Synthesis of nοvel artemisinin dimers with polyamine linkers and evaluation of their potential as anticancer agents

Article abstract of DOI:10.1016/j.bmc.2017.05.018

The natural product artemisinin and derivatives thereof are currently considered as the drugs of choice for the treatment of malaria. At the same time, a significant number of such drugs have also shown interesting anticancer activity. In the context of the present research work, artemisinin was structurally modified and anchored to naturally occurring polyamines to afford new artemisinin dimeric conjugates whose potential anticancer activity was evaluated. All artemisinin conjugates tested were more effective than artemisinin itself in decreasing the number of MCF7 breast cancer cells. The effect required conjugation and was not due to the artemisinin analogue or the polyamine, alone or in combination. To elucidate potential mechanism of action, we used the most effective conjugates 6, 7, 9 and 12 and found that they decreased expression and secretion of the angiogenic growth factor pleiotrophin by the cancer cells themselves, and inhibited angiogenesis in vivo and endothelial cell growth in vitro. These data suggest that the new artemisinin dimers are good candidates for the development of effective anticancer agents.

Full text of DOI:10.1016/j.bmc.2017.05.018

Accepted Manuscript
Synthesis of Νοvel Artemisinin Dimers with Polyamine Linkers and Evaluation
of Their Potential as Anticancer Agents
George E. Magoulas, Tzoanna Tsigkou, Lina Skondra, Margarita Lamprou,
Panagiota Tsoukala, Vassiliki Kokkinogouli, Evangelia Pantazaka, Dionissios
Papaioannou, Constantinos M. Athanassopoulos, Evangelia Papadimitriou
PII:
S0968-0896(17)30111-6
DOI:
http://dx.doi.org/10.1016/j.bmc.2017.05.018
BMC 13736
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
18 January 2017
25 April 2017
8 May 2017
Please cite this article as: Magoulas, G.E., Tsigkou, T., Skondra, L., Lamprou, M., Tsoukala, P., Kokkinogouli, V.,
Pantazaka, E., Papaioannou, D., Athanassopoulos, C.M., Papadimitriou, E., Synthesis of Νοvel Artemisinin Dimers
with Polyamine Linkers and Evaluation of Their Potential as Anticancer Agents, Bioorganic & Medicinal
Chemistry (2017), doi: http://dx.doi.org/10.1016/j.bmc.2017.05.018
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. 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.

1
SYNTHESIS OF ΝΟVEL ARTEMISININ DIMERS WITH POLYAMINE LINKERS AND
EVALUATION OF THEIR POTENTIAL AS ANTICANCER AGENTS
1
,┴,#
2,#
2
2
George E. Magoulas , Tzoanna Tsigkou , Lina Skondra , Margarita Lamprou , Panagiota
1
1
2
1
Tsoukala , Vassiliki Kokkinogouli , Evangelia Pantazaka , Dionissios Papaioannou ,
1
,
2,
Constantinos M. Athanassopoulos *, Evangelia Papadimitriou **
1
Laboratory of Synthetic Organic Chemistry, Department of Chemistry, University of Patras,
GR-26504, Patras, Greece
2
Laboratory of Molecular Pharmacology, Department of Pharmacy, University of Patras, GR-
2
┴
6504, Patras, Greece
Current address: National Hellenic Research Foundation, Institute of Biology, Medicinal
Chemistry and Biotechnology, 48 Vassileos Constantinou Avenue,
GR-11635 Athens, Greece
#
They have equally contributed to this work
*
*
Corresponding author. Tel: +302610 997909, e-mail: kath@chemistry.upatras.gr
* Corresponding author. Tel: +30 2610 962336, e-mail: epapad@upatras.gr

2
Abstract
The natural product artemisinin and derivatives thereof are currently considered as the drugs of
choice for the treatment of malaria. At the same time, a significant number of such drugs have
also shown interesting anticancer activity. In the context of the present research work, artemisinin
was structurally modified and anchored to naturally occurring polyamines to afford new
artemisinin dimeric conjugates whose potential anticancer activity was evaluated. All artemisinin
conjugates tested were more effective than artemisinin itself in decreasing the number of MCF7
breast cancer cells. The effect required conjugation and was not due to the artemisinin analogue
or the polyamine, alone or in combination. To elucidate potential mechanism of action, we used
the most effective conjugates 6, 7, 9 and 12 and found that they decreased expression and
secretion of the angiogenic growth factor pleiotrophin by the cancer cells themselves, and
inhibited angiogenesis in vivo and endothelial cell growth in vitro. These data suggest that the
new artemisinin dimers are good candidates for the development of effective anticancer agents.
Keywords: angiogenesis; artemisinin; breast cancer; conjugates; dimers; endothelial cells;
polyamines

3
1
. Introduction
Artemisinin (ART, 1), a natural product isolated from the Chinese plant Artemisia Annua, and its
semisynthetic derivatives dihydroartemisinin (DHA, 1a), artemether (1b) and artenusate (1c) are
1
-4
a group of drugs that are standard treatment worldwide for P. falciparum malaria. Chemically,
ART is a sesquiterpene lactone containing an unusual peroxide bridge, whose functionality is
5
-7
believed to be responsible for the drug's mechanism of action. Besides antimalarial activity,
recent studies report that artemisinins are also efficacious for improving the immune system and
8
-11
for curing other diseases, including cancers.
Their reported anticancer activity includes strong
1
2-14
7,15,16
selectivity,
reversal of multidrug resistance
and sensitization of cancer cells to
7
,17-19
20,21
radiation
and chemotherapy,
and attracts extensive attention. Many ART derivatives
have been synthesized as anticancer agents and studied in vitro and/or in vivo in breast, prostate,
ovarian, head and neck, lung, gastric, hepatocellular, pancreatic and kidney cancers,
7
-11
osteosarcomas, glioblastomas and leukemias.
Although production of reactive oxygen species
(ROS) is believed to play a role in the anticancer activity of artemisinins, other mechanisms that
are not necessarily connected to ROS production have been also suggested, such as apoptosis or
autophagy induction, cell cycle arrest and reduced cell proliferation, alteration of tumour-
7
,11
associated gene expression, and inhibition of tumour angiogenesis.
Pharmacological shortcomings of ART and derivatives include low water/oil solubility, poor
bioavailability and a short half-life in vivo. An emerging problem with treatments containing
ART and ART derivatives (ART combination therapies, ACTs) is the continuous increase of
resistance to ART and ACTs. To overcome these problems, semisynthetic and fully synthetic
ART-like compounds have been developed and tested. Moreover, there are studies showing that
dimeric and trimeric ART derivatives have much higher antitumor activity than their monomeric

