Innovative chemical synthesis and conformational hints on the lipopeptide liraglutide

Article abstract of DOI:10.1002/psc.2890

Liraglutide is a new generation lipopeptide drug used for the treatment of type II diabetes. In this work, we describe new approaches for its preparation fully by chemical methods. The key step of these strategies is the synthesis in solution of the Lys/γ-Glu building block, Fmoc-Lys-(Pal-γ-Glu-OtBu)-OH, in which Lys and Glu residues are linked through their side chains and γ-Glu is N^α-palmitoylated. This dipeptide derivative is then inserted into the peptide sequence on solid phase. As liraglutide is obtained with great purity and high yield, our approach can be particularly attractive for an industrial production. We also report here the results of a circular dichroism conformational analysis in a membrane mimetic environment that offers new insights into the mechanism of action of liraglutide. Copyright

Full text of DOI:10.1002/psc.2890

Research Article
Received: 3 March 2016
Revised: 30 March 2016
Accepted: 4 April 2016
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI 10.1002/psc.2890
Innovative chemical synthesis and
conformational hints on the
lipopeptide liraglutide
a,b
a
a
c
Ivan Guryanov, * Alex Bondesan, Dario Visentini, Andrea Orlandin,
c
c
c
a
Barbara Biondi, Claudio Toniolo, Fernando Formaggio, Antonio Ricci,
a
a
Jacopo Zanon and Walter Cabri
Liraglutide is a new generation lipopeptide drug used for the treatment of type II diabetes. In this work, we describe new
approaches for its preparation fully by chemical methods. The key step of these strategies is the synthesis in solution of the
Lys/γ-Glu building block, Fmoc-Lys-(Pal-γ-Glu-OtBu)-OH, in which Lys and Glu residues are linked through their side chains and
α
γ-Glu is N -palmitoylated. This dipeptide derivative is then inserted into the peptide sequence on solid phase. As liraglutide is
obtained with great purity and high yield, our approach can be particularly attractive for an industrial production. We also report
here the results of a circular dichroism conformational analysis in a membrane mimetic environment that offers new insights into
the mechanism of action of liraglutide. Copyright © 2016 European Peptide Society and John Wiley & Sons, Ltd.
Keywords: circular dichroism; conformational analysis; lipopeptide; liraglutide; peptide synthesis
Moreover, it induces weight loss, decreases the level of lipids, such
as triglycerides and cholesterol, and has cardioprotective effects
Introduction
A worldwide increase of diseases caused by glycemic balance
disorders, such as diabetes, makes extremely important the devel-
opment of new drugs as well as their production at an industrial
scale. Among the different types of diabetes, type II comprises the
majority of cases [1]. It is accompanied by many dysfunctions, such
as insulin resistance and disruption of renal re-adsorption, and it is
complicated by the risk of cardiovascular diseases. In recent years,
several pharmaceutical agents have been developed for the
treatment of this type of diabetes, such as metformin, α-glucosidase
inhibitors, and thiazolidinediones [2]. Recently, a new class of drugs
was introduced into the clinical practice for incretin-based therapy
[12].
Lipid moieties bound to peptides are known to protect the latter
from enzymatic degradation [13–15]. This resistance is observed in
liraglutide as well, thus making it suitable to be used as injectable
drug. Increased albumin affinity and aggregates formation,
promoted by the lipid conjugation, are likewise responsible for
the long half-life of liraglutide in blood and for its delayed absorp-
tion that allows for once daily administration [9,16–18]. In a detailed
study of a series of GLP-1 analogs, including liraglutide, it was
shown that both lipid moiety and linker structure influence albumin
binding [19]. As a consequence, the peptide is protected against
the metabolic degradation by dipeptidyl peptidase IV and neutral
endopeptidases [20]. Notably, liraglutide half-life increases to
11–15 h after subcutaneous injection, whereas GLP-1 lasts only a
few minutes under physiological conditions. Interestingly, the
recently introduced GLP-1 analog semaglutide exhibits an even
more prolonged half-life and a reduced body clearance [19]. This
[3,4]. The peptide liraglutide belongs to this class. It is a derivative of
glucagon-like peptide-1 (GLP-1), a naturally occurring peptide
hormone with one of the most potent insulinotropic activities.
GLP-1 is part of a longer peptide produced and processed to
GLP-1 in the pancreas and the intestine [5].
Liraglutide (Figure 1, panel A), the sequence of which was first
reported in patent WO 9808871 (Novo Nordisk), encompasses
segment 7-37 of GLP-1, but differs from the original peptide in
α
26
the attachment of a N -palmitoyl-γ-Glu moiety on the Lys side
34
chain and for the replacement of Lys by Arg [6–9].
*
Correspondence to: Ivan Guryanov, Fresenius Kabi Anti-Infectives Srl, 23 via San
Leonardo, Villadose (RO) 45010, Italy. E-mail: ivan.guryanov1@gmail.com
The amino acid sequence of liraglutide displays 97% identity to
that of the native hormone, although the presence of the palmitoyl
moiety would play a crucial role in its in vivo behavior. Interestingly,
liraglutide has the same physiological activity (GLP-1 receptor
agonist) as that of the parent peptide [10]. It stimulates insulin
secretion in a glucose dependent manner, inhibits glucagon activ-
ity, and reduces the level of all types of blood glucose. It improves
also the activity of β-cells and prevents their degeneration [11].
a Fresenius Kabi Anti-Infectives Srl, 45010, Villadose (RO), Italy
b Institute of Chemistry, St. Petersburg State University, 198504, Petrodvorets, St.
