The synthesis of 4,6-diaryl-2-pyridones and their bioactivation in CYP1 expressing breast cancer cells

Article abstract of DOI:10.1016/j.bmcl.2019.03.030

As part of a programme to develop anticancer prodrugs which are activated by cytochrome P450 (CYP)1B1, a library of 4,6-diaryl-2-pyridones was synthesised in yields of 6–60% from the corresponding chalcones. A number of these derivatives showed promising antiproliferative activities in human breast cancer cell lines which express CYP1B1 and CYP1A1, while showing little toxicity towards a non-tumour breast cell line with no CYP expression. Metabolism studies provided evidence supporting the involvement of CYP1 enzymes in the bioactivation of these compounds.

Full text of DOI:10.1016/j.bmcl.2019.03.030

Accepted Manuscript
The synthesis of 4,6-diaryl-2-pyridones and their bioactivation in CYP1 ex-
pressing breast cancer cells
Ketan C. Ruparelia, Sabahat Lodhi, Dyan N. Ankrett, Nicola E. Wilsher,
Randolph R.J. Arroo, Gerard A. Potter, Kenneth J.M. Beresford
PII:
S0960-894X(19)30169-6
https://doi.org/10.1016/j.bmcl.2019.03.030
BMCL 26344
DOI:
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
14 January 2019
18 March 2019
22 March 2019
Please cite this article as: Ruparelia, K.C., Lodhi, S., Ankrett, D.N., Wilsher, N.E., Arroo, R.R.J., Potter, G.A.,
Beresford, K.J.M., The synthesis of 4,6-diaryl-2-pyridones and their bioactivation in CYP1 expressing breast cancer
cells, Bioorganic & Medicinal Chemistry Letters (2019), doi: https://doi.org/10.1016/j.bmcl.2019.03.030
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.

The synthesis of 4,6-diaryl-2-pyridones and their bioactivation in CYP1 expressing breast
cancer cells
Ketan C. Ruparelia, Sabahat Lodhi, Dyan N. Ankrett, Nicola E. Wilsher, Randolph R. J. Arroo,
Gerard A. Potter and Kenneth J. M. Beresford*
Leicester School of Pharmacy, Faculty of Health and Life Sciences, De Montfort University, The
Gateway, Leicester, LE1 9BH, UK.
*Corresponding author. Tel.: +44-116-250-6356; e-mail address: kberesford@dmu.ac.uk
ABSTRACT
As part of a programme to develop anticancer prodrugs which are activated by cytochrome P450
(CYP)1B1, a library of 4,6-diaryl-2-pyridones was synthesised in yields of 6-60% from the
corresponding chalcones. A number of these derivatives showed promising antiproliferative
activities in human breast cancer cell lines which express CYP1B1 and CYP1A1, while showing little
toxicity towards a non-tumour breast cell line with no CYP expression. Metabolism studies provided
evidence supporting the involvement of CYP1 enzymes in the bioactivation of these compounds.
Keywords: 4,6-diaryl-2-pyridones, chalcones, antiproliferative, prodrug, CYP1B1
Cytochrome P450 (CYP) expression has been studied in a variety of human tumours and it is clear
that many have distinct CYP profiles overexpressing specific CYP sub-families [1, 2]. The
therapeutic action of a number of clinically used anticancer drugs has been shown to involve
metabolism by CYPs and several anticancer prodrugs have been designed to be activated by these
enzymes [3]. Although the expression levels of many CYPs differ between tumour and
corresponding normal tissue CYP1B1 is one of the few CYP subfamilies which is significantly and
consistently overexpressed in tumours and is capable of metabolising a structurally diverse range of
anticancer drugs leading to drug resistance in certain cancer cell lines [4]. Because of this, and its
role in the activation of procarcinogens, CYP1B1 is seen as an important target for anticancer drug
development both in terms of inhibitors for this enzyme and in prodrug activation [5-7]. We have
designed several compounds based on the chalcone and stilbene structures to be activated by
CYP1B1 [8-10]. A number of these have also shown potential as chemopreventative agents
including the chalcone DMU-135 (Fig. 1) which was found to inhibit Apc(Min/+) mouse

