Factors influencing the cytotoxicity of α-methylene-γ-hydroxy esters against pancreatic cancer

Article abstract of DOI:10.1016/j.bmcl.2015.07.087

A systematic study to identify the factors influencing the cytotoxicity of α-methylene-γ-hydroxy esters against three pancreatic cancer cell lines (Panc-1, MIA-PaCa-2, and BxPC-3) has established that, in addition to Michael acceptor abilities, the possibility to lactonize to α-methylene-γ-butyrolactones is as important. The substitution pattern and the number of carbons between the hydroxy and ester moieties also influence the bio-activity.

Full text of DOI:10.1016/j.bmcl.2015.07.087

Accepted Manuscript
Factors influencing the cytotoxicity of α-methylene-γ-hydroxy esters against
pancreatic cancer
P. Veeraraghavan Ramachandran, Matthew A. Helppi, Arlie Lehmkuhler,
Jennifer M. Marchi, C. Max Schmidt, Michele T. Yip-Schneider
PII:
S0960-894X(15)00806-9
DOI:
http://dx.doi.org/10.1016/j.bmcl.2015.07.087
BMCL 22976
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
25 June 2015
24 July 2015
27 July 2015
Please cite this article as: Veeraraghavan Ramachandran, P., Helppi, M.A., Lehmkuhler, A., Marchi, J.M., Max
Schmidt, C., Yip-Schneider, M.T., Factors influencing the cytotoxicity of α-methylene-γ-hydroxy esters against
pancreatic cancer, Bioorganic & Medicinal Chemistry Letters (2015), doi: http://dx.doi.org/10.1016/j.bmcl.
2015.07.087
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.

Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com
Factors influencing the cytotoxicity of α-methylene-γ-hydroxy esters against
pancreatic cancer
∗
P. Veeraraghavan Ramachandrana, , Matthew A. Helppia, Arlie Lehmkuhlera, Jennifer M. Marchia, C. Max
Schmidtb,c, and Michele T. Yip-Schneiderb,c
aHerbert C. Brown Center for Borane Research, Department of Chemistry, Purdue University 560 Oval Drive, West Lafayette, IN 47907-2084, USA
bDepartment of Surgery, Indiana University School of Medicine, 980 West Walnut Street, Indianapolis, IN 46202, USA
cDepartment of Biochemistry and Molecular Biology, Indiana University School of Medicine, 980 West Walnut Street, Indianapolis, IN 46202, USA
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
A systematic study to identify the factors influencing the cytotoxicity of α-methylene-γ-hydroxy
esters against three pancreatic cancer cell lines (Panc-1, MIA-PaCa-2, and BxPC-3) has
established that, in addition to Michael acceptor abilities, the possibility to lactonize to α-
methylene-γ-butyrolactones is as important. The substitution pattern and the number of carbons
between the hydroxy and ester moieties also influence the bio-activity.
2009 Elsevier Ltd. All rights reserved.
Keywords:
pancreatic
pro-drug
lactonization
cytotoxicity
hydroxy esters
———
∗Corresponding author. e-mail: chandran@purdue.edu

