Design and synthesis of dipeptide mimetics of the brain-derived neurotrophic factor

Article abstract of DOI:10.1134/S1068162012030053

Low-molecular-weight mimetics of loops 1 and 4 of the brain-derived neurotrophic factor (BDNF) have been designed and synthesized. The compounds represent monomeric and dimeric amides of N-acyldipeptides. Their dipeptide fragments coincide in sequence with the central regions of beta-turns of the corresponding neurotrophin loops, and acyl groups are the bioesosteres of preceding amino acid residues. Hexa-or heptamethylenediamines were used as spacers to link the C-terminal regions of dipeptides in dimeric mimetics of BDNF. These compounds were synthesized by classical methods of peptide synthesis in solution and received the laboratory codes GSB-104 (HO-Suc-Ser-Lys-NH2), GSB-106 {[HO-Suc-Ser-Lys-NH-(CH2)3-]2}, GSB-207 (HO-Suc-Met-Ser-NH2), and GSB-214 ([HO-Suc-Met-Ser-NH-(CH2)7/2-]2). It was shown using immortalized hippocampal cells of the HT22 line under conditions of oxi-dative stress that the dimeric mimetics of both loops at concentrations of 10-5-10-8 M possess a neuropro-tective activity. The monomeric loop 1 mimetic GSB-207 in the same concentration range is inactive, and the monomeric loop 4 mimetic GSB-104 at a concentration of 10-7 impairs the survival of neurons. The find-ing that only dimeric mimetics possess the neuroprotective activity is consistent with the data indicating that BDNF is active in the homodimeric form. As opposed to the dimeric loop 1 mimetic GSB-214, the dimeric loop 4 mimetic GSB-106 exhibits the antidepressant activity typical for BDNF in the Porsolt test on rats at doses of 0.1 and 1 mg/kg injected intraperitoneally. This suggests that the antidepressant activity of BDNF is related to its 4th loop. We believe that the compounds obtained will be useful in studies of the mechanism of action of BDNF and may form the basis for the design of a novel group of drugs with antidepressant and neu-roprotective activities. Pleiades Publishing, Ltd., 2012.

Full text of DOI:10.1134/S1068162012030053

ISSN 1068ꢀ1620, Russian Journal of Bioorganic Chemistry, 2012, Vol. 38, No. 3, pp. 243–252. © Pleiades Publishing, Ltd., 2012.
Original Russian Text © T.A. Gudasheva, A.V. Tarasyuk, S.V. Pomogaibo, I.O. Logvinov, P.Yu. Povarnina, T.A. Antipova, S.B. Seredenin, 2012, published in Bioorganicheskaya
Khimiya, 2012, Vol. 38, No. 3, pp. 280–290.
Design and Synthesis of Dipeptide Mimetics
of the BrainꢀDerived Neurotrophic Factor
T. A. Gudasheva1, A. V. Tarasyuk, S. V. Pomogaibo, I. O. Logvinov, P. Yu. Povarnina,
T. A. Antipova, and S. B. Seredenin
Zakusov Research Institute of Pharmacology, Russian Academy of Medical Sciences,
ul. Baltiiskaya 8, Moscow, 125315 Russia
Received June 1, 2011; in final form, July 18, 2011
Abstract—Lowꢀmolecularꢀweight mimetics of loops 1 and 4 of the brainꢀderived neurotrophic factor
(BDNF) have been designed and synthesized. The compounds represent monomeric and dimeric amides of
N
ꢀacyldipeptides. Their dipeptide fragments coincide in sequence with the central regions of betaꢀturns of
the corresponding neurotrophin loops, and acyl groups are the bioesosteres of preceding amino acid residues.
Hexaꢀ or heptamethylenediamines were used as spacers to link the Cꢀterminal regions of dipeptides in
dimeric mimetics of BDNF. These compounds were synthesized by classical methods of peptide synthesis in
solution and received the laboratory codes GSBꢀ104 (HO–SucꢀSerꢀLys–NH2), GSBꢀ106 {[HO–SucꢀSerꢀ
Lys–NH–(CH2)3–]2}, GSBꢀ207 (HO–SucꢀMetꢀSer–NH2), and GSBꢀ214 ([HO–SucꢀMetꢀSer–NH–
(CH2)7/2–]2). It was shown using immortalized hippocampal cells of the HT22 line under conditions of oxiꢀ
dative stress that the dimeric mimetics of both loops at concentrations of 10–5–10–8 M possess a neuroproꢀ
tective activity. The monomeric loop 1 mimetic GSBꢀ207 in the same concentration range is inactive, and
the monomeric loop 4 mimetic GSBꢀ104 at a concentration of 10–7 impairs the survival of neurons. The findꢀ
ing that only dimeric mimetics possess the neuroprotective activity is consistent with the data indicating that
BDNF is active in the homodimeric form. As opposed to the dimeric loop 1 mimetic GSBꢀ214, the dimeric
loop 4 mimetic GSBꢀ106 exhibits the antidepressant activity typical for BDNF in the Porsolt test on rats at
doses of 0.1 and 1 mg/kg injected intraperitoneally. This suggests that the antidepressant activity of BDNF is
related to its 4th loop. We believe that the compounds obtained will be useful in studies of the mechanism of
action of BDNF and may form the basis for the design of a novel group of drugs with antidepressant and neuꢀ
roprotective activities.
Keywords: BDNF, mimetics, dipeptides, antidepressant activity, neuroprotective activity
DOI: 10.1134/S1068162012030053
1
INTRODUCTION
(dophaminergic neurons of the black substance). In
addition, BDNF was shown to play an important role
in the etiology of Huntington’s disease and depression
[1, 2].
The brainꢀderived neurotrophic factor belongs to
the family of neurotrophins, which involves the nerve
growth factor, neurotrophinꢀ3, and neurotrophinꢀ4/5.
Owing to the ability to increase the survival of neuꢀ
rons, neurotrophins are considered as promising
antineurodegenerative drugs. BDNF is particularly
attractive in this respect since it improves the survival
and prevents the degeneration of neuronal populations
involved in diseases such as amyotrophic lateral scleꢀ
rosis (motoneurons), sensory neuropathies (sensory
neurons), Alzheimer’s disease (basal cholinergic neuꢀ
rons of the forebrain), and Parkinson’s disease
The biological activity of BDNF and other neuꢀ
rotrophins is effected through two types of transmemꢀ
brane receptors. The binding of neurotrophins to a
specific member of the Trk family of tyrosine kinase
receptors (BDNF interacts with TrkB and does not
interact with TrkA and TrkC) results in the
homodimerization of the receptor followed by selfꢀ
phosphorylation and the triggering of a great many
pathways of transduction of the biological response,
including those that lead to neuronal survival. Conꢀ
versely, glycoprotein p75 acts as a common lowꢀaffinꢀ
ity receptor for all neurotrophins. It is assumed to be
involved in the regulation of apoptosis and the moduꢀ
lation of the signaling cascade associated with the Trk
receptor [2].
Abbreviations: BDNF, brainꢀderived neurotrophic factor;
BuONO, butyl nitrite; DCU, 1,3ꢀdicyclohexylurea; DIEA,
N,N
ꢀdiisopropylethylamine; DMAPA, 3ꢀ(dimethylamino)proꢀ
pylamine; HOSu, ꢀhydroxysuccinimide; HONp, 4ꢀnitropheꢀ
nol; IBCF, isobutyl chloroformate; NGF, nerve growth factor;
NMM, ꢀmethylmorpholine; TEA, triethylamine; TFA, trifluꢀ
N
N
oroacetic acid; Z(Cl), 2ꢀchlorbenzyloxycarbonyl.
Corresponding author: phone: (495) 601ꢀ22ꢀ46; fax (499) 151ꢀ
12ꢀ61; eꢀmail: tataꢀsosnovka@mail.ru.
