Design, synthesis of methotrexate-diosgenin conjugates and biological evaluation of their effect on methotrexate transport-resistant cells

Article abstract of DOI:10.1016/j.steroids.2016.10.006

A series of methotrexate-diosgenin conjugates was designed and synthesized to enhance the passive internalization of methotrexate (MTX) into transport-resistant cells. The inhibitory effects of these conjugates on dihydrofolate reductase (DHFR), and their anti-proliferation behaviors against a transport-resistant breast cancer cell line, MDA-MB-231, were investigated. All of the synthesized conjugates retained an ability to inhibit DHFR after the diosgenin substitution. The MTX conjugates were much more potent against methotrexate-resistant MDA-MB-231 cells than MTX. Conjugate 18, containing a disulfide bond, exhibited the most potent anti-proliferative and DHFR inhibitory effects (IC₅₀?=?4.1?μM and 17.21?nM, respectively). Anti-proliferative activity was higher in the conjugate with a longer space linker (conjugate 21) than those with shorter linkers (conjugates 19 and 20). These results suggest that diosgenin conjugation of MTX may be an effective way to overcome its transport resistance in cancer cells.

Full text of DOI:10.1016/j.steroids.2016.10.006

Steroids 116 (2016) 45–51
Contents lists available at ScienceDirect
Steroids
journal homepage: www.elsevier.com/locate/steroids
Design, synthesis of methotrexate-diosgenin conjugates and biological
evaluation of their effect on methotrexate transport-resistant cells
Bangrong Cai a, Aimei Liao a, Kyung-Ku Lee b, Jae-Sam Ban b, Hyun-Sam Yang c, Young Jun Im a,
,
⇑
ChangJu Chun a,
⇑
a Research Institute of Drug Development, College of Pharmacy, Chonnam National University, Gwangju 61186, Republic of Korea
b Biomedical Material and Component Service Center, Gwangju Technology Park, Gwangju 62253, Republic of Korea
c Department of Automotive & Mechanical Engineering, Nambu University, Gwangju 506-706, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of methotrexate-diosgenin conjugates was designed and synthesized to enhance the passive
internalization of methotrexate (MTX) into transport-resistant cells. The inhibitory effects of these con-
jugates on dihydrofolate reductase (DHFR), and their anti-proliferation behaviors against a transport-
resistant breast cancer cell line, MDA-MB-231, were investigated. All of the synthesized conjugates
retained an ability to inhibit DHFR after the diosgenin substitution. The MTX conjugates were much more
potent against methotrexate-resistant MDA-MB-231 cells than MTX. Conjugate 18, containing a disulfide
Received 16 June 2016
Received in revised form 23 September
2016
Accepted 15 October 2016
Available online 19 October 2016
bond, exhibited the most potent anti-proliferative and DHFR inhibitory effects (IC50 = 4.1 lM and
Keywords:
Methotrexate
Diosgenin
Transport resistance
Methotrexate-diosgenin conjugates
MDA-MB-231 cells
17.21 nM, respectively). Anti-proliferative activity was higher in the conjugate with a longer space linker
(conjugate 21) than those with shorter linkers (conjugates 19 and 20). These results suggest that dios-
genin conjugation of MTX may be an effective way to overcome its transport resistance in cancer cells.
Ó 2016 Elsevier Inc. All rights reserved.
1. Introduction
Considering the limitations described above, we have assumed
that naturally occurring hydrophobic molecules through the
Methotrexate (MTX, compound 2, Fig. 1) is a major chemother-
apeutic agent used in the treatment of various types of cancers
such as malignant lymphoma, osteosarcoma, and breast cancer
[1,2]. However, therapy with MTX has several limitations such as
myelosuppression, nephrotoxicity, and gastrointestinal mucositis
[3]. Cellular factors that impede its clinical efficacy include dimin-
ished glutamate polymerization, decreased dihydrofolate reduc-
tase (DHFR) activity, elevated DHFR expression, and lack of
expression of a reduced folate carrier (RFC) that is a main trans-
porter for folate analogs [3–5]. A deficiency in RFC-mediated trans-
port is the primary and frequent mechanism of resistance to MTX,
and the main reason for discontinuation of treatment with this
drug [6]. Extensive efforts have been made to overcome folate
transporter-related resistance of MTX [1]. Lipophilic modification
hydrophobic modification of MTX would probably overcome the
RFC-mediated transport resistance. Diosgenin (compound 1,
Fig. 1), a naturally occurring steroidal saponin, is mainly isolated
from legumes and yams, and exhibits high biocompatibility and
various pharmacological activities such as anti-inflammation [7].
immune-protection [8], and anticancer [9] including that against
colon carcinoma, HCT-116 cells [10], and breast cancer cells [11].
DG is a biocompatible, biodegradable, and non-toxic compound
based on a previous toxicological and pharmacokinetic study
[12]. Besides, DG is a strong hydrophobic molecule (logP = 5.7)
[13] with cholesterol-like properties. Furthermore, DG was
employed recently to modify various therapeutic agents [14–16].
Herein, we report the synthesis and biological evaluation of new
series of MTX-DG conjugates via various polyamine linkers (Fig. 2).
of MTX at the c-carboxylic group of glutamate shows considerable
potential to overcome these limitations [7–9]. However, the thera-
peutic efficacy of the resulting conjugates is always limited by their
decreased solubility, reduced DHFR affinity resulting from the lipo-
philic substitutions of MTX, and inefficient release of MTX.
2. Experimental section
2.1. Materials and methods
Methotrexate (MTX, purity > 98%), anhydrous N-methylmor-
pholine (NMM), triphosgene, O-(benzotriazol-1-yl)-N,N,N0,N0-
tetramethyluronium tetrafluoroborate (TBTU), anhydrous dichlor-
omethane, and N-carboethoxyphthalimide were obtained from
⇑
Corresponding authors.
