Concept and synthetic approach for a kilogram scale synthesis of octa-D-arginine amide nonahydrochloride salt

Article abstract of DOI:10.1016/j.tet.2005.07.037

Oligomers of arginine, such as octa-D-arginine amide, are excellent transporters for active drugs through cell membranes and tissue. The synthesis of octa-D-arginine amide, as the nonahydrochloride salt, was approached via a solution phase synthetic route

Full text of DOI:10.1016/j.tet.2005.07.037

Tetrahedron 61 (2005) 8821–8829
Concept and synthetic approach for a kilogram scale synthesis of
octa-D-arginine amide nonahydrochloride salt
*
J. C. Califano, A. Uribe, J. L. Chang, C. L. Becker, J. J. Napier, V. Kishore, D. Zhou, G. Love,
K. Gernhardt and J. C. Tolle
Abbott Laboratories, GPRD Process Research and Development, Dept R-450, Bldg R13/4, 1401 Sheridan Road, North-Chicago, IL,
60064-6290, USA
Received 4 April 2005; revised 1 July 2005; accepted 8 July 2005
Available online 28 July 2005
Abstract—Oligomers of arginine, such as octa-D-arginine amide, are excellent transporters for active drugs through cell membranes and
tissue. The synthesis of octa-D-arginine amide, as the nonahydrochloride salt, was approached via a solution phase synthetic route involving
the preparation of an octa-D-ornithine intermediate, which was then converted into the desired octa-D-arginine compound through a
guanidinylation step. The multi-step synthesis was carried out at pilot scale, resulting in the preparation of 700 g of the target molecule. No
chromatographic purification was needed at any step of the process.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
once the peptide is cleaved from the resin. A solution
phase synthesis appeared to be a more attractive
approach for the preparation of our target molecule,
octa-D-arginine amide nonahydrochloride salt2–11 (6)
(Scheme 1) in terms of robustness, scalability, and the
potential to minimize labor-intensive purification. The
limited availability and cost of protected D-arginine
derivatives led us to explore the development of a
synthesis in which D-ornithine is substituted for the
D-arginine residues. A per-guanidinylation step would then
Timing and development of scalable processes are
critical to the successful production of a peptide-based
drug.1 Although solid phase synthesis is a common
approach used to prepare oligopeptides, this method can
have limitations, such as the high cost of excess
reagents and protected amino acids, the high volume
of solvents used, potential limited scalability, and the
usual requirement for a chromatographic purification
Scheme 1. Octa-D-arginine amide hydrochloride.
Keywords: Amino acids and derivatives; Arginine; Guanidinylation; Peptide transport; Peptide process development; Minimal isolation peptide synthesis.
*
Corresponding author. Tel.: C1 847 935 0590; fax: C1 847 938 0850; e-mail: jean.califano@abbott.com
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.07.037

8822
J. C. Califano et al. / Tetrahedron 61 (2005) 8821–8829
be used for conversion of the 8 ornithine residues to the
corresponding arginine residues.12–18
treatment of the desired octa-D-arginine amide hydrochlo-
ride salt (6) in water was sufficient for purification.
The synthetic process, as described in Scheme 2, resulted in
the preparation of 0.7 kg of the desired compound, isolated
as a hydrochloride salt (6). The main features of the process
were the synthesis of a key crystalline tetrapeptide ester
intermediate (1) with only one isolation, followed by the
preparation of two crystalline tetrapeptide intermediates (2)
and (3) from the common tetrapeptide ester (1). Coupling of
the N-terminal deprotected (2) with C-terminal deprotected
(3) gave the fully protected octa-D-ornithine intermediate
(4), which upon global removal of 8 amine-protecting
groups followed by global guanidinylation of the free
amines in one step gave the desired D-arginine residues. No
chromatographic purification was used at any step of the
process. At the final step a simple activated carbon
2. Results and discussion
In step 1, Z-[D-Orn(Boc)]4-OMe (1) (Scheme 3), was
prepared following a continuous process (MIPS, minimal
isolation peptide synthesis) to minimize the isolation steps
of the intermediate dipeptide and tripeptide.
The general procedure for conducting the MIPS process is
described in Scheme 4 below. Excess N-benzyloxycarbo-
nyl-protected amino acid N-hydroxysuccinimide ester19
(Z-AA-OSu) is used to drive the acylation to completion.
Residual Z-AA-OSu is captured/scavenged by the addition
of amine-derivatized insoluble resin or silicagel,20,21 which
Scheme 2. Octa-D-arginine amide—overall process.

J. C. Califano et al. / Tetrahedron 61 (2005) 8821–8829
8823
Scheme 3. M.I.P.S process step 1. Z-[D-Orn(Boc)4-OMe.
Scheme 4. MIPS process cycle.
is then removed from the reaction mixture by filtration.
Subsequent N-deprotection of the soluble peptide by
hydrogenolysis completes one MIPS cycle. The process
can be repeated (nK1) times for a peptide of (n) residues in
length (Scheme 4).
Orn(Boc)]4-OMe (1) was converted to the corresponding
C-terminal amide, Z-[D-Orn(Boc)]4-NH2 (2), via treatment
with a solution of anhydrous ammonia in methanol. The
ammonolysis was performed at a temperature of no more
than 10 8C in order to minimize racemization and impurity
formation, which were observed at room temperature.
Following distillation of the ammonia, the material was
isolated by crystallization from methanol/isopropyl acetate/
heptane.
