The 1,1-dioxobenzo[b]thiophene-2-ylmethyloxycarbonyl (Bsmoc) amino- protecting group

Article abstract of DOI:10.1021/jo982140l

Full details are presented for use of the Bsmoc amino-protecting group for both solid phase and rapid continuous solution syntheses. Application to the latter methodology represents a significant improvement over the corresponding Fmoc-based method for rapid solution synthesis due to the opportunity to use water or saturated sodium chloride solution rather than an acidic phosphate buffer to remove all byproducts, with consequent cleaner phase separation and higher yields of the growing peptide. Comparison of the Bsmoc and Bspoc functions showed that the former, because of steric hindrance, does not suffer from the competitive or premature deblocking observed with the Bspoc system. Because of its in corporation of a styrene chromophore, resin loading of Bsmoc amino acids could be followed as has previously been shown for the Fmoc analogues. Applications of Bsmoc chemistry to peptide sequences incorporating the base sensitive Asp-Gly unit gave less contamination due to aminosuccinimide formation than comparable syntheses involving standard Fmoc chemistry because a weaker or less concentrated base could be used in the deblocking step. Experimental details are presented for building up peptides in solution via the continuous methodology. Deblockings involved the use of insoluble piperazino silica as well as the polyamine TAEA which simplified aqueous separation of the growing, but nonisolated peptide product, from excess acylating agent and other side products formed in the deblocking process. By the appropriate choice of base, one can act selectively at either site of a molecule which incorporates both β- elimination and Michael acceptor sites as protective units (Bsmoc vs Fm and Fmoc vs Bsm).

Full text of DOI:10.1021/jo982140l

4324
J . Org. Chem. 1999, 64, 4324-4338
Th e 1,1-Dioxoben zo[b]th iop h en e-2-ylm eth yloxyca r bon yl (Bsm oc)†
Am in o-P r otectin g Gr ou p
Louis A. Carpino,* Mohamed Ismail, George A. Truran, E. M. E. Mansour,‡ Shin Iguchi,
Dumitru Ionescu,§ Ayman El-Faham,‡ Christoph Riemer, and Ralf Warrass
Department of Chemistry, Box 34510, University of Massachusetts Amherst, Massachusetts 01003-4510
Received October 23, 1998
Full details are presented for use of the Bsmoc amino-protecting group for both solid phase and
rapid continuous solution syntheses. Application to the latter methodology represents a significant
improvement over the corresponding Fmoc-based method for rapid solution synthesis due to the
opportunity to use water or saturated sodium chloride solution rather than an acidic phosphate
buffer to remove all byproducts, with consequent cleaner phase separation and higher yields of the
growing peptide. Comparison of the Bsmoc and Bspoc functions showed that the former, because
of steric hindrance, does not suffer from the competitive or premature deblocking observed with
the Bspoc system. Because of its incorporation of a styrene chromophore, resin loading of Bsmoc
amino acids could be followed as has previously been shown for the Fmoc analogues. Applications
of Bsmoc chemistry to peptide sequences incorporating the base sensitive Asp-Gly unit gave less
contamination due to aminosuccinimide formation than comparable syntheses involving standard
Fmoc chemistry because a weaker or less concentrated base could be used in the deblocking step.
Experimental details are presented for building up peptides in solution via the continuous
methodology. Deblockings involved the use of insoluble piperazino silica as well as the polyamine
TAEA which simplified aqueous separation of the growing, but nonisolated peptide product, from
excess acylating agent and other side products formed in the deblocking process. By the appropriate
choice of base, one can act selectively at either site of a molecule which incorporates both
â-elimination and Michael acceptor sites as protective units (Bsmoc vs Fm and Fmoc vs Bsm).
In tr od u ction
Recently a new base-sensitive amino-protecting group,
the 1,1-dioxobenzo[b]thiophene-2-ylmethyloxycarbonyl
(Bsmoc) function 1, was described in a preliminary
report.1 In the present paper we provide new findings
regarding the practical use of this system as well as full
experimental details on the preparation of the key
reagents needed for its application to both solid phase
and solution peptide synthesis. We also update the
previous lists (Tables 1 and 2) of Bsmoc-amino acids and
amino acid fluorides synthesized to date and add detailed
directions for the synthesis of three trifunctional acids,
namely the Bsmoc derivatives of Asn and Gln which are
side-chain-protected by the dimethyl(cyclopropyl)methyl
residue2 and the Pbf derivative of arginine. Finally,
methods are described for the selective deblocking of
systems built from 9-fluorenemethyl and benzothiophene-
sulfonemethyl residues.
Th e Qu estion of P r em a tu r e or Com p etitive De-
block in g. The very high base-lability of the Bsmoc-
relative to that of the Fmoc residue raised the question
of possible premature deblocking during the coupling
step. During a study of the “parent” 2-(tert-butylsulfonyl)-
2-propenoxycarbonyl (Bspoc) residue 2, such a reaction
was noted during acylation of phenylalanine tert-butyl
ester by means of Bspoc-Phe-Cl.3 The more-hindered
† Common name: benzo[b]thiophenesulfone-2-methyloxycarbonyl.
Other abbreviations used: ACP ) acyl carrier peptide; BOC ) tert-
butyloxycarbonyl; Bsm ) 1,1-dioxobenzo[b]thiophene-2-ylmethyl; Bsm-
pip ) 1,1-dioxobenzo[b]thiophene-2-ylmethylpiperidine; Bspoc ) 2-tert-
butylsulfonyl-2-propenoxycarbonyl; DCM ) dichloromethane; DIEA )
diisopropylethyl-amine; DMAP ) 4-(dimethylamino)pyridine; DMD-
MAP ) 2,6-dimethyl-DMAP; Dmcp ) dimethyl(cyclopropyl)methyl;
DMF ) dimethylformamide; DMSO ) dimethyl sulfoxide; EEDQ )
1-(ethyloxycarbonyl)-2-(ethyloxy)-1,2-dihydroquinoline; Fm ) 9-fluo-
renylmethyl; Fmoc ) 9-fluorenylmethyloxycarbonyl; HATU ) N-[(dim-
ethylamino)-1H-1,2,3-triazolo[4,5-b]pyridin-1-ylmethylene]-N-methyl-
methanaminium hexafluorophosphate N-oxide; HBTU
) N[(1H-
benzotriazol-1-yl)(dimethylamino)methylene]-N-methylmethanamin-
ium hexafluorophosphate N-oxide; HOAt ) 7-aza-1-hydroxybenzot-
riazole; HOSu ) N-hydroxysuccinimide; Marfey’s reagent ) 1-fluoro-
2,4-dinitrophenyl-5-L-alanine amide; Pbf ) 2,2,4,6,7-pentamethyldihydro-
benzofuran-5-sulfonyl; pGlu ) pyroglutamic acid; PAC-PEG-PS )
peptide acid linker on poly(ethylene glycol)/polystyrene support; TAEA
) tris(2-aminoethyl)amine; TBTU ) N-[(1H-benzotriazol-1-yl)(dim-
ethylamino)methylene]-N-methylmethanaminium tetrafluoroborate N-
oxide; TFA ) trifluoroacetic acid; TFFH ) tetramethylfluoroforma-
midinium hexafluorophosphate; TLC ) thin-layer chromatography;
ToppipU ) 2-(2-oxo-1(2H)-pyridyl)-1,1,3,3-bispentamethyleneuronium
tetrafluoroborate; Trt ) trityl ) triphenylmethyl; Z ) benzyloxycar-
bonyl.
‡ On leave of absence from the Department of Chemistry, Faculty
of Science, Alexandria University, Alexandria, Egypt.
§ Fulbright Scholar. On leave of absence from the Department of
Organic Chemistry, Faculty of Chemistry, University of Bucharest,
70346 Bucharest, Romania.
(2) Carpino, L. A.; Chao, H.-G.; Ghassemi, S.; Mansour, E. M. E.;
Riemer, C.; Warrass, R.; Sadat-Aalaee, D.; Truran, G. A.; Imazumi,
H.; El-Faham, A.; Ionescu, D.; Ismail, M.; Kowaleski, T. L.; Han, C.-
H.; Wenschuh, H.; Beyermann, M.; Bienert, M.; Shroff, H.; Albericio,
F.; Triolo, S. A.; Sole, N. A.; Kates, S. A. J . Org. Chem. 1995, 60, 7718.
(3) Carpino, L. A.; Philbin, M. J . Org. Chem. 1999, 64, 4315.
(1) Carpino, L. A.; Philbin, M.; Ismail, M.; Truran, G. A.; Mansour,
E. M. E.; Iguchi, S.; Ionescu, D.; El-Faham, A.; Riemer, C.; Warrass,
R.; Weiss, M. S. J . Am. Chem. Soc. 1997, 119, 9915.
10.1021/jo982140l CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/19/1999

Bsmoc Amino-Protecting Group
J . Org. Chem., Vol. 64, No. 12, 1999 4325
Ta ble 1. Syn th esis of Bsm oc-Am in o Acid s a n d Der iva tivesa
compound
prep method
recryst solventb
specific rotation, [RD], t, °C
mp (°C)
168-169
98-100
155-156
175-176
172-173
43-45
104-106
163-164
149.5-151
foam
108-109
amorphous
133-134
109-111
oil
204-206 (dec)
144-146
166-167
139 (dec)
127.5-129
128-130
amorphous
amorphous
oil
yield (%)c
Bsmoc-Gly-OH
1
88.7 (80.0)3
86.0
Bsmoc-Ala-OH
1
-17° (c 1.0 DMF), 23
-9.4° (c 1.0 DMF), 20
-0.37 (c 1.0 DMF), 20
+0.37 (c 1.0 DMF), 20
-11.7 (c 1.0 EtOAc), 20
-5.0 (c 0.5 DMF), 23
+3.0 (c 1.0 DMF), 20
Bsmoc-Val-OH
1
DCM/hexane
DCM/hexane
DCM/hexane
89.0 (80.7)3
94.0 (90.0)3
85.0
Bsmoc-Phe-OH
1
Bsmoc-D-Phe-OH
Bsmoc-Leu-OH
1
1
91.0
Bsmoc-Ile-OH
Bsmoc-Thr(t-Bu)-OH
Bsmoc-Aib-OH
Bsmoc-Asp(O-t-Bu)-OH
Bsmoc-Glu-(O-t-Bu)-OH
Bsmoc-Trp-OH
1
82.0
1
93.0
1
82.8 (31)3
80.7
2
1
+6.2 (c 0.5 CHCl3), 22
-40.4 (c 0.5 DMF), 22
-33.4 (c 1.0 DMF), 22
+12.4 (c 0.5 CHCl3), 22
52.7
2
92.0
Bsmoc-Tyr(t-Bu)-OH
Bsmoc-Met-OH
2
Et2O/hexane
CH2Cl2/hexane
92.0 (70)1
58.0
2
Bsmoc-Pro-OH
2
90.0
Bsmoc-Asn-OH
3
EtOH/H2O
EtOH
90.4
Bsmoc-Gln-OH
3
84.2
Bsmoc-Asn(Trt)-OH
Bsmoc-Gln(Trt)-OH
Bsmoc-Phg-OH
2
DCM/hexaned
DCM/hexaned
-20.0 (c 0.5 CMF), 20
-3.2 (c 0.5 DMF), 20
+79.2 (c 0.5 DMF), 23
-79.8 (c 0.5 DMF), 23
-2.0 (c 1.0 EtOAc), 23
82.8
2
91.0
1
79.0 (80.0)3
75.0
Bsmoc-D-Phg-OH
Bsmoc-Lys(Boc)-OH
Bsmoc-Orn(Boc)-OH
Bsmoc-Ser(t-Bu)-OH
Bsmoc-Asn-ONp
Bsmoc-Gln-ONp
Bsmoc-Asp(O-t-Bu)-OH‚DCHA
Bsmoc-Gly-Gly-OH
Bsmoc-Gln-OPfp
Bsmoc-Asn-OPfp
Bsmoc-Gln(Dod)-OH
Bsmoc-Asn(Dod)-OH
Bsmoc-Ala-NHDmcp
1
1
80.0
1
80.0
1
75
DCC
DCC
-
EtOH
159-160
144-147
90 dec
50
CH3CN/Et2O
Et2O/Skelly B
MeOH
63
87
-
200-203
83
DCC
DCC
e
115
85
138
74
THF/hexane
DCM/hexane
Et2O/hexane
96-98
71
e
f
147-149
88-90
80
97
a
All compounds were characterized on the basis of elemental analyses ((0.3%, C, H, N) and consistent IR and 1H NMR spectral data.
Most of these acids and intermediates are available from Oryza Laboratories, Inc., Chelmsford, MA 01824, FAX (978)-256-7434; Web:
http://www.oryzalabs.com; e-mail: oryza@world.std.com. b All compounds were recrystallized from EtOAc/hexane unless otherwise indicated.
c The numbers in parentheses refer to the yields of the same compounds obtained by the alternate technique indicated by the superscript.
d
In these cases a relatively small amount of hexane was used. e Via the acid, 4,4′-dimethoxybenzhydrol and a catalytic amount of H2SO4
in HOAc. f Via Bsmoc-Ala-F and the amine hydrochloride in the presence of DIEA in DCM. Chromatographed over SiO2 with elution by
EtOAc prior to recrystallization.
Bsmoc residue has been found not to undergo an analo-
gous reaction. Thus HPLC examination of the reaction
shown in eq 1 indicated that 98.5% of the dipeptide 4
was formed along with 1.3% of Bsmoc-Phe-OH.
observed in the glycine case, such reactions are unlikely
in the case of
(
)
Presumably the latter had been present in the original
sample of 3 or had been formed by hydrolysis during the
reaction or on workup of the reaction mixture. No other
side product was observed whereas in the case of the
Bspoc system there was evidence for the formation of
2.1% of 5 and 0.3% of 6.3 The best chance of observing
such a side reaction would involve the least hindered
amino acid
)
(
higher, more-hindered amino acids. Based on this result,
general application of Bsmoc chemistry to solution and
solid-phase peptide synthesis was deemed safe. The
preliminary publication has outlined the results, and
additional details are presented here.
Ap p lica tion of Bsm oc Ch em istr y to Solid -P h a se
Syn th eses. For solid-phase syntheses one could choose
as a starting point a commercial resin already loaded
with a particular amino acid. On the other hand it was
considered of importance to learn whether Bsmoc amino
acids could themselves be loaded onto normal resins by
the usual techniques. In view of the stepwise nature of
1
the deblocking process as evidenced by H NMR exami-
glycine. In fact, in a comparable Bsmoc test with glycine
ethyl ester there was no evidence for the formation of
either 8 or 9. Authentic reference samples of these two
compounds were synthesized according to eq 2. If not
nation,4 one expects that UV analysis5 of the deblocking
by piperidine of a Bsmoc substrate will involve first a
drop in absorption followed by a rise as the weak