E-mail addresses: imyoungjun@jnu.ac.kr (Y.J. Im), cchun1130@jnu.ac.kr
(C. Chun).
http://dx.doi.org/10.1016/j.steroids.2016.10.006
0039-128X/Ó 2016 Elsevier Inc. All rights reserved.

46
B. Cai et al. / Steroids 116 (2016) 45–51
Fig. 1. Structures of diosgenin (1) and methotrexate (2).
reduced pressure and added 200 mL of hot ethanol with stirring.
After stirring for 5 h, the precipitate was filtered, collected, and
dried to give compound 7 in 95% yield. The spectrum pure com-
pound 7 can be obtained by flash chromatography eluting with
ethyl acetate. 1H NMR (300 MHz, CDCl3): d = 7.73 (m, 4H), 7.68
(m, 4H), 3.84–3.38 (t, J = 6.01, 4H), 3.30–2.97 (t, J = 6.01, 4H).
2.2.2. Synthesis of compound 8
Fig. 2. The polyamines used as linkers.
To a solution of N-carboethoxyphthalimide (5.0 g, 22.8 mmol)
and 6.35 mL of TEA in 50 mL of acetonitrile was added a solution
of compound 6 (1.33 g, 11.4 mmol) in 10 mL of acetonitrile at
30 °C. The mixture was stirred for 2 h and then concentrated under
vacuum to afford the bisphthalimide 8 in 65%. 1H NMR (300 MHz,
CD3OD): d = 7.89–7.84 (m, 4H), 7.82–7.79 (m, 4H), 3.90 (t,
J = 5.68 Hz, 4H), 3.30–3.12 (t, J = 5.68 Hz, 4H), 3.01 (s, 4H).
Tokyo Chemical Industry (Tokyo, Japan). Cystamine dihydrochlo-
ride, phthalic anhydride, anhydrous pyridine, di-tert-butyl dicar-
bonate, and anhydrous N,N-dimethylformide (DMF) were ordered
from Alfa Aesar (Haverhill, Massachusetts, USA). Trifluoroacetic
acid (TFA), 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetra-
zolium bromide (MTT), dihydrofolate reductase assay kit, tri-
ethylenetetramine (TETA), diethylenetriamine (DETA), diosgenin
(DG), and ethylenediamine (EDA) were purchased from Sigma-
Aldrich (St. Louis, MO, USA). RPMI 1640 medium, fetal bovine
serum, and an antibiotic-antimycotic (100ꢀ) solution were
obtained from Life Technologies (Carlsbad, CA, USA). All solvents
and reagents were chemical grade unless otherwise stated.
Thin layer chromatography with silica gel 60 F254 (Merck) was
performed to monitor the progress of each reaction. Spots were
visualized with ultraviolet light, a 5% sulfuric acid solution in etha-
nol, a ninhydrin solution, or iodine vapors. Flash chromatography
was carried out on Merck silica gel 60 (70–230 mesh) with the elu-
ent system presented below. Purification and purity determina-
tions of the prepared conjugates were performed by a Skyam
high-performance liquid chromatography (HPLC) system on a Luna
2.2.3. Synthesis of compound 9
To a suspension mixture of compound 7 (8 g, 22 mmol) and
K2CO3 (7.5 g, 55 mmol) in THF was added di-tert-butyl dicarbonate
(5.7 g, 26.4 mmol) at ice cooled bath. The reaction mixture was
warmed to room temperature and stirred for 10 h. The solvent
was removed and the residue was dissolved in 200 mL of DCM.
The organic layer was washed with H2O (2 ꢀ 200 mL), dried with
Na2SO4, filtered and concentrated. The crude product was recrys-
tallized from a mixture of methanol/n-hexane or using flash chro-
matography eluting with 10% ethyl acetate in hexane to give 8 as a
white power. 1H NMR (300 MHz, CDCl3): d = 7.83 (m, 4H), 7.68 (m,
4H), 3.84 (t, J = 6.01 Hz, 4H), 3.55 (t, J = 6.01 Hz, 4H), 1.06 (s, 9H).
2.2.4. Synthesis of compound 10
reversed-phase C18 column (250 ꢀ 4.6 mm, 5
lm), using a gradient
To a mixture suspension of compound 8 (6.0 g, 14.8 mmol) in
100 mL was added di-tert-butyl dicarbonate (7.1 g, 32.6 mmol) at
ice cold bath. The reaction mixture was stirred for 10 h at room
temperature. The solvent was removed and the residue was dis-
solved in 100 mL of DCM. The organic layer was washed with
H2O (2 ꢀ 200 mL) twice, dried with Na2SO4, and concentrated.
The crude product was purified by flash chromatography with elu-
tion of ethyl acetate/methanol/NH3ꢁH2O = 50/1/0.5 or recrystal-
of 0.05% TFA aqueous solution (A) and acetonitrile or methanol (B)
as the mobile phase at a constant flow rate of 1 mL/min. Detection
was at 302 nm. The elution gradient started at 70% A and ended at
30% A during 45 min, then decreased to 20% eluent B during the
next 15 min for acetonitrile system. For methanol system, the elu-
tion gradient (A) started at 90% and ended at 5% in within 10 min
and then constant elution was used for next 45 min. Test samples
were dissolved in 10% TFA in methanol or methanol alone and fil-
lized from
a mixture of methanol and hexane to afford
tered through a 0.2 lm nylon filter before injection. Structural
compound 10 in 90% yield as a white power. 1H NMR (300 MHz,
CDCl3): d = 7.84 (m, 4H), 7.67 (m, 4H), 3.83–3.79 (t, J = 5.79, 4H),
3.51 (t, J = 5.79, 4H), 3.34–3.27 (t, J = 5.79, 4H), 1.23 (s, 18H).
characterization of the synthetic products was done by a Varian
Unity Plus 400 MHz spectrometer in CDCl3 or DMSO-d6. All chem-
ical shifts (d) are reported in parts per million (ppm) and multiplic-
ity (s = singlet, d = doublet, dd = doublet of doublet, t = triplet,
q = quartet, and m = multiplet). The molecular weight of the final
conjugates was confirmed on a Shimadzu ultra-fast liquid chro-
matograph mass spectrometer (UFLC-MS) with the electron spray
ionization (ESI) method.
