Facile solid-phase synthesis of sulfated tyrosine-containing peptides: Total synthesis of human big gastrin-II and cholecystokinin (CCK)-39

Article abstract of DOI:10.1021/jo000895y

Chemical synthesis of tyrosine O-sulfated peptides is still a laborious task for peptide chemists because of the intrinsic acid-lability of the sulfate moiety. An efficient cleavage/deprotection procedure without loss of the sulfate is the critical difficulty remaining to be solved for fluoren-9-ylmethoxycarbonyl (Fmoc)-based solid-phase synthesis of sulfated peptides. To overcome the difficulty, TFA-mediated solvolysis rates of a tyrosine O-sulfate [Tyr(SO₃H)] residue and two protecting groups, ^tBu for the hydroxyl group of Ser and 2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl (Pbf) for the guanidino group of Arg, were examined in detail. The desulfation obeyed first-order kinetics with a large entropy (59.6 J·K^-1·mol^-1) and enthalpy (110.5 kJ·mol^-1) of activation. These values substantiated that the desulfation rate of the rigidly solvated Tyr(SO₃H) residue was strongly temperature-dependent. By contrast, the S_N1-type deprotections were less temperature-dependent and proceeded smoothly in TFA of a high ionizing power. Based on the large rate difference between the desulfation and the S_N1-type deprotections in cold TFA, an efficient deprotection protocol for the sulfated peptides was developed. Our synthetic strategy for Tyr(SO₃H)containing peptides with this effective deprotection protocol is as follows: (i) a sulfated peptide chain is directly constructed on 2-chlorotrityl resin with Fmoc-based solid-phase chemistry using Fmoc-Tyr(SO₃Na)-OH as a building block; (ii) the protected peptide-resin is treated with 90% aqueous TFA at 0 °C for an appropriate period of time for the cleavage and deprotection. Human cholecystokinin (CCK)-12, mini gastrin-II (14 residues), and little gastrin-II (17 residues) were synthesized with this method in 26-38% yields without any difficulties. This method was further applied to the stepwise synthesis of human big gastrin-II (34 residues), CCK-33 and -39. Despite the prolonged acid treatment (15-18 h at 0 °C), the ratios of the desulfated peptides were less than 15%, and the pure sulfated peptides were obtained in around 10% yields.

Full text of DOI:10.1021/jo000895y

VOLUME 66, NUMBER 1
J ANUARY 12, 2001
©
Copyright 2001 by the American Chemical Society
Articles
F a cile Solid -P h a se Syn th esis of Su lfa ted Tyr osin e-Con ta in in g
P ep tid es: Tota l Syn th esis of Hu m a n Big Ga str in -II a n d
Ch olecystok in in (CCK)-391
,2
,
†
†
†
,‡
§
Kouki Kitagawa,* Chikako Aida, Hidetoshi Fujiwara, Takeshi Yagami,* Shiroh Futaki,
|
|
|
Masashi Kogire, J un Ida, and Kazutomo Inoue
Niigata College of Pharmacy, Kamishin’ei-cho 5-13-2, Niigata 950-2081, J apan, National Institute of
Health Sciences, Setagaya-ku, Tokyo 158-8501, J apan, Institute for Chemical Research, Kyoto University,
Uji, Kyoto 611-0011, J apan, and Faculty of Medicine, Kyoto University, Sakyo-ku, Kyoto 606-8501, J apan
kouki@niigata-pharm.ac.jp
Received J une 12, 2000
Chemical synthesis of tyrosine O-sulfated peptides is still a laborious task for peptide chemists
because of the intrinsic acid-lability of the sulfate moiety. An efficient cleavage/deprotection
procedure without loss of the sulfate is the critical difficulty remaining to be solved for fluoren-9-
ylmethoxycarbonyl (Fmoc)-based solid-phase synthesis of sulfated peptides. To overcome the
difficulty, TFA-mediated solvolysis rates of a tyrosine O-sulfate [Tyr(SO
3
H)] residue and two
protecting groups, Bu for the hydroxyl group of Ser and 2,2,4,6,7-pentamethyldihydrobenzofuran-
-sulfonyl (Pbf) for the guanidino group of Arg, were examined in detail. The desulfation obeyed
t
5
-
1
-1
-1
first-order kinetics with a large entropy (59.6 J ‚K ‚mol ) and enthalpy (110.5 kJ ‚mol ) of
activation. These values substantiated that the desulfation rate of the rigidly solvated Tyr(SO H)
residue was strongly temperature-dependent. By contrast, the S 1-type deprotections were less
temperature-dependent and proceeded smoothly in TFA of a high ionizing power. Based on the
large rate difference between the desulfation and the S 1-type deprotections in cold TFA, an efficient
deprotection protocol for the sulfated peptides was developed. Our synthetic strategy for Tyr(SO H)-
3
N
N
3
containing peptides with this effective deprotection protocol is as follows: (i) a sulfated peptide
chain is directly constructed on 2-chlorotrityl resin with Fmoc-based solid-phase chemistry using
Fmoc-Tyr(SO Na)-OH as a building block; (ii) the protected peptide-resin is treated with 90%
3
aqueous TFA at 0 °C for an appropriate period of time for the cleavage and deprotection. Human
cholecystokinin (CCK)-12, mini gastrin-II (14 residues), and little gastrin-II (17 residues) were
synthesized with this method in 26-38% yields without any difficulties. This method was further
applied to the stepwise synthesis of human big gastrin-II (34 residues), CCK-33 and -39. Despite
the prolonged acid treatment (15-18 h at 0 °C), the ratios of the desulfated peptides were less
than 15%, and the pure sulfated peptides were obtained in around 10% yields.
In tr od u ction
tyrosine O-sulfated proteins have been identified from a
wide variety of natural sources.3 This modification is
catalyzed by tyrosylprotein sulfotransferase, a membrane-
bound enzyme within the trans-golgi compartment. The
Sulfation of a tyrosine residue is one of the ubiquitous
posttranslational modifications, and a vast number of
1
0.1021/jo000895y CCC: $20.00 © 2001 American Chemical Society
Published on Web 12/13/2000

2
J . Org. Chem., Vol. 66, No. 1, 2001
Kitagawa et al.
tyrosine sulfation is suggested to be involved in many
biological functions of secretory proteins. Tyrosine O-
peptide chain was constructed using a nonsulfated Tyr
derivative as a building block, it must be sulfated later
without affecting the other susceptible residues (post-
assembly sulfation). This approach requires special man-
ipulations to achieve the selective sulfation of a Tyr
residue. The amino groups of the N-terminus and Lys
residues, the alcoholic hydroxyl groups of Ser and Thr
residues, and the phenolic hydroxyl groups of Tyr resi-
dues that do not need to be sulfated often require
3
sulfate [Tyr(SO H)] residues are also present in a few
4
biologically active peptides such as gastrin-II, cholecys-
tokinin (CCK),5 caerulein,6 leucosulfakinins, leech-
7
8
derived anticoagulant hirudin, and a higher plant-
derived growth factor phytosulfokines.9 The negative
3
charge of a Tyr(SO H) residue would be important not
only for peptide folding but also for specific ligand-
receptor interaction.3f
protection depending on the sulfating reagent used. The
11
A prerequisite for investigating the biological functions
of peptides and proteins is that efficient methods for
obtaining a large amount of them are available; however,
applied protecting groups, in addition, should be removed
without deteriorating the sulfated Tyr residue. For
example, Sakakibara and co-workers1 employed an acid-
stable and base-labile phenoxyacetyl group for protection
of the Ser residue in nonsulfated porcine CCK-33 and
selectively sulfated the Tyr residue using pyridinium
acetyl sulfate1 in TFA. During this reaction, the amino
groups did not need to be protected. In a synthesis of
0c
chemical synthesis of a Tyr(SO
still a great challenge. The major difficulty lies in the
intrinsic acid-lability of a Tyr(SO H) residue. Although
3
H)-containing peptide is
3
1d
many synthetic studies on sulfated peptides have been
reported,10 a general and facile synthetic strategy with
wide applicability has not been established yet. If a
1
0d
human CCK-33 reported by Yajima and co-workers,
a
sequential protection and deprotection procedure was
employed: (i) complete deprotection of nonsulfated hu-
man CCK-33 constructed by a solution-phase method, (ii)
reprotection of the free R- and ꢀ-amino groups with Fmoc
group, (iii) preferential masking of the alcoholic hydroxyl
groups with tert-butyldiphenylsilyl (TBDPS) group, (iv)
*
To whom correspondence should be addressed. Fax: 81-25-268-
1
230.
†
Niigata College of Pharmacy.
National Institute of Health Sciences.
Institute for Chemical Research, Kyoto University.
Faculty of Medicine, Kyoto University.