In the present case, the hydrochloride salt of H-D-Orn(Boc)-
OMe (a1) was coupled with a 5% molar excess of the
preactivated N-hydroxysuccinimide ester Z-D-Orn(Boc)-
OSu (b1), in N-methylpyrrolidinone in the presence of
DIEA, to form Z-[D-Orn(Boc)]2-OMe (c1). Once the
coupling was complete, the excess of Z-D-Orn(Boc)-OSu
was scavenged with an aminosilica gel, which was then
removed by filtration. The remaining solution was then
hydrogenated in the presence of p-toluenesulfonic acid,
giving H-[D-Orn(Boc)]2-OMe$pTSA (d1). The same pro-
cess described above was repeated to obtain the tripeptide
intermediate, Z-[D-Orn(Boc)]3-OMe (e1). The solution was
carried through an additional hydrogenolysis/coupling
reaction. No scavenging reaction was necessary for removal
of the excess active ester at this stage. The NMP solution of
the fully-protected tetrapeptide (1) was diluted with
isopropyl acetate and the resulting mixture was washed
with aqueous solutions, followed by distillation of the
organic layer. The fully-protected tetrapeptide was then
crystallized from methanol/isopropylacetate/heptane.
Saponification of the second half of step 1 material led to
Z-[D-Orn(Boc)]4-OH (3) (step 3; Scheme 2). The saponifica-
tion was carried out with lithium hydroxide in a mixture of
tetrahydrofuran and water. The temperature was maintained
at no more than 10 8C to control racemization of the
C-terminal residue, which was observed at room tempera-
ture. The reaction was then quenched by adjusting to pH 4
with aqueous HCl solution. The product was extracted
into isopropyl acetate. After a series of aqueous washes,
the organic layer was distilled and Z-[D-Orn(Boc)]4-OH
(3) was crystallized from methanol/isopropylacetate/
heptane.
These conditions of saponification were chosen in order to
avoid pre-mature removal of the benzyloxycarbonyl
protecting group under basic aqueous conditions. A
summary of the different conditions tried during lab studies
is reported in Table 1.
To proceed to step 2 (Scheme 2), half of step 1, Z-[D-

8824
J. C. Califano et al. / Tetrahedron 61 (2005) 8821–8829
Table 1. Conditions examined for removal of methyl ester
Saponification
Temperature (8C)
Z-[D-Orn(Boc)]4-OH HPLC peak
area (%)
NMP/NaOH (1 N) (1.3 equiv)
MeOH/NaOH (1 N) (1.3 equiv)
NMP/LiOH (1 N) (1.3 equiv)
MeOH/LiOH (1 N) (1.3 equiv)
THF/LiOH (1 N) (1.3 equiv)
THF/LiOH (0.5 N) (1.5 equiv)
Not complete
Not complete
Complete (slow overnight)
Complete (slow overnight)
Complete (4 h)
0
0
0
0
0
10
35
95
35
95
99.7
99.9
Complete (4 h)
Catalytic hydrogenation of Z-[D-Orn(Boc)]4-NH2 (2) was
carried out in dimethylformamide in the presence of 5%
palladium on alumina (35 psi, 25 8C). Complete removal of
the benzyloxycarbonyl group went smoothly to afford a
dimethylformamide solution of the tetrapeptide H-[D-
Orn(Boc)]4-NH2.
reaction completion, isopropyl alcohol was added to
precipitate the product, Z-(D-Arg)8-NH2$8HCl (5).
While most of the diisopropylethylamine hydrochloride
salt (DIEA$HCl), formed during the guanidinylation
was removed, some of this salt (3.5 wt%) remained
trapped with the solids. Z-(D-Arg)8-NH2 (5) was
precipitated from a methanol solution by addition of
isopropanol to give product with trace (0.01 wt%)
DIEA$HCl.
The free amine solution was coupled with Z-[D-Orn(Boc)]4-OH
(3) in the presence of N-ethyl-N-(3-dimethylaminopropyl)-
carbodiimide hydrochloride (EDAc$HCl) and N-hydroxy-
5-norbornene-2,3-dicarboximide (HONB)22,23 to obtain the
fully protected octapeptide, Z-[D-Orn(Boc)]8-NH2 (4),
which precipitated from the reaction mixture by addition
of water. The crude, protected octapeptide was isolated by
filtration. The wet cake was redissolved in a mixture of
acetone and water and the desired product was precipitated
by additional water. This re-precipitation procedure resulted
in an increase in purity from 89.5 to 92.5%.
The intermediate at step 4, Z-[D-Orn(Boc)]8-NH2 (4),
was obtained with a purity of O92 pa%, as judged by
HPLC. Several impurities formed during step 5, that is
the synthesis of the protected octa-arginine amide Z-
(D-Arg)8-NH2 (5). The impurities are primarily attrib-
uted to the global-Boc removal step. A summary of
these impurities, which were then carried to the
guanidinylation step and indentified by LC/MS, is
reported below:
HONB was chosen as a coupling auxiliary because of the low
levels ofracemization(0.3%)thatwereobserved inlab studies
on this segment condensation. Other coupling conditions that
were attempted for the condensation of Z-[D-Orn(Boc)]4-OH
with H-[D-Orn(Boc)]4-NH2 are summarized in Table 2.
