Basic N-interlinked imipramines show apoptotic activity against malignant cells including Burkitt's lymphoma

Article abstract of DOI:10.1016/j.bmcl.2013.01.020

We here report the synthesis of ethylene glycol N-interlinked imipramine dimers of various lengths from the tricyclic antidepressant desipramine via an amide coupling reaction followed by reduction with lithium aluminium hydride. The target molecules were found to be potent inhibitors of cellular viability while inducing cell type specific death mechanisms in three cancer cell lines including a highly chemoresistant Burkitt's lymphoma cell line. Basic amine analogues were found to be important for increased potency. Imipramine and desipramine were also tested for apoptotic activity and were found to be much less active than the novel dimeric compounds. Imipramine dimers were only found to be moderate inhibitors of the human serotonin transporter (hSERT) having IC₅₀ values in the micromolar region whilst the induction of cell death occurred independently of hSERT expression. These results demonstrate the potential of newly designed and synthesised imipramines derivatives for use against malignant cells, including those resistant to standard chemotherapy.

Full text of DOI:10.1016/j.bmcl.2013.01.020

Bioorganic & Medicinal Chemistry Letters 23 (2013) 1220–1224
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Basic N-interlinked imipramines show apoptotic activity against
malignant cells including Burkitt’s lymphoma
b
c
c
a
Sandra A. Bright a, , Anne Brinkø , Maja Thim Larsen , Steffen Sinning , D. Clive Williams ,
⇑
Henrik H. Jensen b
a School of Biochemistry & Immunology, Trinity Biomedical Sciences Institute, Trinity College Dublin, Ireland
b Department of Chemistry, Aarhus University, Langelandsgade 140, 8000 Aarhus C, Denmark
c Aarhus University Hospital, Skovagervej 2, 8240 Risskov, Denmark
a r t i c l e i n f o
a b s t r a c t
Article history:
We here report the synthesis of ethylene glycol N-interlinked imipramine dimers of various lengths from
the tricyclic antidepressant desipramine via an amide coupling reaction followed by reduction with lith-
ium aluminium hydride. The target molecules were found to be potent inhibitors of cellular viability while
inducing cell type specific death mechanisms in three cancer cell lines including a highly chemoresistant
Burkitt’s lymphoma cell line. Basic amine analogues were found to be important for increased potency.
Imipramine and desipramine were also tested for apoptotic activity and were found to be much less active
than the novel dimeric compounds. Imipramine dimers were only found to be moderate inhibitors of the
human serotonin transporter (hSERT) having IC50 values in the micromolar region whilst the induction of
cell death occurred independently of hSERT expression. These results demonstrate the potential of newly
designed and synthesised imipramines derivatives for use against malignant cells, including those resis-
tant to standard chemotherapy.
Received 14 November 2012
Revised 1 January 2013
Accepted 4 January 2013
Available online 12 January 2013
Keywords:
Tricyclic antidepressant
Anti-cancer
Serotonin transporter
Imipramine
Ó 2013 Elsevier Ltd. All rights reserved.
Antidepressants are able to increase the synaptic concentration
of neurotransmitter monoamines serotonin, dopamine and norepi-
nephrine eventually lifting the symptoms associated with depres-
sion.1 Still used in the clinic in cases of severe depression are
tricyclic antidepressants of which imipramine2 is a prominent rep-
resentative. This drug has a wide range of targets and inhibits the
human serotonin transporter (hSERT) and to a lesser extent the hu-
man norepinephrine transporter (hNET).
Psychoactive drugs like antidepressants (fluoxetine, clomipra-
mine and maprotiline)3,4 and drugs of abuse (amphetamine
analogues)5–7 have previously been evaluated as potential anti-
cancer compounds and found to induce apoptosis in several
malignant Burkitt’s lymphoma cell lines (Scheme 1). This activity
is, however believed to occur independently of hSERT inhibition
despite the presence of these proteins in B cell malignancies.3,4
Imipramine itself has also been found to induce apoptosis in some
cancer cell lines.8 In addition, it was recently found that a triphenyl
methane analogue of imipramine, imipramine blue, has been
shown to have great potential as a new anti-cancer therapy.9 How-
ever, to date there has been no data on the effect of imipramine or
imipramine derivatives against Burkitt’s lymphoma.