4
counterparts, with the nature of the linker potentially playing an important role in the anticancer
1
1,22,23
activity.
Polyamines (PAs) are ubiquitous organic polycationic molecules present in all living organisms,
being essential in the regulation of cell proliferation and differentiation. PA transport systems
control cellular import and export of PAs and have been exploited as potential targets for
therapeutic intervention in cancers. Many cancer cells, such as neuroblastoma, melanoma, human
lymphocytic leukemic, colonic and lung cancer cell lines, exhibit elevated PA import activity,
24,25
probably due to their enhanced need for the growth supporting effect of PAs.
This property
has been exploited in many published papers on PA conjugates with cytotoxic drugs, which may
2
4,26
thus possess improved efficacy and cell selectivity.
Such a conjugate of the PA spermine
2
7
28
(
spm) with epipodophyllotoxin has proved highly effective in human breast and ovarian
2
7
cancer xenografts, in leukemia cells, in pediatric neuroblastoma and high grade glioma cell
2
9
30
lines, in head and neck squamous cell carcinoma cell lines and in non-small cell lung cancer
3
1
32
cells, and is being evaluated in phase I trials for leukemias. Naphthalimide coupled with PAs
3
3-34
exhibits antineoplastic activity and excellent selectivity for transformed versus normal cells.
Other examples of PA conjugates for anticancer lead discovery include dimeric quinoline,
35
36
37
cinnoline and phthalimide moieties,
all-trans retinoic acid,
chloramphenicol
and
3
8
minoxidil. Recently, conjugates of ART with the natural PA spermidine (spd) have been
3
9
synthesized and their antimalarial and anticancer activity was evaluated.
Prompted by the significant anticancer activity of both ART and PA analogs and conjugates, we
decided to further explore the potentiality of ART-PA conjugates. We, therefore, describe herein
the chemical modification of ART at positions 10 and 11 in order to bear suitable linkers, the
application of the thus modified ART (compounds 2-4) to the synthesis of a series of symmetric

5
ART-PA conjugates (compounds 5-13, Fig. 1), and finally the evaluation of the latter as potential
anticancer agents. The natural PAs putrescine (put), spd and spm were employed as the PA
counterpart of the conjugates. The modified ART moiety was attached to the primary amino
functions through a urethane functionality, which is hydrolytically more labile than the
corresponding amide bond, and the secondary amino functions were masked with the also
hydrolytically relatively labile methoxycarbonyl moiety to increase lipophilicity and protect
4
0
conjugates of the 8/11A type (see Fig. 1) from decomposition. Finally, the diethylene glycol
linker was employed to increase water solubility compared to the hydroxypropyl linker. With
these new conjugates, we wished to explore the role of (a) the length of the PA chain (compare
for example compounds 5, 8 and 11), (b) the modified ART nucleus (compare for example
compounds 5-7) and (c) the nature of the linker and its position (10 or 11) on the ART nucleus
(compare for example compounds 5 and 6, and 5 and 7, respectively).

6
O
O
O
O
O
O
O
O
O
2
1
1
11
O
O 10
O 10
OH
O
O
X
O
1
1
1
a: X = OH
b: X = OMe (10S)
c: X = OCO(CH ) CO H (10S)
1
A
2
2
2
O
O
O
O
O
H
R
N
O
O
O
N
R
3
O
4
H
O
O
HO
N
O
5
: R = A
O
OH
6: R = B
7: R = C
B
C
O
O
O
O
H
H
N
R
N
O
R
N
O
O
N
N
R
O
N
N
H
R
H
O
O
O
O
O
O
8
9
: R = A
: R = B
1
1: R = A
12: R = B
3: R = C
1
0: R = C
1
Fig. 1. Structures of compounds encountered in the present work.
2
. Results and discussion
2
.1. Chemistry
Dimers 5-13 consist of a PA backbone bearing two ART derivatives at their primary amino
functions through a urethane bond. The parent compound, ART, was modified at positions 10
and 11. That way, three hydroxyl-functionalized ART derivatives (2-4) were synthesized as

7
described below, which upon activation and reaction with selectively protected PAs provided the
projected dimers with the desired urethane linkages.
2
.1.1. Activated 10-oxo-ART analogue 2
4
1
ART was subjected to NaBH -mediated reduction to afford DHA (1a) in quantitative yield. The
4
latter was etherified with diethylene glycol in the presence of catalytic amount of BF ∙Et O to
3
2
furnish alcohol 2, which was finally activated through its reaction with 4-nitrophenyl
chloroformate in the presence of Et N to give the expected mixed carbonate active ester 14 in
3
9
0% yield (Scheme 1).
o
Scheme 1. Reagents and conditions: (i) NaBH , MeOH, 0 C, 3 h, 100%; (ii) diethylene glycol,
4
o
o
BF ∙Et O, 0 to 25 C, 1.5 h, 74%; (iii) 4-nitrophenyl chloroformate, Et N, CH Cl , 25 C, 24 h,
3
2
3
2
2
9
0%.

8
2
.1.2. Activated 10-carba-ART analogue 3
DHA (1a) was initially reacted with benzoyl chloride in the presence of pyridine to afford the
corresponding C-10 benzoate (15) in quantitative yield. This, in turn, was treated with
allyltrimethylsilane, using anhydrous ZnCl as Lewis acid, in the presence of 4 Å molecular
2
4
2
sieves to give 10β-allyldeoxoartemisinin (16) in 92% yield. The latter was then subjected to a
hydroboration-oxidation sequence of reactions to furnish the primary alcohol 3 in 85% yield over
5
two steps. Finally, the required mixed carbonate active ester 17 was successfully obtained
through activation of alcohol 3 with 4-nitrophenyl chloroformate, as described previously, in
8
4% yield (Scheme 2).
o
Scheme 2. Reagents and conditions: (i) Benzoyl chloride, pyridine, CH Cl , 0 C, 16 h, 100%; (ii)
2
2
o
allyltrimethylsilane, ZnCl , 4 Å mol. sieves, DCE, 0 C, 3 h, 92%; (iii) (a) BH ∙SMe , THF, -
2
3
2
o
o
2
0 C, 2 h; (b) H O , Na CO , H O, 25 C, 30 min, 85%; (iv) 4-nitrophenyl chloroformate, Et N,
2 2 2 3 2 3
o
CH Cl , 25 C, 24 h, 84%.
2
2

9
2
.1.3. Activated 11-aza-ART analogue 4
The insertion of a nitrogen atom at position 11 of ART involves ring opening from a suitable
amine, serving as a nucleophile, and then an acid-catalysed lactam ring formation according to a
4
3,44
published procedure.
For that purpose, ART was reacted with 2-(2-aminoethoxy)ethanol and
the thus obtained intermediate was treated with SiO /20% H SO to provide the desired alcohol 4
2
2
4
in 50% yield. The latter was activated, similarly to alcohols 2 and 3, to afford the mixed
carbonate active ester 18 in 85% yield (Scheme 3).
o
Scheme 3. Reagents and conditions: (i) (a) 2-(2-aminoethoxy)ethanol, CHCl /MeOH, 0 to 25 C,
3
o
1
.5 h; (b) SiO /20% H SO , di-tert-butylphenol, CHCl , 0 to 25 C, 16 h, 50% (ii) 4-nitrophenyl
2
2
4
3
o
chloroformate, Et N, CH Cl , 25 C, 24 h, 85%.
3
2
2
2
.1.4. Selectively protected PAs
1
8
45
1
12
46
N ,N -ditritylspermidine (19) and N ,N -ditritylspermine (20) were orthogonally protected at
their secondary amino functions with the methoxycarbonyl group, upon treatment with methyl
chloroformate in the presence of Et N. From the thus obtained fully protected PA derivatives 21
3

1
0
and 22, respectively, the anticipated compounds 23 and 24 were obtained, in the form of their
corresponding bistrifluoroacetate salts, upon TFA-mediated deprotection of the former (Scheme
4
).
o
Scheme 4. Reagents and conditions: (i) methyl chloroformate, Et N, CH Cl , 0 to 25 C, 24 h,
3
2
2
o
9
8% (for 21) or 85% (for 22); (ii) 45% TFA, CF CH OH, CH Cl , 0 to 25 C, 1.5 h, 98% (for 23
3
2
2
2
and 24).
2
.1.5. Synthesis of ART-PA conjugates 5-13
Finally, the assembly of the projected ART-PA conjugates 5-13 was successfully accomplished
upon coupling of active esters 14, 17 and 18 with the PA put, and the spd and spm derivatives 23
i
and 24, respectively, in the presence of Pr NEt in yields ranging from 20 to 56% (Scheme 5).
2

