Synthesis and biological activity of cyclolinopeptide A analogues modified with γ4-bis(homo-phenylalanine)

Article abstract of DOI:10.1016/j.bmc.2017.05.063

Cyclolinopeptide A (CLA), an immunosuppressive nonapeptide derived from linen seeds, was modified with S or R-γ⁴-bis(homo-phenylalanine) in positions 3 or 4, or both 3 and 4. These modifications changed the flexibility of new analogues and distribution of intramolecular hydrogen bonds. Analogues 11 c(Pro¹-Pro²-Phe³-S-γ⁴-hhPhe⁴-Leu⁵-Ile⁶-Ile⁷-Leu⁸-Val⁹), 13 c(Pro¹-Pro²-S-γ⁴-hhPhe³-R-γ⁴-hhPhe⁴-Leu⁵-Ile⁶-Ile⁷-Leu⁸-Val⁹) and 15 c(Pro¹-Pro²-R-γ⁴-hhPhe³-Phe⁴-Leu⁵-Ile⁶-Ile⁷-Leu⁸-Val⁹) existed as a mixture of stable cis/trans isomers of Pro-Pro peptide bond. The comparison of the relative spatial orientations in crystal state of the two carbonyl groups, neighboring γ-amino acids, revealed conformational similarities to α-peptides. The addition of two -CH₂- groups in γ-amino acids led to a more rigid conformation, although a more flexible one was expected. A significant difference in the relative orientation of the carbonyl groups was found for cyclic γ-peptides with a dominance of an antiparallel arrangement. As carbonyl groups may be engaged in the interactions with plausible receptors through hydrogen bonds, a similar biological activity of the modified peptides was expected. Our biological studies showed that certain cyclic, but not the corresponding linear peptides, lowered the viability of peripheral blood mononuclear cells (PBMC) at 100?μg/mL concentration. The proliferation of PBMC induced by phytohemagglutinin A (PHA) was strongly inhibited by cyclic peptides only, in a dose-dependant manner. On the other hand, lipopolysaccharide (LPS)-induced tumor necrosis factor alpha (TNF-α) production in whole blood cell cultures was inhibited by both linear and cyclic peptides. Peptide 15 c(Pro¹-Pro²-R-γ⁴-hhPhe³-Phe⁴-Leu⁵-Ile⁶-Ile⁷-Leu⁸-Val⁹) blocked the expression of caspase-3, inhibited the expression of caspases-8 and -9 in 24?h culture of Jurkat cells, and caused DNA fragmentation in these cells, as an indicator of apoptosis. Thus, we revealed a new mechanism of immunosuppressive action of a nonapeptide.

Full text of DOI:10.1016/j.bmc.2017.05.063

Accepted Manuscript
Synthesis and biological activity of cyclolinopeptide A analogues modified with
4
γ -bis(homo-phenylalanine)
Karol Jędrzejczak, Paweł Hrynczyszyn, Małgorzata Szczesio, Jolanta Artym,
Tomasz Jastrząbek, Maja Kocięba, Marek Gł ówka, Krzysztof Huben, Iwona
Kochanowska, Michał Zimecki, Janusz Zabrocki, Stefan Jankowski, Beata
Kolesińska
PII:
S0968-0896(16)31140-3
DOI:
http://dx.doi.org/10.1016/j.bmc.2017.05.063
BMC 13782
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
6 November 2016
28 May 2017
30 May 2017
Please cite this article as: Jędrzejczak, K., Hrynczyszyn, P., Szczesio, M., Artym, J., Jastrząbek, T., Kocięba, M.,
Gł ówka, M., Huben, K., Kochanowska, I., Zimecki, M., Zabrocki, J., Jankowski, S., Kolesińska, B., Synthesis and
4
biological activity of cyclolinopeptide A analogues modified with γ -bis(homo-phenylalanine), Bioorganic &
Medicinal Chemistry (2017), doi: http://dx.doi.org/10.1016/j.bmc.2017.05.063
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Synthesis and biological activity of cyclolinopeptide A analogues modified with
4
γ -bis(homo-phenylalanine)
a, *
a
c
b
Karol Jędrzejczak , Paweł Hrynczyszyn , Małgorzata Szczesio , Jolanta Artym , Tomasz
a
b
c
a
b
Jastrząbek , Maja Kocięba , Marek Główka , Krzysztof Huben , Iwona Kochanowska ,
b
a
a
a
Michał Zimecki , Janusz Zabrocki , Stefan Jankowski , Beata Kolesińska
a
Institute of Organic Chemistry, Faculty of Chemistry, Lodz University of Technology, Żeromskiego 116, 90-
24, Łódź, Poland
9
b
Institute of Immunology and Experimental Therapy, Polish Academy of Science, R. Weigla 12, 53-114,
Wrocław, Poland
c
Institute of General and Ecological Chemistry, Faculty of Chemistry, Lodz University of Technology,
Żeromskiego 116, 90-924, Łódź, Poland
*
Corresponding author:
E-mail address: jendrzejczak@interia.pl (Karol Jędrzejczak).
Abstract
Cyclolinopeptide A (CLA), an immunosuppressive nonapeptide derived from linen seeds, was
4
modified with S or R-γ -bis(homo-phenylalanine) in positions 3 or 4, or both 3 and 4. These
modifications changed the flexibility of new analogues and distribution of intramolecular
1
2
3
4
4
5
6
7
8
9
hydrogen bonds. Analogues 11 c(Pro -Pro -Phe -S-γ -hhPhe -Leu -Ile -Ile -Leu -Val ), 13
1
2
4
3
4
4
5
6
7
8
9
1
2
4
c(Pro -Pro -S-γ -hhPhe -R-γ -hhPhe -Leu -Ile -Ile -Leu -Val ) and 15 c(Pro -Pro -R-γ -
3
4
5
6
7
8
9
hhPhe -Phe -Leu -Ile -Ile -Leu -Val ) existed as a mixture of stable cis/trans isomers of
Pro-Pro peptide bond. The comparison of the relative spatial orientations in crystal state of the
two carbonyl groups, neighboring γ-amino acids, revealed conformational similarities to α-
peptides. The addition of two -CH - groups in γ-amino acids led to a more rigid conformation,
2
although a more flexible one was expected. A significant difference in the relative orientation
of the carbonyl groups was found for cyclic γ-peptides with a dominance of an antiparallel
arrangement. As carbonyl groups may be engaged in the interactions with plausible receptors
through hydrogen bonds, a similar biological activity of the modified peptides was expected.
Our biological studies showed that certain cyclic, but not the corresponding linear peptides,
lowered the viability of peripheral blood mononuclear cells (PBMC) at 100 μg/mL
concentration. The proliferation of PBMC induced by phytohemagglutinin A (PHA) was
strongly inhibited by cyclic peptides only, in a dose-dependant manner. On the other hand,
lipopolysaccharide (LPS)-induced tumor necrosis factor alpha (TNF-α) production in whole
1
2
blood cell cultures was inhibited by both linear and cyclic peptides. Peptide 15 c(Pro -Pro -R-
4
3
4
5
6
7
8
9
γ -hhPhe -Phe -Leu -Ile -Ile -Leu -Val ) blocked the expression of caspase-3, inhibited the
expression of caspases-8 and -9 in 24h culture of Jurkat cells, and caused DNA fragmentation
1