Petersburg, Russia
c ICB, Padova Unit, CNR, Department of Chemistry, University of Padova, 35131,
Padova, Italy
J. Pept. Sci. 2016

GURYANOV ET AL.
pH turns to higher values. Therefore, additional information is
needed to unravel the shape of the aggregates and their pH
dependence.
The excellent biological properties of liraglutide, outlined earlier,
underline the importance of this peptide as a drug. For this reason,
several methods for its synthesis at industrial scale are described in
the literature (Table 1).
The first strategy reported includes the use of a recombinant
peptide precursor, which is further chemically modified by
20
acylating the side chain of Lys with γ-(Pal-Glu-OtBu), and finally
deprotected and purified (Table 1, WO2004/029077) (Figure 1,
panel B).
It is worth mentioning also the production of a fusion peptide,
obtained by recombinant DNA, containing an affinity tag, a
cleavable tag and the desired peptide [24]. This method allows
for an easy purification of the product by means of affinity
chromatography. In the last step, chemical agents act on the
cleavable tag to liberate the target peptide.
The chemical synthesis usually foresees the preparation of the
peptide backbone by conventional solid-phase strategies using
orthogonal protecting groups on the Lys residue. Then, after
selective removal of the side-chain protecting groups, the synthesis
is continued on the solid support or in solution to achieve the target
branched structure. It is worth pointing out that the branched
structure of liraglutide and its low aqueous solubility dramatically
hamper its synthesis and purification.
To identify an easier way for the chemical synthesis of liraglutide,
amenable to industrial application, we investigated several
different approaches and introduced a novel strategy. The synthe-
ses reported here were designed in the frame of a Master Thesis
in Industrial Chemistry, University of Padova, and published in
December 2014 [25]. In December 2015, when we were conducting
Figure 1. (A) The amino acid sequence of liraglutide. The changes with
respect to glucagon-like peptide-1 (7-37) are shown in green; (B) A
scheme of preparation of liraglutide by chemical modification of a
recombinant peptide precursor (OSu, 1-oxysuccinimidoyl).
improved behavior was ascribed to a favorable albumin binding
mediated by the fatty acid chain.
Understanding the nature of the aggregates and their role in the
bioactivity mechanism appear to be an important issue as well. By
analytical ultracentrifugation techniques, it was observed that
liraglutide is aggregated even at 0.001 mM concentration [21,22].
In the range of 0.001–1.2 mM at pH 8.0 (10 mM Tris/perchlorate
buffer) and in 100 mM NaCl apparently, it forms heptameric struc-
tures. More recently, it was also found that the composition of
liraglutide oligomers is influenced by pH [23]. In this same study,
it was reported that in sodium phosphate buffer micelle-like
clusters of eight molecules initially form, but then they reorganize
into assemblies of 12 liraglutide molecules when the pH is dropped
below 6.9. This transformation has a slow kinetics as it takes several
days to the system to reach the equilibrium. Moreover, it was
shown that this rearrangement is accompanied by loss of the
α-helical structure, but the 3D-structural change is reversed when
Table 1. Liraglutide preparation patent landscape
Invention
number
Year of
publication
Company/Institution
Title of the invention
Brief description of the synthesis
WO2004/029077
2004
2013
Novo Nordisk
A/S, Denmark
Method for producing
acylated peptides
Attachment of lipophilic moiety to peptides in
aqueous solution containing less than 10%
of aprotic polar solvent
WO2013/037266
Hybio
Method for solid phase
synthesis of liraglutide
Solid-phase synthesis using the Alloc
protecting group for Lys, its removal in the
presence of Pd catalyst with the following
Pharmaceutical
Co. Ltd., China
α
coupling of N -palmitoyl glutamate
CN103864918
2014
2014
2015
2015
Harbin
Solid-phase synthesis
method for liraglutide
Solid-phase synthesis using the Mtt
protecting group for Lys, its removal with
TFA with the following coupling of
Jixianglong
Biotech Co. Ltd.,
China
α
N -palmitoyl glutamate
WO2014/199397
WO2015/100876
Mylan
Process for the
Solid-phase synthesis using the Dde
protecting group for Lys, its removal with
hydrazine with the following coupling of
Laboratories Ltd.,
India
preparation of liraglutide
α
N -palmitoyl glutamate
α
Adlai Nortye
Method for preparing
liraglutide
Preparation of the N -palmitoylated
Pharmaceutical
Co. Ltd., China
protected dipeptide and its incorporation into
the liraglutide sequence during solid-phase
peptide synthesis
CN104650219
University of
Method for preparing
liraglutide by
Synthesis on solid phase of four liraglutide
segments and their couplings in solution
Lanzhou, China
convergent synthesis
TFA, trifluoroacetic acid.
wileyonlinelibrary.com/journal/jpepsci
J. Pept. Sci. 2016

INNOVATIVE CHEMICAL SYNTHESIS AND CONFORMATIONAL HINTS ON LIRAGLUTIDE
the circular dichroism (CD) analysis and working on this manuscript,
a Chinese peptide group of the Adlai Nortye Pharmaceutical
Company patented a strictly related method of synthesis (Table 1,
WO2015/100876).
Agilent Eclipse XDB-C8 21.2× 250 mm column and the same
eluants.