4326 J . Org. Chem., Vol. 64, No. 12, 1999
Carpino et al.
Ta ble 2. Syn th esis of Bsm oc-Am in o Acid F lu or id esa
Ta ble 3. Exten t of Loa d in g a n d Level of Ra cem iza tion
for Atta ch m en t to a P AC-P EG-P S Resin a
recryst
yield
compound
Bsmoc-Gly-F
Bsmoc-Ala-F
Bsmoc-Phe-F
solventb
mp (°C)
(%)
loading
coupling
reagent
(mequiv/g)
(% D-form)
132-133
82.7
79.9
85.6
solvent
base (equiv)
118-120
foam or
Bsmoc-Leu-F DCM
Bsmoc-Leu-F DCM
Bsmoc-Leu-F DCM
3,4-lutidine (4) 0.17 (<0.1)
amorphous solidc
collidine (4)
DIEA (4)
0.17 (<0.1)
Bsmoc-D-Phe-F
foam or
amorphous solid
114-115
Et2O/hexane 107-110
amorphous powder
80.0
0.16 (0.21)
Bsmoc-Val-F
Bsmoc-Val-F
Bsmoc-Phg-F DCM
Bsmoc-Phg-F DCM
Bsmoc-Gly-F
Bsmoc-Gly-F
DCM
DCM
3,4-lutidine (4) 0.14 (<0.1)
collidine (4) 0.11 (<0.1)
3,4-lutidine (4) 0.13 (13.7)
Bsmoc-Val-F
81.7
79.7
75.0
93.0
79.7
66.7
90.0
85.0
95.0
78.0
88.9
87.0
80.0
78.0
81.0
82.0
75.0
87.8
93.7
Bsmoc-Asp(O-t-Bu)-F
Bsmoc-Glu-(O-t-Bu)-F
Bsmoc-Lys(Boc)-F
Bsmoc-Tyr(t-Bu)-F
Bsmoc-Aib-F
Bsmoc-Leu-F
Bsmoc-Ile-F
Bsmoc-Pro-F
Bsmoc-Thr(t-Bu)-F
Bsmoc-Asn(Trt)-F
Bsmoc-Gln(Trt)-F
Bsmoc-Trp-F
collidine (4)
collidine (4)
0.09 (9.5)
0.11
foam
foam
126-127
oil
68-70
oil
118-120
foam
foam
foam
foam
foam
amorphous
amorphous
amorphous solid
amorphous solid
DCM
DMF/DCM (1/9) NMM (4)
0.13
a
The resin (200 mg) was reported as having a loading of 0.24
mequiv of OH/g. All experiments were performed with 4 equiv of
the amino acid derivative in 1.0 mL of solvent for reaction times
of 3 h.
Ta ble 4. Reten tion Tim es for Ma r fey’s Ad d u cts
Bsmoc-Phg-F
adduct
retention time (min)a
Bsmoc-D-Phg-F
Bsmoc-Ser(t-Bu)-F
Bsmoc-Met-F
Bsmoc-Asn(Dmcp)-F
Bsmoc-Gln(Dmcp)-F
Leub
D-Leu
Phg
32.3 (22.8)
36.1 (22.7)
29.1 (22.5)
34.0 (22.5)
26.8 (20.4)
31.5 (20.3)
33.2 (21.8)
24.4 (22.5)
21.9
D-Phg
Val
a
All crystalline acid fluorides were characterized on the basis
of elemental analyses ((0.3%, C, H, N) and consistent IR and 1H
NMR spectral data. Foams or amorphous solids were characterized
D-Valb
piperidine
morpholine
reagent (after hydrolysis)
b
as the methyl esters where necessary. All acid fluorides were
recrystallized from 20% hexane in DCM unless otherwise indi-
cated. c Also obtained as a white solid, mp 132-133 °C (see
Experimental Section).
Asp
D-Asp
Ala
D-Ala
Asnc
D-Asnc
Phe
18.8 (21.6)
20.1 (21.8)
21.2 (21.1, coelutes)
24.4 (20.8)
17.6 (20.9)
17.6 (21.0)
32.2 (22.4)
35.3 (22.4)
chromophore of the intermediate 11 is converted to the
strong styrene chromophore of the stable adduct 12.
Precisely this effect is observed as Bsmoc-Gly-O-t-Bu is
D-Phe
a
The tR shown in parentheses is due to the hydrolysis product
of Marfey’s reagent which serves as a convenient tR standard.
b
These amino acid adducts can be separated from the piperidine
adduct by eluting with an isocratic system consisting of 10%
MeCN/H2O (0.1% TFA) for 5 min followed by a gradient system
(10% to 60% over 55 min). c Asn diastereomers can be separated
by isocratic elution with an aqueous buffer system involving 8%
MeCN/0.02 M NaOAc, pH 4; retention time L,L ) 14.8 min, D,L )
29.5 min.
calculated loading was only 0.140 mequiv/g, but after the
solution was allowed to stand for an additional 16 h, the
loading was measured as 0.168 mequiv/g in agreement
with the figure observed for the 20% piperidine run.
Thus, for accurate analytical results, one must use
appropriate deblocking conditions. Real time monitoring
during a synthesis, without some modification of the
methodology, would not be feasible in the case of de-
blockings carried out with low concentrations of piperi-
dine.
)
(
treated with a low concentration, e.g., 2%, of piperidine
in DCM (Figure 2a,b, Supporting Information). Because
the isomerization of 11 to 12 is slow under these
conditions, quantitative analysis is not possible. On the
other hand if 20% piperidine in DMF is used, deblocking
and subsequent isomerization to 12 is rapid. As an
example, a PEG-PS resin loaded with Bsmoc-Leu-OH
was treated with 20% piperidine in DMF for 30 min,
leading to a calculated loading of 0.168 mequiv/g. The
calculated loading did not change after the resin had been
subjected to the same deblocking reagent for a period of
16 h. On the other hand, after treatment of the same
resin with 2% piperidine in DMF for 5 min, the initial
Having established an analytical method to study the
loading process, we examined the applicability of various
bases and conditions in connection with the problem of
racemization (Table 3). The levels of racemization given
were determined by application of Marfey’s reagent
(Table 4).6 Model studies of similar reactions carried out
in solution have been published.7 These studies showed
(4) Reference 1, Supporting Information.
(5) In view of the presence of a strong styrene chromophore in the
Bsmoc residue, one would expect that methods similar to those used
for the Fmoc group to follow resin loading, the coupling and deblocking
steps, etc., could be used in this case as well. For the Fmoc methods,
see Meienhofer, J .; Waki, M.; Heimer, E. P.; Lambras, T. L.; Makofske,
R. C.; Chang, C.-D. Int. J . Pept. Protein Res. 1979, 13, 35.
(6) Marfey, P. Carlsberg Res. Commun. 1984, 49, 591.
(7) Granitza, D.; Beyermann, M.; Wenschuh, H.; Haber, H.; Carpino,
L. A.; Truran, G. A., Bienert, M. J . Chem. Soc., Chem. Commun. 1995,
2223. It is not certain whether this technique is feasible in the case of
Bsmoc amino acids, some of which, e.g., that of glycine, are sensitive
to DMAP.