in these cells, as an indicator of apoptosis. Thus, we revealed a new mechanism of
immunosuppressive action of a nonapeptide.
Keywords: Cyclolinopeptide A, γ-amino acids, cyclic peptides, immunosuppression,
apoptosis, PBMC, caspase.
Abbreviations:
Boc, tert-Butoxycarbonyl; C8, octyl stationary phase; C18, octadecyl stationary phase; CLA,
cyclolinopeptide A; COSY-DQF, double quantum filtered proton spin correlation; CsA,
cyclosporine A; DIPEA, N,N-diisopropylethylamine; DMF, dimethylformamide; DMSO,
dimethyl sulfoxide; hhPhe, γ-bis(homo-phenylalanine); HOBt, 1-hydroxybenzotriazole
hydrate; Ile, isoleucine; Leu, leucine; LPS, lipopolysaccharide; MTT, 93-[4,5-
dimethylthiazol-2-yl]-2,5-diphenyltetrazolium
bromide;
PBMC,
peripheral
blood
mononucleated cell; PHA, phytohemagglutinin A; Phe, phenylalanine; Pro, proline; TBTU,
O-(Benzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium tetrafluoroborate; TFA, trifluoroacetic
acid; TFMSA, trifluoromethanesulfonic acid; TNF-α, tumor necrosis factor alpha; Val, valine.
1. Introduction
Cyclolinopeptide A (CLA), bearing the sequence c(Pro-Pro-Phe-Phe-Leu-Ile-Ile-Leu-
Val), a naturally occurring peptide from linseed oil [1] was shown to display an
immunosuppressive activity comparable to that of cyclosporine A [2, 3]. As proposed, its
immunosuppressive activity was related to inhibition of cyclophilin A (peptidyl-prolyl cis -
trans isomerase). [4]. Nevertheless, the affinity of CLA to cyclophilin was lower in
comparison to CsA [5].
Studies on structure-activity relationship revealed that -Pro-Pro-Phe-Phe- fragment was
responsible for the immunosuppressive activity of CLA [6]. The modifications of this peptide
fragment had an impact on immunosuppressive activity, suggesting an association with
conformational flexibility [7]. Preliminary studies with non-physiological amino acids
showed that the modifications with β or homo -amino acids caused increases in the
immunosuppressive and anti-inflammatory activities [8, 9, 10]. Of interest, the analogues of
CLA exhibited also synergic immunosuppressive action with methotrexate [11].
3
Our recent report, describing CLA analogues modified with γ -bis(homo-phenylalanine)
residues, indicated that the peptide backbone’s elongation by an ethylene fragment may
2

increase the immunosuppressive activity. However, such a modification was also associated
with a decreased viability of peripheral blood mononuclear cells through the new analogues.
1
2
3
3
4
5
6
7
8
9
Only one analogue c(Pro -Pro -Phe -S-γ -hhPhe -Leu -Ile -Ile -Leu -Val ) did not impact on
cell viability. It appeared that the lack of cell death was correlated with its capability to form
cis, trans isomers within the -Pro-Pro- fragment [12].
1
4
A H NMR examination of Boc-γ -bis(homo-phenylalanine) showed that it was possible
to stop the rotation in the NH-CO region of the amino acid, which caused two possible
1
syn/anti rotamers of the amide bond, visible in the NMR spectrum [13]. H NMR analysis of
that molecule shed light on the conformation in solution. However, to fully examine three-
dimensional features we performed an X-ray analysis and compared it with all available X-ray
data of peptides containing γ-amino acids to find some common conformational patterns.
In order to confirm the influence of ethylene bridges incorporated into the peptide
backbone by γ-amino acid and to evaluate the biological activity, 8 new linear and 8 new
cyclic analogues of CLA have been synthesized. In these new peptides one or both
4
phenylalanines were replaced by one or two S or R-γ -bis(homo-phenylalanine) residues
(
Fig. 1). In addition, we intended to determine the crystal structures of N-protected γ-amino
acids, applied in the study, for evaluation of conformational preferences of γ-amino acids in
comparison to respective α-amino acids, based on crystallographic data.
6
7
8
9-
1
2
4
3
4
4
5
1
2
3
4
5
6
7
8
H-Ile -Ile -Leu -Val Pro -Pro -S-γ -hhPhe -S-γ -hhPhe -Leu -OH
6
7
8
9-
1
2
4
3
4
5
H-Ile -Ile -Leu -Val Pro -Pro -S-γ -hhPhe -Phe -Leu -OH
6
7
8
9-
1
2
3
4
4
5
H-Ile -Ile -Leu -Val Pro -Pro -Phe -S-γ -hhPhe -Leu -OH
6
7
8
9
1
2
4
3
4
4
5
H-Ile -Ile -Leu -Val -Pro -Pro -R-γ -hhPhe -S-γ -hhPhe -Leu -OH
6
7
8
9
1
2
4
3
4
4
5
H-Ile -Ile -Leu -Val -Pro -Pro -S-γ -hhPhe -R-γ -hhPhe -Leu -OH
6
7
8
9
1
2
4
3
4
4
5
H-Ile -Ile -Leu -Val -Pro -Pro -R-γ -hhPhe -R-γ -hhPhe -Leu -OH
6
7
8
9
1
2
4
3
4
5
H-Ile -Ile -Leu -Val -Pro -Pro -R-γ -hhPhe -Phe -Leu -OH
6
7
8
9
1
2
3
4
4
5
H-Ile -Ile -Leu -Val -Pro -Pro -Phe -R-γ -hhPhe -Leu -OH
1
2
4
3
4
4
5
6
7
8
9
9
1
1
1
1
1
c(Pro -Pro -S-γ -hhPhe -S-γ -hhPhe -Leu - Ile -Ile -Leu -Val )
1
2
4
3
4
5
6
7
8
9
0 c(Pro -Pro -S-γ -hhPhe -Phe -Leu - Ile -Ile -Leu -Val )
1
2
3
4
4
5
6
7
8
9
1 c(Pro -Pro -Phe -S-γ -hhPhe -Leu - Ile -Ile -Leu -Val )
1
2
4
3
4
4
5
6
7
8
9
2 c(Pro -Pro -R-γ -hhPhe -S-γ -hhPhe -Leu - Ile -Ile -Leu -Val )
1
2
4
3
4
4
5
6
7
8
9
3 c(Pro -Pro -S-γ -hhPhe -R-γ -hhPhe -Leu - Ile -Ile -Leu -Val )
1
2
4
3
4
4
5
6
7
8
9
4 c(Pro -Pro -R-γ -hhPhe -R-γ -hhPhe -Leu - Ile -Ile -Leu -Val )
3

1
2
4
3
4
5
6
7
8
9
1
1
5 c(Pro -Pro -R-γ -hhPhe -Phe -Leu - Ile -Ile -Leu -Val )
1
2
3
4
4
5
6
7
8
9
6 c(Pro -Pro -Phe -R-γ -hhPhe -Leu - Ile -Ile -Leu -Val )
Fig. 1. The sequences of linear (1-8) and cyclic (9–16) analogues of CLA modified with the
4
use of (4S) and/or (4R)-Boc-γ -hhPhe 17.
2
. Materials and methods
2
.1. General remarks
2
.1.1. Chemistry
All solvents were purified by conventional methods. Evaporations were carried out under
reduced pressure. Melting points were determined in a capillary melting point apparatus
Büchi SMP-20. The optical rotation was measured in a 1 dm cell (1 mL) on a PolAAr 3001
automatic polarimeter at 589 nm. For thin layer chromatography 254 nm silica gel on TLC
plates (Fluka) was used with a following solvent system: ethyl acetate:hexane (4:1). The
chromatograms were visualized by KMnO or ninhydrin. HPLC was performed on a Thermo
4
Spectra System, a Vydac C8 column (0.46 × 25 cm): flow 1.5 mL/min or 1mL/min, detection
at 220 nm and eluents (A) 0.05% trifluoroacetic acid in water and (B) 0.038 % trifluoroacetic
acid in acetonitrile/water 90:10 with a gradient application. The purification of peptides was
performed by the preparative reversed-phase HPLC on Gilson 322 pump and UV/Vis-152
detector on a Vydac C8 column (2.21×25 cm): flow rate 20 mL/min, UV detection at 220 nm.
In the peptides cyclisation process Ascor AP22 pump was used. The identities of the pure
peptides were confirmed by Maldi-TOF mass spectrometry on Voyager Elite, Perseptive
Biosystems using α-cyano-4-hydroxy-cinnamic acid as a matrix. The one- and two-
1
dimensional H NMR spectra were recorded on a Bruker Avance II Plus spectrometer at
7
2
00.4 MHz in DMSO-d
6
, CDCl
3
, MeOD or CD
3
CN using as an internal solvent signals of
.50 ppm, 7.26 ppm, 4.87 ppm, CD
3
CN 1.94 ppm, respectively. Spin-lock time in TOCSY
experiments was 80 ms and ROESY experiments were recorded with 300 ms mixing time. 2D
NMR spectra (COSY, ROESY, TOCSY, HSQC) were processed with TopSpin 2.1 (Bruker)
4