Mass spectrometry analyses
The key step of this new strategy, which we report in this contri-
Mass spectra were acquired on a Mariner API-ToF workstation
20
26
bution, is the preparation in solution of the Lys (Lys in GLP-1)
palmitoylated dipeptide and its subsequent incorporation into the
peptide chain directly on the resin to generate the branched
structure. This method allows for a significant reduction of costs
as it limits the use of expensive Lys derivatives. In addition, incom-
plete couplings, often associated with the final steps, are avoided,
thus markedly decreasing the synthesis time and the number of
side products. We believe that for the industrial production of
liraglutide our chemical approach would be economically more
attractive, as compared with the biotechnological methods.
In addition to the new chemical synthesis, here we report the
results of a CD analysis of liraglutide in the presence of phospho-
lipid bilayers. We believe that this environment, much closer to
the physiological conditions as compared with buffered aqueous
systems, may contribute to better description of the mechanism
of action of liraglutide. Indeed, the amphiphilic nature of the pep-
tide and the presence in the biological fluids of membranes of var-
ious types, including those of blood cells and vessels, make
unavoidable the interaction of liraglutide with the phospholipids.
This event might change the peptide conformational preferences
and, therefore, its biological activity.
(PerSeptive Biosystems, Framingham, MA) and Agilent 6530 mass
accuracy Q-TOF, operating in the positive mode.
Nuclear magnetic resonance spectrometry
1
H NMR spectra were recorded on a Bruker AC spectrometer
(200 MHz) in deuterated chloroform (99.96% d; Sigma Aldrich).
Circular dichroism spectroscopy
Circular dichroism spectra were obtained using a JASCO J-715
spectropolarimeter. All spectra were recorded at r.t., using Hellma
quartz cells with Suprasil® windows, optical path-length of
0.02 cm, a bandwidth of 2 nm, and a time constant of 2 s at a scan
speed of 50 nm/min. The signal-to-noise ratio was improved by
accumulating four scans. The values are expressed in terms of
2
À1
R
[θ] , the total molar ellipticity [degcm dmol ] divided by the
number of amino acid residues. Spectrograde TFE and aqueous
solution of 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) and
SDS were used as solvents.
Synthesis of Fmoc-Lys(Pal-γ-Glu-OtBu)-OH
Pal-Glu-OtBu
(
1) A solution of HOSu (2.7 g, 23.4mmol) in 30 ml of MeCN was
Materials and Methods
added to a solution of Pal-OH (6 g, 23.4mmol) in 90 ml of DCM,
followed by DIC (3.62 ml, 23.4mmol). The reaction mixture was
stirred for 38 h, and then additional 0.3eq of HOSu and DIC were
added. The precipitate was filtered off; the reaction mixture was
concentrated until formation of the precipitate and kept overnight
at 4 °C. The product was filtered off, the surnatant was concentrated
to 50 ml, and the resulting material was crystallized by adding
1
1
2
Materials
Iris Biotech: N,N-dimetylformamide (DMF), dichloromethane (DCM)
(
peptide grade), N-methyl-2-pyrrolidone, N,N-diisopropylethylamine
DIPEA), trifluoroacetic acid (TFA), piperidine, N,N′-diisopropylcarb-
odiimide (DIC), 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo
4,5-b]pyridin-3-oxy hexafluorophosphate, H-Glu-OtBu (OtBu,
(
[
00ml of petroleum ether. (2) To a solution of Pal-OSu (4.65 g,
3.2 mmol), prepared in step (1), in 90 ml of DCM, DIPEA (4.15 ml,
3.8 mmol) and H-Glu-OtBu (3.08 g, 17.2mmol) were added. The
tert-butoxy). Sigma Aldrich: acetonitrile (MeCN) for mass spectrome-
try (>99.9%), TFA for mass spectrometry (>99.9%), ethyl acetate,
methyl tert-butylether (MTBE), acetone, triisopropylsilane (TIS),
reaction mixture was stirred for 48 h. Then, the solvent was evapo-
rated, the solid residue was dissolved in 150 ml of ethyl acetate, and
the solution was washed with 3 × 100ml of 10% KHSO
2,2,2-trifluoroethanol (TFE), N-hydroxysuccinimide (HOSu), 1-hy-
droxy-1,2,3-benzotriazole (HOBt), 2,4,6-trimethylpyridine (collidine),
sodium dodecyl sulfate (SDS), 2-oleoyl-1-palmitoyl-sn-glycero-3-
phosphocholine, petroleum ether (40–60), anhydrous MgSO , NaCl.
4
Carbosynth: N,N,N′,N′-tetramethyl-O-(1H-benzotriazol-1-yl)uronium
hexafluorophosphate (HBTU). CBL Biopharma: 2-chlorotrityl
chloride resin. Chengdu Baishixing Science and Technology
4
in water
and 100 ml of water. Then, the organic layer was dried over MgSO
4
,
concentrated until precipitation of the product and kept overnight
at 4 °C. The product was filtered off, the surnatant was concen-
trated, and crystallization was repeated twice. The product was
obtained as a solid in an overall yield of 5 g (85%).
