Synthesis, biological activity, distribution and membrane permeability of novel spiro-thiazines as potent neuroprotectors

Article abstract of DOI:10.1016/j.ejmech.2014.02.052

New spiro-derivatives of 1,3-thiazine - potential neuroprotectors have been synthesized. It has been determined that the obtained compounds are biologically active and capable of blocking the glutamate-induced calcium ion uptake into synaptosomes of rat brain cortex. The inhibitory activity of the test substances was shown to depend on the chemical nature and structure of the substituents bound with an exocyclic nitrogen atom. Non-polar alkyl and polar radicals with halogen, oxygen and nitrogen atoms were used as substituents. It is typical of the active spiro-thiazines to have alkyl substituents in ortho- and para-position of the benzene ring. Among the investigated spiro-thiazines it is the derivatives with ethyl- and isopropyl-groups in the aril part of the molecules that are the lead-compounds with a high inhibitory ability. We measured the distribution coefficients of the substances in octanol/buffer and hexane/buffer systems and made conclusions about the ability of the investigated drug-like compounds to penetrate the biological membranes. By using the parabolic model we derived a quadratic equation that allowed us to evaluate quantitatively the inhibitory activity of spiro-thiazines with hydrophobic substituents based on lipophilicity data. We also studied the permeability through the phospholipidic membrane and introduced a correlation equation describing the dependence of the investigated spiro-thiazines activity on the descriptors characterizing the donor-acceptor properties.

Full text of DOI:10.1016/j.ejmech.2014.02.052

European Journal of Medicinal Chemistry 77 (2014) 8e17
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Synthesis, biological activity, distribution and membrane permeability
of novel spiro-thiazines as potent neuroprotectors
,
a
a
Svetlana V. Blokhina a , Tatyana V. Volkova , Marina V. Ol’khovich ,
*
Angelica V. Sharapova a, Alexey N. Proshin b, Sergey O. Bachurin b, German L. Perlovich a
,
b
a Institute of Solution Chemistry, Russian Academy of Sciences, 153045 Ivanovo, Russia
b Institute of Physiologically Active Compounds, Russian Academy of Sciences, 142432 Chernogolovka, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
New spiro-derivatives of 1,3-thiazine e potential neuroprotectors have been synthesized. It has been
determined that the obtained compounds are biologically active and capable of blocking the glutamate-
induced calcium ion uptake into synaptosomes of rat brain cortex. The inhibitory activity of the test
substances was shown to depend on the chemical nature and structure of the substituents bound with an
exocyclic nitrogen atom. Non-polar alkyl and polar radicals with halogen, oxygen and nitrogen atoms
were used as substituents. It is typical of the active spiro-thiazines to have alkyl substituents in ortho-
and para-position of the benzene ring. Among the investigated spiro-thiazines it is the derivatives with
ethyl- and isopropyl-groups in the aril part of the molecules that are the lead-compounds with a high
inhibitory ability. We measured the distribution coefficients of the substances in octanol/buffer and
hexane/buffer systems and made conclusions about the ability of the investigated drug-like compounds
to penetrate the biological membranes. By using the parabolic model we derived a quadratic equation
that allowed us to evaluate quantitatively the inhibitory activity of spiro-thiazines with hydrophobic
substituents based on lipophilicity data. We also studied the permeability through the phospholipidic
membrane and introduced a correlation equation describing the dependence of the investigated spiro-
thiazines activity on the descriptors characterizing the donoreacceptor properties.
Received 3 December 2013
Received in revised form
19 February 2014
Accepted 20 February 2014
Available online 21 February 2014
Keywords:
New spiro-derivatives of 1,3-thiazine
Inhibitory activity
Distribution coefficients
Lipophilicity
Permeability
Correlation structureeactivity
Ó 2014 Elsevier Masson SAS. All rights reserved.
1. Introduction
potentiating the toxicity of neurotransmitter amino-acids, as well
as initiating apoptose processes. At present a whole series of ap-
In accordance with the contemporary conceptions of the Alz-
heimer disease pathogenesis the key factor in the development of
neurodegenerative processes in the brain is formation of a patho-
logical form of amyloid peptide. In elderly the mechanism of
cleavage of this protein changes leading to a reduction in the level
proaches have been proposed to prevent the cellular destruction
and to correct neurodegenerative processes, among them devel-
opment of potential neuroprotectors with a high biological activity
e glutamate-inducing calcium channel-blocking agents [1,2]. While
designing new drug compounds it should be taken into account
that drugs with neuroprotective action should easily pass from the
blood flow to the spinal fluid and brain i.e. to overcome the blood
brain barrier (BBB). The main parameter for predicting whether the
substance can pass through the BBB is the partition coefficient in
the hexane/buffer system [3]. Besides, drug compounds should
possess good lipophilicity. It is well known that the distribution
coefficient in the octanol/buffer system is a generally accepted
parameter of lipophilicity. The distribution coefficients in the above
mentioned system are viewed as the most significant physico-
chemical parameters in all investigations connected with toxico-
logical properties and pharmacokinetic behavior of the substances
[4,5]. Another important stage in drug design is investigating
bioavailability that indicates whether the molecules are able to
of the soluble form and the formation of a peptide
aggregation of -amyloid results in creation of insoluble fibrils,
participating in senile plaque formation. Neurodegenerative
mechanisms caused by -form of the peptide involve disturbing the
b-form. Further
b
b
calcium homeostasis, strengthening the oxidative stress,
Abbreviations: K43/21, Biological Coefficient Describing Ability to Inhibit Glu-Ca-
uptake; IC50, concentration of inhibition; DO/B, octanolebuffer coefficient; DH/B
partition coefficient; logDO/H, difference between the partition coefficients octa-
nolebuffer and hexane/buffer; Papp, apparent permeability; SCad , descriptor e
,
D
/a
ratio the sum of H-bond donoreacceptor factors in molecule to polarizability.
* Corresponding author.
E-mail address: svb@isc-ras.ru (S.V. Blokhina).
http://dx.doi.org/10.1016/j.ejmech.2014.02.052
0223-5234/Ó 2014 Elsevier Masson SAS. All rights reserved.

S.V. Blokhina et al. / European Journal of Medicinal Chemistry 77 (2014) 8e17
9
Table 1
penetrate through the semipermeable phospholipid cell mem-
The structural formulae of the R substituents of compounds studied.
branes in order to interact with intercellular receptors.
Searching for correlations between biological activity and lip-
ophilicity, membrane permeability and other molecular charac-
teristics is the base of the scientific concept named QSAR
(Quantitative structureeactivity relationship) in modern medicinal
chemistry. The methodology of finding the correlations between
the structure and biological activity of compounds is developing
intensively as the impact of individual substituents on biological
activity is being investigated [6]. The above mentioned in-
vestigations open opportunities for predicting the physiological
activity and rational synthesis of the drug compounds. These in-
vestigations make possible to predict the physiological activity of
the substances and the efficient synthesis of the drugs for
perspective development of the pharmaceutics and clinical
chemistry.
Compound
eR
Compound
eR
1
eCH3
12
2
13
3
4
eH
14
15
By analyzing the accumulated experimental bioscreening re-
sults of a huge array of substances, data on their medical applica-
tions and structure we managed to single out a number of
structures often used in drug substances [7]. Previously the authors
[8] had proved the ability of isothiourea derivatives to act as neu-
roprotectors and cognitive stimulators. We developed a method of
synthesizing new 2-aminospirothiazines e cyclic derivatives of
isothiourea without substitutions on the bicyclic fragment of the
molecule. The present study is aimed at investigating the ability of
the synthesized spiro-thiazines to inhibit glutamate-inducing cal-
cium ions uptake and revealing the correlations between the bio-
logical activity and the structure, the lipophilicity and the
membrane permeability of the substances. The present study is a
continuation of our previous investigations into the biological ac-
tivity, solubility, lipophilicity and membrane permeability of drugs
and drug-like compounds [9].
5
6
16
17
7
8
18
19
2. Results and discussion
In order to develop drug substances with neuroprotective ac-
tivity we synthesized a number of 1,3-thiazine spiro-derivatives
whose molecular structure is represented in Fig. 1.
We have shown that these compounds can modulate glutamate-
dependent uptake of calcium ions in rats’ cerebral cortex synap-
tosomes. The structural formulae of the R substituents are pre-
sented in Table 1. The results of the biological activity studies of the
investigated compounds are reported in Table 2.