‡
§
|
(1) Abbreviations used are as follows: All amino acids are of the
L-configuration. AcOH, acetic acid; AcONH
4
, ammonium acetate; Boc,
tert-butoxycarbonyl; Bu, tert-butyl; CCK, cholecystokinin; Clt resin,
-chlorotrityl chloride resin; DIPCDI, N,N-diisopropylcarbodiimide;
sulfation of the Tyr residue using a pyridine-SO
3
t
11f
complex, and (v) simultaneous deprotection of the Fmoc
2
1
2
and the TBDPS protecting groups using tetra-n-butyl-
EDT, 1,2-ethanedithiol; Fmoc, fluoren-9-ylmethoxycarbonyl; FT-IR,
Fourier transform infrared spectrometry; HFIP, hexafluoro-2-propanol;
HOBt, 1-hydroxybenzotriazole; HOOBt, 3,4-dihydro-3-hydroxy-4-oxo-
ammonium fluoride. We have also reported an orthogonal
10g,i
procedure for the selective sulfation of a Tyr residue.
1
,2,3-benzotriazine; LSIMS, liquid secondary-ion mass spectrometry;
The alcoholic hydroxyl groups and amino groups on a
peptide chain were masked with safety-catch-type pro-
tecting groups, and the Tyr residue was selectively
MALDI-TOFMS, matrix-assisted laser desorption/ionization time-of-
flight mass spectrometry; NMM, N-methylmorpholine; PAL, 5-(4′-
aminomethyl-3′,5′-dimethoxyphenyl)valeryl; Pbf, 2,2,4,6,7-pentame-
thyldihydrobenzofuran-5-sulfonyl; Pfp, pentafluorophenyl; PyBOP,
benzotriazolyloxytris(pyrrolidino)phosphon-
sulfated using a DMF-SO
3
complex.13 Several sulfated
ium hexafluorophosphate; TFE, 2,2,2-trifluoroethanol.
peptides were prepared with the postassembly sulfation
method; however, this approach is generally time-
consuming and reduction of reactivity is anticipated in
the sulfation of a large peptide.
(2) Preliminary reports: (a) Kitagawa, K.; Aida, C.; Fujiwara, H.;
Yagami, T.; Futaki, S. Tetrahedron Lett. 1997, 38, 599-602. (b)
Kitagawa, K.; Aida, C.; Fujiwara, H.; Yagami, T.; Futaki, S. In
Peptides-Frontiers of Peptide Science (Proceedings of the 15th American
Peptide Symposium ); Tam, J . P., Kaumaya, P. T. P., Eds.; Kluwer/
Escom: Dordrecht, The Netherlands, 1999; pp 295-296. (c) Kitagawa,
K.; Aida, C.; Fujiwara, H.; Yagami, T.; Futaki, S.; Kogire, M.; Ida, J .;
Inoue. K. In Peptide Science-Present and Future (Proceedings of the
On the other hand, recent advances in the Fmoc-based
solid-phase peptide synthesis (Fmoc-SPPS)14 enabled us
1
st International Peptide Symposium); Shimonishi, Y., Ed.; Kluwer
Academic Publishers: Dordrecht, The Netherlands, 1999; pp 525-526.
3) General reviews on tyrosine sulfation in proteins: (a) Huttner,
(10) Solution-phase: (a) Moroder, L.; W u¨ nsch, E. In Natural Product
Chemistry; Attar-ur-Rahman Ed.; Springer-Verlag: Berlin, 1986; vol.
26, pp 255-280 and references cited therein. (b) Ondetti, M. A.;
Pluscec, J .; Sabo, E. F.; Sheehan, J . T.; Williams, N. J . Am. Chem.
Soc. 1970, 92, 195-199. (c) Kurano, Y.; Kimura, T.; Sakakibara, S. J .
Chem. Soc., Chem. Commun. 1987, 323-325. (d) Fujii, N.; Futaki, S.;
Funakoshi, S.; Akaji, K.; Morimoto, H.; Doi, R.; Inoue, K.; Kogire, M.;
Sumi, S.; Yun, M.; Tobe, T.; Aono, M.; Matsuda, M.; Narusawa, H.;
Moriga, M.; Yajima, H. Chem. Pharm. Bull. 1988, 36, 3281-3291.
Solid-phase: (e) Penke, B.; Rivier, J . J . Org. Chem. 1982, 52, 1197-
1200. (f) Penke, B.; Ngyerges, L. Pept. Res. 1991, 4, 289-295. (g)
Futaki, S.; Taike, T.; Akita, T.; Kitagawa, K. Tetrahedron 1992, 48,
8899-8914. (h) Yagami, T.; Shiwa, S.; Futaki, S.; Kitagawa, K. Chem.
Pharm. Bull. 1993, 41, 376-380. (i) Kitagawa, K.; Futaki, S.; Yagami,
T.; Sumi, S.; Inoue, K. Int. J . Pept. Protein Res. 1994, 43, 190-200. (j)
Han, Y.; Bontems, S. L.; Hegyes, P.; Munson, M. C.; Minor, C. A.;
Kates, S.; Albericio, F.; Barany, G. J . Org, Chem. 1996, 61, 6326-
6339.
(
W. B. Nature 1982, 299, 273-276. (b) Huttner, W. B. Methods Enzymol.
1
3
984, 107, 200-223. (c) Huttner, W. B. Annu. Rev. Physiol. 1988, 50,
63-376. (d) Huttner, W. B.; Baeuerle, P. A. In Modern Cell Biology;
Satir, B., Ed.; Alan R. Liss Inc.: New York, 1988; Vol. 6, pp 97- 140.
(
1
7
e) Niehrs, C.; Beiâwanger, R.; Huttner, W. B. Chem. Biol. Interact.
994, 92, 257-271. (f) Kehoe, J . W.; Bertozzi, C. R. Chem. Biol. 2000,
, R57-R61.
(4) (a) Gregory, H.; Hardy, P. M.; J ones, D. S.; Kenner, G. W.;
Sheppard, R. C. Nature 1964, 204, 931-933. (b) Bentley, P. H.; Kenner,
G. W.; Sheppard, R. C. Nature 1966, 209, 583-585.
(
5) (a) Mutt, V.; J orpes, J . E. Eur. J . Biochem. 1968, 6, 156-162.
(
b) Mutt, V.; J orpes, J . E. Biochem. J . 1971, 125, 57P-58P. (c) Rehfeld,
J . F. J . Biol. Chem. 1978, 253, 4022-4030. (d) Tatemoto, K.; J o¨ rnvall,
H.; Siimesmaa, S.; Halld e´ n, G.; Mutt, V. FEBS Lett. 1984, 174, 289-
2
93. (e) Reeve, J . R., J r.; Eysselein, V.; Walsh, J . H.; Ben-Avram, B.
M.; Shively, J . E. J . Biol. Chem. 1986, 261, 16392-16397. (f) Eysselein,
V. E.; Eberlein, G. A.; Schaeffer, M.; Grandtt, D.; Goebell, H.; Niebel,
W.; Rosenquist, G. L.; Meyer, H. E.; Reeve, J . R., J r. Am. J . Physiol.
(11) (a) Reitz, H. C.; Ferrel, R. E.; Frankel-Conrat, H.; Olcott, H. S.
J . Am. Chem. Soc. 1946, 68, 1024-1031. (b) Reitz, H. C.; Ferrel, R.
E.; Olcott, H. S.; Frankel-Conrat, H. J . Am. Chem. Soc. 1946, 68, 1031-
1035. (c) Gilbert, E. E. Chem. Rev. 1962, 62, 549-589. (d) Penke, B.;
Zar a´ ndi, M.; Kov a´ cs, K.; Rivier, J . In Peptides 1984 (Proceedings of
the 18th European Peptide Symposium); Ragnarsson, U. Ed.; Almqvist
and Wiksell, Int.: Stockholm, 1985; pp 279-283. (e) Borin, G.;
Calderan, A.; Ruzza, P.; Moroder, L.; Knaup, G.; G o¨ hring, W.; W u¨ nsch,
E. In Peptide Chemistry 1987; Shiba, T., Sakakibara, S., Eds.; Protein
Research Foundation: Osaka, 1988; pp 179-182. (f) Sisler, S. S.;
Audrieth, L. A. Inorg. Synth. 1946, 2, 173.
1
990, G253-G260.
6) Anastasi, A.; Erspamer, V.; Endean, R. Arch. Biochem. Biophys.
968, 125, 57-68.
7) (a) Nachman, R. J .; Holman, G. M.; Cook, B. J .; Haddon, W. F.;
(
1
(
Ling, N. Biochem. Biophys. Res. Commun. 1986, 140, 357-364. (b)
Nachman, R. J .; Holman, G. M.; Haddon, W. F.; Ling, N. Science 1986,
2
34, 71-73.
8) (a) Dodt, J .; Muller, H.-P.; Seemuller, U.; Chang, J . Y. FEBS
(
Lett. 1984, 165, 180-183. (b) Dodt, J .; Seemuller, U.; Maschler, R.;
(12) Ueki, M.; Amemiya, M. Tetrahedron Lett. 1987, 28, 6617-6620.
