Antitumoural sulphur and selenium heteroaryl compounds: Thermal characterization and stability evaluation

Article abstract of DOI:10.3390/molecules22081314

The physicochemical properties of a compound play a crucial role in the cancer development process. In this context, polymorphism can become an important obstacle for the pharmaceutical industry because it frequently leads to the loss of therapeutic effectiveness of some drugs. Stability under manufacturing conditions is also critical to ensure no undesired degradations or transformations occur. In this study, the thermal behaviour of 40 derivatives of a series of sulphur and selenium heteroaryl compounds with potential antitumoural activity were studied. In addition, the most promising cytotoxic derivatives were analysed by a combination of differential scanning calorimetry, X-ray diffraction and thermogravimetric techniques in order to investigate their polymorphism and thermal stability. Moreover, stability under acid, alkaline and oxidative media was tested. Degradation under stress conditions as well as the presence of polymorphism was found for the compounds VA6E and VA7J, which might present a hurdle to carrying on with formulation. On the contrary, these obstacles were not found for derivative VA4J.

Full text of DOI:10.3390/molecules22081314

molecules
Article
Antitumoural Sulphur and Selenium Heteroaryl
Compounds: Thermal Characterization and
Stability Evaluation
Verónica Alcolea 1,†, Pablo Garnica , Juan A. Palop , Carmen Sanmartín *,
1,†
1
1,
Elena González-Peñas , Adrián Durán and Elena Lizarraga 1
1
2
1
Department of Organic and Pharmaceutical Chemistry, Faculty of Pharmacy and Nutrition,
University of Navarra, Irunlarrea 1, 31008 Pamplona, Spain; valcolea@alumni.unav.es (V.A.);
pgarnica@alumni.unav.es (P.G.); jpalop@unav.es (J.A.P.); mgpenas@unav.es (E.G.-P.);
elizarraga@unav.es (E.L.)
2
Department of Chemistry, Faculty of Sciences, University of Navarra, Irunlarrea 1, 31008 Pamplona, Spain;
adrianduran@unav.es
*
†
Correspondence: sanmartin@unav.es; Tel.: +34-948-425600
Authors have contributed equally to this article.
Received: 29 June 2017; Accepted: 4 August 2017; Published: 8 August 2017
Abstract: The physicochemical properties of a compound play a crucial role in the cancer
development process. In this context, polymorphism can become an important obstacle for the
pharmaceutical industry because it frequently leads to the loss of therapeutic effectiveness of some
drugs. Stability under manufacturing conditions is also critical to ensure no undesired degradations
or transformations occur. In this study, the thermal behaviour of 40 derivatives of a series of
sulphur and selenium heteroaryl compounds with potential antitumoural activity were studied.
In addition, the most promising cytotoxic derivatives were analysed by a combination of differential
scanning calorimetry, X-ray diffraction and thermogravimetric techniques in order to investigate their
polymorphism and thermal stability. Moreover, stability under acid, alkaline and oxidative media
was tested. Degradation under stress conditions as well as the presence of polymorphism was found
for the compounds VA6E and VA7J, which might present a hurdle to carrying on with formulation.
On the contrary, these obstacles were not found for derivative VA4J.
Keywords: thermal analysis; activity; stability; polymorphism; differential scanning calorimetry;
thermogravimetry; X-ray diffraction; HPLC
1
. Introduction
Cancer is a leading cause of death worldwide; the World Health Organization estimates the
growth rate of new cases to peak at 70% in the next twenty years [
territory, an estimated 1650 deaths per day will be due to cancer [ ]. Many treatment patterns include
some kind of chemotherapy or molecularly-targeted therapy drugs. Cancer research is nowadays
encountering two major obstacles, side-effects and tumour resistance, both intrinsic and acquired [ ].
1]. If we draw the focus to US
2
3,4
Therefore, the synthesis of new, safer and more effective drugs is urgent in order to be able to overcome
those previously mentioned difficulties developed by some of the present treatments.
Among the arsenal of new anti-cancer agents, selenium (Se)-containing compounds stand out
for their anti-tumour potency and safety [
and dose used [ ]. For the past decade, our group has focused on the synthesis of sulphur (S) and
Se-containing molecules as antitumour agents with encouraging results [ ]. In the present study, the
homology between those two atoms will be explored in order to widen the variety of chemical entities.
5]. These properties strongly depend on the chemical form
6
7,8
Molecules 2017, 22, 1314; doi:10.3390/molecules22081314
www.mdpi.com/journal/molecules

Molecules 2017, 22, 1314
2 of 22
The introduction of this structural duality will enable us to contrast Se and S influence in terms of
anti-cancer activity, stability and thermal characterization.
It has been found that pharmaceutical compounds such as cephalexin, chloramphenicol, digoxin,
spironolactone, sulfathiazole, carbamazepine and retroviral agents [9–12] undergo transitions during
drying processes, spraying, grinding, compression, excipient addition and the preparation of tableting.
Polymorphism in pharmaceutical solids can be a critical manufacturing variable and therefore has to
be investigated [13].
Physicochemical properties such as bioavailability and stability are directly related to the crystal
structure and possible polymorphism [14,15]. Therefore, the prediction of polymorphism in the early
stages of drug development provides valuable information to selectively synthesize or isolate the
crystalline structure with the best characteristics. In addition, polymorphic transformations can occur
during some steps of the synthesis including purification, crystallization, drying and storage [16,17].
Furthermore, this property influences several parameters such as solubility, shelf life, and formulation
processes [18].
Investigating the relative thermal stability of polymorphs is also essential because the instant
appearance or disappearance of a crystalline structure can lead to serious consequences if the
polymorphic transformation occurs in the dosage form [19]. Da Silva et al. [20] have found that
the solubility and dissolution properties of finasteride, which is widely used for prostate cancer
treatment, were significantly affected by polymorphism. This disparity in bioavailability is related to
the diverse dissolution rate of the different polymorphic forms. In some cases, the metastable form of
some active ingredients may be twice as active as the stable form [21]. In addition, Yoshida et al. [22
]
have observed that the degradation products after drug stress conditions are related to some events
that appeared in the calorimetric curves.
The study of the physical stability of drugs should be properly conducted in order to determine the
characteristics of the proposed formulation and predict possible changes in the solid state of the drug.
Certain pharmaceutical processes (granulation, drying, and compression) can lead to polymorphic
transformations and thus the critical steps in this process must be identified and controlled to ensure
stability until the end of the drug’s shelf life [23].
If we narrow the view to Se and S derivatives, those properties have proved to be critical.
Some Se-containing structures i.e., ebselen, produce two different polymorphs when crystallized
from different solvents that differ in the antioxidant activity related to the ability of Se–N bond
cleavage as well as the potential for Se–O interactions [24]. Moreover, S derivative compounds such as
dalcetrapib [25], probenecid [26] or tauroursodeoxycholic acid [27] showed the ability to crystalize in
different structures, an ability that dramatically affects their biological activity profile.
Those heteroaryl derivatives that exhibited promising cytotoxic activity (VA7J, VA6O and VA5P)
in previous publications from the group [28,29] have the potential to continue the drug discovery
process. With the objective of predicting the non-desired transitions or degradation processes that may
occur along with the already mentioned pharmaceutical manufacturing processing, we proceeded to
the thermal characterization and preliminary stability studies of the leader compounds. In addition,
the biological and thermal characterization of the novel compounds recently developed by our group
have also been included in this work. Inactive analogues were also studied following the same
methodology in order to be able to establish a comparison.
2
. Results and Discussion
2
.1. Synthesis and Selection Criteria
The selected compounds for this study were synthesized in the Organic and Pharmaceutical
Synthesis Unit of the University of Navarra. A total of 40 compounds were included in this work.
They were divided into 12 series (series B to P, Figure 1) according to their central skeleton. They are
hybrid derivatives containing diverse substituents with S or Se and several active heterocyclic scaffolds.