2.2.5. Synthesis of compound 11
Compound 9 (8.5 g, 18.7 mmol) was suspended in 200 mL of
ethanol and hydrazine monohydrate (18 mL, 20 equiv.) and stirred
for 1 day at room temperature. After the reaction, the precipitate
was filtered, and the filtrate was removed under reduced pressure
and diluted with DCM (100 mL). The organic phase was washed
with a small amount of saturated NaHCO3 solution, dried over
MgSO4, filtered and concentrated to give 11 as light yellow oil in
76% yield, which was purified by flash chromatography with elu-
tion of 5–20% methanol in chloroform containing 0.5% aqueous
ammonia. 1H NMR (300 MHz, CDCl3): d = 3.34 (br s, 4H), 2.93 (m,
4H), 1.46 (s, 9H).
2.2. Chemistry
2.2.1. Synthesis of compound 7
To a solution of compound 5 (3.20 g, 31.0 mmol) in glacial
acetic acid (200 mL) was added phthalic anhydride (11.0 g,
37.2 mmol) in solid form. The reaction mixture was refluxed at
80 °C for 2 h. The solvent was completely evaporated under

B. Cai et al. / Steroids 116 (2016) 45–51
47
2.2.6. Synthesis of compound 12
time = 41.37 min. Purity = 97%. 1H NMR (400 MHz, DMSO-d6):
d = 8.56 (s, 1H), 8.13 (d, J = 7.54 Hz, 1H), 7.73–7.71 (d, J = 8.90 Hz,
2H), 7.41 (s, 1H), 7.23 (s, 1H), 6.83–6.81 (d, J = 8.90 Hz, 2H), 6.61
(s, 2H), 5.32 (d, J = 5.14 Hz 1H), 4.78 (s, 2H), 4.31 (m, 3H), 3.39,
(dd, 2H), 3.20 (s, 3H), 2.76 (t, 4H), 2.28 (m, 2H), 2.18 (m, 2H),
0.96 (s, 3H), 0.90 (d, 3H), 0.73 (s, 6H). 13C NMR (100 MHz, DMSO-
d6): d = 173.71, 172.86, 166.59, 165.83, 163.30, 163.17, 156.14,
155.64, 151.33, 149.61, 146.46, 140.23, 129.48, 128.93, 122.34,
122.13, 121.90, 111.63, 108.87, 80.65, 79.63, 73.52, 66.39, 62.26,
56.17, 55.34, 53.96, 49.87, 41.56, 38.74, 38.44, 37.97, 37.65,
37.01, 32.63, 31.92, 31.40, 30.28, 28.95, 28.80, 28.31, 22.33,
20.83, 19.47, 17.55, 16.46, 15.12. HRMS (ESI): m/z 1027.6061
(Mꢂ1), calculated (C52H72N10O8S2) 1028.50.
Compound 10 (8.0 g, 13.3 mmol) and hydrazine monohydrate
(12 mL, 233 mmol) in 100 mL ethanol were stirred for 24 h at room
temperature. The precipitate was removed by filtration and filtrate
was concentrated under reduced pressure, and then diluted with
80 mL of DCM. The combined organic layers were washed with a
small amount of brine solution and then dried and filtered to give
compound 12 as light yellow oil in 80% yield. TLC (DCM/methanol/
NH3ꢁH2O = 10/3/0.01) the crude product was further purified by
flash chromatography with elution of 5% methanol in DCM and
trace of NH3ꢁH2O. 1H NMR (300 MHz, CD3OD): d = 3.59 (t, 4H),
3.40 (t, 4H), 3.19 (t, 4H), 1.38 (s, 18H).
2.2.7. General procedure for preparation of compound 14–17
To a stirred solution of amines 3, 4, 11 or 12 (10 equiv., 8 equiv.,
5 equiv. and 6 equiv., respectively) in 5 mL anhydrous DCM was
slowly added diosgenyl chloroformate (compound 13) solution
(500 mg, 1.05 mmol, 1 equiv.) obtained according to previously
reported method [19] (see details in Supporting information S1)
in 20 mL anhydrous DCM at ice cooled bath over 10 h, the reaction
mixture was allowed to warm to room temperature for 3 h. The
resulting mixture was diluted with 25 mL DCM, sequentially
washed with water, saturated aqueous NaCl solution (2 ꢀ 25 mL)
and dried over anhydrous MgSO4. The resulting mixture was fil-
trated and evaporated to afford crude products as a white solid.
The crude product was further purified through flash chromatogra-
phy with 0.5–1% methanol and aqueous ammonia in ethyl acetate
to give target compounds.
2.2.9. Synthesis of conjugate 19
The target compound was synthesized from MTX and com-
pound 15 using the same procedure described for conjugate 18.
