Lanthionine-somatostatin analogs: Synthesis, characterization, biological activity, and enzymatic stability studies

Article abstract of DOI:10.1021/jm960850i

A series of cyclic somatostatin analogs containing a lanthionine bridge have been subjected to studies of structure-activity relationships. A direct synthesis of the thioether bridged analog (1) of sandostatin (SMS 201,995) and several lanthionine hexa-, hepta-, and octapeptides was carried out by using the method of cyclization on an oxime resin (PCOR) followed by condensation reactions in solution. The structures of the target peptides were analyzed by liquid secondary ion mass spectrometry (LSIMS) and subjected to high-energy collision-induced dissociation (CID) studies after opening of the peptide ring by proteolytic cleavage. The biological activities of these compounds have been evaluated by assaying their inhibitory potencies for the release of growth hormone (GH) from primary cultures of rat anterior pituitary cells, as well as by their binding affinities to cloned somatostatin receptors (SSTR1-5). The structural modification of sandostatin by introducing a lanthionine bridge resulted in a significantly increased receptor binding selectivity. The lanthionine octapeptide with C-terminal Thr-ol (1) showed similar high affinity for rat SSTR5 compared to somatostatin[1-14] and sandostatin. However, it exhibits about 50 times weaker binding affinity for mSSTR2b than sandostatin. Similarly, the lanthionine octapeptide with the C-terminal Thr-NH₂ residue (2) has higher affinity for rSSTR5 than for mSSTR2B. Both peptides (compounds 1 and 2) have much lower potencies for inhibition of growth hormone secretion than sandostatin. This is consistent with their affinities to SSTR2, the receptor which is believed to be linked to the inhibition of growth hormone release by somatostatin and its analogs. The metabolic stability of lanthionine- sandostatin and sandostatin have been studied in rat brain homogenates. Although both compounds have a high stability toward enzymatic degradation, the lanthionine analog has a 2.4 times longer half-life than sandostatin. The main metabolites of both compounds have been isolated and identified by using an in vivo technique (cerebral microdialysis) and mass spectrometry.

Full text of DOI:10.1021/jm960850i

J . Med. Chem. 1997, 40, 2241-2251
2241
La n th ion in e-Som a tosta tin An a logs: Syn th esis, Ch a r a cter iza tion , Biologica l
Activity, a n d En zym a tic Sta bility Stu d ies
George O¨ sapay,†,‡ Laszlo Prokai,§ Ho-Seung Kim,§ Katalin F. Medzihradszky,| David H. Coy, George Liapakis,#
Terry Reisine,# Giuseppe Melacini,† Qin Zhu,† Susan H.-H. Wang,† Ralph-Heiko Mattern,† and
Murray Goodman*,†
Department of Chemistry and Biochemistry, University of California at San Diego, La J olla, California 92093-0343,
Center for Drug Discovery, College of Pharmacy, University of Florida, J . Hillis Miller Health Center,
Gainesville, Florida 32610, Department of Pharmaceutical Chemistry, University of California San Francisco,
San Francisco, California 94143, Department of Medicine, Tulane University Medical Center, New Orleans, Louisiana 70112,
and Department of Pharmacology, University of Pennsylvania, Philadelphia, Pennsylvania 12104
Received December 17, 1996X
A series of cyclic somatostatin analogs containing a lanthionine bridge have been subjected to
studies of structure-activity relationships. A direct synthesis of the thioether bridged analog
(1) of sandostatin (SMS 201,995) and several lanthionine hexa-, hepta-, and octapeptides was
carried out by using the method of cyclization on an oxime resin (PCOR) followed by
condensation reactions in solution. The structures of the target peptides were analyzed by
liquid secondary ion mass spectrometry (LSIMS) and subjected to high-energy collision-induced
dissociation (CID) studies after opening of the peptide ring by proteolytic cleavage. The
biological activities of these compounds have been evaluated by assaying their inhibitory
potencies for the release of growth hormone (GH) from primary cultures of rat anterior pituitary
cells, as well as by their binding affinities to cloned somatostatin receptors (SSTR1-5). The
structural modification of sandostatin by introducing a lanthionine bridge resulted in a
significantly increased receptor binding selectivity. The lanthionine octapeptide with C-terminal
Thr-ol (1) showed similar high affinity for rat SSTR5 compared to somatostatin[1-14] and
sandostatin. However, it exhibits about 50 times weaker binding affinity for mSSTR2b than
sandostatin. Similarly, the lanthionine octapeptide with the C-terminal Thr-NH2 residue (2)
has higher affinity for rSSTR5 than for mSSTR2B. Both peptides (compounds 1 and 2) have
much lower potencies for inhibition of growth hormone secretion than sandostatin. This is
consistent with their affinities to SSTR2, the receptor which is believed to be linked to the
inhibition of growth hormone release by somatostatin and its analogs. The metabolic stability
of lanthionine-sandostatin and sandostatin have been studied in rat brain homogenates.
Although both compounds have a high stability toward enzymatic degradation, the lanthionine
analog has a 2.4 times longer half-life than sandostatin. The main metabolites of both
compounds have been isolated and identified by using an in vivo technique (cerebral
microdialysis) and mass spectrometry.
In tr od u ction
actions of somatostatin are mediated via membrane
bound receptors. The heterogeneity of its biological
functions together with its short half-life has challenged
many researchers to develop more stable compounds.
This has led to studies to identify structure-activity
relationships of peptides at the somatostatin receptors
which resulted in the discovery of five somatostatin
receptor subtypes.5,6 These receptor subtypes [SSTR1-
5] have been cloned and characterized.6,7 Their specific
functional roles are under extensive investigation.
Recent pharmacological studies have demonstrated a
definite correlation between the receptor affinity for
mSSTR2b and the inhibition of growth hormone secre-
tion by somatostatin analogs.7 The receptor SSTR5 has
been proposed to have a limited role in growth hormone
regulation but instead may have a selective function in
controlling insulin secretion.8,9 In drug research, it is
a key issue to minimize side effects by developing highly
potent and selective drug molecules. In our studies, we
set the goal to design and synthesize new and selective
somatostatin analogs.
Somatostatin, a key regulatory hormone, was isolated
from bovine hypothalami in 1973.1 Biological studies
revealed that native somatostatin occurs in two biologi-
cally active forms, a tetradecapeptide somatostatin-14
(Figure 1) and a 28-residue peptide, somatostatin-28.
Both are derived from the same polypeptide precursor:
prosomatostatin.2
Somatostatin (both the 14-mer and the 28-mer pep-
tides) exhibits diverse physiological activities.3 In the
central nervous system, somatostatin has a neurotrans-
mitter function, modulates locomotor activity, and has
cognitive functions. At a cellular level, it inhibits the
adenylylcyclase activity and calcium ion conductance
and potentiates K+ currents. Through these functions
somatostatin acts as an endocrine hormone, as a para-
crine substance, and as an autocrine secretagogue.4 The
* Author to whom correspondence should be addressed.
† University of California at San Diego.
‡ Present address: College of Medicine, University of California
Irvine, Irvine, CA 92697-4800.
§ University of Florida.
| University of California San Francisco.
Tulane University Medical Center.
Extensive structure-activity studies are important
in order to discover possible therapeutic applications of
#
University of Pennsylvania.
X Abstract published in Advance ACS Abstracts, J une 15, 1997.
S0022-2623(96)00850-3 CCC: $14.00 © 1997 American Chemical Society

2242 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 14
O¨ sapay et al.
F igu r e 1. Structures of natural somatostatin, synthetic sandostatin, and lanthionine-sandostatin analogs.
somatostatin analogs. Structural modifications of na-
tive somatostatin such as the incorporation of D-amino
acids, peptidomimetics, and cyclization have led to the
discovery of analogs with extended half-life and in-
creased biological activities.10,11 Veber et al. suggested
that the amino acid sequence Phe7-Trp8-Lys9-Thr10 (the
superscript numbers here denote the relative positions
of the amino acids in the native somatostatin), kept in
â-II’ turn conformation, it is necessary for biological
activity.11 This hypothesis led Veber and his col-
leagues12 to the design and synthesis of the highly active
cyclic hexapeptide somatostatin analog c[Pro6-D-Trp8-
Lys9-Thr10-Phe11].
If either Phe7 or Phe11 is replaced by an Ala residue
in the parent compound, the in vitro activity for the
inhibition of growth hormone release drops by about
90%.13 Furthermore, the free ꢀ-amino group of a Lys
residue has been shown to be necessary for eliciting
bioactivity.14 In addition, the D-Trp at position 8 is
highly sensitive to substitution and modification. Re-
placement of D-Trp with D-Phe causes a drop of 2 orders
of magnitude in the inhibition of growth hormone
release. Substitution of Pro6 and Thr10 affects the
biological activity to a much lower extent.15 Therefore,
only residues Phe7, D-Trp8, Lys9, and Phe11 are thought
to be primarily responsible for somatostatin activity.
modification of the cyclic hexapeptide c[Cys-Phe-D-Trp-
Lys-Thr-Cys] led to the synthesis of the highly active
somatostatin analog, sandostatin (SMS 201, 995), H-D-
Phe-c[Cys-Phe-D-Trp-Lys-Thr-Cys]-Thr-ol (Figure 1).18
On the basis of conformational analysis of sandostatin,
the authors suggested that the disulfide bridge might
not be sufficient to constrain the cyclic backbone in the
proposed bioactive conformation.19 They supposed that
Thr-ol, as the C-terminal residue and D-Phe, as the
N-terminal residue might have additional stabilizing
effect on the active conformation through the formation
of an antiparallel â-sheet-like structure.20
However, the exocyclic residues (D-Phe and Thr-ol)
have only moderate conformational stability which
results in a decreased stability of the bioactive confor-
mation of sandostatin. In order to introduce further
conformational constraints into the ring skeleton, we
have designed a series of somatostatin analogs in which
the disulfide bridge has been replaced by a lanthionine
monosulfide bridge (Figure 1).21 The goal of the present
work focuses on structural modifications of sandostatin
to obtain an improved pharmacological profile. These
modifications led to the development of compounds with
improved selectivity at SSTR5.