1
1
i
o
Scheme 5. Reagents and conditions: (i) active ester 14 or 17 or 18, Pr NEt, CH Cl , 0 to 25 C,
2
2
2
overnight, 20-56% yield.
2
2
.2. Pharmacology
.2.1. Effect of ART and its conjugates on the number of MCF7 breast cancer cells
We first investigated the effect of different concentrations of ΑRT (1) and ART-PA conjugates 5-
3 on MCF7 cell number, 48 h after the addition of the tested agents into the cell culture medium.
1
As shown in Fig. 2 and Table 1, all conjugates decreased the number of MCF7 cells in a
concentration-dependent manner, with the maximum effect being observed at the concentration
of 10 μM. Higher concentrations could not be used because of solubility issues. ART itself had a
minor effect on the number of MCF7 cells. The IC values and the maximum effect (efficacy) of
5
0
the tested agents, as interpolated from graphical data, are shown in Table 1. All the tested
conjugates had similar IC values but based on their efficacy and the concentration-dependency,
50
the most interesting seem to be the conjugates 6, 7, 9 and 12.
These results suggest that conjugation of ΑRT with PAs leads to a significant enhancement of the
anticancer efficacy of the parent compound ART and that the most effective combinations of PA
chain, modified ART and linker are the longer spd or spm chains, containing additional amino

1
2
functions compared to the put chain, the 10-carba-ART or the 11-aza-ART nucleus and the
hydroxypropyl or the diethylene glycol linker attached at position 11 of 11-aza-ART.
3
3
2
2
1
1
5000
0000
5000
0000
5000
0000
35000
30000
25000
20000
15000
10000
5000
5
000
0
0.001 0.01
0.1
1
10
0
0.001 0.01
0.1
1
10
ART (μΜ, n=8)
Conjugate 5 (μΜ, n=3)
3
5000
0000
5000
0000
5000
0000
35000
30000
25000
20000
15000
10000
5000
3
2
2
1
1
*
*
*
*
**
*
*
*
*
**
5
000
0
0.001 0.01
0.1
1
10
0
0.001 0.01
0.1
1
10
Conjugate 7 (μΜ, n=4)
Conjugate 6 (μΜ, n=4)

1
3
3
3
2
2
1
1
5000
0000
5000
0000
5000
0000
35000
30000
25000
20000
15000
10000
*
**
*
**
5
000
5000
0
0.001 0.01
0.1
1
10
0
0.001 0.01
0.1
1
10
Conjugate 9 (μΜ, n=6)
Conjugate 10 (μΜ, n=3)
3
3
2
2
1
1
5000
35000
30000
25000
20000
15000
10000
0000
5000
0000
5000
0000
*
*
*
**
*
5000
5000
0
0.001 0.01
0.1
1
10
0
0.001 0.01
0.1
1
10
Conjugate 11 (μΜ, n=3)
Conjugate 12 (μΜ, n=4)
3
5000
0000
5000
0000
5000
0000
3
2
2
1
1
5000
0
0.001 0.01
0.1
1
10
Conjugate 13 (μΜ, n=3)
Fig. 2. Effect of ART and its conjugates on the number of MCF7 cells 48 h after addition of the
tested agents into the cell culture medium. Results are expressed as mean ± S.E.M. of the number

1
4
of cells in treated compared with untreated cells. Asterisks denote statistical significance from
untreated cells. *P<0.05, **P<0.01, ***P<0.001.
Table 1. IC values and efficacy (maximum percent decrease in the number of cells) of ART and
5
0
its conjugates in decreasing the number of MCF7 cells. IC50 values were calculated from
interpolations of the graphical data presented in Fig. 2.
IC (μΜ)
Efficacy (%)
14 ± 0.7
5
0
-4
ART
1.4x10 ± 0.0164
1 ± 0.097
Conjugate 5
Conjugate 6
Conjugate 7
Conjugate 8
Conjugate 9
Conjugate 10
Conjugate 11
Conjugate 12
Conjugate 13
50 ± 20.5
61 ± 16.35
52 ± 15.6
11.2 ± 13
76 ± 13.7
39 ± 11.3
64 ± 17.9*
70 ±18.9
1 ± 0.000006
0.999 ± 006
n.d.
1 ± 000018
0.999 ± 0.164
1 ± 0.046
1 ± 0.000006
1 ± 0.013
48 ± 11.0
n.d.: not determined; due to the very low efficacy, a concentration-dependent assay was not
performed for this analogue.
*
Although efficacy of this conjugate is high, there is no concentration-dependency of its effect;
therefore, it is not considered physiologically significant.
Interestingly, the mere combination of ART derivative 3 and spd, co-administered at similar
concentrations, did not mimic the effect of conjugate 9 on the number of MCF7 cells (Fig. 3),
suggesting that conjugation plays a decisive role in the observed efficacy of this compound. If
conjugate 9 were considered as ART dimer, our results parallel the data from other research
groups showing that dimeric ART is more effective than ART itself in various biological
2
3
systems.

1
5
3
3
2
2
1
1
5000
0000
5000
0000
5000
0000
*
**
5
000
0
untr
3
spd 3+spd
9
Fig. 3. MCF7 cells were treated with conjugate 9 or the unconjugated ART derivative 3, spd or
their combination (3+spd), all used at 10 μM. Results are expressed as mean ± S.E.M. of the
number of cells compared with the untreated cells (untr). Asterisks denote a statistically
significant difference from untreated cells. ***P<0.001.
2
.2.2. Efforts to elucidate the mechanism through which ART and its conjugates 6, 7, 9, and 12
affect the number MCF7 breast cancer cell
It is well known that the antimalarial action of ART and analogs thereof implicates the formation
of free radicals, which mediate eradication of the Plasmodium species and the same has been also
7
,11
suggested as a possible mechanism for its anticancer activity.
In the present study, we
investigated the effect of ART and its most interesting PA conjugates on ROS production by
MCF7 breast cancer cells. As shown in Supplementary Fig. S1, ART and conjugates 6 and 7, and
to a lesser extend conjugate 12, increased ROS production by MCF7 cells, while conjugate 9,
which is the most effective in inhibiting MCF7 cell growth, had no effect, suggesting that the
observed inhibitory effect of the tested conjugates on the number of MCF7 cells is unrelated to a

1
6
possible increased ROS production and thus, to oxidative stress caused by ART conjugates. This
7
,11
is in line with observations by other groups, adding to the notion that the mechanism of action
of ART and derivatives is multivariate and may be cell type-specific.
ART and derivatives have been previously reported to induce apoptosis of several types of cancer
1
1
cells. In the present study, we investigated whether the decrease in the number of MCF7 cells
caused by conjugates 6, 7, 9 and 12 was, at least partly, due to induction of apoptosis. As shown
in Supplementary Fig. S2, neither ART nor its conjugates had a significant effect on apoptosis or
necrosis of MCF7 cells, as estimated by the relative amount of the annexin V and propidium
iodide-stained cells in the population. These data suggest that the observed decrease in the
number of MCF7 cells may be due to inhibition of cancer cell proliferation and are in line with
previously published data showing that the anticancer effects of artesunate on MCF7 and MDA-
4
7,48
MB-231 breast cancer cells were exerted through cell cycle arrest or an autophagy pathway.
Since neither ROS production nor apoptosis or necrosis induction seems to be the mechanism,
through which the tested ART conjugates affect the growth of MCF7 breast cancer cells, we
investigated whether ART or its conjugates 6, 7, 9 and 12 had any effect on the expression and
secretion of the growth factor pleiotrophin (PTN). We chose PTN because it has been implicated
4
9
in cancer cell growth and has been previously shown to be expressed by MCF7 breast cancer
50
cells and act in an autocrine way to confer a growth advantage on MCF7 tumors As shown in
Fig. 4, all tested conjugates significantly decreased expression and secretion of PTN, while the
effect of ART was marginal and not statistically significant. Conjugate 9 seems to be the most
effective in decreasing PTN expression and secretion.

