Synthesis and biological evaluation of cis-restrained carbocyclic combretastatin A-4 analogs: Influence of the ring size and saturation on cytotoxic properties

Article abstract of DOI:10.1016/j.bmc.2019.07.048

Combretastatin A-4 (CA-4) is a highly cytotoxic natural product and several derivatives have been prepared which underwent clinical trial. These investigations revealed that the cis-stilbene moiety of the natural product is prone to undergo cis/trans isomerization under physiological conditions, reducing the overall activity of the drug candidates. Herein, we report the preparation of cis-restrained carbocyclic analogs of CA-4. The compounds, which differ by the size and hybridization of the carbocyclic ring have been evaluated for their cytotoxic properties and their ability to inhibit tubulin polymerization. Biological data, supported by molecular docking studies, identified cyclobutenyl and cyclobutyl derivatives of the natural product as highly promising drug candidates.

Full text of DOI:10.1016/j.bmc.2019.07.048

Journal Pre-Proof
Synthesis and Biological Evaluation of cis-Restrained Carbocyclic Combretas-
tatin A-4 Analogs: Influence of the Ring Size and Saturation on Cytotoxic
Properties
Christina Nowikow, Rita Fuerst, Maria Kauderer, Christian Dank, Walther
Schmid, Marian Hajduch, Jiri Rehulka, Sona Gurska, Olena Mokshyna, Pavel
Polishchuk, Istvan Zupko, Petr Dzubak, Uwe Rinner
PII:
S0968-0896(19)30372-4
https://doi.org/10.1016/j.bmc.2019.07.048
BMC 15032
DOI:
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
31 March 2019
19 July 2019
28 July 2019
Please cite this article as: Nowikow, C., Fuerst, R., Kauderer, M., Dank, C., Schmid, W., Hajduch, M., Rehulka, J.,
Gurska, S., Mokshyna, O., Polishchuk, P., Zupko, I., Dzubak, P., Rinner, U., Synthesis and Biological Evaluation
of cis-Restrained Carbocyclic Combretastatin A-4 Analogs: Influence of the Ring Size and Saturation on Cytotoxic
Properties, Bioorganic & Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.bmc.2019.07.048
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will
undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing
this version to give early visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier Ltd.

JOURNAL PRE-PROOF
Synthesis and Biological Evaluation of cis-
Restrained Carbocyclic Combretastatin A-4
Analogs: Influence of the Ring Size and Saturation
on Cytotoxic Properties
Christina Nowikow,a Rita Fuerst,a,b Maria Kauderer,a Christian Dank,a Walther Schmid,a,‡
Marian Hajduch,c Jiri Rehulka,c Sona Gurska,c Olena Mokshyna,c Pavel Polishchuk,c Istvan
Zupko,d Petr Dzubak,c,* Uwe Rinnera,e,*
aInstitute of Organic Chemistry, University of Vienna, Währinger Straße 38, 1090 Vienna,
Austria.
bInstitute of Organic Chemistry, Graz University of Technology, Stremayrgasse 9, 8010
Graz, Austria
cInstitute of Molecular and Translational Medicine, Faculty of Medicine and Dentistry,
Palacky University in Olomouc, Hnevotinska 5, Olomouc, Czech Republic.
dDepartment of Pharmacodynamics and Biopharmacy, University of Szeged, Eötvö u 6, 6720
Szeged, Hungary
eDepartment of Life Sciences, University of Applied Sciences Krems, Piaristengasse 1, 3500
Krems, Austria
‡ Deceased
* Corresponding author for Chemical Part (U.R.): Phone: +43 2732 802175; E-mail:
uwe.rinner@fh-krems.ac.at; Corresponding Author for Biological Part (P.D.): Phone: +420
585 632 150; E-mail: petr.dzubak@upol.cz
1

JOURNAL PRE-PROOF
ABSTRACT Combretastatin A-4 (CA-4) is a highly cytotoxic natural product and several
derivatives have been prepared which underwent clinical trial. These investigations revealed
that the cis-stilbene moiety of the natural product is prone to undergo cis/trans isomerization
under physiological conditions, reducing the overall activity of the drug candidates. Herein,
we report the preparation of cis-restrained carbocyclic analogs of CA-4. The compounds,
which differ by the size and hybridization of the carbocyclic ring have been evaluated for
their cytotoxic properties and their ability to inhibit tubulin polymerization. Biological data,
supported by molecular docking studies, identified cyclobutenyl and cyclobutyl derivatives of
the natural product as highly promising drug candidates.
1. Introduction
The South African willow tree Combretum caffrum has served as a source of medication
against a variety of ailments in traditional African medicine.1 Incipient investigations of plant
extracts showed significant activity against the murine P388D1 lymphoma cell line and a
subsequent study of this plant species by Pettit and coworkers resulted in the isolation of
several biologically active natural products (shown in Figure 1), most of which were
characterized by a highly oxygenated stilbene motif.2-9 An extensive study and biological
evaluation of this group of compounds as antimitotic agents showed combretastatin A-4 (CA-
4, 7) to be the most potent of the series and the natural product soon evolved into a highly
promising anti-tumor drug candidate.3
Since its discovery in 1989, numerous derivatives of CA-4 (7) have been synthesized and
evaluated as potential anti-tumor agents.10-18 CA-4´s biological activity was found to be based
(among other mechanisms)19 on the induction of apoptosis by selectively binding to the
colchicine binding site of tubulin, thereby inhibiting tubulin polymerization and ultimately
2

JOURNAL PRE-PROOF
leading to cell cycle arrest at the transition of meta- to ana- phase. Several analogs were
employed in clinical studies, namely CA-4 phosphate (fosbretabulin; 12),20 AVE-802 (13)21
and Oxi 4503 (14)22
OH
MeO
MeO
MeO
MeO
R1O
R2O
OH
OH
2
OMe
OMe
OMe
OH
OR3
OMe
OMe
OMe
R1, R2= -CH2- R3=H
combretastatin A-2 (
R1=R3=H, R2=CH3
combretastatin A-3 (
1
combretastatin (
)
combretastatin A-1 (
)
3
4
)
)
R1O
R2O
R1O
MeO
OMe
OMe
R4
R2
HO
OMe
OMe
OR3
OR3
MeO
OMe
R1=R2=CH3, R3=H, R4=OH R1=CH3, R2=R3=H
OMe
OMe
combretastatin A-6 (
9
)
5
7
)
combretastatin B-1 (
)
combretastatin A-4
R1=R2=H, R3=CH3
combretastatin A-5 (
(
R1=R3=R4=H, R2=CH3
6
8
)
combretastin B-2 (
)
OH
OH
O
O
O
O
O
O
O
10
11
)
combretastatin D-1 (
)
combretastatin D-2 (
Figure 1. Combretastatin family.
As shown in Figure 2, CA-4 (7) is characterized by a cis-stilbenoid core structure with a
3,4,5-trimethoxy substitution pattern on ring A and a 3-hydroxy-4-methoxy structural motif
on ring B, thus offering the possibility for structural modifications on both aromatic rings, A
and B, as well as on the unsaturated two-carbon bridge connecting the aromatic moieties
(structural motif C). Structure-activity-relationship (SAR) studies illustrated the importance
of the trimethoxy motif on ring A.8, 23 Even small modifications of this structural motif (with
the exception of some analogs bearing a halide instead of the methoxy functionality)24-25
result in a significant decrease of the cytotoxicity.23 Additionally, the hydrogen bonding
donor – a free phenol or an amine26-27 – on ring B is of great significance for high levels of
cytotoxicity. Furthermore, the length and also the cis-geometry of the double bonded
3