Bsmoc Amino-Protecting Group
J . Org. Chem., Vol. 64, No. 12, 1999 4327
that both general base8 and nucleophilic9 catalysis were
important. The importance of the latter is emphasized
by the difference between 3,4-dimethyl- and 2,6-dimeth-
ylpyridine.
whether the 20% morpholine system would provide a
purer product than 2% piperidine when used with Bsmoc
chemistry for the assembly of Field’s â-isomer of 13.15
Bsm oc-Ba sed Ra p id Con tin u ou s Solu tion Syn -
th esis of P ep tid es. Previously we devised a method of
using Fmoc chemistry in the rapid synthesis of short
peptides in solution.16 The key step involved removal of
the TAEA deblocking byproduct 14 from the
Having succeeded in loading appropriate resins with
leucine, we first attempted the assembly of leucine
enkephalin via Bsmoc chemistry. These syntheses pro-
ceeded well using N-[[(dimethylamino)-1H-1,2,3-triazolo-
[4,5-b]pyridin-1-yl]methylene]-N-methylmethanamini-
um hexafluorophosphate N-oxide (HATU)10 as coupling
reagent with a 15-min coupling time and deblocking
conditions which varied from 20% piperidine in DMF for
10 min (standard Fmoc conditions) to 2% piperidine for
7 min. When the weaker base morpholine was substi-
tuted for piperidine in the deblocking step, a 5% solution
in DMF gave very poor syntheses, whereas 20% morpho-
line in DMF gave products of acceptable purity. The ACP
decapeptide, a common test sequence,11 was assembled
similarly with good results using either HATU or tetra-
methylfluoroformamidinium hexafluorophosphate(TFFH)12
as coupling reagents.
growing peptide by buffer extractions at pH 5.5 It has
now been found that the process can be simplified by
switching to Bsmoc chemistry since the byproduct adduct
15 formed in this case is soluble in water, thus avoiding
the need for extraction with an acidic buffer. This results
in fewer complications with emulsions and loss of growing
peptide into the aqueous phase. A second byproduct,
derived from excess acylating agent, is the amide 16
which must also be removed in the
Most interesting was the case of hexapeptide (H-Val-
Lys-Asp-Gly-Tyr-Ile-OH) 13 [Toxin 2(1-6) of the scorpion
Androctonus australis Hector].13 In the preliminary com-
munication of this work1 it was described how the â-Asp
form of this peptide14 could be assembled by standard
Fmoc and standard Bsmoc methodology with the result
that the latter method gives a significantly smaller
amount of the aminosuccinimide byproduct according to
HPLC analysis. In the additional work described in the
present paper, the R-analogue was assembled by the
same two methods. Then, following Quibell et al.,13 the
Fmoc product, still attached to the resin, was treated with
20% piperidine in DMF for 19 h in order to simulate a
longer synthesis (Figure 1c). The extended treatment
leads to the formation of large amounts of the R- and
â-piperidides and their D-isomers as well as the initial
cyclization product, the aminosuccinimide derivative.
Similar treatment with 20% morpholine/DMF for 19 h
was unaccompanied by any such byproducts (Figure 1d).
Using Bsmoc amino acid fluorides as coupling reagents
with 2% piperidine/DMF, the deblocking reagent previ-
ously used in the synthesis of the â-Asp system, a small
amount of the aminosuccinimide was formed (Figure 1a).
Therefore Bsmoc chemistry using 20% morpholine for
deblocking was examined for the assembly of 13. The
results were excellent (Figure 1b). On the other hand for
the synthesis of leucine enkephalin via Bsmoc chemistry
an attempt to use 2-5% morpholine/DMF gave little, if
any, of the desired product. It was not determined
aqueous washings. In this case, solubility in water is
determined by the nature of the R group. It was noted
that for glutamine, if the hydrophobic trityl group is used
for N-amide side chain protection, the resulting amide
(16, R ) CH2CH2CONHC(C6H5)3) is not completely
removed in the washing process.17 Water solubility is
enhanced, however, by switching to the lower-molecular-
weight, less-hydrophobic dimethylcyclopropylcarbinyl func-
tion 172 for side chain amide protection (16, R )
CH2CH2CONHC(C3H5)Me2). Results for the asparagine
analogue (16, R ) CH2CONHC(C3H5)Me2) are compa-
rable, and similar solutions are being sought for the His
(15) For recent alternative approaches to minimize â-catalyzed
aspartimide formation during Fmoc assembly and references to earlier
work, see: (a) Karlstro¨m, A.; Unde´n, A. Int. J . Pept. Protein Res. 1996,
48, 305. (b) Zhang, H.; Wang, Y.; Voelter, W. Chem. Lett. 1995, 1111.
(c) Yang, Y.; Sweeney, W. V.; Schneider, K.; Tho¨rquist, S.; Chait, B.
T.; Tam, J . P. Tetrahedron Lett. 1994, 35, 9689. (d) Do¨lling, R.;
Beyermann, M.; Haenel, J .; Kernchen, F.; Krause, E.; Franke. P.;
Brudel, M.; Bienert, M. J . Chem. Soc., Chem. Commun. 1994, 853 and
ref 13.
(8) (a) J encks, W. P.; Carriuolo, J . J . Am. Chem. Soc. 1961, 83, 1743.
(b) Bondarenko, L. I.; Kirichenko, A. I.; Litvinenko, L. M.; Dmitrenko,
I. N.; Kobets, V. D. Zh. Org. Khim. 1981, 17, 2588.
(9) Fersht, A. R.; J encks, W. P. J . Am. Chem. Soc. 1970, 92, 5432.
(10) (a) Carpino, L. A. J . Am. Chem. Soc. 1993, 115, 4397. (b)
Carpino, L. A.; El-Faham, A.; Minor, C. A.; Albericio, F. J . Chem. Soc.,
Chem. Commun. 1994, 201. (c) For a description of the current view
of the structure of such coupling reagents, see: Carpino, L. A.; El-
Faham, A.; Albericio, F. J . Org. Chem. 1995, 60, 3561.
(11) (a) For a brief review of the use of the acyl carrier protein
decapeptide (65-74) as a test sequence for solid phase assembly, see:
Adams, J . H.; Cook, R. M.; Hudson, D.; J ammalamaddaka, V.; Lyttle,
M. H.; Songster, M. F. J . Org. Chem. 1998, 63, 3706, footnote 21. (b)
Reference 10b.
(12) Carpino, L. A.; El-Faham, A. J . Am. Chem. Soc. 1995, 117, 5401.
(13) Quibell, M.; Owen, D.; Packman, L. C.; J ohnson, T. J . Chem.
Soc., Chem. Commun. 1994, 2343.
(14) Lauer, J .; Fields, C. G.; Fields, G. B. Lett. Pept. Sci. 1994, 1,
197.
(16) Carpino, L. A.; Sadat-Aalaee, D.; Beyermann, M. J . Org. Chem.
1990, 55, 1673.
(17) On the other hand, Bsmoc-Gln(Trt)-OH has previously been
used successfully in rapid solution syntheses (see refs 1 and 2 and the
synthesis of heptapeptide 23 described herein), possibly because earlier
syntheses have involved a greater excess of TAEA than needed or used
in current work. The presence of excess TAEA modifies the solvent
properties of the aqueous phase during the extraction process but may
also serve to remove some of the desired growing peptide as well as
the byproducts. It should be noted that structures 14-16 are assigned
on the basis of analogy to the corresponding products derived from
piperidine. None of these materials has been isolated and characterized
unequivocally.

4328 J . Org. Chem., Vol. 64, No. 12, 1999
Carpino et al.
F igu r e 1.
and Cys cases. An outline of the Bsmoc rapid solution
methodology is given in Scheme 1.
As noted earlier1 the Bsmoc deblocking process is easily
monitored by H NMR techniques. The methylene pro-
1

Bsmoc Amino-Protecting Group
J . Org. Chem., Vol. 64, No. 12, 1999 4329
Sch em e 1. Bsm oc Ra p id Con tin u ou s Solu tion
Syn th esis of P ep tid es
and as expected causes deblocking at a significantly
higher rate whereas just the opposite effect is seen for
the rearrangement step. A possible rationale is that the
piperazine adduct 19 is extremely insoluble in DMF, and
its precipitation may drive the reaction to completion. A
similar solubility effect has previously been observed for
the corresponding piperazine adduct formed upon Fmoc
deblocking.20 In addition, or alternatively, conversion of
the initial adduct 18 to 19 may be accelerated via the
cyclic transition state implied in 18.
Assem bly of th e Octa p ep tid e LED-CC-II. Although
the piperazine-induced deblocking/rearrangement pro-
cess described above could be of special use for Bsmoc
deblocking, initial studies were carried out with TAEA
since good results had previously been obtained with this
base in the Fmoc case. In the preliminary report1 of this
work, application of the TAEA deblocking technique to
a number of short peptides was described. Here experi-
mental details are presented for application to an oc-
tapeptide 20 (pGlu-Leu-Thr-Phe-Thr-Pro-Asn-Trp-NH2),
one among many related insect hormones.21 Application
of the Bsmoc protocol proceeded well except at the stage
of the heptapeptide where unacceptable losses of the
Bsmoc-deblocked, free-amino-containing intermediate
into the aqueous phase occurred. To avoid the need for
aqueous extractions at this step, an insoluble piperazino-
substituted silica reagent20 24 was substituted for TAEA.
In this way simple filtration and evaporation of solvent
gave the free-amino segment 25 ready for the next, in
this case final, step. The complete outline is presented
in Scheme 2.
Since the octapeptide occurs as a peptide amide, the
assembly was started with the C-terminal function of
tryptophan amide modified as the N-dimethylcyclopro-
pylmethyl (Dmcp) derivative2 in order to maintain solu-
bility of the growing peptide in the nonpolar, water-
immiscible solvent dichloromethane used in the extraction
procedures needed to effect byproduct removal. Two
different methods were used to assemble the chain, each
being equally satisfactory. In one case isolated amino acid
fluorides22 were used throughout for coupling. In a second
run, the HATU reagent10a,b was used for the first seven
coupling steps, and when the switch was made from
TAEA to piperazino silica as deblocking agent, the
appropriate acid fluoride was substituted. Prior to our
extensive use of Bsmoc amino acid fluorides for coupling
purposes it was considered important to determine
whether any racemization might occur during the cou-
pling process. Although many other urethane-protected
amino acid fluorides (Fmoc, Z, Boc) have been used
without any problems, the Bsmoc group is unique in
incorporating a residue, the sulfone function, which
tons adjacent to the 2-position of the benzothiophene-
sulfone ring show chemical shifts in the range 5.0-5.2
ppm. Following the initial deblocking step, conversion to
the vinyl intermediate 11 is signaled by a shift of these
protons to 6.3-6.5 ppm. As this intermediate is isomer-
ized to 12, the new methylene protons, being now
adjacent to nitrogen, are shifted upfield to 3.4-3.6 ppm.
Thus both deblocking and rearrangement steps are easily
monitored separately by using a limited amount (e.g., 2
equiv) of the deblocking base. To gain a general picture
of the ease of the deblocking process, a number of simple
deblocking amines was examined, and the following
reactivity order observed: piperidine > piperazine >
morpholine ≈ ethanolamine.18 The deblocking rates in
DMF roughly parallel the pKa values for these amines
except in the case of ethanolamine (pKa 9.51) which was
only about as effective as morpholine (pKa 8.33) despite
its greater basicity. A possible explanation is the in-
creased rate of addition to a Michael acceptor of a
secondary vs a primary amine.19 As previously suggested,
this effect may be the result of steric effects due to
increased hydrogen bonding in the case of the primary
system.
The difference between piperidine and piperazine is
particularly interesting in that the former is more basic
(20) Carpino, L.A.; Mansour, E. M. E.; Knapczyk, J . J . Org. Chem.
1983, 48, 666.
(21) Gaede, G.; Keller, R. Peptides 1989, 10, 1287.
(22) Compare: (a) Carpino, L. A.; Sadat-Aalaee, D.; Chao, H.-G.;
DeSelms, R. H. J . Am. Chem. Soc. 1990, 112, 9651. (b) Carpino, L. A.;
Mansour, E. M. E.; Sadat-Aalaee, D. J . Org. Chem. 1991, 56, 2611.
(18) The pKa values, taken from Perrin, D. D. The Dissociation
Constants of Organic Bases; Butterworths: London, 1965, are piperi-
dine (11.12), piperazine (9.81, 5.55), morpholine (8.33), and ethanola-
mine (9.51).
(19) McDowell, S. T.; Stirling, C. J . M. J . Chem. Soc. (B) 1967, 343.

4330 J . Org. Chem., Vol. 64, No. 12, 1999
Carpino et al.
Sch em e 2
were unsuccessful. On the other hand the corresponding
Boc derivative 2924 was easily converted to the acid
fluoride 26 by a modification of the standard procedure.
Following the directions of Schro¨der and Klieger, Boc-
pyroglutamic acid was synthesized in the form of its
crystalline dicyclohexylamine salt. Since the technique
described for converting the salt to the free acid invari-
ably gave an oil which could not be crystallized, direct
conversion of the salt to the acid fluoride was examined.
In fact, despite its potential, as a secondary amine, to
destroy the acid fluoride, dicyclohexylamine, presumably
because of its exceptionally hindered structure, did not
in fact interfere in the reaction. Boc-pyroglutamic acid
fluoride 26 was obtained in a yield of 85%. The same
technique, using the dicyclohexylamine salt of Fmoc-Phe-
OH, was shown to be applicable to Fmoc-Phe-F suggest-
ing its probable generality. Treatment of the free-amino
heptapeptide 25 with acid fluoride 26 gave somewhat
unstable protected octapeptide 27 in 58% yield. For proof
of structure, a freshly prepared sample of 27 was purified
by silica gel chromatography and then subjected to
trifluoroacetic acid deblocking. The free octapeptide 20
showed a similar instability although a freshly prepared
sample could be characterized by HPLC, amino acid
analysis, and mass spectral examination. Although only
used in one step of the present synthesis, the technique
involving resin-based deblocking promises to be generally
applicable to every step, and if the method is modified
to include resin-based coupling for each amino acid, a
contemporary solution to the problem of clean “two
polymer”25 or inverse Merrifield synthesis of peptides
could be contemplated.
An important aspect of the use of acid-sensitive amino
protecting groups is the opportunity to distinguish some
such functions on the basis of the acidity of the deblocking
reagent. Thus, under appropriate conditions, one can
easily remove trityl protection in the presence of a tert-
butyl-based system.26 In the preliminary description of
the present work it was similarly demonstrated that two
base-sensitive groups could be distinguished. Thus the
Bsmoc group could be deblocked in the presence of an
Fm ester residue. Full experimentatal details for as-
sembly of Bsmoc-Leu-Leu-Leu-OFm from Bsmoc-Leu-F
and H-Leu-OFm are appended here. For the dipeptide
Bsmoc-Leu-Phe-OFm, experimental conditions are given
for the selective deblocking of either the Bsmoc- or the
Fm-residue.
Another example, not previously discussed, based on
the reverse of the selectivity outlined in the preliminary
report, namely removal of Fmoc protection in the pres-
ence of a Bsm ester residue, is presented here.
Preliminary experiments with various simple amines
allowed the identification of hindered amines which could
effect â-elimination of the Fmoc residue under conditions
which avoided extensive Michael-like addition to the Bsm
residue. The most effective reagents were tert-butylamine
and N-methyl-tert-butylamine (see Table 5). The latter
was more selective with less than 3% removal of the Bsm
exhibits a high electron-withdrawing inductive effect
which might conceivably promote base-catalyzed abstrac-
tion of the proton R- to the carbonyl fluoride residue with
consequent loss of configuration.23 In fact, coupling of the
highly sensitive amino acid, R-phenylglycine, as its acid
fluoride with alanine methyl ester occurred without any
loss of configuration according to 1H NMR analysis of the
methyl ester peaks of the two expected diastereomers.
It is not expected therefore that other less-sensitive
amino acids would be at risk. Loss of configuration is even
less likely in the presence of coupling reagents such as
HATU or HBTU.
The protected heptapeptide 23, assembled in a single
series of steps over a period of 7 h, was obtained in a
yield of 37% for the purified peptide (column chromatog-
raphy). Piperazino silica deblocking of 23 gave the free-
amino segment 25, ready for the final coupling. The first
approach to introduce the N-terminal amino acid, pyro-
glutamic acid, involved prior conversion of the acid 28
to
(23) Compare: Carpino, L. A.; Ionescu, D.; El-Faham, A.; Henklein,
P.; Wenschuh, H.; Bienert, M.; Beyermann, M. Tetrahedron Lett. 1998,
39, 241. The increased reactivity of Bsmoc amino acids over the Fmoc
analogues has been attributed to a similar effect.1,4
(24) Schro¨der, E.; Klieger, E. Liebigs Ann. Chemie 1964, 673, 196.
(25) Carpino, L. A.; Cohen, B. J .; Lin, Y.-Z.; Stephens, K. E., J r.;
Triolo, S. A. J . Org. Chem. 1990, 55, 251.
(26) Barlos, K.; Chatzi, O.; Gatos, D.; Stauropoulos, G. Int. J . Pept.
Protein Res. 1991, 37, 513.
its acid fluoride. Even though it is internally “protected”
as a lactam, attempts to prepare the acid fluoride of 28