Figure 1. Incorporation of a hererocyclic ring into chalcone and stilbene structures.
gastrointestinal adenoma formation, reducing adenoma multiplicity by 46% compared to controls.
DMU-135 was well tolerated and induced no systemic side-effects [11-13]. Our group has also
demonstrated that CYP1B1 and CYP1A1 catalyse the conversion of certain naturally occurring
methoxylated flavones with good bioavailability but relatively little biological activity into
hydroxylated metabolites with strong antiproliferative activity [14-16].
The therapeutic use of most chalcones and stilbenes is hampered by their high lipophilicity which
results in low water solubility and poor pharmaco-kinetics [17]. In addition, solutions of some
chalcones are susceptible to light induced isomerisation about the carbon–carbon double bond
resulting in mixtures of isomers [18]. One potential solution to these problems is to integrate the key
structural features of the molecule into a heterocyclic system. This approach has been used with
some success by Wang et al who prepared analogues of the cytotoxic stilbene combretastatin A-4 1
(Fig. 1) in which the key cis double bond was replaced by a 4,5-disubstituted imidazole ring. The
incorporation of this ring prevented isomerisation about the double bond and the resulting
analogues 2 and 3 (Fig. 1) had greatly improved pharmacokinetics, longer half-lives and excellent
bioavailability [17]. For DMU-135 a heterocyclic ring could be incorporated into the chalcone
structure as shown in 4 (Fig 1), where the dotted line represents the heterocyclic ring. This would fix
the important geometry about the 2-propenone portion of the molecule and, depending on the
nature of the ring system, could enable formulation as a pharmaceutically acceptable water soluble
addition salt.
As a starting point, it was decided to incorporate a 2-pyridone ring into the chalcone structure.
The 2-pyridone nucleus occurs widely in biologically important natural alkaloids and is a useful
intermediate for the generation of other heterocyclic systems which will allow for further elaboration
of analogues generated [19]. Chalcones are readily accessible in high yield via the Claisen-Schmidt

O
O
-H2O
-BtH
O
Bt
i)
NH
Ar'
OH
NH
Ar'
Ar
Ar'
Ar
Ar
6
5
7
Scheme 1. Reagents and conditions: i) BtCH2CONH2, NaOH, EtOH, reflux.
condensation reaction and are therefore an attractive starting material [10]. Several methods have
been reported for the preparation of 3-unsubstituted 2-pyridones from chalcones, notable among
these is the benzotriazole-assisted preparation developed by Katritzky et al (Scheme 1) [19]. The
chalcone 5 undergoes a tandem reaction with 2-(benzotriazol-1-yl)acetamide to yield a 3-
(benzotriazol-1-yl)-4,6-diaryl-2-piperidinone 6 as an intermediate. Loss of benzotriazole (BtH) and
water generates the pyridone ring 7. A modification of this approach using 2-acetamidoacetamide
(N-acetylglycinamide) in place of 2-(benzotriazole-1-yl)acetamide offers advantages of
commercially available starting reagents and shorter reaction time while retaining high
regioselectivity and good yields [20].
A series of chalcones 5a-q were prepared in yields of 51-94% as described by us previously [10]
by reacting the appropriate benzaldehyde and acetophenone with aqueous sodium hydroxide
solution in methanol at room temperature (Scheme 2). Conversion of the chalcones into the
corresponding 4,6-diaryl-2-pyridones 7a-q was achieved using a modification of the method of
Wang et al [20]. A mixture of the chalcone, 2-acetamidoacetamide and cesium carbonate in DMF
was heated at 150˚C for 1 h (Scheme 2). The crude materials were purified by either flash
chromatography or trituration with ethyl acetate affording the desired products in yields of 6-60%.
Analytical data were fully consistent with the proposed structures. A combinatorial chemistry
approach was adopted for these syntheses with no effort being made to optimise individual yields.
The antiproliferative activities of the pyridones 7a-r were investigated using a human breast cell
line panel which has been characterised for CYP expression in our laboratories [14-16]. This panel
consists of MCF-7, an ER positive tumour cell line with very low levels of CYP expression, MCF-7
induced with TCDD (2,3,7,8-tetrachlorodibenzo-p-dioxin) which expresses high levels of CYP1A1,
MDA-MB-468, an ER negative tumour cell line which constitutively expresses both CYP1B1 and
O
O
NH
Ar'
(i)
(ii)
Ar
Ar'
+
ArCHO
CH3COAr'
Ar
7
5
Scheme2. Reagents and conditions: (i) NaOH(aq), MeOH, rt, 51-94% yield. (ii)
CH3CONHCH2CONH2, Cs2CO3, DMF, 1500C, 1h, 6-60% yield.