JOURNAL PRE-PROOF
bridge28-31 was shown to assume a crucial constitutional function in placing the aromatic
moieties at appropriate distances and dihedral angles to interact with the target tubulin.32
C
MeO
MeO
MeO
MeO
A
B
OMe
OMe
OPO3Na2
OH
OMe
OMe
7
12
)
combretastatin A-4 ( )
combretastatin A-4 phosphate (
fosbretabulin
MeO
MeO
OPO3Na2
OPO3Na2
MeO
OMe
O
MeO
OMe
N
H
OH
NH2HCl
OMe
OMe
14
Oxi 4503 (
)
13
AVE-8062 (
)
umbrabulin
Figure 2. Structural motifs in CA-4 (7) and derivatives used in clinical trials.
Despite its highly promising in vitro activity, CA-4 (7) displayed reduced levels of in vivo
activity, since the stilbenic scaffold is prone to rapidly undergo cis/trans isomerization under
33-35
physiological conditions.24,
Thus, the installation of structural features locking the
scaffold in its active cis configuration is a promising approach to enhance in vivo activity of
novel CA-4 analogs.32
Numerous publications describe the synthesis of cis-constrained derivatives of CA-4 (7),
along with their biological evaluation.10-18, 36-37 The majority of these compounds features
heterocyclic moieties to replace the two-carbon bridge.32, 36, 38 So far, of all carbocyclic
intermediates described herein, only cyclopentene 20 has been prepared before by another
group using 1,2-dibromocyclopentene as starting material. The elegant route, featuring two
Suzuki coupling reactions, allowed the isolation of the stilbene material which was used in a
direct comparison with a series of analogeously prepared heterocyclic derivatives.39
While heterocycles are suited to prevent the system from cis/trans isomerization, the
electronic properties and overall polarity of the compound is significantly altered, which
might be reflected in a different affinity towards tubulin. Furthermore, most studies do not
4

JOURNAL PRE-PROOF
investigate the effect of spatial arrangements of the aromatic rings on the biological activity
of the compound.
In a previous publication, we proposed the synthesis and biological evaluation of
derivatives of CA-4 containing carbocycles of different ring size and hybridization to lock the
cis configuration.40 The overall polarity of these compounds very much resembles that of the
natural compound, while different ring sizes and degrees of saturation allow fine-tuning of
the spatial arrangement of the aromatic rings (see Figure 3). In this respect, derivatives of
combretastatin containing a cyclopropyl moiety have been prepared and the biological
activity was found to be in the submicromolar range when tested against a mini panel of
human cancer cell lines.40 Data obtained from the biological evaluation of the cyclopropyl
deriviatives were highly promising but nevertheless, the activity was lower than that of CA-4.
Thus, it seems that the narrow-angle defined by the three-membered ring seems not ideally
suited to allow binding to tubulin. Ty et al. prepared cyclopropyl derivative 16 in
enantiomerically pure form via a cycloropanation of a (Z)-alkenylboron compound bearing a
chiral auxiliary and the researchers were able to obtain similar activity profiles.41
Interestingly, Ty and co-workers obtained similar activity for both enantiomers of the chiral
cyclopropyl derivative of CA-4; thus, validating the idea of preparing racemic mixtures of
carbocyclic analogs of CA-4.
MeO
MeO
MeO
MeO
OMe
OMe
R
R
7
OMe
16
OMe
17
R=OH;
R=OH; rac-
R=NH2;rac-
15
OMe
OMe
R=NH2;
MeO
MeO
MeO
MeO
MeO
MeO
OMe
20
OMe
22
OMe
18
OH
OH
OH
OH
OH
OH
OMe
OMe
OMe
OMe
MeO
MeO
MeO
MeO
MeO
MeO
OMe
21
OMe
23
OMe
19
rac-
rac-
rac-
5

JOURNAL PRE-PROOF
Figure 3. cis-Restrained CA-4 analogs for biological evaluation.
Herein, we expand our previously presented approach by describing the syntheses and a
more detailed biological evaluation of CA-4 derivatives equipped with carbocycles of ring
sizes ranging from four to six-membered carbocycles, and different degrees of saturation
(Figure 3). The saturated derivatives were prepared in a racemic fashion, since at this point,
we were mainly interested in evaluating the influence of the angle and spatial arrangements
of the aromatic portions. All newly prepared compounds, along with previously synthesized
racemic cyclopropyl derivatives 16 and 17 and CA-4 (7) as positive control were evaluated
against a panel of human cancer cell lines.
2. Results and discussion
2.1. Chemistry
The synthesis of the cyclobutyl series of CA-4 analogs is outlined in Scheme 1. As the
elaboration of an unfunctionalized four-membered ring from the corresponding stilbene
derivatives by means of a [2+2] cycloaddition reaction seemed unfeasible, we decided to
access 18 and 19 via a palladium catalyzed coupling reaction of a boronic ester derived from
vinyl triflate 30 and the requisite bromide 25. Cyclobutanone 29 should be available from
oxaspiropentane 28 via a ring expansion reaction.42
The route commenced with a Wittig reaction of trimethoxybenzaldehyde (26) and bromo-
propyltriphenylphosphonium bromide, which afforded cycloproylidene 27. Exposure of this
material to m-chloroperbenzoic acid delivered the corresponding oxirane which smoothly
underwent the aforementioned ring expansion reaction and cyclobutanone 29 was obtained in
good overall yield. Enolisation of the cyclic ketone and trapping of the intermediary formed
enol anion as its triflate gave compound 30, which was converted to the corresponding
6

JOURNAL PRE-PROOF
boronic ester in a Miyaura type borylation using bis(pinacolato)diborane and a palladium
catalyst.43 To establish the stilbenic core structure of the phenol series, 5-bromo-2-
methoxyphenol was MOM-protected and used as the coupling partner for the cyclobutyl
fragment 30. A Suzuki coupling afforded the desired MOM-protected unsaturated
combretastatin analog 31, which was treated with hydrochloric acid in methanol to release the
phenol group in 18. Hydrogenation of the double bond afforded the saturated cyclobutyl
derivative 19 as racemic mixture.
O
OMe
O
MeO
MeO
MeO
MeO
MeO
MeO
OR
b
c
67%
60%
OMe
OMe
OMe
Br
24
25
a;
95%
26
27
28
R = H
R = MOM
MeO
MeO
MeO
MeO
MeO
MeO
MeO
MeO
d
e, f
g
O
OTf
92%
52%
73%
OMe
OMe
OH
OR
OMe
OMe
OMe
OMe
19
31
18
30
29
rac-
R = MOM
R = H
f; quant.
Reagents and conditions: a)
e)
-78 °C, 73%; [B(pin)]2, PdCl2dppf, KOAc;
b)
b)
d)
MOMCl, DIPEA;
f) 25
95%(BrPr)Ph3PBr, NaH, 60%;
mCPBA, 67%;
LHMDS, Comins' reagent,
g)
h)
HCl, MeOH, quant.; H2, Pd/C, 92%;
, Pd(Ph3)4, DME, EtOH, 52% (two steps);
Scheme 1. Preparation of cyclobutyl derivatives 18 and 19.
The strategies for the elaboration of the cyclopentyl derivatives 20 and 21 as well as the
preparation of the cyclohexyl derivatives 22 and 23 are closely related. For both series, we
decided to employ a route featuring two sequential Suzuki coupling reactions to connect the
carbocycle with the requisite aromatic moieties.
The preparation of cyclopentyl derivatives 20 and 21 is outlined in Scheme 2.39 Starting
from cyclopentanone (32), enolization with LHMDS and triflation with the Comins´reagent44
gave vinyl triflate 33. The coupling partner, boronic ester 35, was derived from 5-bromo-
1,2,3-trimethoxybenzene (34) via a Miyaura type borylation with n-butyllithium and 2-
isopropoxy-4,4,5,5-tetramethyl-1,3,2-dioxaborolane. With both fragments in hand, the first
Suzuki coupling was carried out, yielding compound 36. Cyclopentene 36 was then subjected
7