1
7
A
B
1
1
1
40
20
00
80
1
00
8
6
4
2
0
0
0
0
0
*
*
*
*
*
***
6
4
2
0
0
0
0
*
**
untr ART
6
7
9
12
untr ART
6
7
9
12
Fig. 4. Effect of ΑRT and its conjugates 6, 7, 9 and 12 on PTN expression and secretion by
MCF7 breast cancer cells. (A) Representative picture of Western blot analysis of PTN protein
levels in the conditioned media (secreted) or the cell lysates (cellular) of MCF7 cells following
treatment with the tested conjugates at the concentration of 10 μM for 24 h. (B) PTN protein
levels were quantified by densitometric analysis of the corresponding band in each lane. Results
in all cases are expressed as mean ± S.E.M. (n=3) of the percent change of the amounts of
secreted PTN protein in treated compared with the untreated cells (untr, set as default =100). The
content of β-actin in the corresponding cells at the time that the conditioned medium was
collected is shown as a measure of normalization for PTN secretion. Asterisks denote a
statistically significant difference from untreated cells. *P<0.05, **P<0.01, ***P<0.001.

1
8
2
.2.3. Effect of ART, DHA and ART conjugates 6, 9 and 12 on angiogenesis in vivo
Besides a direct effect on cancer cells, ART and DHA also inhibit angiogenesis, thus indirectly
7
inhibiting cancer growth. In the present work, we studied the effect of ART, DHA and its
conjugates that were more effective in inhibiting MCF7 cancer cells growth on angiogenesis in
vivo, using the chick embryo chorioallantoic membrane (CAM) assay. As shown in Fig. 5,
conjugates 9 and 12 showed a small but statistically significant anti-angiogenic effect in the
chicken embryo CAM model. The decrease in the total vessel length at the dose of 10 nmol/egg
reached 12 ± 3.7% for conjugate 9 and 22 ± 2.4% for conjugate 12. Conjugate 6 at the dose of 10
nmol/egg also caused a statistically significant 14.3 ± 2.6% decrease in the total CAM vessel
length (P<0.01), while neither ART not DHA have any significant effect (6.3 ± 7.9% and 2.8 ±
9
.3% decrease compared with the control respectively).

1
9
1
00
100
90
80
*
*
*
**
8
6
4
2
0
0
0
0
0
70
60
50
40
30
20
10
0
0
2.5
5
10
0
2.5
5
10
Conjugate 12 (nmol/egg)
Conjugate 9 (nmol/egg)
Fig. 5. Effect of ART conjugates 9 and 12 on angiogenesis in the chick embryo CAM. Different
amounts of the tested compounds were applied on the CAM and 48 h later, the total vessel length
was estimated using image analysis software. Results are expressed as mean ± S.E.M. of the
percent change of the total vessel length in treated compared with the untreated tissue (control)
from at least four independent experiments. Asterisks denote a statistically significant difference
from control. **P<0.01, ***P<0.001. The pictures are representative, showing the vessel network
of the chicken embryo chorioallantoic membrane after treatment with the tested conjugates (10
nmoles/egg).
2
.2.4. Effect of ART, DHA and ART conjugates 6, 7, 9 and 12 on the number of human
endothelial cells
Previous studies have shown that ART and DHA have a direct effect on endothelial cells,
5
1-53
decreasing their proliferation.
In the present work, we studied whether the tested ART
conjugates also had a direct effect on endothelial cells, by measuring their effect on the number
of human umbilical vein endothelial cells (HUVEC), as described above for MCF7 breast cancer
cells. As shown in Fig. 6, ART and DHA significantly decreased HUVEC proliferation in a
5
1-53
similar manner, in line with published data.
All tested conjugates also decreased the number
of human endothelial cells in a concentration-dependent manner, with the maximum effect being
observed at the concentration of 10 μM. Higher concentrations could not be used because of

2
0
solubility issues. The IC50 values and the efficacy of the tested agents, as interpolated from
graphical data, are shown in Table 2. All tested agents had similar efficacy and IC values, with
5
0
conjugate 9 being slightly more effective. It is interesting to note that although ART and DHA
significantly affected HUVEC proliferation, they had no effect on CAM angiogenesis, at least at
the concentrations used, while the ART-PA conjugates were effective in vivo, suggesting that
they are either more effective than the parent compounds or that they also affect other pathways
yet to be identified.
1
00
100
80
60
40
20
0
*
8
6
4
2
0
0
0
0
0
***
*
*
0
0.001 0.01 0.1
1
10
0
0.001 0.01 0.1
1
10
ART (μM, n=4)
DHA (μM, n=4)
1
00
100
80
60
40
20
0
*
*
*
8
6
4
2
0
0
0
0
0
**
0
0.001 0.01 0.1
1
10
0
0.001 0.01 0.1
1
10
Conjugate 6 (μM, n=4)
Conjugate 7 (μM, n=4)

2
1
1
00
100
80
60
40
20
0
8
6
4
2
0
0
0
0
0
*
*
*
0
0.001 0.01 0.1
1
10
0
0.001 0.01 0.1
Conjugate 12 (μM, n=3)
1
10
Conjugate 9 (μM, n=4)
Fig. 6. Effect of ART, DHA and conjugates 6, 7, 9 and 12 on the number of human endothelial
cells 48 h after addition of the tested agents into the cell culture medium. Results are expressed as
mean ± S.E.M. of the percent number of cells in treated compared with untreated cells (set as
default 100%). Asterisks denote statistical significance from untreated cells. *P<0.05, **P<0.01,
*
**P<0.001.
Table 2. IC values and efficacy (maximum percent decrease in the number of cells) of ART,
5
0
DHA and conjugates 6, 7, 9 and 12 in decreasing the number of human endothelial cells. IC50
values were calculated from interpolations of the graphical data presented in Fig. 6.
Conjugate
IC (μΜ)
Efficacy (%)
22 ± 1.3
28 ± 2.6
27± 2.5
50
ART
DHA
6
1.1. ± 0.0006
1.4 ± 0.0035
3.6 ± 0.0030
2.3 ± 0.00015
1.2± 0.0003
7.5± 0.0020
7
23± 2.3
9
35± 3.7
1
2
27± 2.5
3
. Conclusion