CYP1A1 and MCF-10A, a non-tumour ‘normal’ cell line which has no basal CYP expression. IC50
values were determined using a standard MTT cytotoxicity assay [10] and are shown in Table 1. To
enable rapid characterization of compounds, two selectivity ratios are included in this table.
Activation factor (AF) quantifies the change in activity in the MCF7 cell line on treatment with TCDD
(AF = IC50 (MCF7)/ IC50 (MCF7 + TCDD)) with values above one indicating an increase in activity.
Tumour selectivity (TS) characterizes the difference in activity between the non-tumour MCF-10A
cells and the MDA-MB-468 breast cancer cell line (TS = IC50 (MCF-10A)/ IC50 (MDA-MB-468)). TS
values above one indicate selectivity for the CYP1 expressing cancer cell line.
Compound 7a, which has the same aromatic substitutions as DMU-135, showed little
antiproliferative activity in the MCF-10A and MCF-7 cell lines both of which have low levels of CYP
expression. In the MDA-MB-468 cell line, which constitutively expresses both CYP1B1 and
CYP1A1, a slightly higher activity was observed. Treatment of the MCF-7 cells with TCDD, to
induce CYP1A1 expression, resulted in a significantly increased potency, the IC50 for 7a decreasing
to 0.2 M with an AF >500. Moving the methoxy groups to the 2,3,4 positions, as in 7b, had the
effect of increasing toxicity to both the cancer cell lines. However, treatment of the MCF-7 cells with
TCDD did not produce the same increase in potency as was observed with 7a suggesting that the
location of the methoxy substituents on the B-ring is key for activity. Little toxicity was seen with the
non-tumour MCF-10A cells. Pyridone 7c, with a 3,5-dimethoxyphenyl B-ring, was highly toxic to all
the cell lines tested. The 3,4-dimethoxy derivative 7d showed little activity in the MCF-7 and MCF-
10A cell lines but relatively high potency in the MDA-MB-468 cells. As this cell line constitutively
expresses CYP1B1 this result could suggest selective activation by this enzyme possibly indicating
that a 3,4-dimethoxy substitution on the B-ring confers selectivity. Compound 7e was highly active
in the MDA-MB-468 cancer cell line with an IC50 of 0.04 M. In the non-tumour cell line this 2,4-
dimethoxy B-ring analogue gave an IC50 of 40 M, a TS of 1000. Some antiproliferative activity was
also seen in the MCF-7 cells, particularly when these were treated with TCDD. The 2,5-dimethoxy
derivative 7f showed similar activity to 7e in the MDA-MB-468 cell line but was more toxic to the
non-tumour cells. In the MCF-7 cell line an IC50 of 2 M was reduced to 0.02 M on treatment of
these cells with TCDD. Thus, as with our trimethoxy analogues, the location of the two methoxy
groups on the B-ring appears to influence both activity and selectivity within the cell lines. The 4-
methoxy derivative 7h showed similar selectivity to 7d for the MDA-MB-468 cell line. The 2-methoxy
analogue 7g was much less selective for the MDA-MB-468 cells being more comparable to 7e in
activity. Replacing the methoxy groups in the B-ring of 7e with chloro groups, as in 7i, resulted in a
derivative with little activity thus confirming the importance of the methoxy groups. The
unsubstituted B-ring analogue 7j showed relatively high toxicity to all the cells tested. The 3,4-
dimethoxy A-ring analogue 7k was prepared to investigate the effect of A-ring substitution on
activity. 7k was less active than the comparable 3,4-methylenedioxy analogue 7e in the MDA-MB-