(13) Futaki, S.; Taike, T.; Yagami, T.; Ogawa, T.; Akita, T.; Kita-
gawa, K. J . Chem. Soc., Perkin Trans. 1 1990, 1739-1744.
(14) Atherton, E.; Sheppard, R. C. In Solid-Phase Peptide Synthe-
sis: A Practical Approach; Oxford University Press: Oxford, 1989.
Fritz, H. Biol. Chem. Hoppe-Seyler 1985, 336, 379-385.
(9) (a) Matsubayashi, Y.; Hanai, H.; Hara, O.; Sakagami, Y. Bio-
chem. Biophys. Res. Commun. 1996, 225, 209-214. (b) Matsubayashi,
Y.; Sakagami, Y. Proc. Natl. Acad. Sci., U.S.A. 1996, 93, 7623-7627.

Sulfated Tyrosine-Containing Peptides
J . Org. Chem., Vol. 66, No. 1, 2001
3
F igu r e 1. Effect of reaction temperature on the desulfation and the deprotections in neat TFA: (a) desulfation of Tyr(SO
3
Na),
t
(b) deprotection of Ser( Bu), and (c) deprotection of Arg(Pbf).
to directly construct a sulfated peptide chain using Fmoc-
Na)-OH1
0e,h
as a building block (preassembly
Tyr(SO
3
sulfation). Here the base-labile Fmoc group is employed
R
for N -protection, and thus repeated acid treatment
3
incompatible with a Tyr(SO H) residue is avoidable
during the peptide chain construction. This approach
seems attractive because of the simplicity and applicabil-
ity to an automated peptide synthesizer. However, there
is a crucial difficulty remaining to be solved in the
synthesis with this approach. The effective final cleavage/
deprotection conditions applied to the constructed pep-
3
tide-resin are generally destructive to a Tyr(SO H)
residue. This becomes a serious problem especially in the
case of peptides having a C-terminal amide structure,
such as gastrin and CCK, because of the lack of a highly
acid-sensitive amide-linker. Therefore, application of the
preassembly sulfation has been limited to the synthesis
3
F igu r e 2. Arrhenius plots for the desulfation rate of Tyr(SO -
Na) and the deprotection rates of Ser( Bu) and Arg(Pbf). All
the reactions were carried out in neat TFA.
t
ate the usefulness of this acid to the cleavage/deprotec-
of relatively small and less-complicated Tyr(SO
containing peptides.
3
H)-
3
tion, TFA-mediated acidolysis rates of Tyr(SO Na) and
t
two protecting groups, Bu for the hydroxyl group of Ser
In the course of our efforts to establish a general
synthetic method for sulfated peptides, we noticed that
TFA, a routinely used reagent for the final cleavage/
deprotection step in the Fmoc-SPPS, was not a destruc-
and 2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl
15
(Pbf) for the guanidino group of Arg, were examined.
These protecting groups were selected because of their
noticeably slow deprotection rates in TFA. The Fmoc-
tive acid to a Tyr(SO
3
H) residue as long as the reaction
t
derivatives, Fmoc-Tyr(SO
3
Na)-OH, Fmoc-Ser( Bu)-OH,
mixture was kept cold.1 Provided that there is a certain
TFA-mediated acidolysis system that is sufficient for the
final cleavage/deprotection but not destructive to a Tyr-
0h
and Fmoc-Arg(Pbf)-OH, were stirred in neat TFA at a
specified temperature (0, 18, or 30 °C), and part of each
solution was periodically withdrawn to determine the
degree of the desulfation or the deprotections by RP-
HPLC. Disappearance of every starting material obeyed
good first-order kinetics (Figure 1). From these results,
the Arrhenius plot (Figure 2) and activation parameters
3
(SO H) residue, sulfated peptides are expected to be
easily prepared with the Fmoc-SPPS. In this article, we
describe such an efficient acidolysis system for Tyr-
3
(SO H)-containing peptides. By combining this deprotec-
tion system with highly acid-sensitive 2-chlorotrityl resin
as a solid support, we established a general synthetic
strategy for sulfated peptides and accomplished the total
synthesis of human big gastrin-II and CCK-39.
(
Table 1) of each reaction were obtained. The striking
feature of the activation parameters determined is that
the desulfation was accompanied by a positive entropy
term (∆S ) +59.6 J ‚K ‚mol ) and the largest enthalpy
term (∆H ) +110.5 kJ ‚mol ). These values were well
reflected by the strong temperature-dependency of the
desulfation rate compared to the rates of the deprotec-
‡
-1
-1
‡
-1
Resu lts a n d Discu ssion
Desu lfa tion vs Dep r otection : Kin etic Stu d ies. We
previously examined the compatibility of several cleav-
age/deprotection systems with a Tyr(SO H)-containing
3
t
tions. The deprotections of Arg(Pbf) and Ser( Bu) were
peptide and found that the desulfation in TFA was
(
15) Carpino, L. A.; Shroff, H.; Triolo, S. A.; Mansour, E.-S. M. E.;
significantly temperature-dependent.1 To further evalu-
0h
Wenschuh, H.; Albericio, F. Tetrahedron Lett. 1993, 34, 6661-6664.

4
J . Org. Chem., Vol. 66, No. 1, 2001
Kitagawa et al.
Ta ble 1. Ra tes a n d Activa tion P a r a m eter s for th e Desu lfa tion a n d th e Dep r otection s in Nea t TF A a t 0 °C
rates and activation parametersa
reaction rate (k
, s-1)
Tyr(SO
Na) f Tyr
Arg(Pbf) f Arg
Ser( Bu) f Ser
t
3
5.30 × 10-6
4.43 × 10-4
9.67 × 10-4
0
-1
activation energy (E
a
, kJ ‚mol
)
112.8
69.1
89.0
‡
-1
enthalpy of activation (∆H , kJ ‚mol
)
110.5
94.3
59.6
14.2
66.8
86.7
‡
-1
free energy of activation(∆G , kJ ‚mol
)
84.2
82.5
entropy of activation(∆S , J ‚K ‚mol-1)
‡
-1
-63.7
-15.2
15.5
‡
entropy of activation(∆S , eu)
3.7
a
Calculated from the equations ∆H ) Ea - RT; ∆S‡ ) (∆H‡ - ∆G )/T; ∆G ) -RT ln(k0h/ κT) at 0 °C. R ) 8.314 J ‚K-1‚mol-1; T )
‡
‡
‡
-
34
-23
-1
-1
-1
2
73.15 K; h ) 6.626 × 10
J ‚s; κ ) 1.380 × 10
J ‚K ; 1 eu ) 4.184 J ‚K ‚mol .
F igu r e 3. Plausible mechanism for proton-catalyzed desulfa-
tion of arylmonosulfates in acidic solution.
1
2 and 56 times faster than the desulfation at 30 °C, but
F igu r e 4. Desulfation and deprotection rates in aqueous TFA
at 0 °C.
these differences were enhanced to 84 and 182 times at
0
°C, respectively (Figure 2).
The feature of the desulfation rate in TFA can be
deprotection reagent for a Tyr(SO
because the high ionizing power of TFA promotes S
type deprotections. Moreover, the rates of the S 1-type
deprotections were less temperature-dependent (Figure
). Thus, by using the marked rate difference between
the desulfation and the S 1-type deprotections in cold
TFA (Figure 2), a large number of protecting groups can
3
H)-containing peptide
interpreted by a mechanism proposed for the deteriora-
H) in acidic solution16
N
1-
tion of arylmonosulfates (ArO-SO
3
N
(
Figure 3). The inherent acid-lability of Tyr(SO
3
H), or the
acid-lability of ArO-SO
3
H, is attributable to the domi-
nant existence of the unstable proton-donating form
2
+
-
N
[
ArO (H)-SO
3
] in acidic solution, which rapidly decom-
. This amphoteric ion is pre-
poses to Ar-OH and SO
3
be removed with minimum damage of a Tyr(SO
residue.
3
H)
dicted to be highly solvated and stabilized by molecules
with a high dipole moment such as TFA and 2,2,2-
trifluoroethanol (TFE). The large entropy of activation
of the desulfation in TFA substantiates the rigidly
solvated amphoteric ion and less-solvated transition
state. Therefore, a higher energy barrier should be passed
over for the desulfation in polar solvents. Acidolysis rates
of arylmonosulfates are actually reported to be affected
In practice, we used aqueous TFA for the deprotection
of a Tyr(SO H)-containing peptide to diminish side reac-
3
tions caused by the released cations. Water is known to
be a good cation scavenger in TFA.14 We examined in
advance the effects of water addition on the rates of the
desulfation and the deprotections. It was found that the
deprotections were gradually retarded with increasing
water content in the TFA. By contrast, the desulfation
rate was not affected significantly (Figure 4). In 90%
aqueous TFA at 0 °C, the deprotection of the Pbf group
was still 40 times faster than the desulfation. From these
results, we concluded that 90-95% aqueous TFA at low
temperature could maintain the large rate difference
1
1d,16,17
by the polarity of the solutions as well as the acidity.