Industries
methyloxycarbonyl), Fmoc-Asp(OtBu)-OH, Fmoc-Arg(Pbf)-OH (Pbf,
,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl), Fmoc-Gln
Trt)-OH (Trt, trityl), Fmoc-Glu(OtBu)-OH, Fmoc-Gly-OH, Fmoc-His
Trt)-OH, Boc-His(Trt)-OH, Fmoc-Ile-OH, Fmoc-Leu-OH, Fmoc-Lys
Boc)-OH (Boc, tert-butyloxycarbonyl), Fmoc-Phe-OH, Fmoc-Ser
Co.,
Ltd:
Fmoc-Ala-OH
(Fmoc,
fluorenyl-9-
Fmoc-Lys(Pal-γ-Glu-OtBu)-OH
2
(
(
(
(
1) A solution of HOSu (1.57 g, 13.5mmol) in 10ml of MeCN was
added to a solution of Pal-Glu-OtBu (5g, 11.3mmol) in 50 ml of
DCM, followed by DIC (2.09 ml, 13.5 mmol). The reaction mixture
was stirred for 48 h. Then, the precipitate was filtered off, and the
solvent was evaporated. The solid residue was dissolved in 60 ml
of ethyl acetate, and the solution was washed with 3 × 100 ml of
in
(tBu)-OH, Fmoc-Thr(tBu)-OH, Fmoc-Trp(N Boc)-OH, Fmoc-Val-OH.
Chromatography
1
0% KHSO
4
in water and 100 ml of water. Then, the organic layer
, concentrated until precipitation of the prod-
Ultra Performance Liquid Chromatography (UPLC) analysis was
performed on an Agilent Technologies 1290 series instrument
using an Agilent Zorbax Ecplipse Plus C8 column (4.6× 50 mm,
was dried over MgSO
4
uct and kept overnight at 4 °C. The product was filtered off, the
surnatant was concentrated, and crystallization was repeated twice.
(2) To a solution of Pal-Glu(OSu)-OtBu (4.3g, 7.8mmol), prepared in
step (1), in 50 ml of DCM, a solution of Fmoc-Lys-OH · TFA (5.76 g,
12 mmol, obtained by treatment of commercial Fmoc-Lys(Boc)-
2
1.8 μm) and eluants A: TFA/H O 0.1% v/v, B: TFA/MeCN 0.07% v/v;
detection at 224 nm. Preparative purification was carried out on
an Agilent Technologies 1260 Infinity chromatograph using an
J. Pept. Sci. 2016
wileyonlinelibrary.com/journal/jpepsci

GURYANOV ET AL.
+
calcd
OH with 30% TFA in DCM) in 100 ml of DCM was added, followed by
DIPEA (6.95 mL, 39.3mmol). The reaction mixture was stirred for
TFA/TIS/water (95:2.5:2.5v/v): ESI-ToF [M+ H]
= 1699.76, [M
+
+ H] = 1699.74.
expt
2
4 h. Then, the reaction mixture was washed with 3 × 150 ml of
Protected segment 17-30: loading of Arg 0.35 mmol/g, total
+
+
10% KHSO in water and 100 ml of water. The organic layer was
yield ESI-ToF [M + H]
= 3194.06, [M + H]
= 3194.14.
4
calcd
expt
dried over MgSO , concentrated until precipitation of the product
and kept overnight at 4 °C. The product was filtered off, the
surnatant was concentrated and crystallization was repeated twice.
4
Synthesis of peptide Fmoc-19-31-OtBu
In 25 ml of DMF, 0.75 g (0.27 mmol) of the protected segment 19-30
was dissolved. A solution of HBTU (100 mg, 0.27 mmol), HOBt
The product was obtained as a solid with a HPLC purity of 95%. The
+
overall yield was 4.2 g (66%). ESI-ToF [M+ H]
= 793.08, [M
calcd
(36mg, 0.27 mmol), and DIPEA (125 μl, 0.71 mmol) in 5 ml of DMF
+
1
+
H]
= 793.11; H NMR (200 MHz, CDCl ): δ = 7.72 (d, 2H, Fmoc),
expt 3
was added, followed by H-Gly-OtBu•HCl (70 mg, 0.41 mmol). The
progress of the reaction was monitored using UPLC-ESI-ToF. After
completion of the reaction, the product was extracted in 20 ml of
ethyl acetate and washed with 3 × 50 ml of 10% KHSO
and 50 ml of water. The organic layer was separated, dried over
7.59 (d, 2H, Fmoc), 7.24-7.41 (m, 4H, Fmoc), 6.73 (t, 1H, NH), 6.58
(
d, 1H, NH), 5.78 (d, 1H, NH), 4.35 (d, 2H, CH ), 4.23 (t, 1H, CH Fmoc),
2
4,20 (t, 1H, CH), 3.80 (t, 1H, CH), 3.26 (m 2H, CH ), 2.24 (m, 4H, 2CH ),
2 2
4
in water
1
.51-2.15 (m, 6H, 3CH ), 1.44 (s, 9H, tBu), 1.24 (m, 26H, 13CH ), 0.87
2
2
3
(t, 3H, CH ).
MgSO
as
4
, and the solvent was evaporated. The product was obtained
a
yellowish solid in 75% yield (0.58 g). ESI-ToF [M
+
+
Synthesis of Liraglutide
+
H]calcd = 2753.53 [M + H]expt = 2753.51.
Solid-phase synthesis
Coupling of segments 11-18 and 19-31
Solid-phase synthesis of both peptides with the full amino acid
sequence and the segments of liraglutide was carried out on a
Biotage Syrowave instrument. In a typical procedure, 0.25 g of
The protected segment 19-31, described earlier, was dissolved in
50 ml of DMF, and 20 ml of piperidine was added. After 2 h, piperi-
dine was evaporated; the Fmoc-deprotected segment 19-31 was
precipitated by adding 100ml of water and filtered. The remaining
solid was washed with 2 × 50 ml of MeCN and dried. The segment
2-chlorotrityl chloride resin with an initial loading of 1.6mmol/g
was used. The resin was swollen in 2 ml of DMF for 30 min, and
the first amino acid (0.8eq) was coupled in presence of DIPEA
11-18 (312 mg, 0.20 mmol), HBTU (115 mg, 0.30 mmol), HOBt
(1.6eq with respect to the amount of amino acid) in 2 h. Then, the
(40 mg, 0.3 mmol), and collidine (40 μl, 0.3mmol) were dissolved
remaining reactive sites of the solid support were capped with
methanol (1.6 eq); the resin was washed with DMF (2 × 2 ml), DCM
in 30 ml of DMF. Then, the Fmoc-deprotected segment 19-31 was
added under stirring. After a week, only traces of the product were
found.