By analyzing the structureeactivity correlation we determined
biologically active compounds. A histogram of the biological ac-
tivity values for the synthesized substances is shown in Fig. 2.
As it follows from the above-mentioned data the inhibiting ac-
tivity depends on the chemical nature and the structure of the
substituents bound to the exocyclic nitrogen atom. Spiro-thiazine
with an unsubstituted phenyl group (15) weakly inhibiting cal-
cium ions uptake was chosen as the fiducial substance to identify
the active and inactive substances which allowed us to divide
conventionally the synthesized compounds into two groups. The
substances (1e14) with halogen, oxygen and nitrogen atoms in the
substituents as well as the compounds in which R is hydrogen or an
9
eC2H5
20
21
22
10
11
alkyl group belong to group I e inactive spiro-derivatives. The
biological activity investigations have shown that the unsaturated
spiro-thiazine (15) and its ethyl analog (9) inhibit calcium ions
uptake poorly, while the methyl derivative (1) even potentiates the
uptake process. For the aryl derivatives the inhibiting activity de-
pends on the kind of substituents of the ring. It is noteworthy that
all the compounds with electron acceptor substituents n the ben-
zene ring (2, 4e8,10e14) displayed insignificant activity in this test.
Moreover, amino-derivative (2) promotes a calcium ion uptake.
Fig. 1. Molecular structure of spiro-compounds.

10
S.V. Blokhina et al. / European Journal of Medicinal Chemistry 77 (2014) 8e17
Table 2
isomers with acetyl-group in the para- and meta-positions of the
benzene ring (13 and 8, respectively) have different inhibiting
properties. Another example of how the steric structure of the
compounds affects the biological activity is represented by the
compound (5) in which the acetyl group is nearer to the spiro ring.
The effectiveness of the action of the above mentioned compound
is lower in comparison with the structurally similar acetyl-
substituted substances.
Coefficient of Biological Activity (K43/21), Concentration of Inhibition (IC50), Con-
stants of Partition Octanol/Buffer (DO/B and Hexane/Buffer (DH/B), Apparent
Permeability Coefficient (Papp) and Descriptor (SCad ) for Spiro-derivatives studied.
)
/
a
ꢁ
N
K43/21 (%)
IC50
(m
M) logDO/B logDH/B
D
logDO/H Papp (106 cm/s) SCad
/a
1
2
3
4
5
6
7
8
9
134.2 ꢂ 13.2
114.7 ꢂ 14.5
91.4 ꢂ 9.8
81.5 ꢂ 0.8
78.1 ꢂ 14.7
76.9 ꢂ 3.1
73.5 ꢂ 12.5
67.3 ꢂ 0.4
65.3 ꢂ 12.1
e
e
e
e
e
e
e
e
e
e
0.88
1.88
1.57
2.03
1.55
2.45
1.96
2.37
1.38
2.53
2.44
0.80
1.97
2.18
1.02
1.56
2.50
2.36
1.88
2.05
2.31
1.57
0.91 ꢀ0.03
2.70 ꢂ 0.23
2.04 ꢂ 0.02
7.33 ꢂ 0.15
7.36 ꢂ 0.25
3.18 ꢂ 0.29
3.10 ꢂ 0.30
1.72 ꢂ 0.14
74.73 ꢂ 0.91
8.78 ꢂ 0.54
90.31 ꢂ 2.41
23.90 ꢂ 0.31
1.11 ꢂ 0.05
2.55 ꢂ 0.31
3.06 ꢂ 0.13
1.81 ꢂ 0.06
5.67 ꢂ 0.20
27.82 ꢂ 1.92
9.04 ꢂ 0.01
21.60 ꢂ 1.54
11.34 ꢂ 0.10
41.02 ꢂ 0.42
15.73 ꢂ 0.98
0.241
ꢀ0.52
2.40
0.35
1.08
0.04
1.21
e
0.347
0.254
0.266
0.231
0.251
0.360
0.287
0.223
0.166
0.162
0.190
0.198
0.170
0.180
0.171
0.170
0.213
0.213
0.225
0.162
0.153
1.22
0.95
1.51
1.24a
e
Group II of the active spiro-thiazines (16, 18e22) is character-
ized by the presence of electron donor alkyl substituents in ortho-
and para-position of the benzene ring. Compound (17) with a
chlorine atom on the phenyl ring also belongs to this group. The
biological activity of these compounds, determined by the K43=21
value, changes within the range of 25 to 0%. The elongation of the
alkyl chain and the replacement of the methyl-derivative (16) with
an ethyl-derivative (21) has strengthened the inhibiting action and
increased the biological activity to the highest value in this test.
Branching of the alkyl chain is usually accompanied by a decrease in
the biological effect as the volume of the molecule enlarges and the
steric hindrance growth impedes the optimal interactions with the
bioreceptor active part. But if the substances under investigation
are spiro-thiazines, the isomerization of the alkyl substituents is a
way to improve their inhibiting properties. It should be noted that
the biological activity of the second group of compounds grows
when the number of carbon atoms in the structure of the sub-
stituents decreases from 4 to 3. As a result along with ethyl-
derivative (21) the best blocker of glutamate-induced calcium
ions uptake is compound (22) with an isopropyl group.
As the efficiency of the drug substance action depends on the
concentration of the substance near the receptor (i.e. passing the
substance through the body membranes), the next part of our study
deals with measuring and analyzing the distribution coefficients in
organic solvent/buffer systems, modeling the passive drug trans-
port through the biological membranes. The two-phase octanol/
buffer system is a model of GI tract membranes and the hexane/
buffer one e of BBB membranes. The distribution coefficients of the
spiro-thiazines in the mentioned systems are given in Table 2.
The analysis of the obtained data showed that logDO/B values
change within the range of 1.9e2.6 and satisfy the well known
requirement [11] that the optimal meanings of the distribution
coefficients should be in the range of 1e3. Such drug compounds
are characterized by a quite good absorption and diffusion (i.e. they
have balanced properties of the hydrophilicelipophilic character-
istics of the molecule), that determine the membrane permeability
processes. Poor lipophilicity is typical of non-aryl derivatives (1, 3
and 9), model compound (15) and spiro-thiazine with a methoxy-
group (12). The halogen-derivatives (6, 10, 11, 14 and 17) belong to
the compounds with high logDO/B values > 2.18. We have not found
any considerable differences in the lipophilicity of the compounds
of group I containing polar (hydrophilic) substituents and group II
with alkyl (hydrophilic) ones. It should be mentioned that the
spiro-derivatives under consideration have a complex molecular
structure with hydrophilic amine and hydrophobic phenyl groups
(Fig. 1). Therefore, the non-fulfillment of the additivity principle of
the impacts of different substituents on the relative hydrophobicity
of the investigated spiro-thiazines can be explained on the basis of
the concept of considerable influence of the steric arrangement of
the substituents in the diphilic compounds [7].
1.31
0.72
1.36
0.91
1.06
0.66
1.17
1.53
10 56.5 ꢂ 4.7
11 45.9 ꢂ 2.3 89.1
12 45.7 ꢂ 1.7 85.1
13 45.1 ꢂ 6.3 81.3
14 42.1 ꢂ 5.3 77.6
15 33.2 ꢂ 6.3 63.1
16 20.4 ꢂ 3.9 31.6
17 12.7 ꢂ 7.4 35.5
1.45 ꢀ0.65
1.88
1.31
0.09
0.87
1.69 ꢀ0.67
1.48
1.34
1.50
1.28
1.48
1.75
1.45
0.08
1.15
0.86
0.60
0.57
0.56
0.12
18
19
20
21
22
5.4 ꢂ 5.4 20.9
3.5 ꢂ 2.7
7.9
9.3
2.6 ꢂ 2.5
0
0
11.7
6.2
a
compound was not dissolved in hexane.
Introducing a CF3-group in the para-position of the phenyl ring (14)
leads to a small decrease in the biological activity when compared
to the naked phenyl substituted spiro-thiazine (15). The inhibiting
ability decreases more significantly when a bromine atom is taken
as a substitute (10) and mono-substituted chlorine and fluorine
derivative is replaced with twice-substituted ones (4 and 6). Pre-
viously we had determined that spiro-compounds containing
halogen atoms in the structure have very low solubility (w10ꢀ7 mol
fraction) in the buffer solution simulating a blood system [10].