If the polarity is low, arylmonosulfates are expected to
rapidly decompose even under weakly acidic conditions.
On the contrary, they would be stable even in strongly
acidic solutions when the polarity is unusually high.
Because of their high polarity, TFA and TFE are sup-
posed to be favorable solvents for the cleavage/deprotec-
tion of sulfated peptides.
N
between the desulfation and the S 1-type deprotections
and can be effectively applied to the deprotection of a
sulfated peptide.
In an acidolysis system for protected peptides, soft
bases are generally dissolved to scavenge the released
cations and to promote the deprotections as S
N
2-type
Esta blish m en t of a F a cile Solid -P h a se Meth od for
th e Syn th esis of Su lfa ted P ep tid es. Previously, we
reactions.18 However, we previously reported that the
10h
addition of a sulfur-containing soft base such as thioani-
prepared CCK-12 with the Fmoc-SPPS using PAL-resin
as a solid support [PAL ) peptide amide linker; 5-(4-(9-
fluorenylmethoxycarbonyl)aminomethyl)-3,5-dimethoxy-
phenoxyvaleric acid linker ]. Despite the use of this
reportedly acid-sensitive linker-resin, the detachment
efficiency of the peptide from the support was no more
than 40% under the cleavage/deprotection conditions
employed. To improve the detachment efficiency, we
sole and ethanedithiol (EDT) significantly accelerated the
desulfation of a Tyr(SO
3
H)-containing peptide.1 The
0h
1
9
deprotections, therefore, have to be promoted without the
addition of sulfur-containing soft bases. It is considered
that the unimolecular deprotection (S
is more advantageous to a Tyr(SO H)-containing peptide
than the bimolecular deprotection (S 2-type reaction)
requiring soft bases. TFA seems again to be a promising
N
1-type reaction)
3
N
(
18) (a) Yajima, H.; Fujii, N. In The Peptides-Analysis, Synthesis,
(
16) Kice, J . L.; Anderson, J . M. J . Am. Chem. Soc. 1966, 88, 5242-
245.
17) Davis, J . M.; Cameron, D. R.; Kubanek, J . M.; Mizuyabu, L.;
Thatcher, R. J . Tetrahedron Lett. 1991, 32, 2205-2206.
Biology; Gross, E., Meienhofer, J ., Eds.; Academic Press: New York,
1983; vol. 5, pp 65-109. (b) Tam, J . P.; Merrifield, R. B. In The
Peptides-Analysis, Synthesis, Biology; Udenfriend, S., Meienhofer, J .,
Eds.; Academic Press: New York, 1987; vol. 9, pp 185-248.
5
(

Sulfated Tyrosine-Containing Peptides
J . Org. Chem., Vol. 66, No. 1, 2001
5
3
F igu r e 5. Synthetic scheme of a Tyr(SO H)-containing pep-
tide with the Fmoc-based solid-phase approach.
decided to use 2-chlorotrityl (Clt) resin20 as a solid
support. Clt resin was developed to provide peptide
segments bearing Bu/Boc side-chain protecting groups,
F igu r e 6. Sequences of human gastrin-II and CCK peptides.
The N-terminal Gln residue of little gastrin-II is posttransla-
tionally cyclized to a Pyr () pyroglutamic acid) residue.
t
and quantitative detachment of a protected peptide from
this linker-resin can be achieved under extremely mild
acidic conditions.2
mediated coupling protocol [PyBOP ) benzotriazolylox-
ytris(pyrrolidino)phosphonium hexafluorophosphate] was
used for the peptide chain elongation.24 For protection
of the Arg residue in CCK peptides, we selected Pbf
group15 that was most sensitive to our deprotection
conditions (90% aqueous TFA, 0 °C) among the various
protecting groups examined. The assembled peptide-resin
corresponding to 2 was subjected to the two-step cleav-
0,21
3
Our strategy for the Fmoc-SPPS of a Tyr(SO H)-
containing peptide is shown in Figure 5. This approach
involves two key features to complete the synthesis in
an efficient manner: (1) utilization of Clt resin as a solid
support to detach a sulfated peptide quantitatively and
(
2) deprotection with 90% aqueous TFA at 0 °C to
minimize the deterioration of a Tyr(SO H) residue. The
effectiveness of this approach was evaluated through the
synthesis of leucine-enkephalin sulfate22 [Tyr(SO
H)-Gly-
Gly-Phe-Leu (1)], human CCK-12 (2), mini gastrin-II (3)
14 residues), and little gastrin-II (4) (17 residues).23 The
2 2
age/deprotection; first with a mixture of HFIP/CH Cl (1:4
3
v/v) for 30 min at 25 °C and then with 90% aqueous TFA
for 8 h at 0 °C. Detachment of the protected peptide from
the Clt resin was quantitative (>95%), and deterioration
of the sulfate during the prolonged deprotection was not
significant (ca. 10% of the peptide was desulfated, Figure
7a). Subsequent purification by RP-HPLC afforded highly
homogeneous 2 in 26% yield from the protected peptide-
resin. Similarly, 3 and 4 were obtained from the corre-
sponding peptide-resin in 34 and 38% yields, respectively
(Figure 7b). Deprotection of gastrin-II peptides was
continued for 5 h at 0 °C because they contain no Arg-
(Pbf) residues. The determined amino acid composition
of each peptide coincided with the theoretical value, and
integrity of the sulfate was ascertained by FT-IR and
liquid secondary-ion mass spectrometry (LSIMS). Espe-
cially, the negative-ion LSIMS spectra gave direct evi-
3
(
amino acid sequences of the CCK peptides and the
gastrin-II peptides are shown in Figure 6.
To overcome the general low recovery of a peptide
amide from the amide-offering linker-resin, the C-termi-
2
0c
nal dipeptide, Fmoc-Asp-Phe-NH
2
,
was linked to Clt
resin through the â-carboxyl group of the Asp residue.
The resulting dipeptide-resin was treated with 20%
R
piperidine/DMF to remove the N -Fmoc group; then each
building block including Fmoc-Tyr(SO
3
Na)-OH was suc-
cessively introduced into the peptide chain. The PyBOP-
(
19) (a) Albericio, F.; Barany, G. Int. J . Pept. Protein Res. 1987, 30,
06-216. (b) Albericio, F.; Kneib-Cordonier, N.; Biancalana, S.; Gera,
L.; Masada, R. I.; Hudson, D.; Barany, G. J . Org. Chem. 1990, 55,
730-3743.
20) (a) Barlos, K.; Gatos, D.; Kallitsis, J .; Papaphotiu, G.; Sotiriu,
P.; Wenqing, Y.; Schafer, W. Tetrahedon Lett. 1989, 30, 3943-3946.
b) Barlos, K.; Gatos, D.; Kapolos, S.; Papaphotiu, G.; Schafer, W.
2
dence of the intactness of the sulfate as previously
reported.2
5,26
The yield of 2 obtained with this method
3
(
(26%) was significantly improved compared to the previ-
10h
ous synthesis
using PAL-resin (7.4% yield from the
(
protected peptide-resin). These results clearly demon-
strated the potential of our novel strategy for the
synthesis of a Tyr(SO H)-containing peptide.
3
Tetrahedron Lett. 1989, 30, 3947-3950. (c) Barlos, K.; Chatzi, O.;
Kapolos, S.; Poulos, C.; Schafer, W.; Wenqing, Y. Int. J . Pept. Protein
Res. 1991, 38, 555-561.
(21) Bollhagen, R.; Schmiedberger, M.; Barlos, K.; Grell, E. J . Chem.
Soc., Chem. Commun. 1994, 2559-2560.
22) Unsworth, C. D.; Hughes, J .; Morley, J . S. Nature 1982, 295,
19-522.
23) At an early stage of this investigation, we used a two-step
cleavage/deprotection protocol for synthesis of the sulfated peptides.
Thus we first examined the extent of desulfation during the cleavage
(
(24) Coste, J .; Le-Nguyen, D.; Castro, B. Tetrahedron Lett. 1990,
31, 205-208.
5
(
(25) Yagami, T.; Kitagawa, K.; Futaki, S. Rapid Commun. Mass
Spectrom. 1995, 9, 1335-1338.
(26) (a) Arlandini, E.; Gioia, B.; Perseo, G.; Vigevani, A. Int. J . Pept.
Protein Res. 1984, 24, 386-391. (b) Gibson, B. W.; Falick, A. M.;
Burlingame, A. L.; Nadasdi, L.; Nguyen, A. C.; Kenyon, G. L. J . Am.
Chem. Soc. 1987, 109, 5343-5348. (c) Gibson, B. W. In Biological Mass
Spectrometry; Burlingame, A. L., McCloskey, J . A., Eds.; Elsevier
Science Publishers B. V.: Amsterdam, 1990; pp 315-336. (d) Carr, S.