Pancreatic adenocarcinoma is the fourth leading cause of cancer-related deaths in the U.S.1 The late detection of this deadly
disease results in only 10% of patients being candidates for surgery, the leading therapeutic approach. The low five year survival
rate (7.2%) has remained relatively unchanged over the past few decades despite 5-flourouracil, gemcitabine and oxaliplatin-based
regimens, all FDA-approved with therapeutic effects.2 Accordingly, new chemotherapeutic treatment options, natural and synthetic,
are actively being pursued and agents directed at single or multiple targets are being sought urgently.
α-Methylene-γ-butyrolactones (AMGBLs, Fig. 1. I), a key functionality found in a large number of natural products, exhibit a
wide range of biological activities including anticancer, anti-inflammatory, antibacterial, anticoagulant, antiviral, and antifungal
properties.3 Four decades ago, molecules bearing these moieties were investigated for their anticancer properties, which were
attributed to their Michael-acceptor capabilities.4 Concerns over the lack of selectivity of the biological Michael reactions and the
associated safety issues stalled further investigations with these molecules. The benefits of parthenolide (PT, Fig. 1, II), isolated
from feverfew (Tanacetum parthenium), as a folk medicine and its potential as an anti-cancer agent5 has contributed to a renewed
interest in AMGBLs and related compounds in medicinal chemistry.
We have been investigating synthetic AMGBLs as potential inhibitors of nuclear factor-kappaB (NF-κB)6 with the goal of
arresting the spread of pancreatic cancer.7 We had reported that substitution at the β- and γ-positions of AMGBLs with aryl groups is
important for increased cytotoxicity. During the course of our investigations, we had noticed that the precursor α-methylene-γ-
hydroxy esters [AMGHEs, Fig. 1, IV(b)] possessed activity similar to those of the corresponding lactones.7c,8
Figure 1. Potential Anti-cancer Agents
We became interested in understanding the origins of the cytotoxicity of AMGHEs. The ease of synthesis of the hydroxy esters
and their increased solubility provided the necessary impetus for our study. In addition to the substituents at the β- and γ-positions,
our attention also focused on the hydroxyl and ester moieties to delineate the link between the structure and activity. We suspected
that the γ-hydroxy ester might act as a pro-drug undergoing lactonization in vivo to the methylene lactones, similar to the conversion
of the aminolactones to the parent AMGBLs, eg. dimethylaminoparthenolide (DMAPT, Fig. 1, III).9 Our study involving synthesis
and bio-assays has revealed that, in addition to Michael acceptor abilities, the possibility to lactonize to α-methylene-γ-
butyrolactones is a necessary condition for AMGHEs to be cytotoxic. The details follow.
On the basis of the reported Michael acceptor ability of AMGBLs,4 our initial focus was on similar properties of AMGHEs. A
study was begun with methyl α-methylene-β,γ-diphenylbutanoate (1a), prepared using known protocols.10 When assayed against the
three selected pancreatic cell lines (Panc-1, MIA PaCa-2, and BxPC-3, in a triplicate growth assay11, 1a revealed cytotoxicity
similar to that of the corresponding AMGBL (cis-α-methylene-β,γ-diphenylbutyrolactone, 2a). This prompted us to examine Morita-
Baylis-Hillman (MBH) adducts [Fig. 1, IV(a)]12 against the same cells since they also possessed, in addition to the hydroxyl group,
a conjugated methylene and ester moieties, the critical units responsible for the Michael addition. Were this the only/key condition
necessary for cytotoxicity, the MBH adduct should be suppressing cell growth, similar to 1a. Our rationale was that if the MBH
adduct fail to show any activity, it could be attributed to the additional carbon in IV(b) (Fig. 1) that will constitute the backbone of
the γ-lactone. A mining of the literature data revealed that MBH adducts have been examined against several cancer cell lines,13
although not against pancreatic cancer, with IC50 values in the low µM range. Also, silylation of the hydroxyl group had no
deleterious effect on the activity. Similarly, we were interested in examining the bio-activity of α-methylene-β-propiolactones (Fig
1, V) and comparing them to the AMGBLs. To the best of our knowledge, α-methylene-β-propiolactones have never been tested
against cancer cells.
Accordingly, the MBH adduct 3a prepared (Scheme 1) from benzaldehyde and methyl acrylate was examined against the three
pancreatic cancer cell lines. There was no effect of this molecule on the cell growth even at 10 μM concentration.14 Substituting the
4-position of the phenyl ring in 3a with an electron-donating methoxy group (3b) or an electron-withdrawing cyano (3c) or nitro
(3d) group had no perceivable effect. However, a 4-trifluoromethyl group (3e) suppressed growth somewhat, particularly that of
the MIA PaCa cells. This prompted the preparation and testing of other fluorinated MBH adducts 3f-j [(4-F-Ph)-, (3-F-Ph)-, (2-F-
Ph)-, C6F5-, and (2-CF3-Ph)-] respectively. However, none of them showed any improvement on the activity (see the table in
supporting information for the summary of the results of MBH adducts).