2
2
A series of new conjugates of the urethane type of ART with the natural PAs put, spd and spm,
have been readily and efficiently synthesized by employing ART derivatives incorporating
hydroxypropyl or diethylene glycol linkers. The biological evaluation of these compounds
revealed that: 1) ART-PA conjugates are more effective than ART alone in inhibiting breast
cancer cell growth; 2) Among all the tested agents, conjugates 6, 7, 9 and 12 were the most
effective in inhibiting human breast cancer cell growth in a dose dependent manner; 3) This
effect of the ART conjugates could not be correlated to ROS production or apoptosis/necrosis
induction; 4) ART conjugates 6, 7, 9 and 12 decreased expression and secretion of the angiogenic
growth factor PTN, inhibited angiogenesis in vivo and decreased endothelial cells growth in
vitro. As concerns their chemical structure characteristics, the most interesting ART conjugates
(9 and 12) appear to combine the longer PA chains of spd or spm with the 10-carba-ART
derivative and the hydroxypropyl spacer at position 10 of the ART nucleus.
4
. Experimental section
4
4
.1. Chemistry
.1.1. General chemistry methods
1
Melting points were determined with a Buchi SMP-20 apparatus and are uncorrected. H NMR
1
3
spectra were recorded at 400.13MHz and CNMR spectra at 100.62 MHz on a Bruker DPX
spectrometer. Tetramethylsilane (TMS) was used as internal standard. Chemical shifts are
reported in δ units, parts per million (ppm) downfield from TMS. Electro-spray ionization (ESI)
mass spectra were recorded on a Micromass-Platform LC spectrometer using MeOH as solvent.
Flash column chromatography (FCC) was performed on Merck silica gel 60 (230-400 mesh) and
TLC on 60 Merck 60 F254 films (0.2 mm) precoated on aluminium foil. Spots were visualized

2
3
with UV light at 254 nm and the ninhydrin or charring agent. All solvents were dried and/or
purified according to standard procedures prior to use. Anhydrous Na SO was used for drying
2
4
o
organic solvents and subsequently solvents were routinely removed at ca. 40 C under reduced
pressure (water aspirator). All reagents employed in the present work were purchased from
commercial suppliers and used without further purification. Reactions were run in flame-dried
glassware under an atmosphere of argon with the exception of those involving aqueous solutions.
4
.1.2. Synthesis of alcohol 2
To an ice-cold solution of DHA 1a (1.0 g, 3.52 mmol) and diethylene glycol (1.68 mL,17.6
mmol), in DCM (30 mL), catalytic amount of BF ∙Et O (0.2 mL) was added dropwise. The
3
2
reaction mixture was stirred at RT for 1.5 h. Then it was diluted with DCM and washed once
with a saturated NaHCO solution and brine. The organic phase was dried over Na SO and
3
2
4
evaporated to dryness. Alcohol 2 was obtained in pure form after FCC purification. Yield: 0.97 g
o
(
[
(
74%); White solid; m.p: 94-96 C; R (PhMe/EtOAc 1:1): 0.21;MS (ESI, 30eV): m/z 411.38
f
1
M+K], 395.45 [M+Na]; H NMR: (CDCl , δ) 5.40 (1H, s), 4.85 (1H, d, J = 3.2 Hz), 3.98-3.90
3
1H, m), 3.75-3.60 (7H, m), 2.67-2.60 (1H, m), 2.39 (1H, td, J =3.6 and 13.6 Hz), 2.04 (1H, dq, J
=
4.5 and 14.0 Hz), 2.08-1.22 (10H, m), 1.45 (3H, s), 0.96 (3H, d, J = 6 Hz), 0.91 (3H, d, J = 7.2
13
Hz) ppm. C-NMR: (CDCl , δ)104.4, 96.4, 87.8, 81.1, 72.2, 70.7, 68.0, 61.8, 52.5, 44.3, 37.5,
3
3
6.4, 34.7, 30.8, 26.0, 24.7, 24.1, 20.4, 13.2 ppm.
4
.1.3. Synthesis of alcohol 3
o
To a cold solution (-20 C) of 10β-allyldeoxoartemisinin 16 (1.54 g, 5.0 mmol) in THF (200 mL),
a 1 M solution of BH ∙SMe in THF (5.0 mL) was added dropwise. The reaction mixture was
3
2
stirred at RT for 2 h. Upon completion, the mixture was quenched with a saturated Na CO
2
3
solution (10 mL), a 30% H O solution (5.0 mL) was added and the resulting mixture was stirred
2
2
for further 30 min. Then, it was evaporated to dryness. The residue was diluted with DCM,
washed once with water and once with brine. The organic phase was dried over Na SO and
2
4

2
4
evaporated to dryness to afford pure compound 3 after FCC purification. Yield: 1.39 g (85%);
colorless oil; R (PhMe/EtOAc 7:3): 0.17;MS (ESI, 30eV): m/z675.45 [2M+Na], 349.55
f
1
[
M+Na], 327.57 [M+H]; H NMR: (CDCl , δ) 5.32 (1H, s), 4.25-4.21 (1H, m), 3.74-3.65 (2H,
3
m), 2.65 (1H, sextet, J = 7.2 Hz),2.32 (1H, td, J = 3.6 and 13.6 Hz), 2.05-2.00 (1H, m), 1.93-1.88
(1H, m), 1.82-1.22 (12H, m), 1.41 (3H, s), 0.96 (3H, d, J = 6.0 Hz), 0.87 (3H, d, J = 7.2 Hz) ppm.
1
3
C-NMR: (CDCl , δ) 103.3, 89.4, 81.3, 75.6, 62.9, 52.4, 44.4, 37.6, 36.7, 34.6, 31.3, 30.7, 26.6,
3
2
6.1, 25.0, 24.9, 20.3, 13.0 ppm.
4
.1.4. Synthesis of alcohol 4
To an ice-cold solution of 2-(2-aminoethoxy)ethanol (14.04 mL, 140.0 mmol) in CHCl /MeOH
3
(
40 mL, 3:7), a solution of ART (2.0 g, 7.10 mmol) in CHCl (10 mL) was added dropwise over
3
o
5
min. The reaction mixture was stirred at 0 C for 2 h, diluted with water and it was extracted
thrice with CHCl . Then, 2,4-bis-tert-butylphenol (130 mg), H SO (130 mL), silica gel (130 g)
3
2
4
were added to the organic phase and the resulting suspension was stirred at RT for 12 h. Upon
completion, the mixture was filtered under vacuo, silica gel was washed thrice with CHCl and
3
the filtrate washed once with water and once with brine. The organic phase was dried over
Na SO and evaporated to dryness to afford pure compound 4 after FCC purification. Yield: 1.32
2
4
g (50%); yellow oil; R (PhMe/EtOAc 1:1): 0.18;MS (ESI, 30eV): m/z 408.29 [M+K], 392.36
f
1
[
M+Na]; H NMR (CDCl , δ):9.35 (1H, s), 5.43 (1H, s), 4.03-3.86 (2H, m), 3.74-3.64 (4H, m),
3
3
1
1
6
.63-3.55 (2H, m), 3.48-3.37 (1H, m), 2.73-2.62 (1H, m), 2.59-2.47 (1H, m), 2.46-2.31 (2H, m),
.90-1.69 (2H, m), 1.66-1.58 (1H, m), 1.38 (3H, s), 1.33-1.22 (1H, m), 1.16 (3H, d, J = 7.2 Hz),
1
3
.07-0.94 (3H, m), 0.97 (3H, d, J = 5.4 Hz); C NMR (CDCl , δ): 172.1, 104.9, 80.1, 79.0, 72.1,
3
9.0, 68.3, 51.3, 45.7, 40.6, 37.1, 36.6, 34.5, 33.1, 25.3, 25.1, 22.7, 20.1, 12.9 ppm.
4
.1.5. General procedure for the synthesis of activated compounds 14, 17, 18
To a solution of 2, 3 or 4 (2.5 mmol) and p-nitrophenyl chloroformate (0.6 g, 3.0 mmol,) in DCM
18 mL), Et N (0.7 mL, 5.0 mmol) was added dropwise. The reaction mixture was stirred at RT
(
3
overnight and then diluted with DCM. The organic phase was washed once with an ice-cold 5%

