Synthesis, Antiproliferative Activity, and Effect on Carcinoma A549 Cell Microtubules of New Tubuloclustin Analogs

Article abstract of DOI:10.1007/s11094-019-02014-y

Combretastatin analogs of the antitumor agent tubuloclustin {N-[7-(adamant-2-yloxy)-7-oxoheptanoyl]-Ndeacetylcolchicine} were prepared via esterification of combretastatin by monoesters of pimelic or adipic acid with adamantan-2-ol or (adamantan-1-yl)methanol. These conjugates were stable and cytotoxic to human lung carcinoma A549 cells (EC₅₀ ≈ 50 – 70 nM) and caused depolymerization of microtubules and slight clustering of tubulin. Tubuloclustin analogs with shortened linkers were prepared via amidation by N-deacetylcolchicine of monoesters of adipic or succinic acids with adamantan-1-ol or (adamantan-1-yl)methanol. The conjugate N-[6-(adamantyl)-6-oxohexanoyl]-N-deacetylcolchicine was more active (EC₅₀ ≈ 4 nM) than tubuloclustin and promoted strong tubulin clusterization. All compounds induced apoptosis of A549 cells. Tests in vivo of N-[6-(adamantyl)-6-oxoheaxnoyl]-N-deacetylcolchicine on carcinoma A549 experimental models were concluded to be promising.

Full text of DOI:10.1007/s11094-019-02014-y

DOI 10.1007/s11094-019-02014-y
Pharmaceutical Chemistry Journal, Vol. 53, No. 5, August, 2019 (Russian Original Vol. 53, No. 5, May, 2019)
SYNTHESIS, ANTIPROLIFERATIVE ACTIVITY,
AND EFFECT ON CARCINOMA A549 CELL
MICROTUBULES OF NEW TUBULOCLUSTIN ANALOGS
N. A. Zefirov,1, 2 Yu. A. Evteeva,1 A. R. Fatkulin,1 S. Schulz,3
S. A. Kuznetsov,3 and O. N. Zefirova1, 2,*
Translated from Khimiko-Farmatsevticheskii Zhurnal, Vol. 53, No. 5, pp. 13 – 19, May, 2019.
Original article submitted May 28, 2018.
Combretastatin analogs of the antitumor agent tubuloclustin {N-[7-(adamant-2-yloxy)-7-oxoheptanoyl]-N-
deacetylcolchicine} were prepared via esterification of combretastatin by monoesters of pimelic or adipic acid
with adamantan-2-ol or (adamantan-1-yl)methanol. These conjugates were stable and cytotoxic to human
lung carcinoma A549 cells (EC50 » 50 – 70 nM) and caused depolymerization of microtubules and slight clus-
tering of tubulin. Tubuloclustin analogs with shortened linkers were prepared via amidation by N-deacetylcol-
chicine of monoesters of adipic or succinic acids with adamantan-1-ol or (adamantan-1-yl)methanol. The con-
jugate N-[6-(adamantyl)-6-oxohexanoyl]-N-deacetylcolchicine was more active (EC50 » 4 nM) than tubulo-
clustin and promoted strong tubulin clusterization. All compounds induced apoptosis of A549 cells. Tests in
vivo of N-[6-(adamantyl)-6-oxoheaxnoyl]-N-deacetylcolchicine on carcinoma A549 experimental models
were concluded to be promising.
Keywords: carcinoma A549, tubuloclustin, colchicine, combretastatin, adamantane.
The antitumor activity of ligands for the colchicine do-
main of the dimeric protein a,b-tubulin is due to their ability
to inhibit its polymerization to microtubules, which play an
important role in the cell life cycle [1, 2]. Although many
such compounds, both natural and synthetic, have now been
prepared and studied, only one example is used in clinical
practice because of high general toxicity [3] or low efficacy
in vitro and/or in vivo [1, 4]. The pairing of molecular frag-
ments has been used extensively in the last decade to design
more active and less toxic ligands of the tubulin colchicine
domain [5, 6], including combining colchicine (reacting
mainly with the protein b-subunit) with groups capable of
binding to the tubulin a-subunit (e.g., Ia-e in Fig. 1) [7-10].