Table 1. Antiproliferative activities of 4,6-diaryl-2-pyridones
Cell line/ IC50 Ma
Cell line/ IC50 Ma
No
Ar (A-ring)
Ar' (B-ring)
AF
TS
MCF-7
MDA-
MCF-7
+
MCF-10A
MB-468
TCDD
7a
7b
7c
7d
7e
7f
3,4-OCH2O-Ph
3,4-OCH2O-Ph
3,4-OCH2O-Ph
3,4-OCH2O-Ph
3,4-OCH2O-Ph
3,4-OCH2O-Ph
3,4-OCH2O-Ph
3,4-OCH2O-Ph
3,4-OCH2O-Ph
3,4-OCH2O-Ph
3,4-(CH3O)2Ph
2,4-(CH3O)2Ph
3,4,5-(CH3O)3Ph
2,3,4-(CH3O)3Ph
3,5-(CH3O)2Ph
3,4-(CH3O)2Ph
2,4-(CH3O)2Ph
2,5-(CH3O)2Ph
2-CH3OPh
>100
12
0.2
10
>500
1.2
2
50
8
>100
>100
0.7
>100
40
>2
>12.5
2
0.4
>100
9
0.2
>100
1.8
0.02
3.5
>100
70
0.35
0.41
0.04
0.02
0.7
1
1
>244
1000
250
40
5
2
100
2
5
7g
7h
7i
7
28
4-CH3OPh
>100
85
1
>100
>100
2.5
50
>100
>1.7
3.1
2,4-(Cl)2Ph
1.2
2
60
7j
Ph
1.6
1
0.8
0.4
4.5
3
0.8
0.5
5
7k
7l
2,4-(CH3O)2Ph
2,4-(CH3O)2Ph
2,4-(CH3O)2Ph
2,4-(CH3O)2Ph
2,4-(CH3O)2Ph
3,4-(CH3O)2Ph
3,4-OCH2O-Ph
2,4-(CH3O)2Ph
2.5
2
100
1
9
5
7m 3,4-O(CH2)2O-Ph
2.5
0.5
45
0.8
2.5
5
0.5
0.08
0.4
0.18
6
6
12
7n
7o
7p
7q
7r
4-CH3OPh
4-PrOPh
0.2
9
1.6
50
20
125
389
13
4-PrOPh
60
20
3
70
2,4-(CH3O)2Ph
3,4-(HO)2Ph
18
1
18
2.1
1
80
1.5
0.002
0.7
0.002
0.5
0.003
1.5
0.0016
3
Combretastatin A-4
0.53
aValues are means two independent experiments in quadruplicate.
468 cell line, but more active in the MCF-7 cells. Changing to a 2,4-dimethoxy substitution on the A-
ring (7l) reduced activity in the cancer cell lines while increasing toxicity in the non-tumour cells.
This could indicate that a 3,4-substitution on the A-ring is important for activity. The 3,4-
ethylenedioxy analogue 7m was significantly less active in the cancer cell lines than its
methylenedioxy counterpart and more toxic to the non-tumour MCF-10A cells. The 4-methoxy A-ring
analogue 7n, with a 2,4-dimethoxyphenyl B-ring, had comparable activity to 7e in the MDA-MB-468
cell line but was significantly more toxic to the MCF-10A cells. It was also more active in the MCF-7
cell line. Interesting results were obtained when the 4-methoxy group on the A-ring was replaced