1
Despite the promising data of preclinical studies,
clinical trials of BDNF in the treatment of amyoꢀ
243

244
GUDASHEVA et al.
trophic lateral sclerosis have not been successful owing geometrically most advantageous for the interaction
to pharmacokinetic limitations, such as a short lifeꢀ with the receptor. Therefore, the sequences of βꢀturns
time in the blood flow (halfꢀlife of neurotrophin in rats ꢀAsp93ꢀSer94ꢀLys95ꢀLys96ꢀ and ꢀAsp30ꢀMet31ꢀSer32ꢀ
is less than 1 min) and a low capacity to permeate Gly33ꢀ were chosen as the basis for modeling.
through the hematoencephalic barrier [3]. In this conꢀ
nection, several laboratories are concerned with the
construction of lowꢀmolecularꢀweight functional
BDNF mimetics. Thus, a team of Australian researchꢀ
ers has designed a number of mimetics based on
BDNF loops: monomeric monocyclic peptides with
the antagonistic activity on the basis of loops 1, 2, and
4 [4, 5]; dimeric bicyclic and tricyclic peptides with
the agonistic activity based on loop 2 [6]; a cyclopenꢀ
tapeptide based on the loop 4 tripeptide fragment
ꢀLysꢀLysꢀArgꢀ, which was created as a ligand for the
p75 receptor and which enhances the survival of sensory
neurons in vitro [7], and its lipophilic derivatives [8].
To reproduce the active conformation, mimetics of
neurotrophin loops are usually obtained by covalent
cyclization of their fragments. We used another
approach, which we have earlier approved to syntheꢀ
size dipeptide mimetics of the NGF [11]. This
approach involves the synthesis of the amides of
N
ꢀacyldipeptides, which often reproduce the
conformation stabilized by the intramolecular hydroꢀ
gen bond [12]. The dipeptide coincided in sequence
with the central fragment of the ꢀturn of the loop,
βꢀturn
β
and the preceding amino acid residue was substituted
for by its bioisostere. In the present study, the aspartic
acid residue Asp93 or Asp30 was substituted for by a sucꢀ
A team of American investigators implanted cinic acid residue. The residue following the central
BDNF fragments in the structure of the NGF and fragment was represented by the amide group.
determined the region of loop 2 involved in the speꢀ
cific interaction of BDNF with TrkB, ꢀSerꢀLysꢀGlyꢀ
GlnꢀLeuꢀ [9]. Based on this structure, a pharmacophꢀ
oric hypothesis was advanced, 35 million described
compounds were virtually screened using this hypothꢀ
esis, and 1785 candidate compounds were identified.
With regard to the criteria of molecular weight and the
accessibility of the pharmacophoric moiety for the
interaction with the receptor, the number of the canꢀ
didates was reduced to 14. Seven of these 14 comꢀ
pounds were commercially available. After in vitro
tests on cell models, one compound based on 1,3,5ꢀ
benzoltricarboxylic acid (LM22Aꢀ4) was chosen. It
was shown in rats that, on intranasal administration,
this compound is capable of restoring the spatial memꢀ
ory impaired by a brain trauma.
As a result, we constructed the mimetic of the
BNDF loop ꢀturn HO–SucꢀSerꢀLys–NH2
(GSBꢀ104) and the mimetic of loop 1 ꢀturn
HO SucꢀMetꢀSer–NH2 (GSBꢀ207).
Because BDNF interacts with the TrkB receptor in
the dimeric form, the agonistic activity was achieved
by dimerizing the mimetics of the ꢀturn by hexaꢀ or
heptamethylenediamine. It has been earlier shown for
similar NGF mimetics that these spacers are optimal
since the reduction in the length to pentamethylenediꢀ
amine led to a decrease and then the reverse of activity,
and the elongation did not enhance the biological
activity [13].
4
β
β
⎯
β
Thus, the BDNF loop 4 mimetic bis(
cinylꢀseryllysine) hexamethylenediamide (GSBꢀ106)
and the BDNF loop 1 mimetic bis( ꢀmonosuccinylꢀ
Nꢀmonosucꢀ
N
Here we describe monomeric and dimeric linear
methionylserine) heptamethylenediamide (GSBꢀ214)
were constructed.
N
ꢀacyldipeptide mimetics of BDNF constructed on
the basis of its 1st and 4th loops. These dipeptides were
chosen since they have a low molecular weight and
better permeate across biological barriers, including
the hematoencephalic barrier. In the gastrointestinal
tract, there is a special system for the transport of
dipeptides, and they can freely penetrate into the
blood. Therefore, it can be expected that, on systemic
administration, including the peroral administration,
they can exhibit activity which is of importance for
their potential therapeutic application.
The mimetics designed were synthesized by the
methods of classical peptide synthesis in solution,
including the methods of activated esters, mixed
anhydrides, and the azide method with the use of the
strategy of Boc/Z protecting groups.
The monomeric dipeptide mimetic of loop
4 GSBꢀ104 was obtained by the condensation of Bocꢀ
Ser(Bzl)–OH with H–Lys(Z(Cl))–NH2 by the
method of mixed anhydrides under Andersen’s condiꢀ
tions [14] followed by the removal of the Bocꢀgroup,
the acylation of the product by succinic anhydride,
and Z(Bzl) deblocking (Scheme 1).
RESULTS AND DISCUSSION
When constructing BDNF mimetics, we relied on
The dimeric dipeptide mimetic of loop 4 GSBꢀ106
the literature data on the crystalline structure of the was obtained by linking the hexamethylenediamine
heterodimer BDNF/NTꢀ3 (pdb ID 1b8m) [10]. Like spacer to BocꢀLys(Z(Cl))–OSu with subsequent elonꢀ
other members of the neurotrophin family, BDNF is a gation of the peptide chain with the residue of the proꢀ
symmetrical homodimer whose monomeric units tected serine by the method of oxysuccinimide esters.
contain seven ꢀstrands linked by three hairpin loops The Boc protecting group was then removed, and the
β
exposed to solvent (loops 1, 2, 4). A visual analysis of bisdipeptide was acylated by succinic anhydride. The
this structure shows that loops 4 and 1 of BDNF are removal of side protecting groups by hydrogenolysis
most exposed to solvent. In addition, the central fragꢀ gave the dipeptide GSBꢀ106 with a total yield of 36%
ments of the ꢀturns of these loops occupy the position (Scheme 1).
β
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 38
No. 3
2012

DESIGN AND SYNTHESIS OF DIPEPTIDE MIMETICS
245
1) IBCF + NMM
TFA/CH2Cl2
DMF, –15°C, 1 h
BocꢀLys(Z(Cl))–OH
BocꢀLys(Z(Cl))–NH2
87%
TFA*H–Lys(Z(Cl))–NH2
99%
Troom, 30 min
2) NH3/DMF
Troom, 1 h
1) IBCF + NMM
DMF, –15°C, 1 h
BocꢀSer(Bzl)–OH
BocꢀSer(Bzl)ꢀLys(Z(Cl))–NH2
60%
2) H–Lys(Z(Cl))–NH2
DMF, Troom, 1 h
TFA
TFA*H–Ser(Bzl)ꢀLys(Z(Cl))–NH2
90%
CH Cl
2, Troom, ~1 h
2
O
O
O
1) TEA 2)
H2, 10% Pd/C
HO–SucꢀSer(Bzl)ꢀLys(Z(Cl))–NH2
88%
HO–SucꢀSerꢀLys–NH2
DMF
DMF, Troom, 12 h
, Troom, ~6–8 h
(GSBꢀ104)
100%
total yield 37%
H2N
NH2
BocꢀLys(Z(Cl))–NH
BocꢀLys(Z(Cl))–OSu
(CH2)6
EtOAc, Troom, 1 h
BocꢀLys(Z(Cl))–NH
85%
BocꢀSer(Bzl)ꢀLys(Z(Cl))–NH
1) TFA, Troom, ~1 h
(CH2)6
2) DIEA
BocꢀSer(Bzl)ꢀLys(Z(Cl))–NH
68%
+ BocꢀSer(Bzl)ꢀOSu
DMF, Troom, 1–2 h
HO–SucꢀSer(Bzl)ꢀLys(Z(Cl))–NH
HO–SucꢀSerꢀLys–NH
HO–SucꢀSerꢀLys–NH
1) TFA,
T
room, ~1 h
H2, 10% Pd/C
(CH2)6
(CH2)6
O
O
MeOH, Troom, ~6–12 h
O
HO–SucꢀSer(Bzl)ꢀLys(Z(Cl))–NH
84%
2) TEA +
(GSBꢀ106)
90%
total yield 36%
DMF, Troom, 3 h
IBCF, isobutyl chloroformate.
Scheme 1. Synthesis of BDNF loop 4 mimetics.
The monomeric dipeptide GSBꢀ207 was syntheꢀ firmed by 1H NMR spectroscopy. The diastereomeric
sized from serine methyl ester and Bocꢀmethionine purity was no less than 98%, as indicated by 1H NMR
pꢀnitrophenyl ester with subsequent ammonolysis, the data.
removal of the protecting group by trifluoroacetic
acid, and acylation by succinic anhydride with a total
yield of 21% (Scheme 2).