Resu lts
Conformational analysis of the bioactive analogs
containing backbone modifications such as retroinverso
modification16 and the completely retro backbone17 has
indicated that the peptide backbone is not directly
involved in binding, but serves mainly as a scaffold
which allows the side chains to adopt the necessary
pharmacophore spatial arrangement.
P ep tid e Design a n d Syn th esis. In the design of
our lanthionine-somatostatin analogs, our strategy was
to constrain the backbone conformation of the biologi-
cally highly active sandostatin without changing the
spatial arrangements of the pharmacophore groups
(Phe7, Trp8, and Lys9) and to enhance the stability of
sandostatin.
The amino acid sequence Phe-D-Trp-Lys-Thr can be
stabilized in the bioactive conformation if a disulfide
bridge is used in place of the Pro-Phe bridge. Stepwise
Molecular modeling studies of sandostatin and lan-
thionine-sandostatin molecules were carried out to
compare the possible and preferred conformations of the

Lanthionine-Somatostatin Analogs
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 14 2243
Sch em e 1. Synthesis of a Lanthionine-Sandostatin
Intermediate on an Oxime Resin
disulfide and the lanthionine-bridged structures in the
cyclic hexapeptides (Figure 1). These investigations
indicated that the removal of one sulfur atom from the
20-membered cyclic hexapeptide will not alter the
orientation of the pharmacophore groups. Furthermore,
we believed that the reduced ring size might lead to a
more stable conformation.
In our synthetic strategy, the fully protected lanthion-
ine hexapeptide (3) was the key intermediate for the
preparation of the hepta- and octapeptides through
addition of the exocyclic residues (D-Phe1, Thr8NH2, or
Thr8(Bzl)-ol). Recently, our laboratory has developed
and applied new synthetic routes for preparation of
lanthionine peptides.22,23
For the preparation of the lanthionine hexapeptide
(3), we extruded a sulfur atom from an existing disulfide
bridge by using the HMPA reagent.24 We have previ-
ously used this methodology for the synthesis of lanthion-
ine enkephalins.25 However, in our present work, these
reactions resulted in extremely low yields (<5%). Under
the conditions used, the formyl protecting group was
cleaved from the indole ring (determined by HPLC and
mass spectrometry analyses) faster than the extrusion
of the sulfur atom occurred. These results support our
previous experience with “sulfur-removal” reactions
which require not only fully protected peptides but the
use of base-stable protecting groups.
A novel synthetic route has been developed in our
laboratory for the synthesis of lanthionine peptides. This
method is based upon the use of an appropriately
protected lanthionine amino acid building block
93.7%), respectively. The final deprotection of the hexa-
and heptapeptides was carried out in two steps: first
the methyl ester group was hydrolyzed with 0.05 N
NaOH/HMPA-water (1/1, v/v) at 10 °C for 1 h, and then
the remaining protecting groups were removed by Na/
NH3 treatment. The crude products were subjected to
Sephadex G-10 gel filtration and HPLC purification.
Two peaks in the HPLC spectrum of the crude heptamer
showed identical UV and mass spectra. Consequently,
the two compounds must be diastereomers in a ratio of
3:2. The diastereomers were separable by HPLC. Two
dimensional NMR spectroscopy was used to make the
configurational assingments at residue 7. The main
product was the desired molecule (9) with the L config-
uration at residue 7. The heptamer 9 exhibits an
[CbzAlaL (BocAlaLOMe)OH]
S
and the method of peptide cyclization on an oxime resin
(PCOR).26,27 [The Cbz and Boc protecting groups can
be selectively removed.] The synthesis was initiated by
the preparation of BocThr(Bzl)-oxime resin (substitution
level: 0.26 mmol/g, determined by picric acid titra-
tion28). The peptide chain was assembled by using Boc
chemistry and BOP activation method (Scheme 1).
Finally, the protected lanthionine residue (4) was
coupled to the peptide chain by using 3-fold molar excess
of the amino acid and BOP reagent. The reaction was
completed within 1 h according to the Kaiser test.29 The
Boc group was then cleaved from the lanthionine
residue followed by the peptide cyclization reaction with
acetic acid catalysis. The reaction was carried out for
24 h according to the general procedure of the PCOR
method.26 The product (3) was obtained after DMF/
water and DMF/ether precipitations in a 65% yield
based upon the starting material, BocThr(Bzl) oxime
resin.
Condensations of the cyclic hexapeptide (3) with the
C- and the N-terminal exocyclic amino acids were
carried out in solution (Scheme 2). The benzyloxycar-
bonyl group was removed from the N-terminus with
HBr/acetic acid treatment (which also cleaved the Bzl
protecting groups). A 5-fold excess of Cbz-D-Phe1-OH
activated with EDC/HOBt was used to form the peptide
bond (yield 86%). In the next step, the heptapeptide
methyl ester was converted to the hydrazide (5, yield
60.9%). The azide (prepared by butyl nitrite/HCl treat-
ment) was allowed to react with a 6-fold excess of
threonine-amide or threonine(Bzl)-ol (yields 97.2% and
7
HR2HR NOE cross peak, which was present in all the
other analogs with an L configuration in the tail unit of
the lanthionine residue. The heptamer 10 does not
show this NOE cross peak. These results allowed us to
7
assign the D configuration to the chiral center CR in
the carboxyl terminal lanthionine unit for heptamer 10.
The hexa- and heptapeptide analogs of sandostatin
with disulfide bridges (12 and 13) were assembled in
an ABI model 433A automatic peptide synthesizer using
standard solid phase procedure (Scheme 3). An HMPB
resin served as solid phase support, and Fmoc chemistry
and BOP/HOBt activation were used in the syntheses.
The disulfide bridge formation was carried out by iodine
oxidation on the resin.30 Finally, Reagent-K was used
to cleave the peptides from the resin with simultaneous
cleavage of the protecting groups. The target disulfide-
bridged peptides were obtained by reversed-phase HPLC
purification.
NMR a n d Ma ss Sp ectr om etr y
FAB mass spectrometry was used to verify the identi-
ties of all intermediates and final compounds. The

2244 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 14
O¨ sapay et al.
Sch em e 2. Synthesis of Lanthionine-Sandostatin Derivativesa
a
t
(a) (1) HBr/AcOH, (2) Cbz-D-PheOH + EDC + HOBt; (b) N2H4; (c) (1) BuNO2 + HCl, (2) H-Thr(Bzl)-ol or H-Thr-NH2; (d) (1) NaOH,
(2) Na/NH3.
Sch em e 3
structural details of each compound were investigated
by 1H NMR spectroscopy. The assignment of every
signal in the 500 MHz NMR spectra was possible with
the use of two-dimensional HOHAHA, COSY, and
ROESY experiments. The intramolecular NOE effects
between the R-protons of the lanthionine residue was
used to determine the chirality of the two somatostatin
CID analysis. Fragmentation observed in these experi-
ments confirmed the amino acid sequence of the pep-
tides and the location of the sulfide linkage. Figure 2
shows the high-energy CID fragmentation of the chy-
motrypsin-digested sulfide-bridged heptapeptide. The
immonium and related ions at m/ z 74; 84, 101, and 129;
120; 130, 159, and 170 indicated the presence of Thr,
Lys, Phe, and Trp residues, respectively.35 Ions y1, y2
and y3 at m/ z 503, 604, and 732, respectively, formed
via peptide bond cleavages with charge retention at the
C-terminus, reveal the amino acid sequence for the
longer chain (nomenclature according to ref 36). The
corresponding N-terminal ions, b2 and b3, were detected
at m/ z 315 and 416, respectively. Loss of Phe residues
from both ends of the shorter chain were observed
yielding y2* ion at 771 and b2* ion at m/ z 753. Some
of the most abundant fragments resulted from double
cleavages. They appear at masses corresponding to
sequential amino acid losses from y2* ion, at m/ z 585
(Trp), 457 (Lys), and 356 (Thr), respectively (labeled
with asterisks in Figure 2).