Table 2. Effect of -naphthoflavone on the activity of 7e
NF/ M
Cell line/ IC50 M
MCF-7
MCF-7 + TCDD
MDA-MB-468
MCF-10A
40
0
1
9
12
1.8
9
0.04
>100
>100
with a propoxy group. These derivatives 7o and 7p showed good selectivity for the MDA-MB-468
cell line with TS values of 125 and 389 respectively. The latter, like 7d, has a 3,4-dimethoxy
substitution on the B-ring again indicating that this substitution may result in selectivity for the
CYP1B1 expressing cell line. To investigate if the position of the nitrogen atom in the pyridone ring
had an influence on antiproliferative activity a derivative of 7e was prepared in which the aromatic
groups were “reversed”. This compound 7q, showed much lower activity in the MDA-MB-468 cell
line when compared with 7e thus indicating that the location of the nitrogen in the pyridone ring is
important.
As a preliminary investigation into the role played by CYP1 enzymes in the mechanism of action
of the 4,6-diaryl-2-pyridones the cytotoxicity assay for 7e was repeated with the addition of the
known CYP1 inhibitor -naphthoflavone (NF) [21]. If CYP1 enzymes are involved in the
bioactivation of this derivative, then their inhibition should result in a reduction in antiproliferative
activity. The analogue 7e was selected because it shows activity in both the cancer cell lines and
low toxicity to the non-tumour MCF-10A cells. In addition, we have previously carried out the same
experiment with the parent chalcone 5e [22]. With 7e in the MDA-MB-468 cell line addition of NF
produced a significant reduction in activity, the IC50 increasing from 0.04 to over 100 M (Table 2).
With the TCDD treated MCF-7 cells, which expresses high levels of CYP1A1, a more modest 5 fold
reduction in activity was observed, the IC50 value obtained being comparable to that of the untreated
MCF7 cells. Both these results are consistent with CYP1 enzymes being involved in the bioactvation
of this compound. The greater activity seen with 7e in the MDA-MB-468 cell line, which expresses
both CYP1B1 and CYP1A1, could suggest that CYP1B1 is playing a more significant role than
CYP1A1 in the activation of this compound. To investigate this further the pyridone 7e was
incubated with SupersomesTM expressing either CYP1A1 or CYP1B1 and NADPH reductase
(20pmol.mL-1 of human CYP expressed in insect cells). Control incubations were carried out using
non-CYP expressing insect SupersomesTM or with the omission of either test compound, NADPH or
the active enzyme. Samples were taken at 5 min intervals in duplicate with the enzymatic reaction
being terminated by the addition of ice-cold acetonitrile. Following centrifugation, aliquots of the
supernatant were analysed by reverse phase HPLC. As can be seen (Fig. 1, Supporting info.), the
pyridone was metabolised more rapidly by CYP1A1 than CYP1B1 with half-lives of 12.9 and 30.1
mins respectively. With CYP1A1 the concentration of 7e appeared to plateau at around 15 mins
which may indicate that this compound is a mechanism based inhibitor of this enzyme.

O
O
NH OCH3
OCH3
NH OCH3
OCH3
(i)
HO
O
OH
O
7r
7e
Scheme 3. Reagents and conditions: (i) BCl3, CH2Cl2, -50C, 1h, 59%.
CYP1A1 produced a single metabolite with 7e which eluted at 10.48 mins (Fig. 2, Supporting
info.). CYP1B1 gave two metabolites, one with the same retention time as that obtained with
CYP1A1 and a second, more polar, metabolite with a retention time of 9.22 mins (Fig. 2, Supporting
info.). Due to the increased polarity of the metabolites compared to 7e, and our previous work with
the corresponding chalcone 5e [22], we anticipated that these metabolites resulted from either: i)
dealkylation of the methylenedioxy group to give a catechol or, ii) demethylation of one of the
methoxy groups to give an alcohol. The catechol metabolite 7r was prepared from 7e in 59 % yield
by reaction with boron trichloride and purified by flash chromatography (Scheme 3) [23]. When
supernatants from each of the enzyme reactions were spiked with this authentic standard co-elution
of 7r with the peaks at 10.48 mins provided evidence that this is the major metabolite produced by
both CYP1A1 and CYP1B1. The second metabolite produced by CYP1B1 is more polar than 7r and
therefore unlikely to result from the demethylation of 7e. It is possible that this metabolite arises
from demethylation of the previously formed catechol 7r. To investigate whether the metabolite 7r
showed antiproliferative activity, it was tested against our human breast cell line panel where it was
found to be toxic to all the cell lines tested, including the non-tumour MCF-10A cells with an IC50 of
1.5 M (Table 1). This result provides further evidence that metabolism by CYP1 enzymes is
involved in the mechanism of action of 7e, with 7r being an active metabolite. The reduced activity
seen with 7r in the MDA-MB-468 cell line when compared to 7e is probably due to its increased
polarity, the catechol group making it more difficult for this molecule to permeate the cell membrane.
In summary, certain 4,6-diaryl-2-pyridones show promising and selective antiproliferative
activities in human breast cancer cell lines which express CYP1B1 and CYP1A1, while showing little
toxicity towards a non-tumour breast cell line with no CYP expression. Metabolism studies have
provided supporting evidence for the involvement of CYP1 enzymes in the bioactivation of these
compounds.
Acknowledgments
De Montfort University for a student bursary (SL). The EPSRC UK National Mass Spectrometry
Faculty at Swansea University.