Probably, this fact accounts for the decrease in the inhibiting bio-
action in a number of halogen-derivatives: CF3(14) > CH3, Cl
(11) > Br (10) > Cl, Cl (6) > F, Cl (4). The influence of the oxygen
atom in the molecule structure on the biological activity has been
studied by the example of the compounds with carbonyl (5, 8 and
13), methoxy (12) and hydroxyl (7) groups. The obtained results
showed that the biological activity is sensitive even to a small
change in the position of the substituents. Thus, the structural
The distribution coefficient in hexane/buffer system is the ratio
of the substance concentrations in the lipophilic phase (hexane) to
its concentration in the hydrophilic medium (phosphate buffer).
The logDH/B values do not exceed 1.9 (Table 2). The obtained data
testify that all the compounds except 2 prefer lipophilic delivery
pathways to penetrate BBB. For all the investigated compounds,
except the following three (1, 12 and 15) with low lipophilicity, the
Fig. 2. Biological activity coefficients (K43/21) of the investigated compounds.

S.V. Blokhina et al. / European Journal of Medicinal Chemistry 77 (2014) 8e17
11
inequality logDO/B > logDH/B is satisfied. In order to predict the BBB
permeability we have calculated logDO/H e parameter of the
spiro-derivatives given in Table 2. An increase in logDO/H is caused,
D
D
as a rule, by a decrease in the permeability and brain distribution
[3]. The values of this parameter for the investigated spiro-
derivatives satisfy the well known requirement that the optimal
DlogDO/H for brain penetration should be less than two units.
The distribution coefficient logarithm in the octanol/buffer
system is the most often used descriptor for estimating quantitative
correlations structureeactivity. Fig. 3 shows the dependences of
the biological activity on the molecular lipophilic properties char-
acterized by logDO/B. As it follows from the represented data for
group I of the compounds with polar substituents, no correlation
was revealed between the activity and distribution coefficients. In
order to explain this fact we used the correlation Equation (1) of
Hansch [6], applied to compounds of a certain kind of activity and
analogous chemical structure:
Fig. 4. Correlation between the membrane permeability coefficients (log(Papp)) and
the distribution coefficients (logDO/B) of spiro-compounds studied.
ꢀ
ꢁ
logð1=kÞ ¼ a0 þa1 logDO=B ꢀa2 logDO=B 2 þa3sþa Es;
(1)
4
as in the first case the compounds reveal a poor absorption in the
lipophilic phase, whereas in the second one, they don’t penetrate
the membranes due to a strong interaction with lipid matrix dis-
played in a poor diffusion. As a result, the dependence of the ac-
tivity of the spiro-thiazines under consideration with hydrophobic
substituents on the distribution coefficient is a parabolic curve with
a maximum. Substances (19e22) belong to the group of leader-
compounds balanced by their lipophilicity/hydrophilicity. The
correlation equation, in which the biological activity acts as a
dependent variable and the distribution coefficient acts as an in-
dependent one, is presented below:
where k is any experimental value characterizing the biological
activity, logDO/B is the distribution coefficient of the substance be-
tween octanol and water phases, parameters
s and Es are electron
and steric influences of the substituents, ai is the constants ob-
tained when the experimental data are processed by the least-
squares method. Since the spiro-derivatives of group I poorly
interact with glutamate receptors, it is probably necessary to take
into consideration the steric impacts of the substituents as well as
the hydrophobic properties in order to estimate the correlations
between the compounds structure and the biological activity.
The dependence of logð1=K43=21Þ of (logDO/B) for the compounds
of group II with alkyl substituents is also represented in Fig. 3. As an
exception this group of biologically active spiro-thiazines also in-
cludes compound (17) containing a chlorine atom. It should be
refined that the most active substances (21 and 22) with K43=21 ¼ 0
are not indicated in Fig. 3. We used the parabolic model to describe
the non-linear correlations lipophilicityeactivity. According to this
model, in order to connect with the target molecules the drug
compounds should have an opportunity, on the one hand, to
circulate in the blood stream (i.e. to be dissoluble in water), and on
the other hand, to penetrate through the cell membranes (i.e. to be
soluble in lipids). In this case the activity of the substances with
very small and very large distribution coefficients, i.e. highly hy-
drophilic (16) or highly lipophilic ones (17 and 18), will be not high
ꢀ
ꢁ
log 1=K43=21 ¼ ꢀ15:43 þ 14:58 log DO=B
ꢀ
ꢁ
2
ꢀ 3:54 log DO=B
ðR ¼ 0:9994Þ
(2)
The statistical criteria of Equation (2) testify that it is possible to
quantitatively assess the inhibiting ability of spiro-thiazines with
hydrophobic substituents on the basis of lipophilic data. The ob-
tained correlations can be used to predict the structure of the
leader compounds with a high biological activity in a series of the
studied spiro-derivatives. The permeability of the compounds
through different membranes determines their bioavailability
which is of principal importance in drug development. It is well
known that most orally administered drugs overcome the intestine
epithelium by passive transport which to a great extent depends on
the drug physicochemical properties. In our study we used Phos-
pholipid vesicle Permeability assay [12] to imitate the intestine
epithelium of the human body. The results of the investigation of
the spiro-derivative membrane permeability are given in Table 2.
As it follows from the permeability coefficient results, all the
compounds under investigation have Papp > 0.9$10ꢀ6 cm/s, there-
fore, they can be classified as substances with a high ability to
permeate through phospholipid membranes. The authors [13]
suggested the hypotheses that cell permeability depends, in the
first place, on the lipophilicity, determined by the distribution co-
efficient logDO/B (pH 7.4), as well as on the polar surface area and
molecular weight of a compound. Fig. 4 presents the correlation
between the logarithms of the permeability coefficients and the
logarithms of the distribution coefficients in buffer/octanol system.
As the experiment results show, in spite of the dependence of the
membrane permeability on the lipophilicity, modeling perme-
ability is a complicated task. It was estimated that for the com-
pounds of groups I and II the permeability increases as the
lipophilicity grows. This is coordinated with literature data: when
Fig. 3. Correlation between the biological activity coefficient (logð1=K43=21Þ) and the
distribution coefficients (logDO/B) of spiro-compounds studied.

12
S.V. Blokhina et al. / European Journal of Medicinal Chemistry 77 (2014) 8e17
logDO/B values are lower than 3.5, the transepithelial permeation
coefficients increase as the lipophilicity rises [14]. Meanwhile, the
low correlation coefficients (R < 0.6) only reveal a qualitative
character of the obtained correlations. The analysis of the results
represented in Fig. 4 shows that the permeability of the studied
spiro-compounds depends on the chemical nature of the functional
groups in the substance molecules. In spite of their small molecular
volume the non-aryl derivative (1), the compound with the
unsubstituted benzene ring (15) and the methoxy-group
substituted one (12) penetrate through the membranes rather
poorly due to their low lipophilicity (logDO/B < 1.4). The maximal
permeability is observed for acetyl- and bromine-substituted aryl
derivatives (8 and 10), which have a high lipophilicity (logDO/
searching for correlations between the structure and activity,
considered to be additive [15]. The most pronounced basic center of
the investigated spiro-derivatives is the nitrogen atom of the sec-
ondary amino-group which can form a hydrogen bond with the
hydrogen atom of the octanol hydroxyl group. This fact
makes it necessary to take into account not only non-specific in-
teractions with the organic solvent but also the interactions
caused by hydrogen bonding when analyzing the activity of the
spiro-thiazines under investigation. That is why we chose the SCad
-parameter to assess the donoreacceptor hydrogen bonding
ability of the molecules, which was calculated by the HYBOT pro-
gram package [16]. The SCad descriptor is the ratio of the sum of H-
bond donoreacceptor factors in the molecule to polarizability. The
statistical criteria of the equations, where SCad act as independent
/
a
/a
>
1.9). It should be mentioned that the correlation line
/a
B
«permeability-lipophilicity» for group II of the compounds with a
high biological activity is located above the analogous line for group
I of the non-active derivatives. This fact indicates that among the
studied substances the biologically active spiro-thiazines can
penetrate through the phospholipid membranes better than the
others.