(2× 2 ml), and dried. Loading of the resin was determined spectro-
photometrically after Fmoc-deprotection of a sample of the resin.
The solid-phase synthesis was carried out using 5 eq of protected
amino acids, 5 eq of HBTU, and 8 eq of DIPEA in 30 min. For the
coupling of Fmoc-Lys(Pal-γ-Glu-OtBu)-OH (3eq), time of the
reaction was increased to 10 h. In case of an incomplete coupling,
the reaction was repeated, and finally, the unreacted peptide
chains were acylated with acetic anhydride (10eq) in the presence
of DIPEA (10 eq). For the Fmoc-deprotection, 20% piperidine in
DMF was used (two cycles for 5 and 20 min). The solid-phase syn-
thesis was repeated to obtain the desired peptide sequence. At
the end of the synthesis, the peptide was cleaved from the resin
with 3 ml of the mixture TFA/TIS/water (95:2.5:2.5v/v) and
precipitated in 5 ml of MTBE. The product was filtered off and dried.
In case of the liraglutide segments, the peptide was cleaved from
the resin using a 1% TFA solution in DCM for 5 min, the resin was
filtered off, and the solution was neutralized with DIPEA. The
solvent was evaporated; the product was washed with MTBE
Results and Discussion
Synthesis
The solid-phase synthesis of branched peptides is particularly
difficult because steric hindrance, which hampers complete
amino acid couplings, induces formation of a variety of impuri-
ties. In the case of liraglutide, an additional issue arises from
the presence of the long aliphatic chain, which can lead to
peptide aggregation. The main method described in the litera-
ture for the fully chemical synthesis of liraglutide involves the
use of a protecting group for the ε-NH group of Lys (e.g., Mtt,
2
4-methyl trityl or Alloc, allyloxycarbonyl) [20,26] that should be
orthogonal to those employed in the Fmoc strategy. In this
way, the ε-NH group of Lys can be deprotected without affect-
2
ing the side-chain protecting groups of the other amino acids.
(2ml) and dried.
Then, it can be acylated with Pal-Glu-OtBu or, in two steps, with
α
α
Liraglutide: loading of Gly 0.7mmol/g, total yield 505 mg
an N -protected Glu-OtBu derivative followed by N -deprotection
and Pal-OH coupling.
+
3+
(
77%); ESI-ToF [M+ H]calcd = 3751.29, [M + 3H]expt = 1251.32
Protected segment 1-10: loading of Val 0.45mmol/g, total
Our approach, which encompasses the use of both solid-phase
and solution syntheses, allowed us to circumvent the drawbacks
associated with the difficult construction of peptide side arms by
conventional solid-phase strategies. We synthesized in solution
the Lys -lipidated segment, Fmoc-Lys(Pal-γ-Glu-OtBu)-OH
(Scheme 1), and subsequently inserted it into the growing polypep-
+
yield 150 mg (74%); ESI-ToF [M + H]calcd = 1808.21, [M
+
+
H]expt = 1808.27.
Protected segment 11-18: loading of Ala 0.8mmol/g, total
+
20
yield 260 mg (84%); ESI-ToF [M + H]calcd = 1542.89, [M
+
+
H]expt = 1542.81.
Protected segment 19-30: loading of Arg 0.35mmol/g, total
tide chain on the solid support.
+
yield 190 mg (80%); ESI-ToF [M + H]calcd = 2752.53, [M
The choice of the experimental conditions (preactivation of
Pal-OH and Pal-Glu-OtBu with formation of the respective -OSu
active esters) allowed us to use H-Glu-OtBu (in the first step) and
Fmoc-Lys-OH without additional protections. The desired
+
+
H]expt = 2752.52.
Protected segment 1-16: loading of Gly 0.9mmol/g, total
yield 440mg (77%); because of solubility problems, the peptide
for characterization was fully deprotected with the mixture
α
N -palmitoylated dipeptide was obtained in excellent purity and
wileyonlinelibrary.com/journal/jpepsci
J. Pept. Sci. 2016

INNOVATIVE CHEMICAL SYNTHESIS AND CONFORMATIONAL HINTS ON LIRAGLUTIDE
Scheme 1. Synthesis of Fmoc-Lys(Pal-γ-Glu-OtBu)-OH.
Figure 2. Ultra performance liquid chromatography profile of Fmoc-Lys(Pal-γ-Glu-OtBu)-OH.
from the resin with TFA (Scheme 2). The UPLC profile of the
crude peptide is shown in Figure 3A. It can be noted that almost
all peaks, including that of liraglutide, are doubled, with the first
peak of each couple having a molecular weight 44 D higher than
that of the second peak. This side product is generated by
formation of N-carboxylated tryptophan (M+ 44), which can be
easily converted to the desired product by lyophilization under
acidic conditions (TFA) (Figure 3B) [27]. Only traces of the
impurity remained, but they were left aside during the following
HPLC purification and lyophilization.
