Proteolytic stability of cyclic α-hydrazino acid containing peptides: a qualitative study

Article abstract of DOI:10.1016/j.tetlet.2016.07.050

Synthesis of cyclic α-hydrazino acid containing tripeptides Phe-X-Val and evaluation their proteolytic stability against digestive enzymes has been reported for the first time. Tripeptides were treated with the digestive juice of mice and incubated in buf

Full text of DOI:10.1016/j.tetlet.2016.07.050

Tetrahedron Letters 57 (2016) 3858–3861
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Proteolytic stability of cyclic
peptides: a qualitative study
a,y,à
a
-hydrazino acid containing
a,y
b
a,
⇑
Indranil Duttagupta
, Jhuma Bhadra , Sanjib Kr. Das , Surajit Sinha
a
Department of Organic Chemistry, Indian Association for the Cultivation of Science, Jadavpur, Kolkata 700 032, India
Taki Government College, Taki, West Bengal 743429, India
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Synthesis of cyclic a-hydrazino acid containing tripeptides Phe-X-Val and evaluation their proteolytic
Received 19 May 2016
Revised 9 July 2016
Accepted 14 July 2016
Available online 17 July 2016
stability against digestive enzymes has been reported for the first time. Tripeptides were treated with
the digestive juice of mice and incubated in buffer at 37 °C for 45 min and was found to be stable when
X = 5,6,7-membered hydrazino acids with d-NH/NHCbz whereas X = proline was digested. Digestion
process was monitored by LC–MS analysis.
Ó 2016 Elsevier Ltd. All rights reserved.
Keywords:
a-Hydrazino acids
Tripeptides
Digestive enzymes
Peptidase stability
LC–MS analysis
There is a general perception that small peptides, equipped
with high selectivity, lower toxicity, and high chemical/biological
diversity than small molecules, might be the best candidates for
therapy of various diseases.1
tion is reversible) or, are expected to interact directly with the tar-
get. However, covalent conjugation usually lowers potency but
compensates by increasing the half life of the peptide. Another rel-
evant approach is to modify the peptide from within using unnat-
4
Despite the selectivity and the perceived efficacy, rapid in vivo
elimination of therapeutic peptides remains one of the major chal-
ural or non-proteinogenic amino acids. The approach exploits the
substrate specificity of enzymes. A slight modification in the amino
acid backbone can render it immune towards proteolytic
degradation.
1
lenges to be addressed. Since most therapeutic peptides are mere
modifications of native sequences, imparting stability to them by a
way of chemical modifications, so that they are less easily
degraded (proteolysis) or less susceptible to renal clearance takes
in a bit of extra effort which in turn has the potential to seriously
alter the potency of the drug candidate. Chemical modifications,
either within the sequence or by conjugation (to polymer, lipid)
has been tried to some extent. A wide range of polymeric conju-
gates are now available,2 including polyethylene glycol (PEG),
hydroxyethyl starch (HES), human serum albumin (HSA), XTEN,
PAS (recombinant polypeptides), polyglutamic acid, and mono-
clonal antibodies. Covalent modifications to create new chemical
entity (NCE) and conjugating them to drug candidates3 are also
known to alter the pharmacokinetics as well as pharmacodynamics
of the original drug. NCE conjugated pharmaceutical ingredients
are either expected to have pro drug characteristics (if the conjuga-
Since the discovery of piperazic acid (Fig. 1) (hexahydropyri-
dazine-3-carboxylic acid) a non-proteinogenic amino acid (from a
5
group of cyclodepsipeptide natural product ) by Hassall and
coworkers as a component of monamycins, it or similar such
motifs has been in focus for its presence in other naturally occur-
ring cyclodepsipeptide.
Peptides containing repeating linear alpha hydrazino carboxylic
acids or azapeptide linkage incorporated into various enzyme inhi-
bitors are known to be stable under proteolytic conditions.4
Thinking alike we hypothesized that their cyclic counterparts
should exhibit similar properties. The d-nitrogen atom of cyclic
c,6
a
-hydrazino acid is expected to play a pivotal role in imparting
stability. The similarity of the cyclic -hydrazino acids with
proline encouraged us to substitute proline with cyclic
a
a-
hydrazino acid (Fig. 1) in a model tri-peptide and probe their
effect on the pharmacokinetics. Herein we wish to report a few
relevant observations.