The dependences of log(1/IC50) and logð1=K43=21Þ of the spiro-
derivatives on their permeability is presented in Fig. 5. For the
biologically active compounds (group II) the dependences have a
parabolic type. It should be noted that compounds 21 and 22 with
K43=21 ¼ 0 are not represented in Fig. 5b, due to mathematical
limitations when using Hansch Equation (1). The obtained results
showed that the poor inhibiting action of the compounds of group I
is not determined by the membrane permeability. For the com-
pounds of group II we revealed a parabolic dependence of activity
on the permeability which can be described by the equation:
variables, were the best among the descriptors. Fig. 6 presents the
dependences of the logarithm of the aryl compound activity on the
SCad/a evalue (non-aryl derivatives 1, 3 and 9 are excluded from
consideration). As wecan seefromFig. 6, the biologicalactivityof the
substances depends on their chemical nature. The activity of spiro-
derivatives of group I with polar substituents is decreasing as the
descriptor grows. The opposite tendency is observed for compounds
of group II with non-polar substituents: the activity rises if the
descriptor increases. It should be mentioned that the model com-
pound (15) with unsubstituted benzene ring is appropriately placed
between the estimated correlation lines of substance groups I and II.
In our opinion, the reason for the revealed correlations is changing
the basicity of the nitrogen atom of the spiro-thiazine amino-group
when polar and non-polar radicals are introduced into the benzene
ring. The electron-acceptor substituents of the spiro-derivatives of
group I (carbonyl, alkoxy, hydroxy and Hal) reduces the basicity of
the nitrogen in the amino-group. The electron-donor substituents
(alkyl) act in the opposite way e they increase the electron density
on the nitrogen atom and raise its basicity. As a result, the electron
pair of the nitrogen atom is held less firmly and interacts more easily
with a proton forming a stronger hydrogen bond. Thus, the obtained
results indicate that the ability of molecules to take part in specific
interactions has an important influence on the biological activity of
the investigated spiro-thiazines. Moreover, it should be taken into
consideration that realizing these interactions in a living organism is
determined bycomplicated processes of bioavailability and receptor
binding activity.
ꢀ
ꢁ
log 1=K43=21 ¼ ꢀ153:88 ꢀ 63:10 log Papp
ꢂ
ꢃ
2
ꢀ 6:49 log Papp
ðR ¼ 0:9347Þ
(3)
As it follows from the presented data, the maximal biological
activity is typical of compounds with a medium permeability. We
can make a conclusion that the high inhibiting ability of the leader
compounds (19e22) is provided by the optimal combination of the
lipophilic and hydrophilic characteristics of the molecules,
permeability through the membranes and geometrical/topological
structure that leads to a better interaction with the receptors.
Lipophilicity of chemical compounds is used as an important
physicochemical parameter practically in all investigations associ-
ated with biological activity of drug substances. It is well known that
the distribution of substances in the two-phase octanol/water sys-
tem, characteristic of lipophilicity, is determined by the contribu-
tions of van-der-Waals forces and tendencies to hydrogen bond
formation. These different effects can be separated and, when
The resulting equations, where the activity acts as the depen-
dent variable and SCad/a e as the independent one are presented
below. For the compounds of group I with electron-acceptor
substituents:
ꢀ
ꢁ
X
log K43=21 ¼ 1:4156 þ 1:6073
Cad= ðR ¼ 0:8130Þ
(4)
a
Fig. 5. Correlation between the biological activity coefficients and the membrane permeability coefficients (logPapp) of spiro-compounds studied: a) e log(1/IC50); b) e
logð1=K43=21Þ.

S.V. Blokhina et al. / European Journal of Medicinal Chemistry 77 (2014) 8e17
13
Fig. 6. Correlation of the biological activity (log(L43/21)) and SCad/a descriptor of spiro-
compounds.
Fig. 7. Comparative analysis of the experimental log(K43/21
)
and predicted log(K43/
exp
)
values for the test set.
21 cal
estimated to be less than 1.9. The data obtained testify that all
spiro-compounds prefer lipid paths of delivery to overcome BBB.
The parabolic model was applied to description of non-linear
For the compounds of group II with electron-donor
substituents:
ꢀ
ꢁ
X
correlations
“lipophilicityeactivity”.
We
experimentally
log K43=21 ¼ 3:5898ꢀ13:9556
Cad= ðR ¼ 0:9605Þ
(5)
a
confirmed the commonly accepted views that leader compounds
should have balanced lipophilicity/hydrophilicity parameters. We
have obtained a second-degree equation allowing quantitative
assessment of the inhibiting activity of spiro-thiazines with hy-
drophobic substituents based on the lipophilicity data.
We have investigated the compounds permeability through the
phospholipidic membranes and made a conclusion that the high
inhibiting ability of the leader compounds is provided by the
optimal combination of lipophilicity, membrane permeability and
interaction with the receptors. It was estimated that the ability of
the investigated spiro-derivatives to take part in specific in-
teractions has a considerable effect on their biological activity. We
have suggested a correlation equation describing the dependence
of such activity of the investigated spiro-thiazines on the de-
scriptors that characterize donoreacceptor properties.
The calculations carried out by Equations (4) and (5) have led us
to the following correlation of the experimental and calculated
activity values:
ꢀ
ꢁ
log K
¼ 0:6369
ð43=21Þcalc
ꢀ
ꢁ
þ0:6414log K
ðR ¼ 0:8014Þ
(6)
ð43=21Þexp
Comparing the experimental and calculated activity values
(Fig. 7) has revealed a rather high prognostic ability of the intro-
duced approach for quantitative description of the structureeac-
tivity correlation of the studied spiro-thiazines.
3. Conclusions
4. Experimental section
We have synthesized new spiro-derivatives of 1,3-thiazine as
potential neuroprotectors. It was estimated that the obtained
compounds are biologically active and can block glutamate-
induced calcium ion uptake into rats’ brain cortex synaptosomes.
It was shown that the inhibiting activity of the tested compounds
depends on the chemical nature and the structure of the sub-
stituents bound with an exocyclic nitrogen atom. The inactive
spiro-thiazine group includes the compounds with halogen, oxygen
and nitrogen atoms in the substituents as well as the non-aryl
derivatives. The group of active spiro-thiazines contains alkyl
substituents in the ortho- and para-position of the phenyl ring.
Among the investigated spiro-thiazines the derivatives with a high
inhibiting activity have ethyl- and isopropyl-radicals on the aryl
part of the molecule.
The ability of drug-like compounds to pass through the barriers
of the body was investigated using the distribution coefficients of
the compounds in octanol/buffer and hexane/buffer systems. The
analysis of the results has shown that logDO/B values for spiro-
derivatives are within the range of 1e3, which allows us to
conclude that all the investigated spiro-thiazines have balanced
characteristics of solubility and permeability by passive diffusion
and, as a consequence, are easily absorbed. The logDH/B values were
4.1. Materials and physical measurements
All chemicals used were purchased from Aldrich (USA). All re-
agents used in current study were of analytical grade. 1H NMR
spectra were recorded on Bruker CXP-200 instrument (Germany) in
CDCl3. Chemical shifts
d are given in the d scale relative to Me4Si.
The solvents were removed using rotary evaporator under water
pump vacuum. Elemental analysis of the compounds obtained was
performed on a CHN analyzer (Carlo Erba). Melting points were
determined by microscope «Leitz Laborlux 12 Pol» with hot stage
«Mettler FP 82». The pH values were measured by the use of a
Toledo MP 220 pH meter (Mettler, USA) standardized with pH 1.68
and 9.22 solutions. Spectrophotometer Spectramax 190 (Molecular
Devices Corporation, California, USA) was used for measuring the
optical density of solutions.
4.2. General procedure for the synthesis of compounds 1e22
Synthetic approaches to novel 22 spiro-derivates were carried
out according to Scheme:

14
S.V. Blokhina et al. / European Journal of Medicinal Chemistry 77 (2014) 8e17
To a stirred solution of 2-(1-cyclohexenyl)ethylamine II (1.25 g,
4.2.5. N-(1-Thia-3-aza-spiro[5.5]undec-2-en-2-yl)-benzamide (5)
Yield 79%, mp 130.4 ꢁC. 1H NMR [200 MHz, DMSO-d6]
: 1.36
0.01 mmol) in diethyl ether (20 mL), a solution of appropriate
isothiocyanate I in diethyl ether (0.01 mmol) was added dropwise.
Then the reaction mixture was stirred for 2e5 h at ambient tem-
perature until formation of precipitate. The precipitate of thiourea
III was filtered off, dried, suspended in 48% aqueous HBr (10 mL),
and refluxed for 5 h; the precipitate dissolved during the reaction.