2
5
aq. solution citric acid and once with brine, dried over Na SO and evaporated to dryness. The
2
4
residues were subjected to FCC to afford pure compounds 14, 17 and 18 respectively.
Compound 14: Yield: 1.21 g (90%); yellow oil; R (PhMe/EtOAc 9:1): 0.15;MS (ESI, 30eV): m/z
f
1
5
9
3
76.33 [M+K], 560.27 [M+Na]; H NMR (CDCl , δ):8.22 (2H, d, J = 9.2 Hz), 7.32 (2H, d, J =
3
.2 Hz), 5.38 (1H, s),4.79 (1H, d, J = 3.6 Hz), 4.36 (2H, t, J = 4.4 Hz), 3.78-3.65 (2H, m), 3.64-
.55 (4H, m), 2.61-2.53 (1H, m), 2.35-2.24 (1H, m), 1.99-1.91 (1H, m), 1.84-1.76 (1H, m), 1.75
(1H, d, J = 3.6 Hz), 1.72-1.64 (1H, m), 1.59-1.51 (1H, m), 1.49-1.34 (2H, m), 1.37 (3H, s), 0.90-
1
3
0
1
3
.78 (2H, m), 0.87 (3H, d, J = 5.6 Hz), 0.85 (3H, d, J = 7.6 Hz); C NMR (CDCl , δ): 155.5,
3
52.5, 145.4, 125.3 (2C), 121.8 (2C), 104.1, 102.2, 87.9, 81.1, 70.7, 68.4, 68.3, 67.3, 52.6, 44.5,
7.5, 36.4, 34.7, 30.9, 26.2, 24.7, 24.4, 20.4, 13.0 ppm.
Compound 17: Yield: 1.03 g (84%); yellow oil; R (PhMe/EtOAc 9:1): 0.12;MS (ESI, 30eV): m/z
f
1
5
9
30.16 [M+K], 514.29 [M+Na]; H NMR (CDCl , δ): 8.21 (2H, d, J = 9.2 Hz), 7.32 (2H, d, J =
3
.2 Hz), 5.25 (1H, s), 4.36-4.24 (2H, m), 4.22-4.15 (1H, m), 2.64-2.54 (1H, m), 2.32-2.24 (1H,
m), 2.06-1.93 (2H, m), 1.90-1.82 (1H, m), 1.81-1.70 (2H, m), 1.65-1.54 (3H, m), 1.54-1.44 (2H,
m), 1.41-1.28 (2H, m), 1.36 (3H, s), 1.23-1.17 (2H, m), 0.92-0.88 (1H, m), 0.90 (3H, d, J = 6.0
1
3
Hz), 0.82 (3H, d, J = 7.6 Hz); C NMR (CDCl , δ):155.6, 152.5, 145.3, 125.3 (2C), 121.8 (2C),
3
1
1
03.1, 89.4, 81.1, 74.2, 69.4, 52.2, 44.1, 37.5, 36.6, 34.4, 30.4, 26.6, 26.0, 25.8, 24.9, 24.7, 20.2,
2.8 ppm.
Compound 18: Yield: 1.14 g (85%); yellow oil; R (PhMe/EtOAc 2:8): 0.21;MS (ESI, 30eV): m/z
f
1
5
7
3
1
1
1
2
73.24 [M+K], 557.18 [M+Na], 535.20 [M+Na]; H NMR (CDCl , δ):8.27 (2H, d, J = 8.8 Hz),
3
.38 (2H, d, J = 8.8 Hz), 5.44 (1H, s), 4.43-4.36 (2H, m), 3.84-3.79 (2H, m), 3.78-3.73 (2H, m),
.72-3.66 (2H, m), 3.33-3.25 (1H, m), 2.46-2.36 (1H, m), 2.04-1.94 (2H, m), 1.83-1.76 (1H, m),
.75-1.69 (1H, m), 1.67-1.62 (1H, m), 1.38 (3H, s), 1.36-1.27 (2H, m), 1.14 (3H, d, J = 7.2 Hz),
1
3
.07-0.94 (3H, m), 0.96 (3H, d, J = 5.4 Hz); C NMR (CDCl , δ):171.9, 155.6, 152.6, 145.5,
3
25.3 (2C), 121.8 (2C), 104.8, 80.2, 79.3, 69.3, 68.4, 68.2, 51.4, 45.8, 40.8, 37.5, 36.7, 33.8, 33.2,
5.5, 25.1, 22.7, 19.9, 12.9 ppm.
4
.1.6. General procedure for the synthesis of fully protected polyamines 21 and 22

2
6
1
8
1
12
To an ice-cold solution of N ,N -ditritylspermidine (19) or N ,N -ditritylspermine (20) (2.0
mmol) and Et N (0.36 mL, 2.6 mmol) in DCM (8.0 mL) methyl chloroformate (1.1 eq per amino
3
o
function) was added. The resulting mixture was left at 0 C for 10 min and for further 1 h at RT.
Then, it was diluted with DCM, washed once with an aqueous solution of 5% NaHCO , once
3
with brine, dried over Na SO and evaporated to dryness. Compounds 21 and 22 were obtained
2
4
pure after FCC purification.
o
Compound 21: Yield: 1.69 g (98%);white solid; m.p. 100-103 C; R (CHCl /MeOH 95:5):
f
3
+
1
0
7
.50;MS (ESI, 30eV): m/z 710.41 [M+Na], 688.41 [M+H], 243.14 [Trt ]; H NMR (CDCl , δ):
3
.40-7.34, 7.21-7.14 and 7.12-7.06 (30H, three m), 3.51 (3H, s), 3.19 (2H, unresolved t), 3.01
1
3
(
2H, unresolved t), 2.03 (4H, q, J = 7.2 Hz), 1.60 (2H, m), 1.42 (4H, m); C NMR (CDCl ,
3
δ):156.8, 146.2 (6C), 128.6 (12C), 127.7 (12C), 126.3 (6C), 70.8 (2C), 52.3, 46.7, 45.4, 43.3,
0.8, 28.0, 26.9, 26.3 ppm.
4
o
Compound 22: Yield: 1.36 g (85%); white solid; m.p. 105-108 C; R (CHCl /MeOH 9:1): 0.18;
f
3
-1
+ 1
IR (KBr, cm ): 2936, 1708, 1218, 704; MS (ESI, 30eV): m/z 804.37 [M+H], 243.14 [Trt ]; H
NMR (CDCl , δ):7.50-7.42, 7.31-7.22 and 7.21-7.13 (30H, three m), 3.60 (6H, s), 3.35-3.20 (4H,
3
1
3
m), 3.19-3.05 (4H, m), 2.14-2.04 (4H, m), 1.75-1.64 (4H, m), 1.49-1.36 (4H, m); C NMR
CDCl , δ):156.7 (2C), 146.2 (6C), 128.6 (12C), 127.8 (12C), 126.2 (6C), 70.9 (2C), 52.4 (2C),
(
3
4
6.8 (2C), 45.0 (2C), 40.9 (2C), 29.9 (2C), 25.4 (2C) ppm.
4
.1.7. General procedure for the selective deprotection of compounds 21 and 22
To a round-bottom flask containing compound 21 or 22 (2.0 mmol) an ice-cold solution of 45%
TFA in anhydrous DCM (6.0 mL) and CF CH OH (0.29 mL, 4.0 mmol) were added. The
3
2
o
resulting mixture was stirred at 0 C for 20 min and at RT for further 1 h. Then, it was evaporated
to dryness. A mixture of Et O and hexane (1:2) was added to the residue and it was refrigerated
2
overnight. The supernatant liquid was decanted to leave pure compounds 23 or 24, respectively,
as oily residues.
Compound 23: Yield: 0.85 g (98%);colorless oil; R (CHCl /MeOH/NH 9:1:0.1): 0.14; MS (ESI,
f
3
3
3
0eV): m/z 227.39 [M+Na], 204.42 [M+H].