This approach was used by us to prepare tubuloclustin
(Id) and its bridgehead analog Ie, which caused tubulin
clusterization in vitro that was uncharacteristic of colchicine
[10] and exhibited high antiproliferative activity against tu-
mor cells in vitro [11] and in vivo [12].
Further attempts to increase the activity and decrease the
general toxicity of tubuloclustin had limited success
[12 – 16]. Therefore, the goals of the present work were to
prepare two pairs of new analogs, to determine their
antiproliferative activity and effect on tumor cell microtubule
networks, and to explain the results using computer molecu-
lar modeling.
Compounds IIa-b and IIIa-b were proposed for synthe-
sis based on previously obtained structure—activity data for
tubuloclustin analogs [11, 13, 15] (Fig. 2). Although all stud-
ied versions of replacing colchicine in Id and Ie by frag-
ments of other ligands of the tubulin colchicine domain with
lower general toxicity than colchicine (2-methoxyestradiol,
combretastatin A-4, etc.) produced compounds with low ac-
tivity [12, 14 – 16], we made yet another attempt in this di-
rection. Pair IIa,b were analogs of starting Id and Ie in
which a combretastatin bonded to a phenol hydroxyl through
an ester bond (this type of linker had not been studied by us
earlier) replaced the colchicine fragment. The overall linker
chain length in IIb did not change compared with lead com-
1
M. V. Lomonosov Moscow State University, Moscow, 119991 Russia.
2
Institute of Physiologically Active Compounds, Russian Academy of Sci-
ences, Chernogolovka, Moscow Oblast, 142432 Russia.
3
Institute of Biological Sciences, University of Rostock, Rostock, Ger-
many.
*
e-mail: olgaz@org.chem.msu.ru
423
0091-150X/19/5305-0423 © 2019 Springer Science+Business Media, LLC

424
N. A. Zefirov et al.
HO(CH2)m
(CH2)n
HO
O
CH3O
CH3O
OH
(CH2)n
OH
O
(CH2)m
H
m =0,1
OCH3 OCH3
O
O
q
IIa,b
O
O
i
ii (from Va or Vb)
V
IV
NH2
Va: n= 5, m =0, 2-adamantyl
Vb:n = 4, m =1, 1-adamantyl
IVa: n = 5
IVb: n = 4
IVc: n = 2
CH3O
CH3O
O
OCH3
OCH3
IIIa,b
Vc: n= 4, m = 0, 1-adamantyl
Vd:n = 2, m = 1, 1-adamantyl
iii (from Vc or Vd)
Scheme 1. Reagents and conditions: i) 4-DMAP, CH Cl , RT, 24 h [Va from IVa and adamantan-2-ol, 71%; Vb from IVb and
2
2
(adamantan-1-yl)methanol, 60%; Vc from IVb and adamantan-1-ol, 49%; Vd from IVc and (adamantan-1-yl)methanol, 50%]; ii)
combretastatin A-4, Va or Vb, DCC, $-DMAP, DMF, RT (IIa from Va, 67%; IIb from Vb, 59%); iii) N-deacetylcolchicine, Vc or Vd, EEDQ,
CH Cl , RT, 12 h (IIIa from Vc, 56%; IIIb from Vd, 25%).
2
2
pound Ie. However, this ester bond was shifted from the ada-
mantane fragment to provide greater conformational flexibil-
ity to the terminal conjugate fragment.
Vc and 172.18 ppm for Vd). The PMR spectrum of Vd ex-
hibited also a characteristic shift of the Ad–CH –O– protons
2
(3.70 ppm) to weak field as compared with those in the start-
ing framework alcohol (3.17 ppm).
The structures of conjugates IIIa,b were proposed based
on published results [13] that showed the antiproliferative
activity could be increased somewhat by using the modified
oxoheptanoyl chain of Ie and that analogs with linkers con-
taining four and less methylenes should be further studied.