Bsmoc Amino-Protecting Group
J . Org. Chem., Vol. 64, No. 12, 1999 4331
Ta ble 5. Deblock in g of F m oc-P h e-Leu -OBsm to
H-P h e-Leu -OBsm a
Con clu sion
This work has described a new base-sensitive protect-
ing group, the Bsmoc residue, for which the deblocking
step involves addition of an amine to a Michael acceptor
site. Relative to the less-hindered Bspoc residue, intro-
duced in the accompanying paper,3 no premature de-
blocking has been noted. Applications of this protectant
are described for both solid phase and rapid continuous
solution synthesis. For the latter system the Bsmoc
methodology shows advantages over the Fmoc system in
that it is simpler, cleaner, and provides higher yields.
Further advantages of Bsmoc chemistry are apparent in
cases where byproduct formation is caused by base-
catalyzed processes. Finally it is shown that Fm- and
Bsm-based protecting groups can be orthogonally de-
blocked by appropriate choice of the deblocking base.
time for
complete Fmoc
deblocking (min)
deblocking at
both Fmoc and
Bsm sites (%)
deblocking
amine/solvent
10% Et2NH/CH3CN
10% N-Methyl-N-
cyclohexylamine/CH3CN
20% Me2NEt/DMF
30
30
48
10
90
90
8
60
5
26
19
5.8
38
2.6
2.8
20% Me2NiPr/DMF
30% t-BuNH2/CH3CNb
20% tert-amylNH2/CH3CN
20% t-BuNHMe/DMF
30% t-BuNHMe/DMF
5
a
To 0.5 mL of the base/solvent combination was added 34 mg
of the protected dipeptide and the progress of the reaction followed
by HPLC analysis, linear gradient 10/90 in 20 min with H2O/
CH3CN/0.1% TFA, retention times: Fmoc-Phe-Leu-OBsm 20.1
min, Fmoc-Phe-Leu-OH 17.1 min, H-Phe-Leu-OBsm 11.9 min,
H-Phe-Leu-OH 7.8 min. b The t-BuNH2 was a sample from Aldrich
Chemical Co. of 99.5% purity. Less pure samples (98%) gave a
less sharp distinction between the two base-sensitive sites.
Exp er im en ta l Section
Ben zoth iop h en e-2-m eth a n ol. Benzothiophene (45.5 g,
0.34 mol) was dissolved in 225 mL of dry THF and the solution
cooled to -30 °C under an atmosphere of N2. n-BuLi (330 mL,
1.6 M in hexane, 1.5 equiv) was added dropwise over 2 h with
mechanical stirring. The dark blue solution was allowed to
come to 0 °C and then recooled to -78 °C. Paraformaldehyde
(72 g, 2.4 mol) was added in small portions over 20 min, and
the mixture was allowed to warm to room temperature
overnight. Caution! A considerable amount of gas is evolved
on warming especially when the temperature reaches 10-20
°C. The apparatus should be well-ventilated and stirring
maintained, otherwise the mixture may boil uncontrollably.
The mixture was acidified to pH about 3 with 3 N hydrochloric
acid, the layers were separated, and the aqueous phase was
extracted twice with 100-mL portions of ether. The solid
remaining in the flask was washed with three 150-mL portions
of ether, all ether extracts were combined and dried over
MgSO4, and the solvent was removed by rotary evaporation.
The residual oil was dissolved in 600 mL of DCM and the
solution washed with saturated NaHCO3 (2 × 100 mL),
saturated NaCl (100 mL), and dried over MgSO4. The solution
was filtered, and the solvent was removed to give an oil which
was recrystallized from DCM/hexane (4:1) to give 42.4 g (70%)
of the alcohol as a white solid, mp 68-70 °C (lit.27 mp 99-100
Sch em e 3
residue occurring after the 5-min deblocking time which
completely removed the Fmoc residue. Although poten-
tially an inexpensive reagent, N-methyl-tert-butylamine
is currently not widely available commercially. tert-
Butylamine, on the other hand, is an inexpensive reagent
and was used in the present studies, although somewhat
less than 6% of the Bsm residue was removed during
deblocking of the Fmoc residue. Other amines of greater
selectivity are being sought. Unsuccessful were N,N-
diethyl- and N,N-diisopropylamine as well as tert-amy-
lamine.
1
°C); IR (KBr): 3277, 1457 cm-1; H NMR (CDCl3): δ 2.24 (s,
1), 4.88 (s, 2), 7.17-7.27 (m, 5). The same alcohol was obtained
in 63% yield, mp 99-100 °C, by treatment of 2-lithioben-
zothiophene with N,N-dimethylformamide followed by reduc-
tion with sodium borohydride in MeOH. Presumably the two
forms, mp 68-70 °C and 99-100 °C, are polymorphic forms,
as the two give identical IR and 1H NMR spectra. The latter
method was based on a literature procedure.28
Ben zoth iop h en esu lfon e-2-m eth a n ol. The previous de-
scription of the oxidation of benzothiophene-2-methanol in-
volved use of magnesium bismonoperoxyphthalic acid. More
recently it has been shown that the method of McKillop and
Kemp29 involving sodium perborate is simpler and less expen-
sive. A general procedure involves portionwise addition over
a period of 20 min of 95.4 g (0.62 mol) of sodium perborate
tetrahydrate to a stirred solution of 0.124 mol of ben-
zothiophene-2-methanol in 500 mL of acetic acid held at 45-
50 °C. Stirring was continued at this temperature until
completion of the oxidation (TLC). The acetic acid was removed
by evaporation under reduced pressure, and the residue was
stirred with 150 mL of water. The precipitated sulfone alcohol
was collected by filtration and recrystallized from chloroform/
hexane or dichloromethane/hexane to give the sulfone alcohol
in about 75% yield. Prior to the development of this general
As an example of the practical utility of this methodol-
ogy, leucine enkephalin C-terminal tripeptide 34 can be
assembled as outlined in Scheme 3.
During the synthesis of 34, protected dipeptide 33 was
isolated and purified before going to the next step. If one
attempts to add a second glycine residue to 34 to give
the corresponding tetrapeptide without purification of the
tripeptide, because of the workup method used (see
Experimental Section), some residual unreacted Fmoc-
Gly-OH remains which then forms a nonvolatile salt with
tert-butylamine, and upon addition of the coupling re-
agent some Fmoc-Gly-NH-t-Bu is formed to contaminate
the desired tetrapeptide. A more hydrophobic Bsm de-
rivative is being sought which would allow an aqueous
workup step to remove byproducts and excess tert-
butylamine, thus allowing a continuous synthesis without
any isolation step.
(27) Blicke, F. F.; Sheets, D. G. J . Am. Chem. Soc. 1949, 71, 2856.
(28) Takeshita, H.; Mametsuka, H.; Motomura, H. J . Heterocycl.
Chem. 1986, 23, 1211.
(29) McKillop, A.; Kemp, D. Tetrahedron 1989, 45, 3299.