HIV infection and is a serious endemic health problem in the
malarial belts of equatorial Africa, north-eastern Brazil and Papua
New Guinea. Some success in the treatment of Burkitt’s lymphoma
in children has been achieved, whereas adults and HIV patients
have a more pessimistic outlook. Accordingly, there is still a need
for developing improved therapeutics, in terms of activity and eco-
nomics against Burkitt’s lymphoma and related diseases.11
Given the recent interesting findings on the anti-cancer effects
of antidepressants, the known activity of polyamines analogues
as anti-cancer agents12,13 and our previous interest in amine termi-
nated tricyclic antidepressants2 we speculated whether a dimer-
ization of imipramine through its amine tail would lead to a
diamine with anti-cancer activity. The aim of this study was there-
fore to investigate how easily prepared dimeric analogues of the
well-known antidepressant imipramine would behave against
malignant cells including Burkitt’s lymphoma. Part of the mecha-
nism of action of these novel compounds was also evaluated.
Novel N-interlinked imipramine dimers were synthesised by
coupling the secondary amine desipramine to ethylene glycol link-
ers having a carboxylic acid functionality in both ends. Ethylene
glycol units of different lengths were chosen as a linker in order
to increase aqueous solubility. The shortest linker, diglycolic acid
(4), is commercially available. The linkers 5 and 6 were prepared
by a procedure described by Fujimoto et al.,14 and Wittmann et
al.,15 using tert-butyl bromoacetate (7) and ethylene glycol (8) with
sodium hydride in DMF followed by acid treatment to give free
Burkitt’s lymphoma is a rare but aggressive B-cell malignancy
that was first documented in 1958.10 It is often associated with
⇑
Corresponding author.
E-mail address: brights@tcd.ie (S.A. Bright).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.01.020

S. A. Bright et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1220–1224
1221
Scheme 1. (A) Structure of some common antidepressants. (B) Synthetic approach for the preparation of imipramine dimers 1–3.
carboxylic acid. In our hands, di-tert-butyl esters 9 and 10 were at-
tained in 35% and 26% yield, respectively. Ester deprotection with
trifluoroacetic acid (TFA)/CH2Cl2 occurred in quantitative yields
to give linkers 5 and 6. The linkers next underwent ethyl dimeth-
ylaminopropyl carbodiimide (EDC) promoted coupling with
desipramine to give the desired compounds in moderate yields
(50–62%). The diamides were then smoothly reduced to the corre-
sponding diamines by reaction with LiAlH4 (Scheme 2).
viability of all three cell lines, a higher drug concentration was re-
quired to inhibit 50% of the cells with IC50 values ranging between
5.8 and 22.9 M. Interestingly, cells were far more sensitive to the
dimeric compounds when compared to the monomeric counter-
parts imipramine and desipramine. Desipramine required higher
l
concentrations to elicit the same effects (7.9–39.8
least potent compound tested, imipramine, showed IC50 values of
14.1 to >50 M (Fig. 1D).
lM) while the
l
The three diamines 1–3, along with the non-basic diamide 12,
were then evaluated for their ability to induce apoptosis in two dif-
ferent Burkitt’s lymphoma cell lines; chemosensitive MUTU-I cells
and chemoresistant DG-75 cells. Additionally, a third cell line,
HeLa, which represents cervical cancer was also was utilised, this
common cell line was used for comparative purposes as it does
not express hSERT. A viability screen initially demonstrated that
all three basic imipramine dimers 1–3 and diamide 12 caused a po-
tent reduction in cell viability. This interesting result was found for
all three cell lines (Fig. 1A–C) with IC50 values of between 1.1 and
Next, we wished to evaluate the ability of the compounds to in-
duce cell death. Apoptosis is a form of programmed cell death that
includes characteristic changes in cellular morphology including
chromatin condensation, DNA fragmentation and membrane bleb-
bing. It can be executed by caspases and proteins known as Bcl-2
proteins. Other forms of cell death include autophagy, whereby a
cell degrades its own components through lysosomal machinery,
and necrosis, which invokes an undesirable inflammatory
response.