1
heptamers (9, 10) separated by HPLC. The H NMR
data of intermediates are listed in the Experimental
Section. Detailed NMR spectra and conformational
analysis of the target compounds are presented in an
accompanying paper.44
In the last decade, mass spectrometry has also
developed as a method of choice for characterizing
synthetic peptides.31 For studying the molecular struc-
ture of our unprotected hexa-, hepta-, and octapeptides,
all samples were subjected to liquid secondary ion mass
spectrometry (LSIMS) analysis. Each compound pro-
duced the expected molecular ion [MH+] as listed in
Table 1. Although high-energy collision-induced dis-
sociation (CID) analysis32,33 of the cyclic peptides pro-
vides some information about their structure, the open
chain structures were expected to show more extensive
fragmentation.34 Thus, the cyclic peptides were sub-
jected to proteolytic digestion with chymotrypsin. The
chymotryptic digestion resulted in ring opening (a
molecular mass increase of 18 Da, see Table 1). The
newly formed species were also subjected to high-energy
A bond cleavage at the sulfur atom, with charge
retention on the more basic (i.e., Lys-containing) peptide
chain yields the ion at m/ z 536 (indicated with an arrow
in Figure 2). The analogous high-energy CID fragmen-
tation of the disulfide-bridged heptapeptide was ob-
served. All fragment ions containing the cross-linkage
are shifted by 32 Da because of the presence of the

Lanthionine-Somatostatin Analogs
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 14 2245
Ta ble 1. Molecular Masses Measured for the Lanthionine and Disulfide-Bridged Peptides before and after Enzymatic Ring Opening
compound
[MH+] theoretical
753.34
[MH+] measured
752.8
[MH+] after chymotryptic digestion
752.8 not cleaved
H-AlaL-Phe-D-Trp-Lys-Thr-AlaL-OH (8)
S
H-D-Phe-AlaL-Phe-D-Trp-Lys-Thr-AlaL-OH (9)
S
900.41
987.48
900.1
987.1
918.5
H-D-Phe-AlaL-Phe-D-Trp-Lys-Thr-AlaL-Thr-ol (1)
1005.5
S
H-D-Phe-AlaL-Phe-D-Trp-Lys-Thr-AlaL-Thr-NH2 (2)
1000.47
785.31
1000.5
785.3
1018.5
S
H-Cys-Phe-D-Trp-Lys-Thr-Cys-OH (12)
785.3 not cleaved
S - S
H-D-Phe-Cys-Phe-D-Trp-Lys-Thr-Cys-OH (13)
S - S
932.38
932.4
950.5
H-D-Phe-Cys-Phe-D-Trp-Lys-Thr-Cys-Thr-ol
1,019.45
1,019.7
1,037.6
S - S
The biological results for the new lanthionine-san-
dostatin analogs showed that only the octamers (1 and
2) inhibit GH release with EC50 values in the nanomolar
range, but they are approximately 100-fold less potent
than somatostatin. These results are in good agreement
with the binding data of these compounds measured on
SSTR1-5 subtypes. Compounds 1 and 2 bind to
mSSTR2b with 13-16 nM affinities. These affinities
are approximately 50-fold weaker than somatostatin-
14. There are no clear pharmacological differences
between rodent and human SSTR1, SSTR2, SSTR3, and
SSTR4. Only SSTR5 exhibits species variation in the
receptor pharmacology.40 Compounds 8-10 did not
bind to SSTR2b and did not affect GH release. Our
lanthionine hexa- and the diastereomeric heptamers
showed no binding to this receptor type. It is worth-
while noting that the lanthionine hexapeptide (8) and
F igu r e 2. High-energy CID spectrum of the chymotrypsin-
digested lanthionine heptapeptide. ([MH+] at m/ z 918.5). The
immonium and related low mass ions are labeled with the one
letter code of the corresponding amino acid. Peptide sequence
ions are labeled according to the nomenclature.36 Fragment
y2* corresponds to the loss of the N-terminal D-Phe residue
from the shorter chain. Ions labeled with asterisks are formed
via double bond cleavages and can be described as sequential
amino acid losses from y2* ion. The fragment formed via
cleavage at the sulfur atom is indicated with an arrow.
7
the heptapeptide (9) containing L-AlaL have weak
activity, but only for rSSTR5, a receptor not associated
with inhibition of growth hormone secretion. The
disulfide-bridged heptapeptide (13) is equally potent on
mSSTR2b, rSSTR5, and hSSTR5.
The two lanthionine octapeptides potently bind to
both rat and human SSTR5. Lanthionine-sandostatin
has similar affinity for rSSTR5 as sandostatin, but it
has approximately 50-fold lower affinity for mSSTR2b
than somatostatin. By comparing the ratios of IC50
values for mSSTR2b/rSSTR5, the receptor binding as-
says of the disulfide- and monosulfide-bridged octapep-
tide alcohols show that the lanthionine analog is about
40 times more selective for rSSTR5. For human SSTR5,
the lanthionine octapeptide amide has affinity similar
to that of the threoninol analog.
The lanthionine analogs showed no affinity (IC50 > 1
µM) for hSSTR1, mSSTR3, and hSSTR4, while soma-
tostatin-14 is active in subnanomolar level at these
receptors. In contrast to lanthionine-sandostatin, san-
dostatin has high affinity for mSSTR3 and hSSTR4
(IC50 ) 3 nM). These results indicate that the introduc-
tion of the lanthionine bridge into sandostatin provides
an altered receptor binding profile, namely a higher
selectivity for SSTR5.
additional sulfur atom. Thus, ions y1, y2, and y3 are at
m/ z 535, 636, and 764, respectively. Similarly, frag-
ment y2* is detected at m/ z 803, while ions formed via
sequential amino acid or peptide segment loss from y2*
ion at m/ z 617 (Trp), 489 (TrpLys), and 388 (TrpLysThr)
(Figure 1S in the Supporting Information). Cleavage
in the disulfide bridge yield ions at m/ z 567 and at m/ z
536.
Biologica l Resu lts. The native somatostatin hor-
mone has a broad variety of biological activities as we
briefly reviewed in the Introduction. In our study, we
examined the ability of the synthetic peptides to inhibit
the release of growth hormone in vitro. The compounds
were also tested for specific binding to the six cloned
somatostatin receptors [hSSTR1, mSTTR2b, mSSTR3,
hSSTR4, rSSTR5, and hSSTR5 (h ) human, m )
mouse, r ) rat)] expressed in CHO cell membranes.37,38
The results are summarized in Table 2. Somatostatin-
14 and its analog, sandostatin, are used as reference
materials in the evaluation of the biological potencies
of the lanthionine-somatostatin analogs. In earlier
studies we have demonstrated that the lanthionine
analog of somatostatin-14 has a similar binding affinity
to mSSTR2b as the natural disulfide-bridged peptide.39
Biod egr a d a tion Stu d ies in Vitr o a n d in Vivo
Biostability is a key issue in peptide drug develop-
ment. For centrally active peptides, degradation in
brain tissue takes on special significance. Cerebral
breakdown may limit the duration of action and also
can change pharmacological profiles. Thus, biological

2246 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 14
O¨ sapay et al.
F igu r e 3. Molecular ion [MH+] regions of the ESI mass spectra form dialysates collected during infusion of sandostatin (A) and
lanthionine-sandostatin (B) into rat brain (striatum).
Ta ble 2. Potencies of Somatostatin Analogs To Inhibit Radioligand Binding to Cloned Somatostatin Receptors and To Inhibit GH
Release from Primary Cultures of Male Rat Anterior Pituary Cells
IC50 (nM)a
Ic50 ratio
inhibn of GH
no.
compound
mSSTR2b mSSTR3
hSSTR4 rSSTR5 hSSTR5 SSTR2b/rSSTR5 Release IC50 (nM)
somatostatin, SRIF[1-14]
sandostatin, SMS-201,995
H-D-Phe-AlaL-Phe-D-Trp-Lys-Thr-AlaL-Thr-ol
0.28
0.28
13.13 >1000
0.08
3
0.08
3
>1000
0.86
1.04
1.29
0.33
0.27
0.52 ( 0.7
0.52 ( 0.7
48 ( 1.8
0.77
4.17
1
2
10.2
3.02
>2
>2
0.75
S
H-D-Phe-AlaL-Phe-D-Trp-Lys-Thr-AlaL-Thr-NH2
16.60 >1000
>1000
5.50
4.92
70 ( 13
S
8
9
H-AlaL-Phe-D-Trp-Lys-Thr-AlaL-OH
>1000
>1000
>1000
>1000
>1000
>1000
>1000
500
500
10.63
N/D
N/D
>1000
>1000
N/D
S
H-D-Phe-Cys-Phe-D-Trp-Lys-Thr-Cys-OH
S - S
H-D-Phe-Cys-Phe-D-Trp-Lys-Thr-Cys-OH
S - S
13
13.62 >1000
10.95
a
Data are expressed as IC50 (nM) ( SEM where available; n ) 5 replicate determinations. None of the compounds, except natural
somatostatin, showed biological activity on hSSTR1.
Ta ble 3. Stability of Lanthionine-Sandostatin (1) Compared
with SMS-201,995 against Degradation by Rat Brain
Homogenate
action may be mediated through conversion to metabo-
lites.41 Biostability in brain tissue is, therefore, an
important property of newly developed, centrally acting
peptide analogs such as those discussed in this paper.
Rat brain homogenate42 is commonly used to evaluate
the possible fate of molecules in the CNS in vitro,
because both soluble and membrane-bound peptidases
are present in these preparations. An in vivo technique
utilizing cerebral microdialysis and mass spectrometry
has also been developed to identify extracellular me-
tabolites of peptides in the central nervous system.43
The enzymatic degradation rates of lanthionine-
sandostatin (1) and sandostatin, as reference material,
have been determined by incubating the peptides with
rat brain homogenates at 37 °C and pH 7.4 for 90 min.
The process of degradation was followed by determining
the remaining peptide concentration by HPLC analysis.
Concentration-time profiles were analyzed by expo-
nential fitting, assuming a pseudo-first-order degrada-
tion process. Half-lives (Table 3) were calculated from
the rate constants (k) as 0.693/k.
standard regression
t1/2
(h)
deviation coefficient
compound
k (1/min)
(1/min)
R2 (1/min)
lanthionine-sandostatin 5.5 × 10-4 30.3 4.5 × 10-5
0.99379
0.99795
sandostatin
1.3 × 10-3 12.5 4.8 × 1--4
high “background” in chromatographic analysis which
prevented us from locating and identifying CNS me-
tabolites from these rather stable peptides in vitro.