1. Partial removal of Z protecting group during the Boc
removal step. During the guanidinylation reaction step,
about 10% of final product, H-(D-Arg)8-NH2 (6), was
observed, as well an over-guanidinylated (a) side
product (0.5–1%) (Scheme 5).
Scheme 5. Over guanidinylated impurity.
The deprotection of Z-[D-Orn(Boc)]8-NH2 (4) was accom-
plished with HCl in ethyl acetate in the presence of
triisopropylsilane as a scavenger at ambient temperature.
This is a solid-to-solid reaction. After completion of the
reaction, the product was filtered and dried under a nitrogen
atmosphere.
2. During the removal of the Boc protecting groups, a
mixture of mono-t-butylated impurities was observed
(b). These impurities were then carried to the next step,
resulting in the formation of a partially guanidinylated
impurity (c) at the same level as (b). An investigation to
improve the Boc deprotection reaction conditions
resulted in a mixture of HCl/EtOAc/triisopropylsilane
showing the most promise. Under these conditions, the
level of t-butylated impurities was decreased from 8 to
2.5% (Scheme 6).
Guanidinylation of Z-D-(Orn)8-NH2$8HCl was achieved
with 1-pyrazole-1-carboxamidine hydrochloride salt5–17,18
and diisopropylethylamine in methanol–water. Upon
Table 2. Racemization study during segment condensation
Temperature (8C)
Racemization level of Z-DDDLDDDD-NH2 [DZD-Orn(Boc)- and
LZL-Orn(Boc)] (%)
DMF/EDAc in water/HOBt$H2O
DMF/EDAc/HOBt
DMF/EDAc/HOBt
DMF/EDAc/HOBt
DMF/EDAc in water/HONB
DMF/EDAc/HONB
K10
K10
0
Room temperature
10
10
1.1
0.7
0.9
3.1
0.3
0.3
The level of racemization was measured by HPLC.

J. C. Califano et al. / Tetrahedron 61 (2005) 8821–8829
8825
Hydrogenolysis of Z-(D-Arg)8-NH2 was carried out in H2O
with 10 wt% of 5% Pd/Al2O3 in the presence of 1 equiv
HCl. The crude product solution was subjected to two
carbon treatments to effect purification. Among the eight
types of carbon evaluated, L3S provided adequate purifi-
cation with best recovery. Laboratory experiments showed
that two carbon treatments (1.8!loading for each) gave
product at 94 pa% and an overall recovery of 65%. The
pyrazole$HCl by-product from the guanidinylation step was
also effectively removed. A summary of these experiments
is reported in Tables 3 and 4.
Scheme 6. tert-Butylation side reaction.
3. During the deprotection, lactam formation occurred at
the C-terminal ornithine residue (d), which led to the
formation of a partially guanidinylated impurity (e)
(Scheme 7).
The solution was then concentrated to a volume of 8 L by
reverse osmosis and the final product was isolated by
lyophilization.
3. Conclusion
A solution phase synthesis of H-(D-Arg)8-NH2$9HCl (6)
was developed and demonstrated at the kilogram scale. The
minimum isolation process for the preparation of tetrapep-
tide (1) was an efficient approach to this key intermediate.
Conversion of the fully protected octa-D-ornithine inter-
mediate (4) to the protected octa-D-arginine intermediate (5)
avoided the need to develop what would be a more costly
and complicated process if protected D-arginine was
the starting material. Given the need and opportunity, the
process as described herein could be used for the
preparation of larger quantities of octa-D-arginine amide
hydrochloride.
Scheme 7. Lactam formation.
4. Impurity (f) formed during the guanidinylation. For-
mation of this impurity is attributed to ammonia serving
as the leaving group in place of pyrazole. This
side reaction was minimized when the guanidinylation
reaction was carried out in methanol/water as the
reaction solvent rather than a N-methylpyrrolidinone/
water mixture (Scheme 8).
Scheme 8. Guanidinylation side reaction.
Table 3. Carbon treatment of H-(D-Arg)8-NH2$9HCla
H-(D-Arg)8-NH2$9HCl HPLC peak area (%)
H-(D-Arg)8-NH2$9HCl recovery (%)
L3S
5SC/G
ENO-PC
CPL
5% Pd/C
10% Pd/C
Darco G60
89.4
88.5
85.7
85.9
86.2
86.4
89.5
91
79
70
71
90
91
85
a Crude H-(D-Arg)8-NH2$9HCl at 77.2 peak area % was used for the study.

8826
J. C. Califano et al. / Tetrahedron 61 (2005) 8821–8829
Table 4. Carbon purification using L3Sa
Scavenging stage for Z-[D-Orn(Boc)]2-OMe (c1). To a
reactor was charged 1.2 kg of Si-amine (amine-derivatized
insoluble silicagel), then the coupling reaction mixture from
above was charged to the same reactor. N-Methylpyrroli-
dinone (5.0 kg) was used as a rinse. The mixture was stirred
at 45G5 8C. An in process sample was pulled after 12 h for
HPLC analysis. The disappearance of Z-D-Orn(Boc)-OSu
was monitored by HPLC. Analytical results indicated that
no starting material, Z-D-Orn(Boc)-OSu, was detected. The
temperature was adjusted to 20G5 8C and the reaction
mixture was filtered through a filter pot to a hydrogenolysis
vessel, which was previously charged with 1.5 kg of
p-toluenesulfonic acid and 0.25 kg of 5% palladium on
alumina. Two rinses with N-methylpyrrolidinone (4.5 kg)
were used, then the mixture carried to the next step.