Flow cytometry was thus performed with the compounds to
establish whether the compounds could induce cell death. Results
further confirmed cellular viability results, demonstrating that the
diamide linked imipramine 12 was significantly less active than
the diamine linked dimers 1–3 and further established the relative
1.9
lM for compound 1, 0.6 and 1.5
lM for compound 2 and 1.2
and 1.9
lM for compound 3 with the MUTU-I cells having the low-
est IC50 values and the DG-75 cells having the highest values.
While the diamide 12 compound caused a potent reduction in cell
Scheme 2. Synthesis of dimeric imipramine analogues.

1222
S. A. Bright et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1220–1224
A
C
B
1
1
2
3
12
100
2
100
50
0
3
12
50
Imipramine
Desipramine
Imipramine
Desipramine
0
-6.0
-5.5
-[log, X]
-5.0
-4.5
-4.0
-6.0
-5.5
-[log, X]
-5.0
-4.5
-4.0
D
1
2
3
12
100
50
0
Imipramine
Desipramine
-6.0
-5.5
-5.0
-4.5
-4.0
-[log, X]
Figure 1. Cell viability test with antidepressants desipramine/imipramine and dimeric analogues. (A) MUTU-I, (B) DG-75 and (C) HeLa cells, respectively. (D) IC50 values in
M of cells treated for 24 h.
l
inactivity of antidepressants imipramine and desipramine when
compounds were treated at a ligand concentration of 2.5
These significant increases in cell death corresponded to a decrease
in the number of cells associated with the G2/M phase of the cell
cycle ( Fig. 2B). Interestingly, while MUTU-I and HeLa cells
underwent a similar extent of cell death when treated with the
common chemotherapeutic and tubulin depolymeriser paclitaxel
(taxol), the chemoresistant DG-75 cells did not exhibit any signif-
icant increase in cell death. Instead however, there was a signifi-
cant increase in the number of cells in the G2/M phase of the cell
cycle (Fig. 2B), this is consistant with previous results demonstrat-
ing that these cells lack the machinery necessary for the execution
of apoptosis.16 In comparison, compounds 12, imipramine and
desipramine did not exhibit any significant effects on any of three
cell lines at this concentration.
lM
( Fig. 2A). Specifically, MUTU-I cells underwent 66.6 4.2%,
56.2 2.8% and 46.3 8.8% cell death when treated with com-
pounds 1, 2 and 3 (compared to vehicle-induced cell death,
3.7 0.4%). Likewise the DG-75 cell line underwent a significant in-
crease in cell death when treated with compounds 2 and 3
(40.9 7.1% and 54.6 10.8%, cell death) when compared to
vehicle treated cells (0.8 0.25%). DG-75 cells treated with 1 also
underwent a significant, albeit lower, increase in cell death upon
treatment (22.8 2.0%). Similarly, HeLa cells underwent 61.7
5.7%, 44.4 6.4% and 46.0 10.9% cell death when treated with 1,
2 and 3, respectively (vehicle treated cells, 7.0 1.9% cell death).
Figure 2. Flow cytometry testing for (A) cell death and (B) the G2/M phase of the cell cycle. Cell death time course against (C) MUTU-I, (D) DG-75 and (E) HeLa cells,
respectively.

S. A. Bright et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1220–1224
1223
The time-frame in which the diamine imipramine dimers (1–3)
were active was next evaluated. Here it was found that the MUTU-I
cells were the most sensitive cell line as a significant percentage of
cell death was observed after only 2 h of treatment (Fig. 2C). DG-75
cells, started to undergo cell death after 24 h of treatment (Fig. 2D),
while HeLa cells started to undergo cell death after 8 h (Fig. 2E).
For MUTU-I and HeLa cells, diamine imipramine dimer-induced
apoptotic cell death was confirmed by cleavage of the enzyme
PARP, a classical hallmark of apoptosis, from a 116 kDa precursor
to an 85 kDa fragment (Fig. 3B) along with activation of caspas-
es-3/7, intracellular cysteine proteases and downstream execu-
tioners of apoptosis ( Fig. 3A). The pro-apoptotic activity of the
original monomeric version has also been reported to induce the
classical intrinsic apoptotic pathway involving caspase activation
and PARP cleavage.17 However in contrast, DG-75 cells did not dis-
play any activation of caspases or PARP cleavage. Diamine N-inter-
linked imipramine dimers (1–3) are thus speculated to be capable
of inducing an alternative form of cell death, known as autophagy
in these cells. These results are consistent with a previous report of
inhibition of the PI3K/Akt/mTor signalling pathway leading to the
subsequent induction of autophagic cell death.18 Compounds are
most likely to impact the same targets with downstream effects
dependent upon available cellular machinery.