Therefore, an in vivo technique involving the introduc-
tion of the peptides into the brain via a microdialysis
probe fitted with a semipermeable polymer membrane
(20 000 Da cutoff) was utilized.43 In this technique, the
drug enters from the probe by diffusion (driven by the
concentration gradient between the inside of the probe
and the extracellular space of the brain), and metabo-
lites are formed in the living tissue. The metabolic
fragments, in turn, distribute themselves into the probe
side solution free of interference by in vitro artifacts and
are collected in the dialysate by the continuous perfu-
Self-proteolytic products of excised brain tissue gave

Lanthionine-Somatostatin Analogs
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 14 2247
sion of the microdialysis device. The identities of the
in vivo metabolites are then determined by ESI mass
spectrometric analysis (Figure 3).
Biological potencies, conformational properties, and
physiological behavior of lanthionine analogs of san-
dostatin have been investigated and compared with the
properties of sandostatin. High biological potencies
were expected since the Phe3-D-Trp4-Lys5-Thr6 tetrapep-
tide, which is considered as the active region, remained
intact and a preferred type II’ â-turn backbone confor-
mation in this tetrapeptide region exists in both mol-
ecules. Our conformational studies44 established these
similarities. In accordance with the conformational
changes, significant biological alterations were found in
in vitro bioactivity, receptor binding, and binding se-
lectivity. The two lanthionine octapeptides (1 and 2)
have reduced affinities for the mSSTR2b receptor (about
1/50 of the binding affinity [IC50] of sandostatin).
However, the affinity of compound 1 for rSSTR5 is
similar to those of somatostatin-14 and sandostatin.
High receptor selectivity is important in order to reduce
side effects. The [IC50]mSSTR2b/[IC50]rSSTR5 affinity ratio
determined for sandostatin and compounds 1 and 2 are
0.3, 10.2, and 2.9, respectively. The lanthionine-
sandostatinamide (2) has a 25% weaker binding to
mSSTR2b than lanthionine-sandostatin (1), which
leads to differences in their in vitro inhibitory activity
in the release of GH from primary cultures of rat
anterior pituitary cells. It is important to highlight that
the two octapeptide lanthionine-sandostatin analogs
have an influence on GH release-inhibition about 2
orders of magnitude weaker than sandostatin. The
lower affinities for mSSTR2b and lower inhibitory
activity on GH release confirms the suggestion7 that this
somatostatin receptor subtype is involved in GH secre-
tion.
The metabolic stability of sandostatin was expected to
be increased by introducing a lanthionine bridge in the
place of the disulfide moiety. Our previous studies on
lanthionine opioids indicated that the lanthionine mol-
ecules are significantly more stable toward enzymatic
degradation than their disulfide-bridged analogs. Bauer
et al. reported in 1982 that sandostatin is many times
more stable than the natural somatostatin-14 in an
ultrafiltrate of rat kidney homogenate.18 In our study,
the in vitro stability of lanthionine-sandostatin (1) was
compared with sandostatin against degradation by rat
brain homogenate. Both compounds have clearly a
prolonged duration of action. However, the lanthionine
analog showed a remarkable 2.4 times longer half life
than sandostatin. In vivo metabolism of sandostatin
and lanthionine-sandostatin in rat brain tissue was
studied by utilizing the cerebral microdialysis method.
Both peptides yielded two extracellular metabolites
detected by ESI-MS from the collected dialysates. The
exocyclic residues (D-Phe and Thr-ol) show enzymatic
instability; they are cleaved enzymatically from the
parent compounds and provide the corresponding hexa-
and heptapeptides as major metabolites. Only small
portions of both drugs degraded in vivo in rat brain,
which highlights the increased stability of these mol-
ecules.
The main degradation product of sandostatin corre-
sponds to the heptapeptide formed by enzymatic cleav-
age of the N-terminal D-Phe [protonated molecule, MH+,
found at m/ z 872.1 in the ESI mass spectrum, expected
m/ z 872.1]. The lanthionine-sandostatin also showed
the loss of the N-terminal D-Phe (MH+, found at m/ z
839.6, expected m/ z 840.0) as a major metabolic path-
way. The loss of the C-terminal Thr-ol (MH+, found at
m/ z 899.9, expected m/ z 900.1) in the ESI mass
spectrum was a minor metabolite. No other metabolites
by cleavage of peptide bonds inside of the disulfide-
bridged and the lanthionine cyclopeptide were detected.
Discu ssion
Our synthetic efforts resulted in a series of lanthion-
ine-bridged sandostatin analogs by utilizing our recently
developed lanthionine ring formation method on an
oxime resin. This method was used to prepare the fully
protected lanthionine-bridged hexapeptide (3) that served
as synthetic precursor for all lanthionine-sandostatin
analogs. Although this strategy requires a preprepared
protected lanthionine unit, the peptide synthesis and
the cyclization reaction proceed smoothly in high yields.
The protecting groups used (Boc, Bzl, For, Tos) are
generally applicable in other PCOR syntheses.39 Chro-
matographic (HPLC analysis/purity angle on a Water’s
Milleneum system) and spectroscopic (1D and 2D-NMR,
LSIMS, and CID studies) analyses of all compounds
established their purities and structures. A comparison
of conformationally relevant NMR parameters of the
molecules was used to determine the chirality of the
7
AlaL residue in a diastereomeric pair (9 and 10). The
7
D-AlaL residue in compound 10 is a consequence of a
partial racemization which occurred during the ester
hydrolysis of compound 4. Low levels of racemization
(<2%) were observed during a similar ester hydrolysis
on the hexapeptide (3).
Earlier studies including the synthesis of a series of
sandostatin analogs with different C-terminal residues18
have shown that, for the strongest somatostatin-like
activity, threonine derivatives must be incorporated as
the C-terminal residue (Thr-NH2 or Thr-ol). Both the
C- and the N-terminal residues play a significant role
in the manifestation of high biological activity. Lack
of the threonine alcoholic hydroxyl group resulted in
reduced activity. In our study, we designed and syn-
thesized compounds in which the peptide ring is more
constrained than in sandostatin. This smaller ring size
should increase the conformational stability of the cyclic
backbone. However, the biological tests showed no or
very little activity for the lanthionine heptapeptides (9,
10, and 11) and both mono- and disulfide-bridged
hexapeptides (8 and 12). These findings suggest that
the two exocyclic residues (D-Phe1 and Thr8-ol or amide)
might have specific function in receptor binding. The
disulfide-bridged heptapeptide (13) shows significant
activity but no selectivity on mSSTR2b and r/hSSTR5.
In addition, the type of the C-terminal end group
slightly affects the binding potency. For example, the
disulfide-bridged hexapeptide acid (12) does not have
somatostatin-like activity; however, its C-terminal amide
analog has weak activity according to the literature.18
Con clu sion
Sandostatin, a highly potent, clinically used soma-
tostatin analog, was the subject of our chemical modi-
fications in order to develop new lead molecules with
improved pharmacological profiles. The ring structure
of sandostatin was reduced by elimination of one sulfur

2248 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 14
O¨ sapay et al.
1H, Hâ2-Lys5), 2.33 (s, 3H, Me-Tos), 2.57 (m, 2H, 2Hꢀ-Lys5),
atom from the disulfide bridge. The synthesis of these
compounds was carried out on an oxime resin by using
the PCOR method. The lanthionine-sandostatin, in
comparison to sandostatin, showed an enhanced recep-
tor selectivity: decreased affinity for mSSTR2b recep-
tors and consequently 2 orders of magnitude weaker
effect on the inhibition of GH release. In addition, the
lanthionine-sandostatin has the same affinity for
rSSTR5 as sandostatin.
The introduction of the lanthionine moiety into the
sandostatin molecule increased its metabolic stability.
The lanthionine-sandostatin has a longer half-life (2.4
times) in rat brain homogenates. In in vivo studies, two
extracellular metabolites of lanthionine-sandostatin
have been determined from the CNS for the first time.
In summary, the newly developed lanthionine-san-
dostatin octapeptides are good drug candidates on the
basis of their high and selective binding affinities for
somatostatin receptor SSTR5 and their increased stabil-
ity against enzymatic degradation.