2
7
Compound 24: Yield: 1.07 g (98%);colorless oil; R (CHCl /MeOH/NH 9:1:0.1): 0.28; MS (ESI,
f
3
3
3
0eV): m/z 342.42 [M+Na], 319.44 [M+H].
4
.1.8. General procedure for the synthesis of ART-PA conjugates 5-13
To an ice-cold solution of put or 23 or 24 (1.0 mmol) in DCM (3.0 mL) activated carbonates 14,
i
1
7 or 18 (2.0 mmol) and Pr NEt (5.0 mmol, 0.87 mL) were added. The reaction mixture was
2
stirred at ambient temperature overnight. Upon completion of the reaction, the mixture was
diluted with DCM and the organic phase was washed once with 5% aqueous citric acid solution,
once with water, once with 5% NaHCO solution, once with brine, dried over Na SO and
3
2
4
evaporated to dryness. The projected compounds were obtained in pure form after FCC
purification.
For compound 5: Yield: 0.48 g (54%); colourless oil; R (PhMe/EtOAc 3:7): 0.20; MS (ESI,
f
1
3
3
2
0eV): m/z 923.25 [M+K], 907.20 [M+Na]; H NMR (CDCl , δ): 5.45 (2H, s), 4.83 (2H, d, J =
3
.6 Hz), 4.25-4.12 (4H, m), 3.96-3.88 (2H, m), 3.69-3.58 (10H, m), 3.18 (4H, unresolved q),
.66-2.57 (2H, m), 2.41-2.29 (4H, m), 2.06-1.98 (2H, m), 1.92-1.83 (3H, m), 1.80 (1H, d, J = 3.6
Hz), 1.78-1.70 (2H, m), 1.66-1.58 (2H, m), 1.55-1.50 (4H, m), 1.48-1.40 (4H, m), 1.42 (6H, s),
1
3
1
.34-1.18 (4H, m), 0.94 (6H, d, J = 7.2 Hz), 0.92-0.85 (2H, m), 0.90 (6H, d, J = 6.0 Hz); C
NMR (CDCl , δ):156.6 (2C), 104.2 (2C), 102.1 (2C), 87.9 (2C), 81.2 (2C), 70.5 (2C), 69.4 (2C),
3
6
2
7.3 (2C), 64.1 (2C), 52.6 (2C), 44.6 (2C), 40.6 (2C), 37.5 (2C), 36.4 (2C), 34.7 (2C), 30.9 (2C),
7.2 (2C), 26.2 (2C), 24.7 (2C), 24.4 (2C), 20.3 (2C), 13.0 (2C) ppm.
For compound 6: Yield: 0.25 g (32%); colourless oil; R (PhMe/EtOAc 1:1): 0.22;MS (ESI,
f
1
3
4
0eV): m/z 831.27 [M+K], 815.34 [M+Na], 793.17 [M+H]; H NMR (CDCl , δ):5.32 (2H, s),
3
.24-4.16 (2H, m), 4.15-4.08 (4H, unresolved t), 3.24-3.14 (4H, unresolved m), 2.73-2.61 (2H,
m), 2.35-2.29 (2H, m), 2.08-2.00 (2H, m), 1.98-1.89 (4H, m), 1.84-1.77 (2H, m), 1.71-1.59 (8H,
m), 1.56-1.52 (4H, m), 1.50-1.42 (4H, m), 1.42 (6H, s), 1.36-1.22 (8H, m), 1.02-0.91 (2H, m),
1
3
0
8
3
1
.97 (6H, d, J = 5.6 Hz), 0.87 (6H, d, J = 7.6 Hz); C NMR (CDCl , δ): 157.0 (2C), 103.2 (2C),
3
9.2 (2C), 81.2 (2C), 74.9 (2C), 65.0 (2C), 52.3 (2C), 44.3 (2C), 40.6 (2C), 37.5 (2C), 36.6 (2C),
4.4 (2C), 30.3 (2C), 27.3 (2C), 27.0 (2C), 26.1 (2C), 24.9 (2C), 24.7 (2C), 21.5 (2C), 20.2 (2C),
2.9 (2C) ppm.

2
8
For compound 7: Yield: 0.18 g (21%); colorless oil; R (PhMe/EtOAc 2:8): 0.14; MS (ESI,
f
1
3
4
2
1
0eV): m/z 917.31 [M+K], 901.37 [M+Na], 879.27 [M+H]; H NMR (CDCl , δ): 5.10 (2H, s),
3
.11-4.05 (4H, m), 3.93-3.80 (4H, m), 3.59-3.48 (8H, m), 3.15-3.06 (4H, m), 2.66-2.58 (2H, m),
.52-2.42 (2H, m), 2.35-2.27 (6H, m), 2.04-2.00 (6H, m), 1.80-1.68 (4H, m), 1.49-1.41 (6H, m),
1
3
.43 (6H, s), 1.15 (6H, d, J = 6.8 Hz), 1.08-0.92 (4H, m), 1.00 (6H, d, J = 6.0 Hz); C NMR
(
(
(
CDCl , δ):162.4 (2C), 156.8 (2C), 105.7 (2C), 80.1 (2C), 78.0 (2C), 70.8 (2C), 69.1 (2C), 67.9
3
2C), 41.3 (2C), 40.8 (2C), 37.3 (2C), 34.5 (2C), 33.9 (2C), 31.9 (2C), 29.9 (2C), 29.7 (2C), 27.1
2C), 26.2 (2C), 22.7 (2C), 20.5 (2C), 20.1 (2C), 16.9 (2C) ppm.
For compound 8: Yield: 0.38 g (38%); colorless oil; R (PhMe/EtOAc 1:9): 0.22; MS (ESI,
f
1
3
0eV): m/z 1038.35 [M+K], 1022.38 [M+Na]; H NMR (CDCl , δ):5.38 (2H, d, J = 3.6 Hz), 4.77
3
(
2H, d, J = 3.6 Hz), 4.20-4.15 (2H, m), 4.15-4.07 (4H, m), 3.90-3.82 (2H, m), 3.70-3.51 (12H,
m), 3.62 (3H, s), 3.18-3.07 (4H, m), 2.60-2.51 (2H, m), 2.38-2.24 (4H, m), 2.01-1.92 (4H, m),
.86-1.77 (4H, m), 1.74 (1H, d, J = 3.6 Hz), 1.72-1.67 (2H, m), 1.67-1.60 (4H, m), 1.59-1.50
5H, m), 1.44-1.34 (2H, m), 1.36 (6H, s), 1.23-1.09 (2H, m), 0.89 (6H, d, J = 7.2 Hz), 0.90-0.82
1
(
(
(
1
3
2H, m), 0.84 (6H, d, J = 6.0 Hz); C NMR (CDCl , δ): 156.5 (3C), 104.1 (2C), 102.1 (2C), 87.9
3
2C), 81.1 (2C), 70.5 (2C), 69.5 (2C), 67.3 (2C), 64.1 (2C), 52.6, 44.5 (2C), 44.1, 40.6, 37.5 (3C),
3
6.5 (2C), 34.7 (3C), 30.9 (3C), 29.7, 27.3, 26.2 (3C), 24.4 (2C), 21.4 (2C), 20.4 (3C), 13.0 (2C)
ppm.
For compound 9: Yield: 0.39 g (43%); colorless oil; R (PhMe/EtOAc 1:1): 0.14; MS (ESI,
f
1
3
4
2
1
0
0eV): m/z 946.38 [M+K], 930.27 [M+Na]; H NMR (CDCl , δ): 5.23 (2H, s), 4.16-4.10 (2H, m),
3
.06-3.97 (4H, m), 3.63 (3H, s), 3.29-3.18 (2H, m), 3.15-3.06 (6H, m), 2.63-2.53 (2H, m), 2.30-
.19 (2H, m), 2.00-1.91 (2H, m), 1.90-1.79 (4H, m), 1.75-1.68 (2H, m), 1.63-1.48 (10H, m),
.43-1.35 (4H, m), 1.33 (6H, s), 1.36-1.22 (12H, m), 0.93-0.81 (2H, m), 0.89 (6H, d, J = 5.6 Hz),
13
.79 (6H, d, J = 7.6 Hz); C NMR (CDCl , δ):162.9 (2C), 156.9,103.2 (2C), 89.2 (2C), 81.2
3
(
2C), 77.2 (2C), 64.9 (2C), 52.7, 52.3 (2C), 44.3 (2C), 40.5 (2C), 37.4 (2C), 36.6 (2C), 34.4 (2C),
1.6, 30.3 (2C), 29.7, 27.2, 27.1, 27.0, 26.1 (2C), 26.0, 25.9, 24.9 (2C), 24.7 (2C), 22.6 (2C), 20.2
2C), 14.1 (2C) ppm.
3
(
For compound 10: Yield: 0.26 g (26%); colorless oil; R (EtOAc): 0.12; MS (ESI, 30eV): m/z
f
1
1
031.96 [M+K], 1016.11 [M+Na], 994.34 [M+H]; H NMR (CDCl , δ):5.32 (2H, s), 4.10-4.05
3
(4H, m), 3.72-3.65 (4H, m), 3.63-3.54 (8H, m), 3.62 (3H, s), 3.26-3.17 (4H, m), 3.15-3.06 (6H,