JOURNAL PRE-PROOF
to a hydroboration/Parikh-Doering oxidation sequence45 to install the required
cyclopentanone scaffold (37). Next, exposure of the ketone to triethyl amine and the Comins’
reagent allowed the isolation of vinyl triflate 38, the substrate for the second Suzuki coupling,
in only moderate yield. The reaction of triflate 38 with boronic ester 40, derived from 4-
bromo-2-methoxyphenol (24) as outlined in Scheme 2, delivered the carbocyclic stilbene
derivative 20 in good yield. The sequence was completed with the hydrogenation of 20,
which afforded the saturated cis-cyclopentyl derivative 21.
MeO
c
a
59%
67%
O
TfO
MeO
33
36
32
OMe
60%
d
O
B
MeO
MeO
Br
MeO
MeO
MeO
MeO
O
b
79%
OH
OMe
OMe
OMe
35
34
37
e
99%
MeO
MeO
MeO
MeO
MeO
g
f
O
OTf
58%
26%
MeO
OMe
OMe
OMe
OH
38
39
20
OMe
h
60%
O
B
MeO
MeO
Br
O
i
60%
MeO
MeO
OMe
OH
OH
40
OH
OMe
24
21
rac-
Reagents and conditions: a)
b)
LHMDS, PhNTf2, THF, -78 °C, 59%;
c)
n-BuLi,
d)
iPrO(pin)B, THF, -78 °C, 79%; Pd(PPh3)4, Na2CO3, H2O, DMF, 67%;
e) f)
BH3.THF, H2O2, NaOH, 60%; DMSO, SO3.py, NEt3, 99%; NEt3, Comins'
g) 40 h)
Pd/C, H2,
reagent, THF, 60 °C, 26%;
, Pd(PPh3)4, Na2CO3, DMF, 58%;
MeOH, 60%; [B(pin)]2, PdCl2dppf, KOAc, 60%.
i)
Scheme 2. Preparation of cyclopentyl derivatives 20 and 21
In order to circumvent the hydroboration reaction in the preparation of cyclopentanone 38,
we also devised an alternative route as outlined in Scheme 3. A Sharpless dihydroxylation of
alkene 36 afforded diol 41 in excellent yield.46 Exposure of this material to zinc iodide then
8

JOURNAL PRE-PROOF
afforded cyclopentanone 38 in 60% yield.47 As the overall yield of 38 could not be increased,
we returned to the hydroboration route for the preparation of the combretastatin analog.
MeO
MeO
MeO
MeO
MeO
MeO
OH
a
b
OH
O
60 %
78 %
OMe
OMe
OMe
36
41
38
Reagents and conditions: a)
b)
ZnI2, 24h,
AD-mix-alpha, 72h, Na2SO3, 78%;
80 °C, 60%.
Scheme 3. Alternative route to intermediate 38.
As pointed out above, the cyclohexyl derivatives 22 and 23 were accessed via a similar
approach, which also featured two successive Suzuki couplings. The route toward 22 and 23
is outlined in Scheme 4. Triflation of cyclohexanone (42) using triflic anhydride and sodium
carbonate in dichloromethane afforded vinyl triflate 43, which was converted to cyclohexene
44 by reaction with boronate 35. Again, a hydroboration/oxidation/triflation sequence was
employed to obtain cyclohexenyl intermediate 47, which served as a direct precursor for the
second Suzuki coupling. The palladium-catalyzed reaction of vinyl triflate 47 and boronic
ester 40 afforded the cyclohexenyl derivative of CA-4 (22) in excellent 90% yield. Finally,
the fully saturated analog (23) was obtained in racemic form via hydrogenation of the double
bond in compound 22.
9

JOURNAL PRE-PROOF
MeO
MeO
a
b
85%
57%
O
TfO
42
43
44
OMe
c
69%
MeO
MeO
MeO
MeO
d
O
OH
45
57%
46
OMe
e
OMe
85%
MeO
MeO
MeO
MeO
f
OTf
90%
OMe
OMe
22
OH
47
OMe
85%
g
Regents and conditions: a)
Tf2O, Na2CO3,
MeO
MeO
b) 35
CH2Cl2, 57%;
, Pd(PPh3)4, Na2CO3,
c)
H2O, DMF, 85%; BH3.THF; H2O2, NaOH,
d) e)
69%;
PhNTf2, 0 °C, 85%;
g)
DMSO, SO3.py, NEt3, 57%;
NaH,
, Pd(PPh3)4, Na2CO3,
H2, Pd/C, MeOH, 85%.
f) 40
OMe
OH
DMF, 90%;
23
OMe
rac-
Scheme 4. Preparation of cyclohexyl derivatives 22 and 23.
2.2. Biological evaluation
All carbocyclic analogs of combretastatin A-4, as well as the natural product (7) and its
amino derivative (15), were evaluated for their in cellulo cytotoxic activity against a panel of
human cancer and two non-tumor fibroblast cell lines by using the standard MTS cell
viability assay.48 The cancer cell lines used in the investigation were derived from acute T-
lymphoblastic leukemia CCRF-CEM, chronic myeloid leukemia K562 and their multidrug-
resistant counterparts (CEM-DNR, K562-TAX) expressing the multidrug resistance-
associated protein 1 (MRP-1) and P-glycoprotein (P-gp) transporter proteins involved in
tumor resistance, respectively.49 Additionally, solid tumor cell lines, including lung (A549)
and colon (HCT116, HCT116p53-/-) adenocarcinomas, an osteosarcoma cell line (U2OS),
and for comparison, two healthy human, non-cancer fibroblast lines (BJ, MRC-5) as well as
an endothelial (HUVEC) cell line were included in the study.
10

JOURNAL PRE-PROOF
The results of this investigation are summarized in Table 1. In analogy to previously
published results,50 we have identified CA-4 (7) to be highly active against the hematological
cancer cell lines CCRF-CEM and K562 with a good therapeutic index (TI) as an indication of
the anticancer activity (TIavg = 36.5) if compared with endothelial HUVEC or fibroblast BJ
and MRC-5 cell lines. Low activity was observed against the lung tumor A549, however,
picomolar activity was measured against the colorectal adenocarcinoma and osteosarcoma
cell lines HCT116 and U2OS, respectively. Surprisingly, even higher activity was observed
against the HCT116 cell line which lacks the p53 gene.
The amino derivative of CA-4 (15) as well as the cyclopropyl derivatives 16 and 17 and
cyclobutene containing compound 18 showed sub-micromolar activities against different
cancer cell lines (Table 1), however, the therapeutic indices are rather poor. In contrast, the
cyclobutane derivative 19 displays promising selectivity for the two leukemia cell lines
CCRF-CEM and K562 with a TIavg of 316 as well as lung adenocarcinoma cell line A549,
while the activity against the multidrug-resistant sublines (CEM-DNR and K562-TAX) was
quite low. On the other side, the cyclopentyl and cyclohexyl derivatives 20, 21, and 22, 23
have moderate to good therapeutic indices (TIavg = 6.59 – 45.3). However, cyclohexene
derivative 22 showed the lowest cytotoxic activity (IC50(CCRF-CEM) = 4.41 moL) of all
compounds tested during the course of this study. This can easily be rationalized as the
spatial arrangement of the aromatic rings significantly varies from all other derivatives
included in this investigation.
Generally, derivatives with larger carbocycles seem to have lower activity than analogs
with smaller carbocycles on the two-carbon bridge connecting the aromatic moieties. These
findings were later confirmed by means of docking studies, as outlined below.
11