The neuroprotective activity of the dipeptides was
revealed using a model of oxidative stress with hydroꢀ
gen peroxide on a culture of immortalized rat hippocꢀ
ampal cell line HT22 [15]. The activity was calculated
in percent of the maximum possible effect, which
enabled one to quantitatively estimate both the neuroꢀ
protective and neurodegenerative (negative neuroproꢀ
tective) activities. The peptides were examined in a
concentration range of 10–5–10–8 M. Both dimeric
mimetics increased the viability of neurons (see table).
The 4 loop mimetic GSBꢀ106 and the 1 loop mimetic
GSBꢀ214 exhibited the neuroprotective activity in a
The dimeric dipeptide GSBꢀ214 was synthesized
by the azide method from Bocꢀmethionylꢀserine
methyl ester by its conversion into hydrazide and then
into azide followed by the condensation of the azide
with heptamethylenediamine. After the removal of the
Boc protection, bis(methionylserine) heptamethylꢀ
enediamide was acylated by succinic anhydride. GSBꢀ
214 was obtained with a total yield of 46% (Scheme 2).
All target compounds were purified by recrystalliꢀ concentration range of 10–5–10–8 M. The monomeric
zation. The products were homogeneous, as indicated loop 4 mimetic GSBꢀ104 at a concentration of 10–7
by the TLC and HPLC data. The structure was conꢀ decreased the neuron viability twofold, i. e., exhibited
M
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 38
No. 3
2012

246
GUDASHEVA et al.
the negative neuroprotective activity. This may be due are consistent with the fact that BDNF exhibits the
to its antagonism toward endogenous BDNF. The activity when interacting with the TrkB receptor in the
monomeric loop 1 mimetic was inactive. These data dimeric form.
BocꢀMetꢀONp HCl ⋅ H–Ser–OMe + TEA BocꢀMetꢀSer–OMe
DMF, Troom, 1 h
81%
1) TFA,
Amberlite IRAꢀ900
NH3/MeOH
room, ~72 h
BocꢀMetꢀSer–NH2
83%
HO–SucꢀMetꢀSer–NH2
T
O
O
2)
O
(GSBꢀ207)
32%
total yield 21%
/DMF
NH2NH2 ⋅ H2O
BocꢀMetꢀSer–OMe
BocꢀMetꢀSer–NH
BocꢀMetꢀSer–N2H3
100%
MeOH, Troom, ~12 h
1) TEA, Amberlite IRAꢀ900
1) 4 H HCl/dioxane
HO–SucꢀMetꢀSer–NH
T
room, ~0.5 h
BuONO, –30°C, 3 min
(CH2)7
(CH2)7
O
O
BocꢀMetꢀSer–NH
23%
O
HO–SucꢀMetꢀSer–NH
2) DIEA/DMF, –20°C
2)
3)
NH2
H2N
(GSBꢀ214)
100%
total yield 46%
DMF, Troom, 2 h
DMF, Troom, 1 h
Scheme 2. Synthesis of BDNF loop 1 mimetics.
Because BDNF is involved in the pathogenesis of pressant effect at doses of 1 and 0.1 mg/kg i.p. (see
depression and possesses the antidepressant activity figure). The magnitude of the effect amounted to
[2, 16], we studied the antidepressant properties of the 30% of the effect of the most optimal dose of the
BDNF dimeric mimetics on a classical “forced swimꢀ standard antidepressant imipramine on this model
ming” model in the original Porsolt test [17] by comꢀ (25 mg/kg i.p.) [18].
parison with the classical antidepressant imipramine.
Presumably, its fourth loop is responsible for the
activity of BDNF. Note that the selectivity associated
with loop structures was also found in the case of
dipeptide mimetics of NGF: loop 4 was responsible for
the neuroprotective activity of this neurotrophin, and
loop 1, for its differentiation activity [11].
Thus, we designed and synthesized for the first time
in vitro and in vivo active BDNF dipeptide mimetics,
which can be used to study the mechanism of action of
neurotrophin and serve as the basis for the developꢀ
ment of novel neuroprotective and antidepressant
drugs.
Only the loop 4 mimetic GSBꢀ106 exhibited the
activity. It produced a statistically significant antideꢀ
35
*
*
30
25
20
15
10
5
EXPERIMENTAL
Commercially available amino acids and their derivꢀ
atives (Fluka, Sigma) were used. Solvents were purified
0
1
10
1
0.10
mg/kg, i.p.)
0.05
and dried by standard methods. H NMR spectra
(δ, ppm) were recorded on a Bruker ACꢀ250 spectromꢀ
Dose
(
eter (Germany) in DMSOꢀd6 solutions with tetrameꢀ
thyl silane as an internal standard. The melting point
was determined in open capillaries and was not corꢀ
rected. Specific optical rotation was measured on a Perꢀ
kinꢀElmerꢀ241 polarimeter (England). TLC was perꢀ
Antidepressant activity of the BDNF dipeptide mimetic
GSBꢀ106 in the Porsolt forced swimming test as compared
with the classical antidepressant imipramine (25 mg/kg,
i.p.). *p < 0.05 relative to the control by the Mann—Whitꢀ
ney ꢀtest.
U
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 38
No. 3
2012

DESIGN AND SYNTHESIS OF DIPEPTIDE MIMETICS
247
Effect of dipeptide mimetics of BDNF on neuron viability under the conditions of oxidative stress
Optical absorption
Compound
GSBꢀ104
Concentration, M
Activity, %
with Н2О2
without Н2О2
10–5
0.114 0.036
0.100 0.020
0.089 0.010*
0.107 0.025
0.106 0.023#
26
–19
–55
3
10–6
10–7
10–8
0 (control)
0.137 0.026
0.124 0.010
0.113 0.007
GSBꢀ106
GSBꢀ207
GSBꢀ214
BDNF
10–5
0.097 0.010*
0.095 0.008*
0.095 0.010*
0.089 0.009
0.085 0.006#
31
27
26
10
10–6
10–7
10–8
0 (control)
10–5
0.081 0.010
0.087 0.010
0.082 0.005
0.084 0.010
0.083 0.005#
–7
13
–3
3
10–6
10–7
10–8
0 (control)
10–5
0.096 0.008*
0.098 0.011*
0.095 0.006*
0.088 0.004*
0.084 0.004#
34
40
31
11
10–6
10–7
10–8
0 (control)
0.119 0.006
0.094 0.004
50 ng/mL (~10–9
0 (control)
)
0.090 0.003
0.076 0.003#
78
Notes: Experiments were carried out on hippocampal neurons of mice, line HT22. Cell viability was determined by the MTT test.
* p < 0.05 by the Student’s tꢀtest relative to the active control (with H O ).
2
2
#
p < 0.05 by the Student’s tꢀtest relative to the passive control (without H O ).
2
2
formed on Kieselgel 60 G/F254 plates (Merck, Germany) TFA (0–25%, 25 min). The flow rate was 1 mL/min,
in solvent systems: chloroform–methanol 9 : 1 (A), and the detection was at 214 nm.
dioxane–water 9 : 1 (B), chloroform–methanol–water–
acetic acid 15 : 10 : 2 : 3 (C), pyridine–water–acetic acid–
Synthesis of HOOC(CH2)2CO
–SerꢀLys–NH2
ethyl acetate 20 : 11 : 6 : 120 (D), hexane–ethyl acetate
4 : 1 (E), and ꢀbutanol–acetic acid–water 4 : 1 : 1 (F).
(GSBꢀ104)
n
BocꢀLys(Z(Cl))–NH2. NMM (0.34 mL, 3.62 mmol)
and IBCF (0.52 mL, 3.62 mmol) were simultaneously
added from two dropping funnels to a solution of Bocꢀ
Lys(Z(Cl))–OH (1.5 g, 3.62 mmol) in DMF (10 mL)
Aminoꢀcontaining compounds were detected by ninꢀ
hydrin; compounds containing amide groups, in
iodine vapors; compounds with an open carboxyl
group, by bromocresol green; and compounds conꢀ
taining aromatic groups, in UV rays. The elemental
analysis was carried out on a device for carbon and
hydrogen determination with four electric ovens
at –10
no higher than –5
°
С
; the temperature was maintained at values
°С. After mixing for 2–3 min, DMF
(10 mL), saturated by cooling (0°С) with ammonia
(pH 8), was added. The reaction mixture was stirred
(600–900 C, model MAꢀG/6P; LETO plant, Russia)
°
in a flow of oxygen and on a device for nitrogen deterꢀ
mination with three similar electric ovens in a flow of
carbon dioxide. The elemental analysis data on the
percent content of C, H, and N of compounds deviate
from the theoretical data by no more than 0.4%.
for 1 h at –10°С and then for 1 h at room temperature.