and analyzed by Sparky software [14]. All amino acid derivatives and peptide bond forming
reagents were purchased from IRIS Biotech (Germany). All reagents were purchased from
Sigma-Aldrich and solvents were purchased from POCh.
2
.1.2. Biology
RPMI 1640 medium, Hanks’ medium and Lymphocyte Separation Medium (LSM) were
from Cytogen Sinn, Germany. MTT (93-[4,5-dimethylthiazol-2-yl]-2,5diphenyltetrazolium
bromide), phytohemagglutinin A (PHA) and other reagents were from Sigma, USA. Fetal calf
serum was from Gibco, USA. TNF-α was measured using ELISA kits manufactured by RD
System, Minneapolis, USA & Canada. In separations of PBMC cells Heraeus Sepatech
Varifuge 3.0R centrifuge was used. The MTT assays were performed with the use of Elisa
reader, Opsys MR Dynex Technologies. Cells were incubated in New Brunswick™ Galaxy®
4
8 R incubator. Density of cells was determined with the use of Bürker hemocytometer cell
counter.
2
.2. Peptide synthesis and purification
The linear peptides 1-8 were synthesized by the manual solid-phase method using
chloromethylated Merrifield resin as a solid support. Attachment of the first Boc-AA-OH
Boc-Leu) to the resin was performed according to the cesium salt procedure [15] and the
(
substitution level was determined by weight gain measurements (Boc-Leu- , 0.62 mmol for
peptides 1-8). Synthesis of the peptides was achieved by stepwise coupling of Boc-amino
acids to the growing peptide chain on the resin. Starting with 0.2 mmol of Boc protected
amino acid attached to the resin, standard single TBTU/HOBt/DIPEA coupling protocol (with
deprotonation steps omitted) was used for all amino acids and was repeated if Kaiser test [16]
or Chloranil test (proline residue) [17] was found positive. In all cases, where the second
coupling test was slightly positive, remaining free amino groups were acetylated with the aid
of acetic anhydride (190 μL) in the presence of diisopropylethylamine (350 μL). After
coupling step the Boc protecting group was removed with 50% TFA in methylene chloride.
The peptide resin was treated with TFMSA (1 mL), TFA (10 mL) and anisole (0.5 mL) at
0
°C and stirred for 60 min at room temperature. The resin had been filtered off and washed
with TFA, and the crude peptide was precipitated upon the concentration of solvents and
addition of diethyl ether as a TFA*peptide salt.
5

Cyclisations of the peptides were achieved with the use of medical syringe pump
equipped with two syringes A and B according to the idea presented in the reference [18] with
some modifications. Syringe A contains solution of 60 mg of the linear peptide dissolved in
1
0ml of DMF, and syringe B contains solution with an equivalent amount of the
HATU/HOAT in 10ml of DMF. Solutions from syringe A and B were injected with a speed
.42 ml/h into a flask containing approximately 15 ml of DMF and 3 eq of DIPEA. Solution
0
in the flask was stirred with a magnetic stirrer. After cyclisation the mixtures were analyzed
by HPLC. DMF was evaporated under reduced pressure and crude peptides were purified by
preparative HPLC.
All the peptides sent to biological trials were purified by the preparative HPLC and once
again were examined by HPLC and additionally by MS techniques. The physicochemical
properties of the synthesized linear and cyclic peptide are summarized in Table 1.
Table 1. Physicochemical properties of peptides 1-16.
Peptide Yield
%]
HPLC
Molecular
formula
Formula
mass
MALDI MS
[
m/z
+
+
t
R
Purity [%]
[M + H]
[M + Na]
(
min)
Crude Purified
a
1
2
3
4
5
6
7
8
9
96
88
12.98 97.20 99.64
C
C
C
C
C
C
C
C
61
59
59
61
61
61
59
59
H
H
H
H
H
H
H
H
95
91
91
95
95
95
91
91
N
N
N
N
N
N
N
N
9
9
9
9
9
9
9
9
O
O
O
O
O
O
O
O
10 1113.72
10 1085.69
10 1085.69
10 1113.72
10 1113.72
10 1113.72
10 1085.69
10 1085.69
1114.18
1086.41
1086.49
1114.54
1114.50
1114.76
-
1136.14
1108.33
1108.42
1136.53
1136.54
1136.76
1109.03
1108.81
1118.40
1090.54
1090.51
1118.58
a
12.76 87.47 99.79
a
100
94
13.38 95.78 97.70
a
13.54 99.79 96.26
a
81
12.76 94.56 99.11
a
70
12.96 82.89 99.00
a
84
13.55 88.37 100.00
b
a
82
12.94 94.59 97.35
1086.88
1096.44
1068.58
1068.52
1096.64
b
45
8.23
9.90
100.00
99.20
C
C
C
C
61
59
59
61
H
H
H
H
93
89
89
93
N
N
N
N
9
9
9
9
O
O
O
O
9
9
9
9
1095.71
1067.68
1067.68
1095.71
b
1
0
1
2
40
b
1
1
22
10.38
10.46
100.00
100.00
b
19
6

b
b
b
b
1
1
1
1
3
4
5
6
61
35
24
23
9.12
9.61
96.20
95.20
96.47
100.00
C
61
C
61
C
59
C
59
H
93
H
93
H
89
H
89
N
9
N
9
N
9
N
9
O
9
O
9
O
9
O
9
1095.71
1095.71
1067.68
1067.68
1096.68
1118.59
1118.98
1090.60
-
*
1097.02
1068.61
1068.98
10.71
10.61
a) before HPLC purification; b) yield of cyclization
1
2
.3. H NMR analysis of cyclolinopeptides 9-16
All NMR spectra (COSY, TOCSY, HSQC and ROESY) of synthesized peptides can be
1
found in supplementary material file as well as H NMR description assigned based on these
spectra.
2
.4. Synthesis procedures
4S)- and (4R)-4-((tert-Butoxycarbonyl)amino-)-4-benzyl-butanoic acid 17 was obtained
(
from L or D-Boc-phenylalanine in parallel synthesis according to known procedure [13] with
some modifications. Boc-phenylalanine was transformed into a mixed anhydride by using
ClCOOEt/TEA and reduced to Boc-phenylalaninol using NaBH in THF. Boc-phenylalaninol
4
was oxidized to Boc-phenylalaninal by using PCC in DCM in a presence of Å4 molecular
sieves. Side products of oxidation were removed by crystallization with EtOAc/hexane
mixture. The solution containing Boc-phenylalaninal after evaporation of the solvents was
used in the next step reaction without additional purification. The product of oxidation was
used in Horner-Wadsworth-Emmons reaction with triethyl phosphonoacetate and KOH as a
base in a mixture of H
2
O/1,4-dioxane (10/90%). Product (E)-4-[(tert-butoxycarbonyl)amine]-
5
-phenyl-2-pentenoic ethyl ester was then hydrogenated by using 10% Pd/C and hydrolyzed
with 2N KOH in methanol giving derivative 17 with S or R configuration. The products were
purified by crystallization from ethyl acetate/hexane to yield crystalline solids. The
enantiomeric purity was determined according to the procedure using N -(2,4-dinitro-5-
α
fluorophenyl)-L-valinamide as a derivatizing reagent [19]. Diastereomeric derivatives were
detected at different retention times (min): 8.84 and 7.45 for (4R)-17 and (4S)-17,
respectively. Samples intended for crystallographic studies were recrystallized from pure
1
13
acetonitrile. All spectra ( H, C, IR), as well as specific rotation and melting points of
synthetized intermidiates and (4R)-17 and (4S)-17, can be found in supplementary material
file.
7

2
.5. Crystallographic studies
Single crystals of the two enantiomers (4S)-17 and (4R)-17 were measured with Bruker
SMART APEX II CCD diffractometer [20] at 100 K with Cu K radiation. Other details
concerning the crystal data and refinement are available in supplementary material section
(
Table 14s,15s). The applied crystallographic programs were as follows: SAINT-Plus for cell
refinement and data reduction [21], SADABS for multi-scan absorption correction [22],
SHELXS97 for structure solution (direct methods) and refinement (full-matrix least-squares
on F2) [23]. The absolute structures were verified with Flack parameter [24] and the material
for publication was made with PLATON [25]. The statistical X-ray analysis was performed
with the use of CCDS conquest software, version 1.18 using Cambridge Structural Database
[
39].
In order to confirm representativeness of the studied single crystals, the powder X-ray
diffraction of (4S)-17 and (4R)-17 was performed using a PANalytical X’Pert Pro MPD
diffractometer in Bragg-Brentano reflecting geometry and CuK radiation at room