HPLC peak
area (%)
Pyrazole peak
area (%)
Recovery (%)
1st Carbon
treatment
2nd Carbon
treatment
90.73
93.62
1.63
81.6
80.5
0.15
a Crude H-(D-Arg)8-NH2$9HCl at 79.9% peak area % and pyrazole at 6.1%
peak area % was used for the study.
4. Experimental
4.1. General
Hydrogenolysis stage, H-[D-Orn(Boc)]2-OMe$pTSA (d1).
The solution was hydrogenated at approximately 40 psi
while maintaining the temperature at 20G5 8C. Every 2 h
the reactor was subjected to 2 vent/purge cycles to remove
CO2 generated from the Z group hydrogenolysis. After 6 h,
an in process sample was pulled to monitor the reaction by
HPLC. Analytical result showed the reaction was complete
(Z-[D-Orn(Boc)]2-OMe (c1)Z0.0%).
All starting materials were obtained from commercial
suppliers and used as received. Ultrapure aminosilica gel
was obtained from SiliCycle, Inc, Que., Canada. Carbon
treatment was performed with carbon L3S from CECA S.A
(France). Triisopropylsilane, 99%, was obtained from
Johnson Mathey. All reactions were performed under an
atmosphere of nitrogen. Analytical thin layer chromato-
graphy was performed on MERCK 5!10 cm SG-60F 250 m
precoated silicagel plates. Visualization was accomplished
with iodine vapor. RP-HPLC analysis were performed on
Agilant 100 Series HPLC, using analytical columns
Chromolith RP-18e, 100!4.6 mm; for compounds 1, 2, 3
and Water Symmetry300 C18, 250!4.6 mm, 5 mm; for
compounds 4, 5, 6; with UV detection (lZ205 nm). The
products were eluted utilizing a solvent gradient [AZ
water:acetonitrile:perchloric acid (850:150:1); BZwater:-
acetonitrile:perchloric acid (150:850:1); for compounds 1,
2, 3 and AZ0.1% perchloric acid in water; BZ0.1%
perchloric acid in water/acetonitrile 5:95%; for compounds
4, 5, 6]. The reverse osmosis was performed on a Millipore
unit with a Nanomax-50 membrane (1.800 cartridge). The
lyophilization was performed with an FTS tray freeze dryer.
NMR spectra were measured on a Varian UI500 magnetic
Coupling stage for Z-[D-Orn(Boc)]3-OMe (e1). The hydro-
genated solution was filtered through a filter pot from the
hydrogenolysis vessel to a reactor, which was previously
charged with 3.3 kg of Z-D-Orn(Boc)-OSu. N-Methylpyrroli-
dinone (4.5 kg) was used as a rinse of the hydrogenolysis
vessel and the filter pot. The solution was mixed for 15 min to
dissolve the solids, then diisopropylethylamine (1.1 kg) was
charged tothe reactor and the mixture was stirred atan internal
temperatureof20G5 8C. Aninprocesssamplewastakenafter
4 h for HPLC analysis. Analytical results indicated that no
starting material, H-[D-Orn(Boc)]2-OMe (d1), was detected.
Scavenging stage for Z-[D-Orn(Boc)]3-OMe (e1). To a
reactor was charged 1.4 kg of Si-amine, then the coupling
reaction mixture was charged to the same reactor. N-
Methylpyrrolidinone (5.0 kg) was used as a rinse. The
mixture was stirred at 45G5 8C. An in process sample was
pulledafter36 hforHPLCanalysis. ThedisappearanceofZ-D-
Orn(Boc)-OSu was monitored by HPLC. Analytical results
indicated that 0.67% of Z-D-Orn(Boc)-OSu was detected and
the reaction was called complete. The temperature was
adjusted to 20G5 8C and the reaction mixture was filtered
through a filter pot to a hydrogenation vessel, which was
previously charged with 0.3 kg of 5% palladium on alumina.
Two rinses with N-methylpyrolidinone (4.5 kg) were used
and the mixture was carried to the next step.
resonance spectrometer (1H NMR spectra at 500 MHz, 13
C
NMR spectra at 125 MHz). Mass spectra were collected
with an Agilent Series 1100 LC/MS using electrospray
(ES-MS). The column was Technikrom Kromasil KR100-
5C18-250A, 250!4.6 nm, (lZ205 nm); elution with AZ
0.1% trifluoroacetic acid in water; BZ0.1% trifluoroacetic
acid in acetonitrile. For compounds 1, 2, 3, and 4, the mass
reported corresponds to the structure with the loss of a Boc
protecting group (100), which is common with ES-MS.
4.1.1. Preparation of Z-[D-Orn(Boc)]4-OMe (1). Coupling
stage for Z-[D-Orn(Boc)]2-OMe (c1). To a reactor were
charged successively 2.0 kg of H-D-Orn(Boc)-OMe$HCl
(a1) and 3.4 kg of Z-D-Orn(Boc)-OSu (b1) (5% molar
excess). N-Methylpyrrolidinone (27.0 kg) was charged to
the same reactor to dissolve the solids under agitation at an
internal temperature of 20G5 8C. Diisopropylethylamine
(1.0 kg) was then charged to the reactor and the mixture was
stirred for 2–3 h. An in process sample was taken for TLC
and HPLC analysis. Analytical results indicated that no
starting material, H-D-Orn(Boc)-OMe$HCl (a1), was
detected and the reaction mixture was used directly in the
next step.