2.68 (m, 1H, Hâ1-Phe3), 2.74 (m, 2H, 2Hâ-AlaL2), 2.77 (m, 2H,
7
Hâ1-AlaL + Hâ2-Phe3), 2.81 (m, 1H, Hâ1-D-Trp4), 2.86 (m, 1H,
Hâ2-AlaL7), 2.99 (m, 1H, Hâ2-D-Trp4), 3.60 (s, 3H, OMe), 3.88
(quintet, 1H, Hâ-Thr6, J ) 5.48 Hz), 4.48 (d, 1H, Hâ1-CH2 of
BzlO-Thr6, J ) 12.3 Hz), 4.51 (d, 1H, Hâ2-CH2 of BzlO-Thr6, J
) 12.3 Hz), 4.58 (m*, 1H, HR-Phe3), 4.67 (m*, 1H, HR-AlaL2),
5.00 (s, 2H, Cbz(CH2)), 6.96-7.02 (m, 5H, C6H5-Phe3), 7.21-
7.28 (m, 5H, C6H5 of BzlO-Thr6), 7.29 (d, 1H, NH-AlaL2, J )
7.8 Hz), 7.29-7.35 (m, 5H, C6H5 of Cbz), 7.35 (d, 2H of Tos
ortho to SO2), 7.36-7.41 (m, 3H, H2 + H5 + H6 of Trp4), 7.42
(t, 1H, NH(ꢀ)-Lys5, J ≈ 5 Hz), 7.53-7.58 (m, 2H, H4 + H7 of
D-Trp4), 7.54 (d, 1H, NH-Thr6, J ≈ 8 Hz), 7.65 (d, 2H, 2H of
Tos meta to SO2), 8.17 (d, 1H, NH-Lys5), 8.31 (d, 1H, NH-Phe3,
J ≈ 8 Hz), 8.33 (d, 1H, NH-AlaL7, J ) 7.8 Hz), 8.41 (d, 1H,
NH-D-Trp4, J ≈ 8 Hz), 9.15 (s, 0.6H#, H(For)), 9.58 (s, 0.4#,
H(For)). *This multiplet appears like a broad doublet of
doublets. #The H of For appears splitted in two singlets,
possibly because of the presence of a cis-trans isomerism
involving the For group.
S
Cbz-D-P h e-c[AlaLP h e-D-Tr p(For )Lys(Tos)Th r (Bzl)AlaL]-
OMe (4). The protected hexapeptide (3, 1.0 g, 0.85 mmol) was
specifically deprotected with 33% HBr/AcOH (6 mL, v/v) at 0
°C for 10 min and at room temperature for 50 min. Absolute
ether (50 mL) was added to the reaction mixture, and the
solution was kept at 0 °C for 1 h. The product was filtered
and dried. Yield: 0.92 g (96.7%). Rf(B): 0.93. Cbz-D-Phe-
OH (3.2 g, 3.29 mmol) and HOBt‚H2O (0.89 g, 6.58 mmol) were
dissolved in DMF (5 mL) and cooled to 0 °C, and EDC‚HCl
(0.69 g, 3.62 mmol), DIEA (1.91 mL, 11.0 mmol), and the HBr
salt of the above peptide (0.91 g, 0.82 mmol) were given to the
reaction mixture. After being stirred for 1 h at 0 °C and
overnight at room temperature, the product was precipitated
by addition of water. The crude product was filtered, washed
with 0.5 N HCl, 5% NaHCO3, and water, and then recrystal-
Exp er im en ta l Section
Protected amino acids were purchased from either Bachem
Bioscience or Novabiochem. All amino acids were of the
L
configuration if not otherwise indicated. p-Nitrobenzophenone
oxime resin was prepared according to the literature.45 ACS
grade DCM, DMF, i-PrOH, EtOAc, CHCl3, and MeOH were
purchased from Fisher Scientific and dried over sodium
aluminosilicate molecular sieves (4 Å nominal pore diameter)
obtained from Sigma. DMF was treated with Amberlist-15
cation-exchange resin. TFA, TFE (Aldrich), and NMP (Applied
Biosystem) were used without further purification. DIEA
(Aldrich) was dried over KOH and distilled from ninhydrin.
Silica gel for flash chromatography was purchased from Baker,
HOBt from Merck, BOP from Novabiochem, thioanisole from
Aldrich, EDC‚HCl from Bachem California, and HBTU from
Applied Biosystem.
lized from DMF-ether. Yield: 0.9 g (86%). RP-HPLC: tR
)
31.8 min (standard analytical conditions described above).
FAB-MS: m/ z 1231 [MH+], theoretically 1230.5.
S
Cbz-D-P h e-c[AlaLP h e-D-Tr p(For )Lys(Tos)Th r (Bzl)AlaL]-
Th r (Bzl)-ol (6). The protected heptapeptide (4, 0.3 g, 0.24
mmol) in DMF (6 mL) was stirred with N2H4‚H2O (0.6 mL)
overnight under a nitrogen atmosphere. The solution was
concentrated under reduced pressure, and then water was
added to the reaction mixture. The precipitated hydrazide (5)
was filtered and dried. Yield: 220 mg (60.9%). Rf(A): 0.37.
RP-HPLC: tR ) 29.3 min (standard analytical conditions
described above). FAB-MS: m/ z 1204 [MH+], theoretically
1202.5.
To a solution of the peptide hydrazide (5, 180 mg, 0.15
mmol) in DMF (2 mL) cooled to -15 °C were added 4 N HCl/
dioxane (0.225 mL) and tert-butyl nitrite (0.042 mL 0.35
mmol). After the mixture was stirred for 25 min at -15 °C,
HCl‚H-Thr(Bzl)-ol (0.21 g, 0.9 mmol) and DIEA (0.195 mL, 1.12
mmol) were added. The reaction mixture was stirred for 1 h
at -15 °C and for 48 h at 4 °C and then diluted with water.
The precipitated product was filtered, washed with 1 N HCl,
and 5% NaHCO3, and dried. The crude material was recrys-
tallized from DMF-ether. Yield: 185 mg (93.7%). Rf(A): 0.61.
RP-HPLC: tR ) 33.1 min (standard analytical conditions
described above). FAB-MS: m/ z 1367 [MH+], theoretically
1365.6. 1H-NMR (DMSO-d6): δ (ppm) 0.87 (q, 2H, 2Hγ-Lys5),
0.98 (q, 3H, Me-Thr6, J ) 6.8 Hz), 1.04 (d, 3H, Me-Thr8, J )
6.8 Hz), 1.19 (m, 2H, 2Hδ-Lys5), 1.31 (br, 1H, Hâ2-Lys5), 1.51
Solid phase peptide syntheses were carried out batchwise
either manually or by using an Applied Biosystem, Inc. Model
431A automated peptide synthesizer.
Crude and purified peptides were analyzed on precoated
silica gel 6 F254 0.25 mm plates (Merck) using (A) CHCl3/
MeOH/AcOH, 12/1/1; (B) EtOAc/BuOH/AcOH/water, 1/1/1/1;
(C) n-BuOH/AcOH/H2O, 4/1/1. Analytical RP-HPLC was
performed on a Vydac C18 analytical (0.46 × 25 cm) column
using 0.1% TFA in acetonitrile-water as the eluant. A linear
gradient from 10 to 90% acetonitrile over 40 min, with a flow
rate of 1.2 mL/min, was employed. For purification of peptides,
a Vydac C18 semipreparative (1.0 × 25 cm) column was used
with the solvent system described above.
S
C b z-c [Ala L P h e -D -T r p (F o r )L y s (T o s )T h r (B zl)Ala L ]-
OMe (3). BocThr(Bzl)-oxime resin (5.4 g, substitution level:
0.26 mmol/g based on picric acid titration) was deprotected
with 25% TFA/DCM (v/v) and neutralized by 5% DIEA/DCM
(v/v). The peptide chain was then assembled by consecutive
addition of 2.5 equiv of BOP esters of BocLys(Tos)OH, Boc-D-
Trp(For)OH, BocPheOH, and ZAlaL(TrtAlaLOMe)OH followed
by deprotection steps. After the Trt group was removed with
25% TFA/DCM (v/v), the peptidyl resin was washed and
neutralized according to the standard oxime resin protocol. The
cyclization reaction was carried out by shaking the peptidyl
resin in DMF-DCM (100 mL, 2/1, v/v) in the presence of 10
equiv of acetic acid at room temperature for 24 h. The cyclic
peptide product was obtained from the filtrate of the reaction
mixture: the solvent was removed under reduced pressure and
the product was precipitated from a mixture of DMF-water.
Yield: 1.07 g (65%). Rf(A): 0.73. RP-HPLC: tR ) 34.0 min
(standard analytical conditions described above). FAB-MS:
m/ z 1173 [MH+], theoretically 1173.4, 1H-NMR (DMSO-d6):
δ (ppm) 0.99 (m, 2H, 2Hγ-Lys5), 1.08 (d, 3H, Me-Thr6, J ) 5.9
Hz), 1.24 (m, 2H, 2Hδ-Lys5), 1.41 (m, 1H, Hâ1-Lys5), 1.59 (br,
(br, 1H, Hâ1-Lys5), 2.36 (s, 3H, Me-Tos), 2.51-2.66 (m, 4H, Hâ2
-
2
7
AlaL and Hâ2-AlaL and 2Hꢀ-Lys5), 2.85-3.04 (m, 4H, Hâ1-D-
Trp4 and Hâ1-AlaL7 and Hâ1-Phe3 and Hâ1-D-Phe1), 3.39 (m, 1H,
Hâ2-Thr8), 3.51 (m, 1H, Hâ1-Thr8), 3.74 (m, 1H, Hâ-Thr8), 3.82
(m, 1H, HR-Thr8), 3.88 (m, 1H, HR-Lys5), 4.03 (m, 1H, Hâ-Thr6),
4.19 (dd, 1H, HR-Thr6), 4.31-4.45 (m, 2H, HR-D-Phe1 and HR-
7
D-Trp4), 4.65-4.76 (m, 4H, HR-AlaL2 and HR-Phe3 and HR-AlaL
and Thr8-ol(OH)), 4.90 (s, 2H, Cbz(CH2)), 4.94 (d, 1H, Thr6-
OH), 6.90-7.34 (m, 21H, NH-Thr6 and NH-D-Phe1 and H2,5,6,7
-
D-Trp(indole) and C6H5-Phe3 and Cbz(C6H5)), 7.36 (d, 2H,
Hortho-Tos, J ) 8.6 Hz), 7.39 (t, 1H, NHꢀ-Lys, J ) 6.05 Hz),

Lanthionine-Somatostatin Analogs
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 14 2249
7.44 (d, 1H, NH-Thr8-ol, J ) 9.5 Hz), 7.50 (d, 1H, H4-D-Trp4-
(indole), J ) 8.6 Hz), 7.65 (d, 2H, Hmeta-Tos, J ) 8.6 Hz), 7.93
(d, 1H, NH-AlaL7, J ) 8.6 Hz), 8.03 (d, 1H, NH-Phe3, J ) 6.9
Hz), 8.35 (d, 1H, NH-AlaL2, J ) 8.6 Hz), 8.40 (d, 1H, NH-Trp4,
J ) 6.91 Hz), 8.44 (d, 1H, NH-Lys5, J ) 6.9 Hz), 10.74 (s, 1H,
H1-(indole)-D-Trp4).
material was purified by gel permeation chromatography
(Sephadex G-10) and then by HPLC separation (on a Vydac
C-18 semipreparative column eluted with isochratic 24%
acetonitrile/water containing 0.1% TFA). Yield: 6.5 mg
(12.2%). Rf(C) ) 0.30. RP-HPLC: tR ) 14.5 min (isochratic
24% acetonitrile/water). FAB-MS: m/ z 914 [MH+] and 936
[MNa+], theoretically 914.4.