Yield = 71%. Analytical HPLC, retention time = 29.71 min. Pur-
ity = 98.7%. 1H NMR (400 MHz, DMSO-d6, ppm): d = 8.55 (s, 1H),
7.91 (s, 1H), 7.77-7.75 (d, J = 8.87, 1H), 7.69–7.65 (d, J = 8.87, 1H),
7.41 (s, 1H), 7.08–7.02 (d,1H), 6.83–6.80 (dd, J = 8.90, 2H), 6.60
(s, 2H), 5.31 (s, J = 5.14 Hz, 1H), 4.78 (s, 2H), 4.29–4.26 (m, 3H),
3.42 (dd, 1H), 3.42–3.00 (m, 4H), 2.25–2.17 (m, 4H), 0.94 (s, 3H),
0.90 (d, 3H), 0.73 (s, 6H). 13C NMR (100 MHz, DMSO-d6):
d = 173.51, 172.92, 166.66, 165.93, 163.19, 163.14, 155.47,
151.21, 149.56, 146.58, 140.26, 129.49, 129.02, 122.28, 122.03,
121.90, 111.61, 108.87, 80.64, 73.40, 66.37, 62.24, 56.16, 55.29,
49.87, 41.54, 38.71, 37.00, 36.65, 32.60, 31.90, 31.39, 30.26,
28.93, 28.29, 27.80, 22.50, 19.43, 17.52, 16.43, 15.09. HRMS (ESI):
m/z 935.6147 (M-1), calculated (C50H68N10O8) 936.52.
2.2.7.1. Compound 14. 1H NMR (300 MHz, CDCl3): d = 5.35 (d,
J = 5.14 Hz, 1H), 4.40 (m, 1H), 4.37–4.31 (q, J = 7.41, 14.70 Hz,
1H), 3.52–3.31 (m, 7H), 3.12 (dt, 2H), 3.06(s, 1H), 2.80 (t,
J = 6.63 Hz, 3H), 2.31 (dd, J = 5.14, 13.03 Hz, 1H), 2.20 (dt, 1H),
0.96 (s, 3H), 0.91 (d, J = 6.86 Hz, 3H), 0.73 (s, 3H)0.71 (3H, d,
J = 6.28 Hz).
2.2.10. Synthesis of conjugate 20
Conjugate 20 was prepared from MTX and compound 16 using
the same procedure described for conjugate 18, followed by
removal of the Boc group under acidic conditions. In detail, the
conjugate 20 intermediate was suspended in dry chloroform, and
TFA (10 equivalents) was added slowly in an ice bath. After 4 hours,
TFA was removed under vacuum and the sample diluted with chlo-
roform, then washed with water and saturated brine. Alternatively,
the reaction mixture was poured into cold acetone and stirred for
30 min. Concentrated HCl solution was added to adjust the pH to 4,
and the solution was centrifuged to afford the crude compound
which was further purified by HPLC to yield the target product as
a yellow solid. Yield = 54%. Analytical HPLC, retention time =
23.82 min. Purity = 97.4%. 1H NMR (400 MHz, DMSO-d6, ppm):
d = 8.57 (s, 1H), 8.21 (s, 1H), 8.18 (d, J = 7.54 Hz, 1H), 7.75–7.73
(d, J = 8.99 Hz, 2H), 7.55 (s, 1H), 7.32 (d, 1H), 6.84–8.81 (d,
J = 8.99 Hz, 2H), 6.78 (s, 2H), 5.31 (d, J = 5.14 Hz, 1H), 4.80 (s, 2H),
4.29 (m, 3H), 0.96 (s, 3H), 0.90 (d, 3H), 0.73 (s, 6H). 13C NMR
(100 MHz, DMSO-d6): d = 174.57, 172.99, 166.60, 163.16, 159.03,
156.34, 154.84, 151.37, 149.56, 146.92, 140.19,, 129.58, 129.34,
122.17, 121.67, 119.12, 111.54, 108.87, 80.64, 73.86, 66.39, 56.17,
55.31, 49.86, 47.03, 41.56, 38.69, 37.25, 36.93, 36.67, 35.63,
32.26, 31.92, 31.40, 31.01, 30.26, 28.26, 28.25, 27.00, 20.31,
19.44, 17.52, 16.42, 15.09. HRMS (ESI): m/z 978. 6692 (Mꢂ1), cal-
culated (C52H73N11O8) 979.56.
2.2.7.2. Compound 15. 1H NMR (400 MHz, CDCl3): d = 5.38 (d,
J = 5.19 Hz, 1H), 4.49 (m, 1H), 4.44–4.38 (q, J = 7.41, 14.75 Hz,
2H), 3.49–3.35 (dd, J = 4.05, 10.67 Hz, 1H), 3.23 (t, J = 5.86 Hz,
2H), 2.86 (t, J = 5.86 Hz, 2H), 2.36 (dd, J = 5.14, 13.03 Hz, 1H), 2.27
(t, J = 11.67 Hz, 1H), 1.03 (s, 3H), 0.96(d, J = 6.86 Hz, 3H), 0.80 (s,
3H), 0.78 (d, J = 6.28 Hz, 3H).
2.2.7.3. Compound 16. 1H NMR (300 MHz, CDCl3): d = 5.36 (d,
J = 5.14 Hz, 1H), 4.37–4.44 (m, 2H), 3.61 (t, 2H), 3.49–3.33 (m,
10H), 3.20 (m, 1H), 2.30 (dd, J = 5.14, 13.03 Hz, 2H), 1.02 (s, 3H),
0.97 (d, 3H), 0.78 (s, 6H).
2.2.7.4. Compound 17. 1H NMR (300 MHz, CDCl3): d = 5.36 (d,
J = 5.14 Hz, 1H), 4.37 4.45 (m, 2H), 3.70 (t, 2H), 3.49 3.30 (m,
12H), 2.30 (dd, J = 5.14, 13.03 Hz, 2H), 1.02 (s, 3H), 0.97 (d, 3H),
0.78 (s, 6H).