References
1.
2.
3.
Go, R.E.; Hwang, K.A.; Choi, K.C. J. Steroid Biochem. Mol. Biol., 2015, 147, 24-30.
Bruno, R.D.; Njar, V.C.O. Bioorg. Med. Chem., 2007, 15, 5047-5060.
Rooney, P.H.; Telfer, C.; McFadyen, M.C.E.; Melvin, W.T.; Murray, G.I. Curr. Cancer Drug
Targets, 2004, 4, 257-265.
4.
5.
6.
7.
8.
9.
Gajjar, K.; Martin-Hirch, P.L.; Martin, F.L. Cancer Lett., 2012, 324, 13-30.
Dutour, R.; Poirier, D. Eur. J. Med. Chem., 2017, 135, 296-306.
D'Uva, G.; Baci, D.; Albini, A.; Noonan, D. M. Cancer Treatment Rev., 2018, 63, 1-18.
McFayden, M.C.E.; Murry, G.I. Future Oncol. 2005, 1, 259-263.
Potter, G. A.; Ijaz, T. U.S. Patent 7,112,698 B2, 2006.
Potter, G. A.; Butler, P. C. U.S. Patent 7,598, 294 B2, 2009.
10. Ruparelia, K. C.; Zeka, K.; Ijaz, T.; Ankrett, D. N.; Wilsher, N. E.; Butler, P. C.; Tan, H. L.;
Lodhi, S.; Bhambra, A. S.; Potter, G. A.; Arroo R. R. J.; Beresford, K. J. M. Med. Chem., 2018,
14, 322-332.
11. Sale, S.; Tunstall ,R. G.; Ruparelia, K. C.; Butler, P. C.; Potter, G. A.; Steward, W. P.;
Gescher, A. J. Invest. New Drugs, 2006, 24, 459-464.
12. Sale, S.; Tunstall ,R. G.; Ruparelia, K. C.; Potter, G. A.; Steward, W. P.; Gescher, A. J. Int. J.
Cancer, 2005, 115, 194-201.
13. Sale, S.; Verschoyle, R.; Boocock, D.; Jones, D. ; Wilsher, N.; Ruparelia, K.; Potter, G.;
Farmer, P.; Steward, W.; and Gescher, Brit. J. Cancer, 2004, 90, 736-44.
14. Androutsopoulos, V.; Wilsher, N.; Arroo, R. R. J.; Potter, G. A. Cancer Lett., 2009 274, 54-60.
15. Androutsopoulos, V. P.; Ruparelia, K.; Arroo, R. R. J.; Tsatsakis, A. M.; Spandidos, D. A.
Toxicol., 2009, 264, 162-170.
16. Androutsopoulos, V.; Arroo, R. R. J.; Hall, J. F.; Surichan, S.; Potter, G. A. Breast Cancer
Res., 2008, 10, R39.
17. Wang, L.; Woods, K. W.; Li, Q.; Barr, K. J.; McCroskey, R. W.; Hannick, S. M.;Gherke, L.;
Credo, R. B.; Hui, Y.; Marsh, K.; Warner, R.; Lee, J. Y.; Zielinski-Mozng, N.; Frost, D.;
Rosenberg, S. H.; Sham, H. L. J. Med. Chem., 2002, 45, 1697-1711.
18. Larsen M.; Kroman H.; Kharazmi A.; Nielsen S. F. Bioorg. Med. Chem. Lett., 2005, 15, 4858-
4861.
19. Kartritzky, A. R.; Belyakov, S. A.; Sorochinsky, A. E.; Henderson, S. A.; Chen, J. J. Org.
Chem., 1997, 62, 6210-6214.
20. Wang, S.; Gang, Y.; Lu, J.; Xiao, K.; Hu, y.; Hu, H. A. Synthesis, 2003, 4, 487-490.
21. Shimada, T.; Yamakazi, H.; Foroozesh, M.; Hopkins, N. E.; Alworth, W. L.; Guengerich, F. P.
Chem. Res. Toxicol., 1998, 11, 1048-1056.
22. Ankrett,
D.N.
PhD
Thesis,
De
Montfort
University,
Leicester,
2008.
https://www.dora.dmu.ac.uk/xmlui/handle/2086/4282

23. Castro; A. Migueldel, J. M.; Gordaliza, C. M.; Grande, C.; Gómez-Zurita, A.; Garcıa-Grávalos,
́
D.; Feliciano, Eur. J. Med. Chem., 2003, 38, 65-74.