JOURNAL PRE-PROOF
Table 1. Cytotoxic/cytostatic activity IC50 (nmoL/L) of 7, 15 – 23 against cancer and healthy
cell lines was measured by MTS assay. The values are the means of three independent
experiments with a standard deviation of maximum 20% of the average value. The
therapeutic index for TIavg was calculated as the median of IC50 value of nontumor cell lines
divided by the median of IC50 value of hematological non-resistant cell lines.
CA-4 (7)
15
16
17
18
19
20
21
22
23
CCRF-CEM 0.708
51.2
15.7
15.7
13.8
3.18
693
223
631
53.5
47.2
515
270
878
215
64.2
96.2
76.3
112
317
14.3
3.17
1530
2.67
430
24.5
1890
23.2
349
110
4410
183
CEM-DNR
K562
319
0.0282
1.75
1710
100
54300
4140
3120
199
K562-TAX
HCT116
1340
1110
45600
58700
2150
1010
0.547
477
126
HCT116
p53-/-
0.0794
3.17
50.2
75.5
15.2
535
164
1500
58900
1280
U2OS
A549
0.766
200
6.00
51.2
11.3
87.3
9.19
57.2
693
130
286
204
66.8
515
89.5
308
134
14.8
96.2
175
141
18.2
415
3.17
1600
137
92.5
24.5
146
70.2
255
853
110
42200
4410
1630
183
HUVEC
BJ
4.17
28.6
6.15
1750
99800
1840
6130 1.17×105 19200
MCR-5
1030
1340
98000
4940
TI
(BJ/CCRF-
CEM)
40.4
7.62
1.71
3.55
0.413 0.598
1.47
12.2
43.2
3.35
2.87
1.16
55.7
26.4
105
TI
(HUVEC/H
CT116)
2.75
1.39
1.58
1.70
1.82
TI (MRC-
5/CCRF-
CEM)
8.69
35.2
0.180 0.294 0.260
0.189
1.07
324
316
10.4
6.59
12.2
29.3
22.2
24.5
27.0
45.3
TIavg*
1.07
0.312 0.254
12

JOURNAL PRE-PROOF
Table 2. Summary of the cell cycle (G0/G1, S, G2/M), apoptosis (sub G1), mitotic marker
(pH3Ser10 positive), DNA and RNA synthesis analyses in CCRF-CEM for cis-restrained
combretastatin analogs 7 (CA-4), 15 – 23.
% of total cellular population
pH3Ser10
Compd. Concentration
control
sub-G1
(apoptosis)
DNA
synthesis synthesis
RNA
G0/G1
S
G2/M
(mitotic
marker)
3.71
32.03
33.57
30.34
33.57
33.52
34.21
31.71
34.51
32.56
34.28
4.74
6.8
40.89
1.37
1.2
46
12.12
60.72
61.1
0.97
46.51
47.84
48.37
49.15
43.35
46.69
43.86
50.68
45.85
48.88
1.14
52.13
25.5
47.37
42.38
45
1xIC50
5xIC50
1xIC50
5xIC50
1xIC50
5xIC50
1xIC50
5xIC50
1xIC50
5xIC50
1xIC50
5xIC50
1xIC50
5xIC50
1xIC50
5xIC50
1xIC50
5xIC50
1xIC50
5xIC50
37.91
37.7
CA-4
(7)
25.58
29.37
28.25
25.19
24.89
28.2
1.19
1.38
1.38
1.09
1.54
1.11
1.33
1.62
41.87
40.88
39.49
39.73
38.49
41.49
41.1
40.2
42.3
41.28
35.04
42.77
37.89
41.49
45.09
33.6
63.77
55.84
60.73
57.43
53.37
65.29
56.01
63.81
13.77
14.83
12.97
16.05
14.08
15.46
12.82
17.44
12.71
14.41
53.2
15
16
17
18
19
20
21
22
23
46.12
50.82
51.4
71
29.6
45.13
41.26
51.45
37.22
23.94
54.87
29.88
35.55
41.87
54.11
38.76
30.97
40.96
42.66
34.58
44.36
44.29
47.54
44.22
47.43
43.04
46.07
42.36
45
29.99
29.72
57.17
55.81
57.18
55.54
56.92
51.14
95.31
49.25
57.91
49.33
1.52
3.8
1.1
4.28
4.16
4.63
4.5
1.67
1.39
1.5
1.15
4.7
1.71
3.42
4.38
2.08
44.31
1.53
13

JOURNAL PRE-PROOF
Figure 4. Cell cycle analysis of the CCRF-CEM cell line treated with 1 x IC50 concentration
of the tested compounds.
Interestingly, together with compound 7 (CA-4), derivatives 19, 20, 21, 22 and 23 are less
active against resistant cell lines CEM-DNR and K562-TAX in comparison to parental
CCRF-CEM / K562 cell lines which indicates a possible transport of these compounds by
MRP-1 and/or P-gp transporters. Remarkably, derivatives 16 and 17 were more toxic to
multidrug-resistant cells. Maybe the relatively narrow arrangement of the aromatic rings
prevents interaction of the substrate with the transporter proteins.
Because CA-4 (7) is a well-known inhibitor of tubulin polymerization and has been
described as mitotic poison,29 all derivatives were tested for the effect on the cell cycle by
flow cytometry (FACS).48 In this analysis, only derivatives 7 (CA-4), 15, 16, 17 and 18
showed the expected inhibition of the cell cycle in G2/M phases after 24 hours in the CCRF-
CEM cell line (Figure 4). The high phosphorylation positivity of histone H3 at Ser10
(pH3Ser10) was used to indicate a mitotic block of the cell cycle (Table 2).51
When exposed to the natural product (7) or its synthetic analogs, the microscopic analysis
of colorectal cancer cells (HCT116) showed evidence of “mitotic catastrophe”,52 namely
14

JOURNAL PRE-PROOF
multipolar or disrupted mitotic spindles (Figure 5A), which are consistent with an alteration
of the microtubular apparatus in the cells treated with the drug candidates. These findings are
also in agreement with data obtained from the cell cycle analysis. Additionally, the DNA
synthesis monitored by the incorporation of 5-bromo-2'-deoxyuridine (BrdU) decreased in
cells exposed to mitotic blockers (7 (CA-4), 15, 16, 17, 18). While the inhibition of DNA is
proportional to the accumulation of the cells in the G2/M cell cycle phase, interference with
the DNA replication machinery is unlikely. Interestingly, derivative 19, which exhibited the
highest therapeutic index, is inhibiting RNA synthesis at 1 x IC50 / 5 x IC50 (37.22 /
23.94%) while the DNA synthesis is not interrupted. No effect of derivative 19 on the cell
cycle was observed This interesting finding deserves further attention in future studies.
To evaluate the direct activity of CA-4 (7) and its carbocyclic analogs on the tubulin
polymerization, an in vitro (non-cellular) tubulin polymerization test on pure tubulin (99%)
was performed at a concentration of 10 moL of the potential drugs.53 With one exception
(22), all tested compounds have shown the expected inhibitory effect on tubulin
polymerization (Figure 5B, C,). Again, the highest inhibition was observed with those
compounds which also showed highest in cellulo activity on the cell cycle, thus, verifying our
previous findings. Derivatives 19, 20, 21 and 23 displayed activity comparable with
nocodazole, a well-known reversible inhibitor of tubulin polymerization, and were highly
cytotoxic to the cells in the 3-day MTS assay (and confirmed findings of the research group
of Tron on cyclopentene 20).39 However, the compounds did not block cells in the mitotic
phase after 24 hours of treatment (Table 2), thus suggesting slower effect on cellular kinetics
or different modes of action.19
15

JOURNAL PRE-PROOF
Figure 5. On the left (A) microtubule staining in the colorectal HCT116 cell line as
visualized by immunofluorescence localization of α-tubulin (green), nuclei are stained with
4′,6-diamidino-2-phenylindole (DAPI, blue). Representative images of mitosis are shown.
Cells were treated for 24 hours with dimethyl sulfoxide (DMSO), 1 moL/L of paclitaxel or
nocodazole and corresponding 1 x IC50 concentration of compounds 7 (CA-4), 15, 16, 17 and
18. Tubulin in vitro assembly (B, C) in the presence of combretastatins, nocodazole, and
paclitaxel (all at concentration 10 µmoL/L) or an equivalent volume of DMSO (control). The
in vitro assay is indicating the impact of tested derivatives on tubulin polymerization. Tests
were done in duplicates.
Table 3. Maximal velocity of polymerization (Vmax) calculated from tubulin polymerization
curves in the presence of 10 µmoL/L combretastatins, nocodazole, paclitaxel or an equivalent
16