Scheme 1. Synthesis of MBH Adducts and Derivatives
The MBH adducts were saponified with LiOH under mild conditions. However, the lactonization of the β-hydroxy acids was
successful only for phenyl substituents bearing the 4-CN or 4-NO2 groups.15 Both of these α-methylene-β-propiolactones (4c and 4d)
were ineffective in suppressing the growth of any of the three pancreatic cells at 10 μM concentration. The failure of the MBH
adducts and the corresponding β-lactones in inhibiting pancreatic cancer cell growth, despite mildly inhibiting several other cancer
cell lines, suggests that a Michael acceptor moiety alone is insufficient to suppress cell growth and other factors in the molecule
influence the cytotoxic activity of AMGHEs. The extra carbon in AMGHE appeared critical for cytotoxicity.
Our attention was now diverted to the structure-activity of AMGHEs and a library of substituted methyl 4-hydroxy-2-
methylenebutanoates were synthesized from methyl acrylate and two different aldehydes (R1 and R2 = H, aliphatic, aromatic, or a
tethered aromatic) according to published protocols (Scheme 2)10 The synthesized AMGHEs (1a-m), as well as their biological
activities at 10 μM concentrations are summarized in Table 1. The assays of PT and 2a were repeated (Table 1, entries 1 and 2) and
selected assays (entries 7, 9, 13, and 17) from our earlier work7a are included for comparison.
O
2
R
O
OH
o
OH
O
R CHO, In
2
HBr/HOAc
O
O
OMe
1
R
OMe
1
1
R
1
R
OMe
OMe
R
H
DABCO
32-91%
or PBr
3
THF/H O
2
Br
78-91%
43-95%
Scheme 2. Synthesis of substituted α-methylene-γ-hydroxyesters
The current study of the activity of AMGHEs revealed several parallels with that of our reported study on AMGBLs.7a
AMGHEs without substitution at the γ-position (1b-d) resulted in poor activity (entries 4-6). The cytotoxicity is poor if substituents
at the both β- and γ-poisitions are aliphatic groups (1f-g, entries 10-11). Among the disubstituted hydroxy esters, a phenyl group at
either the β- or γ-position revealed better activity, especially against MIA PaCa-2 and BxPC-3 (1a, 1h-l, entries 12, 14-16, 18).
Except for lactones 2d and 2h, the hydroxy esters are slightly better than the lactones at suppressing cell growth (1a/2a, 1d-e/2d-e,
1h/2h, and 1k/2k, entries 2-3, 6-9, 12-13, and 16-17). Lactones 2e and 2k (entries 9 and 17) are almost completely inactive while
the corresponding hydroxy esters, 1e and 1k (entries 8 and 16), show relatively good activity. Interestingly, an aliphatic tether
could be inserted with no significant loss of activity (1m, entry 19). Substituted aromatics were not pursued in this study with the
hydroxy esters since we expected their activity to be similar to those of the corresponding lactones.
The similarity in the activity of both the open and cyclic esters supported our assumption that AMGHEs might be undergoing
cyclization in vivo before acting on the cells. Suppressing the cyclization was then targeted to substantiate this hypothesis. The
literature had reported that protecting the hydroxyl group of the MBH adducts with a silyl group did not affect the bio-activity.13
The γ-hydroxyl group of the AMGHE 1a was protected as a methyl and TBS-ether (1a’ and 1a”, respectively). Methylation was
carried out in the presence of silver oxide16 and TBS-protection was achieved with excess TBSCl and imidazole. Surprisingly, the
cytotoxic activity of 1a was completely blocked in ethers 1a’ and 1a” (Table 2, entries 1 and 2), backing our hypothesis that
cyclization to AMGBL might be a necessary condition for bio-activity. Additional support was sought by modifying the ester
moiety. Rather than carrying out the synthesis with different acrylates to prepare the corresponding AMGHEs, we resorted to
saponification of the 1a”, followed by Steiglich esterification17 with the corresponding alcohols and removal of the silyl group.
Thus, the ethyl, benzyl, and phenyl esters (5-7) were prepared and examined. There is little effect of the ester moiety in suppressing
cell growth (Table 2, entries 3-5).