A.; Hemling, M. E.; Bean, M. F.; Roberts, G. D. Anal. Chem. 1991, 63,
2802-2824. (e) Gibson, B. W.; Cohem, P. Method Enzymol. 1990, 193,
480-501. (f) Talbo, G.; Roepstorff, P. Rapid Commun. Mass Spectrom.
1992, 7, 201-204.
with
propanol (HFIP)/CH
peptide. These experiments showed that about 1.8% of the peptide was
desulfated during the cleavage with AcOH/TFE/CH Cl (25 °C, 2 h)
and that no desulfated peptide was generated during the treatment
a
mixture of AcOH/TFE/CH
2 2
Cl (1:1:3 v/v) or hexafluoro-2-
R
2
Cl (1:4 v/v) using N -Fmoc protected 1 as a model
2
2
2
R
with HFIP/CH
Fmoc protected peptide-resin was treated with 20% piperidine/DMF
then with HFIP/CH Cl (0.5 h). Pure 1 was obtained in 58% yield from
the peptide-resin after RP-HPLC purification.
2 2
Cl (25 °C, 6 h). For the characterization of 1, the N -
2
2

6
J . Org. Chem., Vol. 66, No. 1, 2001
Kitagawa et al.
F igu r e 8. HPLC chromatograms of (a) crude 5 and (b)
purified 5. An asterisk in (a) shows the desulfated peptide (big
gastrin-I) produced during the deprotection. [HPLC condi-
tions: column, Cosmosil 5C -AR (4.6 × 150 mm); elution
F igu r e 7. HPLC chromatograms of crude (a) CCK-12 2 and
b) little gastrin-II 4. An asterisk in each chromatogram shows
the desulfated peptide produced during the deprotection step.
HPLC conditions: column, Cosmosil 5C18-AR (4.6 × 150 mm);
elution system, a linear gradient of CH CN (B) in 0.1 M
AcONH (A); flow rate, 1 mL/min].
(
1
8
system, a linear gradient of CH CN (B) in 0.1 M AcONH (A);
3
4
[
flow rate,1 mL/min].
3
4
peptide chain using Fmoc-His(Boc)-OH14 as a building
block. The assembled peptide-resin was subjected to the
cleavage/deprotection with 90% aqueous TFA. Consistent
with the results of our kinetic studies, deprotection with
Syn th esis of Hu m a n Big Ga str in -II. Gastrins are
a family of peptide hormones that stimulate gastric acid
4
secretion. They contain many troublesome residues for
9
0% aqueous TFA at 0 °C for 8 to 9 h was satisfactory
chemical synthesis (Figure 6). Several synthetic achieve-
ments of gastrin-I peptides, the nonsulfated counterparts
for completing the removal of the protecting groups.
Despite this prolonged acid treatment, loss of the sulfate
was about 15% (Figure 8a); on the other hand, even brief
treatment with 90% aqueous TFA at room-temperature
resulted in a drastic production of the desulfated peptide
of gastrin-II, were reported with solution- or solid-phase
methods.1
0a,27
It has been recognized that the homoge-
neous products are difficult to obtain with a conventional
solid-phase approach due to several critical side reactions.
Sheppard and co-workers pointed out that the Trp
residues in gastrin peptides tended to be alkylated by
(big gastrin-I, 6). After purification of the crude peptide
by RP-HPLC, highly homogeneous 5 (Figure 8b) was
obtained in 13% yield from the protected peptide-resin.
The amino acid composition of the acid hydrolysate
coincided with the theoretical value. In the positive- and
negative-ion LSIMS spectra, this product exhibited the
t
Bu cations generated from the protecting groups during
the acidolysis.27c,d Nucleophilic addition of the Trp residue
to the polymer support also reduces the yield of the
desired peptide.28 Various scavenger cocktails, such as
+
pronounced peaks corresponding to [M + H] and [M -
2
9a
TFA-phenol-thioanisole-EDT-H
2
O
and TFA-thio-
-
30
H] , respectively. Moreover, amino acid analyses of two
fragments (N-16 and C-17 in Figure 9) generated by a
lysyl endopeptidase digestion supported the structure of
this final product.
anisole-EDT-anisole,2 have been effectively used to
reduce such side reactions during the cleavage/deprotec-
tion. However, these cocktails including sulfur-containing
soft bases are not applicable to the deprotection of a
9b
Syn th esis of Big-Molecu la r -F or m CCK P ep tid es,
CCK-33 a n d -39. Cholecystokinin (CCK) is a group of
peptide hormones that is distributed both in the brain
and gastrointestinal tract and participates in the central
1
0h
sulfated peptide as already discussed. We attempted
the synthesis of human big gastrin-II (5, Figure 6) to
demonstrate the usefulness of our new strategy.
The C-terminal dipeptide was attached to Clt resin,
and it was elongated to the 34-residue peptide in a
stepwise manner. A His residue was introduced into the
5
and pancreatic regulations. This peptide is also known
to exist in several molecular forms (Figure 6), such as
5a,b
5e
5f
CCK-8, -22, -33,
-39, and -58, but the significance
of this molecular diversity and the physiological roles of
each form remain to be investigated. Every form has a
Tyr(SO H) residue at the seventh position from the
3
C-terminus that is crucial for the biological activities.
We further demonstrated the effectiveness of our
strategy through the synthesis of CCK peptides, CCK-
(
27) (a) Beacharm, J .; Bentley, P. H.; Gregory, R. A.; Kenner, G.
W.; MacLeod, J . K.; Sheppard, R. C. Nature 1966, 209, 585-586. (b)
J aeger, E.; Thamm, P.; Schmidt, I.; Knof, S.; Moroder, L.; W u¨ nsch, E.
Hoppe-Seyler’s Z. Physiol. Chem. 1978, 359, 155-164. (c) Brawn, E.;
Sheppard, R. C.; Williams, B. J . J . Chem. Soc., Perkin Trans. 1 1983,
7
5-82. (d) Brawn, E.; Sheppard, R. C.; Williams, B. J . J . Chem. Soc.,
Perkin Trans. 1 1983, 1161-1167. (e) Tam, J . P.; Merrifield, R. B. Int.
J . Pept. Protein Res. 1985, 26, 262-273. (f) Kneib-Cordonier, N.;
Albericio, F.; Barany, G. Int. J . Pept. Protein Res. 1990, 35, 527-538.
2
2 (7), -33 (8), and -39 (9). Their synthesis, however, was
(28) Atherton, E.; Cameron, L. R.; Sheppard, R. C. Tetrahedron
1
988, 44, 843-857.
29) (a) King, D. S.; Fields, C. G.; Fields, G. B. Int. J . Pept. Protein
Res. 1990, 36, 255-266. (b) Sol e´ , N. A.; Barany, G. J . Org. Chem. 1992,
7, 5399-5403.
(30) Although the desulfated [M + H - SO
as a prominent peak on the positive-ion LSIMS spectra of a Tyr(SO
3
]+ is usually detected
H)-
containing peptide, the [M + H] peak becomes dominant over the [M
(
3
+
+
3
+ H - SO ] peak as the peptide chain is elongated.
31
5

Sulfated Tyrosine-Containing Peptides
J . Org. Chem., Vol. 66, No. 1, 2001
7
(Figure 11). These fragments were identified as YIQQARK
(10), APSGRMSIVK (11), and CCK-22 (7). All the ana-
lytical data supported the structural correctness of the
final products. The insulinotropic activity of the CCK
peptides was examined using the isolated pancreatic
islets (male Wister rats) pretreated with glucose (11.1
mM). Synthetic human CCK-39 increased the insulin
release to the same extent as CCK-8 and -33 at doses of
-
8
-9
1
0
and 10 M (Table 2). In this way, CCK peptides
were synthesized with our new approach without any
difficulties.
Con clu sion s
A facile and efficient strategy for the Fmoc-SPPS of a
Tyr(SO H)-containing peptide was established and suc-
3
cessfully applied to the synthesis of various sizes of
human gastrin-II and CCK peptides. The vital points of
our strategy are the use of an extremely acid-sensitive
Clt resin as a solid support and the S
tions promoted in TFA having a high ionizing power. To
minimize the deterioration of a Tyr(SO H) residue, the
N
1-type deprotec-
F igu r e 9. HPLC chromatogram of a lysyl endopeptidase
digest of 5. [HPLC conditions: column, Cosmosil 5C18-AR (4.6
3
final deprotection with TFA must be carried out at low
temperature without addition of sulfur-containing soft
bases. To our knowledge, the synthesis of big gastrin-II
and CCK-39 has not been achieved so far. Our novel
strategy has general applicability to sulfated peptides
regardless of their sizes. The establishment of a facile
synthetic protocol would also be helpful for biological
studies of sulfated peptides and proteins.
×
150 mm); elution system, a linear gradient of CH
.1% TFA (A); flow rate, 1 mL/min].