Molecules 2017, 22, 1314
3 of 22
Six novel derivatives (VA6G, VA6H, VA6I, VA2J, VA4J and VA6J) were incorporated to complete the
series. The compounds VA6G, VA6H, VA6I and VA6J were synthesized according to a previously
published procedure by reaction of the corresponding alkyl halides with potassium thiocyanate in
acetone under reflux conditions [30]. To obtain compound VA2J, 2,3-dichloroquinoxaline was treated
with thiourea in a 1:2.2 (reagent/thiourea) molar ratio at reflux, using absolute ethanol as solvent. VA4J
was synthesized by the treatment of derivative VA2J with a potassium hydroxide solution in DMSO
and stirring at room temperature (r.t.), with a posterior methylation with methyl iodide. The final
compound was purified taking advantage of its solubility properties. The synthesis of the remaining
3
4 derivatives has been previously published by our group [28,29].
Figure 1. Structures of S and Se heteroaryl compounds studied in this work. The compounds VA6G,
VA6H, VA6I, VA2J, VA4J and VA6J are novel molecules. The structures of series 7 and 8 have been
previously described [29] and the remaining ones have been extracted from a previous work from our
group [28].
In order to perform the thermal characterization and preliminary stability studies of those
compounds with the best drug-likeness profile, the cytotoxic potency was evaluated by the MTT
(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay.
The novel synthetic compounds VA2J, VA4J, VA6G, VA6H, VA6I and VA6J were tested in breast
adenocarcinoma (MCF-7) and non-malignant mammary-gland-derived (184B5) cell lines to be able
to establish their potency and selectivity. The reference drug used was methylseleninic acid (MSA)
as a Se-containing structure with promising in vitro and in vivo results against different tumour
types [31
concentration that inhibits completely cell growth (TGI) and the concentration that kills 50% of cells
LD ). The selectivity index (SI) is calculated as GI (184B5)/GI (MCF-7).
,32]. Results are expressed as the concentration that inhibits 50% cell growth (GI ), the
50
(
50
50
50
As shown in Table 1, five out of the six tested compounds (VA2J, VA4J, VA6G, VA6H and VA6I)
exhibited GI50 values under 10 M. If we move to selectivity values, the compound VA4J emerges as
the most selective derivative with an SI over 50 times higher than MSA.
For the remaining compounds of series and , cytotoxic activity had already been
evaluated following the same methodology [28]. Cell viability assays for compounds of series and
µ
1
,
2,
3,
4
,
5
6
7
8
had already been screened in a panel of six human cancer cell lines: 1205Lu (melanoma), A549 (lung),
DU145 (prostate), HCT116 (colon), PANC-1 (pancreas) and BxPC3 (pancreas) [29].

Molecules 2017, 22, 1314
4 of 22
Table 1. Average values of GI , TGI and LD (µM) and calculated SI.
5
0
50
Cell Lines
Compound
MCF-7
TGI
184B5
TGI
SI
GI50
LD50
GI50
LD50
VA2J
VA4J
VA6G
VA6H
VA6I
VA6J
2.20
1.02
3.59
8.53
4.51
18.50
13.90
>100
23.95
14.83
>100
50.55
44.02
>100
67.27
24.18
>100
83.93
12.45
52.45
14.29
27.76
2.26
15.97
>100
34.32
75.84
>100
69.80
20.48
>100
64.84
>100
>100
94.62
5.66
51.25
3.99
3.25
0.50
2.15
39.77
MSA
1.28
3.43
>100
1.79
3.84
>100
1.40
Based on all the synthetic data and antitumour preliminary activity evaluation, the following
criteria were established to select the most promising structures to proceed to their thermal
characterization and preliminary stability studies:
-
Synthetic yield has to be higher than 40% as lower yields might present limitations to a future
production scale-up.
-
High activity against cancer cells must be exhibited.
Taking into account these criteria, the structure VA4J was chosen as the most promising one of the
newly studied structures with a yield of 53%, GI50 value against MCF-7 cells under 10
of 51.25.
µM and an SI
Derivative VA7J was selected from series
7 and 8 with 82% yield and the most potent activity
profile. Even though the MTT assay conditions were slightly different in this work, a clear leader can
be established. Compound VA7J was selected from this group of structures as it shows IC50 values
under 10 µM in all six cell lines even at short time points.
Focusing on the compounds remaining of series 1, 2, 3, 4, 5 and 6, compounds VA6O and VA5P
were selected in the referenced work as the most potent ones. Nevertheless, VA6O and VA5P had
to be dismissed for this study as VA6O has an oily physical appearance and VA5P showed fusion
with decomposition in a preliminary test. Therefore, we followed the criteria of SI, potency and
potential safety, to select the representative compound for these structures. According to these criteria,
derivatives VA2E and VA6E were chosen. Compound VA2E did not meet the yield condition with a
5% yield, so it was ruled out. Finally, the compound VA6E was selected as it has a reasonable yield,
high potency and it is fourteen times more selective than MSA [28].
2
.2. Compound Thermal Characterization
The combination of several techniques is necessary in order to properly confirm polymorphism
presence and thermal stability. The methodology employed to perform the complete thermal
characterization of a compound include IR spectroscopy, powder X-ray diffractometry (PXRD) and
thermal methods, especially differential scanning calorimetry (DSC) and thermogravimetry (TG).
2
.2.1. Compound VA6E
The DSC results of compound VA6E are shown in Figure 2a and Table 2. During the first thermal
scan (curve A), three processes were observed. The first corresponds with a typical endothermic
process of the original product (Form I), followed by a recrystallization process to another form (Form
II), which then melts. After cooling, in the sample—through a polymorphic phase transition—Form
II changes into another form which shows fusion processes (Form III, curve B). In the third scan a
new endothermic process happened at a lower temperature (curve C), with the appearance of a new
polymorph (Form IV). The complete thermogram (Figure 2b) shows that other processes at another
temperature (higher or lower) can be excluded.

Molecules 2017, 22, 1314
5 of 22
(
a)
(
b)
Figure 2. DSC and TG of compound VA6E (
a,
b
) respectively. (a) Curve A: first scan (fusion Form I and
Form II); curve B: second scan (fusion of Form III) and curves C and D: third and fourth scan (fusion of
◦
Form IV); (b) TG of compound VA6E at 10 C/min.
Table 2. Fusion–recrystallization of compound VA6E.
Form I
Recrystallization
New Polymorphs
Scan
st
nd
rd
th
◦
−1
◦
−1
◦
−1
∆H (J g )
Tonset ( C)
∆H (J g
)
Tonset ( C)
∆H (J g
)
Tonset ( C)
1
97.2
7.9
98.5
−8.8
101.9
100.7
100.1
99.9
75.6
(Form I)
(Form II)
2
-
-
-
-
-
-
-
-
-
-
-
-
80.3
76.7
76.0
(Form III)
3
(
Form IV)
Form IV)
4
(

Molecules 2017, 22, 1314
6 of 22
Marked differences were found in the positions of the peaks between the diffractograms of the
different forms. In Figure 3, the diffractogram in black corresponds to the sample after melting and
the subsequent recrystallization. Some of the peaks matched with those from the original sample
(
(
diffractogram in red). However, other peaks only appeared after melting and recrystallization
2θ = 6.7 , 13.4 , 20.3 , 22.8 , 29.2 , 33.9 and 34.1 ). In this case, we think Form I has partially
◦
◦
◦
◦
◦
◦
◦
transformed into another polymorphic form (Form II), while preserving part of the original form.
Figure 3. X-ray diffractograms for compound VA6E. Form I shows the diffractogram of the original
polymorph and Form I + II, the one obtained after one cycle of melting-recrystallization.
Figure 4 shows the X-ray diffractograms after different and consecutive melting-recrystallization
cycles. As discussed before, after one cycle, a mixture of Forms I and II can be observed. The following
cycle provoked the appearance of a new polymorphic form (Form III), but the original Forms I and
◦ ◦
II are also present (diffractogram in red) as the match of the position of the peaks 2θ = 20.3 , 20.8 ,
◦
◦
◦
θ = 20.4 that was not
2
3.3 , and 24.9 shows. Form III (diffractogram in blue) showed a peak at 2
detected in Forms I and II, as well as the absence of other signals present in Forms I and II (peaks
◦ ◦
θ = 7.2 and 34.8 ). Finally, Form IV was observed after another cycle (diffractogram in green).
at 2
The differences between the resulting form from this last cycle and the previous ones are highlighted
◦
◦
◦
in yellow dotted lines. Peaks at 2
θ
= 20.4 , 29.8 and 30.8 are present in Form III but not in the last
◦
◦
polymorphic form. Peaks at 7.2 and 34.8 , present in Form I + II, do not appear in Form IV. After
successive heating–cooling processes the same Form IV was isolated. Taken together, the results of the
X-ray diffraction indicate that all four forms observed by DSC were indeed distinct.