Table 1.
Cytotoxic activity of α-methylene-γ-hydroxy esters against Panc-1, MIA PaCa-2, and BxPC-3 cell lines.
% Cell Growth
Entry Structure
1
(#) Panc-1
MIA PaCa-2
11
BxPC-3
12
PT
22
2
3
2a
1a
12
14
9
9
3
2
120
164
102
31
124
98
29
14
14
106
147
77
32
14
32
24
29
81
16
16
126
86
60
40
1
4
5
1b
1c
6
1d
2da
1e
7
67
8
9
2ea
1f
135
110
145
91
136
122
120
49
41
38
27
23
95
8
10
11
12
13
14
1g
1h
2ha
1i
62
104
96
OH
CO Me
2
15
16
17
18
19
1j
Ph
83
1k
2ka
1l
104
33
14
6
1m
a) Results from ref 7a.
In support of our hypothesis that the AMGHEs are acting as prodrugs, converting themselves to AMGBLs in vivo, we could
predict that the corresponding α-methylene-γ-hydroxy ketone to be inactive, in spite of its increased Michael acceptor properties.18
Accordingly, we prepared 5-hydroxy-3-methylene-4,5-diphenylpentan-2-one (8) using a procedure similar to the one in Scheme 2,

starting with methyl vinyl ketone in place of methyl acrylate. Expectedly, the hydroxy ketone 8 showed no cytotoxicity (Table 2,
entry 6), confirming that 1,4-addition alone is insufficient for suppressing cell growth.
Scheme 3. Synthesis of various esters
Table 2.
Anti-pancreatic cancer activities of α-methylene carbonyls against Panc-1, MIA-PaCa-2, and BxPC-3 cell lines.
% Cell Growth
Entry Structure
1
(#)
Panc-1
123
MIA PaCa-2
102
BxPC-3
94
1a’
135
5
110
13
12
8
117
3
2
3
4
5
6
7
1a”
5
6
2
6
9
1
7
105
124
53
59
104
129
8
9
105
87
117
8
10
The effectiveness of AMGHEs might be attributed to the binding of the α-methylene-γ-lactones, formed in vivo, to the active site.
The failure of α-methylene-β-hydroxy esters (MBH adducts) to suppress growth compared to the corresponding γ-hydroxy esters
persuaded us to compare the δ-hydroxy esters as well. A representative δ-hydroxy ester, methyl 5-hydroxy-2-methylene-4,5-
diphenylpentanoate (9, Table 2, entry 7) was synthesized from benzyl phenyl ketone via allylation followed by Luche reduction
(Scheme 4). Cell proliferation assay revealed poor activity. The δ-hydroxy ester should be kinetically very slow to lactonize as
compared to the γ-hydroxy ester, which might be responsible for the poor activity. We had recently demonstrated this to be the case
when a δ-hydroxy ester did not undergo aminolactonization, contrary to γ-hydroxy esters.7c
Scheme 4. Synthesis of δ-hydroxy ester
In our earlier report, we had shown that, compared to AMGBLs, an α-ethylidene-γ-hydroxy lactone is inactive for growth
suppression of pancreactic cancer cells.7a This was attributed to decreased Michael acceptor capabilities. In this study, we prepared
a benzylidene-γ-lactone (cis-α-benzylidene-β,γ-diphenylbutanoate, 10) and examined its activity.19 The benzylidene moiety is
slightly electronically activated, but is sterically hindered, so it is not surprising that it is inactive (Table 2, entry 8).
In conclusion, we have demonstrated that the biological activity of α-methylene hydroxy esters depends on several factors. In
addition to Michael acceptor properties, the ability to cyclize to the corresponding lactone appears to be crucial for the activity.