Literature study of previously reported enzymatic studies on
short peptide sequences indicated that tripeptides with the general
⇑
Corresponding author. Tel.: +91 33 2473 4971; fax: +91 33 2473 2805.
E-mail address: ocss5@iacs.res.in (S. Sinha).
Authors equally contributed.
y
à
Present address: Institute of Radical Chemistry, Aix-Marseille University,
Marseille Cedex 20, France.
http://dx.doi.org/10.1016/j.tetlet.2016.07.050
0
040-4039/Ó 2016 Elsevier Ltd. All rights reserved.

I. Duttagupta et al. / Tetrahedron Letters 57 (2016) 3858–3861
3859
COOH
O
O
COOH
COOH
NH2
1
, n = 1, 2, 3
HN
NH
NH
NH
NH
HO C
N
X = CH , NH, NCbz
2
NH
NH
2
Proline analogue,
X = CH , n = 1
2
X
Ph
n
Figure 1. Structure of cyclic a-hydrazino acids.
Figure 2. General structure of model tripeptide.
structure Phe-
Accordingly tripeptide 1 was chosen as our model substrate
Fig. 2). The cyclic -hydrazino acid in the central position was
varied (analogues and enantiomers). The proteolytic stability was
W
Pro-Val are recognized by the enzymes.7a,d
MeO
O
(
a
NH
compared to the stability of proline (when X = CH
containing peptide under similar conditions.
2
,
n = 1)
NCbz
2
H
N
H
N
MeO C
Methyl ester of mono Cbz protected d-azaproline (Aza-Pro)
MeO C
O
N
2
O
2
27b–d
(racemic) was used as the starting material for the
synthesis of the d-azaproline containing protected tripeptide 3
according to our previously reported procedures7 (Fig. 3).
O
O
Ph
Ph
NHFmoc
N
NCbz
NHFmoc
c
NCbz
Piperazic (Piz) acid and its seven membered (diazepane car-
boxylic acid) analogue containing tripeptides were synthesized
from the dipeptides 4 and 5 which in turn were synthesized using
our previously reported procedure.7b D/L Configuration of the
3
a
3b
Figure 3. d-Azaproline containing tripeptides.
7b
hydrazino acids were established previously. Compounds 4 and
7
b
5
on Boc deprotection followed by selective Cbz protection
Ph
Ph
yielded 6 and 7, respectively. The diastereomeric mixtures of
mono protected dipeptides (6a/6b, 7a/7b) were nicely separated
by flash column chromatography using silica gel. Diastereomers
were then separately converted to the corresponding protected
tripeptides by treating with Fmoc-Val-Cl and AgCN in benzene to
obtain 8a, 8b and 9a, 9b, respectively (Scheme 1).
The protected tripeptides were then deprotected to obtain the
model peptides. Compounds 3a, 8a, and 9a were deprotected par-
tially to prepare the Cbz protected tripeptides 10a, 11a, 12a
CO
2
Me
CO Me
2
HN
O
HN
O
1
. TFA (20%), DCM, 4 h
*
*
2
. CbzCl, Et N, DCM
NBoc
3
NH
o
-15 C, 30 min
NBoc
n
NCbz
n
4, n = 1
, n = 2
6
a (44%), 6b (39%), n = 1
7
a (46%), 7b (37%), n = 2
5
Ph
CO Me
2
(Scheme 2) and the compounds were treated with TFA to protonate
O
O
N
Fmoc-L-Val-Cl, AgCN
60 oC, C H , 30 min
the free amine in order to eliminate the chances of intramolecular
HN
7
c
cyclization during the storage or handling.
*
NHFmoc
NCbz
6
6
To synthesize the completely deprotected tripeptides, com-
pounds 3, 8, and 9 were subjected to sequential hydrolysis, hydro-
geno-lysis and Fmoc deprotection. Complete deprotection of
individual isomers yielded the tripeptides 13a, 13b, 14a, 14b,
and 15a, 15b as their respective piperidine salts (Scheme 3).