Hydrogenolysis stage, H-[D-Orn(Boc)]3-OMe (f1). The
solution was hydrogenated at approximately 40 psi while
maintaining the temperature at 20G5 8C. After 2 h the
reactor was subjected to 2 vent/purge cycles to remove CO2
generated from the Z group hydrogenolysis. After 3 h, an in
process sample was pulled to monitor the reaction by HPLC.
The analytical result showed the reaction was complete
(Z-[D-Orn(Boc)]3-OMe (e1)Z0.0%).
Coupling stage for Z-[D-Orn(Boc)]4-OMe (1). The hydro-
genated solution was filtered through a filter pot from the

J. C. Califano et al. / Tetrahedron 61 (2005) 8821–8829
8827
1
hydrogenolysis vessel to a reactor, which was previously
charged with 3.24 kg of Z-D-Orn(Boc)-OSu. N-Methylpyr-
rolidinone (4.5 kg) was used as a rinse of the hydrogenolysis
vessel and the filter pot. The mixture was stirred at an
internal temperature of 20G5 8C. An in process sample was
taken after 3 h for HPLC analysis. Analytical results
indicated that no starting material, H-[D-Orn(Boc)]3-OMe
(f1), was detected.
purity of 97% by HPLC. MS 910 [MCHK(Boc)]C. H
NMR (500 MHz, DMSO-d6) d 1.35 (s, 36H), 1.36 (m, 8H),
1.46–1.61 (m, 8H), 2.89 (m, 8H), 3.98 (td, JZ8.54, 4.73 Hz,
1H), 4.21–4.25 (m, 3H), 4.97–5.01 (m, JZ12.50 Hz, 1H),
5.01–5.05 (m, JZ12.50 Hz, 1H), 6.73–6.76 (m, 4H), 7.01–
725 (s, 2H), 7.31–7.36 (m, 5H), 7.39 (m, 1H), 7.73 (d, JZ
7.63 Hz, 1H), 7.91 (d, JZ7.78 Hz, 2H). 13C NMR
(100 MHz, DMSO-d6) d 25.9–26.1, 28.5, 29.3–29.6, 39.5,
51.9, 52.2–52.3, 54.4, 65.4, 77.4, 127.6–128.3, 136.9, 155.5,
155.9, 171.1, 171.5, 171.8, 173.3. HRMS (EI) m/z
1008.5967 (calcd for C48H82N9O14 1008.5976).
The reaction mixture was diluted with 116 kg of isopropyl
acetate and the resulting solution was washed with purified
water (89 kg). The organic layer was then distilled under
vacuum at a jacket temperature of 45 8C to a volume of
approximately 40 L. To the reactor was charged 11 kg of
methanol and the mixture was stirred at about 40 8C until all
solids dissolved. The temperature was then adjusted to
about 20 8C and 124 kg isopropyl acetate was charged to the
reactor. Crystalline solids formed and after 6 h heptane
(18 kg) was charged to the reactor to complete the
crystallization. Solids were filtered and rinsed with
isopropyl acetate (30 kg). The wet cake was dried under
vacuum at 50 8C for 8 h (sample tested for loss on drying
showed 0% LOD) to give 5.26 kg (72%) of 1, with a purity
4.1.3. Preparation of Z-[D-Orn(Boc)]4-OH (3). To a
reactor were charged 2.5 kg of Z-[D-Orn(Boc)]4-OMe (1),
tetrahydrofuran (22 kg), and purified water (6.7 kg). The
contents of the reactor were mixed at room temperature until
solids dissolved. The internal temperature was adjusted to
10 8C and 7.33 kg of 0.5 M lithium hydroxide solution was
charged. The solution was mixed while maintaining the
temperature at no more than 15 8C. An in process sample
was pulled after 2 h for HPLC analysis. Analytical results
indicated that 1.0% starting material, Z-[D-Orn(Boc)]4-OMe
(1), remained and the reaction was called complete. While
maintaining the internal temperature at no more than 15 8C,
37.5 kg of 0.1 N HCl solution was charged to the reactor,
resulting in a solution at pH!4. The reaction mixture was
diluted with 30 kg of isopropyl acetate. The aqueous layer
was separated and was back-extracted with 30 kg of
isopropyl acetate. The organic layers were combined and
washed with purified water (30 kg). The organic layer was
then distilled under vacuum at 45 8C to a volume of
approximately 15 L. To the reactor was charged 5 kg of
methanol and the mixture was stirred at 45 8C until solids
dissolved. The temperature was then adjusted to 20 8C and
52.5 kg of isopropyl acetate was charged to the reactor. The
mixture was stirred for 12 h to effect crystallization of Z-[D-
Orn(Boc)]4-OH (3). Heptane (20.4 kg) was charged to the
reactor to complete the crystallization. After 3 h the solids
were filtered and rinsed with a mixture of methanol (1.2 kg)/
isopropyl acetate (14 kg)/heptane (6 kg). The wet cake was
dried under vacuum at 50 8C for 8 h to an LOD of 0% (limit
2%), to give 1.84 kg (77%) of product, with a purity of
1
of 95.4% by HPLC. MS 924 [MCHK(Boc)]C. H NMR
(500 MHz, DMSO-d6) d 1.35 (s, 36H), 1.39 (m, 8H), 1.45–
1.69 (m, 8H), 2.89 (m, 8H), 3.60 (s, 3H), 3.97 (td, JZ8.54,
4.73 Hz, 1H), 4.21–4.25 (m, 3H), 4.96–5.01 (m, JZ
12.50 Hz, 1H), 5.00–5.04 (m, JZ12.70 Hz, 1H), 6.72–
6.76 (m, 4H), 7.31–7.36 (m, 5H), 7.39 (m, 1H, N), 7.85–
7.88 (m, 2H), 8.20 (d, JZ7.17 Hz, 1H). 13C NMR
(100 MHz, DMSO-d6) d 28.2, 29.3–29.6, 39.3–39.5, 51.7,
51.6, 51.9–52.0, 54.4, 65.4, 77.4, 127.6–128.3, 137.0,
155.5–155.6, 155.9, 171.6, 171.2, 171.7, 173.3. HRMS
(EI) m/z 1023.5953 (calcd for C49H83N8O15 1023.5972).