temperature (Fig. 115s). Due to the very small amount of enantiomeric samples available, the
examination was done with a Silicon single crystal substrate. The data were collected in the
range 2-50° 2Ɵ with step of 0.0167° and exposition per one step of 50 s.
2
2
.6. Biological assays
.6.1. Preparation of the compounds for biological assays
The compounds were dissolved in DMSO (5 mg/200 μL) and subsequently diluted to 5
mL of RPMI-1640 medium. As a control, appropriate dilutions of DMSO in RPMI-1640
medium were used.
2
.6.2. Isolation of the peripheral blood mononuclear cells (PBMC)
Venous blood from a single donor was withdrawn into heparinized syringes and diluted
twice with PBS. PBMC were isolated by centrifugation on LSM and centrifuged at 800 × g
for 20 min at 4 C. The interphase cells were then washed three times with Hanks’ medium
and re-suspended in a culture medium, consisting of RPMI-1640, supplemented with 10%
8

fetal calf serum, L-glutamine, sodium pyruvate, 2-mercaptoethanol and antibiotics (referred to
6
below as the culture medium), at density of 2×10 cells/mL.
2
.6.3. Phytohemagglutinin A (PHA)-induced proliferation of human peripheral blood
mononuclear cells
PBMC were distributed into 96-well flat-bottom plates in 100 µL aliquots (2×10
5
cells/well). PHA was added at a concentration of 5 µg/mL. The compounds were tested at
concentrations of 1, 10 and 100 µg/mL. DMSO at appropriate dilutions served as control.
After fourth-day incubation in a cell culture incubator, the proliferative response of the cells
was determined by the colorimetric MTT method [26]. The data are presented as a mean
optical density (OD) value from quadruplicate wells ± standard error (SE).
2
.6.4. Effects of the compounds on viability of human blood mononuclear cells
5
PBMC at density of 3×10 /100 µL/well, re-suspended in the culture medium, were
cultured for 24h in a cell culture incubator with the preparations at 1, 10 and 100 μg/mL
concentrations. Cell survival was determined by MTT colorimetric method. The results are
given in percentage of viable cells as compared with appropriate DMSO controls.
2
.6.5. Lipopolysaccharide (LPS)-induced tumor necrosis factor alpha (TNF-α) production in
whole blood cell culture
The test was performed as described elsewhere [27]. In brief, venous blood from a
single donor was diluted 10× with RPMI-1640 medium and distributed in 1 mL aliquots in
2
4-well culture plates. The cultures were stimulated by addition of 1 µg/mL of LPS from E.
coli, serotype O:111:B4. The compounds were added to the cultures at concentrations of 1
and 10 µg/mL. Higher concentrations of the compounds could not be used because of
inhibitory effects on TNF-α production by corresponding DMSO (the solvent) dilutions.
Appropriate dilutions of DMSO served as controls (representing 100% cytokine production).
After overnight incubation in a cell culture incubator, the supernatants were harvested and
frozen at -20 C until cytokine determination by immunoassay.
2
.6.6. Cultures of Jurkat cells and total RNA isolation
6
Jurkat cells (10 /mL) in the culture medium were cultured overnight with the compound
1
5 (10 μg/mL). Total RNA isolation was carried with TRIzol Reagent (Ambion) accordingly
6
to manufacturer's recommendations. The cell pellet (2×10 cells) was suspended in 1 mL of
9

TRIzol Reagent, shaken, incubated for 10 min at room temperature (RT), supplemented with
0
.2 mL of chloroform, shaken vigorously for 15 sec, incubated for 3 min at RT and
o
centrifuged at 12000 × g for 15 min at 4 C. The water phase was collected, transferred to a
new tube, supplemented with 0.5 mL of isopropanol, incubated at RT for 10 min and
o
centrifuged at 12000 × g for 10 min at 4 C. The RNA pellet was washed with 1 mL of 75%
ethanol, dried in air and dissolved in 20-30 μL of sterile diethyl pyrocarbonate-treated Mili-Q
o
water. RNA samples were stored at -20 C.
2
.6.7. Reverse transcription
Single stranded complementary DNA (cDNA) was synthesized with oligo (dT)12-18
primers from 5 μg of total RNA using Novazym VerteKit, accordingly to the manufacturer's
instruction.
2
.6.8. Quantitation of gene expression by Real Time PCR
Expression of the genes, i.e. β-actin, caspase-3, caspase-8 and caspase-9, was measured
using APA SYBR FAST qPCR Kit. The sequences of primers are enclosed in the
supplementary data. The reaction was performed in Applied Biosystems ViiA7 thermocycler
o
starting with 5 min of preincubation at 95 C followed by 35 amplification cycles as follows:
o
o
9
5 C for 30 sec and simultaneous annealing-extension-data acquisition for 45 sec and 60 C.
Beta-actin was used as a housekeeping gene for arbitrary unit calculation for every tested gen.
2
.6.9. Analysis of DNA fragmentation
6
o
Approximately 2×10 cells were lysed overnight in 37 C, in 500 µL 50 mM Tris-HCl
buffer pH 8.5 containing 10 mM EDTA, 5% SDS and 500 µg/mL proteinase K with
o
occasional shaking. Proteinase K was deactivated by 30 min heating in 80 C, than samples
were cooled on ice, shortly centrifuged, supplemented with RNase (50 µg/mL) and incubated
o
for 1h at 37 C. DNA was precipitated with 750 µL of isopropanol, and after centrifugation at
o
1
2 000×g at 4 C for 15 min, washed with 75% ethanol, and dissolved in water. The cleavage
of DNA fragmentation was analyzed by electrophoresis fractionation in 2% agarose gel.
3
. Results and discussion
3
.1. Chemistry
10

4
All analogues were synthesized using Boc-γ -hhPhe-OH. The resin was exactly the same
as in the reference [12] to give the readers possibility to compare the yield of the synthesis.
This amino acid was incorporated into 3 or 4 and both 3 and 4 position of CLA. Linear
precursors of CLA 1-8 were synthesized according to SPPS procedure based on Boc amine
protecting group. Each analogue was synthesized at scale 0.2 mmol. The removal of the Boc
amino protecting group was achieved using 50% TFA in DCM through 15 min. In the
coupling reactions TBTU was used as a coupling reagent, DIPEA as an amine and HOBt as
an antiracemic additive. After each step (coupling reaction, acylation of unreacted amine
groups, deprotection of amine) a resin was washed three times with DCM, MeOH and DCM
each. In the coupling reactions standard Boc-protected amino acids were used in 3 eq, and
4
Boc-γ -hhPhe-OH was used in 1.5 eq to the resin. 1 mL of TFMSA, 10 mL of TFA and 0.5
mL of anisol were used for 30 min to cleavage the peptide from the resin. After filtration, the
obtained solution was concentrated under vacuum and peptide was precipitated by diethyl
ether addition. In the cyclization reaction, to promote high dilution of reacting peptides, a
syringe pump was used. The reaction of cyclization took place through 24h with yield from
1
9% for analogue 12 to 61% for analogue 13. Prior to biological assays all peptides were
purified by preparative HPLC.
3
.2. NMR measurements
1
H NMR spectra of new analogues 9-16 recorded in DMSO-d at 300 K had better
6
resolution in comparison with CLA, which probably correlated with its lower exchange rate
between conformers in the NMR scale [28]. However, by the subtraction of COSY spectra
from ROESY spectra we observed large number of long-range correlations in comparison to
CLA. That fact confirmed us that the structures of new analogues are more rigid in
comparison to CLA molecule (please see supporting material file).
The NH resonances were located within the narrow range of 0.5-0.8 ppm, as compared to
the range of 0.9, 1.0 and 1.4 ppm for peptides 10, 11a and 16. The resolution of most of the
NH signals enabled to determine the NH-CH coupling constants.
In order to determine the geometries of peptide bonds for CLA analogues 9-16 the
ROESY spectra were recorded. In the fragment Pro-Pro, the geometry was also confirmed by
1
3