Target conjugates IIa,b and IIIa,b were synthesized in
steps. First, the reaction of polyanhydrides of pimelic or
adipic acids or succinic anhydride (IVa-c, prepared by react-
ing the corresponding dibasic acids with acetic anhydride)
with adamantan-2-ol, (adamantan-1-yl)methanol, or adaman-
tan-1-ol in the presence of a catalytic amount of 4-dimethyl-
aminopyridine (4-DMAP) produced monoesters Va-d in
49-71% yields (Scheme 1). 13C NMR spectra of previously
unreported Vc and Vd (see Experimental section) showed a
characteristic resonance for ester C atoms (173.55 ppm for
Target pair IIa,b was synthesized by esterification of
monoesters of carboxylic acids Va,b with combretastatin A-4
in the presence of N,N-dicyclohexylcarbodiimide (DCC) and
4-DMAP. Formation of the esters was confirmed by two res-
onances for ester C atoms in 13C NMR spectra of conjugates
IIa,b at 171.42 and 172.97 ppm for IIa and 172.24 and
173.63 ppm for IIIa. MALDI-TOF mass spectra exhibited
peaks for a molecular ion with m/z 592 [M+]. It is noteworthy
that compounds IIa,b were demonstrated to be stable in
CDCl solution for 3 d according to NMR spectroscopy al-
3
though combretastatin A-4 esters are rather stable [17].
N-Deacetylcolchicine was synthesized in three steps as
before [18, 19] to produce the target pair of colchicine conju-
gates IIIa,b. Amidation of monoesters Vc and Vd by
TABLE 1. Biological Test Results for Human Lung Carcinoma A549 Cell Line
Compound
Cytotoxicity* EC50, nM
70 ± 6
Effect on microtubule network**
IIa
10 mM: 24 h, microtubule depolymerization; 72 h, point clusterization in several cells
10 mM: 24 h, microtubule depolymerization, point clusterization in several cells
IIb
IIIa
53 ± 5
100 nM: 24 h, complete microtubule depolymerization; 800 nM: 24 – 72 h, strong
clusterization
4.2 ± 0.5
IIIb
13 ± 2
6 [11]
1 mM: 24 h, complete microtubule depolymerization; 72 h, moderate clusterization
1 mM: 24 h, depolymerization and strong clusterization
Id (tubuloclustin)
II
[R= –CH2–(CH2)5C(O)O-2Ad]
1040 [15]
10 mM: 24 h, microtubule depolymerization; 72 h, point clusterization in several cells
[15]
Ie
11 [11]
10 ± 2
31 ± 5
1 mM: 24 h, depolymerization and strong clusterization
CA-4
1 mM: 24 – 72 h, microtubule depolymerization and no clusterization
1 mM: 24 – 72 h, microtubule depolymerization and no clusterization
Colchicine
*
Average of three independent experiments;
**
effect on microtubular network at the given concentrations and incubation periods.

Synthesis, Antiproliferative Activity, and Effect on Carcinoma A549
425
O
O
CH3O
CH3O
Ia: R=
NH
Cl
O
R
O
O
O
F
N
N
O
O
Ic:R=
ONO2
N
CH3O
Ib: R=
NH
O
Colchicine
fragment
OCH
3
(CH2)5
NH
Id: R=
O
(CH2)5
NH
Ie:R=
O
O
O
O
O
I
Fig. 1. Reported conjugates of colchicine and groups interacting with tubulin a-subunit; Id is tubuloclustin.
N-deacetylcolchicine in the presence of N-ethoxycarbonyl-
2-ethoxy-1,2-dihydroquinoline (EEDQ) afforded conjugates
IIIa,b in 56 and 25% yields. PMR spectra of IIIa,b showed
(10 mM): 24 h, 18%; 48, 42; 72, 48; IIb: 24 h, 42%; 48, 47;
72, 60].
Immunofluorescent microscopy gave interesting results
(Fig. 3, a-d). Both IIa and IIb depolymerized completely
A549 cell microtubules and caused small point clusters of
tubulin to form in several of them (Fig. 3c ). This was charac-
teristic of several tubuloclustin analogs and for less cytotoxic
conjugate II [R = –CH –(CH ) C(O)O–2Ad] with a similar
colchicine C resonances shifted to weaker field (4.65 ppm)
7
relative to the resonance of the same proton in the spectrum
of N-deacetylcolchicine (~3.71 ppm). This confirmed that
the amide bond formed. The carboxyl C atom resonance of
starting acids Va,b (178.42 – 179.52 ppm) disappeared in
13C NMR spectra of IIIa,b. Mass spectrometry and elemen-
tal analysis confirmed the structures of all previously unre-
ported compounds.