4.1.2. Preparation of Z-[D-Orn(Boc)]4-NH2 (2). To a
reactor were charged 2.5 kg of Z-[D-Orn(Boc)]4-OMe (1)
and 22 kg of methanol. The mixture was stirred at 45 8C
until solids were dissolved. The temperature was then
adjusted to 5 8C and 7.5 kg of anhydrous ammonia was
charged to the reactor while maintaining the temperature at
no more than 15 8C. At the end of the addition, the
temperature was adjusted to 10 8C and the solution was
mixed. An in process sample was pulled after 43 h for
HPLC analysis. Analytical results indicated that 0.9%
starting material, Z-[D-Orn(Boc)]4-OMe (1), remained and
the reaction was called complete (in-process limit for Z-[D-
Orn(Boc)]4-OMe (1) was set at 1.0%).
97.2% by HPLC analysis. MS 909 [MCHK(Boc)]C. H
1
NMR (500 MHz, DMSO-d6) d 1.35 (s, 36H), 1.36 (m, 8H),
1.47–1.62 (m, 8H), 2.89 (m, 8H), 3.98 (m, 1H), 4.14 (m,
1H), 4.24–4.27 (m, 2H), 4.96–5.01 (m, JZ12.50 Hz, 1H),
5.01–5.06 (m, JZ12.50 Hz, 1H), 6.72–6.77 (m, 4H), 7.30–
7.35 (m, 5H), 7.38 (m, 1H), 7.85–7.89 (m, 2H), 8.03 (m,
1H), 12.54 (s, 1H). 13C NMR (100 MHz, DMSO-d6) d 25.9–
26.1, 28.3, 28.5–29.6, 39.5, 51.7, 51.9–52.1, 54.4, 65.4,
77.4, 127.7–128.3, 137.0, 155.6, 155.9, 171.3, 171.4, 171.8,
173.3. HRMS (EI) m/z 1009.5796 (calcd for C48H81N8O15
1009.5816).
The temperature was adjusted to 20 8C and the contents of
the reactor were distilled under vacuum at this temperature
for about 40 min. The distillation was then carried out at
45 8C under vacuum until the mixture reached 8 L. To the
reactor was charged 5 kg of methanol and the mixture was
stirred at a 45 8C until solids dissolved. The temperature
was then adjusted to 20 8C and 54 kg of isopropyl acetate
was charged to the reactor. The mixture was stirred for 12 h
to crystallize Z-[D-Orn(Boc)]4-NH2 (2). Heptane (14 kg)
was charged to the reactor to complete the crystallization.
After 3 h, the solids were filtered and rinsed with a mixture
of methanol (1.5 kg)/isopropyl acetate (17 kg)/heptane
(4.3 kg). The wetcake was dried under vacuuum at 50 8C
for 8 h (LOD 0%) to give 1.92 kg (76%) of product, with a
4.1.4. Preparation of Z-[D-Orn(Boc)]8-NH2 (4). To a
reactor were charged 1.75 kg of Z-[D-Orn(Boc)]4-NH2 (2),
0.175 kg of palladium on alumina, and dimethylformamide
(30 kg). The solution was then hydrogenated at 40 psi and
20 8C. An in-process sample was pulled after 3 h for HPLC
analysis. Analytical results indicated that no starting
material, Z-[D-Orn(Boc)]4-NH2 (2), was detected and the
reaction was called complete. The catalyst was removed by

8828
J. C. Califano et al. / Tetrahedron 61 (2005) 8821–8829
filtration and the filter cake was washed with dimethylfor-
mamide (15 kg).
isopropyl alcohol (30 kg). The wet cake was dried under
vacuum at 40 8C for 8 h to give 2.1 kg of product. The
product was dissolved in methanol (16.5 kg). The solution
was slowly poured into 63 kg of isopropyl alcohol under
high mixing. The reactor was rinsed with methanol
(1.6 kg) and the rinse was added to the IPA slurry, the
precipitate was filtered and rinsed with isopropyl
alcohol (8 kg). The wet cake was dried under vacuum
at 55 8C for 12 h to give 1.44 kg (69%) product with a
purity of 77% by HPLC. MS 701 [M/2CH]C (as free
base), 467 [M/3CH]C (as free base). 1H NMR
(500 MHz, DMSO-d6) d 1.51 (m, 16H), 1.59–1.71 (m,
16H), 3.13 (m, 16H), 4.17 (m, 1H,), 4.27 (m, 7H),
4.97–5.01 (m, JZ12.50 Hz, 1H), 5.01–5.06 (m, JZ
12.50 Hz, 1H), 7.14–7.51 (s, 2H), 7.31–7.36 (m, 5H)
7.03–7.47 (m, 24H), 7.50 (m, 1H), 7.84 (m, 8H), 8.22
(m, 7H). 13C NMR (100 MHz, DMSO-d6) d 25.0–25.1,
28.9–29.0, 40.3–40.4, 52.1, 52.1–52.5, 65.5, 127.7–
128.4, 136.9, 157, 171.4–171.5, 171.8, 173.4. HRMS (EI)
m/z for [MC3H]3C 467.6313 (calcd for C56H108N33O10
467.6314).