The solvent was removed on a rotor evaporator, and
the oily residue was dissolved in chloroform (30 mL)
and successively washed with water, 5% NaHCO3, 1
N
HCl, and again with water (30 mL each). Then the
solution was evaporated, the oily residue was triturated
with diethyl ether, and the precipitate was filtered and
dried in a desiccator over P2O5 (15 mmHg). A chroꢀ
matographically homogenous white powder was
The analytical HPLC of the dipeptide GSBꢀ106
was performed using a Wellchrom 2001 chromatoꢀ
graphic system (KNAUER, Germany) on a Diasorbꢀ
130ꢀC16T column (4.6
×
150 0 mm, 7 m) in a gradiꢀ
μ
obtained. Yield: 1.3 g (87%); Rf 0.49 (A); mp 105–
ent of concentrations of isopropyl alcohol in 0.5%
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 38
No. 3
2012

248
GUDASHEVA et al.
20
1
20°С and evaporated. The crystallization of the resiꢀ
106°С, [α]D –8.5° (c 0.4, chloroform); H NMR:
due under diethyl ether yielded 265 mg (90%) of the
1.36 (9 H, s, ОC(CH3)3), 1.17–1.65 (6 H, m,
20
β
γ
δ
ε
С Н2С Н2С Н2 Lys), 2.97 (2 H, m, C H2 Lys), 3.80
product as white crystals; mp 153–155°С, [α]D
α
1
γ
0.4, DMF); H NMR: 1.27 (2 H, m, C H2
(1 H, m, С Н Lys), 5.07 (2 H, s, –ОСН2С6Н4Cl),
+6.25° (
c
6.73 (1 H, d,
two s, NH2 amide), 7.37 (1 H, t,
7.30–7.50 (4 H, m, ОСН2С6Н4Cl).
CF3COOH H–Lys(Z(Cl))–NH2. A solution of
J 7.7 Hz, NH Lys), 6.96 and 7.25 (2 H,
δ
Lys), 1.39 (2 H, m, C H2 Lys), 1.54 and 1.65 (2 H,
two m, C H2 Lys), 2.97 (2 H, m, C H2 Lys), 3.67 (2 H,
m, C H2 Ser), 4.11 (1 H, m, C H Ser), 4.24 (1 H, m,
C H Lys), 4.54 (2 H, s, –СН2С6Н5), 5.07 (2 H, s,
ОСН2С6Н4Cl), 7.14 and 7.45 (9 H, two m,
СН2С6Н5, –OСН2С6Н4Cl), 7.35 (1 H, t, 5.7 Hz,
N H Lys), 8.25 (3 H, s, br, N+H3 Ser), 8.57 (1 H, d,
ε
J
5.6 Hz, N H Lys),
β
ε
β
α
⋅
α
BocꢀLys(Z(Cl))–NH2 (400 mg, 1.00 mmol) in TFA
(3 mL) and dichloromethane (3 mL) was stirred for
30 min at room temperature. The solvent was removed
in vacuo, the residue was evaporated twice with diethyl
ether (10 mL) and dried over P2O5 in vacuo
(15 mmHg). The product was obtained as a trifluoroꢀ
⎯
⎯
J
ε
J
7.7 Hz, NH Lys).
HOOC(CH2)2CO–Ser(Bzl)ꢀLys(Z(Cl))–NH2. TEA
(0.06 mL, 0.44 mmol) was added to a solution of
CF3COOH Н–Ser(Bzl)ꢀLys(Z(Cl))–NH2 (240 mg,
acetate salt. Yield: 410 mg (99%); Rf 0.05 (F); mp
γ
109–111
(2 H, m, C H2 Lys), 1.70 (2 H, m, C H2 Lys), 2.99
°С
; 1H NMR: 1.31 (2 H, m, C H2 Lys), 1.40
δ β
·
0.40 mmol) in DMF (4 mL), the mixture was stirred
for 15 min, and succinic anhydride (60 mg, 0.6 mmol)
was added. The reaction mixture was stirred for 12 h,
and after the termination of the reaction (TLC conꢀ
trol) water (25 mL) was added. The white precipitate
was filtered, washed with water, and dried in vacuo
(15 mmHg) over P2O5. The yield of the product was
207 mg (88%); Rf 0.28 (A); mp 158–166°С; 1H NMR:
ε
α
(2 H, m, C H2 Lys), 3.68 (1 H, m, C H Lys), 5.08
(2 H, s, ОСН2С6Н4Cl), 7.31–7.52 (4 H, m,
ОСН2С6Н4Cl), 7.57 and 7.88 (2 H, two s, NH2 amide),
8.09 (3 H, s, NH+3 Lys).
BocꢀSer(Bzl) Lys(Z(Cl))–NH2. NMM (0.09 mL,
0.82 mmol) was added to a solution of BocꢀSer(Bzl)–
OH (240 mg, 0.82 mmol) in DMF (10 mL) at –15°С
after which IBCF (0.13 mL, 0.82 mmol) was added.
After 2–3 min, a solution of trifluoroacetic salt
Lys(Z(Cl))–NH2 (350 mg, 0.82 mmol) and NMM
γ
δ
1.25 (2 H, m, C H2 Lys), 1.35 (2 H, m, C H2 Lys),
1.52 and 1.68 (2 H, two m, C H2 Lys), 2.41 (4 H, m,
HOOCCH2СН2), 2.94 (2 H, m, C H2 Lys), 3.60 (2 H,
β
ε
β
α
(0.09 mL, 0.82 mmol) in DMF (10 mL) was added, m, C H2 Ser), 4.13 (1 H, m, C H Lys), 4.47 (1 H, m,
α
which was prepared on cooling (
reaction mixture was stirred for 1 h at –5
0
°
С
) beforehand. The
C H Ser), 4.48 (2 H, s,
ОСН2С6Н4Cl), 7.06 and 7.14 (2 H, two s, NH2
amide), 7.20–7.50 (9 H, m, СН2С6Н5,
ОСН2С6Н4Cl), 7.34 (1 H, t, 5.7 Hz, N H Lys),
7.7 Hz, NH Lys), 8.25 (1 H, d, 8.0 Hz,
⎯ СН2С6Н5), 5.07 (2 H, s,
…
–10°С and
⎯
for 1 h at room temperature. The resulting reaction
mixture was diluted with ethyl acetate (200 mL), sucꢀ
cessively washed with water (50 mL), 2% H2SO4
⎯
ε
⎯
J
7.92 (1 H, d,
NH Ser).
J
J
(30 mL), 3% K2CO3 (30 mL), and water (
2 50 mL),
×
and dried over Na2SO4. The solvent was removed in
vacuo, the residue was triturated with diethyl ether,
and the precipitate was filtered. A white powder was
obtained (510 mg) in which two components were
HOOC(CH2)2CO–SerꢀLys–NH2 (GSBꢀ104). To
a solution of HOOC(CH2)2CO–Ser(Bzl)ꢀLys(Z(Cl))–
NH2 (180 mg, 0.30 mmol) in DMF (10 mL), 10%
Pd/C (180 mg) was added, and the reaction mixture
was stirred in an atmosphere of hydrogen for 10 h until
the initial compound disappeared (TLC control).
Then, the catalyzer was filtered, and the filtrate was
evaporated. The product was obtained as a white powꢀ
identified by TLC, with
Rf 0.55 (A) and
Rf 0.60 (A).