2.2.8. Synthesis of conjugate 18
To a solution of MTX (100 mg, 0.22 mmol) and N-methylmor-
pholine (0.05 mL, 0.44 mmol) in 5 mL of anhydrous DMF, were
slowly added TBTU (70.6 mg, 0.44 mmol) in 5 mL of DMF over
45 min at ice cooled bath. Compound 14 (143.7 mg, 0.25 mmol)
was then added as a solid form. The resulting solution was stirred
for additional 45 min at 0 °C, then warmed to room temperature
for 5 days under nitrogen with protection from light. The reaction
solution was diluted with 30 mL of chloroform and washed with a
1 N HCl solution and distilled water, then dried over MgSO4. After
concentration, the crude product was further purified by flash
chromatography using 5–10% methanol in chloroform. The final
product was a yellow solid. Yield = 65%. Analytical HPLC, retention
2.2.11. Synthesis of conjugate 21
Conjugate 21 was prepared from MTX and compound 17
according the method described for conjugate 20. The final product
was a light yellow solid obtained with 51% overall yield. Analytical
HPLC, retention time: 18.45 min. Purity = 98.4%. 1H NMR
(300 MHz, DMSO-d6, ppm): d = 8.56 (s, 1H), 8.07 (d, J = 7.54 Hz,
2H), 7.92 (s, 1H), 7.81 7.74 (d, J = 8.99 Hz, 2H), 7.42 (s, 2H), 7.25
(s, 1H), 6.82–6.81 (d, J = 8.99 Hz, 2H), 6.61 (s, 2H), 5.32 (d,
J = 5.14 Hz, 1H), 4.87 (s, 2H), 4.28 (m, 2H), 4.21 (m, 1H), 3.20 (s,

48
B. Cai et al. / Steroids 116 (2016) 45–51
3H), 2.74 (m, 3H), 2.26–2.14 (m, 4H), 0.96 (s, 3H), 0.92 (d, 3H), 0.73
(s, 6H). 13C NMR (100 MHz, DMSO-d6): d = 174.12, 172.99, 166.77,
163.09, 159.02, 158.79, 156.55, 155.32, 151.55, 149.24, 145.94,
140.14, 129.61, 129.44, 122.53, 122.17, 121.69, 118.54, 116.58,
111.19,108.87, 80.61, 73.90, 66.35, 62.19, 56.12, 55.94, 49.83,
45.92, 43.02, 41.51, 37.10, 37.04, 36.63, 35.50, 32.32, 31.86,
31.35, 30.20, 28.89, 28.80, 20.78, 19.42, 17.50, 16.40, 15.07. HRMS
(ESI): m/z 1021.7123 (Mꢂ1), calculated (C54H78N12O8) 1022.61.
instructions of the DHFR assay kit (Product Code CS0340, Saint
Louis, Missouri 63103 USA). Briefly, 0.002 units of DHFR and 6
l
l
L
L
of NADPH solution were added into 1ꢀ assay buffer with 5–l0
of test compound in 10 nM PBS (the controls contained buffer solu-
tion only) to achieve final drug concentrations between 0.5 and
1000 nM for the inhibition assay, or without inhibitors for the
activity assay. The assay mixture was incubated for 5 min at room
temperature, and the reaction was started by adding 5 lL of dihy-
drofolic acid in a final volume of 1 mL. The change in the absor-
bance at 340 nm and 22 °C was measured using a kinetic model
with reading intervals of 15 s for 20 min. This is sufficient time
for DHFR activity to give a linear reaction slope. Results are given
as percent inhibition of enzyme activity versus the control. IC50
values for the DHFR inhibition and cell proliferation assays are
summarized in Table 1.
2.3. Cell growth inhibition assays
The human breast cancer cell lines, MCF-7 and MDA-MB-231,
were obtained from the Korean Cell Line Bank and cultured in RPMI
1640 media supplemented with 10% fetal bovine serum and 1%
penicillin–streptomycin in humidified air with 5% CO2 at 37 °C.
Cells (105/well) were seeded on 96-well plates. Once the cells
reached confluency, they were treated with the test compounds
2.5. Stability of the MTX-DG conjugates in the presence of glutathione
at concentrations ranging from 0.01 to 100
lM for the MTX conju-
gates, and 1 M to 5 mM for MTX. After 48 h incubation, 20
MTT solution (5 mg/mL in phosphate-buffered saline (PBS)) was
added and the cells were incubated for another 4 h. The culture
medium was then discarded and replaced with 200 lL of iso-
propanol at room temperature. The absorbance of each well was
determined by a microplate reader at a wavelength of 570 nm.
The IC50 of each sample was calculated by linear regression analy-
sis of the concentration-response curves obtained from the MTT
assay. IC50 values are reported in Table 1.
l
l
L of
To assess the basis for differences in the anti-proliferative activ-
ity of conjugates 18 and 21, we evaluated their stability in the pres-
ence of GSH. A solution of GSH in PBS at 7.4 was added into a
solution of conjugate 18 or 21 in methanol. The final concentra-
tions of the MTX-DG conjugates and GSH were 1 mg/mL and
10 mM, respectively. The reaction mixtures were incubated at
37 °C for 5 h. Then, 0.2 mL of the reaction mixture was removed
and analyzed by HPLC. MTX was used as a control and all tests
were repeated three times.
2.4. DHFR inhibition assay
3. Results and discussion
The inhibition of DHFR was evaluated by monitoring NADPH
oxidation at 340 nm in a microplate reader according to the
3.1. Chemistry
Scheme 1 shows the selective protection of primary amino
groups of polyamines (5, 6) using Nefkens’ reagent to afford the
corresponding terminally N-protected analogs, 7 or 8 [17]. This
was followed by treatment with di-tert-butyl dicarbonate (Boc)
anhydride to give compounds 9 and 10, and then de-protection
with hydrazine monohydrate to afford compounds 11 and 12
[18]. Scheme 2 shows the synthesis methods for MTX-DG conju-
gates. Briefly, a solution of compound 13, prepared by a reported
method [19] (see details in Supporting information S1). was added
into an excess of amine (3, 4, 11, or 12) solutions to give DG amine
conjugates 14–17. Compounds 14–17 were reacted with MTX due
to the presence of the coupling reagent, O-(benzotriazol-1-yl)-N,N,
N0,N0-tetramethyluroniumtetrafluoroborate (TBTU), to yield final
MTX-DG conjugates (18–21). TBTU was employed as the coupling
Table 1
Anti-proliferative and DHFR inhibitory IC50 values of MTX conjugates to MDA-MB-
231 cells.