The stability of these compounds in the presence of enzymes
was evaluated against the standard proline based tripeptide 18 in
order to compare and test the efficacy of the enzymes. Compound
8
9
a (86%), 8b (80%), n = 1
a (69%), 9b (73%), n = 2
n
Scheme 1. Synthesis of protected tripeptides 8 and 9.
CO Me
2
1
. iPrOH, THF, H O
2
O
Ph
O
LiOH, CaCl , rt. 45 min
HN
2
1
8 was prepared by sequential peptide coupling from N-Boc-
L-pro-
N
2. Piperidene (20%),
DMF, rt., 12 h
line. N-Boc- -proline was subjected to EDC coupling with -pheny-
L
L
3
a, n = 1
8a, n = 2
a, n = 3
NHFmoc
NCbz
lalanine methyl ester to yield the protected dipeptide 16. This
dipeptide on N-Boc deprotection followed by EDC coupling with
N-Boc L-valine yielded the protected tripeptide 17. The protected
n
3
. TFA (10%) in
9
o
DCM, 0 C, 10 min
CO -
2
tripeptide 17 on ester hydrolysis and N-Boc deprotection yielded
the TFA salt of the control tripeptide 18 (Scheme 4). TFA salt was
O
Ph
O
Overall yield:
HN
1
3
confirmed by
C
NMR as peak corresponding at 161.9
10a (65%), n = 1
N
1
1a (62%), n = 2
2a (61%), n = 3
NH3+
(
q, JC,F = 36 Hz) and 115.9 (q, JC,F = 288 Hz) are for TFA.
NCbz
1
n
Proteolytic stability of cyclic
peptides
a-hydrazino acid-containing
Scheme 2. Synthesis of Cbz-protected tripeptides 10a, 11a, 12a.
Proteolytic cleavage breaks down proteins in food into smaller
peptides and amino acids so that they can be absorbed and used
by an organism. Similar fate is met by the therapeutic peptides
when taken orally; most of the bioactive peptides undergo prote-
olytic degradation into many smaller peptides and amino acids.
Instead of choosing individual enzymes based on their selectiv-
ity of substrates, a broad range of them ought to be used in our
study. Consequently Swiss Albino mice were chosen as model
mammals for this study.
Inbred mice were purchased and maintained in the laboratory
at 24 °C, 12 h day night cycle and fed on chick peas twice daily.
The mice were anaesthetized before being sacrificed. Extraction
of the digestive enzymes from mouse was done following the
8
reported procedure of Rao et al. The abdomen was opened and
the duodenum and a part of the proximal ileum was removed
under sterile condition and kept in 0.05 M Tris–HCl buffer (pH 9)
at 4 °C for immediate processing. The alimentary canal was slit
opened and washed with 0.05 M Tris–HCl buffer (pH 9) to recover

3
860
I. Duttagupta et al. / Tetrahedron Letters 57 (2016) 3858–3861
CO Me
Table 1
2
1
. LiOH, CaCl ,
2
Relative qualitative analysis of proteolysis
O
Ph
O
THF, iPrOH, H
2
O
HN
Stabilityi
Entry
1)
(2)
Peptides
N
2. Pd/C (10%), H
NHFmoc 3. Piperidine (20%), DMF
2
, MeOH
(
18
Phe-Pro-Val-OH
ꢀ
+
+
3
8
9
a, n = 1
a, n = 2
a, n = 3
NCbz
10a
11a
12a
13a
13b
14a
14b
15a
15b
Phe-(
Phe-(
Phe-(
Phe-(
Phe-(
Phe-(
Phe-(
Phe-(
Phe-(
L
L
L
L
)-AzaPro(Cbz)-Val-OH
)-Piz(Cbz)-Val-OH
)-Diazepane(Cbz)-Val-OH
)-AzaPro-Val-OH
n
(
(
(
(
3)
4)
5)
6)
+
++
++
+
+
++
++
CO H
2
D
)-AzaPro-Val-OH
)-Piz-Val-OH
)-Piz-Val-OH
)-Diazepane-Val-OH
D)-Diazepane-Val-OH
CO H
2
Ph
(7)
(8)
L
O
Ph
HN
O
O
D
HN
O
(
(
9)
10)
L
N
N
NH NH2
i
NH NH2
n
(++) = Stable with negligible proteolysis; (+) = stable with slight proteolysis;
ꢀ) = complete proteolysis, a/b indicates the diastereomers.
n
13a (53%), n = 1 from 3a
14a (53%), n = 2 from 8a
15a (43%), n = 3 from 9a
(
1
3b (71%), n = 1 from 3b
4b (65%), n = 2 from 8b
5b (57%), n = 3 from 9b
1
1
Scheme 3. Synthesis of unprotected tripeptides 13, 14, 15.
the pancreatic enzyme. The extracts were centrifuged, the super-
natant (enzyme solutions) were separately stored at 4 °C until
the experiment.