Molecules 2017, 22, 1314
7 of 22
Figure 4. X-ray diffractograms of the forms of VA6E obtained after successive cycles of
melting–recrystallization.
2
.2.2. Compound VA4J
DSC results (Figure 5a and Table 3) show that the compound VA4J does not alter its thermal
behaviour after an initial fusion–recrystallization cycle. Under these conditions there is no evidence
of polymorphism. During the first thermal curve of compound VA4J (curve A in Figure 5a), a classic
endothermic process of Form I was discerned (Table 3), typical of a melting process. After cooling
the sample, the melted form recrystallized in the same form, as indicated by the same endothermic
process in curve B.
(
a)
Figure 5. Cont.

Molecules 2017, 22, 1314
8 of 22
(
b)
Figure 5. DSC (a) and TG (b) of compound VA4J. (a) Curve A: first scan (fusion of Form I) and curve B:
◦
second scan (fusion of recrystallized form); (b) TG of compound VA4J at 10 C/min.
Table 3. Fusion–recrystallization of compound VA4J.
◦
−1
)
Tonset ( C)
∆H (J g
1
st scan
119.84
119.59
93.58
93.91
Form I
Form I
2
nd scan
To further confirm the absence of polymorphism, both samples were analysed by using PXRD.
An excellent match of the position of the peaks in both diffractograms was observed (see peaks at 2
θ
◦
◦
◦
◦
◦
◦
◦
◦
at 11.5 , 11.9 , 20.9 , 24.1 , 26.2 , 27.6 , 29.3 and 33.0 in Figure 6). The differences in the relative
intensities of some of the peaks were probably due to some preferred orientation in the samples
derived from the absence of random orientation of crystal grains in space.
Figure 6. X-ray diffractograms of the original form of VA4J (in black) and those obtained after successive
cycles of melting–recrystallization (in red).

Molecules 2017, 22, 1314
9 of 22
2
.2.3. Compound VA7J
In the first heating scan, the original form of compound VA7J melts with desolvation.
A new endothermic fusion of the anhydrous product generated confirms the polymorphism
pseudopolymorphism or phase transition of the hydrated form to the corresponding anhydride)
(
of this compound (curve A, Figure 7a). After cooling the sample, it can be observed (curve B) that the
subsequent endothermic process occurs at the same temperature (curve B, Figure 7a, Table 4).
(
a)
(
b)
Figure 7. DSC (
second scan; (b) TG of compound VA7J at 10 C/min.
a
) and TG (
b
) of compound VA7J. (
a
) Curve A: first scan (desolvation) and curve B:
◦

Molecules 2017, 22, 1314
10 of 22
Table 4. Fusion–recrystallization of compound VA7J.
◦
−1
)
Behaviour III
st scan
nd scan
Tonset ( C)
∆H (J g
1
2
73.8
18.2
−88.4
Form I
Form II
1
18.7
2
219.7
17.4
Form II
Two different compounds were detected in the sample VA7J by using PXRD (Figure 8). Several
differences were found between the peak positions in the raw sample (diffractogram in black)
and the sample after one desolvation/phase transition process (diffractogram in red). Peaks at
◦
◦
◦
◦
◦
◦
◦
◦
2
(
θ
= 19.3 , 21.2 , 22.1 , 23.8 , 24.7 , 26.6 , 28.4 and 32.7 were present in the original compound
Form I) but disappeared in Form II of anhydre compound. This confirms the appearance of a new
polymorphic form.
Figure 8. X-ray diffractograms of the forms of VA7J obtained after successive cycles of
melting-recrystallization.
The diffractogram collected after a melting scan and a subsequent recrystallization process of
Form II of the compound (diffractogram in green) is very similar to the one obtained from the first
cycle (in red). Therefore, a new polymorphic form is not present. In Figure 8, the majority of the
◦ ◦ ◦ ◦ ◦ ◦
θ position (see peaks at 2θ = 21.9 , 23.4 , 26.1 , 26.4 , 27.9 and 31.2 )
diffraction peaks match in the 2
and the observed differences between both diffractograms are in terms of the relative intensities.
2
.3. Compound Stablity Tests
2
.3.1. Thermal Stability
Thermogravimetric analyses were performed to evaluate the thermal stability of the compounds
(
Figures 2b, 5b and 7b, Table 5. The TG analysis of the compounds VA4J and VA6E show that the
◦
◦
decomposition or desolvation processes in the temperature range studied (100–135 C and 94–108 C,
respectively) can be excluded. On the other hand, the thermogravimetric curve of the compound VA7J
shows a debromhidration process at 173 C, and when further heated, decomposition of desolvated
◦
◦
compound from 228 C.

Molecules 2017, 22, 1314
11 of 22
Table 5. Thermal data obtained by DSC and TG.
DSC
TG
Ref.
R
◦
−1
◦
Scan
To/ C
∆H /J g
Behaviour
To/ C
f
Series B
VA2B
VA4B
SH
1st
1st
-
-
-
V
V
263.3
103.5
SCH3
-
Series C
1
st
175.5
-
91.3
-
VA1C
SeH
I
218.4
2
nd
1
st
178.8
178.6
85.9
86.6
VA2C
VA4C
SH
III
V
224.7
131.8
2
nd
SCH3
1st
-
-
Series D
VA1D
VA6D
VA8D
SeH
SCN
1st
1st
1st
-
-
-
V
V
*
244.7
205.0
193.1
-
SC(NHNH2)
196.7
−90.3
Series E
1
st
148.8
-
34.0
-
I
VA2E
VA6E
SH
266.2
184.1
2
nd
97.2
7.8
−8.8
75.6
80.3
76.7
76.0
1
st
98.4
1
01.9
100.7
100.1
99.9
SCN
IV
2
nd
3
4
th
th
1
st
2nd
th
183.8
177.2
-
21.5
28.1
-
VA7E
VA8E
SeC(NHNH2)
SC(NHNH2)
II
I
205.2
215.1
3
1
st
196.1
208.5
-
33.8
−37.4
-
2
nd
Series F
VA7F
VA8F
SeC(NHNH2)
SC(NHNH2)
1st
1st
214.4
−63.6
−64.6
*
*
214.7
190.8
198.3
Series G
-
VA6G
VA7G
VA8G
SCN
1st
1st
1st
-
-
-
V
V
V
238.3
198.2
243.7
SeC(NHNH2)
SC(NHNH2)
-
-
Series H
1
st
54.9
-
75.7
-
VA4H
VA6H
VA7H
VA8H
SCH3
I
I
I
I
105.6
139.1
145.8
190.2
2
2
2
2
nd
1
st
78.9
-
116.6
-
SCN
nd
1
st
113.7
-
42.0
-
SeC(NHNH2)
SC(NHNH2)
nd
1
st
176.2
-
49.3
-
nd