To a second reactor were charged 1.67 kg of Z-[D-
Orn(Boc)]4-OH (3), 0.34 kg of HONb and the hydrogenated
solution from above. The mixture was stirred until all solids
dissolved, then the internal temperature was adjusted to
10 8C. To the same reactor was added a solution of
EDAc$HCl (0.50 kg) in purified water (0.50 kg). The
mixture was stirred at 10 8C. An in process sample was
pulled after 6 h for HPLC analysis. Analytical results
indicated that 0.33% starting material, Z-[D-Orn(Boc)]4-OH
(3), remained and the reaction was called complete. The
internal temperature was adjusted to 20 8C. Under high
agitation, 48 kg of purified water was charged to the
mixture. The solution was stirred for 1 h. The precipitate
was filtered and rinsed with a mixture of dimethylforma-
mide (5 kg) and water (5 kg), followed by a rinse with 10 kg
of water. The material was blown dry with nitrogen for 2 h,
then the wet cake was recharged to the same reactor,
followed by the addition of acetone (42 kg) and purified
water (16 kg). The mixture was stirred until solids were
dissolved, then 31 kg of purified water was slowly charged
to the reactor. The mixture was stirred for 1 h. The
precipitate was filtered and rinsed with a mixture of acetone
(5 kg) and water (6 kg). The material was blown dry with
nitrogen for 2 h, then the wet cake was dried under vacuum
at 50 8C to an LOD of 0%. Upon completion of isolation and
drying, 2.29 kg (71%) was obtained, with a purity of 92.5%
4.1.6. Preparation of H-(D-Arg)8-NH2$9HCl (6). To a
reactor were charged 2.0 kg of Z-(D-Arg)8-NH2$8HCl (5),
0.2 kg of 5% palladium on alumina, water (25 kg), and
0.115 kg of 12 N hydrochloric acid. The mixture was then
hydrogenated at 30 psi and 20 8C. An in process sample was
pulled after 4 h for HPLC analysis. Analytical results
indicated that no starting material, Z-(D-Arg)8-NH2$8HCl
(5), was detected. The catalyst was filtered and washed
with water (25 kg). The hydrogenated solution was
charged to a reactor followed by a rinse with water
(3 kg). L3S activated carbon (2.4 kg) was charged and
the mixture was stirred at ambient temperature. After
4 h the mixture was filtered and the carbon was washed with
water (3!3 kg).
1
(HPLC area %). MS/ES 1765 [MCHK(Boc)]C. H NMR
(500 MHz, DMSO-d6) d 1.35 (s, 72H), 1.36 (m, 16H), 1.62–
1.46 (m, 16H), 2.89 (m, 16H), 3.98 (m, 1H), 4.14 (m, 1H),
4.21 (m, 6H), 4.97–5.01 (m, JZ12.50 Hz, 1H), 5.01–5.05
(m, JZ12.50 Hz, 1H), 6.71–6.76 (m, 8H), 7.01–7.25 (s,
2H), 7.31–7.35 (m, 5H), 7.40 (m, 1H), 7.72 (m, 1H), 7.89
(m, 6H). 13C NMR (100 MHz, DMSO-d6) d 25.9–26.1,
28.2, 29.4–29.6, 39.5, 52.0, 52.2, 52.4, 65.4, 77.4, 127.6–
128.2, 136.9, 155.5, 156.0, 171.1, 171.4, 173.3. HRMS (EI)
m/z for [MC2H]2C 933.0669 (calcd for C88H153N17O26
933.0659).
The combined filtrates were transferred to a reactor
followed by a water rinse (3 kg). To the same reactor was
charged 1.4 kg of L3S activated carbon and the mixture
was stirred at ambient temperature. After 4 h the mixture
was filtered and the carbon was washed with water (3!
3 kg). The combined filtrates were concentrated by
reverse osmosis from a volume of approximately 45–
8 L. The concentrated solution was then frozen and
lyophilized to give H-(D-Arg)8-NH2$9HCl (0.7 kg) with
a purity of 94% by HPLC. MS 634 [M/2CH]C (as free
base), 623 [M/3CH]C (as free base). 1H NMR
(500 MHz, DMSO-d6) d 1.51 (m, 16H), 1.60–1.71 (m,
16H), 3.13 (m, 16H), 4.17 (m, 1H), 4.27 (m, 7H), 7.04–
7.49 (m, 24H), 7.50 (m, 2H), 7.14–7.52 (s, 2H), 7.84–
7.93 (m, 8H), 8.02 (m, 1H), 8.23 (m, 4H), 8.33 (m, 1H),
8.76 (m, 1H). 13C NMR (100 MHz, DMSO-d6) d 24.2,
25.0, 28.8–29.0, 40.3–40.4, 52.1–52.5, 157.0, 169.8, 171.1,
171.2, 171.5, 173.4. HRMS (EI) m/z for [MC3H]3C
422.9527 (calcd for C48H102N33O8 422.9524).