The involvement of reactive oxygen species was also investi-
gated, however results suggested reactive oxygen species were
not involved in the induction of diamine imipramine dimer-in-
duced cell death as pre-treatment with N-acetyl cysteine (NAC), a
reactive oxygen species inhibitor did not prevent cell death. This
was in contrast to a previous report suggesting the production of
reactive oxygen species and loss of mitochondrial membrane po-
tential were critical for imipramine-induced cell death.19
Other suggested targets of imipramine and imipramine
derivatives have included the potassium ion channel hEAG20 while
phospholipidosis, the excessive accumulation of intracellular phos-
pholipids has also been shown to inhibit cellular proliferation.21
Some of these pathways and targets may also be involved in the
pro-apoptotic induction of cancer cell lines by novel basic imipra-
mine dimers, however further investigations will be required to
determine the exact mechanism.
As a known target of imipramine, we next decided to investi-
gate, whether the compounds were also active against hSERT. All
three diamine imipramine dimers (1–3) were moderate inhibitors
of hSERT in the micromolar range; however they were two orders
of magnitude less potent than imipramine which binds to hSERT
with high affinity (Table 1). This suggests that the pro-apoptotic
activity of imipramine dimers 1–3 is independent of hSERT inhibi-
tion, an observation that is consistent with previous reports.4
These results are also in agreement with the viability and apoptosis
assay results listed above which suggest the effects of amine N-
interlinked imipramine dimers to be independent of the hSERT
expression levels of the tested cell lines. Diamide imipramine 12
was found to be inactive against hSERT in accordance with the
binding model by Sinning et al., where salt bridge formation be-
tween a ligand amine and a protein aspartic acid is vital for bind-
ing.2 Imipramine also binds hNET, albeit to a lesser extent than
hSERT and thus may also be a potential target for diamine imipra-
mine dimers.
In conclusion, we have demonstrated that novel N-interlinked
imipramine dimers are able to induce cell death, including apopto-
sis in several cancer cell lines and are more potent than their
monomeric counterparts. While the effects of imipramine has pre-
viously been documented in HeLa cells,22 this is the first account of
the potential anti-cancer effects of imipramine, and its more po-
tent dimer derivatives, against Burkitt’s lymphoma. One amide
analogue was furthermore tested and found to be less potent than
basic amine analogues indicating the necessity of a basic nitrogen
atom for increased potency. The mechanism with which these ac-
tive compounds exert their activity towards malignant cells in-
cludes induction of caspase activation and PARP cleavage for two
of three cell lines tested, while the involvement of reactive oxygen
species and hSERT inhibition is unlinked to apoptotic activity. In
the third cell line tested, the highly chemoresistant DG-75 cell line,
diamine imipramine dimers were also able to circumvent the
intrinsic resistance to apoptosis displayed by these cells by activat-
ing an alternative cell death pathway. These results highlight the
different mechanisms of actions of diamine imipramine derivatives
depending on cell type and available cellular machinery. Planned
future work involves alkylation of the parent imipramine to give
Figure 3. The involvement of (A) caspases, (B) PARP and (C) reactive oxygen species in dimer-induced cell death. Left—MUTU-I cells, centre—DG-75 cells and right column—
HeLa cells.

1224
S. A. Bright et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1220–1224
2. Sinning, S.; Musgaard, M.; Jensen, M.; Severinsen, K.; Celik, L.; Koldsø, H.;
Table 1
Meyer, T.; Bols, M.; Jensen, H. H.; Schiøtt, B.; Wiborg, O. J. Biol. Chem. 2010, 285,
8363.