Molecules 2017, 22, 1314
12 of 22
Table 5. Cont.
DSC
∆H /J g
TG
Ref.
R
◦
−1
◦
Scan
To/ C
Behaviour
To/ C
f
Series I
1
st
234.9
-
110.8
-
VA6I
VA7I
VA8I
SCN
I
*
*
246.9
226.4
272.4
2
nd
1
st
st
230.1
30.1
−14.4
39.6
−36.7
SeC(NHNH2)
SC(NHNH2)
2
34.4
263.2
65.4
1
2
Series J
VA2J
VA4J
VA6J
SH/SC(NHNH2)
SCH3
1st
206.5
−63.9
93.58
93.91
*
199.2
119.84
197.42
I
1
st
2nd
st
119.59
167.6
173.8
SCN
−227.0
*
165.0
1
−88.4
218.2
18.5
VA7J
VA8J
SeC(NHNH2)
SC(NHNH2)
IV
I
From 228
2
3
nd
th
219.7
219.6
17.4
17.4
1
st
222.9
224.1
-
31.3
−95.3
-
251.05
2
nd
Series M
1
st
113.6
110.2
68.1
70.7
VA2M
VA6M
SH
III
I
166.0
164.8
2
nd
1
st
143.3
-
50.8
-
SCN
2
nd
1
st
211.6
34.4
−68.7
4.4
10.3
14.8
216.0
VA7M
VA8M
SeC(NHNH2)
SC(NHNH2)
220.0
219.7
219.8
IV
I
255.8
233.1
2
2
2
nd
th
3
1
st
228.7
-
116.7
-
nd
Series O
1
st
67.7
65.6
76.0
66.2
VA5O
VA7O
SeCN
III
V
141.7
185.9
nd
SeC(NHNH2)
1st
st
2nd
rd
-
-
1
194.7
199.8
-
21.1
45.8
-
VA8O
SC(NHNH2)
II
201.8
3
Series P
VA5P
VA7P
SeCN
1st
1st
161.3
79.6
*
*
*
149.4
175.8
199.8
SeC(NHNH2)
180.2
−84.7
1.7
−60.4
1
st
200.6
210.1
VA8P
SC(NHNH2)
2
nd
*
fusion with decomposition.
2
.3.2. Stability after Stress Conditions
The stability of compounds VA4J (without polymorphic behaviour) and VA6E and VA7J (with
polymorphic behaviour) was investigated after the application of diverse stress conditions: acid

Molecules 2017, 22, 1314
13 of 22
hydrolysis (HCl 1 M), basic hydrolysis (NaOH 0.1 M) and oxidation (H O 3%). Chromatographic
2
2
studies using HPLC/UV-DAD were performed with this purpose.
During calibration sample analysis, compound VA4J had a retention time of 4.9 min, compound
VA6E of 7.3 min (for both Form I and IV), and compound VA7J of 1.1 min (for both form I and II).
The linearity showed an adequate relationship between the instrumental response (peak areas) and
the respective sample concentration (x). Intra-day precision (Relative Standard Deviation (RSD) < 7%
was satisfactorily obtained in the ranges evaluated for the three compounds (Table 6). When new
peaks appeared after stress conditions, they were adequately separated from the parent peak.
)
Table 6. Limit of Quantification (LOQ), linearity data and within-day precision.
tR
min)
LOQ
(µg/mL)
Range
(µg/mL)
Curve
Equation
Slope Interval
(p = 95%)
Within-Day Precision
r2
(
[
µg/mL]
%RSD (n = 3)
VA4J
VA6E
VA7J
4.9
7.3
1.1
0.8
3.0
0.8
0.8–59.5
3.0–119.0
0.8–59.5
y = 97.7x − 51.1
y = 26.2x − 59.2
y = 97.5x + 37.1
0.999
96.0–99.9
25.3–27.1
94.9–100.0
0.8
5.6
5.5
4.6
2.6
0.5
1
6
.0
.0
2
5
3.8
9.5
0.996
0.999
3.0
.0
0.8
2.4
3.9
1.5
0.7
6
1
4
1.9
7.6
1
19.0
0.8
6.5
3.9
10.5
1.7
1
6
.0
.0
2
5
3.8
9.5
0.9
Compound VA6E (Form I and IV)—similar stability for the two polymorphs was found after
application of different stress conditions (Figures 9 and 10). The highest degradation occurred in the
alkaline medium, as both forms degraded almost completely after 4 h of exposure to NaOH. Although
the stability was higher, degradation product peaks were also detected after acid hydrolysis (t 3.0 min)
R
and oxidation (t 3.0, 3.5 and 3.8 min).
R
Figure 9. VA6E (form I and IV) chromatogram before and after 24 h application of diverse stress
conditions: acid and alkaline hydrolysis and oxidation.

Molecules 2017, 22, 1314
14 of 22
Figure 10. Stability of compound VA6E (polymorph I and IV) after 0, 4 and 24 h of exposition to HCl,
NaOH and H O .
2
2
Compound VA4J—The compound was stable after all the stress conditions studied. The VA4J
chromatogram before and after 24 h application of acid and alkaline hydrolysis and oxidation showed
no change in the peak area at 4.9 min t and no new peaks were detected at other retention times (data
R
not shown).
Compound VA7J—The compound was stable to acid hydrolysis and only a minor degradation
due to alkaline conditions was detected. However, it totally degraded in the presence of H O , as no
2
2
peak was detected at its retention time at zero time of exposure to oxidation (Figure 11). A new
chromatographic peak was detected at t of 1.4 min, corresponding to an oxidized form of VA7J,
R
probably an N-oxide derivative.
Figure 11. Stability of compound VA7J (polymorph I and II) after application of stress conditions.
2
.4. Polymorphic Behaviours
To establish general polymorphic behaviours, samples of all the compounds synthetized (Figure 1)
were exposed to successive cycles of melting–recrystallization using DSC. Calorimetric analyses were
performed before the beginning of the degradation process (Table 6). The analysis of the thermal data
revealed interesting calorimetric performance for some of these compounds. The results showed that
there are five types of calorimetric behaviours:
•
Behaviour I—Compounds that show a polymorphic phase transition into an amorphous or mainly
amorphous solid form after a first melting–cooling scan. This behaviour was shown by several
compounds: VA1C, VA2E, VA8E, series H, VA6I, VA8J, VA6M and VA8M. DSC and PXRD results
of compound VA7H are shown as supplementary material (Figures S1 and S2 and Table S1).

Molecules 2017, 22, 1314
15 of 22
•
•
Behaviour II—Compounds that change into a new polymorphic form with a different Tonset
of fusion than the one observed at the first melting–recrystallization scan after cooling. These
compounds are VA7E and VA8O. DSC and PXRD results of compound VA7E are shown as
supplementary material (Figures S3 and S4, Table S2).
Behaviour III—These compounds show a fusion process without modifications in the thermal
behaviour after an initial fusion–recrystallization cycle. Consequently, there is no evidence of
polymorphic behaviour. The compounds VA2C, VA2M, VA4J and VA5O belong to this group.
DSC and PXRD results of compound VA2M are shown as supplementary material (Figures S5
and S6, Table S3).
•
•
Behaviour IV—Compounds that show recrystallization or other types of phase transition of
the original form into another crystalline one in the first heat scan. In the successive cycles of
melting–recrystallization, the new form may not alter its crystalline form or transit to another
polymorphic form. Compounds VA6E, VA7J and VA7M show this behaviour.
Behaviour V—Compounds without fusion processes at the range of temperatures studied. The
original compound shows an amorphous form that does not recrystallize into any crystalline
form in the successive melting–recrystallization cycles. The compounds that showed this thermal
behaviour are VA2B, VA4B, VA4C, VA1D, VA6D, series G, and VA7O.
Polymorphic behaviour is affected by reaction conditions and purification processes, as they
include temperature modifications that can lead to modifications in the final solid state structure.
General patterns are therefore difficult to extract as these factors vary among the 12 series.
The main skeleton of the structure differs in each of these series and it has a crucial effect on the
internal organization.
Some general patterns within each series of compounds can be found. All the derivatives from
series B, D and G share the same behaviour (V), those compounds do not suffer transformations
due to heating as they only exhibit one polymorphic form (except for compound VA8D that melts
with decomposition). In the same line, compounds of series H exhibit behaviour I, which leads us
to think that the conditions for the original crystalline structure to transform to the polymorph were
probably not met during the synthetic process. Therefore, the structural nucleus is the main factor that
determines internal organization in this series.
Interestingly, the polymorphic behaviour of some of the series changes depending on the
substituent. For example, series E shows three different polymorphic behaviours depending on the
attached substituent in position 5. Therefore, the substituents play an important role in the calorimetric
performance of the resulting derivatives. However, no general patterns could be established in terms
of the effect of substituents on internal organization, probably because of the strong effect that main
nucleus modifications have among them. Thus, each series should be evaluated independently.
2
.5. Comparative Thermal Stability
The values of thermal degradation temperatures obtained for the S and Se heteroaryl compounds
are also shown in Table 6. Thermal stability is affected by intramolecular interactions and electron
delocalization inside the core of the structure, as well as by interactions that might exist between the
core and the substituents.
A comparison between compounds substituted with the same chemical group is hard to establish
since only compounds 6, 7 and 8 are present in most of the mentioned series. Moreover, the variability
of the chemical nature of the heterocycles used enables us to observe only general trends in the analysis.
◦
For example, quinolone-based structures (series H) are less stable with Tonset values of 139.1 C,
◦
◦
145.8 C and 190.2 C for the corresponding derivatives
6,
7
and
8
, respectively. On the other hand,
the structures of series I are among the most stable with Tonset over 220 C in all three cases and ranking
among the two most stable compounds in derivatives 6, 7 and 8.
◦