Thus, we reduced significantly the time required for the
solid-phase synthesis and obtained a high yield of the crude
α
product. The truncated, N -acetylated segment 21-31, lacking
the palmitoylated moiety, has a quite different retention time
in the reverse-phase HPLC separation, which simplified peptide
purification. Liraglutide was obtained with a purity greater than
98% after a chromatographic purification by means of prepara-
tive reverse-phase HPLC. Its identity was assessed by mass
spectrometry and by comparison with a commercially available
sample.
We also investigated two synthetic variants for the preparation of
liraglutide (Schemes 3 and 4) that comprise the syntheses of the
side-chain protected segments of liraglutide and their couplings
in solution.
Scheme 2. The solid-phase approach used to perform the synthesis of
liraglutide.
To this purpose, we prepared on solid phase a number of
segments, namely 1-10, 11-18, 19-30, 1-16, and 17-30 in order to
combine them by classical solution peptide synthesis following
the general schemes {1-10 + [11-18 + (19-30 + 31)]} and [1-16
+ (17-30 + 31)] (Table 2).
high yield by performing extractions and crystallizations only
(Figure 2) (Materials and Methods). Thus, this approach can be
particularly convenient for the industrial production of this
dipeptide.
After insertion of Fmoc-Lys(Pal-γ-Glu-OtBu)-OH into the pep-
tide chain and capping of the unreacted peptide with acetic
anhydride, the conventional solid-phase synthesis was continued
to produce the full-sequence liraglutide. Then, all side-chain
protecting groups were removed, and the peptide was cleaved
Unfortunately, we faced severe solubility problems with the
protected segment 1-10 in all organic solvents commonly used in
peptide synthesis, because of its extreme hydrophobicity and
tendency to aggregate. Therefore, we were forced to give up with
the strategy [1-16 + (17-30 + 31)].
J. Pept. Sci. 2016
wileyonlinelibrary.com/journal/jpepsci

GURYANOV ET AL.
Figure 3. Ultra performance liquid chromatography profiles of crude liraglutide (A), after treatment with trifluoroacetic acid and lyophilization (B), and after
preparative HPLC purification (for a better resolution a slower gradient was used) (C). In the inset of (C), the ESI-MS spectrum is shown (for the experimental
conditions, see Materials and Methods).
Scheme 3. The combined solid-phase/solution syntheses of liraglutide using segments strategy {1-10 + [11-18 + (19-30 + 31)]}.
wileyonlinelibrary.com/journal/jpepsci
J. Pept. Sci. 2016

INNOVATIVE CHEMICAL SYNTHESIS AND CONFORMATIONAL HINTS ON LIRAGLUTIDE
Scheme 4. The combined solid-phase/solution syntheses of liraglutide using segments strategy [1-16 + (17-30 + 31)].
Table 2. Primary structures of the segments of liraglutide
1
10
Segment 1-10
Fmoc-His(Trt) -Ala-Glu(OtBu)-Gly-Thr(tBu)-Phe-Thr(tBu)-Ser(tBu)-Asp(OtBu)- Val -OH
1
1
18
Segment 11-18
Segment 19-30
Segment 1-16
Segment 17-30
Fmoc-Ser(tBu) - Ser(tBu)-Tyr(tBu)-Leu-Glu(OtBu)-Gly-Gln(Trt)- Ala -OH
19
in
30
Fmoc-Ala -Lys(Pal-Glu-OtBu)-Glu-(OtBu)-Phe-Ile-Ala-Trp(N Boc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf) -OH
1
16
Boc-His(Trt) -Ala-Glu(OtBu)-Gly-Thr(tBu)-Phe-Thr(tBu)-Ser(tBu)-Asp(OtBu)-Val-Ser(tBu)-Ser(tBu)-Tyr(tBu)-Leu-Glu(OtBu)- Gly -OH
17
in
30
Fmoc-Gln(Trt) -Ala-Ala-Lys(Pal-Glu-OtBu)-Glu-(OtBu)-Phe-Ile-Ala-Trp(N Boc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf) -OH
Figure 4. Ultra performance liquid chromatography profile of the reaction mixture at the end of the coupling reaction between the protected segment 19-
30 and H-Gly-OtBu (because the peak at 7.9 min does not have a total ion current signal, it was not integrated).
The first step of the synthetic approach {1-10 + [11-18 + (19-30
31)]} (coupling of the C-terminal Gly) afforded the desired
and 19-31. Unfortunately, coupling of these two segments of
liraglutide was unsuccessful. Even after a week time of reaction,
only traces of the desired product were found.
+
protected peptide Fmoc-19-31-OtBu (Figure 4).
α
However, we also obtained the N -acetylated peptide Ac-19-
3
1-OH that was formed because of the partial Fmoc-
CD Conformational Analysis
deprotection during solid-phase synthesis of the segment 19-
19
30 because of the long reaction time of Ala coupling with fol-
The remarkable resistance to enzymatic hydrolysis of liraglutide
has been the subject of several investigations and inspiration
for new, long-lasting analogs in humans [19]. Understanding
the origin of this positive behavior would allow one to apply
the same approach to a number of other peptides of similar
lowing capping with acetic anhydride of the deprotected pep-
tide. Clearly, this impurity cannot react with the segment 11-18
in the following step. Therefore, the synthesis was continued
to the next step of coupling of the protected segments 11-18
J. Pept. Sci. 2016
wileyonlinelibrary.com/journal/jpepsci

GURYANOV ET AL.