4332 J . Org. Chem., Vol. 64, No. 12, 1999
Carpino et al.
method, a small scale test of a similar procedure, but without
removal of acetic acid prior to workup, showed that the
benzothiophenesulfone alcohol was obtained in a yield of about
60%. It is expected that the yield will be improved upon
following the general procedure described. The alcohol was
obtained after recrystallization from CHCl3/hexane (4:1) in the
form of white crystals, mp 113-114 °C; IR (KBr): 3358, 1289,
1148 cm-1; 1H NMR (CDCl3): δ 2.86 (bs, 1), 4.69 (s, 2), 7.05-
7.71 (m, 5). Anal. Calcd for C9H8O3S: C, 55.09; H, 4.11; S,
16.33. Found: C, 54.86; H, 4.11; S, 16.21.
nitromethane, mp 234-235 °C. Anal. Calcd for C13H11NO5S:
C, 53.32; H, 3.78; N, 4.78. Found: C, 53.13; H, 3.78; N, 4.85.
N-(Ben zoth iop h en esu lfon e-2-m eth yl)p ip er id in e (12).
To an ice-cold solution of Bsmoc-Cl (1.22 g, 5 mmol) in 15 mL
of DMF was added piperidine (1.5 mL, 15 mmol) dropwise over
several minutes. After stirring for 1 h, the mixture was poured
slowly into 150 mL of cold water. The resulting precipitate
was washed with water and recrystallized from MeOH to give
473 mg (36%) of the piperidine derivative as white crystals,
1
mp 122.5-124.5 °C (lit.31 mp 124-126 °C); H NMR (CDCl3):
Ben zoth iop h en esu lfon e-2-m eth yl Ch lor ofor m a te. To a
stirred solution of 30.8 g (0.16 mol) of benzothiophenesulfone-
2-methanol in 400 mL of dry THF cooled to -30 °C in a dry
ice-acetone bath was added in one portion 104 g (80 mL, 1.05
mmol) of phosgene which had been condensed into a graduated
cylinder. The solution was allowed to come to room-tempera-
ture overnight. Excess phosgene and solvent were removed
with the aid of a water aspirator, the process being repeated
using additional solvent: 200 mL of THF, 200 mL of THF-
ether (1:1), and 200 mL of ether. Recrystallization of the
residue from dry ether gave 36.1 g of the chloroformate as
δ 1.3-1.9 (m, 6), 2.3-2.7 (m, 4), 3.4-3.6 (d, 2), 6.85-7.10 (t,
1), 7.15-7.9 (m, 4); UV (MeOH) 310.5 nm (log ꢀ 3.484), (DCM)
311.5 nm (log ꢀ 3.454).
Gen er a l P r oced u r es for t h e P r ep a r a t ion of Ben -
zoth iop h en esu lfon e-2-m eth oxyca r bon yl Am in o Acid s.
F or Ava ila bility See F ootn ote a , Ta ble 1. (Mtd 1) Bolin
Tech n iqu e32 via Bsm oc-Cl. To a suspension of 3.87 mmol
of an amino acid in 20 mL of DCM was added in one portion
0.98 mL (7.73 mmol) of chlorotrimethylsilane. The mixture was
then refluxed for 1 h and cooled in an ice bath. Diisopropyl-
ethylamine (1.3 mL, 7.31 mmol) was added slowly followed
by 1 g (3.87 mmol) of benzothiophenesulfone-2-methyl chlo-
roformate. The reaction mixture was allowed to stand at 0 °C
for 20 min and then for 1-1.5 h at room temperature. The
solvent was removed in vacuo and the resulting oil distributed
between 40 mL of ether and 80 mL of 2.5% NaHCO3 solution.
The combined aqueous layers were acidified to pH 2 with
concentrated HCl and extracted with three 30-mL portions of
EtOAc. The extracts were combined and washed with 30 mL
of saturated NaCl and 30 mL of water, dried over MgSO4, and
filtered, and solvent was evaporated. The white residue or oil
was recrystallized from the appropriate solvent or solvent
mixture to give the corresponding Bsmoc-protected amino
acids. See Table 1.
(Mtd 2) Active Ester Tech n iqu e via Bsm oc-OSu in
Aceton itr ile-Wa ter in th e P r esen ce of Tr ieth yla m in e.
To a suspension of 2.96 mmol of the amino acid in 20 mL of
acetonitrile/water (1/1) was added an amount of triethylamine
to give an apparent pH of 9.0. The suspension became clear.
The pH was kept at 8.5-9.0, after the addition of 1 g (2.96
mmol) of Bsmoc-OSu, by adding triethylamine. The uptake of
base ceased after about 15 min. The reaction mixture was
stirred at room temperature for 40-45 min, acidified to pH 5
with 0.1 N HCl or 10% KHSO4, and concentrated in vacuo.
The mixture was diluted with about 5 mL of water and
acidified to pH 2 with 0.1 N HCl. The mixture was extracted
with four 25-mL portions of ethyl acetate. The combined ethyl
acetate extracts were washed several times with saturated
NaCl solution, dried over MgSO4, and filtered, the solvent was
removed in vacuo, and the residue was treated as in method
1.
(Mtd 3) Active Ester Tech n iqu e via Bsm oc-OSu in
Aceton e-Wa ter in th e P r esen ce of Sod iu m Bica r bon a te.
To a solution of the amino acid (1.48 mmol) and NaHCO3 (2.95
mmol) in 10 mL of water was added a solution of ben-
zothiophenesulfone-2-methyl N-succinimidyl carbonate (1.48
mmol) in 10 mL of acetone. The reaction mixture was stirred
overnight at room temperature, diluted with water, and
extracted twice with DCM to remove a small amount of
benzothiophenesulfone-2-methanol and unreacted Bsmoc-OSu.
The aqueous layer was cooled in an ice bath and acidified with
concentrated HCl to pH 2. The resulting white precipitate or
oil was extracted with three 25-mL portions of ethyl acetate.
The combined organic layer was washed with 30 mL of
saturated NaCl solution and 30 mL of water and dried over
MgSO4, the solvent was removed in vacuo, and the residue
was treated as in method 1.
1
white crystals, mp 76-77 °C; IR (KBr): 1772 cm-1; H NMR
(CDCl3): δ 5.3 (d, 2), 7.29 (bs, 1), 7.3-7.89 (m, 4). The
chloroformate is unstable on TLC plates (SiO2). Anal. Calcd
for C10H7ClO4S: C, 46.43; H, 2.73; Cl, 13.71. Found: C, 46.42;
H, 2.69; Cl, 13.82.
N -(Be n zot h iop h e n e su lfon e -2-m e t h yl)-N -su ccin im -
id yl Ca r bon a te. Bsmoc-Cl (8.98 g, 34.8 mmol) was dissolved
in 100 mL of dry DCM. The dicyclohexylamine salt of N-
hydroxysuccinimide30 (10.3 g, 34.8 mmol) was added with
stirring to the solution which was protected from moisture
(CaSO4). After stirring for 20 h, the solution was filtered and
the amine salt washed with two 75-mL portions of DCM. The
combined organic filtrates were washed with saturated NaH-
CO3 (2 × 100 mL) and saturated NaCl (2 × 100 mL), dried
(MgSO4), filtered, and treated with decolorizing carbon. Filtra-
tion and evaporation of solvent gave 9.6 g (81%) of the crude
carbonate which on recrystallization from DCM/hexane or
EtOAc/hexane gave 8.6 g (74%) of the carbonate as white
crystals, mp 170-172 °C; IR (KBr): 1818, 1790, 1742 cm-1
;
1H NMR (CDCl3 + TFA): 2.87 (s, 4), 5.45 (d, 2), 7.73 (m, 5).
Anal. Calcd for C14H11NO7S: C, 49.85; H, 3.28; N, 4.15.
Found: C, 49.61; H, 3.28; N, 4.12.
Ben zoth iop h en esu lfon e-2-m eth yl N-(p-Ch lor op h en yl)-
ca r ba m a te. To a solution of 1 g (3.87 mmol) of Bsmoc-Cl in
20 mL of dry DCM was added at 0 °C portionwise 0.98 g (7.74
mmol) of p-chloroaniline. A white precipitate separated at once.
After addition was complete, the mixture was stirred at 0 °C
for 10 min and for 2 h at room temperature. The white
precipitate was filtered and washed with DCM. The combined
DCM solution was washed with two 25-mL portions of 5% HCl
followed by 25 mL of water and dried over MgSO4. After
filtration, evaporation, and recrystallization from DCM/hexane
there was obtained 1.15 g (85%) of the carbamate as a white
solid, mp 165-166 °C; IR (KBr): 3339, 1700, 1305, 1151 cm-1
;
1H NMR (CDCl3/DMSO-d6): δ 5.25 (s, 2), 7.22-7.76 (m, 9),
9.96 (bs, 1). Anal. Calcd for C18H12ClNO4S: C, 54.94; H, 3.46;
N, 4.00; Cl, 10.13. Found: C, 54.74; H, 3.35; N, 3.90; Cl, 10.35.
The same compound was obtained in 88.6% yield from p-
chlorophenyl isocyanate and benzothiophenesulfone-2-metha-
nol in benzene after 20 h of refluxing.
N-(Ben zot h iop h en esu lfon e-2-m et h oxy)su ccin im id e.
2-(Chloromethyl)benzothiophenesulfone (310 mg, 1.46 mmol)
was dissolved in 4 mL of DMF and 0.3 mL (1.72 mmol) of
DIEA, 0.23 g (1.96 mmol) of N-hydroxysuccinimide was added,
and the mixture was stirred for 3.5 h. Dilution with 80 mL of
water gave a solid which was filtered and washed with H2O,
MeOH-H2O, and acetone to give 278 mg (65%) of the succin-
imide as white crystals, mp 234-235 °C, IR (KBr): 3490, 3049,
2991, 2951, 1710, 1451, 1373, 1305, 1210, 1190, 1148, 1108,
Gen er a l P r oced u r e for P r ep a r a tion of N-(Ben zoth io-
p h en esu lfon e-2-m eth oxyca r bon yl)a m in o Acid Flu or id es.
A solution (or suspension) of Bsmoc-amino acid (1 mmol) in
dry DCM (10 mL) was treated under a nitrogen atmosphere
1
1076; H NMR (CDCl3 + TFA): δ 2.81 (s, 4), 5.05-5.2 (d, 2),
7.2-7.9 (m, 5). The analytical sample was recrystallized from
(31) Bordwell, F. G.; Ross, F.; Weinstock, J . J . Am. Chem. Soc. 1960,
82, 2878.
(30) Paquet, A. Can. J . Chem. 1982, 60, 976.