Molecules 2017, 22, 1314
16 of 22
Patterns relating to the substituent nature and thermal stability of the compounds can be spotted
when focusing on those compounds with a common core structure (Series B, C, D, E, F, G, H, I, J, M, O
and P).
Stability is enhanced in structures with the ability to establish hydrogen intermolecular
interactions. Overall, SH (compound
2) or SeH (compound 1) groups improve stability dramatically.
If we look at other substituents that can establish hydrogen interactions such as –Se–C(NH )=NH or
2
–
S–C(NH )=NH, we can also observe an increase in the thermal stability. Although compounds of
2
series 7 and 8 are homologs except for the presence of Se or S atoms, S derivatives are more stable than
Se-substituted compounds in general.
Considering the other substituents, compounds with a nitrile (CN; compounds
5 and 6) or methyl
(
CH ; compounds ) group showed lower Tonset values of degradation, with this effect being more
4
3
pronounced for the methyl group. Methyl-substituted structures might be less stable because they do
not have the ability to establish either hydrogen interaction.
3
. Materials and Methods
3
.1. Chemistry
1
13
Proton ( H) and carbon ( C) NMR spectra were recorded on a Brüker 400 Ultra-shield
™
spectrometer (Brüker, Rheinstetten, Germany) using DMSO-d or CDCl as solvent. IR spectra
6
3
were recorded on a Thermo Nicolet FT-IR Nexus spectrophotometer (Thermo Nicolet, Madison,
WI, USA) using KBr pellets for solid samples or NaCl plates for oil compounds. Elemental analysis
was performed on a LECO CHN-900 Elemental Analyzer (LECO, Saint Joseph, MI, USA). The purity
of all final compounds was 99% or higher. Chemicals were purchased from E. Merck (Darmstadt,
Germany), Panreac Química S.A. (Barcelona, Spain), Sigma-Aldrich Quimica, S.A. (Madrid, Spain)
and Acros Organics (Janssen Pharmaceuticalaan, Geel, Belgium). The synthesis of 34 out of the
4
0 compounds analysed in this work has been previously described [28,29].
3
.1.1. General Procedure for the Synthesis of Compounds VA6G, VA6H, VA6I and VA6J
The synthesis (Scheme 1) was carried out from the corresponding alkyl-halide reagent and
potassium-thiocyanate in acetone at reflux, as previously described. The corresponding halide reagent
1 mmol) was added to a mixture of 1.2 mmol of potassium thiocyanate in dry acetone (20 mL).
(
The mixture was stirred at reflux for 3.5–4 h and the product was isolated by filtration after the
addition of 50 mL of water and purified by washing or recrystallization.
Scheme 1. Reaction scheme to yield compounds VA6G, VA6H, VA6I and VA6J.
3
.1.2. 3-Mercaptoquinoxalin-2-yl Carbamimidothioate Hydrochloride (VA2J)
To obtain compound VA2J, 2,3-dichloroquinoxaline was treated with thiourea in a 1:2.2
(reagent/thiourea) molar ratio (Scheme 2) at reflux for 4 h using absolute ethanol as solvent (20 mL).
The product was purified by recrystallization from methanol. An orange powder was obtained. Yield:
◦
1
4
2%; melting point (m.p.): 197–198 C (direct combustion). H-NMR (400 MHz, DMSO-d ):
δ
7.54
6
(
t, 1H, J = 7.3 Hz, H6), 7.66–7.72 (m, 2H, H5 + H7), 7.90 (d, 1H, J = 8.1 Hz, H8), 9.76 + 9.95 (bs + bs, 4H,
13
NH + NH + HCl), 15.13 ppm (bs, 1H, SH). C-NMR (100 MHz, DMSO-d6):
δ
117.4 (C6), 127.4 (C7),
2
1
28.7 (C5), 131.9 (C8), 132.0 (C4a), 135.6 (C8a), 160.7 (C3), 165.4 (C2), 171.4 ppm (S–C–(NH)(NH )).
2
−
1
IR (KBr): νe 3380–3200 (s; N–H, N–H ), 1625 cm (s; C=N). MS [m/z (% abundance)]: 194 (100), 161
2