C chemical shifts between C and C in prolines. According to the literature the differences
of about 9 ppm are characteristic for trans Xxx-Pro, whereas about 5 ppm are characteristic
for cis Xxx-Pro geometries [29]. In DMSO solution, with the use of NMR spectroscopy, it
was found that all cyclic peptides, except peptides 11, 13 and 15, existed as single isomers
11

with one cis peptide bond in the -Pro-Pro- fragment. In case of peptides 11, 13 and 15 the
content of cis/trans isomers was 50/50, 70/30 and 70/30, respectively.
2
In almost all cyclic analogues the chemical shifts of the  protons of Pro were very
4
similar. The differences were found in analogues 14, 15a and 16 modified with R-γ -hhPhe or
4
3
4
4
1
2 containing fragment of R-γ -hhPhe -S-γ -hhPhe . In case of these peptides one of 
protons exhibited a high-field shift up to 0.86 ppm (12), 0.87 (11a), 0.95 (14), 1.02 ppm (15a)
2
and 0.69 ppm (16). Notably, up-field shift was not as large as in native CLA (Pro  proton
3
2
observed at 0.33 ppm), but still indicated a proximity of the Phe aromatic rings and Pro side
chains in these peptides [30].
The signals of aromatic rings of both phenylalanine residues were located in a narrow
range for cyclopeptides 11a, 12, 14, 15a, 16, but for analogues 9, 10, 11b, 13, 15b a wide
range was observed, allowing a possibility for arrangement of the phenyl ring to the edge-to-
face geometry. The observed arrangement of phenyl rings for these analogues corresponded to
the geometry of native CLA [28, 31, 32] or analogues modified with tyrosine [33] with
characteristic pattern of the aromatic range (Fig. 52s, 53s) containing one signal shift to high-
field (lower ppm unit).
3
1
Vicinal coupling constants JNHCH were determined directly from H NMR or from
3
COSY-DQF spectra (peptide 11 and 15) and are summarized in Table 2. Almost all JNHCH
vicinal couplings constants were about 8 Hz. For two analogues 11 and 15 vicinal coupling
3
constants JNHCH were about 10~12 Hz suggesting an almost perpendicular arrangement [34].
3
Table 2. Vicinal coupling constants JNHCH [Hz] of peptides 9-16 in DMSO-d
6
at 300 K and
3
48 K.
3
4
5
6
7
8
9
Peptide
Phe or
Phe or
Leu
Ile
Ile
Leu
Val
4
3
4
4
γ -hhPhe γ -hhPhe
9
8
9
.4
.3
8.8
8.4
9.3
8.8
8.4
8.4
1
0
7.9
8.4
7.9
8.3
7.9
9.2
7.7
6.6
9.3
1
1a
1b
8.9
14.2
10.7
12.2
1
11.6*
8.8*
8.8
10.6*
8.7
9.6*
9.0
8.2*
8.2
7.7*
8.7
13.7*
8.4
1
2
8
.2
12

1
3a
8
8
.8
.6
8.1
7.8
8.0
8.1
7.1
8.7
1
4
8.2
8.2
8.4
7.8
8.2
8.0
1
5a
5b
8.3*
7.3
10.0*
10.0*
9.7*
10.3*
9.1*
10.8*
1
8.6
8.0
-
8.8
8.4
7.7
8.6
6.5
7.6
8.0
-
1
6
8
.4
7.8
*
vicinal coupling constants 3JNHCH [Hz] of peptides determined from COSY-DQF spectra.
The presence of intramolecular hydrogen bonds was determined by measuring the NH’s
temperature coefficients (/T) (Table 3).
Table 3. Temperature dependence of the NH chemical shifts (-/T, ppb/K) of peptides
9
-16 in DMSO-d , in the range of 300-340 K.
6
3
4
4
4
Phe or γ - Phe or γ -
Solvent
Ref.
5
6
7
8
9
Peptide
Leu
Ile
Ile
Leu Val
3
4
hhPhe
hhPhe
CLA
CLA
CLA
9
-2.8
-6.6
a
-6.2
-5.4
0.8
0.9
1.2
b
-4.1 -4.7 -4.7 -2.8
DMSO
[37]
[35]
[28]
-6.8 -2.5 -4.2 -1.8 DMSO/CDCl
3
2.8
1.6
2.0
0.9
CDCl
3
1.6
2.1
3.4
5.1
3.1
4.8
4.8
6.0
6.7
5.5
2.4
4.9
5.7
4.8
5.4
0.1
3.0
4.3
5.7
0.3
4.7
1.8
5.3
3.5
2.8
2.7
3.1
3.1
4.9
0.4
1.9
3.7
2.0
5.0
3.9
3.8
5.0
2.3
3.7
2.5
3.1
3.2
3.9
3.0
4.4
4.0
4.7
0.2
-
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
1
1
0
2
1
3a
1
4
1
5a
1
6
The temperature dependence of the NH chemical shifts for cyclopeptides 11, 13b and 15b were unable to determine and is omitted.
a - non-linear, b - absent
13

According to literature, the coefficient (/T < 3 ppb/K) in DMSO is characteristic for
the presence of intramolecular hydrogen bonds [36]. For the natural CLA molecule, the
4
7
8
9
intramolecular hydrogen bonds involve NH’s of Phe , Ile , Leu , and Val [28], whereas when
3
5
9
dissolved in DMSO, the NH’s of Phe , Leu and Val were engaged in intramolecular
4
hydrogen bonds [37]. In the case of the CLA analogues modified with  -hhPhe (analyses in
DMSO) different types of controllers determining their stability were found. Analogue 10
4
3
5
7
9
contained four hydrogen bonds covering S- -hhPhe , Leu , Ile , Val residues, for peptide 14
6
7
8
three hydrogen bonds were observed (Ile , Ile , Leu ), for peptides 9, 12, 13, 15 in
4
3
stabilization two hydrogen bonds were involved, including NH protons from (S- -hhPhe ,
6
7
8
4
6
5
9
Ile ), (Ile , Leu ), (Phe , Ile ) and (Leu , Val ) respectively. For peptide 16, it was not possible
9
to determine temperature dependence of the NH of Val residue, what suggests that it is the
least stabilized analogue.
4
The replacement of one or both Phe residues by incorporation of  -hhPhe residue(s)
changed the network of stabilizing hydrogen bonds relative to native CLA. This relation was
observed for all analyzed analogues. Only peptide 14 contained the same number of
stabilizing hydrogen bonds as native CLA in DMSO, but they interacted with different NH’s
6
7
8
residues, i.e. Ile , Ile , Leu . Peptide 10 had the same pattern of hydrogen bond as native CLA,
but additional, Ile7 residue was involved in the structure’s stabilization.
All cyclic peptides, except 11, 13 and 15, existed as isomers with the cis configuration
4
3
peptide bond between the Pro-Pro motif. Peptide 11, containing only one residue S-γ -hhPhe ,
existed as a mixture of equal amount of cis and trans geometry isomers. In case of two
4
3
4
4
4
3
peptides, 13 modified with S-γ -hhPhe -R-γ -hhPhe and 15 contained single R-γ -hhPhe
residue, two conformers with major cis isomer in 70% of population were observed.
3
.3. Crystallography
4
Resolution of the crystal structures of the Boc-γ -hhPhe-OH derivatives (4R)-17 and (4S)-
7 allowed to determine their absolute configurations and their conformations (Fig. 2).
1
Torsion angles listed in Table 4 showed the presence of trans geometry of the amide bond for
both enantiomers and very similar conformations of their main chains. The orientation of the
carboxyl group was defined by respective torsion angle t1. The differences were related to
different hydrogen bond networks having an impact on different packing of (4R)-17 and (4S)-
1
7 in crystal state (Fig. 3). In Table 4 we compared torsion angles of the known γ-amino acids
obtained from X-ray analysis. The restricted number of observed values of torsion angles, also
14

in other structures than (4R)-17 and (4S)-17, suggested a rather rigid conformation of the
fragment, despite the cis/trans isomerism. In the known crystals of γ-amino acids presented in
Table 4 only intermolecular hydrogen bonds were observed as well as in (4R)-17 and (4S)-17.
4
The enantiomeric Boc-γ -hhPhe-OH derivatives 17 were characterized by inversed
polymorphs. This observation was confirmed by different melting points of derivatives, 108-
1
10 °C for (4S)-17 enantiomer and 122-124 °C for the (4R)-17 enantiomer. Literature values
were reported for (4R)-17 crystals, 132-133 °C by Loukas et al. [13] and 114 °C by Hanessian
Yang [38]. They all might be still other polymorphs of 17, because the crystallization
&
conditions in these three cases were different. However, the differences in crystal state, to
some degree, could result in the presence of other factors, such as small amounts of impurities
indicated by powder diffraction profile of (4R)-17 (Fig. 115s).
4
Fig. 2. The molecular structures of Boc-γ -hhPhe-OH 17 in their enantiomeric crystals.
15