Bsmoc Amino-Protecting Group
J . Org. Chem., Vol. 64, No. 12, 1999 4333
with cyanuric fluoride (295 µL, 3.5 mmol) and pyridine (80.8
µL, 1 mmol). The reaction mixture was stirred at room
temperature for 1-1.5 h. During the reaction, a white pre-
cipitate separated. The reaction mixture was extracted with
two 20-mL portions of ice-water. The organic layer was dried
over MgSO4 and filtered and solvent removed to give a residue
which was crystallized from DCM/hexane to give the corre-
sponding Bsmoc-amino acid fluoride. The purity of the acid
fluorides was easily checked by converting them to their
methyl esters with dry methanol and checking the product by
TLC. Exceptions to this general procedure were made in the
cases of Bsmoc-Asn(Trt)-F and Bsmoc-Gln(Trt)-F. In both of
these cases it was necessary to use 2 equiv of pyridine and a
lower temperature (-20 °C). Unfortunately, some of the
Bsmoc-amino acid fluorides were not easily crystallized. In
such cases after removal of the DCM, dry ether was added,
the ether removed in vacuo, and the resulting foam (oil) dried
in a vacuum pump for 24 h prior to storage. Storage vials were
flushed with dry N2 and placed in capped bottles flushed with
dry N2 containing a desiccant (CaSO4 or P2O5). Failure to dry
these materials well leads to formation of the corresponding
free acid on attempted long-term storage. See Table 2.
extracted with three 30-mL portions of saturated NaCl, dried
over MgSO4, and filtered and the solvent removed in vacuo to
give a pale yellow oil. A mixture of 511 mg (0.85 mmol) of
Bsmoc-Asn(Trt)-OH and 324 mg (0.857 mmol) of Toppipu in 3
mL of DCM in the presence of 136 µL of DIEA was allowed to
stand for 3 min. The resulting mixture was added to the oily
tryptophan amide and the resulting mixture stirred at room
temperature for 3 h. The solvent was evaporated in vacuo to
give a yellow oil which was chromatographed on silica gel
using EtOAc/CHCl3 (75:25) as an eluent to give 538.5 mg (80%)
of the desired product as an off-white solid. Recrystallization
from Et2O/CH2Cl2 (4:1) gave 444.2 mg (66%) of the dipeptide
amide; mp 138-140 °C; IR (KBr): 3404, 1734, 1670, 1303,
1151 cm-1; 1H NMR (CDCl3): δ 0.12 (m, 5), 0.95 (d, 6), 2.69-
3.26 (m, 4), 4.47 (m, 2), 5.03 (dd, 2), 5.45 (s, 1), 6.54 (d, 1),
6.86-8.15 (m, 28); [R] -28° (c ) 0.5, DMF). Anal. Calcd for
C
50H49N5O7S: C, 69.50; H, 5.72; N, 8.10. Found: C, 69.33; H,
5.49; N, 7.96.
Bsm oc-P h e-P h e-Val-Gly-Leu -Met-OBn . Bsmoc-Leu-F (260
mg, 0.73 mmol) was added to a solution of H-Met-OBn‚TsOH
(200 mg, 0.49 mmol) in 5 mL of dry DCM in the presence of
170 µL of DIEA. The resulting mixture was stirred for 15 min
and then treated with tris(2-aminoethyl)amine (1.5 mL, 20
equiv). The solution was stirred for 10 min, diluted with 30
mL of DCM, extracted with three 30-mL portions of saturated
NaCl, dried over MgSO4, and filtered, and the solvent was
removed in vacuo. The resulting oil was dissolved in 5 mL of
DCM and a new cycle begun. The procedure was repeated four
times using Bsmoc-Gly-F (175 mg, 0.58 mmol), Bsmoc-Val-F
(200 mg, 0.5 mmol), and Bsmoc-Phe-F (twice with 292 mg, 0.58
mmol, in each cycle). After the last coupling, the solvent was
removed, and the resulting oil was chromotagraphed on silica
gel using CHCl3/t-BuOH (6:1) to give 187 mg (37.2%) of the
protected hexapeptide as a white solid, mp 224-225 °C; 1H
NMR (CDCl3): δ 0.88 (m), 1.6-1.9 (m), 2.4-2.58 (m), 3.0 (m),
3.5 (m), 4.7 (m), 5.1 (bs), 7.1-7.6 (m); [R] -38.4° (c ) 0.5, DMF).
Anal. Calcd for C53H64N6O11S2: C, 62.09; H, 6.29; N, 8.20.
Found: C, 62.28; H, 6.49; N, 8.05.
N-(Ben zot h iop h en esu lfon e-2-m et h oxyca r b on yl)p h e-
n yla la n in e ter t-Bu tyl Ester . A solution of 0.5 g (1.93 mmol)
of benzothiophenesulfone-2-methyl chloroformate and 0.497 g
(1.93 mmol) of tert-butyl phenylalaninate hydrochloride in 15
mL of DCM was stirred with 0.324 g of NaHCO3 in 10 mL of
water at room temperature for 3 h. The aqueous phase was
separated, and the organic phase was washed with two 30-
mL portions of 5% HCl. The organic layer was dried over
MgSO4 and filtered, and the solvent was removed in vacuo to
give a colorless oil which was recrystallized from MeOH/H2O
to give 0.708 g (82.8%) of the phenylalanine derivative as white
crystals, mp 112.5-114 °C; IR (KBr): 3421, 1714, 1313, 1155
1
cm-1; H NMR (CDCl3): δ 1.39 (s, 9), 3.06 (d, 2), 4.49 (m, 1),
5.07 (d, 2), 5.38 (d, 1), 7.07-7.70 (m, 10); [R] -15.2° (c ) 1,
DMF). Anal. Calcd for C23H25NO6S: C, 62.29; H, 5.68; N, 3.16.
Found: C, 62.34; H, 5.64; N, 3.13.
Bsm oc-Tyr (t-Bu )-Gly-Gly-P h e-Leu -O-t-Bu . The rapid
solution synthesis was carried out as described above using
Fmoc amino acid fluorides as coupling reagents starting with
1 mmol of leucine tert-butyl ester hydrochloride. After the last
coupling with Bsmoc-Tyr(t-Bu)-F, the solvent was removed in
vacuo and the resulting oil chromatographed on silica gel using
EtOAc as eluent to give 436.5 mg (49.1%) of the protected
pentapeptide as a white solid, mp 133.5-135 °C; 1H NMR
(CDCl3): δ 0.86 (d, 6), 1.3 (s, 9), 1.5 (m, 12), 3.0 (m, 4), 3.9 (m,
4), 4.6 (m, 2), 5.1 (s, 2), 6.6-7.7 (m, 17); [R] -27.2° (c ) 0.5,
DMF). Anal. Calcd for C46H59N5O11S: C, 62.07; H, 6.68; N.
7.87. Found: C, 61.88; H, 6.93; N, 7.73.
Bsm oc-Gln (Dm cp )-OH. To a solution of 15.6 g (42 mmol)
of Z-Glu-OBn in 150 mL of acetonitrile there were added 12.53
g (46 mmol) of Dmcp-NH2‚TsOH, 13.5 g of TBTU (42 mmol),
and 21.9 mL (125 mmol) of DIEA. The mixture was stirred
overnight at room temperature, the solvent removed in vacuo,
and the residue dissolved in 500 mL of EtOAc. The solution
was washed with 5% NaHCO3 and saturated NaCl solution
and dried (MgSO4), and the solvent was removed in vacuo to
give 24 g of the ester as a colorless oil which without
purification was dissolved in MeOH (150 mL) and 0.5 g of 5%
Pd/C catalyst was added. The mixture was shaken under an
atmosphere of H2 (40 psi) for 24 h. Filtration and washing with
MeOH followed by evaporation of solvent gave a residue which
was triturated with ether and filtered to give 9.6 g (∼100%)
of the crude deblocked acid (H-Gln(Dmcp)-OH). To a suspen-
sion of 10 g (43.8 mmol) of the crude acid in 300 mL of
acetonitrile-H2O (1:1) there was added an amount of triethy-
lamine to give an apparent pH of 9.0. The suspension became
clear. There was added 14.8 g (43.8 mmol) of Bsmoc-OSu and
the pH kept at 9.0 by the addition of triethylamine. The uptake
of base ceased after about 15 min. After stirring for 2 h, the
reaction mixture was acidified to pH 5.0 with 10% citric acid
or 0.1 N HCl and extracted with 3 100-mL portions of EtOAc.
The combined organic extracts were washed with saturated
NaCl solution (3 × 50 mL) and dried (MgSO4), and the solvent
Deblock in g of Bsm oc-P h e-OCMe3. A solution of 1 g (2.25
mmol) of tert-butyl benzothiophenesulfone-2-(methoxycarbo-
nyl)phenylalaninate in 10 mL of 50% DCM/TFA was stirred
at room temperature for 4 h. Excess TFA and solvent were
removed in vacuo from a water bath at 45 °C. The resulting
oil was recrystallized from DCM to give 0.739 g (84.9%) of the
colorless acid. The melting point and spectral data (IR, 1H
NMR) obtained for this material agreed with that obtained
directly from phenylalanine (see Table 1).
Meth yl Ben zoth iop h en esu lfon e-2-(m eth oxyca r bon yl)-
p h en ylglycyla la n in a t e. Prepared from Bsmoc-Phg-F as
described in the Supporting Information for the Phe-Ala
analogue in 90% yield, mp 179-180 °C; IR (KBr): 3327, 1736,
1
1654, 1307, 1152 cm-1; H NMR (CDCl3): δ 1.37 (d, 3), 3.64
(s, 3), 4.5 (m, 1), 5.10 (d, 2), 5.25 (d, 1), 6.25 (two d, 2), 7.1-
7.75 (m, 10); [R]+34.4° (c ) 0.5, DMF). Examination of the
1
region near δ 3.73 and 1.24 by 300 MHz H NMR analysis of
the crude sample prior to recrystallization gave no evidence
for the presence of the corresponding DL-diastereomer de-
scribed in the Supporting Information. The methodology and
detection limits have been described previously.33 Anal. Calcd
for C22H22N2O7S: C, 57.63; H, 4.84; N, 6.11. Found: C, 57.54;
H, 4.61; N, 6.04.
Bsm oc-Asn (Tr t)-Tr p -NH-Dm cp . There was added 0.4 g
(0.93 mmol, 1.2 equiv.) of Bsmoc-Trp-F to a solution of 105
mg (0.78 mmol) of dimethylcyclopropylcarbinylamine hydro-
chloride in 5 mL of DCM in the presence of 267 µL of DIEA.
The resulting mixture was stirred for 15 min and treated
directly with tris(2-aminoethyl)amine (2.3 mL, 15 equiv). The
mixture was stirred for 10 min. A white precipitate was formed
which dissolved during subsequent extractions with saturated
NaCl. The reaction mixture was diluted with 40 mL of DCM,
(32) Bolin, D. R.; Sytwu, I.; Humeic, F.; Meienhofer, J . Int. J . Pept.
Protein Res. 1983, 33, 353.
(33) (a) Carpino, L. A.; Chao, H. G.; Nowshad, F.; Shroff, H. J . Org.
Chem. 1988, 53, 6139. (b) Carpino, L. A. J . Org. Chem. 1988, 53, 875.

4334 J . Org. Chem., Vol. 64, No. 12, 1999
Carpino et al.
was removed in vacuo. The residue was flash chromatographed
(Merck Silica Gel 60) using a mixture of EtOAc-MeOH (1:1)
as eluent to give 16.2 g (82.1%) of the Bsmoc acid as a white
solid having an indistinct melting point. Examination of the
NMR spectrum showed that the solid material had crystallized
with 0.5 mol of H2O and 0.5 mol of EtOAc. When attempts
were made to dry the sample, the material became oily as the
EtOAc and water were removed. The NMR and MS samples
Tr eatm en t of 2-(Ch lor om eth yl)ben zoth ioph en esu lfon e
w ith Eth yl Glycin a te. A solution of 0.8 g (3.73 mmol) of
2-(chloromethyl)benzenethiophenesulfone, 1 g (7.17 mmol) of
ethyl glycinate hydrochloride, and 0.93 g (7.17 mmol) of DIEA
in 5 mL of DMF was stirred at room temperature for 3 h. The
reaction mixture was poured into 100 mL of ice cold water and
extracted with four 25-mL portions of EtOAc. The combined
organic extracts were dried (MgSO4), solvent was removed in
vacuo, and the residual oil was purified by column chroma-
tography using EtOAc/CHCl3 (1:1) as eluent. The first fractions
(Rf ) 0.7) were combined, evaporated, and the residue recrys-
tallized from EtOAc/hexane (1:1) to give 480 mg (45.8%) of
adduct 9, as white crystals, mp 78-79 °C; IR (KBr): 3336
1
were analyzed as such with the EtOAc and water present. H
NMR (CDCl3) δ 0.24 (br s, 5), 1.13 (s, 6), 2.1-2.3 (m, 4), 4.14
(m, 1), 4.8-5.3 (m, 2), 7.25-7.60 (m, 5): HRFABMS: calcd
(M + H)+ 451.1539; found: 451.1548.
Bsm oc-Asn (Dm cp )-OH. The asparagine derivative was
obtained as described for the glutamine analogue. In this case
also the solid acid was obtained (yield 87.8%) as a solvate
containing 0.5 mol of water and 0.5 mol of EtOAc which was
(NH), 1734 (CO, ester), 1295, 1147 (SO2) cm-1 1H NMR
;
(CDCl3): δ 1.2 (t, 3), 3.45 (s, 2), 3.87 (s, 2), 4.15 (q, 2), 7.0-7.2
(m, 5). Anal. Calcd for C13H15NO4S: C, 55.50; H, 5.37; N, 4.98.
Found: C, 55.65; H, 5.18; N, 4.95.
1
of indistinct melting point, H NMR (CDCl3) δ 0.22 (br s, 5),
0.94 (s, 6), 2.75 (br s, 2), 4.4 (br s, 1), 4.98 (m, 2), 7.1-7.45 (m,
A second fraction (Rf ) 0.5) was collected and following
recrystallization from Et2O/hexane gave 100 mg of the unre-
arranged adduct 8, mp 124-125 °C; IR (KBr): 3364 (NH), 1742
5); HRFABMS: calc (M + H)+ 437.1383; found 437.1370.
Bsm oc-Ar g(P bf)-OH. To a suspension of 2.33 g (5.46 mmol)
of H-Arg(Pbf)-OH in 34 mL of acetonitrile-H2O (1:1) an
amount of triethylamine was added to give an apparent pH of
9.0. The suspension became clear and was treated with 1.83 g
(5.46 mmol) of Bsmoc-OSu. While stirring, the pH was kept
at 9.0 by the addition of triethylamine. The uptake of base
ceased after about 15 min. The reaction mixture was stirred
at room temperature for 2 h and then acidified to pH 5 with
10% citric acid. Acetonitrile was removed in vacuo at a bath
temperature of 30-35 °C, the mixture cooled in an ice bath,
and the pH brought to 2 with 10% citric acid. The mixture
was extracted with EtOAc (3 × 30 mL), the extracts were
washed with saturated NaCl (2 × 50 mL) and dried (MgSO4),
and the solvent was removed in vacuo. The residue was
recrystallized twice from EtOAc-ether. The acid separated
first as an oil which solidified upon trituration with ether while
cooling. There was obtained 3.3 g (93.1%) of the acid as a white
solid, mp 118-126 °C. As in the cases of the Asn and Gln
derivatives discussed above, the indistinct melting point was
associated with loss of solvent. In this case the solid crystal-
(CO, ester), 1636 (CdC), 1270, 1164 (SO2) cm-1 1H NMR
;
(CDCl3): δ 1.25 (t, 3), 3.64 (s, 1), 4.03 (s, 2), 4.19 (q, 2), 7.26-
7.72 (m, 6). Anal. Calcd for C13H15NO4S: C, 55.50; H, 5.37; N,
4.98. Found: C, 55.26; H, 5.32; N, 4.88.
Loa d in g of Bsm oc Am in o Acid s on to Hyd r oxym eth yl
Resin s. A sample of hydroxyl resin (170 or 340 mg) was
weighed into a tared 5-mL polypropylene syringe fitted with
a coarse polypropylene frit which was mounted on an adapter
equipped with a Teflon stopcock connected to a vacuum source.
The resin was swollen and washed three times with 4 mL of
DMF and three times with 4 mL of DCM. Bsmoc-AA-F (4 eq
based on the claimed functionality of resin used) or Bsmoc-
AA-OH (4 equiv) and an appropriate coupling reagent were
weighed into a 4-mL sample vial and dissolved in solvent (1.0
or 2.0 mL depending on the weight of resin used), and the
calculated volume (liquids) or weight (solids) of base (4 equiv)
was added to the solution. The contents of the vial were
dissolved and added to the resin for a given time period. The
resin was washed with 4-mL volumes of DMF three times,
MeOH once, and DCM three times. The entire sample in the
syringe was then transferred to a vacuum desiccator and dried
overnight. Loadings and levels of racemization observed are
given in Table 3.
1
lized with 0.5 mol of ethyl ether. H NMR (CDCl3): δ 1.42 (s,
6), 1.43-2.0 (m, 4), 2.03 (s, 3), 2.45 (s, 3), 2.50 (s, 3), 2.9 (s, 2),
3.16 (s, 2), 4.24 (s, 1), 5.05 (t, 2), 7.18-7.64 (m, 5); HR-
FABMS: calc (M + H)+ 649.2002; found: 649.1987.
P r ep a r a tion of Bsm oc-Gly-Gly-OEt. A Test for P r e-
m a tu r e Deblock in g. There was added 300 mg (1 mmol) of
benzothiophenesulfone-2-(methoxycarbonyl)glycyl fluoride to
a solution of 117 mg (0.83 mmol) of glycine ethyl ester
hydrochloride in 7 mL of dry DCM in the presence of 0.3 mL
of DIEA. The reaction mixture was stirred for 10 min and the
solvent removed in vacuo to give a pale yellow oil which was
purified by column chromatography using EtOAc with 1% of
HOAc as eluent to give 276 mg (87%) of the desired dipeptide
ester as a colorless oil; IR (neat): 3359 (NH), 1731 (CO, ester,
urethane), 1679 (CO, amide), 1302, 1150 (SO2) cm-1; 1H NMR
(CDCl3): δ 1.25 (t, 3), 3.93 (d, 2), 3.99 (d, 2), 4.15 (q, 2), 5.15
(s, 2), 5.79 (bs, 1), 6.8 (bs, 1), 7.1-7.72 (m, 5). HPLC analysis
of the crude reaction product showed, in addition to the
dipeptide (84.8%, tR ) 15.45), 3.66% of Bsmoc-Gly-OH, pre-
sumably formed via hydrolysis of the acid fluoride and other
minor unidentified impurities but none of the expected adducts
8 (tR ) 18.63 min) and 9 (tR ) 11.08) which would have been
present had the amino acid ester caused deblocking during
the coupling process. Anal. Calcd for C16H18N2O7S: C, 50.26;
H, 4.74; N, 7.32. Found: C, 50.21; H, 4.88; N, 7.03.
Rea ction of Bsm oc-OSu w ith N-Hyd r oxysu ccin im id e
An ion . Bsmoc-OSu (337 mg, 1.0 mmol), HOSu (165 mg, 1.5
mmol), and DIEA (260 µL, 1.5 mmol) were added to 10 mL of
MeCN/H2O (10:1), and the mixture was stirred for 17 h. The
resulting white solid was filtered from the reaction mixture
and washed with water and acetone to give 143 mg (49%) of
Bsm-OSu as a white solid, mp 234-235 °C. The solid was
identified by comparison of its IR and 1H NMR spectra with
the spectra of authentic Bsm-OSu prepared from 2-(chlorom-
ethyl)benzothiophenesulfone as described above.
UV Absor p tion Ch a r a cter istics of Bsm -P ip . N-(Ben-
zothiophenesulfone-2-methyl)piperidine (Bsm-pip) (263.35 mg,
1.0 mmol) was weighed into a tared 10-mL volumetric flask
and diluted to volume with methanol. One milliliter of this
solution was transferred to a 250-mL volumetric flask with a
volumetric pipet, diluted to volume, and used as a stock
solution (standard 0.4 mmol/L). From the stock solution,
standards in the range of 0.08-0.4 mmol were prepared by
serial dilution into 25-mL volumetric flasks. Standard curves
obtained were linear (R2 > 0.999) in absorbance vs concentra-
tion in the ranges of 0.15-1.6 and 0.08-0.4 (mmol/L), respec-
tively. The average slope of three calibration curves (three
separate weighings) was used to determine the absorption
data: λmax ) 310.5 nm, log ꢀ ) 3.484 (MeOH), λmax ) 311.5
nm, log ꢀ ) 3.454 (DCM), R2 > 0.999 for both solvents.
Meth od for Deter m in a tion of Resin Loa d in g. A thor-
oughly dried sample of loaded resin (20-45 mg depending on
expected loading) was weighed into a fritted polypropylene
syringe. The sample was deblocked with 1.00 mL of 2 or 20%
piperidine (v/v) in DMF for 5 or 30 min, respectively. The 2%
solution was used to check for completion of the deblocking/
isomerization process using short reaction times at low base
concentrations. The solution was drained into a 25-mL volu-
metric flask, the resin and syringe were washed with methanol
(20 mL) into the flask, and the solution was diluted to volume.
A solvent blank was prepared by diluting 1.0 mL of deblocking
solution to volume in a 25-mL flask. Absorbance was deter-
mined at 310.5 nm (MeOH) and the loading calculated using
the formula (A/ꢀ × 25 mL)/1000 mL/L/wt of resin in g ) loading
in mmol/g (ꢀ is taken as 3.048 for a concentration of c ) mmol/
L).