S
S
Cbz-D-P h e-c[AlaLP h e-D-Tr p(For )Lys(Tos)Th r (Bzl)AlaL]-
Th r -NH2 (7). The peptide hydrazide (5, 180 mg, 0.15 mmol)
was converted to azide as described above, and then HCl‚H-
Thr-NH2 (0.139 g, 0.9 mmol) and DIEA (0.195 mL, 1.12 mmol)
were added to the reaction mixture. After the reaction mixture
was stirred for 1 h at -15 °C and 48 h at 4 °C, the product
was obtained as described in the synthesis of compound 6.
Yield: 188 mg (97.2%). Rf(A): 0.34. RP-HPLC: tR ) 32.2 min
(standard analytical conditions described above). FAB-MS:
m/ z 1290 [MH+] and 1312 [MNa+], theoretically 1288.5.
S
H-D-P h e-Ala L-P h e-D-Tr p -Lys-Th r -Ala L-Th r -ol (1). The
protected octapeptide (6, 30.0 mg, 0.022 mmol) was deprotected
with sodium (15 mg) in liquid ammonia (20 mL) according to
the protocol described for compound 8. The crude material
was purified by gel permeation chromatography (Sephadex
G-10) and then by HPLC separation (on
a Vydac C-18
semipreparative column eluted with isochratic 24% acetoni-
trile/water containing 0.1 % TFA). Yield: 13.4 mg (62%).
RP-HPLC: tR ) 16.4 min (standard analytical conditions
described above). FAB-MS: m/ z 987.1 [MH+], theoretically
987.5. NMR data are published in our accompanying paper.44
S
H-Ala L-P h e-D-Tr p -Lys-Th r -Ala L-OH (8). The protected
hexapeptide (3, 117.3 mg, 0.1 mmol) was dissolved in hexam-
ethylphosphoramide (2.5 mL) and cooled to 15 °C, and 0.5 N
NaOH (1.0 mL) was added dropwise over 30 min. The reaction
mixture was stirred for further 30 min at this temperature,
then diluted with water (25 mL), and acidified to pH 2.5 with
2.5 N HCl at 0 °C. After being kept for 1 h at this temperature,
the product was filtered, dried, and recrystallized from DMF-
H-D-P h e-AlaL-P h e-D-Tr p-Lys-Th r -AlaL-Th r -NH2 (2). The
protected octapeptide amide (7, 30.0 mg, 0.023 mmol) was
deprotected by sodium (15 mg) in liquid ammonia (20 mL).
After removal of the ammonia the residue was desalted by gel
permeation chromatography (1.5 × 75 cm Sephadex G-10
eluted with 10% acetic acid) followed by RP-HPLC purification
on a Vydac C-18 semipreparative column. The pure peptide
fractions were pooled and lyophilized. Yield: 15.7 mg (68%).
RP-HPLC: tR ) 16.2 min (standard analytical conditions
described above). FAB-MS: m/ z 1000.5 [MH+], theoretically
1000.5. NMR data are published in our accompanying paper.44
H-c[CysP h e-D-Tr p LysTh r Cys]-OH (12) a n d H-D-P h e-
c[CysP h e-D-Tr p LysTh r Cys]OH (13). The synthesis started
from Fmoc-Cys(Trt)-HMPB resin (0.75 g, 0.3 mmol/g substitu-
tion level). Assembly protocol was as follows: deprotection
(20% piperidine/NMP, 20 + 2 × 2.5 min), wash (NMP), single
coupling (4 equiv of HBTU-HOBt-DIEA, 20 min), wash
(NMP). In successive couplings, FmocThr(tBu)OH, FmocLys-
(Boc)OH, Fmoc-D-Trp(Boc)OH, FmocPheOH, FmocCys(Trt)OH,
and Boc-D-PheOH were used. Cyclization: Both resin-bound
peptides (hexamer 14 and heptamer 15) were treated with 10
equiv of I2/NMP for 1 h, then washed (NMP, DCM, EtOH),
and dried. Cleavage-deprotection: Samples (0.5 g) of both
substances were subjected to final deprotection. Hexamer 14
was treated first with 20% piperidine/DMF for 25 min, and
then both peptidyl resins were treated with Reagent-K (3 mL)
at 0 °C for 1 h. The resins were filtered and washed with TFA.
The crude products were isolated from the liquid phases by
ethereal precipitation and centrifugation. The peptides were
purified by RP-HPLC on a Vydac C-18 semipreparative column
(eluent: acetonitrile-water mixture containing 0.1% TFA;
conditions: linear gradient from 18 to 24% of acetonitrile
during 20 min).
ether. Yield: 110 mg (97%). Rf(A) ) 0.32. RP-HPLC: tR
)
13.7 min (column and eluent above; conditions: linear gradient
from 50 to 75% of acetonitrile during 20 min). FAB-MS: m/ z
1131 [MH+], theoretically 1131.4.
This material was deprotected with sodium (50 mg) in liquid
ammonia (75 mL). After removal of the ammonia, the residue
was dissolved in water (2 mL) and the pH was adjusted to 5
with acetic acid. The solution was subjected to gel permeation
chromatography (1.5 × 75 cm Sephadex G-10 eluted with 10%
acetic acid) followed by RP-HPLC purification on a Vydac C-18
semipreparative column eluted with isochratic 22% acetoni-
trile/water containing 0.1% TFA. The pure peptide fractions
were pooled and lyophilized. Yield: 38 mg (51%). RP-HPLC:
tR ) 14.5 min (standard analytical conditions described above).
FAB-MS: m/ z 752.8 [MH+], theoretically 753.3. NMR data
are published in our accompanying paper.44
S
H-D-P h e-Ala L-P h e-D-Tr p -Lys-Th r -Ala L-OH (9) a n d H-D-
S
P h e-Ala L-P h e-D-Tr p -Lys-Th r -D-Ala L-OH (10). The pro-
tected heptapeptide (4, 90.0 mg, 0.073 mmol) was dissolved
in hexamethylphosphoramide (1.8 mL) and cooled to 15 °C,
and 0.5 N NaOH (0.75 mL) was added dropwise over 30 min.
The reaction mixture was stirred for further 30 min at this
temperature, then diluted with water (20 mL), and acidified
to pH 2.5 with 2.5 N HCl at 0 °C. After being kept for 1 h at
this temperature, the product was filtered, dried, and recrys-
tallized from DMF-ether. Yield: 77.2 mg (89%). Rf(A): 0.22.
FAB-MS: m/ z 1189 [MH+], theoretically 1188.5.
This material was deprotected with sodium (35 mg) in liquid
ammonia (50 mL). After removal of the ammonia, the rest
was dissolved in water (2 mL) and the pH was adjusted to 5
with acetic acid. The solution was subjected to gel permeation
chromatography (1.5 × 75 cm Sephadex G-10 eluted with 10%
acetic acid) followed by RP-HPLC purification on a Vydac C-18
semipreparative column eluted with isochratic 24% acetoni-
trile/water containing 0.1% TFA. The two peptide fractions
pooled separately and lyophilized showed identical mass
spectra.
Hexa p ep tid e 12. Yield: 16 mg (17.5%). RP-HPLC: tR
)
14.6 min (standard analytical conditions described above).
FAB-MS: m/ z 785.3 [MH+], theoretically 785.3.
Hep ta p ep tid e 13. Yield: 20 mg (17.1%). RP-HPLC: tR
) 16.5 min (standard analytical conditions described above).
FAB-MS: m/ z 932.4 [MH+], theoretically 932.4. NMR data
are published in our accompanying paper.44
Ma ss Sp ectr om etr y. LSIMS and high-energy CID spectra
were recorded on a Kratos Concept IIHH four sector tandem
mass spectrometer equipped with a charge-coupled device
(CCD) detector and a continuous flow probe for sample
introduction. A mixture of 5% acetonitrile, 5% thioglycerol,
0.1% TFA in water was used as a liquid matrix.46,47 For the
high-energy CID experiments, He was used as the collision
gas, and its pressure was adjusted to attenuate the precursor
ion abundance to 30% of its initial value. The collision energy
was 4 keV.