Compound
Cytotoxicitya IC50
(lM)
DHFRb IC50 (nM)
DG
MTX
18
19
20
73.98 6.84
714.54 14.3
4.1 2.25
89.2 7.19
59.67 1.59
24.64 7.86
NDc
3.39 0.28
17.21 3.34
167.32 6.01
99.31 0.91
58.31 3.36
21
a
IC50 values for cytotoxicity were obtained by the MTT assay for each compound
after 48 h incubation. All samples were analyzed in triplicate and repeated three
independent times (average s.d.).
b
reagent because of the preferential modification of the
tuted isomer over the -substituted isomer [20].
c-substi-
DHFR assay was carried out using a kit.
Non-detectable (no activities toward DHFR were observed even above 1 lM).
c
a
Scheme 1. Synthesis of tert-butoxycarbonyl (Boc) protected polyamine linkers 11–12. Reagents and conditions: (a) phthalic anhydride, CH3COOH, 90 °C, 2 h; (b) N-
carboethoxyphthalimide, triethylamine, CH3CN; (c) di-tert-butyl dicarbonate, K2CO3 tetrahydrofuran; (d) hydrazine monohydrate, ethanol.

B. Cai et al. / Steroids 116 (2016) 45–51
49
Scheme 2. Synthesis of methotrexate (MTX)-diosgenin conjugates 18–21. Reagents and conditions: (a) triethanolamine, dichloromethane; (b) MTX, N-methylmorpholine
(NMM), O-(benzotriazol-1-yl)-N,N,N0,N0-tetramethyluroniumtetrafluoroborate (TBTU), dimethyl formamide (DMF); (c) 1) MTX, NMM, TBTU, DMF; 2) trifluoroacetic acid,
CHCl3.
3.2. DHFR inhibition assay
Among the tested compounds, conjugate 18, bearing a disulfide
bond, demonstrated the highest anti-proliferative activity with an
The synthesized MTX conjugates were tested for their inhibi-
tory activities against DHFR using an assay kit. All conjugates
retained inhibitory activity toward DHFR, although they were less
potent compared to MTX itself with an IC50 value of 3.39 nM. This
result was consistent with previous studies showing that most of
the resultant conjugates had a reduced affinity to the target
enzyme after chemical modifications at the glutamate residue of
MTX [21,22]. The inhibitory activity of the MTX conjugates
increased with an extension of the space length, except for conju-
gate 18, in the order of 19 < 20 < 21, with IC50 values of 167, 99, and
58 nM, respectively. This result indicated that the length of the
linkages was important for maintaining affinity to the enzyme.
Interestingly, conjugate 18 displayed the highest affinity against
DHFR with an IC50 value of 17.2 nM, even though it had a space
length similar to conjugate 21. The different pattern in the enzyme
inhibition assay may be attributed to the presence of a disulfide
bond in conjugate 18.
IC50 value of 4.1
anti-proliferative effects with space lengths increasing from 2 to
8 atoms; their IC50 values were 89.2, 59.7, and 24.6 M, respec-
lM. Conjugates 19, 20, and 21 showed increased
l
tively, showing the same trend as the results of their DHFR inhibi-
tory activities (Table 1). Conjugate 21 showed higher cellular
growth inhibition than compounds 19 and 20. This could be
explained by its increased DHFR inhibitory activity due to the
longer space length, thereby decreasing steric hindrance of the
bulky DG to the active site of DHFR. Alternatively, conjugate 21
may be more permeable to cell membranes owing to the presence
of two amine groups. The amine groups are positively charged in
physiological conditions and cationic molecules tend to penetrate
cell membranes [24]. The lower potency of conjugate 19 could be
explained by its poor affinity to the target enzyme. In addition, it
is possible that DG acts mainly as an inhibitor after entering tumor
cells because it is cytotoxic when its concentration exceeds 65 lM.
Moreover, a combination of both possibilities may lead to the
observed results.
Based on the space lengths of conjugates 20, and 21, compared
to conjugate 18, the biological activity of conjugate 18 was
expected to be between that of conjugates 20 and 21. However,
conjugate 18 had both the highest DHFR inhibition and anti-prolif-
eration activities. There are two potential explanations for this dif-
ference. First, the higher anti-proliferative activity of conjugate 18
is mainly due to its strong inhibition of DHFR because it is about
3.4-fold more potent to the target enzyme relative to conjugate
21. Second, cancer cells contain abundant glutathione (GSH) with
concentrations ranging from 2 to 20 mM, which are more than
100 times higher than those in the extracellular matrix and blood
plasma that range from 2 to 20 lM [25,26]. Thus, the redox-trig-
gered rapid release of MTX from conjugate 18 may have con-
tributed to this phenomenon.
Compared to the anti-proliferative activity of conjugate 21, con-
jugate 18 was approximately 6-fold more active in MDA-MB-231
cells. As noted above, this increased activity could be explained
by the redox-triggered rapid release of an active substance from
3.3. Cell growth inhibition assay
The synthesized compounds were tested for their anti-prolifer-
ative activities using a human breast cancer line, MDA-MB-231.
This cell line, isolated from an MTX-treated patient, is a well-char-
acterized model for the study of MTX transport resistance [21,22].