Bsmoc Amino-Protecting Group
J . Org. Chem., Vol. 64, No. 12, 1999 4335
Exten t of Ra cem iza tion Du r in g Resin Loa d in g of
Bsm oc Am in o Acid s. Standard solutions of Marfey’s adducts
were prepared by mixing 0.4 mL of 10 mM Marfey’s reagent6
in acetone with 0.4 mL of 10 mM amino acid in 0.1 N NaHCO3.
The same samples of amino acid were used to prepare the
Marfey’s standards as those used for the preparation of the
Bsmoc-protected amino acids and for the subsequent racem-
ization studies. The mixtures were placed in 4-mL capped
sample vials, sonocated in a 40 °C water bath for 1 h,
neutralized with 0.2 mL of 0.2 N HCl, and stored at -5 °C. If
a solid precipitated upon storage, the samples are warmed
prior to use. The standards were diluted with 2 parts of
acetonitrile prior to injection onto a Waters C18 Nova-pak
column, 5-µL injection volume, sensitivity ) 0.3, flow ) 1.0
mL/min, using an acetonitrile/water gradient (0.1% in TFA)
from 10% to 60% acetonitrile over 50 min, with detection at
340 nm. No D,L diastereomer was detected in any of the L,L
standards. Standards of the deblocking base adducts (from
piperidine and morpholine) were also prepared using the same
method.
A weighed sample of loaded resin (20-50 mg, depending
on the determined loading) in a fritted 5-mL syringe was
deblocked for 30 min using 20% piperidine or 20% morpholine
(v/v) in DMF (the retention times of both diasteriomers
expected are checked in relation to the piperidine and mor-
pholine standards as these peaks may overlap). The resin was
washed with DMF, MeOH, DMF, MeOH, DCM, MeOH, and
finally DCM twice. The N-deblocked amino acid was then
cleaved from the resin using neat TFA for 90 min. The TFA
solution of the free amino acid was drained into a small round-
bottomed flask and the syringe and resin washed with several
portions of TFA. The combined TFA filtrates were then
evaporated to dryness, the residue was dissolved in 400 µL of
10 mM NaHCO3, and the solution was diluted with 200 µL of
acetone and treated with 200 µL of 10 mM Marfey’s reagent
in acetone. The mixture was sonocated and neutralized exactly
as described for the standards above. The samples were
injected directly with no prior dilution for examination by
HPLC. The amount of racemization was determined by locat-
ing the retention times of the L,L- and D,L-diasteriomers by
comparison with the standards and integrating the corre-
sponding peaks from the resin samples followed by coinjection
of the D,L-standard to confirm the position of the D,L-peak (see
Table 4). If the piperidine adduct had the same or a close
retention time relative to either diasteriomer, morpholine was
substituted as deblocking base since the corresponding adduct
has a considerably shorter retention time in the separation
system used. Attempts to remove totally the deblocking base
from the resin prior to cleavage by additional washes was not
possible, a fact which has been noted in the literature
previously for piperidine.34 The amount of racemization is
expressed as % D.
Gen er a l P r oced u r e for th e Solid -P h a se Syn th esis of
Mod el P ep tid es. Manual syntheses were performed in 5-mL
fritted polypropylene syringes. A 200-mg sample of resin was
weighed into a syringe which was connected to a stoppered
vacuum source. The resin was swollen and washed with DCM
(3 × 4 mL) and DMF (3 × 4 mL). The solvent was removed by
aspiration and the protecting group attached to the first amino
acid deblocked by means of a solution of amine in DMF or
DCM (4 mL). At no time during the deblocking process was
the resin drained dry of deblocking solution. The deblocking
solution and wash solvents are most conveniently handled in
500-mL wash bottles. The resin was washed five times with
4-mL portions of DMF, the duration of each wash being 1 min.
The Bsmoc-AA-F or the combination of Bsmoc-AA-OH’s and
coupling reagent (4 equiv) were preweighed into 4-mL sample
vials and then dissolved with a freshly prepared 1.0-mL stock
solution of DMF containing the desired amount of tertiary
amine catalyst. The contents of the vial were thoroughly mixed
and pipetted into the syringe. If uronium salts were used as
coupling reagents the coupling solution was allowed to pre-
activate in the vial prior to addition to the resin. The coupling
was performed for a given time and the resin again was
washed as described above. At this point the cycle of deblock-
ing/washing/coupling/washing was repeated until the desired
sequence of amino acids was obtained. The Bsmoc group on
the amino terminus was then deblocked as described above.
The resin was thoroughly washed with DMF (3 × 4 mL),
MeOH (4 mL), DMF (3 × 4 mL), and DCM (4 × 4 mL), the
duration of each wash being 1.5 min. The peptide was cleaved
from the resin along with any side chain protecting groups by
90-min treatment with 10% TFA/H2O. The TFA solution of
the peptide was drained into a 50- or 100-mL round-bottomed
flask and the syringe and resin washed with TFA (3 × 3 mL).
The solvent was removed from the combined filtrate and
washes in vacuo without heating. The concentrated peptide
solution was cooled in a dry ice bath, and cold anhydrous ether
was added to precipitate the TFA salt of the peptide. If no
precipitate appeared the ether was evaporated and the pre-
cipitation process repeated. The precipitated peptide was
separated from ether by centrifugation in a tared centrifuge
tube and the resulting peptide pellet washed with several
portions of fresh ether by breaking up the pellet with a Pasteur
pipet in order to resuspend it in fresh ether. Finally the peptide
was dried in a desiccator. The crude peptide was used directly
for HPLC studies. No corrections were made for different
epsilon values of the components in the crude HPLC profile
(λ ) 220 nm), although the HPLC of the crude peptide was
also examined at longer wavelengths (λ ) 280 or 310 nm) in
order to search for the presence of residual protecting groups
or premature deblocking products.
Com p a r ison of th e Bsm oc Deblock in g/Rea r r a n gem en t
P r ocess for Va r iou s Am in es. Stock solutions of organic
bases (piperidine, piperazine, morpholine, and ethanolamine)
were prepared by weighing 0.733 mmol of base into tared 2.00-
mL volumetric flasks and diluting to volume with CDCl3 or
DMF. Each solution (0.60 mL, 0.22 mmol base) was pipetted
into a vial containing 44 mg (0.1 mmol) of Bsmoc-Phe-O-t-Bu
UV Mon itor in g of th e Deblock in g/Sca ven gin g P r ocess
for Bsm oc-Gly-O-t-Bu w ith P ip er id in e. The extinction
coefficient of Bsmoc-Gly-O-t-Bu (ꢀ ) 2611, DCM) at 310 nm
was determined as described for Bsm-Pip. To three 25-mL
volumetric flasks were added stock solutions of Bsmoc-Gly-
O-t-Bu (5.00 mL, 1.97 mmol/L) in DCM. Thereafter, 5-mL
portions of stock solution containing 1.9, 3.8, or 19 equiv of
piperidine were added to each flask which were then diluted
to volume. Using the corresponding concentration of piperidine
solution as a reference blank, the solution was scanned from
245 to 320 nm to monitor the deblocking/scavenging process.
Deblocking was noted by an initial decrease in absorbance in
the region of 310 nm followed by a slow increase as the initial
deblocking intermediate isomerized to the final product. The
extent of isomerization was estimated by calculating the
expected absorbance of a 0.3936 mmol/L solution of Bsm-Pip
at λ ) 311.5 nm (the initial reaction mixture is 0.3936 mmol/L
in Bsmoc-Gly-O-t-Bu). For the results, see Figure 2a,b in the
Supporting Information.
1
and H NMR analysis used to follow the deblocking/rearrange-
ment process on a Hitachi R-1200 spectrometer at 60 MHz.
In the case of runs in DMF the solvent proton peak at δ 8.05
was used as reference rather than TMS which was used in
CDCl3. Deblocking was considered complete when the meth-
ylene protons of the Bsmoc residue (δ 5.075, 5.095, d, 2) were
no longer observed and rearrangement when the terminal
vinyl protons (δ 6.3-6.5, m, 2) of the labile intermediate were
no longer observed. The results are given qualitatively in the
text.
Assem bly of Toxin 2 of A. a n d r octon u s Hector (13). (1)
Via Bsm oc Ch em istr y. A PAC-PEG-PS resin (2.5 g, 0.325
mmol) with a claimed functionality of 0.13 mmol/g was washed
and swollen in DCM. A solution of 460 mg (1.3 mmol) of
Bsmoc-Ile-F and collidine (0.17 mL, 1.3 mmol) dissolved in
sufficient DCM to cover the swollen resin was added. The
mixture was kept for 19 h with occasional agitation. The resin
was washed with DCM (5 × 35 mL) and capped with DCM/
(34) Adamson, J . G.; Hoang, T.; Crivici, A.; Lajoie, G. A. Anal.
Biochem. 1992, 202, 210.