After the completion of the reaction, the mixture was cooled to
ambient temperature, diluted with water (20 mL) and dichloro-
methane (50 mL), and then saturated aqueous NaHCO3 was care-
fully added until the solution was basic. The organic layer was
separated and dried with Na2SO4. The drying agent was filtered off
and the solvent was removed. The residue was recrystallized from
propane-2-ol to give N-substituted 1-thia-3-aza-spiro[5.5]undec-
2-en-2-ylamine (IV) [17].
d
(1H, m, C(8)HH), 1.68 (7H, m, C(6)HH, C(7)H2, C(8)HH, C(9)H2, C(10)
HH), 2.00 (4H, m, C(5)H2, C(6)HH, C(10)HH), 3.64 (2H, m, C(4)H2),
7.40 (3H, m, ArH), 8.08 (2H, m, ArH),11.52 (1H, br s, NH). Anal. Calcd
for C16H20N2OS: C, 66.63; H, 6.99; N, 9.71. Found: C, 66.55; H, 6.85;
N, 9.70; Calcd. Mass: 288.4146; Found: 288.4146.
4.2.6. (3,4-Dichloro-phenyl)-(1-thia-3-aza-spiro[5.5]undec-2-en-
2-yl)-amine (6)
Yield 75%, mp 151.5 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.20 (1H, m,
C(8)HH), 1.48 (7H, m, C(6)HH, C(7)H2, C(8)HH, C(9)H2, C(10)HH),
1.81 (4H, m, C(5)H2, C(6)HH, C(10)HH), 3.50 (2H, t, C(4)H2,
J ¼ 5.8 Hz), 6.60 (2H, m, ArH), 7.25 (1H, m, ArH). Anal. Calcd for
C15H18Cl2N2S: C, 54.71; H, 5.50; N, 8.51. Found: C, 54.66; H, 5.51; N,
8.56; Calcd. Mass: 329.2961; Found: 329.2959.
4.2.1. Methyl-(1-thia-3-aza-spiro[5.5]undec-2-en-2-yl)-amine (1)
Methyl-(1-thia-3-aza-spiro[5.5]undec-2-en-2-yl)-amine
was obtained from methyl-isothiocyanate in a similar way in 72%
yield; mp 72e73 ꢁC. 1H NMR [200 MHz, CDCl3]
: 1.30 (m, 1H, CH),
(1)
4.2.7. (4-Hydroxy-phenyl)-(1-thia-3-aza-spiro[5.5]undec-2-en-2-
yl)-amine (7)
Yield 63%, mp 192.9 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.44 (8H, m,
d
1.52 (m, 1H, CH), 1.59 (m, 6H, 6CH), 1.71 (t, 2H, J ¼ 5.8 Hz, NCH2CH2),
1.88 (m, 2H, 2CH), 3.66 (C, 3H, CH3), 3.66 (t, 2H, J ¼ 5.8 Hz,
NCH2CH2). Anal. Calcd for C15H20N2S: C, 60.54; H, 9.17; N, 14.12.
Found: C, 60.52; H, 9.18; N, 14.11; Calcd. Mass: 198.3862; Found:
198.3860.
C(6)HH, C(7)H2, C(8)H2, C(9)H2, C(10)HH), 1.80 (2H, m, C(6)HH,
C(10)HH), 1.99 (2H, t, J ¼ 5.6 Hz, C(5)H2), 3.60 (2H, t, J ¼ 5.6 Hz, C(4)
H2), 6.78 (2H, d, J ¼ 8.8 Hz, ArH), 6.96 (2H, d, J ¼ 8.8 Hz, ArH), 10.60
(1H, br.s, OH). Anal. Calcd for C15H20N2OS: C, 65.12; H, 7.24; N,10.13;
Found: C, 65.17; H, 7.21; N, 10.11; Calcd. Mass: 276.4030; Found:
276.4034.
4.2.2. 4-Aminophenyl(1-thia-3-azaspiro[5.5]undec-2-en-2-yl)
amine (2)
4.2.8. 3-Acetylphenyl(1-thia-3-azaspiro[5.5]undec-2-en-2-yl)
The overall yield for the two steps was 67%, mp 161 ꢁC. 1H NMR
amine (8)
Yield 63%, mp 138.2 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.29 (1H, m,
[200 MHz, CDCl3] d: 1.31 (1H, m, C(8)HH); 1.59 (7H, m, C(6)HH, C(7)
C(8)HH); 1.56 (7H, m, C(6)HH, C(7)H2, C(8)HH, C(9)H2, C(10)HH);
1.85 (2H, t, C(5)H2, J ¼ 5.9 Hz); 1.88 (2H, m, C(6)HH, C(10)HH); 3.61
(2H, t, C(4)H2, J ¼ 5.9 Hz); 2.51 (3H, s, C(O)CH3); 6.28 (1H, br.s, NH);
7.46 (1H, m, ArH); 7.56 (1H, m, ArH), 7.79 (1H, m, ArH); 8.60 (1H, s,
ArH). Anal. Calcd for C17H22N2OS: C, 67.51; H, 7.33; N, 9.26. Found:
C, 67.40; H, 7.45; N, 9.28; Calcd. Mass: 302.4417; Found: 302.4422.
H2, C(8)HH, C(9)H2, C(10)HH); 1.89 (4H, m, C(5)H2, C(6)HH, C(10)
HH); 3.59 (2H, m, C(4)H2); 4.94 (2H, br.s, NH2); 6.20 (1H, br.s, NH);
6.64 (2H, d, ArH, J ¼ 8.6 Hz); 6.99 (2H, d, ArH, J ¼ 8.6 Hz). Anal. Calcd
for C15H21N3S: C, 65.35; H, 7.62; N,15.25. Found: C, 65.24; H, 7.67; N,
15.31; Calcd. Mass: 275.4191; Found: 275.4193.
4.2.3. 1-Thia-3-aza-spiro[5.5]undec-2-en-2-ylamine (3)
4.2.9. Ethyl-(1-thia-3-aza-spiro[5.5]undec-2-en-2-yl)-amine (9)
Yield 67%; mp 104e106 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.27 (m,
Yield 62%, mp 107 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.16 (t, 3H,
1H, CH), 1.56 (m, 7H, 7CH), 1.67 (t, 2H, J ¼ 5.9 Hz, NCH2CH2), 1.85 (m,
2H, 2CH), 3.59 (t, 2H, J ¼ 5.9 Hz, NCH2CH2), 3.66 (C, 2H, NH2). Anal.
(C9H16N2S) C,H,N. Anal. Calcd for C9H2S: C, 58.66; H, 8.74; N, 15.20.
Found: C, 58.63; H, 8.73; N, 15.24; Calcd. Mass: 184.2913; Found:
184.3051.
J ¼ 7.2 Hz, CH3), 1.32 (m, 1H, CH), 1.52 (m, 1H, CH),1.59 (m, 6H, 6CH),
1.72 (t, 2H, J ¼ 5.8 Hz, NCH2CH2), 1.88 (m, 2H, 2CH), 3.26 (q, 2H,
J ¼ 7.2 Hz, NCH2CH3), 3.64 (t, 2H, J ¼ 5.8 Hz, NCH2CH2). Anal. Calcd
for C11H20N2S: C, 62.22; H, 9.48; N, 13.19. Found: C, 62.27; H, 9.46;
N, 13.21; Calcd. Mass: 212.3413; Found: 212.3416.
4.2.4. (3-Chloro-4-fluoro-phenyl)-(1-thia-3-aza-spiro[5.5]undec-
2-en-2-yl)-amine (4)
4.2.10. (4-Bromo-phenyl)-1-thia-3-aza-spiro[5.5]undec-2-en-2-
yl)-amine (10)
Yield 71%, mp 154.6 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.23 (1H, m,
Yield 72%, mp 175.3 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.32 (1H, m,
C(8)HH), 1.40 (7H, m, C(6)HH, C(7)H2, C(8)HH, C(9)H2, C(10)HH),
1.83 (4H, m, C(5)H2, C(6)HH, C(10)HH), 3.45 (2H, m, C(4)H2), 6.82
(1H, m, ArH), 6.94 (1H, m, ArH), 7.10 (1H, m, ArH). Anal. Calcd for
C(8)HH), 1.61 (7H, m, C(6)HH, C(7)H2, C(8)HH, C(9)H2, C(10)HH),
1.91 (4H, m, C(5)H2, C(6)HH, C(10)HH), 3.59 (2H, m, C(4)H2), 7.08
(2H, d, J ¼ 8.6 Hz, ArH), 7.39 (2H, d, J ¼ 8.6 Hz, ArH). Anal. Calcd for
C
15H18ClFN2S: C, 57.59; H, 5.79; N, 8.95. Found: C, 57.55; H, 5.80; N,
C15H19BrN2S: C, 53.10; H, 5.64; N, 8.26. Found: C, 53.30; H, 5.65; N,
8.98; Calcd. Mass: 312.8421; Found: 312.8419.