The composition of the compound with
Rf 0.55 was
not determined. After separation on a column of silica
gel (Kieselgel 130–270, Aldrich), the elution with the
system chloroform—ethanol 9 : 1 was carried out. The
final product was obtained as a white powder. Yield:
20
der. Yield: 95 mg (100%, total yield 37%); Rf 0.11 (F);
370 mg (60%); Rf 0.60 (A); mp 127–129
°
С, [α]D
2.5° (c 0.4, DMF); H NMR: 1.15–1.85 (6 H, m,
β γ
20
mp 136–138
°
С, [α]D –15.75
°
(c
0.4. DMF). Found, %:
1
⎯
⎯
C 46.80; H 7.39; N 16.68; C13H24N4O6. Calculated, %:
δ
С Н2С Н2С Н2ꢀLys), 1.37 (9 H, s, ОC(CH3)3), 2.93
1
C 46.99; H 7.23; N 16.87. H NMR: 1.29 (2 H, m,
(H, s, –СН2С6Н5), 5.06 (1 H, s, ОСН2С6Н4Cl), 7.05
(1 H, d, 6.8 Hz, NH Ser), 7.09 and 7.25 (2 H, two s,
NH2 amide), 7.2–7.5 (9 H, m, СН2С6Н5,
5.6 Hz, N H Lys).
HꢀSer(Bzl)ꢀLys(Z(Cl))–NH2.
solution of BocꢀSer(Bzl)ꢀLys(Z(Cl))–NH2 (287 mg,
0.5 mmol) in TFA (2 mL) was stirred for 30 min at NH Ser).
γ
β
C H2 Lys), 1.53 (2 H, m, C H2 Lys), 1.79 (2 H, m,
J
δ
C H2 Lys), 2.51 (4 H, m, HOOCCH2СН2), 3.91 and
4.16 (2 H, two m, C H2 Ser), 4.40 (1 H, m, C H Lys),
4.55 (1 H, m, C H Ser), 7.21 (2 H, s, NH2 amide),
8.00 (1 H, d,
⎯
β
α
ε
ОСН2С6Н4Cl), 7.45 (1 H, t,
CF3COOH
J
α
⋅
A
J
7.6 Hz, NH Lys), 8.31 (1 H, d,
J
8.1 Hz,
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 38
No. 3
2012

DESIGN AND SYNTHESIS OF DIPEPTIDE MIMETICS
249
Synthesisof(HOOC(CH2)2CO
–
SerꢀLys
–
NH
–
)2(CH2)6
2CF3COOH ⋅ (H–Lys(Z(Cl))–)2(CH2)6. TFA
(GSBꢀ106)
(15 mL) was poured to (BocꢀLys(Z(Cl))–NH–)2(CH2)6
(1.42 g, 1.71 mmol), and the solution was stirred for 2 h.
After the termination of the reaction (TLC control), the
BocꢀLys(Z(Cl))–OSu. ꢀHydroxysuccinimide
N
(2.08 g, 18 mmol) was added to a solution of Bocꢀ
Lys(Z(Cl))–OH (7.60 g, 17 mmol) in THF (150 mL),
the solution was cooled to –5оС, and a solution of
DCC (3.50 g, 17 mmol) in THF (20 mL) was added
dropwise for 30 min; care was taken that the temperaꢀ
solution was evaporated with diethyl ether (
The resulting foam (solidified oil) of the product was tritꢀ
urated with diethyl ether. Yield: 1.2 g (90%); f 0.10 (C),
mp 87–90
1H NMR: 0.95–1.75 (20 H, m,
2
×
20 mL).
R
°С;
δ
β
γ
2
⎯
C H2C H2C H2Lys, –NHCH2(CH2)4CH2NH–),
ture was no higher than
0°С
. Then the reaction mixꢀ
2.98 (4 H, m, –NHCH2(CH2)4CH2NH–), 3.09 (4 H,
ture was stirred for 30 min at –5
°С
and for 4 h at room
ε
α
m, 2 C H2 Lys), 3.70 (2 H, m, 2 C H Lys), 5.07 (4 H,
s, 2 –ОСН2С6Н4Cl), 7.37 (2 H, t, 4.9 Hz,
temperature. The white DCU precipitate was filtered,
and the filtrate was evaporated. The resulting viscous
oil was dissolved in dichloromethane (30 mL). The
J
⎯
NH(CH2)6NH–), 7.45 (8 H, m, 2 –ОСН2С6Н4Cl),
8.23 (6 H, m, 2 N+H3 Lys), 8.59 (2 H, t,
J 6.0 Hz,
solution was kept for 24 h at
0 C, the DCU precipitate
°
ε
2 N H Lys).
was filtered, and the filtrate was evaporated in vacuo.
A chromatographically homogeneous oily product was
obtained. Yield: 6.80 g (84%); Rf 0.40 (A).
(BocꢀSer(Bzl)ꢀLys(Z(Cl))–NH)2(CH2)6. To a soluꢀ
tion of 2CF3COOH (Н–Lys(Z(Cl))–NH–)2(CH2)6
·
(1.13 g, 1.25 mmol) in DMF (25 mL), first DIEA
(0.46 mL, 2.64 mmol) and then BocꢀSer(Bzl)–OSu
(2.48 g, 2.60 mmol) were added. The reaction mixture
was stirred for 1 h at room temperature, DMAPA
(0.5 mL) was added, and the mixture was stirred for
15 min. The reaction mixture was diluted with ethyl
acetate (150 mL), successively washed with water, 3%
H2SO4, 2% K2CO3, and again with water (40 mL
each), dried over Na2SO4, and evaporated. The recrysꢀ
BocꢀSer(Bzl)–OSu.
(1.17 g, 11 mmol) was added to a solution of Bocꢀ
Ser(Bzl)–OH (2.89 g, 10 mmol), the solution was
NꢀHydroxysuccinimide
cooled to –5°С, and a solution of DCC (2.07 g,
10 mmol) in THF (20 mL) was added dropwise for
20 min. The reaction mixture was stirred for 3 h at
room temperature. The white DCU precipitate was
filtered, the filtrate was evaporated, and the resulting
oil was crystallized from isopropyl alcohol. The crysꢀ
tals that precipitated were filtered and dried in vacuo
tallization of the crystalline residue from methanol
yielded 1.25 g (75%) of the product as a white powder;
(15 mmHg) over CaCl2. Yield: 3.20 g (83%);
R
f 0.60 (А),
20
20
Rf 0.40 (А), Rf 0.60 (E); mp 143–151
0.4; DMF)
ОC(CH3)3,
NHCH2(CH2)4CH2NH–), 2.93 (8 H, m, 2 C H2
°
С, [α]D –2.5
°
Rf 0.70 (E); mp 106–108
°
С, [α]D –7.0
°
(с
1;
1H NMR: 0.93–1.83 (38 H, m,
1
(
2
⎯
с
⎯
;
DMF).; H NMR: 1.39 (9 H, s, –ОC(CH3)3), 2.42
β γ δ
β
2
–C H2C H2C H2ꢀ
Lys,
ε
(4 H, m, HOOC–CH2СН2), 3.70 (2 H, m, C H2 Ser),
α
4.39 (1 H, m, C H Ser), 4.50 (2 H, s, СН2С6Н5),
7.20–7.50 (5 H, m, –СН2С6Н5), 7.72 (1 H, d, J 8.0 Hz,
NH Ser).
(BocꢀLys(Z(Cl))–NH–)2(CH2)6. A solution of
hexamethylenediamine (380 mg, 3.30 mmol) in ethyl
acetate (5 mL) was added to a solution of Bocꢀ
Lys, –NHCH2(CH2)4CH2NH–), 3.59 (4 H, m,
β
α
α
2 C H2 Ser), 4.20 (4 H, m, 2 C H Lys, 2 C H Ser),
4.47 (4 H, s, 2 –СН2С6Н5), 5.07 (4 H, s,
2
⎯
ОСН2С6Н4Cl), 7.00 (2 H, d,
J 8.1 Hz, 2 NH Ser),
7.19–7.50 (20 H, m, 2 –СН2С6Н5, 2 –ОСН2С6Н4Cl,
ε
Lys(Z(Cl))–OSu (3.50 g, 6.80 mmol) in ethyl acetate –NH(CH2)6NH–), 7.77 (2 H, t,
J 5.8 Hz, 2 N H
(40 mL). The reaction mixture was stirred for 1 h and
evaporated. The residue was dissolved in ethyl acetate
(100 mL) and washed successively with 1 N H2SO4
Lys), 7.91 (2 H, d,
J
7.8 Hz, 2 NH Lys).
(H–Ser(Bzl)ꢀLys(Z(Cl))–NH–)2
2CF3COOH
⋅
(CH2)6. (BocꢀSer(Bzl)ꢀLys(Z(Cl))–NH–)2 (CH2)6
(1.00 g, 0.84 mmol) was dissolved in TFA (3 mL), and
the solution was stirred for 40 min and evaporated.
(50 mL), 3% K2CO3 (50 mL), and water (
2 × 50 mL).
The filtrate was dried over Na2SO4 and evaporated.