2
2 5
structure that was prepared by us earlier [16] (Table 1). First,
these data indicated that hydrolysis of the ester bond at the
phenol hydroxyl of IIa, b by nonspecific cellular esterases, if
it occurred at all, was insignificant. This meant that a large
number of unhydrolyzed molecules remained in the cells
(because the hydrolysis products did not possess clustering
effects). Also, the results confirmed the hypothesis proposed
by us previously that tubuloclustin analogs with various lig-
ands of the tubulin colchicine domain could exhibit cluster-
ing effects [15].
Antiproliferative activity (cytotoxicity) of target combre-
tastatin conjugates IIa,b and colchicine conjugates IIIa,b
against human lung carcinoma A549 cells was determined
using the standard colorimetric assay with 3-(4,5-dime-
thylthiazol-2-yl)-2,5-diphenyltetrazole hydrobromide (MTT
[20]) and direct counting of A549 cells under a microscope
24 and 48 h after treatment with the tested compounds. The
ability of the synthesized compounds to affect A549 cell
microtubular networks and stimulate apoptosis was studied
using immunofluorescent spectroscopy (Table 1, Fig. 3, a-f).
The MTT assay showed that combretastatin conjugates
IIa,b possessed high cytotoxicity against carcinoma A549
Computer docking of IIa, b and II [R = –CH –
2
(CH ) C(O)O–2Ad] into three-dimensional models of com-
2 5
plexes of the tubulin dimer with various ligands of the
colchicine domain (PDB ID: ISAO, 4O2B) using CLC Drug
Discovery Workbench (Version 1.5) revealed a version for
positioning three of these molecules in the 4O2B model
(from several with minima in estimating functions) that ex-
plained the obtained test results (Fig. 4).
cells with EC values of 50-70 nM, i.e., approximately
50
equal to EC of colchicine and slightly greater than EC of
50
50
combretastatin A-4 (Table 1). Direct cell counting showed
that cell proliferation was completely inhibited 24 h after
treatment with IIa,b (10 mM). Both conjugates also induced
apoptosis of tumor cells [% cells with fragmented nuclei
characteristic of apoptosis (Fig. 3e ) after treatment with IIa
Figure 4 shows that the linkers to adamantane in IIa, b
had similar positions that were different from those of II
[R = –CH –(CH ) C(O)O–2Ad]. However, the adamantane
2
2 5
framework in all three molecules occupied lipophilic regions
Colchicine fragment
Combretastatin fragment
(CH2)5
OCH3
(CH2)4
R=
O
R=
O
O
CH3O
CH3O
O
O
O
O
OR
NHR
OCH3
IIa
IIIa
CH3O
CH3O
CH3O
(CH2)4
R=
O
R=
O
(CH2)2
O
O
O
O
OCH
3
IIb
IIIb
II
III
Fig. 2. Structures of new analogs Ia and Ib with the colchicine fragment replaced by combretastatin and a modified linker.

426
N. A. Zefirov et al.
a
b
c
d
e
f
Fig. 3. Tubulin accumulation in carcinoma A549 cells treated with tested or control compounds: whole microtubules (negative control, 0.5%
DMSO) (a); diffuse distribution of tubulin after depolymerization of microtubules (CA-4, positive control) (b); weak (point) tubulin
clusterization (c); strong tubulin clusterization analogous to the effect of tubuloclustin (d); moderate tubulin clusterization (e); normal
unfragmented nuclei (top), nuclear fragmentation characteristic of apoptotic cells (bottom). Scale, 10 mm.
In general, the results demonstrated that tubuloclustin
analogs with the colchicine fragment replaced by a less toxic
tubulin colchicine site ligand could remain highly cytotoxic
against tumor cells and stimulated further research in this
area. The cytotoxicity of IIIa against human lung carcinoma
A549 cells was comparable to that of one of the most effec-
tive tubulin-directed clinically used antitumor agents, taxol.
Therefore, IIIa was interesting for further in vivo testing.
of the protein a-subunit that were close to amino-acid resi-
due aTyr224. This was characteristic of compounds capable
of causing the clustering effect characteristic of tubuloclustin
and its analogs [10, 15]. The locations in the protein of the
combretastatin fragments of the examined molecules were
close to that for free combretastatin for conjugates IIa, b and
distinctly different from it for II [R = –CH –(CH ) C(O)O–
2Ad]. Apparently, limited free rotation around the phenol es-
ter bond in IIa, b was responsible for the difference because
the carbonyl O atom of this group was not involved in
H-bonding with protein amino-acid residues according to the
modeling data.