4.1.5. Preparation of Z-(D-Arg)8-NH2$8HCl (5). To a
reactor were charged 2.5 kg of Z-[D-Orn(Boc)]8-NH2 (4)
and a mixture of ethyl acetate (60 kg)/anhydrous HCl
(7.3 kg)/triisopropylsilane (10.6 kg). The slurry was stirred
at 20 8C. An in process sample was pulled after 90 min for
HPLC analysis. Analytical results indicated that no starting
material, Z-[D-Orn(Boc)]8-NH2 (4), was detected. The
solids were filtered and rinsed with ethyl acetate (30 kg),
then the wet-cake was blown dry with nitrogen for 12 h to
give Z-(D-Orn)8-NH2$8HCl (2.5 kg). The dried Z-(D-Orn)8-
NH2$8HCl was charged to a reactor, followed by 3.4 kg
of 1H-pyrazole-1-carboxamidine hydrochloride, water
(2.7 kg), and methanol (16.2 kg). The mixture was stirred
until solids dissolved. Diisopropylethylamine (4.2 kg) was
charged and the mixture was stirred. An in process sample
was pulled after 4 h for HPLC analysis. Analytical results
indicated that no starting material, Z-(D-Orn)8-NH2$8HCl,
was detected. The reaction mixture was slowly poured into
144 kg of isopropyl alcohol under high agitation to
precipitate the product. The reactor was rinsed with 10 kg
of methanol and the rinse was charged to the IPA slurry.
After 20 min, the precipitate was filtered and rinsed with
Acknowledgements
1
We thank Ian Marsden for his critical H NMR analysis.

J. C. Califano et al. / Tetrahedron 61 (2005) 8821–8829
8829
12. Bernatowicz, M. S.; Wu, Y.; Matsueda, M. S. J. Org. Chem.
1992, 57, 2497–2502.
References and notes
13. Bodansky, M. Principles of Peptide Synthesis; Springer: New
York, 1984; pp 137–141, see also 192.
1. Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149–2154.
2. Kirshberg, T. A.; Thorsten, A.; VanDeusen, C. L.; Rothbard,
J. B.; Yang, M. A.; Wender, P. A. Org. Lett. 2003, 5,
3459–3462.
14. Bredereck, H.; Effenberger, F.; Hajek, M. Chem. Ber. 1965,
98, 3178.
15. Wojiechowska, H.; Zgoda, W.; Dziegielewski, K.; Drze-
winski, W.; Lutinska, I.; Schoeffer, A.; Borowski, E. Pol.
Patent 99496, 15 Nov, 1978.
3. Rothbard, J. B.; Kreider, E.; VanDeusen, C. L.; Wright, L. R.;
Wylie, B. L.; Wender, P. A. J. Med. Chem. 2002, 45,
3612–3618.
16. Kim, K.; Lin, Y. T.; Mosher, H. S. Tetrahedron Lett. 1988, 29,
3183.
4. Wender, P. A.; Jessop, T. C.; Pattabiraman, K.; Pelkey, E. T.;
VanDeusen, C. L. Org. Lett. 2001, 3, 3229–3232.
5. Wender, P. A.; Mitchel, D. J.; Pattabiraman, K.; Pelkey, E. T.;
Steinman, L.; Rothbard, J. B. Proc. Natl. Acad. Sci. U.S.A.
2000, 97, 13003–13008.
17. Luedtke, N. W.; Baker, T. J.; Goodman, M.; Tor, Y. J. Am.
Chem. Soc. 2000, 122, 12035–12036.
18. Feichtinger, K.; Zapf, C.; Sings, H. L.; Goodman, M. J. Org.
Chem. 1998, 3804–3805.
6. Mitchel, D. J.; Kim, D. T.; Steinman, L.; Fathman, C. G.;
Rothbard, J. B. J. Peptide Res. 2000, 55, 318–325.
7. Prochiantz, A. Curr. Opin. Cell Biol. 2000, 12, 400–406.
8. Lindgren, M.; Hallbrink, M.; Prochiantz, A. Langel Trends
Pharmacol. Sci. 2000, 21, 99–102.
19. McComas, W.; Chen, L.; Kyungjin, K. Tetrahedron Lett.
2000, 41, 3573–3576.
20. Ultrapure aminosilica gels are commercially avalaible from
SiliCycle, Inc, Que., Canada.
21. Anderson, G. W.; Zimmermann, J. E.; Callahan, F. M. J. Am.
Chem. Soc. 1963, 85, 3039.
9. Schwartz, J. J.; Zhang, S. G. Curr. Opin. Mol. Theor. 2000, 2,
162–167.
22. Fujino, M.; Kobayashi, S.; Obayashi, M.; Fukuda, T.;
Shinagawa, S.; Nishimura, O. Chem. Pharm. Bull. 1974, 22,
1857–1863.
10. Scharze, S. R.; Dowdy, S. F. Langel Trends Pharmacol. Sci.
2000, 21, 45–48.
11. Scharze, S. R.; Hruska, K. A.; Dowdy, S. F. Langel Trends
Cell Biol. 2000, 10, 290–295.
23. Kitada, C.; Fujino, M. Chem. Pharm. Bull. 1978, 26, 585–590.