JOURNAL PRE-PROOF
volume of DMSO and values of scoring function from molecular docking simulations. There
was found a significant positive correlation between the values of VinaLigand efficiency and
Vmax with R = 0.77.
Vmax
(mOD/min)
AutoDock
score
VinaLigand efficiency(LE)
Compound
DMSO
16.8
13.0
4.4
Nocodazole
-7.8
-8.8
-8.4
-0.56
-0.49
-0.47
CA-4 (7)
15
5.2
16
5.6
-8.1 (-7.9; -8.3)
-7.8 (-7.8; -7.8)
-9.7
-0.43
-0.41
-0.49
17
5.4
18
9.0
19
10.4
10.6
12.0
16.2
11.6
84.0
-7.05 (-6.6*; -7.5) -0.35
-8.8 -0.42
-7.45 (-6.6*; -8.3) -0.35
-7.7 -0.35
-7.95 (-7.9; -8.0) -0.36
20
21
22
23
Paclitaxel
* scores of poses which did not correspond to the pose of combretastatin A-4 (7)
17

JOURNAL PRE-PROOF
CA-4 (7)
15
17
19
21
23
16
18
20
22
Figure 6. Top docked poses in comparison with pose of the reference crystallized ligand (in
grey color).
18

JOURNAL PRE-PROOF
The X-ray structure of a tubuline-combretastatin A-4 complex (5LYJ) was used for docking
studies.54 Self-docking of CA-4 (7) successfully reproduced the ligand pose (RMSD =
0.42 Ǻ). Since compounds 16, 17, 19, 21, 23 were synthesized and tested as racemic
mixtures, both enantiomers were used for docking studies and the scores were averaged. A
moderate correlation between the computed docking scores and measured activity values is
observed for compounds 7 (CA-4), 15-23 with a Spearman correlation coefficient of 0.49
(Table 3). Docking poses of almost all studied compounds were similar to the pose of CA-4
except for one of the enantiomers of compound 19 and 21 (Figure 6). The interaction patterns
were almost identical. The only notable difference was a greater number of hydrophobic
contacts for compounds with the larger carbocyclic ring systems in place of the stilbene
moiety. This could create a bias in the estimated docking scores. Therefore, we calculated a
weighted score by the division of the docking score of the particular compound by the
number of its carbon atoms. This is one of so-called ligand efficiency (LE) measures which
can be used to rank compounds in docking studies.55 Ligand efficiency correlates much better
with the observed activity values, and a Spearman’s correlation coefficient of R = 0.77
indirectly confirms the validity of the found docking poses.
3. Conclusion
Summarizing, several carbocyclic ring containing analogs of the anti-tumor drug CA-4 (7)
have been prepared in short synthetic routes from readily available starting materials. The
cytotoxic properties of CA-4 (7) and its analogs were evaluated, along with their potential to
act as inhibitors for tubulin polymerization. According to docking simulations all compounds
(at least one of enantiomers) may adopt the binding site cavity and demonstrated similar
orientation and interaction pattern to the reference compound – combretastatin A-4 (7).
19

JOURNAL PRE-PROOF
Cyclobutene derivative 18 was identified as the promising drug candidate with high
activity against several cancer cell lines. The excellent anticancer properties are still
compromised with relatively low selectivity against healthy human fibroblasts, although, the
selectivity against HUVEC cells is better than in the case of the natural product. Cyclobutane
19 exhibits activity comparable to the natural product against CCRF-CEM and K562 cell
lines with low selectivity against non-cancerous cell lines and thus has a remarkable TI.
Therefore, our future studies will be aimed towards detailed in vivo evaluation (safety and
efficacy) of cyclobutene 18 and cyclobutane 19 as potential novel drug candidates.
4. Experimental protocols
4.1. Chemistry
General methods: All non-aqueous reactions were carried out under a positive pressure of
argon using oven-dried (100 °C) or flame-dried glassware (under vacuum), unless noted
otherwise. Tetrahydrofuran (THF) was dried by distillation from potassium under argon.
Diethyl ether, dimethoxyethane and toluene were purified by distillation and dried by
distillation from sodium/benzophenone ketyl under argon. Dimethyl sulfoxide (DMSO) and
N,N-dimethylformamide were dried by distillation from calcium hydride under reduced
pressure. Dichloromethane (DCM) was purified by distillation and dried by distillation from
phosphor pentoxide and passage over aluminum oxide, neutral activity. Dry solvents were
stored under an argon atmosphere over molecular sieve (4 Ǻ). Triethyl amine,
diisopropylethylamine and diisopropylamine were distilled from calcium hydride under an
atmosphere of argon prior to use. All other commercially available reagents were used
without further purification. Except if indicated otherwise, reactions were magnetically
stirred and monitored by thin layer chromatography using silica gel 60-F254 glass plates. The
plates were developed with a mixture of hexane/EtOAc or toluene/EtOAc. Unless the
20

JOURNAL PRE-PROOF
compound was colored, UV-active spots were detected at longwave UV (254 nm) or
shortwave (180 nm). Most plates were additionally treated with one of the following
visualization reagents: CAM [H2SO4 (conc., 22 mL), phosphormolybdic acid (20 g),
Ce(SO4)2 (0.5 g), 378 mL H2O)] or silica gel impregnated with iodine. Flash column
chromatography was performed with silica gel 60 (0.040-0.063 μm, 240-400 mesh). Optical
rotations were measured at the sodium D line with a 100 mm path length cell, and are
reported as follows: [α]T , concentration (g/100 mL), and solvent. NMR spectra were
D
recorded either on a 400 or 600 MHz spectrometer. Unless stated otherwise, all NMR spectra
were measured in CDCl3 solutions and referenced to the residual CDCl3 signal (1H, δ = 7.26,
13C, δ = 77.16). All 1H and 13C shifts are given in ppm (s = singlet, d = doublet, dd = doublet
of doublets, t = triplet, q = quartet, quint = quintet, m = multiplet, br = broadened signal).
Coupling constants J are given in Hz. Assignments of proton resonances were confirmed,
when possible, by correlated spectroscopy (COSY, HSQC, HMBC, TOCSY, NOESY). IR
spectra are reported in wave numbers (cm-1). All compounds were measured using a single
reflection monolithic diamond ATR module. High-resolution mass spectra were performed
on a mass spectrometer using ESI-mode and a UHR-TOF (Qq-TOF) mass analyzer
(acetonitrile/MeOH 1:1, +1% H2O). The purity of all tested compounds was above 95%
which was confirmed by HPLC analysis and NMR spectroscopy. All compounds subjected to
biological evaluation have been examined for known classes of assay interference. The
software tool used for this examination (http://zinc15.docking.org/) did not discover any
compounds with potential PAINS liability.
4.1.1. 4-Bromo-1-methoxy-2-(methoxymethoxy)benzene (25). To a solution of bromide 24
(9.9 mmol, 2.0 g) in CH2Cl2 (10 mL) at 0 °C was added DIPEA (49.5 mmol, 8.6 mL) and
MOM-Cl (29.7 mmol, 2.3 mL; caution - powerful carcinogenic agent). The reaction mixture
21