Sa m p le P r ep a r a tion s. About 0.1 mg of the sample was
dissolved in 50 µL of 20% acetonitrile/0.1% TFA/H2O. A 2 µL
aliquot of the sample solutions was mixed with 20 µL of liquid
matrix (5% acetonitrile/5% thioglycerol/0.1% TFA in water).
Then 5 µL of these solutions was injected for the LSIMS. For
the high-energy CID experiments, 15 µL of the original peptide
Com p ou n d 9. Yield: 17.1 mg (23.4%). Rf(C): 0.40. RP-
HPLC: tR ) 16.9 min (standard analytical conditions described
above). FAB-MS: m/ z 900.1 [MH+], theoretically 900.4.
Com p ou n d 10. Yield: 12.8 mg (13.7%). RP-HPLC: tR
)
16.1 min (standard analytical conditions described above).
NMR data are published in our accompanying paper.44
S
H-D-P h e-Ala L-P h e-D-Tr p -Lys-Th r -Ala L-N2H3 (11). The
protected heptapeptide hydrazide (5, 70.0 mg, 0.058 mmol) was
deprotected with sodium (75 mg) in liquid ammonia (50 mL)
according to the protocol described for compound 8. The crude

2250 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 14
O¨ sapay et al.
solutions was dried, redissolved in 18 µL of 100 mM NH4HCO3
buffer, and incubated with chymotrypsin overnight. Two
microliters of the digests was diluted with the liquid matrix
and analyzed by LSIMS and high-energy CID as described
above.
tapeptide (3 pages). Ordering information is given on any
current masthead page.
Refer en ces
(1) Brazeau, P.; Vale, W.; Burgus, R.; Ling, N.; Butcher, M.; Rivier,
J .; Guillemin, R. Hypothalami polypeptide that inhibits the
secretion of immunoreactive pituitary growth hormone. Science
1973, 179, 77-79.
(2) Pless, J .; Bauer, W.; Briner, U. Chemistry and pharmacology of
SMS 202-995, a long-acting octapeptide analogue of somatosta-
tin. Scand. J . Gastroenterol. 1986, 21, Suppl. 119, 65-72.
(3) Epelbaum, J . Somatostatin in the central nervous system:
physiology and pathological modifications. Prog. Neurobiol.
1986, 27, 63-100.
(4) Bloom, S. R.; Polak, J . M. Somatostatin. Br. Med. J . 1987, 295,
288-290.
(5) Yamada, Y.; Post, S.; Wang, K.; Tager, H.; Bell, G. I.; Seino, S.
Cloning and functional characterization of a family of human
and mouse somatostatin receptors expressed in brain, gas-
trointestinal tract and kidney. Proc. Natl. Acad. Sci. U.S.A.
1992, 89, 251-255.
(6) Raynor, K.; O’Carroll, A.; Kong, H.; Yasuda, K.; Mahan, L. C.;
Bell, G. I.; Reisine, T. Characterization of Cloned Somatostatin
Receptors SSTR4 and SSTR5. Mol. Pharmacol. 1993, 44, 385-
392.
En zym a tic Degr a d a tion Stu d ies. Approximately 100
nmol of peptide was added to 1 mL of rat brain homogenate
(10% w/w, in artificial cerebrospinal fluid, pH 7.4), and the
mixture was incubated at 37 °C in a temperature-controlled,
shaking water bath. Aliquots (100 µL) were removed after 5,
15, 30, 45, 60, and 90 min of incubation, respectively, and
transferred to 1.5 mL plastic centrifuge tubes containing 200
µL of ice-cold 1 M aqueous acetic acid. After centrifugation
at 12500g for 15 min, the supernatant was removed and
analyzed by microbore high-performance liquid chromatogra-
phy. Analyses were done on a 1 mm × 300 mm i.d. Supelcosil
LC-18 (5 µm) reversed-phase column (Supelco, Bellefonte, PA),
at 216 nm. A mobile phase of water-acetonitrile containing
0.1% TFA was used in a gradient elution. HPLC calibration
curves for determining concentrations were obtained by adding
known amount of peptide into aliquots of brain homogenate
transferred into ice-cold 1 M acetic acid solution and analyzing
the supernatant after centrifugation.
(7) Raynor, K.; Murphy, W. A.; Coy, D. H.; Taylor, J . E.; Moreau, J .
P.; Yasuda, K.; Bell, G. I.; Reisine, T. Cloned Somatostatin
Receptors: Identification of Subtype-Selective Peptides and
Demonstration of High Affinity Binding of Linear Peptides. Mol.
Pharmacol. 1993, 43, 838-844.
(8) Rossowski, W.; Coy, D. H. Potent inhibitory effects of type four
receptor selective somatostatin analog on rat insulin release.
Biochem. Biophys. Res. Commun. 1993, 197, 366-371.
(9) Rossowski, W.; Coy, D. H. Specific inhibition of ratpancreatic
insulin and glucagon release by receptor-selective somatostatin
analogs. Biochem. Biophys. Res. Commun. 1994, 205, 341-346.
(10) Rivier, J .; Brown, M.; Vale, W. D-Trp8-somatostatin, analog of
somatostatin more potent than the native molecule. Biochem.
Biophys. Res. Commun. 1975, 65, 746-751.
(11) Veber, D. F.; Holly, F. W.; Nutt, R. F.; Bergstrand, S. J .; Brady,
S. F.; Hirschmann, R.; Glitzer, M. S.; Saperstein, R. Highly active
cyclic and bicyclic somatostatin analogs of reduced ring size.
Nature 1979, 280, 512-514.
(12) Veber, D. F.; Freidinger, R. M.; Perlow, D. S.; Palevada, W. R.
J .; Holly, F. W.; Strachan, R. G.; Nutt, R. F.; Arison, B. J .;
Homnick, C.; Randall, W. C.; Glitzer, M. S.; Saperstein, R.;
Hirschmann, R. A potent cyclic hexapeptide analogue of soma-
tostatin. Nature 1981, 292, 55-58.
(13) Vale, W. W., J r.; Vaughan, J .; Yamamoto, G.; Bruhn, T. O.;
Douglas, C.; Dalton, D.; Rivier, C. L.; Rivier, J . Assay of
corticotropin releasing factor. Methods Enzymol. 1983, 103,
565-577.
(14) Nutt, R. F.; Veber, D. F.; Curley, P. E.; Saperstein, R.; Hir-
schmann, R. F. Somatostatin analogs which define the role of
the lysine-9 amino group. Int. J . Pept. Protein Res. 1983, 21,
66-73.
(15) Freidinger, R. M.; Perlow, D. S.; Randall, W. C.; Saperstein, R.;
Arison, B. H.; Veber, D. F. Conformational modifications of cyclic
hexapeptide somatostatin analogs. Int. J . Pept. Protein Res.
1984, 23, 142-150.
(16) Mierke, D. F.; Pattorni, C.; Deleat, N.; Toy, A.; Goodman, M.;
Tancredi, T.; Motta, A.; Temussi, P. A.; Moroder, L.; Bovermann,
G.; Wu¨nsch, E. Cyclic hexapeptides related to somatostatin -
Conformational analysis employing 1H NMR and molecular
dynamics. Int. J . Pept. Protein Res. 1990, 36, 418-432.
(17) Freidinger, R. M.; Veber, D. F. Conformationally directed drug
design. ACS Symposium Series, No. 251 1984, 169-187.
(18) Bauer, W.; Briner, U.; Doepfner, W.; Haller, R.; Huguenin, R.;
Marbach, P.; Petcher, T.; Pless, J . SMS 201-995: a very potent
and selective octapeptide analogue of somatostatin with pro-
longed action. Life Sci. 1982, 31, 1133-1140.
(19) Wynants, C.; van Binst, G.; Loosli, H. R. Hexapeptide analogue
of somatostatin, Sandoz 201-456. Int. J . Pept. Protein Res. 1985,
25, 622-627.
(20) Wynants, C.; van Binst, G.; Loosli, H. R. SMS 201-995, an
octapeptide somatostatin analogue. Int. J . Pept. Protein Res.
1985, 25, 615-621.
(21) The lanthionine sandostatin has already been noted in the
literature34 as a minor side product from the production of
sandostatin.
(22) Polinsky, A.; Cooney, M. G.; Toy-Palmer, A.; O¨ sapay, G.;
Goodman, M. Synthesis and conformational properties of the
lanthionine-bridged opioid peptide [D-AlaL2, D-AlaL5]enkephalin
as Determined by NMR and Computer Simulations. J . Med.
Chem. 1992, 35, 4185-4194.