2
2 5
EXPERIMENTAL CHEMICAL PART
The course of reactions and purity of compounds were
monitored by TLC on Silufol-UV254 plates. Column chro-
matography used Acros silica gel (40 – 60 mm). PMR and
13C NMR spectra were recorded at 28°C on a Bruker Avance
400 spectrometer (400 and 100 MHz, respectively). Chemi-
cal shifts were given on the d scale (ppm) vs. residual solvent
resonances. Spin—spin coupling constants were given in Hz.
Elemental analyses were performed on a Vario micro cube
CHN analyzer. MALDI-TOF mass spectra were recorded on
a Bruker Autoflex II instrument. Colchicine (Sigma-Aldrich,
95% by HPLC) and tubuloclustin synthesized by the litera-
ture method [11] (purity >95% by PMR spectroscopy and el-
emental analysis) were used in the work.
According to MTT assays of colchicine conjugates with
shortened linker chains (IIIa, b), both were highly cytotoxic
against tumor cells with an EC value for IIIa that was an
50
order of magnitude greater than that of colchicine. Moreover,
this compound was also more cytotoxic than tubuloclustin Id
and its bridgehead analog Ie (Table 1). It caused complete
tubulin depolymerization already at a concentration of
100 nM (Table 1) and gave a strong clustering effect (forma-
tion of point clusters began at much lower concentrations) at
800 nM (Fig. 3d ). Both compounds at a concentration of
1 mM inhibited completely cell growth and induced apop-
tosis (determined from fragmented nuclei, Fig. 3e ).

Synthesis, Antiproliferative Activity, and Effect on Carcinoma A549
427
7-(2-Adamantyloxy)-7-oxoheptanoic acid (Va) was
prepared in 71% yield by the literature method [11].
6-(Adamant-1-ylmethoxy)-6-oxohexanoic acid (Vb)
was prepared in 60% yield by the literature method [13].
6-(Adamant-1-yl)-6-oxohexanoic acid (Vc). Adaman-
tan-1-ol (0.217 g, 1.31 mmol) in CH Cl (5 mL) was treated
2
2
with adipic acid polyanhydride (0.500 g, 3.9 mmol) and a ca-
talytic amount of DMAP (0.01 g). The reaction mixture was
stirred at room temperature for 48 h. The solvent was evapo-
rated in a rotary evaporator. The solid was chromatographed
[eluent EtOAc–petroleum ether (40 – 70°C), 1:51:3 gradi-
ent] to afford Vc (0.180 g, 49%) as a colorless oil. PMR
spectrum (CDCl , d, ppm, J/Hz): 1.51 – 1.52 (6H, m),
3
1.63 – 1.74 (10H, m), 1.98 (3H, m), 2.34 – 2.40 (4H, m),
10.70 (1H, br.s, COOH). 13C NMR spectrum (CDCl , d,
3
ppm): 24.06, 24.32, 27.96, 33.08, 33.61, 33.88, 36.89, 39.21,
173.55 [C(O)O], 179.52 (COOH). C H O .
Fig. 4. Positioning of structures of IIa (light-gray), IIb (black), and
2
II [R = –CH –(CH ) C(O)O– Ad] (gray) in a, b-tubulin dimer
16 24
4
2 2 5
(b-subunit shown on the right; a-subunit, on the left; H-bonds are
4-(Adamant-1-ylmethoxy)-4-oxobutanoic acid (Vd).
Adamantan-1-ylmethanol (0.341 g, 2.05 mmol) in THF
(5 mL) was treated with succinic anhydride (0.612 g,
6.12 mmol) and a catalytic amount of DMAP (0.01 g). The
reaction mixture was refluxed for 24 h. The solvent was
evaporated. The residue was chromatographed [eluent
EtOAc–petroleum ether (40 – 70°C) 1:71:4 gradient] to af-
ford Vd (0.270 g, 50%) as a colorless oil. 13C NMR spectrum
shown as dotted lines; most H atoms were omitted for clarity).