3
CN (B) in
0
supposed much more difficult than that of big gastrin-II
because they contain multiple Arg residues and oxida-
tion-susceptible Met residues in addition to an alkylation-
sensitive Trp residue (Figure 6). To minimize the number
of protecting groups, Asn and Gln residues were intro-
duced without masking the side-chain. For this purpose,
Fmoc-Asn/Gln-OPfp32 (Pfp ) pentafluorophenyl) were
Exp er im en ta l Section
used as acylation reagents with the help of 3,4-dihydro-
Gen er a l P r oced u r es. Fmoc-amino acid derivatives, Py-
BOP reagent, and Clt resin (substituted level; 1.47 mmol/g
resin) were purchased from Watanabe Chemical Co., Ltd.
3
3
3
-hydroxy-4-oxo-1,2,3-benzotriazine (HOOBt). Each con-
structed peptide-resin was subjected to the cleavage/
deprotection with 90% aqueous TFA at 0 °C. Due to the
presence of multiple Pbf groups in the protected peptides,
the deprotection had to be prolonged (10 h for 7 and 15
to 18 h for 8 and 9). Nevertheless, deterioration of the
(
Hiroshima, J apan). Fmoc-His(Boc)-OH was obtained from
Calbiochem-Novabiochem J apan, Ltd. (Tokyo, J apan). Lysyl
endopeptidase from Achromobacter lyticus M497-1 (EC 3.4.21.50)
was purchased from Wako Pure Chemicals Co., Ltd. (Osaka,
J apan). Other chemicals were of analytical grade and were
used without further purification. Acid hydrolysis was carried
out at 110 °C for 24 h with a mixture of propionic acid and 12
N HCl (1:1 v/v) for the resin-bound peptides, or with 6 N HCl
containing a few drops of phenol for the purified peptides, and
the hydrolysate was subjected to an amino acid analysis. For
LSIMS analysis, glycerol, thioglycerol, and m-nitrobenzyl
alcohol were used as a matrix. Sinapinic acid was used as a
matrix for MALDI-TOFMS analysis.
3
4
3
Tyr(SO H) residues was not significant (Figure 10a,c).
After one-step HPLC purification, highly homogeneous
CCK peptides were obtained (Figure 10b,d). The yields
from the protected peptide-resin were not excellent (9.7%
for 8 and 7.8% for 9), but these figures would be
satisfactory for a total solid-phase synthesis of a sulfated
peptide of 35 to 40 residues in length. The amino acid
compositions of their acid hydrolysates coincided with the
theoretical values, and intactness of the sulfates was
directly evidenced on LSIMS or matrix assisted laser
desorption/ionization time-of-flight mass spectrometry
Deter m in a tion of th e Ra te Con sta n ts a n d Kin etic
P a r a m eter s. Fmoc-Tyr(SO Na)-OH (4.28 mg, 8.47 µmol),
3
t
Fmoc-Ser( Bu)-OH (3.25 mg, 8.47 µmol), and Fmoc-Arg(Pbf)-
OH (5.50 mg, 8.47 µmol) were separately dissolved in TFA (1.0
mL) and stirred at a specified temperature (0, 18, or 30 °C).
At appropriate intervals, 5.0 µL of each solution was with-
(MALDI-TOFMS). Lysyl endopeptidase digestion pro-
duced two and three fragments from 8 and 9, respectively
drawn and diluted with a mixture of CH
3
CN and phosphate
buffer (0.2 M, pH 7.0) (3:7 v/v, 1 mL). Part of this solution
(
31) (a) Yagami, T.; Kitagawa, K.; Aida, C.; Fujiwara, H.; Futaki,
(
200 µL) was analyzed by RP-HPLC [column, Wakosil-II 5C18
4.6 × 150 mm); elution system, a linear gradient of CH CN
(30 to 70% in 20 min); flow rate, 1 mL/min;
absorbance was detected at 300 nm]. Under these chromato-
graphic conditions, the t of the starting compounds and the
reaction products were as follows: 4.8 and 7.8 min for Fmoc-
Tyr(SO Na)-OH and Fmoc-Tyr-OH; 10.8 and 6.3 min for Fmoc-
S. In Peptide Science-Present and Future (Proceedings of the 1st
International Peptide Symposium); Shimonishi, Y., Ed.; Kluwer Aca-
demic Publishers: Dordrecht, The Netherlands, 1999; pp 370-372. (b)
Yagami, T.; Kitagawa, K.; Aida, C.; Fujiwara, H.; Futaki, S. J . Pept.
Res. 2000, 56, 239-249.
(
3
in 0.1 M AcONH
4
R
(
(
32) Kisfauldy, L.; Sch o¨ n, I. Synthesis 1983, 325-327.
33) (a) Atherton, E.; Holder, J . L.; Meldal, M.; Sheppard, R. C.;
3
Valerio, R. M. J . Chem. Soc., Perkin Trans. 1 1988, 2887-2894. (b)
Peters, S.; Bielfeldt, T.; Meidal, M.; Bock, K.; Paulsen, H. J . Chem.
Soc., Perkin Trans. 1 1992, 1163-1171.
t
Ser( Bu)-OH and Fmoc-Ser-OH; 14.1 and 6.1 min for Fmoc-
Arg(Pbf)-OH and Fmoc-Arg-OH, respectively. These experiments
were conducted in triplicate and the rates and activation
parameters determined from the results are summarized in
Table 1.
(
34) We observed a significant difference in acid-sensitivity of Tyr-
(
SO
3
H) residues in gastrin-II and CCK peptides; the Tyr(SO H) residue
3
in gastrin-II peptides was more sensitive to acidic conditions compared
to that in CCK peptides. It seems very likely that this difference is
due partly to the ionic interactions between the Tyr(SO
3
H) residue
Effects of H
2
O Ad d ition on th e Rea ction Ra tes. Aci-
3
1b
t
and Arg residues in CCK peptides.
dolysis rates of Fmoc-Tyr(SO Na)-OH, Fmoc-Ser( Bu)-OH, and
3

8
J . Org. Chem., Vol. 66, No. 1, 2001
Kitagawa et al.
F igu r e 10. HPLC chromatograms of (a) crude 8, (b) purified 8, (c) crude 9, and (d) purified 9. Asterisks in (a) and (c) show the
desulfated peptides produced during the deprotection. [HPLC conditions: column, Cosmosil 5C18-AR (4.6 × 150 mm); elution
system, a linear gradient of CH
3
CN (B) in 0.1 M AcONH
4
(A); flow rate, 0.8 mL/min for (a), (c), and (d), 1.0 mL/min for (b)].
R
2
(
0 min. For deprotection of the N -Fmoc group of Gln and Glu-
t
O Bu) residues, the concentration of piperidine was reduced
to 10%. After the Fmoc cleavage, the peptide-resin was washed
with DMF (× 6). The next residue was then incorporated with
the DIPCDI-HOBt coupling protocol [Fmoc-amino acid (3
equiv), DIPCDI (3 equiv), and HOBt (3 equiv)] or the PyBOP-
mediated coupling protocol [Fmoc-amino acid (3 equiv), PyBOP
reagent (3 equiv), and N-methylmorpholine (NMM) (9 equiv)].
Asn and Gln residues were exceptionally coupled with an
active ester protocol [Fmoc-Asn/Gln-OPfp (5 equiv), HOOBt
(5 equiv), and NMM (5 equiv)]. After gentle agitation (1.5 h)
and washing with DMF (× 6), part of the peptide-resin was
subjected to the Kaiser test.35 On completion of the assembly,
the peptide-resin was successively washed with DMF (× 5),
MeOH (× 5), and ether (× 5) and then dried in vacuo.
Syn th esis of Ga str in -II a n d CCK P ep tid es by th e
F a cile Solid -P h a se Meth od . P ep tid e Ch a in Assem bly.
Fmoc-Asp-Phe-NH
2
was prepared by a solution-phase method
and attached to Clt resin according to the procedure of Barlos
and co-workers.2 The resulting Fmoc-Asp(Clt resin)-Phe-NH
0c
2
(
307 mg, 100 µmol) was used for the starting dipeptide-resin,
F igu r e 11. HPLC chromatograms of a lysyl endopeptidase
digest of (a) 8 and (b) 9. [HPLC conditions: column, Cosmosil
and each Fmoc-amino acid derivative was manually incorpo-
rated into it. For the synthesis of 4 and 5, the N-terminal
pyroglutamic acid (Pyr) was incorporated into the peptide
5
C
18-AR (4.6 × 150 mm); elution system, a linear gradient of
R
CH CN in 0.1 M AcONH , 5 to 65% in 60 min; flow rate, 1
3
4
chains without N -protection.
mL/min; absorbance was detected at 225 nm]. Arrows in (a)
and (b) show the eluting position of 8 and 9, respectively.