2
9
m), 2.39-2.27 (4H, m), 2.00-1.88 (6H, m), 1.73-1.54 (10H, m), 1.37-1.25 (6H, m), 1.29 (6H, s),
1
3
1
1
4
2
.06 (6H, d, J = 7.2 Hz), 1.01-0.88 (4H, m), 0.92 (6H, d, J = 6.0 Hz); C NMR (CDCl , δ):171.9,
3
71.1, 162.7, 162.6, 155.4, 107.1, 104.7, 80.2, 79.3, 69.4, 69.3, 69.2, 69.1, 64.2, 64.1, 54.9, 45.7,
4.2, 43.2 (2C), 41.3 (2C), 40.8, 40.6, 39.7, 37.5, 37.3, 36.6 (2C), 31.1 (2C), 29.9, 29.7, 27.2,
6.7, 25.4 (2C), 25.1 (2C), 24.3 (2C), 21.3, 21.1, 20.5 (2C), 20.1 (2C), 14.2 (2C) ppm.
For compound 11: Yield: 0.41 g (37%); colorless oil; R (EtOAc): 0.14;MS (ESI, 30eV): m/z
f
1
1
4
3
153.23 [M+K], 1138.30 [M+Na]; H NMR (CDCl , δ): 5.43 (2H, s), 4.83 (2H, d, J = 3.6 Hz),
3
.22-4.14 (4H, m), 3.96-3.90 (2H, m), 3.68 (6H, s), 3.68-3.59 (12H, m), 3.33-3.19 (4H, m), 3.17-
.11 (8H, m), 2.65-2.57 (2H, m), 2.41-2.28 (2H, m), 2.07-1.98 (4H, m), 1.92-1.82 (3H, m), 1.80
(1H, d, J = 3.6 Hz), 1.78-1.58 (10H, m), 1.51-1.40 (2H, m), 1.42 (6H, s), 1.25-1.18 (6H, m), 0.94
1
3
(
6H, d, J = 6.0 Hz), 0.92-0.82 (2H, m), 0.90 (6H, d, J = 7.2 Hz); C NMR (CDCl , δ):156.5 (4C),
3
1
04.1 (2C), 102.1 (2C), 87.8 (2C), 81.3 (2C), 70.4 (2C), 69.6 (2C), 64.0 (2C), 61.7 (2C), 52.5
(
2C), 51.4 (2C), 44.5 (4C), 44.1 (2C), 37.4 (2C), 36.4 (2C), 34.7 (2C), 34.2 (2C), 30.9 (2C), 30.1
2C), 29.7 (2C), 26.1 (2C), 24.7 (2C), 24.4 (2C), 20.5 (2C), 12.9 (2C) ppm.
(
For compound 12: Yield: 0.20 g (20%); colorless oil; R (PhMe/EtOAc 3:7): 0.17;MS (ESI,
f
1
3
0eV): m/z 1061.53 [M+K], 1045.40 [M+Na]; H NMR (CDCl , δ):5.23 (2H, s), 4.15-4.08 (2H,
3
m), 4.06-3.99 (4H, m), 3.62 (6H, s), 3.27-3.20 (4H, m), 3.16-3.04 (8H, m), 2.63-2.54 (2H, m),
2
1
0
.27-2.20 (2H, m), 1.99-1.93 (2H, m), 1.90-1.80 (4H, m), 1.76-1.67 (4H, m), 1.67-1.50 (10H, m),
.45-1.32 (4H, m), 1.34 (6H, s), 1.27-1.20 (12H, m), 0.96-0.83 (2H, m), 0.89 (6H, d, J = 6.0 Hz),
1
3
.79 (6H, d, J = 7.6 Hz); C NMR (CDCl , δ):156.9 (4C), 103.2 (2C), 89.1 (2C), 81.1 (4C), 64.8
3
(2C), 52.7 (2C), 52.3 (2C), 46.5 (2C), 44.3 (2C), 44.1 (2C), 37.5 (2C), 36.6 (2C), 34.5 (2C), 30.3
(2C), 29.7 (2C), 28.2 (2C), 27.1 (2C), 26.1 (2C), 26.0 (2C), 24.9 (2C), 24.7 (2C), 21.5 (2C), 20.2
(2C), 12.9 (2C) ppm.
For compound 13: Yield: 0.62 g (56%); colorless oil; R (EtOAc): 0.11; MS (ESI, 30eV): m/z
f
1
1
147.31 [M+K], 1131.18 [M+Na], 1109.27 [M+H]; H NMR (CDCl , δ): 5.39 (2H, s), 4.16-4.09
3
(4H, m), 3.77-3.68 (4H, m), 3.66 (3H, s), 3.65-3.58 (8H, m), 3.31-3.20 (6H, m), 3.17-3.07 (7H,
m), 3.04-2.98 (1H, m), 2.43-2.33 (4H, m), 2.02-1.92 (4H, m), 1.78-1.57 (12H, m), 1.50-1.41 (9H,
1
3
m), 1.34 (6H, s), 1.11 (6H, d, J = 7.2 Hz), 1.00-0.89 (4H, m), 0.96 (6H, d, J = 6.0 Hz); C NMR
(
CDCl , δ):171.8 (2C), 157.0 (2C), 156.4 (2C), 104.7 (2C), 80.1 (2C), 79.3 (2C), 69.3 (2C), 69.1
3
(2C), 64.2 (2C), 51.3 (2C), 46.4 (2C), 45.8 (2C), 44.1 (2C), 40.9 (2C), 37.5 (2C), 36.6 (2C), 29.8