Molecules 2017, 22, 1314
17 of 22
(
16), 150 (43), 134 (25). Elemental analysis calculated (%) for C H N S2·HCl: C: 39.63, H: 3.33, N: 20.54;
9
8
4
found: C: 39.92, H: 3.61, N: 20.19.
Scheme 2. Synthesis schematics for compounds VA2J and VA4J.
3
.1.3. 2,3-Bis(methylsulfanyl)quinoxaline (VA4J)
VA4J was synthesized by adding 1 mmol of VA2J to a suspension of 3.6 mmol of potassium
hydroxide in DMSO (5 mL) and stirring for 1 h at room temperature (r.t.). Then, 4 mmol of methyl
iodide were added and stirred for additional 4 h at r.t. (Scheme 2). The reaction medium was then
poured into ice and the resultant precipitate was filtered and solved in dichloromethane (40 mL).
The insoluble fraction was discarded and the solvent was removed under vacuum by rotatory
◦
1
evaporation. A yellow powder was obtained. Yield: 53%; mp: 118–119 C. H-NMR (400 MHz,
CDCl ): 2.75 (s, 6H, (CH3) 2), 7.58–7.61 (m, 2H, H6 + H7), 7.91–7.96 ppm (m, 2H, H5 + H8).
C-NMR (100 MHz, DMSO-d6): 13.7 ((CH3) 2), 128.1 (C5 + C8), 129.4 (C6 + C7), 139.9 (C4a + C8a),
δ
×
3
1
3
δ
×
−
1
1
2
55.3 ppm (C2 + C3). IR (KBr): νe 2919 (m; C–Haliph), 1615 cm (s; C=N). MS [m/z (% abundance)]:
+
•
22 (52; M ), 207 (100), 192 (31), 175 (8), 160 (20), 102 (17). Elemental analysis calculated (%) for
C H N S ·1/2H O: C: 51.87, H: 4.76, N: 12.10; found: C: 51.63, H: 4.56, N: 12.46.
1
0
10
2
2
2
3
.1.4. (9,10-Dioxo-9,10-dihydroanthracen-2-yl)methyl Thiocyanate (VA6G)
From 2-(chloromethyl)-9,10-dioxo-9,10-dihydroanthracen. Conditions: 3.5 h at reflux;
recrystallized from ethanol. A yellow powder was obtained. Yield: 46%; m.p.: 174–175 C. H-NMR
400 MHz, DMSO-d6): 4.58 (s, 2H, –CH –), 7.93–7.96 (m, 3H, H1 + H3 + H4), 8.20–8.29 ppm (m, 4H,
H5 + H6 + H7+ H8). 13C-NMR (100 MHz, DMSO-d6):
28.0 (C1), 128.3 (C4), 133.4 (C3), 133.8 (C4a), 134.1 (C6 + C7), 135.4 (C8a + C10a), 135.7 (C9a), 144.1
◦
1
(
δ
2
δ
37.1 (–CH –), 113.7 (–SCN), 127.6 (C5 + C8),
2
1
−
1
(
2
C2), 183.0 ppm (C9 + C10). IR (KBr): νe 2149 (m, C
≡
N), 1671 cm (s, C=O). MS [m/z (% abundance)]:
+
•
79 (5; M ), 221 (100), 193 (51), 165 (27). Elemental analysis calculated (%) for C H NO S: C: 68.80,
16 9 2
H: 3.25, N: 5.01; found: C: 68.59, H: 3.31, N: 4.82.
3
.1.5. Quinolin-2-ylmethyl Thiocyanate (VA6H)
The commercially available 2-(chloromethyl)quinoline hydrochloride was treated with basic water
50 mL) in order to obtain the free base, which was used for the synthesis of VA6H. Conditions: 4 h at
reflux. The precipitate was washed with ethyl ether (20 mL). A brown powder was obtained. Yield:
(
◦
1%; m.p.: 80–81 C. H-NMR (400 MHz, DMSO-d ): δ 4.74 (s, 2H, –CH –), 7.63–7.66 (m, 2H, H3 + H6),
6 2
1
4
7
13
.80 (t, 1H, J = 7.8 Hz, H7), 8.01 (d, 2H, J = 8.6 Hz, H5 + H8), 8.44 ppm (d, 1H, J = 8.5 Hz, H4).
C
NMR (100 MHz, DMSO-d6):
δ
40.0 (–CH –), 113.9 (–SCN), 121.8 (C3), 127.8 (C6), 127.9 (C4a), 128.8
2
(
(
1
C5), 129.3 (C8), 131.0 (C7), 138.2 (C4), 147.8 (C8a), 156.4 ppm (C2). IR (KBr): νe 3058 (m; C–Harom), 2946
42 (100), 128 (9), 115 (47). Elemental analysis calculated (%) for C H N S 1/2H O: C: 63.08, H: 4.30,
2
− +•
1
m; C–Haliph), 2147 (s; C N), 1655 cm (m; C=N). MS [m/z (% abundance)]: 200 (100; M ), 174 (6),
≡
·
11
8
2
N: 13.99; found: C: 62.96, H: 4.42, N: 13.15.
3
.1.6. (6,7-Dimethoxy-2-oxo-2H-chromen-4-yl)methyl Thiocyanate (VA6I)
From 4-(bromomethyl)-6,7-dimethoxy-2-oxo-2H-chromene. Conditions: 4 h at reflux. The product
was washed with ethyl ether and dichloromethane (20 mL of each). A white powder was obtained.
◦
1
Yield: 70%; m.p.: 204–205 C. H-NMR (400 MHz, DMSO-d ):
δ
3.83 + 3.88 (s, 6H, (–OCH3)
×
2), 4.56
6
13
(
s, 2H, –CH –), 6.41 (s, 1H, H3), 7.13 (s, 1H, H5), 7.38 ppm (s, 1H, H8). C-NMR (100 MHz, DMSO-d ):
2
6

Molecules 2017, 22, 1314
18 of 22
δ
33.6 (–CH –), 57.1 + 57.3 ((–OCH )
×
2), 101.5 (C8), 107.2 (C5), 110.0 (–SCN), 113.3 (C3), 114.0 (C4a),
2
3
1
46.9 (C6), 150.4 (C8a), 150.7 (C7), 153.8 (C4), 160.8 ppm (C2). IR (KBr): νe 3065 (w; C–Harom), 2999–2849
27), 191 (69), 147 (18). Elemental analysis calculated (%) for C H NO S: C: 56.31, H: 4.00, N: 5.05;
13 11 4
− +•
1
(
m; C–Haliph), 2154 (s; C N), 1711 cm (C=O). MS [m/z (% abundance)]: 277 (100; M ), 234 (6), 219
≡
(
found: C: 56.40, H: 3.98, N: 4.88.
3
.1.7. Quinoxalin-2,3-diyldimethanediyl Bisthiocyanate (VA6J)
From 2,3-bis(bromomethyl)quinoxaline. Conditions: 4 h. at reflux. The product was washed with
◦
ethyl ether (20 mL). A brown powder was obtained. Yield: 88%; m.p.: 158–159 C (direct combustion).
1
H-NMR (400 MHz, DMSO-d ):
δ
4.98 (s, 4H, (–CH –)
×
2), 7.91–7.93 (m, 2H, H6 + H7), 8.11–8.14 ppm
37.7 ((–CH –) 2), 112.0 ((–SCN) 2), 128.4
6
2
(
(
(
m, 2H, H5 + H8). 13C-NMR (100 MHz, DMSO-d6):
δ
×
×
2
C5 + C8), 128.4 (C6 + C7), 143.7 (C4a + C8a), 160.3 ppm (C2 + C3). IR (KBr): νe 2926 (m; C–Haliph), 2155
− +•
1
s; C N), 1563–1485 cm (m; C–Carom). MS [m/z (% abundance)]: 272 (36; M ), 245 (14), 214 (100),
≡
1
87 (15), 170 (13), 155 (30), 129(13). Elemental analysis calculated (%) for C H N S : C: 52.92, H: 2.96,
12 8 4 2
N: 20.57; found: C: 52.7, H: 3.07, N: 20.77.
3
.2. Biological Evaluation
The antitumoural evaluation of 34 out of the 40 compounds analysed in this work has been
previously described [28,29].
3
.2.1. Cell Culture
Cell lines were purchased from the American Type Culture Collection (ATCC, Barcelona, Spain).
MCF7 cell lines were grown in Roswell Park Memorial Institute (RPMI) medium (Gibco, Madrid, Spain)
supplemented with 10% fetal bovine serum (FBS; Gibco), 100 units/mL penicillin and 100 mg/mL
streptomycin (Gibco, Madrid, Spain). 184B5 cells were grown in DMEM/F12 medium supplemented
with 5% FBS, 1x ITS (Lonza, Barcelona, Spain), 100 nM hydrocortisone (Sigma Aldrich, Madrid, Spain),
2
mM sodium pyruvate (Lonza), 20 ng/mL EGF (Sigma Aldrich, Madrid, Spain), 0.3 nM trans-retinoic
acid (Sigma-Aldrich) 100 units/mL penicillin and 100 mg/mL streptomycin. Cells were maintained at
◦
3
7 C and 5% CO .
2
3
.2.2. Cell Viability
Compounds were dissolved in DMSO at a concentration of 0.01 M. Serial dilutions were prepared
with non-supplemented medium. The cytotoxic effect of each compound was tested at eight different
concentrations ranging between 0.01 and 100 µM.
4
A total of 10 cells/well in 96 well-plates were treated with either DMSO or increasing
concentrations of the corresponding compound for 72 h and cell viability was determined using
the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) (Sigma Aldrich, Madrid,
Spain) assay [33]. Briefly, cells were incubated with 50
The medium was then removed by aspiration and the formazan crystals were dissolved in 150
DMSO. The absorbance was measured at 550 nm in a microplate reader (Sunrise, Tecan, Männedorf,
Switzerland). Results are expressed as GI , the concentration that reduces by 50% the growth of
µL of MTT (2 mg/mL stock) for 4 h.
µL of
50
treated cells with respect to untreated controls, TGI, the concentration that completely inhibits cell
growth, and LC , the concentration that kills 50% of the cells. Data were obtained from at least three
50
independent experiments performed in quadruplicate.
3
.3. Thermal Analysis
The thermogravimetric (TG) studies were carried out with a Perkin–Elmer TGA-7 (Perkin Elmer,
◦
−1
min . The calibration
Inc. Watham, MA, USA) which was calibrated with alumel and nickel at 10 C
·
of the oven temperatures was carried out automatically and mass calibrated with a certified sample of