Solutions of peptides were treated with the enzyme solution9
and the mixtures were analyzed with LCMS (Table 1).
Analysis of the LCMS of control tripeptide, 18
The control tripeptide 18 (Phe-Pro-Val-OH) is expected to
undergo proteolytic degradation since it contains only naturally
occurring amino acids (L-Phe, L-Pro, L-Val). The degradation will
also indicate the activity of extracted digestive enzymes. The frag-
ment at 263.8 corresponds to the Phe-Pro dipeptide whereas the
peak at 118.0 corresponds to the released Val (Fig. 4a). It has
shown the efficacy of the extracted digestive enzymes, as well as
demonstrated the fate of the peptide (or protein) based drugs
under digestive conditions.
Figure 4. Digestion of (a) Control peptide 18 (b) Phe-(
L
)-AzaPro(Cbz)-Val-OH, 10a
⁄
(
c) Phe-(
L)-AzaPro-Val-OH, 13a. ( ) Denotes metabolites and (^) denotes peptides.
Analysis of the LCMS of
L-d-azaproline containing tripeptide
Phe-( )-Aza Pro-Val-OH, 13a
L
Analysis of the LCMS of Cbz protected d-L-azaproline containing
tripeptide 10a
L
-Configuration of d-azaproline in 13a is the closest to that of L-
proline and consequently, stability of this tripeptide is of utmost
interest. Without any protection as that of 10a, peptide 13a is
expected to be prone to proteolysis.
Contrary to the expected degradation, compound 13a showed
remarkable stability towards proteolysis which is evident from
the ion peak at 363.5 (Fig. 4c).
Compound 10a, being protected is not expected to undergo
hydrolysis. The Cbz group at the d-nitrogen is expected to inhibit
the formation of enzyme substrate complex. The molecular ion
peak of 10a at 495.2 and the presence of very little related metabo-
lites at 396.3 indicated its resistance towards proteolytic enzymes
(Fig. 4b).
Analysis of the LCMS of
D-d-azaproline containing tripeptide,
Phe-( )-AzaPro-Val-OH, 13b
D
MeO C
O
L-Phe-OMe.HCl,
EDC,
2
Boc
N
D
-Amino acids are reported to enhance the pharmacokinetics of
N
H
CO H
peptides principally by exploiting the substrate specificity of
enzymes, and hence it was anticipated that the tripeptide 13b
would not undergo proteolysis under these conditions. LCMS of
the mixture showed the anticipated results and the molecular
ion peak at 363.1 corresponding to protonated tripeptide was
observed.
N
2
HOBT, DCM,
DIPEA, 12 h, 89%
Boc
Ph
16
1
0
. TFA (20%), DCM,
CO Me
2
o
C, 4 h
O
O
Ph
HN
2
. Boc-Val-OH, EDC,
N
17
HOBT, DIPEA, DCM,
rt. 12 h, 73%
NHBoc
Analysis of the LCMS of Cbz protected
containing tripeptide, 11a
L-piperazic acid
CO H
2
1
. LiOH, THF, H O
2
O
O
Similar to compound 10a (Cbz protected d-azaproline) 11a was
also expected to resist proteolytic degradation, which was con-
firmed from LCMS analysis with a major peak at 511.4 (M+H )
Ph
HN
18
2
. TFA (20%), DCM
N
71%
+
NH .TFA
2
and slightly hydrolysed products confirmed by corresponding
peaks at 412.2 and 119.4.
Scheme 4. Synthesis of control tripeptide 18.