Clea va ge a n d Dep r otection . The protected peptide-resin
was treated with a mixture of HFIP/CH
proximately 5 mL for 100 mg of the peptide-resin) for 0.5 h at
5 °C and then filtered. The combined filtrate and washing
were condensed using a N stream; then dry ether (50 mL)
2 2
Cl (1:4 v/v, ap-
Ta ble 2. Glu cose-Dep en d en t In su lin otr op ic Effect of th e
Syn th etic CCK P ep tid es
2
2
n
insulin(µU/h/islet)a
was added to precipitate the peptides. After centrifugation,
the collected precipitate was dried in vacuo. The cleavage
efficiency was usually about 95% based on the amino acid
analysis of the residual resin. The protected peptide thus
detached was treated with precooled 90% aqueous TFA (ap-
proximately 2 mL for 100 mg of the peptide-resin) at 0 °C for
an appropriate period of time; then dry ether (50 mL) was
added. The formed precipitate was collected by centrifugation
and washed twice with ether and then dissolved in 0.025 M
glucose (3.3 mM)
glucose (11.1 mM)
3
4
7
7
7
7
7
9.3 ( 2.1
14.5 ( 0.9
19.5 ( 4.2
16.7 ( 2.8
19.5 ( 2.6
16.6 ( 2.6
21.6 ( 2.6
-
8
glucose (11.1 mM) + CCK-8 (10 M)
-
-
-
9
glucose (11.1 mM) + CCK-33 (10 M)
8
glucose (11.1 mM) + CCK-33 (10 M)
9
glucose (11.1 mM) + CCK-39 (10 M)
glucose (11.1 mM) + CCK-39 (10- M)
8
a
Mean ( SD.
NH
sample.
4 3
HCO (50 mL) and lyophilized to give a crude peptide
Fmoc-Arg(Pbf)-OH at 0 °C in aqueous TFA (95, 90, 85, or 80%
TFA) were determined as described above. These experiments
were conducted in triplicate, and the results are depicted in
Figure 4.
For the synthesis of big gastrin-II (5), CCK-22 (7), -33 (8),
and -39 (9), the cleavage from the peptide-resin and the
deprotection of the protecting groups were concurrently con-
ducted with 90% aqueous TFA at 0 °C for an appropriate
period of time.
F m oc-Ba sed Solid -P h a se P ep tid e Syn th esis (gen er a l
p r oced u r es). Solid-phase peptide synthesis was conducted
manually. Side-chain protecting groups used in the synthesis
P u r ifica tion by P r ep a r a tive RP -HP LC a n d P u r ity
Assessm en t by An a lytica l HP LC. A crude peptide sample
t
were as follows: Bu for Asp, Glu, Ser, and Tyr; Boc for Lys
R
and His; Pbf for Arg. Fmoc groups for N -protection were
cleaved by 1 min treatment with 20% piperidine in DMF
followed by the second treatment with the same reagent for
(35) Kaiser, E.; Colescott, R. L.; Bossinger, C. D.; Cook, P. I. Anal.
Biochem. 1970, 34, 595-598.

Sulfated Tyrosine-Containing Peptides
J . Org. Chem., Vol. 66, No. 1, 2001
9
obtained after deprotection was purified by RP-HPLC using a
using a linear gradient of B in A (23 to 30% in 60 min). After
lyophilization, 1.8 mg of pure 5 was obtained (13% yield from
the protected peptide-resin). This sample exhibited a sharp
column of µBondasphere 5C18 100 Å (19 × 150 mm). A solvent
4
system consisting of solvent A (0.1 M AcONH ) and solvent B
(CH
3
CN) at a flow rate of 3 mL/min was used for elution, and
R
single peak at t 16.4 min on an analytical HPLC chromato-
the absorbance was detected at 235 or 275 nm depending on
the peptide. The solvent was removed by lyophilization, and
the residue was again dissolved in 0.025 M NH HCO (10 mL).
4 3
gram (Figure 8b). Amino acid ratios in the acid hydrolysate:
Asp 2.09 (2), Ser 1.05 (1), Glu 7.57 (8), Pro 4.97 (5), Gly 3.90
(4), Ala 2.00 (2), Val 0.91 (1), Met 0.72 (1), Leu 2.71 (3), Tyr
0.89 (1), Phe 1.00 (1), Lys 2.12 (2), His 0.94 (1), Trp not
Lyophilization afforded a fluffy powder as a final purified
peptide. The purity of the purified material was assessed by
analytical HPLC using a column of Cosmosil 5C18-AR (4.6 ×
determined (2). LSIMS m/z: calcd. for C176
H
251
+
N
43
O
56
S
2
3929.3
([M], average mass); found 3929.8 [M + H] in the positive-
-
1
5
50 mm) except for 3 and 7, in which a column of Cosmosil
ion mode, and 3927.8 [M - H] in the negative-ion mode.
C
18-AR (6.0 × 150 mm) was used. A solvent system consisting
Lysyl En d op ep tid a se Digestion of 5. Purified 5 (100 µg)
of solvent A (0.1 M AcONH
4
) and solvent B (CH CN) at an
3
4 3
was digested by lysyl endopeptidase in 0.1 M NH HCO (pH
appropriate flow rate was used for elution, and the absorbance
was detected at 225 or 275 nm.
8.2, 250 µL) at 37 °C for 4 h. The weight ratio of the enzyme
to the substrate was 1:200. The enzyme digest of 5 showed
CCK-12 (2). Starting with the protected peptide-resin (150
mg, 25.5 µmol), 40.6 mg of a crude peptide (Figure 7a) was
obtained after cleavage and deprotection (8 h at 0 °C). A
portion of this sample (22.5 mg) was purified by preparative
HPLC using a linear gradient of B in A (25 to 35% in 60 min).
After lyophilization, 5.9 mg of pure 2 was obtained (26% yield
from the protected peptide-resin). This sample exhibited a
two peaks (t
(Figure 9). These two peaks were identified to be the N-
terminal 16-mer (N-16, t )21.8 min) and the C-terminal 17-
mer (C-17, t )32.9 min) from amino acid analyses of their
acid hydrolysates: N-16, Asp 0.98 (1), Ser 0.94 (1), Glu 1.91
(2), Pro 4.36 (4), Gly 2.00 (2), Ala 0.98 (1), Val 0.95 (1), Leu
1.90 (2), Lys 0.97 (1), His 0.96 (1); C-17, Asp 1.03 (1), Glu 5.80
(6), Pro 0.80 (1), Gly 2.25 (2), Ala 1.07 (1), Met 0.92 (1), Leu
1.00 (1), Tyr 0.95 (1), Phe 1.20 (1), Trp not determined (2).
Nonsulfated counterpart, big gastrin-I (6), was similarly
digested by lysyl endopeptidase. The resulting digest exhibited
R
)21.8 and 32.9 min) on an HPLC chromatogram
R
R
sharp single peak at t
R
11.4 min on an analytical HPLC
chromatogram [elution system, a linear gradient of B in A (20
to 35% in 30 min); flow rate, 1 mL/min; absorbance was
detected at 225 nm]. Amino acid ratios in the acid hydrolysate
were as follows (theoretical values are given in parentheses):
Asp 2.76 (3), Ser 0.94 (1), Gly 1.00 (1), Met 1.78 (2), Ile 0.89
two peaks (t
the first eluting peak coincided with the t
R
) 21.8 and 34.3 min) on an HPLC chromatogram;
R
of N-16, while the
(
(
1), Tyr 0.90 (1), Phe 1.05 (1), Trp not determined (1), Arg 0.93
1). LSIMS m/z: calcd for C68 1612.6 ([M - H],
last eluting peak was slightly retarded compared with the t
of C-17.
R
94 17 23 3
H N O S
-
monoisotopic mass); found 1612.5 [M - H] . FT-IR: νmax (KBr)
049 and 1260 cm-1
Min i Ga str in -II (3). Starting with the protected peptide-
Syn th esis of Hu m a n CCK-33 a n d -39. Each residue was
incorporated stepwise into Fmoc-Asp(Clt resin)-Phe-NH (307
1
.
2
mg, 100 µmol). There were no doubts about the completion of
each coupling step, and repeated coupling was not required.
After introduction of the Asn residue (twenty seconds from the
resin (50.0 mg, 10.0 µmol), 19.0 mg of a crude peptide was
obtained after cleavage and deprotection (5 h at 0 °C). This
sample was purified by preparative HPLC using an elution
system; an isocratic elution (22% B in A) for 10 min followed
by a linear gradient of B in A (22 to 32% in 60 min). After
lyophilization, 6.5 mg of pure 3 was obtained (34% yield from
the protected peptide-resin). This sample exhibited a sharp
single peak at t 15.5 min on an analytical HPLC chromato-
R
gram [elution system, a linear gradient of B in A (15 to 45%
in 30 min); flow rate, 1 mL/min; absorbance was detected at
R
C-terminus), the N -Fmoc group was removed, and the result-
ing peptide-resin was dried in vacuo. Part of this peptide-resin
(50 mg) was used for the preparation of CCK-22, and the rest
was further elongated to obtain the longer CCK peptides.