The resulting solid crystalline residue was dried in
vacuo (15 mmHg) over CaCl2 to yield 2.40 g (85%) of
20
The residue was reevaporated with diethyl ether (
2
×
15 mL). After the trituration of oil with diethyl ether,
0.92 g (92%) of a chromatographically homogeneous
beige powder was obtained; Rf 0.33 (А), Rf 0.50 (C);
20
the product; Rf 0.45 (A), mp 101–103°С, [α]D –7.3°
γ
1
(
c
0.5, DMF); H NMR: 1.29 (8 H, m, 2 C H2 Lys,
mp 117–122
1.15–1.75 (20 H, m, 2
NHCH2(CH2)4CH2NH–), 3.00 (8 H, m, 2 C H2 Lys,
°
С, [α]D +4.75° (с
0.4; DMF); 1H NMR:
β γ δ
⎯
2
⎯
NH(CH2)2CH2CH2(CH2)2NH–), 1.42 (18 H, s,
δ
⎯
C H2C H2C H2ꢀLys,
⎯
ОC(CH3)3), 1.55 (8 H, m, 2 C H2 Lys,
ε
NHCH2CH2(CH2)2CH2CH2NH–), 1.79 (4 H, m,
⎯
β
ε
β
2 C H2 Lys), 2.96 (4 H, m, 2 C H2 Lys), 3.20 (4 H, m, –NHCH2(CH2)4CH2NH–), 3.90 (4 H, m, 2 C H2
α
α
α
–NHCH2(CH2)4CH2NH–), 4.53 (2 H, m, 2 C H Lys), Ser), 4.25 (4 H, m, 2 C H Lys, 2 C H Ser), 4.46 (4 H,
s, 2 СН2С6Н5), 5.10 (4 H, s, 2 –ОСН2С6Н4Cl), 7.04
(6 H, s, 2 N+H3 Ser), 7.20–7.50 (20 H, m,
СН2С6Н5, 2 –ОСН2С6Н4Cl, NH(CH2)6NH–),
⎯
5.34 (4 H, c, 2 СН2С6Н4Cl), 7.10 (2 H, d,
J
7.8 Hz,
5.7 Hz, 2 N H Lys), 8.01
4.7 Hz, NH(CH2)6NH).
ε
2NH Lys), 7.39 (2 H, t,
(2 H, t,
J
J
2
⎯
⎯
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 38
No. 3
2012

250
GUDASHEVA et al.
ε
J 5.6 Hz, 2 N H Lys), 7.88 (2 H, d,
Synthesis of HOOC(CH2)2CO–MetꢀSer
–NH2
7.74 (2 H, t,
J
(GSBꢀ207)
7.6 Hz, 2 NH Lys).
(HOOC(CH2)2CO–Ser(Bzl)ꢀLys(Z(Cl))–NH–)2
(CH2)6. TEA (0.1 mL, 1.50 mmol) was added to a susꢀ
pension of 2 CF3COOH (Н–Ser(Bzl)ꢀLys(Z(Cl))–
BocꢀMetꢀSer–OMe. TEA (5.7 mL, 41 mmol) was
added under stirring to a solution of H–Ser–OMe
hydrochloride (6.38 g, 41 mmol) and BocꢀMet–ONp
(15.8 g, 41 mmol) in DMF (50 mL). The reaction
mixture was kept for 18 h at room temperature. After
the termination of the reaction (TLC control),
DMAPA (2 mL) was added, and the solution was
stirred for 30 min and evaporated. Ethyl acetate
(150 mL) and water (100 mL) were added to the resiꢀ
due. The organic phase was successively washed with
·
NH)2(CH2)6 (0.80 g, 0.72 mmol) in DMF (5 mL), the
solution was stirred for 30 min, and succinic anhydride
(0.408 g, 4.08 mmol) was added. The reaction mixture
was stirred for 3 h. After the disappearance of the startꢀ
ing compound (TLC control), the reaction mixture
was poured into cold water, and the white precipitate
was filtered, washed with water, and dried over CaCl2
in vacuo (15 mmHg). The yield of the product was
0.72 g (91%); Rf 0.40 (D), Rf 0.95 (C), mp 128–
20
3% K2CO3 (3
×
60 mL) and 2% HCl (100 mL), dried
over Na2SO4, and evaporated. The residue was disꢀ
solved in diethyl ether (50 mL), and heptane (25 mL)
was added to turbidity. The crystals that precipitated
were filtered, washed with heptane, and dried. The
yield of the product was 11.6 g (80.7%); Rf 0.80 (C),
20
130
0.4; DMF);
δ
°
С
(from ethyl acetate), [α]D –11.25
°
(
с
1H NMR: 0.95–1.78 (20 H, m,
β
γ
2
⎯
C H2C H2C H2 Lys,
⎯NHCH2(CH2)4CH2NH–),
R
f 0.67 (D); mp 65–68
°
С, [α]D –20.75 (с 0.4,
°
2.42 (8 H, m, 2 HOOCCH2СН2), 2.97 (8 H, m,
methanol); 1H NMR: 1.36 (9 H, s, –OС(СН3)3), 1.81
ε
2 C H2 Lys, ⎯NHCH2(CH2)4CH2NH–), 3.60 (4 H, m,
β
(2 H, m, С Н2 Met), 2.07 (3 Н, s, –SCH3), 2.45 (2 H,
β
α
2 C H2 Ser), 4.14 (2 H, m, 2 C H Lys), 4.44 (2 H, m,
2 C H Ser), 4.45 (4 H, s, 2 –СН2С6Н5), 5.07 (4 H, s,
СН2С6Н4Cl), 7.20–7.49 (20 H, m, 2 –СН2С6Н5,
2 –СН2С6Н4Cl, –NH(CH2)6NH–), 7.53 (2 H, t,
γ
t,
J 5.3 Hz, С Н2 Met), 3.62 (3 Н, s, –OCH3), 3.70
α
β α
(2 H, m, C H2 Ser), 4.08 (1 H, m, С Н Met), 4.33
2
⎯
α
(1 H, m, C H Ser), 5.08 (1 H, m, –OH Ser), 6.98
(1 H, d,
J 9.2 Hz, NH Met), 8.07 (1 H, d, J 8.0 Hz,
ε
J
5.6 Hz, 2 N H Lys), 7.89 (2 H, d,
J 7.3 Hz, 2 NH
NH Ser).
Lys), 8.24 (2 H, d, 8.1 Hz, 2 NH Ser).
J
BocꢀMetꢀSer–NH2. A solution of BocꢀMetꢀSerꢀ
OMe (4 g, 11.4 mmol) in methanol (10 mL) was satuꢀ
(HOOC(CH2)2COꢀSerꢀLys–NH–)2(CH2)6 (GSBꢀ
106). To a solution of (HOOC(CH2)2CO–Ser(Bzl)ꢀ
Lys(Z(Cl))–NH)2(CH2)6 (600 mg, 0.50 mmol) in
methanol (50 mL), 10% Pd/C (500 mg) was added,
and the solution was stirred in an atmosphere of
hydrogen for 6–12 h until the starting compound
completely disappeared (TLC control). After the terꢀ
mination of hydration, the catalyzer was filtered, the
filtrate was evaporated, and the solid residue was
washed with diethyl ether and dried in vacuo
(15 mmHg). If the product was isolated as oil, the resꢀ
idue was then triturated with diethyl ether. The yield of the
product was 450 mg (90%, total yield 23%); Rf 0.10 (F),
20
rated with ammonia at
0 С and kept for 3 days at room
°
temperature in a closed flask. Then the solution was
evaporated, and the residue was crystallized with a
mixture of diethyl ether (30 mL) and ethanol (5 mL).
The crystals formed were filtered off and dried. The
yield of the product was 3.19 g (83.4%); Rf 0.71 (C),
20
R
f 0.57 (D); mp 120–121
°
С, [α]D –12.25
0.4; МеОН); H NMR: 1.38 (9 H, s, OC(СН3)3),
1.82 (2 H, m, С Н2 Met), 2.03 (3 Н, s, –SCH3), 2.46
°
1
(с
⎯
β
γ
β
(2 H, t, J 5.2 Hz, С Н2 Met), 3.56 (2 H, m, C H2 Ser),
α α
4.00 (1 H, m, С Н Met), 4.17 (1 H, m, C H Ser), 7.12
and 7.23 (2 H, two s, NH2 amide), 7.14 (1 H, d,
J
9.1 Hz, NH Met), 7.65 (1 H, d,
J 8.1 Hz, NH Ser).