Adamant-1-ylmethoxy-6-{2-methoxy-5-[(Z)-3,4,5-tri-
methoxystyryl]phenyl}hexanedioate (IIb) was synthesized
analogously to IIa from Vb (0.021 g, 0.071 mmol),
combretastatin A-4 (0.020 g, 0.063 mmol), and DCC
(0.028 g, 0.135 mmol) to afford IIb (0.022 g, 59%) as a yel-
(CDCl , d, ppm): 27.96, 28.92, 29.03, 33.13, 36.87, 39.13,
3
74.39 (AdCH ), 172.18 (COO), 178.42 (COOH). C H O .
2
15 22
4
low oil. PMR spectrum (CDCl , d, ppm, J/Hz): 1.53 (6H, m);
1-Adamantan-2-yl-7-{2-methoxy-5-[(Z)-3,4,5-trimetho-
xystyryl]phenyl}heptanedioate (IIa). Compound Va
(0.020 g, 0.068 mmol) in CH Cl (10 mL) was treated with
3
1.60 – 1.78 (9H, m); 1.98 (3H, m); 2.34 – 2.40 (2H, m,
J = 6.9 Hz); 2.57 (2H, t, J = 6.9 Hz); 3.68 (2H, s, AdCH );
2
2
2
3.71 (6H, s, OCH ); 3.80 (3H, s, OCH ); 3.84 (3H, s, OCH );
combretastatin A-4 (0.20 g, 0.063 mmol), DCC (0.028 g,
0.135 mmol), and a catalytic amount of DMAP (0.01 g). The
mixture was stirred at room temperature for 12 h, treated
with HOAc (5 – 10 mL), and evaporated after 15 min. The
precipitate was dissolved in EtOAc (10 – 20 mL) and stored
at 0 – 4°C for 2 – 3 h. The resulting precipitate of N,N¢-di-
cyclohexylurea was filtered off and rinsed with cold EtOAc
(2 ´ 10 mL). The filtrate was washed with saturated NaCl
solution (10 mL) and H O (10 mL), dried over Na SO , and
3
3
3
6.45 (2H, d, CH=CH, J = 1.3 Hz); 6.50 (2H, s, Ar); 6.85 (1H,
d, Ar, J = 8.5 Hz); 7.00 (2H, d, Ar, J = 2.1 Hz); 7.12 (1H, dd,
Ar, J = 2.1, 8.5 Hz). 13C NMR spectrum (CDCl , d, ppm):
3
24.30, 24.42, 27.99, 33.11, 33.52, 33.89, 36.91, 39.23, 55.83,
55.87, 60.84, 73.95, 105.77, 111.88, 123.07, 127.60, 128.54,
129.44, 130.01, 132.41, 137.08, 139.39, 150.16, 152.91,
171.17 (C=O), 173.49 (C=O). Mass spectrum (MALDI-
TOF) m/z: 592 [M+K]+.
2
2
4
N-[6-(Adamantyl)-6-oxohexanoyl]-N-deacetylcolchici-
ne (IIIa). Compound Vc (0.050 g, 0.15 mmol) in CH Cl
evaporated. The residue was chromatographed [eluent
EtOAc–petroleum ether (40 – 70°C), 1:8 – 1:5] to afford IIa
2
2
(10 mL) was stirred, treated with EEDQ (0.037 g,
0.15 mmol) and N-deacetylcolchicine (0.48 g, 0.13 mmol),
stirred at room temperature for 24 h, and evaporated. The
residue was chromatographed [eluent EtOAc—petroleum
ether (40 – 70°C), 1:5; then CH Cl —MeOH, 30:1] to afford
(0.025 g, 67%) as a colorless oil. PMR spectrum (CDCl , d,
3
ppm, J/Hz): 1.46 (2H, m); 1.58 (3H, m); 1.67 – 1.78 (3H, m);
1.84 (6H, m); 1.99 (3H, m); 2.36 (2H, m, J = 6.9 Hz); 2.43
(1H, m, J = 6.9 Hz); 2.55 (2H, t, J = 7.5 Hz); 3.68 (2H, s,
AdCH ); 3.71 (6H, s, OCH ); 3.80 (3H, s, OCH ); 3.84 (3H,