Figure 5. Circular dichroism spectra of liraglutide in 2,2,2-trifluoroethanol (TFE), 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC), and sodium dodecyl
-
4
sulfate (SDS) at 10 M peptide concentration (A); circular dichroism spectra as a function of time of liraglutide in a pH 6.7 phosphate buffer/TFE mixture at
-4
a 7 : 3 ratio (peptide concentration 10 M) (B) and in the presence of DOPC vesicles (C).
chemical structure. Lipidation of liraglutide is believed to be the
Conclusions
key feature responsible for its proteolytic resistance. In a recent
CD and light scattering investigation, it was observed that
liraglutide forms aggregates at pH values lower than 6.9, at
variance from slightly alkaline pH values [23]. In addition, it
was seen that aggregation is fully reversible. However, these
studies were performed in buffered aqueous solutions, not fully
representative of the complex surrounding found in a living
organism. Besides the previously mentioned role of albumin, also
micelles or phospholipid bilayers of the biological fluids might
affect conformation and bioactivity of liraglutide, as it was
shown, for example, for lipidated vasoactive intestinal peptide
Liraglutide, an important lipopeptide drug used for the treatment
of type II diabetes, can be chemically synthesized by means of a
new, convenient, solid-phase procedure. The key step is the
insertion of a protected, lipidated dipeptide directly on solid phase.
Our conformational analysis, performed by CD in membrane
mimetic environments, suggests that the remarkable resistance to
enzymatic degradation of liraglutide may not be ascribed solely
to aggregate formation. Indeed, we observed that micelles and
phospholipid vesicles do not allow the peptide to form aggregates.
Therefore, additional studies are required to fully disclose the
mechanism of action of liraglutide.
[28]. Therefore, we decided to study the conformational
preferences of liraglutide in membrane-like environments. The
CD spectra of liraglutide in SDS micelles, DOPC vesicles, and
TFE are reported in Figure 5.
References
In all three environments, the peptide adopts a predominantly
α-helical conformation (Figure 5A). Although TFE as well would
induce helix formation [29–32], its effect is less evident, as
compared with those of SDS and DOPC. Also, it can be noted
that the α-helix is less pronounced in the case of SDS micelles
with respect to DOPC vesicles, probably because both SDS and
liraglutide are characterized by an overall negative charge, and
therefore, a destabilization (although modest) of the peptide
1
World Health Organization. Fact Sheet No. 312 “Diabetes”. Available at:
http://www.who.int/entity/mediacentre/factsheets/fs312/en/index.html
2 Ryan GJ, Hardy Y. Liraglutide: once-daily GLP-1 agonist for the treatment
of type 2 diabetes. J. Clin. Pharm. Ther. 2011; 36: 260–274.
3
4
Robertson C. Incretin-related therapies in type 2 diabetes: a practical
overview. Diabetes Spectr. 2011; 24: 26–35.
Masband S. Exenatide and liraglutide: different approaches to develop
GLP-1 receptor agonists (incretin mimetics) – preclinical and clinical
results. Best Pract. Res. Clin. Endocrinol. Metab. 2009; 23: 463–477.
Brubaker PL, Drucker DJ. Structure-function of the glucagon receptor
family of G protein–coupled receptors: the glucagon, GIP, GLP-1, and
GLP-2 receptors. Recept. Channels 2002; 8: 179–188.
3D-structure is reasonably expected.
5
To check whether a time-dependent conformation change of
liraglutide would take place in a hydrophobic environment, as
reported for buffered aqueous solutions [23], we measured the
CD spectra as a function of time at pH6.7 in phosphate
buffer/TFE solution and in the presence of DOPC vesicles
6 Knudsen LB, Nielsen PF, Huusfeldt PO, Johansen NL, Madsen K,
Pedersen FZ, Thøgersen H, Wilken M, Agerso H. Potent derivatives of
glucagon-like peptide-1 with pharmacokinetic properties suitable for
once daily administration. J. Med. Chem. 2000; 43: 1664–1669.
(Figures 5B and C). No noticeable changes occurred up to 100
7
Madsen K, Knudsen LB, Agerso H, Nielsen PF, Thøgersen H, Wilken M,
Johansen NL. Structure-activity and protraction relationship of long-
acting glucagon-like peptide-1 derivatives: importance of fatty acid
length, polarity, and bulkiness. J. Med. Chem. 2007; 50: 6126–6132.
Rossi MC, Nicolucci A, Nauck MA. Liraglutide in type 2 diabetes: from
pharmacological development to clinical practice. Acta Biomed. 2009;
and 48h, respectively. Moreover, we did not observe any
increase of the 222nm/208 nm ellipticity ratio, indicative of pep-
tide aggregation [32]. Apparently, the absence of conformational
transitions over time, reported for aqueous solutions [23], can be
ascribed to the presence of the vesicles or of the membrane
mimetic solvent TFE. Thus, given the complex composition of
the biological fluids, we are inclined to conclude that aggrega-
tion and albumin interactions might not be the only factors
responsible for the prolonged resistance in vivo of liraglutide.
Peptide stabilization through interaction with biological
membranes may play a significant role as well.
8
8
0: 93–101.
9 Victoza (package insert). Novo Nordisk Inc., Princeton, NJ, 2010.
10 Cada DJ, Levien TL, Baker DE. Formulary drug reviews. Liraglutide. Hosp.
Pharm. 2010; 45: 552–563.