I. Duttagupta et al. / Tetrahedron Letters 57 (2016) 3858–3861
3861
Analysis of the LCMS of piperazic acid-containing tripeptides, 14
Siddhartha S. Jana, Department of Biological Chemistry, IACS for
helping in mice experiment.
Although piperazic is a non-proteinogenic naturally occurring
amino acid the tripeptides 14 containing both
did not undergo proteolysis in the presence of mammalian diges-
tive enzymes. Peaks were obtained for L-isomer: 377.2 (M+H) ,
L and D-isomers
Supplementary data
+
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2016.07.
+
2
1
78.2 for fragments and for
52 (for piperazic acid).
D-isomer: 377.1 (M+H) , 278.3 and
0
50.
Analysis of the LCMS of Cbz protected
hydrazino acid-containing tripeptides, 12a
L-7-membered cyclic a-
References and notes
1
.
.
(a) Fosgerau, K.; Hoffmann, T. Drug Discovery Today 2015, 20, 122–128; (b)
Jenssen, H.; Aspmo, S. I. Methods Mol. Biol. 2008, 494, 177–186.
Canalle, L. A.; Löwik, D. W. P. M.; van Hest, J. C. M. Chem. Soc. Rev. 2010, 39, 329–
Similar to the previous two tripeptides (10a and 11a) com-
pound 12a also resists proteolysis as the major molecular ion peak
2
353.
+
was obtained at 525.4 (M+H) except two additional short peaks at
3. Ryan, S. M.; Mantovani, G.; Wang, X. X.; Haddleton, D. M.; Brayden, D. J. Expert
Opin. Drug Delivery 2008, 5, 371–383.
2
79.1 and 426.3 correspond to the fragments.
4.
(a) Gentilucci, L.; De Marco, R.; Cerisoli, L. Curr. Pharm. Des. 2010, 16, 3185–
203; (b) Goodwin, D.; Simerska, P.; Toth, I. Curr. Med. Chem. 2012, 19, 4451–
3
Analysis of the LCMS of
acid-containing tripeptides, 15a/b
L
and
D
-7-membered cyclic
a
-hydrazino
4461; (c) Gante, J. Angew. Chem., Int. Ed. 1994, 33, 1699–1720; (d) Chakraborty,
T. K.; Ghosh, S.; Jayaprakash, S. Curr. Med. Chem. 2002, 9, 421–435; (e) Kotha, S.
Acc. Chem. Res. 2003, 36, 342–351.
5
.
.
Bevan, K.; Davies, J. S.; Hassall, C. H.; Morton, R. B.; Phillips, D. A. S. J. Chem. Soc.
(C) 1971, 514.
(a) Oehme, P.; Katzwinkel, S.; Vogt, W. E.; Niedrich, E. Experientia 1973, 29,
The structures closely resembling proline (d-azaproline) and
naturally occurring piperazic acid containing tripeptides were
stable in the presence of mammalian digestive enzymes and anal-
6
9
2
47–948; (b) Dutta, A. S.; Giles, M. B. J. Chem. Soc., Perkin Trans. 1 1976, 244–
48; (c) Greenlee, W. J.; Thorsett, E. D.; Springer, J. P.; Patchett, A. A. Biochem.
ogously structurally more varied 7-membered cyclic
a-hydrazino
Biophys. Res. Commun. 1984, 122, 791–797; (d) Magrath, J.; Abeles, R. H. J. Med.
Chem. 1992, 35, 4279–4283; (e) Gante, J.; Krug, M.; Lauterbach, G.; Weitzel, R.;
Hiller, W. J. Pept. Sci. 1995, 1, 201–206; (f) Xing, R.; Hanzlik, R. P. J. Med. Chem.
acid-containing tripeptides 15a/b was anticipated to be resistant
as well. This was confirmed by LCMS analysis from a major peak
1998, 41, 1344–1351; (g) Venkatraman, S.; Kong, J.; Nimkar, S.; Wang, Q. M.;
+
at 391.3 (M+H) (15a and a minor metabolite peak at 292.5) and
Aube, J.; Hanzlik, R. P. Bioorg. Med. Chem. Lett. 1999, 9, 577–580; (h) Witherell,
G. Curr. Opin. Invest. Drugs 2001, 2, 340–347; (i) Wieczerzak, E.; Drabik, P.;
Lankiewicz, L.; Oldziej, S.; Grzonka, Z.; Abrahamson, M.; Grubb, A.; Brömme, D.