Rf 0.40 (C),
τ
= 15
1 min; mp 143–145°C, [α]D
HOOC(CH2)2CO–MetꢀSer–NH2 (GSBꢀ207).
A
⎯
24.7 0.4; DMF). Found, %: C 51.54; H 8.03;
° (с
solution of BocꢀMetꢀSerꢀNH2 (2.99 g, 8.9 mmol) in
TFA (15 mL) was stirred for 1 h and evaporated. The
residue was triturated with diethyl ether, and the crysꢀ
tals formed were filtered and washed with diethyl
ether. Then the crystals were dissolved in 50% ethanol
(20 mL), treated with Amberlite IRAꢀ900 resin (–OH
form) to pH ~ 9, filtered, and evaporated. The crystals
were evaporated again with the addition of DMF. The
residue was dissolved in DMF (10 mL), and succinic
anhydride (1 g, 10 mmol) was added. The reaction
mixture was stirred for 12 h, and water (25 mL) was
added. The white precipitate was filtered, washed with
water, and dried. The yield of the product was 0.96 g
N 15.18; C32H58N8O12. Calculated, %: C 51.47;
γ
H 7.77; N 15.01. 1H NMR: 1.10 (4 H, m, 2 C H2 Lys),
δ
β
1.34 (4 H, m, 2 C H2 Lys), 1.70 (4 H, m, 2 C H2 Lys),
1.00–1.75 (8 H, m, NHCH2(CH2)4CH2NH–),
⎯
2.42 (8 H, m, 2 HOOCCH2СН2), 2.74 (4 H, m,
ε
2 C H2 Lys), 3.00 (4 H, m, –NHCH2(CH2)4CH2NH–),
β
α
3.63 (4 H, m, 2 C H2 Ser), 4.14 (2 H, m, 2 C H Lys),
α
4.27 (2 H, m, 2 C H Ser), 7.68 (2 H, t,
⎯
J
4.9 Hz,
NH(CH2)6NH–), 7.72 (4 H, s br, 2 N H2 Lys), 7.95
(2 H, d, 7.8 Hz, 2 NH Lys), 8.10 (2 H, d, 8.1 Hz,
2 NH Ser).
ε
J
J
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 38
No. 3
2012

DESIGN AND SYNTHESIS OF DIPEPTIDE MIMETICS
251
β
2 ⎯ОС(СН3)3), 1.80 (4 H, m, 2 С Н2 Met), 2.02 (6 H,
s, 2 –SCH3), 2.44 (4 H, t,
(32%; total yield 21.5%); Rf 0.45 (C), Rf 0.11 (D);
γ
J 5.5 Hz, 2 С Н2 Met), 3.02
20
mp 101–105
°
С, [α]D –14.25
°
(
с
0.4; МеОН)
.
.
(4 H, m, –NHCH2(CH2)5CH2NH–), 3.53 (4 H, m,
Found, %: C 42.85; H 6.50; N 12.68; C12H21N3O6S
Calculated, %: C 42.99; H 6.27; N 12.54. H NMR:
β
α
2 C H2 Ser), 3.99 (2 H, m, 2 С Н Met), 4.17 (2 H, m,
1
α
β
2 С Н Ser), 7.12 (2 H, d,
(2 H, d, 8.0 Hz, 2 NH Ser), 7.71 (2 H, t,
NH(CH2)7NH–).
J
9.1 Hz, 2 NH Met), 7.68
1.75 and 1.90 (2 H, two m, С Н2 Met), 2.03 (3 Н, s,
J
J
5.5 Hz,
⎯
SCH3), 2.41 (4 H, m, HOOCCH2СН2), 2.44(2 H, t,
⎯
γ
β
J
5.2 Hz, С Н2 Met), 3.58 (2 H, m, C H2 Ser), 4.31
α α
(1 H, m, С Н Met), 4.16 (1 H, m, C H Ser), 4.90
(1 H, m, –OH Ser), 7.08 and 7.17 (2 H, two s, NH2
(HOOC(CH2)2CO–MetꢀSer–NH–)2(CH2)7 (GSBꢀ
214). (BocꢀMetꢀSer–NH)2(CH2)7 (3.14 g, 4.1 mmol)
was dissolved in TFA (20 mL), and the solution was
stirred for 30 min and evaporated. The residue was tritꢀ
urated with diethyl ether, decanted, dissolved in 50%
ethanol (50 mL), and treated with Amberlite IRAꢀ900
resin (OHꢀform) to pH ~ 10. The solution was evapoꢀ
rated and evaporated again with DMF (40 mL). The
residue was dissolved in DMF (10 mL), and succinic
amide), 8.16 (1 H, d,
d, 8.2 Hz, NH Ser).
J 9.1 Hz, NH Met), 7.75 (1 H,
J
Synthesis of (HOOC(CH2)2CO–MetꢀSer–NH–)2(CH2)7
(GSBꢀ214)
BocꢀMetꢀSer–N2H3. Hydrazine hydrate (3 mL)
was added to a solution of BocꢀMetꢀSerꢀOMe (6.31 g, anhydride (1 g, 10 mmol) was added. The reaction
18 mmol) (obtained earlier during the synthesis of mixture was stirred for 2 h and evaporated. The residue
GSBꢀ207) in MeOH (10 mL), and the solution was was triturated with acetone and filtered. Ethanol
kept for 2 h. After the disappearance of the starting
compound (TLC control), water (10 mL) was added
to the reaction mixture, and the mixture was kept at
(30 mL) was added to the solid residue, and the mixꢀ
ture was boiled for 30 min (the residue does not disꢀ
solve completely), kept for 30 min at –20°С, filtered
⎯
20°C for 18 h. The precipitate was filtered, washed
off, and successively washed with cold ethanol, diethyl
ether, and hexane. After drying, 1.75 g of the product
(100%, total yield 46%) was obtained; Rf 0.57 (C),
20
with cooled aqueous methanol, and airꢀdried. The
yield of the product was 6.31 g (100%); Rf 0.66 (C),
20
Rf 0.47 (D); mp 160–162
°
С, [α]D –21.75
°
(с
0.4;
Rf 0.0 (D); mp 162–163
°
С, [α]D –9.00 (с 0.4;
°
1
DMF). Found, %: C 48.72; H 7.30; N 10.88;
C31H54N6O12S2. Calculated, %: C 48.56; H 7.05;
N 10.97. 1H NMR: 1.20–1.40 (10 H, m,
МеОН); H NMR: 1.37 (9 H, s, –OС(СН3)3), 1.80
β
(2 H, m, С Н2 Met), 2.02 (3 H, s, –SCH3), 2.43 (2 H,
γ
β
t,
(1 H, m, С Н Met), 4.19 (1 H, m, С Н Ser), 4.39
(2 H, d, 4.0 Hz, –NH–NH2), 4.88 (1 H, t, 4.0 Hz,
–OH Ser), 7.07 (1 H, d, 9.1 Hz, NH Met), 7.69
(1 H, d, 8.0 Hz, NH Ser), 9.02 (1 H, t,
NH–NH2).