Fig. 3. Packing of the molecules in the enantiomeric crystals of the 17 amino acid.
Table 4. Conformation of the Boc-N-protected γ-amino acids in the crystal state (this study
and CSD [39]). The definitions of the torsion angles are as follows: HO-C1-C2-C3 (t1), C1-
C2-C3-C4 (t2), C2-C3-C4-N5 (t3), C3-C4-N5-C6 (t4), C4-N5-C6-O7 (t5) and N5-C6-O7-C8
(
t6). The values of the torsion angles are in degrees and rounded. The compounds comprising
substituents participating in strong hydrogen bonds or rigid systems incorporated into their
main chain were omitted in the analysis. All structures were enantiomeric.
Structure/
Substituents t1
t2 t3 t4
t5
t6
HB type
Literature
REFCODE
(a) cis/
(b)
trans
(
(
4S)-17, Z’=2 4-Bz
94 -176 -59 127 176 -179 aa(A)+aa(B)
cc(AB)
-
179 173 -60 131 170 -177
4R)-17
4-Bz
-86 -176 -62 111 177 177
ac(en)
1
1
8 HOMYOZ 3-Bz
-174 -168 71 90
-14 177
-169
ac(en)
ac(en)
[40]
[41]
9 NIFPOH 2-Me; 4-Me 119 -71 -69 122
3
Notations: (a) The torsion angles were unified by making t4 positive; (b) Hydrogen bond network notation (aa, cc and ac) refers respectively
to amide-amide (-N-H···O=C-NH···), carboxyl-carboxyl (-C-OH···O=C-OH···), amide-carboxyl (-N-H···O=C-OH···) hydrogen bonds. In
the case of enantiomeric structure 1 (Z’=2), A and B are used to distinguish the symmetrically independent molecules.
The observed torsion angles (Table 4) indicated that the conformation of N-protected γ-
amino acids is rather conservative, despite the presence of a very flexible fragment
comprising three single bonds which separate amine and carboxylic groups. Thus, it seemed
interesting to compare the conformational predisposition of γ-amino acids with α-amino acids
in the peptides according to available X-ray structures. Some light on the subject gives a
comparison of preferred orientations of the carbonyl groups participating in the hydrogen
bonds in the α-peptides with peptides containing γ-amino acids [39] (Fig. 4).
The distribution of  angles (Fig. 116s) between C-O vectors representing two
neighbouring carbonyl groups (being H acceptors) in α-peptides was significantly different
from peptides comprising γ-amino acids (Fig. 4a,b). The range of preferable values for
peptides derivatives of γ-amino acids was extremely narrow and the average value is much
higher (Fig. 4b). For 2230 appropriate fragments found in Crystallographic Structural Data
[
39] for peptides built from α-amino acids (excluding glycine), the 56% of the  angles were
16

in the range of 20-90° (Fig. 4a). For this range the average angle was equal to 39°. In the case
of peptides built from γ-residue (Fig. 4b) the 88% (136 examples) of the  angles were in the
narrow range of 40-60°, and the average value was equal to 49° [39]. Additionally, much
smaller maximum of  values of γ-peptides was observed for antiparallel orientation of
carbonyl groups (Fig. 4b).
Fig. 4. Distribution of  angles for structures deposited in the CSD [-peptides – a (linear and
cyclic peptides), c (linear peptides), e (cyclic peptides); γ-peptides – b (linear and cyclic
peptides), d (linear peptides), f (cyclic peptides)]. In each graph X axis represents  angle
o
o
(
range 0 – left, 180 –right) and Y axis represents a number of the analysed X-ray structures.
However, comparison of linear and cyclic peptides showed more clearly the differences
in distribution of  angles (Fig. 4c-f). In the case of -amino acids, creating linear peptides,
the  angles were predominantly in the range of 20-80° (Fig. 4c), whereas in cyclic -peptides
the orientation was more balanced between antiparallel and parallel orientation (Fig. 4e).
Even more dramatic change in distribution of  angles was visible for the γ-peptides (Fig.
4
d,f). In rare cyclic peptide examples all the  angles were over 70°, most of them over 110°
(
Fig. 4f). The difference in  values between peptides containing γ- and α-amino acids could
be one of the factors causing different PHA-induced PBMC proliferation between new linear
and cyclic analogues of CLA reported in this work.
3
.4. Biology
17

3
5
.4.1. Effects of the compounds on viability of PBMC.
The effects of the linear and cyclic compounds on the cell viability are presented in (Fig.
a and 5b). The compounds were tested at concentration of 1, 10 and 100 μg/mL in a 24h
culture of PBMC and the cell viability was determined by the MTT colorimetric method.
Fig. 5A.
Fig. 5B.
18

Fig. 5. The effects of linear peptides 1 – 8 (A) and cyclic peptides 9 – 16 (B) on the viability
of human peripheral blood mononuclear cells. The peptides were used at concentrations of 1,
1
0 and 100 μg/mL. The viability of cells upon culture with the compounds was compared to
the effects of DMSO used as solvent (filled bars). The open bar represents PHA-stimulated
cultures without DMSO addition.
The linear peptides (Fig. 5a) did not lower cell viability at the studied concentrations but
the cyclic ones showed some inhibitory effects on the cell survival at 100 μg/mL, in particular
compounds 9, 10 and 11 (Fig. 5b).
3
.4.2. Effects of the compounds on PHA-induced PBMC proliferation
Fig. 6A.
19

Fig. 6B.
Fig. 6. Effects of linear peptides 1 – 8 (A) and cyclic peptides 9 – 16 (B) on PHA-induced
PBMC proliferation. The peptides were used at concentrations of 1, 10 and 100 μg/mL. The
viability of cells upon culture with the compounds was compared to the effects of DMSO
20

used as solvent (filled bars). The open bar represents PHA-stimulated cultures without DMSO
addition. “Ctrl” – no PHA added.
The effects of the compounds on PHA-induced proliferation of human PBMC in the
concentration range 1-100 μg/mL are presented in Fig 6ab. The linear peptides (Fig. 6a) were
not suppressive, whereas the cyclic peptides (Fig. 6b) showed differential, dose–dependent
ability to inhibit the mitogen-induced cell proliferation. Considerable inhibitory effects were
already observed at concentration of 10 μg/mL for almost all cyclic peptides. Compound 15
was the most suppressive taking into account its best effects at all tested concentrations in
comparison to other compounds.
3
.4.3. Effects of the compounds on LPS-induced TNF-α production in whole blood cultures
The effects of the peptides on LPS-induced TNF-α production in whole blood cell cultures are
shown in (Fig. 7a and 7b). The linear peptides exhibited about 45% inhibitory effects on the
cytokine production at 10 μg/mL concentration with exception of peptide 2. The cyclic
compounds also inhibited TNF-α production at 10 μg/mL. In addition, peptide 14 was
strongly inhibitory at both concentrations and peptide 13 also at 10 μg/mL. The inhibitory
effects of the peptides on TNF-α production correlate well with their antiproliferative actions
since TNF-α augments mitogen-induced PBMC proliferation [42].
Fig. 7A.
21

Fig. 7B.
Fig. 7. Effects of linear peptides 1 – 8 (A) and cyclic peptides 9 – 16 (B) on LPS-induced
TNF-α production in human whole blood cultures. The peptides were used at concentrations
22

of 1 and 10 μg/mL. The filled bars represent solvent (DMSO) controls. The oper bar (“LPS”)
–
the cultures with addition of LPS only. “Ctrl” – no LPS added.
3
.4.4. Description of structure-activity relationships in the previously investigated CLA
analogues
In an effort to increase the suppressive potency of CLA, natural, modified or not physiologic
amino acids were substituted outside or within the “active” Pro-Pro-Phe-Phe sequence of the
cyclolinopeptide. The actions of such peptides were compared with native CLA and/or CsA.
Three tyrosine analogues were synthesized by replacement of phenylalanine in the -Pro-Pro-
Phe-Phe- fragment of the CLA: IV -Pro-Pro-Phe-Tyr-, V -Pro-Pro-Tyr-Phe- and VI -Pro-Pro-
Tyr-Tyr- [43]. The analogues were tested for their effects on the inductive phase of the
delayed type hypersensitivity (DTH) in mice. It appeared that compounds IV and V were
weaker but peptide VI was a stronger inhibitor than the parent CLA peptide. It seems,
therefore, that the replacement of both phenylalanines with tyrosines may convert the parent
cyclopeptide into a more potent inhibitor. In this model native CLA displayed a stronger
immunosuppressive action than CsA. In the same experimental model [44] several CLA
analogues were tested with substitution of threonine in the -Leu-Ile-Ile-Leu-Val- fragment.
The analogues with the following sequences: c(Thr-Ile-Ile-Leu-Val-Pro-Pro-Phe-Phe), c(Leu-
Ile-Thr-Leu-Val-Pro-Pro-Phe-Phe) and c(Leu-Ile-Ile-Leu-Thr-Pro-Pro-Phe-Phe) displayed
stronger immunosuppressive activities than CsA in the DTH model. In another attempt CLA
2
3
was modified with (S)-β -iso-proline or with (S)-β -homoproline [10]. In the mouse models of
the proliferative response of lymphocytes to mitogens and in the humoral, secondary immune
1
3
2
response in vitro, a peptide bearing the Pro -β hPro
fragment exhibited a strongest
immunosuppressive action. In another approach, [8] the authors found that a CLA analogue
3
3
4
containg Phe -β Phe fragment appeared to be the most effective (better than CsA) in
inhibition of the inductive phase of DTH to ovalbumin. A comparison of three CLA,
3
4
3
4
3
4
analogues bearing: Hphe -Phe , Phe -Hphe and Hphe -Hphe substitutions, showed that the
first analogue exhibited the strongest inhibitory activities in the models of mitogen-induced
splenocyte proliferation and secondary humoral immune response in vitro [9]. Lastly, two
cyclic analogues of CLA were synthesized, containing conformationally restricted dipeptide
fragment Phe-Phe or d-Phe-d-Phe with ethylene brigde -CH
2
-CH - between phenylalanine
2
3
4
3
4
nitrogens [11] of the following sequences: -Pro -Pro -Phe-Phe- and -Pro -Pro -D-Phe-D-Phe-.
The peptides were tested in the model of the secondary humoral immune response suppressed
23