Molecules 2017, 22, 1314
19 of 22
1
0 mg (ASTM E617, provided by Perkin Elmer, Inc. Watham, MA, USA). TG analyses were performed
−
1
◦
under a nitrogen atmosphere with a gas flow of 40 mL
1
Tmax of the degradation process as well as any associated mass loss.
·
min , from 50 to 350 C, at a heating rate of
◦
−1
0 C min , using a sample of approximately 3 mg, in order to obtain the values of the Tonset and
·
For calorimetric studies a Perkin–Elmer DSC diamond calibrated with indium and zinc (provided
by Perkin–Elmer (Perkin Elmer, Inc. Watham, MA, USA) and fabricated according to guideline ISO35)
◦
−1
−1
min was employed. The gas connected to the equipment
at 10 C
·
min and nitrogen flow of 20 mL
·
was nitrogen with a purity of 99.999%. Calorimetric analyses were carried out in aluminium capsules
for volatilities of 10
in order to establish the Tonset, Tmax and the enthalpy of fusion ∆H .
◦
−1
µ
L, at a heating rate of 10 C min , using a sample of approximately 3 mg,
·
f
The calorimetric analyses started with the study of the thermal behaviour of the compounds
before the process of degradation in order to evidence the possible existence of polymorphism. For this
preliminary study, samples of all the compounds were exposed to successive cycles of heating–cooling.
◦
◦
All samples were heated from 50 C to temperatures 15–20 C below Tonset of the degradation process,
in order to ensure that compounds were not degraded (Table 6). After melting the samples, they were
left at room temperature for enough time to allow the compounds to recrystallize before the successive
thermal process. Additional cycles of heating–cooling were performed if polymorphism was detected.
3
.4. Powder X-ray Diffraction
The X-ray diffraction experiments were performed by using a Bruker D8 Advance diffractometer,
◦
◦
◦
θ < 40 . The equipment
under the conditions of a step size of 0.02 and 1 s time/step, from 5 < 2
consisted of an X-ray generator with a Cu anode (radiation Cu K = 1.54 Å, 40 kV, 30 mA) and a
α
scillintator detector. Experiments were performed in reflectance mode. Measurements were carried out
on samples laid on a glass slide without additional manipulation. The preferential grain orientations
that are likely to exist in these compounds could not be avoided due to the absence of grinding of the
samples and also to the impossibility of using spinning capillary in our equipment.
3
.5. Chromatography
Chromatographic studies were performed using an HPLC/UV-DAD (HP 1200, Agilent
Technologies. Santa Clara, CA, USA) with a C18 column (Gemini-NX 150
Phenomenex (Phenomenex. Torrance, CA, USA). The mobile phase was a mixture of acetonitrile/water
80:20; v/v) for compound VA4J, (50:50; v/v) for VA6E and (80:20, 0.1% formic acid; v/v) for compound
×
4.6 mm, 5 µm) from
(
VA7J. The injection volume was 50
performed at 35 C. UV-DAD detection was established at λ = 254 nm.
µ
L and the flow rate was 1 mL/min. Chromatography was
◦
In the assessment of linearity, calibration curves were plotted in the range of 3–119 µg/mL for
VA6E and 0.8–59.5 µg/mL for VA4J and VA7J. Calibration curves were evaluated by the analysis of the
2
distribution properties of the residuals, correlation coefficient r > 0.990, slope of the linear calibration
curve statistically different from 0 (p = 95%) and the intercept not statistically different from 0 (p = 95%).
Precision (as RSD%) of the method was evaluated by analysing three replicate calibration samples at
five concentration values (including lower, medium and higher value) of the calibration curve.
Stock standard solutions containing 1 mg/mL of compound VA4J, VA6E and VA7J were prepared
by diluting each compound in a mixture of acetonitrile/water (50:50; v/v). Calibration samples in the
mobile phase were prepared by the dilution of the stock standard solutions.
For stability evaluation in stress conditions, working solutions of 47.6
diluting 20 L of each standard solution in 200 L of acetonitrile and 200
NaOH and (c) 3% H O . Data were recorded at zero time and after 4 and 24 h of exposure.
µg/mL were prepared by
µ
µ
µL of (a) 1 M HCl, (b) 0.1 M
2
2
4
. Conclusions
The thermal characterization of compounds with antitumour potency is crucial to further continue
the drug discovery process. In order to predict the non-desired transformations that might occur in

Molecules 2017, 22, 1314
20 of 22
the pharmaceutical manufacturing process, the polymorphism of the most promising compounds has
been studied.
Regarding the results of differential scanning calorimetry, powder X-ray diffraction and the
stability studies under stress conditions:
•
Compound VA6E: temperature conditions have to be carefully selected as solid state behaviour
includes four different polymorphs. In terms of stability, alkaline, acid and oxidation media are
not suggested as they cause degradation.
•
•
Compound VA7J: oxidation conditions and desolvation have to be taken into account as they lead
to immediate degradation and crystal structure modification, respectively.
Compound VA4J: there is no evidence of polymorphism and stress conditions do not affect the
solid-state structure.
Taken together we can conclude that compounds VA6E and VA7J present some obstacles to
further development as drugs as both temperature and stress conditions affect the original structure.
On the other hand, compound VA4J proved to be safe to work with under the conditions studied.
Therefore, this compound is of special interest for further development as it shows promising activity,
selectivity and stability.
In addition, it has been found that the polymorphic behaviour as well as the thermal stability of
these compounds can be modified by the introduction of different functional groups in the heterocycle
under consideration. According to our data, the incorporation of a selenol or thiol group improves the
thermal stability.
Supplementary Materials: supplementary material are available online.
Acknowledgments: The authors wish to express their gratitude to the Plan de Investigación de la Universidad de
Navarra, PIUNA (Ref 2014-26), for financial support for the project. V. Alcolea and P. Garnica wish to express their
gratitude to the Asociación de Amigos de la Universidad de Navarra for the pre-doctoral fellowship. The research
leading to these results has received funding from “la Caixa” Banking Foundation.
Author Contributions: Carmen Sanmartín, Elena Lizarraga and Juan Antonio Palop conceived and designed
the experiments; Verónica Alcolea performed the synthesis of the compounds; Pablo Garnica performed the
biological evaluation, Elena Lizarraga the experiments in thermal analysis; Adrian Durán the experiments in
X-ray diffraction and Elena González Peñas designed and performed the experiments in HPLC. Elena Lizarraga,
Elena González Peñas, Adrián Durán, Pablo Garnica, Verónica Alcolea and Carmen Sanmartín analyzed the data;
all the authors wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
World Health Organization. Cancer Fact Sheet February 2017; World Health Organization: Geneva,
Switzerland, 2017.
Siegel, R.L.; Miller, K.D.; Jemal, A. Cancer statistics, 2017. CA Cancer J. Clin. 2017, 67, 7–30. [CrossRef]
[PubMed]
Holohan, C.; Van Schaeybroeck, S.; Longley, D.B.; Johnston, P.G. Cancer drug resistance: An evolving
paradigm. Nat. Rev. Cancer 2013, 13, 714–726. [CrossRef] [PubMed]
Miller, K.D.; Siegel, R.L.; Lin, C.C.; Mariotto, A.B.; Kramer, J.L.; Rowland, J.H.; Stein, K.D.; Alteri, R.; Jemal, A.
Cancer treatment and survivorship statistics, 2016. CA Cancer J. Clin. 2016, 66, 271–289. [CrossRef] [PubMed]
Wallenberg, M.; Misra, S.; Bjornstedt, M. Selenium cytotoxicity in cancer. Basic Clin. Pharmacol. Toxicol. 2014
14, 377–386. [CrossRef] [PubMed]
,
1
Fernandez-Herrera, M.A.; Sandoval-Ramirez, J.; Sanchez-Sanchez, L.; Lopez-Muñoz, H.; Escobar-Sanchez, M.L.
Probing the selective antitumour activity of 22-oxo-26-selenocyanocholestane derivatives. Eur. J. Med. Chem.
2
014, 74, 451–460. [CrossRef] [PubMed]
7
.
Romano, B.; Plano, D.; Encio, I.; Palop, J.A.; Sanmartin, C. In vitro radical scavenging and cytotoxic activities
of novel hybrid selenocarbamates. Bioorg. Med. Chem. 2015, 23, 1716–1727. [CrossRef] [PubMed]