CCK-22 (7). The protected peptide-resin (50 mg, 7.5 µmol)
was treated with 90% aqueous TFA (10 h at 0 °C) for cleavage/
deprotection. The resulting crude peptide (19.0 mg) was
purified by preparative HPLC using a linear gradient of B in
A (25 to 32% in 60 min). After lyophilization, 2.8 mg of pure 7
was obtained (13.4% yield from the protected peptide-resin).
2
(
0
75 nm]. Amino acid ratios in the acid hydrolysate: Asp 1.04
1), Glu 4.74 (5), Gly 1.00 (1), Ala 1.00 (1), Met 0.65 (1), Leu
.94 (1), Tyr 0.87 (1), Phe 0.95 (1), Trp not determined (2).
LSIMS m/z: calcd. for C85 1912.0 ([M - H],
average mass); found 1911.9 [M - H] . FT-IR: νmax (KBr) 1050
R
This sample exhibited a sharp single peak at t 16.6 min on
H
108
N
17
O
30
S
2
an analytical HPLC chromatogram [elution system, a linear
gradient of B in A (20 to 45% in 30 min); flow rate, 1 mL/min;
absorbance was detected at 230 nm]. Amino acid ratios in the
acid hydrolysate: Asp 5.26 (6), Ser 1.84 (2), Glu 1.00 (1), Pro
0.91 (1), Gly 1.00 (1), Met 1.83 (2), Ile 0.76 (1), Leu 1.81 (2),
Tyr 0.89 (1), Phe 1.05 (1), His 0.88 (1), Trp not determined
-
-
1
and 1241 cm
.
Little Ga str in -II (4). Starting with the protected peptide-
resin (102 mg, 20.0 µmol), 43.4 mg of a crude peptide (Figure
7
b) was obtained after cleavage and deprotection (5 h at 0 °C).
A portion of this sample (20.4 mg) was purified by preparative
HPLC using an elution system; an isocratic elution (22% B in
A) for 10 min followed by a linear gradient of B in A (22 to
(1), Arg 1.78 (2). LSIMS m/z: calcd. for C117
([M - H], average mass); found 2788.9 [M - H] .
172 35 39 3
H N O S 2789.0
-
CCK-33 (8). The protected peptide-resin (36 mg, 3.8 µmol)
was treated with 90% aqueous TFA (15 h at 0 °C) for cleavage/
deprotection. The resulting crude peptide (15.0 mg, Figure 10a)
was purified by preparative HPLC using an elution system;
an isocratic elution (27% B in A) for 10 min followed by a linear
gradient of B in A (27 to 30% in 60 min). After lyophilization,
1.5 mg of pure 8 was obtained (9.7% yield from the protected
peptide-resin). This sample exhibited a sharp single peak at
2
7% in 60 min). After lyophilization, 7.9 mg of pure 4 was
obtained (38% yield from the protected peptide-resin). This
sample exhibited a sharp single peak at t 15.4 min on an
R
analytical HPLC chromatogram [elution system, a linear
gradient of B in A (15 to 30% in 30 min); flow rate, 1 mL/min;
absorbance was detected at 275 nm]. Amino acid ratios in the
acid hydrolysate: Asp 1.02 (1), Glu 5.89 (6), Pro 1.31 (1), Gly
2
.00 (2), Ala 1.00 (1), Met 0.31 (1), Leu 1.00 (1), Tyr 0.78 (1),
R
t 18.6 min on an analytical HPLC chromatogram (Figure 10-
Phe 0.95 (1), Trp not determined (2). LSIMS m/z: calcd. for
b). Amino acid ratios in the acid hydrolysate: Asp 5.53 (6), Ser
3.18 (4), Glu 1.03 (1), Pro 2.12 (2), Gly 2.00 (2), Ala 0.89 (1),
Val 0.66 (1), Met 2.53 (3), Ile 1.52 (2), Leu 1.80 (2), Tyr 0.93
(1), Phe 1.00 (1), Lys 1.91 (2), His 0.91 (1), Trp not determined
(1), Arg 2.72 (3). Recoveries of Val and Ile were slightly low
due to steric hindrance of the Ile-Val bond. LSIMS m/z: calcd
C
97
H
122
N
20
-
O
34
S
2
2176.2 ([M - H], average mass); found 2176.3
-
[M - H] , 2096.7 [M - H - SO ] . FT-IR: νmax (KBr) 1049
3
-
1
and 1239 cm
.
Syn th esis of Big Ga str in -II. Big Ga str in -II (5). Each
residue was incorporated stepwise into Fmoc-Asp(Clt resin)-
Phe-NH
2
(155 mg, 50.0 µmol) according to the general proce-
263 51 52 4
for C167H N O S 3945.4 ([M], average mass); found 3946.5
[M + H] in the positive-ion mode, and 3944.6 [M - H] in
the negative-ion mode.
CCK-39 (9). The protected peptide-resin (50 mg, 5.1 µmol)
was treated with 90% aqueous TFA (18 h at 0 °C) for cleavage/
deprotection. The resulting crude peptide (23.6 mg, Figure 10c)
+
-
dures for the Fmoc-SPPS. The resulting protected peptide-
resin (60.0 mg, 10.2 µmol) was treated with 90% aqueous TFA
(8 h at 0 °C) for cleavage/deprotection. The crude peptide (37.0
mg) was obtained after lyophilization (Figure 8a). A portion
of this sample (13.6 mg) was purified by preparative HPLC

1
0
J . Org. Chem., Vol. 66, No. 1, 2001
Kitagawa et al.
was purified by preparative HPLC using an elution system;
an isocratic elution (27% B in A) for 10 min followed by a linear
gradient of B in A (27 to 30% in 60 min). After lyophilization,
and they coincided with peaks of authentic H-YIQQARK-OH
(10), 11, and 7, respectively.
Biologica l Activity of Syn th etic CCK P ep tid es. Pan-
creatic islets were isolated from male Wister rats (about 210
to 240 g) by collagenase digestion and dextran gradient
separation. Approximately 400 to 600 islets were obtained from
each pancreas. Eight to 10 islets were incubated at 30 °C for
60 min in Krebs-Ringer bicarbonate buffer (pH 7.4, 1 mL)
containing glucose (11.1 mM) with or without a synthetic CCK
peptide. Insulin released into the medium by the islets (Table
2) was assayed with an enzyme immunoassay kit (Sanyo,
Kyoto, J apan).
1
.8 mg of pure 9 was obtained (7.8% yield from the protected
peptide-resin). This sample exhibited a sharp single peak at
26.8 min on an analytical HPLC chromatogram (Figure 10-
t
R
d). Amino acid ratios in the acid hydrolysate: Asp 5.95 (6),
Ser 4.07 (4), Glu 3.04 (3), Pro 1.65 (2), Gly 1.85 (2), Ala 1.77
(2), Val 0.67 (1), Met 2.09 (3), Ile 2.29 (3), Leu 2.00 (2), Tyr
1
.74 (2), Phe 1.06 (1), Lys 1.79 (2), His 0.97 (1), Trp not
determined (1), Arg 3.52 (4). Low recovery of Met was due to
the oxidation (Met sulfoxide), and recoveries of Val and Ile
were low as in the case of CCK-33. MALDI-TOFMS m/z: calcd
for C201
H
316
+
N
62
O
61
S
4
4705.3 ([M], average mass); found 4706.0
Ack n ow led gm en t. This work was supported by a
grant from the Promotion and Mutual Aid Corporation
for Private Schools of J apan and a Grant-in-Aid for
Scientific Research (C) (no. 10672008) from the J apan
Society for the Promotion of Science.
-
[
M + H] in the positive-ion mode, and 4704.0 [M - H] in
the negative-ion mode.
Lysyl En d op ep tid a se Digestion of 8 a n d 9. Purified 8
(
100 µg) was digested by lysyl endopeptidase in 0.1 M NH
HCO (pH 8.2, 250 µL) at 37 °C for 4 h. The weight ratio of
the enzyme to the substrate was 1:200. The enzyme digest of
showed two peaks (t ) 21.5 and 46.2 min) on an HPLC
chromatogram (Figure 11a). The first eluting peak (t ) 21.5
min) was indistinguishable from a peak of H-APSGRMSIVK-
OH (11), and the second eluting peak (t ) 46.2 min) coincided
4
-
3
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails for the desulfation rates with the cleavage reagents using
8
R
R
1
1
as a model sulfated peptide and preparation of compound 6,
0, and 11; the positive- and negative-ion mode LSIMS spectra
R
of 5 and 8; the positive- and negative-ion mode MALDI-
TOFMS spectra of 9. This material is available free of charge
via the Internet at http://pubs.acs.org.
with a peak of CCK-22 (7). Purified 9 was similarly digested
by lysyl endopeptidase, and the digest was analyzed on HPLC
using the same chromatographic conditions (Figure 11b).
Three peaks with t
R
) 17.5, 21.5, and 46.2 min were detected,
J O000895Y