Additionally, the substitution pattern and the number of carbons are both critical to obtain good activity. The hydroxy esters are
easier to prepare and offer improved solubility over the lactones. Further work to better understand the activity of these compounds,
and to optimize the structures is ongoing.
Acknowledgments
We sincerely thank the Herbert C. Brown Center for Borane Research for the funding of this work.
References and notes
1. http://www.cancer.org/cancer/pancreaticcancer/detailedguide/pancreatic-cancer-key-statistics
2. Gemcitabine is marketed as Gemzar® and is a registered trademark of Eli Lilly and Co.
3. (a) Kitson, R. A.; Millemaggi, A.; Taylor, R. J. K. Angew. Chem. Int. Ed. 2009, 48, 9426; (b) Hoffmann, H. M. R.; Rabe, J. Angew. Chem. 1985,
97, 96; (c) Hoffman, H. M. R.; Rabe, J. Angew. Chem. Int. Ed. Engl. 1985, 24, 94.
4. (a) Kupchan, S. M. Trans. N.Y. Acad. Sci. 1970, 32, 85; (b) Kupchan, S. M.; Fessler, D. C.; Eakin, M. A.; Ciacobbe, T. J. Science 1970, 168, 376;
(c) Kupchan, S. M.; Eakin, M. A.; Thomas, A. M. J. Med. Chem. 1971, 14, 1147.
5. Ghantous, A.; Sinjab, A.; Herceg, Z.; Darwiche, N. Drug Discov. Today 2013, 18, 894.
6. Ramachandran, P. V.; Yip-Schneider, M. T.; Schmidt, C. M. Future Med. Chem. 2009, 1, 179.
7. (a) Ramachandran, P. V.; Pratihar, D.; Nair, H. N. G.; Walters, M.; Smith, S. Yip-Schneider, M. T.; Wu, H.; Schmidt, M. C. Bioorg. Med. Chem.
Lett. 2010, 20, 6620; (b) Ramachandran, P. V.; Nicponski, D. R.; Nair, H. N. G.; Helppi, M. A.; Gagare, P. D.; Schmidt, C. M.; Yip-Schneider,
M. T. Bioorg. Med. Chem. Lett. 2013, 23, 6911; (c) Ramachandran, P. V.; Nicponski, D. R. Chem. Comm. 2014, 50, 15216.
8. Neamati, N.; Kabalka, G. W.; Venkataiah, B.; Dayam, R. US Patent 2007/0203224. Aug 30, 2007.
9. DMAPT (LC-1) is being developed as an anti-cancer drug by Leuchemix Inc. Guzman, M. L.; Rossi, R. M.; Neelakantan, S.; Li, X.; Corbett, C.
A.; Hassane, D. C.; Becker, M. W.; Bennett, J. M.; Sullivan, E.; Lachowicz, J. L.; Vaughan, A.; Sweeney, C. J.; Matthews, W.; Carroll, M.;
Liesveld, J. L.; Crooks, P. A.; Jordan, C. T. Blood 2007, 110, 4427.
10. Kim, K. H.; Lee, H. S.; Kim, S. H.; Lee, K. Y. Lee, J.; Kim, J. N. Bull. Korean Chem. Soc. 2009, 30, 1012.
11. Triplicate cell growth assay; Panc-1 (1.5x103 cells/well), MIA PaCa-2 (7.5x102 cells/well), or BxPc-3 (103 cells/well) were plated in 96 well
plates. The following day, (day 1), DMSO-treated (vehicle), 1 or 10 µM solutions of PT or desired compounds were added. 72 h later (day 3), cell
growth was determined by the addition of Cell Titer 96 Cell Proliferation reagent (Promega, Madison, WI), and the absorbance was read at 450
nm. Cell growth was then determined by normalization to 100% of the growth of vehicle treated cells.
12. Basavaiah, D.; Rao, K. V.; Reddy, R. J. Chem. Soc. Rev. 2007, 36, 1581.
13. Kohn, L. K.; Pavam, C. H.; Veronese, D.; Coelho, F.; De Carvalho, J. E.; Almeida, W. P. Eur. J. Med. Chem. 2006, 41, 738.
14. A concentration of 10 µM was set as the upper limit for minimum activity in our studies.
15. It is known that thermal decarboxylation occurs for electron neutral and rich aromatic substituents. Martínez, I.; Andrews, A. E.; Emch, J. D.;
Ndakala, A. J.; Wang, J.; Howell, A. R. Org. Lett. 2003, 5, 399.
16. Bouzide, A. Org. Lett. 2002, 4, 1347.
17. Neises, B.; Steglich, W. Angew. Chem. Int. Ed. 1985, 17, 522.
18. Mulliner, D.; Wondrousch, D.; Schuurmann, G. Org. Biomol. Chem. 2011, 9, 8400-8413.
19. Kim, S. H.; Kim, K. H.; Lee, H. J.; Kim, J. N. Tetrahedron Lett. 2013, 54, 329.
Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Factors influencing the cytotoxicity of α-
methylene-γ-hydroxy esters against
pancreatic cancer
Leave this area blank for abstract info.
∗
P. Veeraraghavan Ramachandrana, , Matthew A. Helppia,
Arlie Lehmkuhlera, Jennifer M. Marchia, C. Max Schmidtb,c, and Michele T. Yip-Schneiderb,c