Molecules 2017, 22, 1314
21 of 22
8
9
.
.
Plano, D.; Sanmartín, C.; Moreno, E.; Prior, C.; Calvo, A.; Palop, J.A. Novel potent organoselenium
compounds as cytotoxic agents in prostate cancer cells. Bioorg. Med. Chem. Lett. 2007, 17, 6853–6859.
[
CrossRef] [PubMed]
Otsuka, M.; Kaneniwa, N. Dehydration of cephalexin hydrates. Chem. Pharm. Bull. 1983, 31, 1021–1029.
CrossRef]
[
1
1
0. Borka, L.; Haleblian, J.K. Crystal polymorphism of pharmaceuticals. Acta Pharm. Jugosl. 1990, 40, 71–94.
1. Takahashi, Y.; Nakashima, K.; Ishihara, T.; Nakagawa, H.; Sugimoto, I. Polymorphism of fostedil:
Characterization and polymophic change by mechanical treatments. Drug Dev. Ind. Pharm. 2008, 11,
1
543–1563. [CrossRef]
1
2. Correa, J.C.R.; Perissinato, A.G.; Serra, C.H.D.; Trevisan, M.G.; Salgado, H.R.N. Polymorphic stability of
darunavir and its formulation. J. Therm. Anal. Calorim. 2016, 123, 2185–2190. [CrossRef]
1
3. Bonfilio, R.; Pires, S.A.; Ferreira, L.M.; de Almeida, A.E.; Doriguetto, A.C.; de Araujo, M.B.; Salgado, H.R.
A discriminating dissolution method for glimepiride polymorphs. J. Pharm. Sci. 2012, 101, 794–804.
[CrossRef] [PubMed]
1
1
1
1
4. Giron, D. Thermal-analysis and calorimetric methods in the characterization of polymorphs and solvates.
Thermochim. Acta 1995, 248, 1–59. [CrossRef]
5. Du, Y.; Xue, J. Investigation of polymorphism and cocrystallization of active pharmaceutical ingredients
using vibrational spectroscopic techniques. Curr. Pharm. Des. 2016, 22, 4917–4928. [CrossRef] [PubMed]
6. Barnes, A.F.; Hardy, M.J.; Lever, T.J. A review of the applications of thermal methods within the
pharmaceutical-industry. J. Therm. Anal. 1993, 40, 499–509. [CrossRef]
7. Wang, L.Y.; Zhu, L.; Sha, Z.L.; Li, X.C.; Wang, Y.F.; Yang, L.B.; Zhao, X.Y. Characterization, thermal stability,
and solid-state phase transition behaviors of gestodene polymorphs and amorphous phase. J. Therm.
Anal. Calorim. 2017, 127, 1533–1542. [CrossRef]
1
8. Waszkielewicz, A.M.; Cegla, M.; Zeslawska, E.; Nitek, W.; Sloczynska, K.; Marona, H.
N-[(2,6-dimethylphenoxy)alkyl]aminoalkanols-their physicochemical and anticonvulsant properties.
Bioorg. Med. Chem. 2015, 23, 4197–4217. [CrossRef] [PubMed]
1
9. Jiang, S.F.; Jansens, P.J.; ter Horst, J.H. Mechanism and kinetics of the polymorphic transformation of
o-aminobenzoic acid. Cryst. Growth Des. 2010, 10, 2123–2128. [CrossRef]
2
0. Da Silva, L.M.; Montanari, C.M.; Santos, O.M.; Cazedey, E.C.; Angelo, M.L.; de Araujo, M.B. Quality
evaluation of the finasteride polymorphic forms i and ii in capsules. J. Pharm. Biomed. Anal. 2015, 105, 24–31.
[
CrossRef] [PubMed]
1. Carrer, H.; Cortez, J.; Frare, L.M.; Costa, M.B.; Bittencourt, P.R.S. Thermal characterization of the bromopride
recrystallized from different solvents and at different temperature conditions. J. Therm. Anal. Calorim. 2016
23, 927–931. [CrossRef]
2
2
,
1
2. Yoshida, M.I.; Gomes, E.C.; Soares, C.D.; Cunha, A.F.; Oliveira, M.A. Thermal analysis applied to verapamil
hydrochloride characterization in pharmaceutical formulations. Molecules 2010, 15, 2439–2452. [CrossRef]
[PubMed]
2
2
3. Huang, L.F.; Tong, W.Q. Impact of solid state properties on developability assessment of drug candidates.
Adv. Drug. Deliv. Rev. 2004, 56, 321–334. [CrossRef] [PubMed]
4. Thomas, S.P.; Satheeshkumar, K.; Mugesh, G.; Guru Row, T.N. Unusually short chalcogen bonds involving
organoselenium: Insights into the Se-N bond cleavage mechanism of the antioxidant ebselen and analogues.
Chem. Eur. J. 2015, 21, 6793–6800. [CrossRef] [PubMed]
2
2
2
5. Neumann, M.A.; Van de Streek, J.; Fabbiani, F.P.; Hidber, P.; Grassmann, O. Combined crystal structure
prediction and high-pressure crystallization in rational pharmaceutical polymorph screening. Nat. Commun.
2
015, 6, 7793. [CrossRef] [PubMed]
6. Nauha, E.; Bernstein, J. “Predicting” polymorphs of pharmaceuticals using hydrogen bond propensities:
Probenecid and its two single-crystal-to-single-crystal phase transitions. J. Pharm. Sci. 2015, 104, 2056–2061.
[
CrossRef] [PubMed]
7. Xu, K.; Zheng, S.; Zhai, Y.; Guo, L.; Tang, P.; Yan, J.; Wu, D.; Li, H. Two solid forms of tauroursodeoxycholic
acid and the effects of milling and storage temperature on solid-state transformations. Int. J. Pharm. 2015
86, 185–194. [CrossRef] [PubMed]
,
4

Molecules 2017, 22, 1314
22 of 22
2
2
3
3
8. Alcolea, V.; Plano, D.; Encio, I.; Palop, J.A.; Sharma, A.K.; Sanmartin, C. Chalcogen containing heterocyclic
scaffolds: New hybrids with antitumoural activity. Eur. J. Med. Chem. 2016, 123, 407–418. [CrossRef]
PubMed]
9. Alcolea, V.; Plano, D.; Karelia, D.N.; Palop, J.A.; Amin, S.; Sanmartin, C.; Sharma, A.K. Novel seleno- and
thio-urea derivatives with potent in vitro activities against several cancer cell lines. Eur. J. Med. Chem. 2016
13, 134–144. [CrossRef] [PubMed]
[
,
1
0. Plano, D.; Baquedano, Y.; Moreno-Mateos, D.; Font, M.; Jiménez-Ruiz, A.; Palop, J.A.; Sanmartín, C.
Selenocyanates and diselenides: A new class of potent antileishmanial agents. Eur. J. Med. Chem. 2011, 46,
3
315–3323. [CrossRef] [PubMed]
1. Wang, L.; Guo, X.; Wang, J.; Jiang, C.; Bosland, M.C.; Lu, J.; Deng, Y. Methylseleninic acid superactivates
p53-senescence cancer progression barrier in prostate lesions of pten-knockout mouse. Cancer Prev. Res.
2
016, 9, 35–42. [CrossRef] [PubMed]
2. Fernandes, A.P.; Gandin, V. Selenium compounds as therapeutic agents in cancer. Biochim. Biophys. Acta
015, 1850, 1642–1660. [CrossRef] [PubMed]
3
3
2
3. Lamberto, I.; Plano, D.; Moreno, E.; Font, M.; Palop, J.A.; Sanmartin, C.; Encio, I. Bisacylimidoselenocarbamates
2
cause g /m arrest associated with the modulation of cdk1 and chk2 in human breast cancer mcf-7 cells.
Curr. Med. Chem. 2013, 20, 1609–1619. [CrossRef] [PubMed]
Sample Availability: Samples of the compounds reported in this paper are available from the authors.
©
2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
CC BY) license (http://creativecommons.org/licenses/by/4.0/).
(