by methotrexate (MTX). It was found that although native CLA augmented the suppressive
action of MTX, the modified peptides counteracted the action of MTX, in particular the
second one. Taken into consideration the structure-activity relationships and suppressive
potencies of the hitherto described CLA analogues, out of the herein presented peptides, 15
appears to be most promising, even we had examined different biological assays. Compound
1
2
4
3
4
5
6
7
8
9
1
5, c(Pro -Pro -R-γ -hhPhe -Phe -Leu -Ile -Ile -Leu -Val ), showed the strongest effect in
the model of the PHA-induced PMBC proliferation, where a complete inhibition of
proliferation was achieved at 10 μg/mL and some, statistically significant inhibition at
1
μg/mL concentration. Such a conclusion may be drawn based on our thorough analysis on
the results cited above and generated exclusively in the Institute of Immunology, Wrocław, in
collaboration with biochemists from other laboratories.
All new analogues had less impact on viability of PBMC than analogues modified with
3
γ -bis(homo-phenylalanine) [12] which is associated with the relocation of the benzyl group.
4
2
4
3
In the case of γ -bis(homo-phenylalanine) analogues, the fragment -Pro -γ -hhPhe - or -Pro-
4
3
Phe-γ -hhPhe - is very similar to the native CLA. That similarity gives a possibility to the
edge-to-face arrangement. In the case of the most active peptide 15 we reported cis/trans
isomerism together with the edge-to-face arrangement of phenyl rings and additional more
rigid structure in comparison to CLA. The most important feature of the new incorporated
modifications and in the previous one [12] is that the modification is incorporated into the
peptide backbone. That induces large conformational changes in comparison to the side chain
modifications and gives new conformations by relocation of stabilizing hydrogen bonds.
In summary, the above described modifications of the native CLA sequence indicate that
some substitutions may lead to synthesis of new analogues of a better immunosuppressive
potency, and that the naturally occurring -Pro-Pro-Phe-Phe- sequence, regarded as essential
for the inhibitory property of native CLA, may be modified to obtain stronger
immunosuppressive effects. In addition, CLA analogues of unexpected immunorestoring
activities may be also obtained.
3
.4.5. Molecular studies
The hitherto conducted studies on a plausible mechanism of action of cyclic peptides have not
provided satisfactory explanation on this matter. As induction of apoptosis in activated T cells
seemed to be very likely, we exposed Jurkat cells for 24h to 15 peptide at concentration of 10
24

μg/mL (at this concentration the viability of PBMC was not affected). DMSO (the solvent)
was used in control culture. RNA and DNA were isolated from the cell pellets to determine
expression of relevant caspases by RT-PCR and DNA fragmentation in gel electrophoresis.
The results shown in Table 5 revealed a block of caspase-3 expression and diminution of
caspase-8 and -9 expressions. The demonstration of inhibition of caspase expression in Jurkat
cells by 15 allows to explain its suppressive effect in PHA-induced proliferation of PBMC,
since caspases are essential for IL-2 release upon T cell activation [45]. In addition, T-cell
receptor-induced NF-κB activation and subsequent IL-2 production are also dependent on
caspase cascade [46]. Inhibition of caspases may also result in an alternative cell death
pathway termed necroptosis [47]. Such a process is not excluded in the case of compound 15
action since DNA fragmentation was observed in Jurkat cells (Fig. 114s). The fragmentation
of DNA in Jurkat cells was described also for another bacteria-derived peptide – nicin [48].
Several compounds are known to interfere with the TNF-α production and several
mechanisms of action was have been considered [47, 49]. Thus, the inhibition of caspase
expression, possibly associated with an inhibition of NF-kappaB activity, could also explain
the inhibition of the TNF-α production by peptide 15.
Table 5. Effects of peptide 15 on expression of caspases in 24h culture of Jurkat cells.
Cell line
Protein mRNA expression
β-actin
DMSO
1
15
1
caspase-3
0.686
0.895
0.829
0
Jurkat
caspase-8
0.591
0.538
caspase-9
3
.4.6. Additional biological studies
For additional studies we chose the most potent cyclic peptide 15 and the native CLA
molecule. We revealed that the peptide did not inhibit the activity of cyclooxygenases
supplementary data, Table 12s.). On the other hand (supplementary data Fig. 112s. and 113s)
(
compound 15 did not inhibit growth of HT-29 colon cell line and A-431 epidermal tumor cell
line at a concentration of 25 μg/mL. CLA, the parent nonapeptide, was still significantly
suppressive in the models at the concentrations of 12 μg/mL and 6 μg/mL, respectively. We
25

hope that these additional biological studies would be helpful information for future applied
studies with CLA or CLA analogues.
4
. Conclusions
Incorporation of one or two γ -hhPhe residues into the peptide chains of CLA changes the
4
length of their backbone by two or four methylene group. This modification dramatically
changes the distribution of intramolecular hydrogen bonds stabilizing conformation of the
1
obtained derivatives. For three peptides, 11, 13 and 15, the inappropriate resolution in H
NMR spectra was correlated with cis/trans isomerism between the two prolines in positions 3
and 4. Peptide 11 exists as an equal mixture of cis and trans isomers, whereas peptides 13 and
1
1
5 exist as a mixture containing 70% of cis isomer. The increase in the resolution of H NMR
spectra is associated with more rigid structure of new analogues. This finding is also
consistent with the statistic examination of  angles in γ-peptides. The result obtained in
4
X-ray studies shows also that in γ-peptides modified with the use of Boc-γ -hhPhe-OH
residue the  angles should be in the narrow ranges. Thus incorporation of γ-bis(homo-
aminoacids) into the peptide sequence in our opinion leads to a more rigid structure of the
modified peptides.
Evaluation of the biological activities revealed that the cyclic peptides, but not linear
ones, were potent inhibitors of mitogen-induced proliferation of human peripheral blood
mononuclear cells. However, both types of peptides demonstrated the ability to suppress
TNF-α production in blood cell cultures. Compound 15, a representative of cyclic peptides,
was shown to inhibit caspase expression and induce DNA fragmentation in Jurkat cells, thus
revealing the alternative mechanism of action of this group of compounds. The CLA
analogues may therefore represent model compounds with a potential therapeutic value in
control of autoimmune diseases or prevention of allogeneic graft rejection and other clinical
states associated with abnormal recruitment and proliferation of T cells.
Acknowledgements
A financial support of the studies was donated by Lodz University of Technology Grant I-
1
8/501/6245/pl1 and I-18/501/6215/pl1.
Supplementary data
Supplementary data associated with this article can be found, in the online version at ;
26