JOURNAL PRE-PROOF
was allowed to warm to RT and was stirred until total consumption of the starting material
(48 h). The reaction was quenched with aqueous sat. NH4Cl (20 mL). The layers were
separated, and the aqueous layer was extracted with CH2Cl2 (2 x 10 mL). The combined
organic extract was dried over Na2SO4, and the solvent was removed under reduced pressure.
The residue was purified by flash column chromatography (hexane/ethyl acetate 19:1)
1
affording the protected product 25 (2.33 g, 95%) as a colorless oil. H NMR (400 MHz,
CDCl3) δ [ppm] 7.29 (d, J = 2.36 Hz, 1H), 7.09 (dd, J = 8.58, 2.36 Hz, 1H), 6.76 (d, J = 8.58
Hz, 1H), 5.21 (s, 2H), 3.85 (s, 3H), 3.51 (s, 3H); 13C NMR (100 MHz, CDCl3) δ [ppm] 149.2,
147.4, 125.2, 119.8, 113.2, 112.8, 95.8, 56.5, 56.2.
4.1.2. 5-(Cyclopropylidenemethyl)-1,2,3-trimethoxybenzene (27). A vigorously stirred
suspension of 3-bromopropyl triphenylphosphonium bromide (2.40 g, 0.80 eq) in THF (50
mL) was charged with NaH (450 mg, 1.6 eq) and 3 drops of dry methanol and the mixture
was stirred at 70°C for 6 h. After cooling to RT, a solution of trimethoxybenzaldehyde
(26;1.27 g, 6.48 mmol) in THF (15 mL) was added and the solution was stirred at 70°C for 2
h. The reaction was cooled to RT, quenched with ice/water (100 g), the layers were separated,
and the aqueous phase was extracted with EtOAc (3 x 25 mL). The combined organic layers
were washed with brine (20 mL) and dried over Na2SO4. After evaporation of the solvent in
vacuo, the crude product was purified by flash column chromatography (hexane/EtOAc 2:1),
1
yielding 856 mg (3.89 mmol, 60%) of product 27. H-NMR (400.13 MHz, CDCl3),  [ppm]
6.78 (s, 2H); 6.65-6.68 (m, 1H); 3.88 (s, 6H); 3.85 (s, 3H); 1.38-1.45 (m, 2H); 1.16-1.22 (m,
2H); 13C-NMR (100.61 MHz, CDCl3),  [ppm] 153.5, 134.2, 123.8, 118.3, 103.9, 61.1, 56.2,
4.2, 0.7; IR  3001, 2959, 2923, 2855, 2362, 2350, 2336, 1578, 1505, 1463, 1416, 1353,
1317, 1239, 1130, 1005, 843, 749, 681, 635 cm-1; HRMS (ESI) (m/z): [M]+ calcd. for
C13H16O3 220.1099; found 220.1105.
22

JOURNAL PRE-PROOF
4.1.3. 2-(3,4,5-Trimethoxyphenyl)cyclobutanone (29). To
a
solution of 5-
(cyclopropylidenemethyl)-1,2,3-trimethoxybenzene (27;1.12 g; 5.0 mmol) in CH2Cl2 (50 mL)
was added m-chloroperbenzoic acid (1.72 g 10.0 mmol) at 0 °C. After 2 h, the reaction
mixture was diluted with CH2Cl2 (150 mL), washed with saturated NaHCO3 solution (2 x 50
mL) and brine (20 mL), dried over Na2SO4 and concentrated in vacuo. The crude product was
purified with flash column chromatography (hexane/EtOAc 5:1), yielding 791 mg (3.35
1
mmol, 67%) of cyclobutanone 29 as a pale-yellow oil (791 mg; 3.35 mmol; 67%). H-NMR
(400.13 MHz, CDCl3),  [ppm] 6.47 (d, 2H, J = 0.40 Hz); 4.40-4.56 (m, 1H); 3.86 (s, 6H);
3.82 (s, 3H); 3.17-3.29 (m, 1H); 2.98-3.09 (m, 1H); 2.47-2.60 (m, 1H); 2.15-2.26 (m, 1H);
13C-NMR (100.61 MHz, CDCl3),  [ppm] 207.3, 153.5, 132.3, 104.2, 64.6, 61.0, 56.3, 44.9,
18.1; IR  3001, 2974, 2938, 2360, 2348, 2338, 1700, 1585, 1507, 1419, 1343, 1262, 1238,
1130, 1001, 910, 769, 749 cm-1; HRMS (ESI) (m/z): [M]+ calcd. for C13H16O4 236.1049;
found 236.1050.
4.1.4. 2-(3,4,5-Trimethoxyphenyl)cyclobut-1-enyl) trifluoromethanesulfonate (30). To a
solution of LHMDS (6.09 mL, 1.0 M, 1.2 eq) in THF was added a solution of cyclobutanone
29 (1.2 g, 5.08 mmol) in THF (10 mL) dropwise at -50 °C. After 20 min, a solution of N-(5-
chloro-2-pyridyl)bis(trifluoromethanesulfonimide) (2.4 g 1.2 eq, 6.09 mmol) in THF (10 mL)
was added slowly to keep the internal temperature at -50 °C. Stirring was continued for 1 h,
after which time TLC showed total consumption of the starting material. The reaction was
quenched by addition of a saturated aqueous NH4Cl solution (30 mL). The aqueous phase
was extracted with diethyl ether (3 x 20 mL), the combined organic layer was washed with
H2O (20 mL) and brine (15 mL) and dried over Na2SO4. The solvent was removed in vacuo
and the crude dark yellow oil was purified by flash column chromatography (pentane/EtOAc
1
2:1) affording triflate 30 (1.37 g; 3.71 mmol; 73%) as a yellow oil. H-NMR (400.13 MHz,
CDCl3),  [ppm] 6.64 (s, 2H); 3.87 (s, 6H); 3.86 (s, 3H); 3.03 (t, 2H, J = 3.42 Hz); 2.52 (t,
23

JOURNAL PRE-PROOF
2H, J = 3.42 Hz); 13C-NMR (100.61 MHz, CDCl3),  [ppm] 153.6, 139.0, 130.3, 128.7,
126.3, 120.2, 104.2, 61.1, 56.3, 31.7, 21.8; IR  2945, 2843, 2363, 2353, 2345, 1690, 1583,
1505, 1466, 1419, 1344, 1234, 1211, 1129, 1083, 1040, 1004, 901, 858, 840, 632, 616 cm-1;
HRMS (ESI) (m/z): [M]+ calcd. for C14H15F3O6S 368.0541; found 368.0546.
4.1.5.
1,2,3-Trimethoxy-5-(2-(4-methoxy-3-(methoxymethoxy)phenyl)cyclobut-1-
enyl)benzene (31). A flame dried Schlenk tube was charged with cyclobutenyl
trifluoromethanesulfonate 30 (250 mg 0.68 mmol), PdCl2(dppf) (50 mg, 0.069 mmol; 10
mol%), bis(pinacolato)diboron (190 mg; 0.75 mmol; 1.1 eq) and KOAc (200 mg, 2.1 mmol;
3.0 eq) and the mixture was flushed with argon three times. THF (20 mL) was added and the
reaction mixture was heated at reflux for 3 h. The reaction mixture was filtered through a
plug of silica gel and washed with EtOAc. The solvents were evaporated and the crude
boronic ester was used without further purification. To a degassed (three freeze-pump-thaw-
cycles) solution of 4-bromo-1-methoxy-2-(methoxymethoxy)benzene 25 (141 mg, 0.57
mmol, 0.83 eq) in DME (5 mL), Pd(PPh3)4 (70 mg; 0.06 mmol; 10 mol%) was added, and the
resulting solution was stirred for 20 min. A solution of the boronic ester in EtOH (2.5 mL)
was added and stirring was continued for 20 min. After the addition of an aqueous solution of
Na2CO3 (0.6 mL; 2 M), the mixture was heated at 75 °C for 30 min. The solution was cooled
to RT, the precipitate was filtered off, washed with EtOAc (10 mL) and the filtrate was
evaporated to dryness. The residue was again dissolved in EtOAc (20 mL), washed with
water (10 mL) and dried over Na2SO4. Flash column chromatography of the residue
1
(hexane/EtOAc 3:1) yielded compound 31 (115 mg; 0.3 mmol; 52%) as a yellow solid. H-
NMR (400.13 MHz, CDCl3),  [ppm] 7.32 (1H, d, J = 2.00 Hz); 7.19 (dd, 1H, J = 8.36; 2.00
Hz); 6.85 (d, 1H, J = 8.40 Hz); 6.75 (s, 2H); 5.17 (s, 2H); 3.88 (s, 3H); 3.85 (s, 3H); 3.82 (s,
6H); 3.48 (s, 3H); 2.72 (s, 4H); 13C-NMR (100.61 MHz, CDCl3),  [ppm] 153.3, 149.7,
146.4, 137.9, 137.1, 129.5, 132.1, 120.9, 114.9, 111.6, 103.4, 95.8, 61.0, 56.4, 56.2, 56.1,
24