The mechanism of resistance involves a deficiency of the reduced
folate carrier (RFC), which is the main transporter for cellular
uptake of reduced folate and its analogs [23]. We confirmed that
MDA-MB-231 cells (IC50 = 715
to MTX than the sensitive cell line, MCF-7 (IC50 = 15.2
All conjugates had a strong anti-proliferative effect against
MDA-MB-231 cells in the concentration range of 3–90
(Table 1). Most were much more potent than MTX, although some
conjugates displayed weak anti-proliferative activities in this con-
centration range. These results suggest that the conjugates have a
promising capability of entering resistant cells after the diosgenin
derivatization.
l
M) were 47-fold more resistant
l
M).
lM

50
B. Cai et al. / Steroids 116 (2016) 45–51
Fig. 3. Stability study of conjugates 18 and 21 in the presence of GSH. HPLC analysis spectra for (a) MTX, (b) conjugate 18, (c) unstable conjugate 18 with GSH (d) conjugate 21
and (e) stable conjugate 21 with GSH.
the conjugate after entering the cells. The released MTX conjugate
with a short-chain cysteamine would further restore its affinity to
the target enzyme [27]. Synergistic effects from released MTX and
DG entities would be produced because DG hits multiple target
molecules in cancer cells [28].
and National Development Institute of Korean Medicine helped
us perform the 1H NMR and mass spectrometry analyses,
respectively.
Appendix A. Supplementary data
We investigated the redox-responsive release of MTX from con-
jugate 18 and conjugate 21 using 10 mM of GSH to mimic the
reductive environment of tumor cells. A rapid release of MTX from
conjugate 18 indeed occurred based on the appearance of a new
HPLC peak with a retention time of 4.76 min (Fig. 3c). The released
MTX showed a slightly different retention time compared with
standard MTX (4.32 min) because the released MTX is covalently
attached to cysteamine through an amide bond (see S1 in Support-
ing information). The new peak at 4.76 min was further analyzed
by UFLC-MS. The mass data (515.3023, M+2) was consistent with
an MTX cysteamine conjugate with a calculated mass of 513.19
(see Fig. S26 in Supporting information). In contrast, conjugate
21 did not show any release when treated with same condition
as conjugate 18, suggesting that conjugate 21 could be relatively
stable under the reductive environment due to the absence of
redox-responsive disulfide bond (Fig. 3e).
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.steroids.2016.10.
006.
References
[1] S.S. Abolmaali, A.M. Tamaddon, R. Dinarvand,
A review of therapeutic
challenges and achievements of methotrexate delivery systems for treatment
of cancer and rheumatoid arthritis, Cancer Chemother. Pharmacol. 71 (5)
(2013) 1115–1130.
[2] Z.A. Khan, R. Tripathi, B. Mishra, Methotrexate: a detailed review on drug
delivery and clinical aspects, Expert Opin. Drug Deliv. 9 (2) (2012) 151–169.
[3] A.K. Fotoohi, F. Albertioni, Mechanisms of antifolate resistance and
methotrexate efficacy in leukemia cells, Leuk. Lymphoma 49 (3) (2008) 26–
410.
[4] J.R. Bertino, E. Goker, R. Gorlick, W.W. Li, D. Banerjee, Resistance mechanisms
to methotrexate in tumors, Oncologist 1 (4) (1996) 223–226.
[5] M.A. Smal, Z. Dong, H.T. Cheung, Y. Asano, L. Escoffier, M. Costello, M.H.
Tattersall, Activation and cytotoxicity of 2-alpha-aminoacyl prodrugs of
methotrexate, Biochem. Pharmacol. 49 (4) (1995) 567–574.
[6] G. Jansen, R. Mauritz, S. Drori, H. Sprecher, I. Kathmann, M. Bunni, D.G. Priest, P.
Noordhuis, J.H. Schornagel, H.M. Pinedo, G.J. Peters, Y.G. Assaraf, A structurally
altered human reduced folate carrier with increased folic acid transport
mediates a novel mechanism of antifolate resistance, J. Biol. Chem. 273 (46)
(1998) 30189–30198.
[7] W.C. Wang, S.F. Liu, W.T. Chang, Y.L. Shiue, P.F. Hsieh, T.J. Hung, C.Y. Hung, Y.J.
Hung, M.F. Chen, Y.L. Yang, The effects of diosgenin in the regulation of renal
proximal tubular fibrosis, Exp. Cell Res. 323 (2) (2014) 255–262.
[8] Z. He, Y. Tian, X. Zhang, B. Bing, L. Zhang, H. Wang, W. Zhao, Anti-tumour and
immunomodulating activities of diosgenin, a naturally occurring steroidal
saponin, Nat. Prod. Res. 26 (23) (2012) 2243–2246.
[9] L. Ghezali, D.Y. Leger, Y. Limami, J. Cook-Moreau, J.L. Beneytout, B. Liagre,
Cyclopamine and jervine induce COX-2 overexpression in human
erythroleukemia cells but only cyclopamine has a pro-apoptotic effect, Exp.
Cell Res. 319 (7) (2013) 1043–1053.
[10] C. Lepage, B. Liagre, J. Cook-Moreau, A. Pinon, J.L. Beneytout, Cyclooxygenase-2
and 5-lipoxygenase pathways in diosgenin-induced apoptosis in HT-29 and
HCT-116 colon cancer cells, Int. J. Oncol. 36 (5) (2010) 1183–1191.
[11] S. Srinivasan, S. Koduru, R. Kumar, G. Venguswamy, N. Kyprianou, C.
Damodaran, Diosgenin targets Akt-mediated prosurvival signaling in human
breast cancer cells, Int. J. Cancer 125 (4) (2009) 961–967.
4. Conclusions
In summary, a new series of MTX-DG conjugates with polya-
mine spacers was designed and synthesized. Biological evaluations
of those conjugates for their abilities to inhibit DHFR and prolifer-
ation of the MTX transport-resistant cell line, MDA-MB-231, were
performed. All MTX conjugates showed stronger anti-proliferative
activities than MTX alone against MDA-MB-231 cells. Extension of
the space lengths affected the biological activity. Conjugate 18,
containing a disulfide bond, was the most active among the pre-
pared conjugates. These results suggest that DG conjugates of
MTX have the capability of entering MTX-resistant cells. The bio-
logical data suggest that MTX-DG conjugates may be promising
anticancer agents and indicate the need for further systematic
optimizations and studies on their applications to various cancers
in vitro and in vivo.