(BocꢀMetꢀSer–NH–)2(CH2)7. To BocꢀMetꢀSer–
N2H3 (5.23 g, 14.9 mmol) in DMF (25 mL) cooled to
–20 , first HCl in dioxane (~4 M, 50 mL) and then
ꢀbutyl nitrite (1.75 mL, 15 mmol) were added; the
temperature was maintained at –20 –30 . The
J 5.2 Hz, С Н2 Met), 3.54 (2 H, m, C H2 Ser), 4.00
⎯
NHCH2(CH2)5CH2NH–), 1.76 and 1.89 (4 H,
α α
β
two m, 2 С Н2 Met), 2.02 (6 H, s, 2 –SCH3), 2.40
J
J
γ
(8 H, m, 2 HOOCCH2СН2), 2.42 (4 H, m, 2 С Н2 Met),
J
J
J 4.0 Hz, 3.01 (4 H, m, –NHCH2(CH2)5CH2NH–), 3.55 (4 H,
β α
⎯
m, 2 C H2 Ser), 4.17 (2 H, m, 2 С Н Ser), 4.30 (2 H,
α
m, 2 С Н Met), 7.61 (2 H, m, –NH(CH2)7NH–), 7.76
(2 H, d,
J 8.1 Hz, 2 NH Ser), 8.16 (2 H, d, J 9.2 Hz,
°С
2 NH Met).
n
Study of the biological activity in vitro. Immortalized
murine hippocampal cells of line HT22 were seeded in
96ꢀwell plates with a density of 3.5 thousand cells per
well in medium DMEM (HyClon) containing 5% calf
°С…
°С
reaction mixture was kept for 3 min after which first
DIEA (7.7 mL, 45 mmol) and then a solution of hepꢀ
tamethylenediamine (0.91 g, 7 mmol) in DMF (3 mL)
were added. The reaction mixture was kept for 12 h at
room temperature after which ethyl acetate (250 mL)
and water (200 mL) were added. The organic phase
fetal serum (Invitrogen) and 2 mM
Lꢀglutamine
(ICN) and incubated at 37 in 5% СО2 until a
°
С
monolayer formed. Oxidative stress was modeled using
H2O2 at a final concentration of 1.5 mM. Cells were
incubated with H2O2 in an atmosphere of 5% СО2 at
was washed with 2% HCl (
2
×
100 mL) and 3%
NaНСO3 (100 mL) and evaporated to a volume of
~60 mL. A gradual settling of a precipitate was
observed, and the amount of the precipitate increased
as the solution was cooled to –20°С. The precipitate
was filtered at room temperature, washed with ethyl
37 С for 30 min. Then the medium was replaced by
°
normal medium, and after 4 h cell viability was deterꢀ
mined using 3ꢀ(4,5ꢀdimethylthiazolꢀ2ꢀyl)ꢀ2,5 dipheꢀ
nyltetrazolium bromide (MTT) (Sigma). Optical
absorption was measured on a Multiscan EX spectroꢀ
photometer (Thermo) at 600 nm. Peptides were introꢀ
duced 24 h prior to the addition of hydrogen peroxide.
BDNF (Sigma) at a final concentration of 50 ng/mL
was used as a positive control.
acetate, and dried. The yield of the product was 3.3 g
20
(57%); Rf 0.77 (C), Rf 0.59 (D); mp 160–162
–28.75
m, NHCH2(CH2)5CH2NH–), 1.37 (18 H, s,
°С, [α]D
°
(с
0.4, methanol); 1H NMR: 1.18–1.40 (10 H,
⎯
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 38
No. 3
2012

252
GUDASHEVA et al.
Each experiment was carried out in 16 replicas immobility (s) in the group Imipramine. For the calꢀ
(16 wells). The neuroprotective activity was deterꢀ culation, the median values of samples were used. The
mined using the average values. The activity in experꢀ differences between control and experimental groups
iments on the counteraction to oxidative stress was were determined using the nonparametric Mann–
calculated by the formula:
Whitney
Uꢀtest.
А
(%) = (Dexp act)/(Dcontr – Dact) × 100,
–
D
where Dexp is the optical absorption of solution in
experiment, Dact is the optical absorption of a solution
of active control (with Н2О2), and Dcontr is the optical
absorption of a solution of passive control (without
Н2О2). The statistical significance of the difference
between the experiment and control was determined
ACKNOWLEDGMENTS
This work was partially supported by the Russian
Foundation for Basic Research, project no. 12ꢀ04ꢀ
01225.
REFERENCES
by the Student’s tꢀtest.
Study of the biological activity in vivo. The activity
was examined in the “forced swimming” test [17].
Balb/c male mice weighing 20–22 g from a nursery of
laboratory animals in Pushchino (10 animals in each
experimental group) were used. Animals were housed
for 7 days before the test in a vivarium of the Zakusov
Research Institute of Pharmacology, Russian Acadꢀ
emy of Medical Sciences, under standard conditions
with free access to water and briquetted feed under a
natural light/dark regime. All experiments were carꢀ
ried out in the autumn–winter period, from 10 to 13 a.
m. On the first day of the experiment, animals were
housed for 10 min in a cylinder (height 30 cm, diameꢀ
1. Zuccato, Ch. and Cattaneo, E., Nature Rev./Neurology
,
2009, vol. 5, pp. 311–322.
2. Schmidt, H.D. and Duman, R.S., Neuropsychopharꢀ
macology, 2010, vol. 35, pp. 2378–2391.
3. (No authors listed), A Controlled Trial of Recombinant
Methionyl Human BDNF in ALS: The BDNF Study
Group (Phase III), Neurology, 1999, vol. 52, pp. 1427–
1433.
4. O’Leary, P.D. and Huges, R.A., J. Neurochem., 1998,
vol. 70, pp. 1712–1721.
5. Fletcher, J.M. and Hughes, R.A., J. Pept. Sci., 2006,
vol. 12, pp. 515–521.
6. O’Leary, P.D. and Huges, R.A., J. Biol. Chem., 2003,
ter 10 cm) filled with water (22 С) to 65%. After
°
vol. 278, pp. 25738–25744.
60 min, peptides were injected to the animals. After
24 h, the animals were placed again in the same condiꢀ
tions, and the time during which the animal retained a
typical form of immobility (refusal from active defenꢀ
sive and exploratory behavior) was registered for
5 min. The antidepressant imipramine at a dose of
25 mg/kg served as a positive control; it was injected
intraperitoneally 60 min prior to the test. The peptides
were injected intraperitoneally in the form of aqueous
solutions (GSBꢀ104 and GSBꢀ106) or a water–Tween
suspension (GSBꢀ207 and GSBꢀ214), 5 mL per 1 kg
of body weight of the mouse. Control animals were
injected with either distilled water or 0.05% aqueous
Tweenꢀ80 (Merck).
7. Fletcher, J.M., Morton, C.M., Zwar, R.A., Murray, S.S.,
O’Leary, P.D., and Huges, R.A., J. Biol. Chem., 2008,
vol. 283, pp. 33375–33383.
8. Fletcher, J.M. and Hughes, R.A., Bioorg. Med. Chem.
2009, vol. 17, pp. 2695–2702.
,
9. Massa, S.M., Yang, T., Xie, Y., Shi, J., Bilgen, M.,
Joyce, J.N., Nehama, D., Rajadas, J., and Longo, F.M.,
J. Clin. Invest., 2010, vol. 120, no. 5, pp. 1774–1785.
10. Robinson, R.C., Radziejewski, C., Spraggon, G.,
Greenwald, J., Kostura, M.R., Burtnick, L.D., Stuꢀ
art, D.I., Choe, S., and Jones, E.Y., Protein Sci., 1999,
vol. 8, pp. 2589–2597.
11. Gudasheva, T.A., Antipova, T.A., and Seredenin, S.B.,
Dokl. Biol. Sci., 2010, vol. 434, pp. 262–265.
The data on the duration of immobility were proꢀ
cessed using nonparametric statistics. The data were
presented, not as arithmetic means, but as median valꢀ
ues of samples representing a value that is in the midꢀ
dle of the series of immobility durations for all animals
in the group in the order of increasing value. To access
the scatter of sampling, instead of rootꢀmeanꢀsquare
deviations, the lower and the upper quartiles (the valꢀ
ues of a parameter being measured that cut off, respecꢀ
tively, one fourth of its values from below and one
fourth from the top in this series) were used. The activꢀ
ity of the compounds was calculated by the formula:
12. Gudasheva, T.A., Lezina, V.P., Kir’yanova, E.P.,
Troitskaya, V.S., Kolik, L.G., and Seredenin, S.B.,
Khim.ꢀFarm. Zh., 2006, vol. 40, no. 7, pp. 21–26.
13. Seredenin, S.B. and Gudasheva, T.A., RF Patent
No. 2410392, Byull. Izobret., No. 3, 2011.
14. Anderson, G.W., Zimmerman, J.E., and Callahan, F.M.,
J. Am. Chem. Soc., 1967, vol. 89, pp. 5012–5017.
15. Jackson, G.R., WerrbachꢀPerez., Ezell, E.L, Post, J.F.,
and PerezꢀPolo, J.R., Brain Res., 1992, vol. 592, no. 1,
pp. 239–248.
16. Altar, C.A., Trends Pharmacol. Sci., 1999, vol. 20, no. 2,
pp. 59–61.
17. Porsolt, R.D., Anton, G., Blavet, N., and Jalfre, M.,
Eur. J. Pharmacol., 1978, vol. 47, pp. 379–391.
18. Holmes, A., Yang, R.J., Murphy, D.L., and Crawley, J.N.,
А
(%) = (Tcontr
where Tcontr is the duration of immobility (s) in the
control group, exp is the duration of immobility (s) in
the experimental group, and Timꢀin is the duration of
– Texp)/(Tcontr – Timꢀin) × 100,
T
Neuropsychopharmacology
, 2002, vol. 27, no. 6,
pp. 914–923.
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 38
No. 3
2012