Inhibition of hSERT by antidepressants desipramine/
imipramine and dimeric analogues
3. Meredith, E. J.; Holder, M. J.; Chamba, A.; Challa, A.; Drake-Lee, A.; Bunce, C. M.;
Drayson, M. T.; Pilkington, G.; Blakely, R. D.; Dyer, M. J.; Barnes, N. M.; Gordon,
J. FASEB J. 2005, 9, 1187.
4. Cloonan, S. M.; Drozgowska, A.; Fayne, D.; Williams, D. C. Leuk. Lymphoma
2010, 51, 523.
IC50 in lM
Desipramine
Imipramine
1
2
3
12
0.56
0.029
6.0
4.8
5.5
5. Bengel, D.; Isaacs, K. R.; Heils, A.; Lesch, K. P.; Murphy, D. L. Neuroreport 1998,
14, 2989.
6. Azmitia, E. C.; Murphy, R. B.; Whitaker-Azmitia, P. M. Brain Res. 1990, 510, 97.
7. McNamara, Y. M.; Cloonan, S. M.; Knox, A. J.; Keating, J. J.; Butler, S. G.; Peters,
G. H.; Meegan, M. J.; Williams, D. C. Bioorg. Med. Chem. 2011, 19, 1328.
8. Jeon, S. H.; Kim, S. H.; Kim, Y.; Kim, Y. S.; Lim, Y.; Lee, Y. H.; Shin, S. Y. Biochem.
Biophys. Res. Commun. 2011, 413, 311.
9. Munson, J. M.; Fried, L.; Rowson, S. A.; Bonner, M. Y.; Karumbaiah, L.; Diaz, B.;
Courtneidge, S. A.; Knaus, U. G.; Brat, D. J.; Arbiser, J. L.; Bellamkonda, R. V. Sci.
Transl. Med. 2012, 4.
46
a monomer with the linker chain attached in order to rule out
potency of the basic linker itself.
We believe that due to the potent anti-cancer activity and ease
of preparation of dimeric imipramines, that the possible use of
these compounds as anti-cancer agents warrants further
investigations.
10. Burkitt, D. Br. J. Surg. 1958, 46, 218.
11. Ferry, J. A. Oncologist 2006, 11, 375.
12. Wallace, H. M.; Fraser, A. V. Biochem. Soc. Trans. 2003, 31, 393.
13. Casero, R. A. J.; Wostera, P. M. J. Med. Chem. 2009, 52, 4551.
14. Fujimoto, K.; Oimoto, N.; Katsuno, K.; Inouye, M. Chem. Commun. (Camb.) 2004,
11, 1280.
15. Wittmann, V.; Takayama, S.; Gong, K. W.; Weitz-Schmidt, G.; Wong, C. J. Org.
Chem. 1998, 63, 5137.
16. Cloonan, S. M.; Williams, D. C. Int. J. Cancer 2011, 128, 1712.
17. Xia, Z.; Bergstrand, A.; DePierre, J. W.; Nässberger, L. J. Biochem. Mol. Toxicol.
1999, 13, 338.
Acknowledgements
We thank the Carlsberg Research Foundation for generously
supporting H.H.J. and Trinity College Dublin for supporting D.C.W.
18. Peng, C. H.; Chiou, S. H.; Chen, S. J.; Chou, Y. C.; Ku, H. H.; Cheng, C. K.; Yen, C. J.;
Tsai, T. H.; Chang, Y. L.; Kao, C. L. Neuropsychopharmacology 2008, 18, 128.
19. Xia, Z.; Lundgren, B.; Bergstrand, A.; DePierre, J. W.; Nässberger, L. Biochem.
Pharmacol. 1999, 57, 1199.
20. Gavrilova-Ruch, O.; Schönherr, K.; Gessner, G.; Schönherr, R.; Klapperstück, T.;
Wohlrab, W.; Heinemann, S. H. J. Membr. Biol. 2002, 188, 137.
21. Piccotti, J. R.; LaGattuta, M. S.; Knight, S. A.; Gonzales, A. J.; Bleavins, M. R. Drug
Chem. Toxicol. 2005, 28, 117.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2013.01.
020.
22. Gupta, R. S.; Murray, W.; Gupta, R. Br. J. Cancer 1988, 58, 441.
References and notes
1. Andersen, J.; Kristensen, A. S.; Bang-Andersen, B.; Strømgaard, K. Chem.
Commun. (Camb.) 2009, 25, 3677.