J. Med. Chem. 2002, 45, 4202–4211; (j) Zega, A. Curr. Med. Chem. 2005, 12, 589–
+
3
91.2 (M+H) for 15b.
In conclusion, the above experiments strongly suggest the capa-
bility of cyclic
proteolytic stability to peptides. After establishing the overall
pharmacokinetics of these peptides, the cyclic -hydrazino acids
a-hydrazino acids (protected or native) in imparting
597; (k) Löser, R.; Frizler, M.; Schilling, K.; Gütschow, M. Angew. Chem., Int. Ed.
2008, 47, 4331–4334; (l) Guido, B. G.; Fässler, A.; Capraro, H. G.; Cozens, R.;
Klimkait, T.; Lazdins, J.; Mestan, J.; Poncioni, B.; Rösel, J.; Stover, D.; Tintelnot-
Blomley, M.; Acemoglu, F.; Beck, W.; Boss, E.; Eschbach, M.; Hürlimann, T.;
Masso, E.; Roussel, S.; Ucci-Stoll, K.; Wyss, D.; Lang, M. J. Med. Chem. 1998, 41,
3387–3401.
(a) Keller, M.; Sager, C.; Dumy, P. L.; Schutkowski, M.; Fischer, G. S.; Mutter, M.
J. Am. Chem. Soc. 1998, 120, 2714–2720; (b) Duttagupta, I.; Goswami, K.; Sinha,
S. Tetrahedron 2012, 68, 8347–8357; Optical purification and absolute
configuration was determined (c) Duttagupta, I.; Goswami, K.; Chatla, P.;
Sinha, S. Synth. Commun. 2014, 44, 2510–2519; (d) Duttagupta, I.; Misra, D.;
Bhunya, S.; Paul, A.; Sinha, S. J. Org. Chem. 2015, 80, 10585–10604.
Rao, R. R.; Patel, K.; Srinivasan, K. Nahrung/Food 2003, 47, 408–412.
a
particularly d azaproline can find their applicability in imparting
the stability to relevant bioactive peptides or peptide drug candi-
7
.
.
dates to the replacement of pseudoproline for the treatment of Alz-
heimer’s disease.1 In addition, synthetic methods of cyclic
0
a
-
hydrazino acids-containing peptides with the selective protection
at the d-N and the peptide elongation at the -N has been reported
a
which could be useful in designing the peptides for biological
interest.
8
9. Experimental procedure: (i) A stock solution of the tripeptides (approx 2 mg/
ml) was prepared in the same buffer (0.05 M Tris–HCl, pH 9).
(
(
ii) The enzyme extracts were mixed and mixed well prior to the experiment.
iii) 50 l of peptide stock solution and 20 l of enzyme solution (to make a
g/70 l) were added into 0.5 ml
Eppendorf tube. The mixture was kept at 37 ± 1 °C for 45 min.
iv) 20 l of trichloroacetic acid (TCA) solution (0.5 g TCA into 0.35 ml d H
Acknowledgments
l
l
l
final peptide concentration of 100
l
a
Financial support from Department of Science and Technology,
Government of India, is gratefully acknowledged. J.B. thankful to
IACS for the fellowship. The use of animal facilities within the Insti-
tute is gratefully acknowledged. All animals were treated accord-
ing to the procedures outlined in the Guide for the Care and Use
of Laboratory Animals (IACS), and experimental procedures were
approved by the Animal Ethics Committee of IACS. We thank Dr.
(
l
2
O)
was added to the above solution and mixed well. The cloudy reaction sample
was cooled (4 °C) for 15 min, then spun at 13,000 g (Eppendorf centrifuge) for
2 min to pellet the precipitated serum proteins.
(
v) The supernatant solution was then analyzed using LCMS.
1
0. Adessi, C.; Frossard, M.-J.; Boissard, C.; Fraga, S.; Bieler, S.; Ruckle, T.; Vilbois, F.;
Robinson, S. M.; Mutter, M.; Banks, W. A.; Soto, C. J. Biol. Chem. 2003, 278,
13905–13911.