8.15; Calcd. Mass: 339.3001; Found: 339.3000.

S.V. Blokhina et al. / European Journal of Medicinal Chemistry 77 (2014) 8e17
15
4.2.11. (3-Chloro-4-methyl-phenyl)-(1-thia-3-aza-spiro[5.5]undec-
Anal. Calcd for C15H19ClN2S: C, 61.10; H, 6.50; N, 9.50. Found: C,
61.22; H, 6.53; N, 9.45; Calcd. Mass: 294.8491; Found: 294.8495.
2-en-2-yl)-amine (11)
Yield 74%, mp 149.3ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.28 (1H, m,
C(8)HH), 1.55 (7H, m, C(6)HH, C(7)H2, C(8)HH, C(9)H2, C(10)HH),
1.87 (4H, m, C(5)H2, C(6)HH, C(10)HH), 2.29 (3H, s, CH3), 3.54 (2H,
m, C(4)H2), 6.88 (1H, dd, J ¼ 1.1 8.1 Hz, ArH), 7.08 (1H, d, J ¼ 8.1 Hz,
ArH), 7.18 (1H, d, J ¼ 1.1 Hz, ArH). Anal. Calcd for C16H21ClN2S: C,
62.22; H, 6.85; N, 9.07. Found: C, 62.30; H, 6.85; N, 9.05; Calcd.
Mass: 308.8762; Found: 308.8766.
4.2.18. (2-Methyl-6-Isopropyl- phenyl)-(1-thia-3-aza-spiro[5.5]
undec-2-en-2-yl)-amine (18)
Yield 75%, mp 98 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.22 (6H, t,
J ¼ 6.8 Hz, CH(CH3)2), 1.31 (1H, m, C(8)HH), 1.56 (7H, m, C(6)HH,
C(7)H2, C(8)HH, C(9)H2, C(10)HH), 1.85 (2H, m, C(6)HH, C(10)HH),
2.01 (2H, t, J ¼ 5.7 Hz, C(5)H2), 2.23 (3H, c, CH3), 3.11 (1H, m,
CH(CH3)2), 3.54 (2H, m, C(4)H2), 6.25 (1H, br s, NH), 7.11 (3H, m,
ArH).
4.2.12. (4-Methoxy-phenyl)-(1-thia-3-aza-spiro[5.5]undec-2-en-2-
yl)-amine (12)
Anal. Calcd for C19H28N2S: C, 72.11; H, 8.91; N, 8.85. Found: C,
72.13; H, 8.90; N, 8.81; Calcd. Mass: 316.4850; Found: 316.4851.
Yield 65%, 115.5 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.33 (1H, m,
C(8)HH), 1.60 (7H, m, C(6)HH, C(7)H2, C(8)HH, C(9)H2, C(10)HH),
1.88 (4H, m, C(5)H2, C(6)HH, C(10)HH), 3.61 (2H, m, C(4)H2), 3.81
(3H, s, OCH3), 4.50 (1H, br s, NH), 6.84 (2H, d, J ¼ 8.8 Hz, ArH), 7.13
(2H, d, J ¼ 8.8 Hz, ArH). Anal. Calcd for C16H22N2OS: C, 66.17; H, 7.64;
N, 9.65. Found: C, 66.22; H, 7.66; N, 9.69; Calcd. Mass: 290.4305;
Found: 290.4310.
4.2.19. (2,6-Diethyl-phenyl)-(1-thia-3-aza-spiro[5.5]undec-2-en-2-
yl)-amine (19)
Yield 69%, mp 119.6 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.11 (3H, t,
J ¼ 7.5 Hz, CH3), 1.28 (1H, m, C(8)HH), 1.42 (7H, m, C(6)HH, C(7)H2,
C(8)HH, C(9)H2, C(10)HH), 1.71 (2H, m, C(6)HH, C(10)HH), 1.84 (2H,
t, J ¼ 5.6 Hz, C(5)H2), 2.44 (4H, q, J ¼ 7.5 Hz, CH2), 3.40 (2H, t,
J ¼ 5.6 Hz, C(4)H2), 6.96 (3H, m, ArH). Anal. Calcd for C19H28N2S: C,
72.11; H, 8.91; N, 8.85. Found: C, 72.09; H, 8.93; N, 8.86; Calcd.
Mass: 316.4850; Found: 316.4848.
4.2.13. 1-[4-(1-Thia-3-aza-spiro[5.5]undec-2-en-2-ylamino)-
phenyl]-ethanone (13)
Yield 69%, mp 163.6 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.29 (1H, m,
C(8)HH), 1.56 (7H, m, C(6)HH, C(7)H2, C(8)HH, C(9)H2, C(10)HH),
1.85 (2H, t, J ¼ 5.9 Hz, C(5)H2), 1.88 (2H, m, C(6)HH, C(10)HH), 2.55
(3H, s, C(O)CH3), 3.61 (2H, t, J ¼ 5.9 Hz, C(4)H2), 6.25 (1H, br s, NH),
7.24 (2H, d, J ¼ 8.6 Hz, ArH), 7.87 (2H, d, J ¼ 8.6 Hz, ArH).
4.2.20. (2,4,6-Trimethyl-phenyl)-(1-thia-3-aza-spiro[5.5]undec-2-
en-2-yl)-amine (20)
Yield 76%, mp 148.1 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.33 (1H, m,
Anal. Calcd for C17H22N2OS: C, 67.52; H, 7.31; N, 9.27. Found: C,
67.46; H, 7.40; N, 9.28; Calcd. Mass: 302.4417; Found: 302.4422.
C(8)HH), 1.55 (7H, m, C(6)HH, C(7)H2, C(8)HH, C(9)H2, C(10)HH),
1.83 (2H, m, C(6)HH, C(10)HH), 2.17 (6H, c, 2CH3), 1.93 (2H, t,
J ¼ 5.7 Hz, C(5)H2), 3.52 (2H, t, J ¼ 5.7 Hz, C(4)H2), 2.29 (3H, c, CH3);
6.21 (1H, br s, NH). 6.86 (2H, c, ArH). Anal. Calcd for C18H26N2S: C,
71.53; H, 8.66; N, 9.27. Found: C, 71.59; H, 8.44; N, 9.22; Calcd. Mass:
302.2479; Found: 302.2480.
4.2.14. (1-Thia-3-aza-spiro[5.5]undec-2-en-2-yl)-(4-
trifluoromethyl-phenyl)-amine (14)
Yield 77%, mp 149.8 ꢁC, 1H NMR [200 MHz, CDCl3]
d: 1.29 (1H, m,
C(8)HH), 1.58 (7H, m, C(6)HH, C(7)H2, C(8)HH, C(9)H2, C(10)HH),
1.85 (4H, t, J ¼ 5.9 Hz, C(6)HH, C(10)HH, C(5)H2), 3.60 (2H, t,
J ¼ 5.9 Hz, C(4)H2), 6.11 (1H, br s, NH), 7.25 (2H, d, J ¼ 8.4 Hz, ArH),
7.49 (2H, d, J ¼ 8.4 Hz, ArH). Anal. Calcd for C16H19F3N2S: C, 58.52; H,
5.83; N, 8.53. Found: C, 58.33; H, 5.85; N, 8.65; Calcd. Mass:
328.4024; Found: 328.4028.
4.2.21. 4-Ethyl-(1-thia-3-aza-spiro[5.5]undec-2-en-2-yl)-amine
(21)
Yield 62%, mp 139.5 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.16 (3H, t,
J ¼ 7.2 Hz, CH3), 1.32 (1H, m, C(8)HH), 1.58 (7H, m, C(6)HH, C(7)H2,
C(8)HH, C(9)H2, C(10)HH), 1.72 (2H, t, J ¼ 5.8 Hz, C(5)H2), 1.88 (2H,
m, C(6)HH, C(10)HH), 3.28 (2H, q, J ¼ 7.2 Hz, NHCH2), 3.61 (1H, br s,
NH), 3.64 (2H, t, J ¼ 5.8 Hz, C(4)H2)0. Anal. Calcd for C11H20N2S: C,
62.22; H, 9.49; N, 13.19. Found: C, 62.33; H, 9.43; N, 13.05; Calcd.
Mass: 212.3595; Found: 212.3598.