References
1
2
2
. Kaufmann HP, Tobschirbel A. An oligopeptide from linseed. Chem. Ber. 1959; 92: 2805-
809.
. Siemion IZ, Cebrat M, Wieczorek Z. Cyclolinopeptides and their analogs – a new family of
peptide immunosuppressants affecting the calcineurin system. Arch. Immunol. Ther. Exp.
1
3
999; 47: 143-153.
. Wieczorek Z, Bengtsson B, Trojnar J, Siemion IZ. Immunosuppressive activity of
cyclolinopeptide A. Pept. Res. 1991; 4: 275-283.
. Gallo P, Rossi F, Saviano M, Pedone C, Colonna G, Ragone R. Specific Interaction
between Bovine Cyclophilin A and Synthetic Analogues of Cyclolinopeptide A. J. Biochem.
4
1
5
998; 124: 880-885.
. Gaymes TJ, Cebrat M, Siemion IZ, Kay JE. Cyclolinopeptide (CLA) mediates its
immunosuppressive activity through cyclophilin-dependent calcineurin inactivation. FEBS
Lett. 1997; 418: 224-227.
6
. Wieczorek Z, Zimecki M, Pędyczak A, Lisowski M, Siemion IZ. Immunosuppressive
activity of alanine analogs of cycloplinopeptide. Arch. Immunol. Ther. Exp. 1993; 41: 291-
2
7
96.
. Naider F, Benedetti E, Goodman M. Conformation of Cyclolinopeptide A Observed by
Circular Dichroism. Proc. Nat. Acad. Sci. USA 1971; 68: 1195-1198.
. Kaczmarek K, Farina KB, Zubrzak P, Jankowski S, Zimecki M, Suder P, Benedetti E,
8
Fattorruss R, Saviano M, Zabrocki J. Synthesis, conformational analysis and immunological
3
activity of  Phe-substituted cyclolinopeptide A analogues. J. Pept. Sci. 2009; 15: 166-174.
9
. Drygała P, Olejnik J, Mazur A, Kierus K, Jankowski S, Zimecki M, Zabrocki J. Synthesis
and immunosuppressive activity of cyclolinopeptide
A
analogues containing
homophenylalanine. Eur. J. Med. Chem. 2009; 44: 3731-3738.
1
0. Katarzyńska J, Mazur A, Bilska M, Adamek E, Zimecki M, Jankowski S, Zabrocki J.
Synthesis and immunosuppressive activity of new cyclolinopeptide A analogs modified with
-prolines. J. Pept. Sci. 2008; 14: 1283-1294.
1. Katarzyńska J, Mazur A, Rudzińska E, Artym J, Zimecki M, Jankowski S, Zabrocki J.

1
Cyclolinopeptide derivatives modify methotrexate-induced suppression of the humoral
response in mice. Eur. J. Med. Chem. 2011; 46: 4608-4617.
27

1
2. Jędrzejczak K, Hrynczyszyn P, Artym J, Kocięba M, Zimecki M, Zabrocki J, Jankowski
3
S. Synthesis and biological activity of cyclolinopeptide A analogues modified with γ -
bis(homophenylalanine). Eur. J. Med. Chem. 2014; 86: 515-527.
1
3. Loukas V, Noula C, Kokotus G. Efficient protocols for the synthesis of enantiopure
γ-amino with proteinogenic side chains. J. Peptide Sci. 2003; 9: 312-319.
1
1
4. Goddard TD, Kneller DG. SPARKY 3, University of California, San Francisco.
5. Wang SS, Gisin BF, Winter DP, Makofske R, Kulesha ID, Tzougraki, C, Meienhofer J.
Facile synthesis of amino acids and peptide esters under mild conditions via cesium salts. J.
Org. Chem. 1977; 42: 1286-1290.
1
6. Kaiser E, Colescott R, Bossinger CD, Cook PI. Color test for detection of free terminal
amino groups in the solid-phase synthesis of peptides. Anal. Biochem. 1970; 34: 595-598.
1
1
7. Vojkovsky T. Detection of secondary amines on solid phase. Pept. Res. 1995; 8: 236-237.
8. Malesevic M, Strijowski U, Bächle D, Sewald N. An improved method for the solution
cyclization of peptides under pseudo-high dilution conditions. J. Biotechnol. 2004; 112: 73-
7
1
2
2
2
2
2
2
2
1
2
7.
9. Marfey P. Determination of D-amino acids. II. Use of a bifunctional reagent, 1,5–difluoro-
-dinotrobenzene. Carlsberg Res. Commun. 1984; 49: 591-596.
0. Bruker APEX2, Bruker AXS Inc., Madison, Wisconsin, USA, 2005.
1. Bruker SAINT-Plus, Bruker AXS Inc., Madison, Wisconsin, USA, 2008.
2. Sheldrick GM. SADABS, University of Göttingen, Germany, 2003.
3. Sheldrick GM. A short history of SHELX Acta Cryst. A. 2008; 64: 112-122.
4. Flack HD. On enantiomorph-polarity estimation. Acta Cryst. A. 1983; 39: 876-881.
5. Spek AL. Structure validation in chemical crystallography. Acta Cryst. D. 2009; 65: 48-
55.
6. Hansen MB, Nielsen SE, Berg K. Reexamination and further development of a precise
and rapid dye method for measuring cell growth/cell kill. J. Immunol. Methods. 1989; 119:
2
2
03-210.
7. Weber-Dąbrowska B, Zimecki M, Mulczyk M. Effective phage therapy is associated with
normalization of cytokine production by blood cell cultures. Arch. Immunol. Ther. Exp. 2000;
4
2
84: 31-37.
8. Di Blasio B, Rossi F, Benedetti E, Pavone V, Pedone C, Temussi PA, Zanotti G, Tancredi
T. Bioactive peptides: solid-state and solution conformation of cyclolinopeptide A. J. Am.
Chem. Soc. 1989; 111: 9089-9098.
28

2
9. Kessler H. Conformation and biological activity of cyclic peptides. Angew. Chem. Int. Ed.
Engl. 1982; 21: 512-523.
0. Dasgupta B, Chakrabarti P, Basu G. Enhanced stability of cis Pro-Pro peptide bond in
Pro-Pro-Phe sequence motif. FEBS Lett. 2007; 581: 4529-4532.
1. Zubrzak P, Leplawy MT, Kowalski ML, Szkudlinska B, Paneth P, Silberring J, Suder P,
3
3
Zabrocki J. Correlating biological activity with calculated geometric motifs in
cyclolinopeptide A analogs J. Phys. Org. Chem. 2004; 17: 625-630.
3
2. Huben K, Jewgiński M, Pabis A, Paluch P, Luy B, Jankowski S. The structure of
cyclolinopeptide A in chloroform refined by RDC measurements. J. Pept. Sci. 2014; 20: 901-
9
3
07.
3. Siemion IZ, Cebrat M, Jankowski A, Lisowski M, Pędyczak A, Wysłouch A. Does the
edge-to-face interaction between aromatic rings occur in cyclolinopeptide A analogues. Int. J.
Pept. Proteins Res. 1994; 44: 61-69.
3
4. Bystrov VF. Spinespin coupling and the conformational states of peptide systems. Prog.
NMR Spectrosc. 1976; 10: 1-41.
5. Matsumoto T, Shishido A, Morita H, Itokawa H, Takeya K. Conformational analysis of
cyclolinopeptide A and B. Tetrahedron 2002; 58: 5135-5140.
3
3
2
3
6. Kessler H. Conformation and biological activity of cyclic peptides. Angew. Chem. 1982;
1: 512-523.
7. Raghothama S, Ramakrishnan C, Balasubramanian D. Conformational Analysis of
Cyclolinopeptide A, a Cyclic Nonapeptide: Nuclear Overhauser Effect and Energy
Minimization Studies. Biopolymers 1989; 28: 573-588.
3
8. Hanessian S, Yang H. Design of a novel secondary structure scaffolding device: Induction
of a reverse turn in tetrapeptides by incorporating a β-amino acid and stereocontrolled free
radical α-substitution reactions in peptide motifs. Tetrahedron Lett. 1997; 38: 3155-3158.
3
9. a) Allen FH. The Cambridge Structural Database: a quarter of a million crystal structures
and rising. Acta Cryst. B. 2002; 58: 380-388; b) Groom CR, Bruno IJ, Lightfoot MP, Ward
SC. The Cambridge Structural Database. Acta Cryst. B. 2016; 72: 171-179.
4
3
0. Jędrzejczak K, Szczesio M, Oracz M, Jankowski S, Glówka ML. Crystal structure of (3R)-
-benzyl-4-[(tert-butoxycarbonyl)amino]•butanoic acid. Acta Cryst. E, 2014; 70: o1081-
o1082.
1. Pihko PM, Koskinen AMP. Synthesis of the C26−C32 Oxazole Fragment of Calyculin C:ꢀ
A Test Case for Oxazole Syntheses. J. Org. Chem. 1998; 63: 92-98.
4
29