2
2
2
3
3
IIIa (0.045 g, 56%) as a yellow oil. PMR spectrum (CDCl ,
s, OCH ); 4.93 (1H, m); 6.46 (2H, d, CH=CH, J = 1.5 Hz);
3
3
6.51 (2H, s, Ar); 6.85 (1H, d, Ar, J = 8.6 Hz); 7.00 (2H, d,
Ar, J = 2.1 Hz); 7.12 (1H, dd, Ar, J = 2.1, 8.4 Hz). 13C NMR
d, ppm, J/Hz): 1.51 – 1.69 (m, 8H), 1.84 – 1.90 (m, 2H), 2.07
(m, 6H), 2.14 (m, 3H, H6colch), 2.18 – 2.29 (m, 5H),
2.37 – 2.42 (m, 1H), 2.51 (dd, 1H, J 13.5, 6.2 Hz, H5colch),
3.65 (s, 3H, OCH ), 3.90 (s, 3H, OCH ), 3.94 (s, 3H, OCH ),
spectrum (CDCl , d, ppm): 24.61, 24.79, 28.50, 31.77, 32.72,
3
33.71, 34.65, 36.30, 37.35, 55.89, 56.07, 60.86, 76.77,
103.35, 105.79, 111.91, 120.28, 123.12, 127.58, 128.57,
129.44, 130.04, 132.42, 139.44, 150.20, 152.92, 171.42
(C=O), 172.97 (C=O). Mass spectrum (MALDI-TOF) m/z:
592 [M+K]+.
3
3
3
3.99 (s, 3H, OCH ), 4.65 (m, 1H, H7colch), 6.53 (s, 1H,
3
H4colch), 6.83 (d, 1H, J 11.0 Hz, H11colch), 6.91 (br.s., 1H,
NH), 7.30 (d, 1H, J 11.0 Hz, H12colch), 7.43 (s, 1H,
H8colch). 13C NMR spectrum (CDCl , d, ppm): 24.58, 24.98,
3

428
N. A. Zefirov et al.
25.83, 28.67, 29.93, 30.26, 33.36, 34.26, 34.28, 34.69, 35.96,
36.98, 51.64 (OCH ), 52.15 (C7colch), 56.12 (OCH ), 56.44
ACKNOWLEDGMENTS
3
3
The work was financially supported by RFBR Grant
18–33–01121 mol a. We thank German Academic Exchange
Service DAAD for supporting an academic exchange under
terms of a collaboration agreement between Moscow and
Rostock Universities.
(OCH ), 61.40 (OCH ), 61.62 (OCH ), 66.91, 107.38,
3
3
3
113.02, 125.56, 130.64, 134.22, 135.82, 138.70, 141.68,
146.41, 152.19, 153.57, 164.02, 172.51 (C=O), 173.02
(C=O), 177.47 (C9colch=O). C H NO .
36 45
8
N-[4-(Adamant-1-ylmethoxy)-4-oxobutanoyl]-N-dea-
cetylcolchicine (IIIb) was synthesized analogously to IIIa
from Vd (0.041 g, 0.15 mmol), EEDQ (0.037 g, 0.15 mmol),
and N-deacetylcolchicine (0.050 g, 0.14 mmol) to afford
REFERENCES
1. B. Kumar, R. Kumar, I. Skvortsova, and V. Kumar, Curr. Can-
cer Drug Targets, 17(4), 357 – 375 (2017).
IIIb (0.021 g, 25%) as a yellow oil. PMR spectrum (CDCl ,
3
2. R. A. Stanton, K. M. Gernert, J. H. Nettles, and R. Aneja, Med.
Res. Rev., 31(3), 443 – 481 (2011).
d, ppm, J/Hz): 1.51 – 1.69 (m, 5H), 1.84 – 1.90 (m, 2H), 2.07
(m, 6H), 2.14 (m, 3H, H6colch), 1.83 (m, 1H), 1.93 – 1.98
(m, 3H), 2.25 (m, 1H), 2.42 (m, 1H), 2.69 – 2.70 (m, 2H),
3.64 (s, 3H, OCH ), 3.91 (s, 3H, OCH ), 3.94 (s, 3H, OCH ),
3. S. Wu, J. Fisher, J. Naci, et al., Chem. Res. Toxicol., 26(12),
1840 – 1861 (2013).
4. N. A. Zefirov and O. N. Zefirova, Zh. Org. Khim., 51(9),
1231 – 1240 (2015).