11 Patent WO 01/66135 A1.
1
2 Davies MJ, Kela R, Khunti K. Liraglutide – overview of the preclinical and
clinical data and its role in the treatment of type 2 diabetes. Diabetes
Obes. Metab. 2011; 13: 207–220.
wileyonlinelibrary.com/journal/jpepsci
J. Pept. Sci. 2016

INNOVATIVE CHEMICAL SYNTHESIS AND CONFORMATIONAL HINTS ON LIRAGLUTIDE
13 Goodwin D, Simerska P, Toth I. Peptides as therapeutics with enhanced
bioactivity. Curr. Med. Chem. 2012; 19: 4451–4461.
Presented at The American Diabetes Association’s 68th Scientific
Sessions; June 6–10, 2008; San Francisco, CA..
1
4 Dalbøge LS, Pedersen SL, van Witteloostuijn SB, Rasmussen JE,
Rigbolt KTG, Jensen KJ, Holst B, Vrang N, Jelsing J. Synthesis
and evaluation of novel lipidated neuromedin U analogs with
increased stability and effects on food intake. J. Pept. Sci. 2015;
23 Wang Y, Lomakin A, Kanai S, Alex R, Benedek GB. Transformation of
oligomers of lipidated peptide induced by change in pH. Mol. Pharm.
2015; 12: 411–419.
24 Patent WO2011/057237.
2
1: 85–94.
25 Bondesan A. “Preparation of the peptide liraglutide via novel synthetic
strategies”. Master Thesis in Industrial Chemistry, University of Padova,
Italy, December 2014.
26 Chhabra SR, Hothi B, Evans DJ, Whitea PD, Bycroft BW, Chan WC. An
appraisal of new variants of Dde amine protecting group for solid
1
5 Santoprete A, Capitò E, Carrington PE, Pocai A, Finotto M, Langella A,
Ingallinella P, Zytko K, Bufali S, Cianetti S, Veneziano M, Bonelli F,
Zhu L, Monteagudo E, Marsh DJ, SinhaRoy R, Bianchi E, Pessi A. DPP-IV-
resistant, long-acting oxyntomodulin derivatives. J. Pept. Sci. 2011; 17:
2
70–280.
phase peptide synthesis. Tetrahedron Lett. 1998; 39: 1603–1606.
27 Franzén H, Grehn L, Ragnarsson U. Synthesis, properties, and use of N -
in
1
6 Li Y, Shao M, Zheng X, Kong W, Zhang J, Gong M. Self-assembling
peptides improve the stability of glucagon-like peptide-1 by
forming a stable and sustained complex. Mol. Pharm. 2013; 10:
Boc tryptophan derivatives. J. Chem. Soc. Chem. Commun. 1984; 24:
1699–1700.
3
356–3365.
28 Longo E, De Santis E, Hussain R, van der Walle CF, Casas-Finet J, Uddin S,
Dos Santos A, Siligardi G. The effect of palmitoylation on the
conformation and physical stability of a model peptide hormone. Int. J.
Pharm. 2014; 472: 156–164.
1
7 Ward BP, Ottaway NL, Perez-Tilve D, Ma D, Gelfanov VM, Tschop MH,
Di Marchi RD. Peptide lipidation stabilizes structure to enhance
biological function. Mol. Metab. 2013; 2: 468–479.
1
8 Patent WO 96/29342.
29 Toniolo C, Falxa ML, Goodman M. Conformational aspects of
polypeptides. XXV. Solvent and temperature effects on the
conformations of copolymers of benzyl and methyl L-aspartate with
nitrobenzyl L-aspartate. Biopolymers 1968; 6: 1579–1603.
30 Goodman M, Verdini AS, Toniolo C, Phillips WD, Bovey FA. Sensitive
criteria for the critical size for helix formation in oligopeptides. Proc.
Natl. Acad. Sci. U. S. A. 1969; 64: 444–450.
31 Vymětal J, Bednárová L, Vondrášek J. Effect of TFE on the helical content
of AK17 and HAL-1 peptides: theoretical insights into the mechanism of
helix stabilization. J. Phys. Chem. B 2016; 120: 1048–1059.
32 Toniolo C, Formaggio F, Woody RW. Electronic circular dichroism of
peptides.In Comprehensive Chiroptical Spectroscopy, Vol. , 2 BerovaN ,
PolavarapuPL , NakanishiK , WoodyRW (eds.). Wiley: New York, 2012;
499–544.
1
9 Lau J, Bloch P, Schäffer L, Pettersson I, Spetzler J, Kofoed J, Madsen K,
Knudsen LB, McGuire J, Steensgaard DB, Strauss HM, Gram DX,
Knudsen SM, Nielsen FS, Thygesen P, Reedtz-Runge S, Kruse T.
Discovery of the once-weekly glucagon-like peptide-1 (GLP-1)
analogue semaglutide. J. Med. Chem. 2015; 58: 7370–7380.
0 Mäde V, Els-Heindl S, Beck-Sickinger AG. Automated solid-phase peptide
synthesis to obtain therapeutic peptides. Beilstein J. Org. Chem. 2014; 10:
2
1197–1212.
21 Knudsen LB. Liraglutide: the therapeutic promise from animal models.
Int. J. Clin. Pract. 2010; 167: 4–11.
2
2 Steensgaard DB, Thomsen JK, Olsen HB, Knudsen LB. The molecular
basis for the delayed absorption of the once-daily human GLP-1
analogue, liraglutide. Diabetes 2008; 57(suppl. 1): A164 Abstract 552-P.
J. Pept. Sci. 2016
wileyonlinelibrary.com/journal/jpepsci