[12] M.N. Cayen, E.S. Ferdinandi, E. Greselin, D. Dvornik, Studies on the disposition
of diosgenin in rats, dogs, monkeys and man, Atherosclerosis 33 (1) (1979) 71–
87.
Acknowledgements
[13] D.H. Kim, B.N. Hong, H.T. Le, H.N. Hong, C.W. Lim, K.H. Park, T.W. Kim, T.H.
Kang, Small molecular weight PEGylation of diosgenin in an in vivo animal
study for diabetic auditory impairment treatment, Bioorg. Med. Chem. Lett. 22
(14) (2012) 4609–4612.
[14] G. Xin, Y. Wang, X. Guo, B. Huang, D. Du, S. He, R. Zhang, Z. Xing, H. Zhao, Q.
Chen, W. Huang, Y. He, Synthesis of diosgenin-ibuprofen derivatives and their
activities against insulin-dependent diabetes mellitus, Chem. Pharm. Bull. 61
(5) (2013) 532–538.
This study was financially supported by the Ministry of Trade,
Industry & Energy, Korea Institute for Advancement of Technology,
and the Honam institute for Regional Program Evaluation through
the Leading Industry Development for Economic Region (grant#:
R0001536). This study was also financially supported by Chonnam
National University, 2011–2012. The Korea Basic Science Institute

B. Cai et al. / Steroids 116 (2016) 45–51
51
[15] B. Huang, D. Du, R. Zhang, X. Wu, Z. Xing, Y. He, W. Huang, Synthesis,
characterization and biological studies of diosgenyl analogues, Bioorg. Med.
Chem. Lett. 22 (24) (2012) 7330–7334.
[16] J.E. Chojnacki, K. Liu, J.M. Saathoff, S. Zhang, Bivalent ligands incorporating
curcumin and diosgenin as multifunctional compounds against Alzheimer’s
disease, Bioorg. Med. Chem. 23 (22) (2015) 7324–7331.
[23] J. Worm, A.F. Kirkin, K.N. Dzhandzhugazyan, P. Guldberg, Methylation-
dependent silencing of the reduced folate carrier gene in inherently
methotrexate-resistant human breast cancer cells, J. Biol. Chem. 276 (43)
(2001) 39990–40000.
[24] N.V. Bhupathiraju, X. Hu, Z. Zhou, F.R. Fronczek, P.O. Couraud, I.A. Romero, B.
Weksler, M.G. Vicente, Synthesis and in vitro evaluation of BBB permeability,
[17] N.K. Sung, D.W. Cho, J.H. Choi, K.W. Choi, U.C. Yoon, H. Maeda, P.S. Mariano, A
facile approach to the preparation of bis-crown ethers based on SET-promoted
photomacrocyclization reactions, J. Org. Chem. 72 (23) (2007) 8831–8837.
[18] S.O. Kang, V.W. Day, K. Bowman-James, Tricyclic host for linear anions, Inorg.
Chem. 49 (18) (2010) 8629–8636.
[19] J. Xu, C.D. Duffy, C.K. Chan, J.D. Sutherland, Solid-phase synthesis and
hybrization behavior of partially 20/30-O-acetylated RNA oligonucleotides, J.
Org. Chem. 79 (8) (2014) 3311–3326.
[20] M.A. Santos, E.A. Enyedy, E. Nuti, A. Rossello, N.I. Krupenko, S.A. Krupenko,
Methotrexate gamma-hydroxamate derivatives as potential dual target
antitumor drugs, Bioorg. Med. Chem. 15 (3) (2007) 1266–1274.
[21] M. Lindgren, K. Rosenthal-Aizman, K. Saar, E. Eiriksdottir, Y. Jiang, M. Sassian,
P. Ostlund, M. Hallbrink, U. Langel, Overcoming methotrexate resistance in
breast cancer tumour cells by the use of a new cell-penetrating peptide,
Biochem. Pharmacol. 71 (4) (2006) 416–425.
tumor cell uptake, and cytotoxicity of a series of carboranylporphyrin
conjugates, J. Med. Chem. 57 (15) (2014) 6718–6728.
[25] S. Borrelli, M.S. Christodoulou, I. Ficarra, A. Silvani, G. Cappelletti, D. Cartelli, G.
Damia, F. Ricci, M. Zucchetti, F. Dosio, D. Passarella, New class of squalene-
based releasable nanoassemblies of paclitaxel, podophyllotoxin, camptothecin
and epothilone A, Eur. J. Med. Chem. 85 (2014) 179–190.
[26] H. Han, Q. Jin, Y. Wang, Y. Chen, J. Ji, The rational design of a gemcitabine
prodrug with AIE-based intracellular light-up characteristics for selective
suppression of pancreatic cancer cells, Chem. Commun. 51 (98) (2015) 17435–
17438.
[27] R. Pignatello, S. Guccione, S. Forte, C. Di Giacomo, V. Sorrenti, L. Vicari, G.
Uccello Barretta, F. Balzano, G. Puglisi, Lipophilic conjugates of methotrexate
with short-chain alkylamino acids as DHFR inhibitors. Synthesis, biological
evaluation, and molecular modeling, Bioorg. Med. Chem. 12 (11) (2004) 2951–
2964.
[22] Z. Wu, A. Shah, N. Patel, X. Yuan, Development of methotrexate proline
prodrug to overcome resistance by MDA-MB-231 cells, Bioorg. Med. Chem.
Lett. 20 (17) (2010) 5108–5112.
[28] B. Sung, S. Prasad, V.R. Yadav, B.B. Aggarwal, Cancer cell signaling pathways
targeted by spice-derived nutraceuticals, Nutr. Cancer 64 (2) (2012) 173–197.