In Vivo Meta bolism of Sa n d osta tin An a logs. Sprague-
Dawley rats (250-300 g body weight) were used to study in
vivo metabolism in brain tissue.43,48 A guide cannula was
implanted stereotaxically under pentobarbital anesthesia into
the striatum (anterior-posterior +2.4 mm and medial-lateral
+3.0 mm with reference to bregma and sagittal suture,
respectively, and dorsal-ventral -6.0 mm in depth from the
surface of the cortex). After 5-7 days of recovery from surgery,
a
CMA/12 microdialysis probe fitted with a 4 mm long
polycarbonate membrane (BAS, West Lafayette, IN) was
inserted through the guide cannula, and a 200 µM solution of
the peptide in artificial cerebrospinal fluid was perfused
through the probe at a flow rate of 0.25 µL/min with a
microliter syringe pump (BAS, West Lafayette, IN). The
unanesthetized animal was placed in a plastic bowl, and
dialysate (10-20 µL) was collected. The effluent solution
containing the peptide and its in vivo metabolites was desalted
on a reversed-phase minicolumn (1 mL Supelclean LC-18,
Supelco, Bellefonte, PA) and analyzed by electrospray ioniza-
tion (ESI) mass spectrometry (Vectec/PerSeptive Biosystems,
Model 200 ES instrument).49,50
Abbreviations. The following abbreviations are used
in the paper: AlaL, one end of a lanthionine unit; Boc,
tert-butyloxycarbonyl; BOP, (benzotriazol-1-yloxy)tris-
(dimethylamino)phosphonium hexafluorophosphate; tBu,
tert butyl; Bzl, benzyl; Cbz, benzyloxycarbonyl; COSY-
DQF, double quantum-filtered correlation spectroscopy;
DCM, dichloromethane; DIEA, N,N-diisopropylethy-
lamine; DMF, N,N-dimethylformamide; DMSO-d6, hexa-
deuterated dimethyl sulfoxide; EDC, 1-ethyl-3-(3′-
(dimethylamino)propyl)carbodiimide; Fmoc, 9-fluoren-
ylmethoxycarbonyl; For, formyl; HMPA, hexamethyl-
phosphoramide; HOBt, N-hydroxybenzotriazole; HO-
HAHA, homonuclear Hartman-Hahn experiment; NMP,
N-methylpyrrolidone; ROESY, rotating frame nuclear
Overhauser enhancement spectroscopy; TFA, trifluoro-
acetic acid; PCOR, peptide cyclization on an oxime resin;
Tos, p-toluenesulfonyl; Trt, trityl or triphenylmethyl.
Ack n ow led gm en t. This research was supported by
the NIH-DK 15410 Grant, NIH-RR 01614, and NSF-
DIR 8700766 (Dr. A. L. Burlingame). The authors
thank F. C. Walls for performing the high-energy CID
experiments. We thank Mr. J . Taulane for his careful
technical assistance.
Su p p or tin g In for m a tion Ava ila ble: High-energy CID
spectrum of the chymotrypsin-digested disulfide-bridged hep-

Lanthionine-Somatostatin Analogs
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 14 2251
(23) O¨ sapay, G.; Wang, S.; Shao, H.; Goodman, M. Synthesis and
characteristics of lanthionine peptides. In Peptide Chemistry
1992. Proceedings of the 2nd J apan Symposium on Peptide
Chemistry; Yanaihara, N., Ed.; ESCOM: Leiden, 1993; pp 152-
154.
(24) Fukase, K.; Kitazawa, M.; Sano, A.; Shimbo, K.; Horimoto, S.;
Fujita, H.; Kubo, A.; Wakamiya, T.; Shiba, T. Synthetic study
on peptide antibiotic nisin 5. Total synthesis of nisin. Bull.
Chem. Soc. J pn. 1992, 65, 2227-2240.
(25) O¨ sapay, G.; Shao, H.; Goodman, M.; Taylor, J . W. Peptide
cyclization on an oxime resin (The PCOR method): Reaction
conditions and applications. In Peptides: Chemistry, Structure
and Biology, Proceedings of the Thirteenth American Peptide
Symposium; Hodges, R. S., Smith, F. A., Eds.; Escom: Leiden,
1994; pp 101-103.
(26) O¨ sapay, G.; Bouvier, M.; Taylor, J . W. Peptide cyclization on
oxime resin (PCOR). In Techniques in Protein Chemistry;
Villafranca, J . J ., Ed.; Academic Press: San Diego, 1991; Vol.
II, pp 221-231.
(27) O¨ sapay, G.; Goodman, M. New application of peptide cyclization
on an oxime resin (the PCOR method): preparation of lanthion-
ine-peptides. J . Chem. Soc., Chem. Commun. 1993, 1599-1600.
(28) Gisin, B. F. Monitoring of reactions in solid phase peptide
synthesis with picric acid. Anal. Chim. Acta 1972, 58, 248-
249.
(29) Kaiser, E.; Colescott, R. L.; Bossinger, C. D.; Cook, P. I. Color
test for detection of free terminal amino groups in the solid-phase
synthesis of peptides. Anal. Biochem. 1970, 34, 595-598.
(30) Kamber, B.; Hartmann, A.; Eisler, K.; Riniker, B.; Rink, H.;
Sieber, P.; Rittel, W. The synthesis of cystine peptides by iodine
oxidation of S-trityl-cysteine and S-acetamidomethyl-cysteine
peptides. Helv. Chim. Acta 1980, 63, 899-915.
(31) Biemann, K.; Scoble, H. A. Characterization by tandem mass
spectrometry of structural modifications in proteins. Science
1987, 237, 992-998.
(32) Martin, S. A.; Biemann, K. A comparison of keV atom bombard-
ment mass spectra of peptides obtained with a two-sector mass
spectrometer with those from a four-sector tandem mass spec-
trometer. Int. J . Mass Spectrom. Ion Processes 1987, 78, 213-
228.
(33) Medzihradszky, K. F.; Burlingame, A. L. The advantages and
versatility of a light energy collision induced dissociation based
strategy for the sequence and structural determination of
proteins. Companion Methods Enzymol. 1994, 6, 284-303.
(34) Bean, M. F.; Carr, S. A.; Escher, E.; Neugebauer, W.; Samanen,
J . Identification of a thioether by-product in the synthesis of a
cyclic disulfide peptide by tandem mass spectrometry. In
Peptides: Chemistry, Structure and Biology, Proceedings of the
Eleventh American Peptide Symposium; Rivier, J . E., Marshall,
G. R., Eds.; ESCOM: Leiden, 1990; pp 443-445.
(35) Falick, A. M.; Hines, W. M.; Medzihradszky, K. F.; Baldwin, M.
A.; Gibson, B. W. Low-mass ions produced from peptides by high-
energy collision-induced dissociation in tandem mass spectrom-
etry. J . Am. Soc. Mass Spectrom. 1993, 4, 882-893.
(36) Biemann, K. Appendix 5. Nomenclature for peptide fragment
ions (positive ions). Methods Enzymol. 1990, 193, 886-887.
(37) Rens-Domiano, S.; Law, S.; Yamada, Y.; Seino, S.; Bell, G.;
Reisine, T. Pharmacological properties of two cloned somatosta-
tin receptors. Mol. Pharmacol. 1992, 42, 28-34.
(38) Yasuda, K.; Rens-Domiano, S.; Breder, C.; Law, S.; Saper, C.;
Reisine, T.; Bell, G. Cloning of a novel somatostatin receptor,
SSTR3, coupled to adenylyl cyclase. J . Biol. Chem. 1992, 267,
20422-20428.
(39) O¨ sapay, G.; Melacini, G.; Zhu, Q.; Tehrani, L.; Goodman, M.
Synthesis and physicochemical characterization of the lanthion-
ine analog of somatostatin[1-14]. Lett. Pept. Sci. 1994, 1, 81-
87.
(40) O’Carroll, A-M.; Raynor, K.; Lolait, S. J .; Reisine, T. Charac-
terization of cloned human somatostatin receptor SSTR5. Mol.
Pharmacol. 1994, 46, 291-298.
(41) Dixon, D. M.; Traynor, J . R. Formation of [Leu5]enkephalin from
dynorphin A(1-8) by rat central nervous tissue in vitro. J .
Neurochem. 1990, 54, 1379-1385.
(42) Robson, L. E.; Gilan, M. G. C.; McKnight, A. T.; Kosterlitz, M.
W. [H3]-Dynorphin A(1-9) binding characteristics and degrada-
tion profile in brain homogenates. Life Sci. 1983, 33 (Suppl. 1),
283-286.
(43) Andre´n, P. E.; Caprioli, R. M. In Vivo metabolism of substance
P in rat striatum utilizing microdialysis/liquid chromatography/
micro-electrospray mass spectrometry. J . Mass Spectrom. 1995,
30, 817-824.
(44) Melacini, G.; Zhu, Q.; O¨ sapay, G.; Goodman, M. A refined model
for the somatostatin pharmacophore: conformational analysis
of lanthionine-sandostatin analogs. J . Med. Chem. 1997, 40,
2252-2258.
(45) DeGrado, W. F.; Kaiser, E. T. Polymer-bound oxime esters as
supports for solid-phase peptide synthesis. Preparation of
protected peptide fragments. J . Org. Chem. 1980, 45, 1295-
1300.
(46) Walls, F. C.; Hall, S. C.; Medzihradszky, K. F.; Burlingame, A.
L. 41st ASMS Conference on Mass Spectrometry and Allied
Topics, 1993.
(47) Burlingame, A. L. Mass spectrometry. In Biological Mass
Spectrometry: Present and Future; Matsuo, T., Caprioli, R. M.,
Gross, M. L., Seyama, Y., Eds.; J ohn Wiley & Sons: Chichester,
1994; pp 147-164.
(48) All surgery procedures and protocols were consistent with the
National Institutes of Health Guidelines in Guide for the Care
of Laboratory Animals and the Federal Animal Welfare Act, and
were approved by the University of Florida Institutional Animal
Care and Use Committee.
(49) Bodor, N.; Prokai, L.; Wu, W.-M.; Farag, H.; J onalagadda, S.;
Kawamura, M.; Simpkins, J . A strategy for delivering peptides
into the central nervous system by sequential metabolism.
Science 1992, 257, 1698-1700.
(50) Prokai, L.; Bodor, M. Chemical delivery system to transport a
pyroglutamyl peptide amide to the central nervous system. J .
Am. Chem. Soc. 1994, 116, 2643-2644.
J M960850I