4.2.15. Phenyl-(1-thia-3-aza-spiro[5.5]undec-2-en-2-yl)-amine
(15)
Yield 79%, mp 113.5 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.22 (1H, m,
C(8)HH), 1.50 (7H, m, C(6)HH, C(7)H2, C(8)HH, C(9)H2, C(10)HH),
1.78 (4H, m, C(5)H2, C(6)HH, C(10)HH), 3.55 (2H, m, C(4)H2), 6.20
(1H, br s, NH), 6.90 (1H, m, ArH), 7.19 (4H, m, ArH). Anal. Calcd for
4.2.22. (4-Isopropyl-phenyl)-(1-thia-3-aza-spiro[5.5]undec-2-en-
2-yl)-amine (22)
Yield 73%, mp 139.7 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.21 (6H, d,
C
15H20N2S: C, 69.19; H, 7.74; N, 10.76. Found: C, 69.30; H, 7.85; N,
10.85; Calcd. Mass: 260.4041; Found: 260.4040.
J ¼ 6.7 Hz, C(CH3)2), 1.30 (1H, m, C(8)HH), 1.56 (7H, m, C(6)HH, C(7)
H2, C(8)HH, C(9)H2, C(10)HH), 1.84 (2H, t, J ¼ 5.9 Hz, C(5)H2), 1.88
(2H, m, C(6)HH, C(10)HH), 2.84 (1H, m, CH(CH3)2), 3.65 (2H, t,
J ¼ 5.9 Hz, C(4)H2), 7.10 (2H, m, ArH). Anal. Calcd for C18H26N2S: C,
71.47; H, 8.66; N, 9.26. Found: C, 71.40; H, 8.85; N, 9.25; Calcd.
Mass: 302.4853; Found: 302.4852.
4.2.16. 4-Methyl-(1-thia-3-aza-spiro[5.5]undec-2-en-2-yl)-amine
(16)
Yield 72%, mp 108.3 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.30 (1H, m,
C(8)HH), 1.58 (7H, m, C(6)HH, C(7)H2, C(8)HH, C(9)H2, C(10)HH),
1.71 (2H, t, J ¼ 5.8 Hz, C(5)H2), 1.88 (2H, m, C(6)HH, C(10)HH), 2.81
(3H, s, CH3), 3.66 (2H, t, J ¼ 5.8 Hz, C(4)H2), 3.88 (1H, br s, NH). Anal.
Calcd for C10H18N2S: C, 60.56; H, 9.15; N, 14.12. Found: C, 60.59; H,
9.33; N, 14.05; Calcd. Mass: 198.3324; Found: 198.3327.
Purity of the spiro-derivatives was 0.98 (mass fraction).
4.3. Permeation studies
Egg phosphatidyl choline, Lipoid E-80 was obtained from Lipoid
(Germany). Culture inserts (Transwell-Clear, d ¼ 6.5 mm) and
plates were purchased from Corning Inc. Corning, USA, filters in
4.2.17. (4-Chloro-phenyl)-1-thia-3-aza-piro[5.5]undec-2-en-2-yl)-
amine (17)
Yield 75%, mp 162.7 ꢁC. 1H NMR [200 MHz, CDCl3]
d: 1.29 (1H, m,
existence were removed, and mixed cellulose ester filters (0.65 mm
C(8)HH), 1.59 (7H, m, C(6)HH, C(7)H2, C(8)HH, C(9)H2, C(10)HH),
1.89 (4H, m, C(5)H2, C(6)HH, C(10)HH), 3.57 (2H, m, C(4)H2), 7.06
(2H, d, m, ArH), 7.35 (2H, m, ArH).
pore size) from Millipore, Billerica, USA were fused on. Phospho-
lipid vesicle-based barriers were prepared according to the proce-
dure described earlier [12].

16
S.V. Blokhina et al. / European Journal of Medicinal Chemistry 77 (2014) 8e17
Table 3
The method of the liposome preparation involves film hydra-
Effect of spiro-derivatives studied on glutamate-induced 45Ca2þ uptake into syn-
aptosomes of rat cortex.
tion/filter extrusion. Egg phosphatidyl choline (1.8 g) was dissolved
in a mixture of chloroform and methanol (2:1) (6 mL) in a round-
bottomed flask. The organic solvent was removed under vacuum
at 55 ꢁC. The deposited lipid film was exposed to vacuum of 55 hPa
at room temperature for an additional period of 3 h to remove
traces of solvent, before hydration with phosphate buffer, con-
taining 10% (v/v) ethanol to obtain a 6% (w/v) liposomal dispersion.
The liposome dispersion was then filter-extruded through 0.8 and
Compound
Concentration,
Percentage 45Ca2þ of control
Inhibition, %
m
M
(control e 100%)
MK-801
100
50
25
10
5
26.9
35.0
46.5
65.2
76.1
97.5
98.4
45.9
69.1
92.9
45.7
68.3
97.0
45.1
63.7
82.3
42.1
63.2
98.8
33.2
54.7
80.4
20.4
31.3
60.4
12.7
33.1
65.7
93.3
5.4
14.8
41.5
72.3
3.5
25.1
28.2
45.7
55.7
2.6
14.2
27.6
50.1
61.9
0.0
20.8
36.1
52.2
69.9
0.0
73.1
65.0
53.5
34.8
23.9
2.5
0.4 mm polycarbonate membrane filters (Millipore) at room tem-
1
0.5
1.6
perature. The extrusion was done by hand with a syringe through
the filters in a 25 mm Swinnex Filter Holder from Millipore. To
obtain liposomes of two different sizes, one portion was extruded
11
12
13
14
15
16
17
100
50
25
100
50
25
100
50
25
100
50
25
100
50
10
100
50
25
100
50
25
10
100
50
25
10
100
50
25
10
5
100
50
25
10
5
100
50
25
10
5
100
50
25
10
5
54.1
30.9
7.1
54.3
31.7
3.0
five times through the 0.8
another portion was extruded five additional passages through
filters with 0.4 m pore size.
mm polycarbonate membrane filter while
m
54.9
36.3
17.7
57.9
36.8
1.2
66.8
45.3
19.6
79.6
68.7
39.6
87.3
66.9
34.3
6.7
94.6
85.2
58.5
27.7
96.5
74.9
71.8
54.3
44.3
97.4
85.8
72.4
49.9
38.1
100.0
79.2
63.9
47.8
30.1
100.0
99.9
95.0
61.8
44.7
Stock solutions of the compounds under study (2.5$10ꢀ5 to
2.5$10ꢀ3 mol Lꢀ1) were prepared by dissolving the drug in phos-
phate buffer solution pH 7.4. Phosphate buffer solution (pH 7.4) was
prepared by mixing the solutions of appropriate phosphoric acid
sodium and potassium salts. Containing KH2PO4, Na2HPO4$12H2O,
NaCl, NaN3, and water, the solution was adjusted to pH 7.4 with
HCl/NaOH if necessary. Concentrations of different drug solutions
were chosen on the assumption of analytical considerations as well
as for obtaining the sink conditions during permeation studies. The
concentrations of different drug solutions had to be high enough
for the obtainment of a definite amount of a drug in an acceptor
chamber during permeation studies to be quantified by means of
UV-absorbance and still be below the solubility limit.
18
19
Permeation studies were performed after loading the inserts
(the facility for bearing the 0.65
filter) with drug solution (100
acceptor compartments containing phosphate buffer solution
(600 L) (pH 7.4). The permeation experiment was carried out at
mm pore size mixed cellulose ester
m
L) and placing them into separate
m
room temperature without agitation, and inserts were moved to
wells containing equal quantities of fresh buffer solution. Taking
into account the low solubility of the compounds under investi-
gation, the operational range of 12e24 h was optimal. In conclu-
20
21
22
sion of the permeation experiment, samples (200 mL) from each
acceptor compartment were transferred into 96-well UV 96-well
black plates (Costar) and drug concentrations were measured
spectrophotometrically at the most appropriate wavelength for
each drug. A blank well was filled with pure phosphate buffer pH
7.4. The electrical resistance of the lipid barriers was measured
(Millicell-ERS, Millipore, USA) immediately after completion of
permeation studies. To find out the resistance of the lipid barrier
0.1
5.0
38.2
55.3
itself, a value of 119
U resulting from the filter characteristics was
subtracted from the observed resistance meaning. Such a differ-
ence was then multiplied by the surface area (0.33 cm2) to be
normalized for the dimensions of the insert. The experiments
were performed at least in triplicate with three inserts in each
parallel for every compound. The mean values and standard de-
viations are reported.