JOURNAL PRE-PROOF
26.9, 26.8; IR  3390, 2985, 2934, 2837, 2348, 2334, 1581, 1508, 1474, 1455, 1371, 1355,
1335, 1255, 1239, 1148, 1127, 1079, 1006, 983, 853, 678, 632 cm-1; HRMS (ESI) (m/z):
[M]+ calcd. for C22H26O6 386.1729; found 386.1736.
4.1.6. 2-Methoxy-5-(2-(3,4,5-trimethoxyphenyl)cyclobut-1-enyl)phenol (18). To a solution
of MOM-ether 31 (245 mg; 0.634 mmol) in MeOH (10 mL) was added an aqueous solution
of HCl (3.0 mL; 3 M HCl) and the mixture was stirred at RT for 48h. Water was added (20
mL) and the reaction mixture was extracted with CH2Cl2 (3 x 10 mL). The combined organic
layer was washed with water (5 mL), dried over Na2SO4 and the solvent was removed in
vacuo. The crude material was purified by HPLC (hexane/EtOAc 4:1), yielding phenol 18
1
(217 mg;0.634 mmol; quant.) as a pale-yellow solid. H-NMR (400.13 MHz, CDCl3), 
[ppm] 7.20 (d, 1H, J = 2.04 Hz); 7.06 (dd, 1H, J = 8.32 Hz, J = 2.04 Hz); 6.81 (d, 1H, J =
8.36 Hz); 6.78 (s, 2H); 5.56 (s, 1H); 3.90 (s, 3H); 3.87 (s, 3H); 3.83 (s, 6H); 2.72 (s, 4H); 13C-
NMR (100.61 MHz, CDCl3),  [ppm] 153.3, 146.3, 145.6, 138.0, 137.1, 132.1, 130.0, 118.5,
112.6, 110.5, 103.4, 61.1, 56.3, 56.1, 26.8, 26.8; IR  3435, 2940, 2911, 2837, 1580, 1504,
1463, 1416, 1355, 1333, 1289, 1262, 1238, 1187, 1126, 1062, 1029, 1003, 950, 923, 881,
842, 832, 796, 649 cm-1; HRMS (ESI) (m/z): [M]+ calcd. for C20H22O5 342.1467; found
342.1473.
4.1.7. 2-Methoxy-5-(2-(3,4,5-trimethoxyphenyl)cyclobutyl)phenol (19). To a solution of
cyclobutene 18 (60 mg; 0.176 mmol) in MeOH (3 mL) was added Pd/C (10 mg; 10%) as
hydrogenation catalyst and the reaction mixture was stirred under an H2 atmosphere until the
total consumption of the starting material (30 min). The mixture was filtered, and the solvent
was evaporated in vacuo, to afford cyclobutene 19 (0.162 mmol; 56 mg; 92%) as a pale-
yellow oil. 1H-NMR (400.13 MHz, CDCl3),  [ppm] 6.64 (d, 1H, J = 2.28 Hz); 6.60 (d, 1H, J
= 8.08 Hz); 6.41 (dd, 1H, J = 8.08 Hz, J = 2.27 Hz); 6.13 (s, 2H); 5.47 (bs, 1H); 3.86-3.88
(m, 2H); 3.78 (s, 3H); 3.74 (s, 3H); 3.67 (s, 6H); 2.25-2.49 (m, 4H); 13C-NMR (100.61 MHz,
25

JOURNAL PRE-PROOF
CDCl3),  [ppm] 152.5, 145.2, 144.7, 137.4, 136.1, 135.0, 119.7, 114.4, 110.3, 105.4, 60.9,
56.1; 45.5, 45.1, 25.0, 24.1; IR  3419, 2940, 2834, 1590, 1512, 1462, 1421, 1348, 1323,
1273, 1239, 1129, 1027, 1009, 828, 630 cm-1; HRMS (ESI) (m/z): [M]+ calcd. for C20H24O5
344.1624; found 344.1623.
4.1.8. Cyclopent-1-en-1-yl trifluoromethanesulfonate (33). A solution of LHMDS (2.52
mL, 1.0 M in THF, 2.52 mmol) was added dropwise to a solution of cyclopentanone (2.8
mmol, 235 mg, 0.25 mL) in THF (5 mL) at -78 °C. After 45 min, a solution of PhNTf2 (2.8
mmol, 0.98 g) in THF (4.5 mL) was slowly added. The mixture was allowed to warm to RT
overnight and the reaction was quenched with aqueous NH4Cl (15 mL). The layers were
separated, and the aqueous layer was extracted with diethyl ether (3 x 30 mL). The combined
organic layer was then washed with aqueous Na2CO3 (5%; 20 mL), dried over Na2SO4 and
the solvent was removed under reduced pressure to afford 33 as colorless liquid (330 mg;
1
59%). H NMR (400 MHz, CDCl3) δ [ppm] 5.62 (q, J = 2.22 Hz, 1H), 2.61-2.52 (m, 2H),
2.45-2.36 (m, 2H), 2.08-1.98 (m, 2H); IR  2928, 1419, 1247, 1212, 1143, 1053 cm-1.
4.1.9. 4,4,5,5-Tetramethyl-2-(3,4,5-trimethoxyphenyl)-1,3,2-dioxaborolane (35). To a
solution of 5-bromo-1,2,3-trimethoxybenzene (34; 40.5 mmol, 10.0 g) in THF (250 mL) n-
butyllithium (44.6 mmol, 2.5 M in hexane, 17.8 mL) was added dropwise at -78 °C. After 1 h
at -78 °C, 2-isopropoxy-1,3,2-dioxaborolane (52.6 mmol, 10.7 mL) was added and the
mixture was allowed to warm to RT overnight. The solution was quenched with NH4Cl (100
mL), the layers were separated, and the aqueous layer was extracted with EtOAc (3 x 40 mL).
The combined organic layer was washed with water (30 mL) and brine (30 mL) and dried
with Na2SO4 before the solvent was removed under reduced pressure affording 35 (9.42 g;
79%) as white crystals. 1H NMR (400 MHz, CDCl3) δ [ppm] 7.03 (s, 2H), 3.90 (s, 6H), 3.87
(s, 3H), 1.34 (s, 12H); 13C NMR (100 MHz, CDCl3) δ [ppm] 153.1, 141.1, 111.5, 104.9, 84.0,
26