3
3
3
4.00 (s, 3H, OCH ), 4.65 (ddd, 1H, J 12.9, 6.7, 5.9 Hz,
3
H7colch), 6.49 (br.s., 1H, NH), 6.53 (s, 1H, H4colch), 6.86
(d, 1H, J 11.0 Hz, H11colch), 7.33 (d, 1H, J 11.0 Hz,
H12colch), 7.50 (s, 1H, H8colch). 13C NMR spectrum
5. K. Nepali, S. Sharma, M. Sharma, et al., Eur. J. Med. Chem., 77,
422 – 487 (2014).
6. E. C. Breen and J. J. Walsh, Curr. Med. Chem., 17, 609 – 639
(2010).
(CDCl , d, ppm): 20.67, 27.97, 29.91, 33.08, 33.48, 35.15,
7. C. Vilanova, S. Daaz-Oltra, J. Murga, et al., J. Med. Chem.,
57(24), 10391 – 10403 (2014).
3
36.90, 37.04, 39.21, 52.28 (C7colch), 56.08 (OCH ), 56.51
3
8. S. K. Kim, S. M. Cho, H. Kim, et al., Exp. Mol. Med., 45, E19
(2013).
(OCH ), 61.34 (OCH ), 61.59 (OCH ), 74.00 (AdCH ),
3
3
3
2
107.37, 112.74, 125.67, 130.65, 134.26, 135.42, 136.65,
141.64, 151.25, 153.52, 153.45, 163.93, 172.24 (C=O),
173.63 (C=O), 179.15 (C9colch=O). C H NO .
9. P. Thomopoulou, J. Sachs, N. Teusch, et al., ACS Med. Chem.
Lett., 7(2), 188 – 191 (2015).
10. O. N. Zefirova, H. Lemcke, M. Lantow, et al., ChemBioChem,
14(12), 1444 – 1449 (2013).
35 43
8
11. O. N. Zefirova, E. V. Nurieva, D. V. Shishov, et al., Bioorg.
Med. Chem., 19(18), 5529 – 5538 (2011).
EXPERIMENTAL BIOLOGICAL PART
12. O. N. Zefirova, E. V. Nurieva, Ya. S. Glazkova, et al.,
Khim.-farm. Zh., 48, No. 6, 19 – 24 (2014); Pharm. Chem. J.,
48(6), 373 – 378 (2014).
The MTT assay for cytotoxicity was conducted on hu-
man lung carcinoma A549 cell line (CCL-185) using the
published method [13]. Each compound was tested in tripli-
cate. Results were processed statistically (nonlinear regres-
13. N. A. Zefirov, M. Hoppe, I. V. Kuznetsova, et al., Mendeleev
Commun., 28(3), 225 – 344 (2018).
14. O. N. Zefirova, Ya. S. Glazkova, E. V. Nurieva, et al., Izv. Akad.
Nauk, Ser. Khim., 5, 1126 – 1129 (2014).
sion, Excel: XLSTAT software). EC values were deter-
50
15. O. N. Zefirova, E. V. Nurieva, B. Wobith, et al., Mol. Diversity,
21(3), 547 – 564 (2017).
mined from sigmoidal curves (values with p < 0.05 were
considered statistically significant).
16. E. V. Nurieva, N. A. Zefirov, A. V. Mamaeva, et al., Izv. Akad.
Nauk, Ser. Khim., 4, 688 – 693 (2018).
Cell growth and tests with immunofluorescent-la-
beled microtubules were studied using the same cell line
and published methods [16]. Differences among effects on
the microtubule network of control and test compounds were
determined visually using a Nikon Diaphot 300 microscope
(Nikon GmbH, Dusseldorf, Germany) equipped with a
SenSys camera (Photometrics, Munich, Germany).
17. B. Coggiola, F. Pagliai, G. Allegrone, et al., Bioorg. Med. Chem.
Lett., 15(15), 3551 – 3554 (2005).
18. L. Lebeau, P. Ducray, and C. Mioskowski, Synth. Commun.,
27(2), 293 – 296 (1997).
19. J. D. Bagnato, A. L. Eilers, R. A. Horton, and C. B. Grissom, J.
Org. Chem., 69(26), 8987 (2004).
20. T. Mosmann, J. Immunol. Methods, 65(1 – 2), 55 – 63 (1983).