flux rates were used to calculate the apparent permeability coeffi-
cient (Papp, cm sꢀ1) by means of the following equation:
The standard time dependences of the cumulative amount of
permeated drug consisted of 6e9 points, and every point repre-
sented the mean value of three parallels. The r2-meanings were
always higher than 0.99. Repeatability corresponded to 3.9e20.1%,
and an intermediate precision of 2.7e29.4% was observed for
different drugs using such a novel method. The linear part of the
slope of the curve of the standard time dependence of the cumu-
lative amount represented the steady-state flux rate. If a lag time
was observed before the steady-state conditions were attained, and
the saturation conditions took place at the end of the experiments,
only the middle points of the cumulative plot were used for the
calculation procedure, based on the steady-state flux. The obtained
Papp ¼ J=ðA$ðCd ꢀ CaÞÞ
(7)
where J is the observed flux rate at steady-state (nmol sꢀ1), A is the
surface area of the insert (cm2), and Cd and Ca are the concentra-
tions of the solutions in the donor and acceptor chambers (nmol/
mL), respectively. The experiments were performed under the sink
conditions; that is, the drug concentration in the acceptor chamber
did not exceed 10% of the drug concentration in the donor chamber
at any time [18].

S.V. Blokhina et al. / European Journal of Medicinal Chemistry 77 (2014) 8e17
17
4.4. Determination of partition coefficients
The concentrations of the compounds and characteristics of
inhibition are presented in Table 3. NMDA-antagonist MK-801 have
been used for positive control. Dizocilpine (MK-801) with purity
98% was purchased from SigmaeAldrich (Saint Louis, MO, USA).
The isothermal saturation method was used to determine the
partition coefficients in n-octanol/buffer (pH 7.4), DO/B, and n-
hexane/buffer (pH 7.4), DH/B, systems. All the experiments were
carried out at 25 ꢁC. The substance concentrations were measured
spectrophotometrically using calibration curves. The partition co-
efficient was calculated using the following equation:
4.6. Calculation procedure
Physicochemical descriptors were calculated with the program
package HYBOT-PLUS (version of 2003) in Windows [16].
DO=B ¼ CO=B=CB=O
(8)
Acknowledgments
where CO/B and CB/O are the molar concentrations of solute in the
mutually saturated phases of octanol and water. The accuracy of the
D value was verified by checking the mass balance of the starting
amount of compound i compared to the total amount of the com-
pound partitioned between the two phases:
ꢁ
This work was supported by the grant RFBR (N 12-03-00019-
a).
References
mi ¼ mO=B þ mB=O
(9)
[1] C.G. Parsons, W. Danysz, G. Quack, Glutamate in CNS disorders as a target for
drug development: an update, Drug News & Perspectives 11 (1998) 523e579.
[2] S.O. Bachurin, Medicinal chemistry approaches for the treatment and pre-
vention of Alzheimer’s disease, Medicinal Research Reviews 23 (2003) 48e88.
[3] E.H. Kerns, L. Di, Drug-like Properties: Concepts, Structure Design and
Methods, Academic Press, 2008.
where mi ¼ CiVi is the starting mass (in moles) of the compound,
mO/B ¼ CO/BVO/B is the mass of the substance dissolved in the water-
saturated octanol phase, and mB/O ¼ CB/OVB/O is the mass of the
substance dissolved in the octanol-saturated water phase.
ꢀ
[4] V. Pliska, Lipophilicity in Drug Action and Toxicology, John Wiley & Sons,
Chichester, 2008.
[5] J. Sangster, Octanol-water Partition Coefficients: Fundamentals and Physical
Chemistry, John Wiley & Sons, Chichester, 1997.
[6] C. Hansch, Quantitative structure-activity relationship in drug design, in:
E.J. Ariens (Ed.), Drug Design, vol. 1Academic Press, New York, 1971, pp. 271e
342.
[7] F.R. Rekker, R. Mannhold, Calculation of Drug Lipophilicity: the Hydrophobic
Fragmental Constant Approach, VCH, Weinheim, 1992.
4.5. Calcium-blocking property experiments
Interaction between the compounds and the glutamate-
dependent calcium uptake system was studied on newborn
(8ꢀ11 days old) rat brain synaptosomal P2-fraction isolated ac-
cording to the following method: synaptosomes were suspended in
incubation buffer A (132 mM NaCl, 5 mM KCl, 5 mM HEPES) pH 7.4
stored at 0 ꢁC during the experimental stage. The aliquots of syn-
[8] G.L. Perlovich, A.N. Proshin, T.V. Volkova, S.V. Kurkov, V.V. Grigoriev,
L.N. Petrova, S.O. Bachurin, Novel isothiourea derivatives as potent neuro-
protectors and cognition enhancers: synthesis, biological and physicochemical
properties, Journal of Medicinal Chemistry 52 (2009) 1845e1852.
[9] G.L. Perlovich, A.N. Proshin, T.V. Volkova, L.N. Petrova, S.O. Bachurin, Novel
1,2,4-thiadiazole derivatives as potent neuroprotectors: approach to creation
of bioavailable drugs, Molecular Pharmaceutics 9 (2012) 2156e2167.
[10] S.V. Blokhina, M.V. Ol’khovich, A.V. Sharapova, A.N. Proshin, G.L. Perlovich,
Thermodynamics of solubility processes of novel drug-like Spiro-Derivatives
in model biological solutions, Journal of Chemical & Engineering Data 57
(2012) 1996e2003.
[11] J.E. Comer, High Throughput Measurement of logD and pKa. Methods and
Principles in Medicinal Chemistry, Wiley-VCH, Weinheim, 2003.
[12] G.E. Flaten, A.B. Dhanikula, K. Luthman, M. Brandl, Drug permeability across a
phospholipid vesicle-based barrier: a novel approach for studying passive
diffusion, European Journal of Pharmaceutical Sciences 27 (2006) 80e90.
[13] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and
computational approaches to estimate solubility and permeability in drug
discovery and development settings, Advanced Drug Delivery Reviews 46
(2001) 3e26.
aptosomes (50
compound and 45Ca-samples. Calcium ion uptake was stimulated
by introducing the glutamate solution (200 M). After 5 min in-
mL) were transposed to buffer A containing test
m
cubation at 37 ꢁC, the process was terminated by the filtration on
GF/B-filters. The sample was washed three times with cold buffer B
(145 mM KCl, 10 mM Tris, 5 mM Trilon B) followed by radioactivity
measuring with scintillator counter SL-4000 Intertechnic. In pre-
liminary experiments, all the compounds were tested for the ability
to inhibit glutamate stimulated Ca uptake at the concentration of
100 mM. If the inhibition of Glu-Ca-uptake was 50% and more, then
further studies were carried K43/21 out to determine the concen-
tration dependence of inhibition and the corresponding value of
was measured according to the following equation:
[14] P. Wils, A. Warnery, V. Phung-Ba, S. Legrain, D. Scherman, High lipophilicity
decreases drug transport across intestinal epithelial cells, Journal of Phar-
macology and Experimental Therapeutics 269 (1994) 654e658.
[15] M.H. Abraham, Scales of solute hydrogen-bonding: their construction and
application to physicochemical and biochemical processes, Chemical Society
Reviews 22 (1993) 3e83.
K43=21 ¼ 100$½ðCa4 ꢀ Ca3Þ=ðCa2 ꢀ Ca1Þꢃ
(10)
[16] Raevsky, O.A., Grigor’ev, V.J., Trepalin, S.V. HYBOT Program Package, Regis-
where Ca1 is the Ca2þ influx in the blank experiment (without
glutamate and test compounds); Ca2 is the Ca2þ influx in the
presence of glutamate only (Glu-Ca-uptake); Ca3 is the Ca2þ influx
in the presence of test compound (without glutamate); and Ca4 is
the Ca2þ influx in the presence of both glutamate and test
compound.
ꢁ
tration by Russian State Patent Agency, N 990090 of 26.02.99.
[17] A.N. Proshin, I.V. Serkov, L.N. Petrova, S.O. Bachurin, New isothiourea spiro-
derivatives in a number of the 1,3-thiazine, Russian Chemical Bulletin 11
(2011) 1e2.
[18] P. Artursson, Epithelial transport of drugs in cell-culture. A model for studying
the passive diffusion of drugs over intestinal absorptive (Caco-2) cells, Journal
of Pharmaceutical Sciences 79 (1990) 476e482.