (2013) Fluorescent substrates for flow cytometric evaluation of
efflux inhibition in ABCB1, ABCC1, and ABCG2 transporters. Anal Biochem 437:77–87.
Szewczyk P, Tao H, McGrath AP, Villaluz M, Rees SD, Lee SC, Doshi R, Urbatsch IL, Zhang Q,
and Chang G (2015) Snapshots of ligand entry, malleable binding and induced helical movement
in P-glycoprotein. Acta Crystallogr D Biol Crystallogr 71:732–741.
Vahedi S, Chufan EE, and Ambudkar SV (2017) Global alteration of the drug-binding pocket of
human P-glycoprotein (ABCB1) by substitution of fifteen conserved residues reveals a negative
correlation between substrate size and transport efficiency. Biochem Pharmacol 143:53–64.
Vahedi S, Lusvarghi S, Pluchino K, Shafrir Y, Durell SR, Gottesman MM, and Ambudkar SV
(2018) Mapping discontinuous epitopes for MRK-16, UIC2 and 4E3 antibodies to extracellular
loops 1 and 4 of human P-glycoprotein. Sci Rep 8:12716.
Verwilst P, Kim HR, Seo J, Sohn NW, Cha SY, Kim Y, Maeng S, Shin JW, Kwak JH, Kang C,
et al. (2017) Rational design of in vivo tau tangle-selective near-infrared fluorophores: expanding
the BODIPY universe. J Am Chem Soc 139:13393–13403.
Weiss J, Kerpen CJ, Lindenmaier H, Dormann SM, and Haefeli WE (2003) Interaction of
antiepileptic drugs with human P-glycoprotein in vitro. J Pharmacol Exp Ther 307:262–267.
Wenger RM (1986) Cyclosporine and analogues--isolation and synthesis--mechanism of action and
structural requirements for pharmacological activity. Fortschr Chem Org Naturst 50:123–136.
Wenger RM (1988) Cyclosporine: conformation and analogues as tools for studying its mechanism
of action. Transplant Proc 20 (2 Suppl 2):313–318.
Yigitaslan S, Erol K, and Cengelli C (2016) The effect of P-glycoprotein inhibition and activation
on the absorption and Serum levels of cyclosporine and tacrolimus in rats. Adv Clin Exp Med 25:
237–242.
Zhao N, Williams TM, Zhou Z, Fronczek FR, Sibrian-Vazquez M, Jois SD, and Vicente MGH (2017)
Synthesis of BODIPY-peptide conjugates for fluorescence labeling of EGFR overexpressing
cells. Bioconjug Chem 28:1566–1579.
}
Szöllosi D, Rose-Sperling D, Hellmich UA, and Stockner
T (2018) Comparison of
mechanistic transport cycle models of ABC exporters. Biochim Biophys Acta Biomembr
1860:818–832.
Address correspondence to: Suresh V. Ambudkar, National Cancer Institute, NIH,
Building 37, Room 2120, 37 Convent Drive, Bethesda, MD 20892-4256. E-mail:
ambudkar@mail.nih.gov
Tanaka K, Hirai M, Tanigawara Y, Yasuhara M, Hori R, Ueda K, and Inui K (1996) Effect of
cyclosporin analogues and FK506 on transcellular transport of daunorubicin and vinblastine via
P-glycoprotein. Pharm Res 13:1073–1077.
Tedesco D and Haragsim L (2012) Cyclosporine: a review. J Transplant 2012:230386.