JOURNAL PRE-PROOF
61.0, 56.3, 25.0; ESI HRMS: [M+Na]+ calcd.: C15H23BO5 294.1639, found 294.1648; IR 
3727, 2978, 2937, 1578, 1451, 1407, 1370, 1276, 1232, 1143, 1125 cm-1.
4.1.10. 5-(Cyclopent-1-en-1-yl)-1,2,3-trimethoxybenzene (36). A three-necked-flask was
charged with an aqueous solution of Na2CO3 (0.92 mmol, 98 mg; 1.0 mL), cyclopent-1-en-1-
yl trifluoromethanesulfonate (33; 0.46 mmol, 135 mg), 4,4,5,5-tetramethyl-2-(3,4,5-
trimethoxyphenyl)-1,3,2-dioxaborolane (35; 0.46 mmol, 100 mg) and DMF (5 mL).
Pd(PPh3)4 (0.046 mmol, 53 mg) was added in an oxygen- free atmosphere. The solution was
stirred at 80 °C for 6 h and further 24 h at RT until total consumption of the starting material.
The reaction was quenched with saturated aqueous NH4Cl (20 mL) and the layers were
separated. The aqueous phase was extracted with EtOAc (3 x 20 mL), the combined organic
layers were washed with brine (10 mL) and dried over Na2SO4, and the solvent was removed
under reduced pressure. The residue was purified by flash chromatography (hexane/EtOAc
19:1) to afford 36 (67.4 mg, 67%) as a yellow oil. 1H NMR (400 MHz, CDCl3) δ [ppm] 6.66
(s, 2H), 6.13-6.08 (m, 1H), 3.88 (s, 6H), 3.84 (s, 3H), 2.73-2.66 (m, 2H), 2.57-2.49 (m, 2H),
13
2.08-1.97 (m, 2H); C NMR (100 MHz, CDCl3) δ [ppm] 153.2, 142.5, 133.0, 126.1, 103.0,
61.0, 56.3, 33.6, 33.5, 23.6; ESI HRMS: [M+Na]+ calcd.: C14H18O3 234.1256, found
234.1246; IR  2939, 2838, 1680, 1586, 1497, 1478, 1468, 1415, 1298, 1255, 1112, 1006 cm-
1
.
4.1.11. 2-(3,4,5-Trimethoxyphenyl)cyclopentanol (37). To a solution of 36 (2.72 mmol, 637
.
mg) in THF (2 mL) was added BH3 THF (4.89 mmol, 1M in THF, 6.3 mL) dropwise at 0 °C.
After 30 min the cooling bath was removed, and the solution was allowed to warm to RT
over a period of 6 h. H2O2 (5.44 mmol, 4.6 mL) and NaOH (3M; 2.72 mmol, 0.9 mL) were
added in quick succession and stirring was continued for 15 h at RT. The solution was
quenched with Na2S2O3 (5 mL) and diluted with water (10 mL). The aqueous and the organic
phases were separated. The aqueous layer was extracted with EtOAc (3 x 10 mL), the
27

JOURNAL PRE-PROOF
combined organic layer was washed with saturated aqueous NH4Cl (10 mL) and brine (10
mL), was dried over Na2SO4 and concentrated in vacuo. The residue was purified by flash
1
chromatography (toluene/EtOAc 5:1) to afford 37 (408 mg, 60%) as a yellow oil. H NMR
(400 MHz, CDCl3) δ [ppm] 6.48 (s, 2H), 4.14 (q, J = 8.18 Hz, 1H), 3.86 (s, 6H), 3.82 (s, 3H),
13
2.86-2.76 (m, 1H), 2.19-2.07 (m, 2H), 1.93-1.63 (m, 5H); C NMR (100 MHz, CDCl3) δ
[ppm] 153.5, 139.0, 136.8, 104.4, 80.5, 61.0, 56.3, 55.0, 34.0, 31.9, 21.7; ESI HRMS:
[M+Na]+ calcd.: C14H20O4 252.1362, found 252.1359; IR  3435, 2940, 2839, 1589, 1509,
1458, 1240, 1125 cm-1.
4.1.12. 2-(3,4,5-Trimethoxyphenyl)cyclopentanone (38). To a solution of 2-(3,4,5-
trimethoxyphenyl)cyclopentanol 37 (0.19 mmol, 50 mg) in DCM (1 mL), DMSO (2.38
.
mmol, 0.17 mL), TEA (1.19 mmol, 0.16 mL) and SO3 py (0.59 mmol, 94 mg) were added at
0 °C. After 30 min the cooling bath was removed, and the resulting mixture was stirred until
total consumption of the starting material (8 h). The solution was quenched with water (5
mL) and the aqueous phase was extracted with EtOAc (3 x 3 mL). The combined organic
layers were washed with brine (2 mL), dried over Na2SO4 and concentrated in vacuo. Flash
column chromatography (hexane/EtOAc 5:1) gave 38 (49 mg, 99%) as a yellow oil. 1H NMR
(400 MHz, CDCl3) δ [ppm] 6.40 (s, 2H), 3.85 (s, 6H), 3.82 (s, 3H), 3.30 (m, 1H), 2.56-2.43
(m, 2H), 2.37-2.25 (m, 1H), 2.22-2.04 (m, 2H), 2.00-1.86 (m, 1H); 13C NMR (100 MHz,
CDCl3) δ [ppm] 218.0, 153.5, 137.2, 134.0, 105.4, 61.0, 56.2, 55.6, 38.5, 32.0, 20.9; ESI
HRMS: [M+Na]+ calcd.: C14H18O4 250.1205, found 250.1208; IR  3597, 2941, 2842, 1732,
1676, 1587, 1457, 1415, 1333, 1126.
4.1.13. 2-(3,4,5-Trimethoxyphenyl)cyclopentyltrifluoromethanesulfonate (39). To a solution
of 38 (1.13 mmol, 283 mg) in THF (1.5 mL), TEA (20 mmol, 2.8 mL) was added dropwise.
Subsequently, a solution of N-(5-chloro-2-pyridyl)bis(trifluoromethanesulfonimide) (1.7
mmol, 660 mg) in THF (1.5 mL) was added. The mixture was stirred at 60 °C until total
28

JOURNAL PRE-PROOF
consumption of the starting material (22 h). The reaction was quenched with saturated
aqueous NH4Cl (10 mL) and diluted with water (10 mL) and the aqueous phase was extracted
with CH2Cl2 (3 x 10 mL). The combined organic layers were washed with brine (5 mL) and
dried over Na2SO4. Flash column chromatography (hexane/EtOAc 9:1) gave 39 (112 mg,
26%) as a yellow oil. 1H NMR (400 MHz, CDCl3) δ [ppm] 6.74 (s, 2H), 3.87 (s, 3H), 3.86 (s,
6H), 2.88-2.77 (m, 4H), 2.14-2.03 (m, 2H); 13C NMR (100 MHz, CDCl3) δ [ppm] 153.0,
142.6, 138.2, 129.0, 123.0, 104.9, 60.9, 56.2, 32.2, 31.9, 19.2; ESI HRMS: [M+Na]+ calcd.:
C15H17F3O6S 382.0698, found 382.0703; IR  2943, 2842, 2349, 1654, 1646, 1582, 1541,
1418, 1346, 1246.
4.1.14. 2-Methoxy-5-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenol (40). A flask was
charged with 5-bromo-2-methoxyphenol 24 (3.94 mmol, 799 mg), bis(pinacolato)diboron
(B2pin2; 4.3 mmol, 1.1 g), dry KOAc (12 mmol, 1.16 g) and DMSO (16 mL). Subsequently,
1,1'-bis(diphenylphosphino)ferrocene (dppf; 0.19 mmol, 109 mg) and PdCl2 (0.19 mmol, 51
mg) were added. The reaction mixture was stirred at 80 °C until total consumption of the
starting material (12 h). The reaction was then quenched with saturated aqueous NH4Cl (10
mL) and diluted with water (30 mL). The aqueous phase was extracted with EtOAc (3 x 10
mL). The combined organic layers were washed with brine (5 mL) and dried over Na2SO4.
The solvent was removed under reduced pressure und the residue was purified by flash
column chromatography (hexane/EtOAc 7:1) to afford 40 (0.59 g, 60%) as white crystals. 1H
NMR (400 MHz, CDCl3) δ [ppm] 7.35 (bs, 1H), 7.34 (dd, J = 7.81, 1.54 Hz, 1H), 6.85 (d, J =
7.81 Hz, 1H), 5.49 (s, 1H), 3.90 (s, 3H), 1.33 (s, 12H); 13C NMR (100 MHz, CDCl3) δ [ppm]
149.3, 145.2, 127.7, 121.9, 120.5, 110.2, 83.8, 83.6, 55.9, 25.2, 25.0; EI HRMS: [M+Na]+
calcd: C13H19BO4 250.1376, found 250.1372; IR  3406, 2978, 2936, 1614, 1439, 1354,
1294, 1267, 1216, 1168, 1127, 1025.
29