4336 J . Org. Chem., Vol. 64, No. 12, 1999
Carpino et al.
IR (neat) 2983 (CH), 1850 (COF), 1795 (CO), 1712 (CO) cm-1
Ac2O/pyridine 5/1/1, 30 mL) for 14 h. The capped resin was
thoroughly washed with DCM (5 × 35 mL), MeOH (2 × 35
mL), and DCM (4 × 35 mL) and dried in a desiccator. Two
samples (42.0 and 42.5 mg) of the resin were deblocked with
1.0 mL of 20% piperidine in DMF for 30 min, and the extent
of loading was determined by UV analysis. The loadings
observed were 0.096 mmol/g for both samples. Following the
general procedure, a 700-mg sample of the above-described
resin was used to assemble the hexapeptide under the follow-
ing conditions: (a) 5-min deblocking with 2% piperidine/DMF;
(b) 30-min coupling via Bsmoc-AA-F (4 equiv) without base;
(c) lysine, tyrosine, and aspartic acid protected via Boc, t-Bu
ether, and t-Bu ester groups, respectively. Before removal of
the Bsmoc group from valine, the resin was divided in half,
one portion being deblocked for 5 min, the other for 19 h with
2% piperidine in DMF. In the latter case (Figure 1a) the major
impurity was the Asu-containing peptide (peak eluting near
16 min). In a second run the couplings were performed using
4 equiv of Bsmoc-AA-F and 1 equiv of collidine in DMF for 20
min. Deblockings were carried out as above except that 20%
morpholine was used for deblocking. An acceptable synthesis
resulted. See Figure 1b.
(2) Via F m oc Ch em istr y. In this case the assembly
modifications were as follows: (a) 10-min deblocking with 20%
piperidine/DMF; (b) 15-min coupling via HATU(4)/Fmoc-AA-
OH(4)/TMP(8) with a 3-min preactivation period. As before,
two samples of the resin were deblocked for 5 min (not shown)
and 19 h. The numerous side products seen after 19 h with
retention times (min) in parentheses included aminosuccin-
imide (16.1), R-L-piperidide (19.2), R-D-piperidide (20.4), â-L-
piperidide (21.3), and â-D-piperidide (22.3) (Figure 1c). A third
sample treated for 19 h with 20% morpholine showed little
effect (Figure 1d) in agreement with the successful use of these
conditions for the Bsmoc synthesis.
Syn th esis of Hep ta p ep tid e 23 via Bsm oc-AA-OH a n d
HATU. To a solution of 67.75 mg (0.5 mmol) of (dimethylcy-
clopropylcarbinyl)amine hydrochloride were added 0.270 mL
(3.1 equiv) of DIEA and 213.12 mg (0.5 mmol) of Bsmoc-Trp-
OH dissolved in 10 mL of dry DCM. After cooling in an ice
bath, 190.5 mg (0.5 mmol) of HATU was added in one batch.
The course of the reaction was monitored by TLC and found
to be complete within 30 min after which 1.5 mL (20 equiv) of
TAEA was poured into the solution. After 10 min, deblocking
was complete, and the mixture was washed with three 10-mL
portions of saturated NaCl solution. After drying over MgSO4
and evaporation of solvent a new coupling cycle was started.
The remaining cycles were executed in the same way except
that only 2.1 eq of DIEA were used considering that in step I
1 equiv of DIEA was used to neutralize the hydrochloride salt.
The next three amino acids were added in the same manner;
however, because of expected loss of material, for the fifth,
sixth, and seventh amino acids, the amount of Bsmoc amino
acid and HATU was progressively reduced to 0.475, 0.47, and
0.45 mmol, respectively. The amount of TAEA used was not
changed.
In its fully protected form the heptapeptide was purified by
column chromatography (40 g of SiO2, 2 × 50 cm glass column).
There was obtained 280 mg (37%) of the heptapeptide as a
white solid, mp 143-145 °C. MS(FAB) m/e calcd 1557.9, found
1557.7 (M + Na)+. A very intense peak at m/e 243.1 was
observed for the stable trityl cation. A second independent MS
determination showed m/e 1535.8, calcd 1535.9 (M + H)+ along
with the trityl cation at 243.1. Amino acid analysis: found
(expected): Asp 0.65 (1.0), Thr 2.15 (2.0), Pro 1.17 (1.0), Leu
0.99 (1.0), Phe 1.04 (1), Trp not determined. Anal. Calcd for
;
1H NMR (CDCl3) δ 1.62 (s, 9), 1.9-2.9 (m, 4 H), 4.85 (t, 1 H).
Meth od B (via BOC-p Glu -OH‚DCHA). A solution of 201
mg (0.4 mmol) of BOC-pGlu-OH‚DCHA was dissolved in 10
mL of dry DCM, and the solution was cooled to -30 °C and
treated with 0.035 mL (1.1 equiv) of pyridine and 0.101 mL (3
equiv) of cyanuric fluoride. After 3 h the mixture was washed
with two 10-mL portions of ice cold water. The fluoride was
obtained as an oil in a yield of 85%, IR(neat): 1850 (COF),
1
1795 (CO), 1712 cm-1 (CO); H NMR (CDCl3): δ 1.62 (s, 9),
1.9-2.9 (m, 4), 4.85 (t, 1). The fluoride was not further purified
but used as such in the acylation of the heptapeptide.
N-(Ben zot h iop h en esu lfon e-2-m et h oxyca r b on yl)p h e-
n yla la n in e Dicycloh exyla m m on iu m Sa lt. A suspension of
0.5 g (1.29 mmol) of Bsmoc-Phe-OH in 20 mL of dry ether and
3 mL of dry THF was treated with 0.308 mL (1.2 equiv) of
dicyclohexylamine dropwise. After a few min the solution
became clear, and a few moments later a white precipitate
separated. After stirring for 4 h at room temperature and
storage in the refrigerator overnight filtration gave 690 mg
(95%) of the salt as a white solid, mp 164 °C; 1H NMR (CDCl3)
δ 1.1-2.35 (bm, 1), 3.2 (d, 2), 3.1 (t, 1), 5.1 (s, 2), 7.0-7.7 (m,
10). Anal. Calcd for C31H40O6N2S: C, 65.47; H, 7.09; N, 4.93.
Found: C, 65.07; H, 7.03; N, 4.90.
N-(Ben zot h iop h en esu lfon e-2-m et h oxyca r b on yl)p h e-
n yla la n in e F lu or id e via Bsm oc-P h e-OH‚DCHA. A solution
of 300 mg (0.528 mmol) of Bsmoc-Phe-OH‚DCHA in 20 mL of
dry DCM was treated with 0.046 mL (1.1 equiv) of pyridine
and 0.134 mL (3 equiv) of cyanuric fluoride. The reaction was
found to be complete within 75 min. After washing with two
20-mL portions of ice cold water, the organic layer was dried
over MgSO4 and evaporated in vacuo and the residue recrys-
tallized from DCM/hexane to give the fluoride in 97% yield as
a white solid, mp 129 °C. The IR and NMR spectral data
agreed with data obtained for a sample of the fluoride obtained
from the free acid (mp 174 °C). In the latter case the acid
fluoride was obtained either as a clear white foam, mp 50 °C
or a white solid, mp 132-133 °C, IR (KBr): 3345 (NH), 1846
1
(COF), 1731 (CO), 1303, 1151 (SO2) cm-1; H NMR (CDCl3):
δ 3.2 (d, 2), 4.8 (m, 1), 5.1 (s, 2), 5.3 (d, 1), 7.1-7.7 (m, 10).
Anal. Calcd for C19H16FNO5S: C, 58.60; H, 4.14; N, 3.60.
Found: C, 58.82; H, 4.26; N, 3.64.
F u lly P r otected Octa p ep tid e, BOC-p Glu -Leu -Th r (t-
Bu )-P h e-Th r (t-Bu )-P r o-Asn (Tr t)-Tr p -NHDm cp (27). To a
solution of 70 mg (0.0456 mmol) of the heptapeptide 23 in 20
mL of acetonitrile there was added 684 mg (15 equiv of NH, 1
mequiv/g of NH) of a piperazine-functionalized silica gel
reagent. The reaction mixture was gently rotated on the shaft
of a rotary evaporator while monitoring the course of the
reaction by TLC analysis (elution with 1% HOAc in EtOAc).
Deblocking was complete within 60 min. The silica gel was
filtered and washed with 100 mL of DCM. Evaporation of the
solvent gave a white solid, which was redissolved in 10 mL of
dry DCM and treated with 0.0087 mL (1.1 equiv, 0.05 mmol)
of DIEA and 5.3 mg (0.5 equiv) of freshly prepared BOC-pGlu-
F. To avoid using an excess of the fluoride, which has not been
obtained in a completely pure state, the needed full equivalent
was approached stepwise. After further additions of 0.4, 0.2,
and 0.2 equiv (total 1.3 equiv) of the fluoride, the coupling was
found to be complete. The solvent was evaporated, the residue
was redissolved in 1 mL of DCM, and the solution was applied
to a preparative TLC plate. The first elution was made with
DCM/acetone (2:1). The collected material still showed an
impurity spot on TLC. A second preparative TLC plate was
loaded and eluted with EtOAc/1% glacial acetic acid. Although
a clear separation was made, the same impurity spot was
observed again when the material was tested a few hours after
the pure product had been collected. Evaporation of the solvent
led to the isolation of 36.8 mg (53%) of the peptide as an off-
white residue. HPLC analysis on a 4 µm, 3.9 × 150 mm, 60 Å
Waters Nova Pak C18 column using as eluent (A) CH3CN/(B)
H2O 0.1% TFA, detection at 220 nm, flow rate 1 mL/min using
the following gradient: (min/A/B) 1/10/90, 20/90/10, 25/10/90,
showed a purity of 89%; tR 26.10 min; MS/(FAB) m/e calcd
1546.9. Found: 1546.8 (M + Na)+. Also observed were the
C
86H106N10O14S: C, 67.25; H, 6.96; N, 9.12. Found: C, 67.00;
H, 7.04; N, 8.91.
Boc-p yr oglu t a m ic Acid F lu or id e 26. Met h od A (via
Boc-p Glu -OH). 210.4 mg (0.918 mmol) of Boc-pGlu-OH was
dissolved in 10 mL of dry DCM and treated with 0.077 mL
(1.05 equiv) of pyridine and 0.234 mL (3 equiv) of cyanuric
fluoride at -30 °C. After 3 h the mixture was washed with
two 20-mL portions of ice cold water. Recrystallization at-
tempts failed, the product being obtained as an oil in 89% yield;

Bsmoc Amino-Protecting Group
J . Org. Chem., Vol. 64, No. 12, 1999 4337
expected trityl cation peak at m/e 243.1 and an (M - BOC)
peak at m/e 1446.7. Amino acid analysis: Found (expected):
Asp 0.83 (1), Glu 1.17 (1), Thr 2.45 (2), Pro 1.35 (1), Leu 1.16
(1), Phe 1.32 (1).
for C39H38N2O7S: C, 69.02; H, 5.60; N, 4.13. Found: C, 68.99;
H, 5.66; N, 4.14.
Bsm oc-Leu -P h e-OH. Treatment of 2 mmol of H-Phe-
OCMe3‚HCl with 4 mmol of DIEA and 2.2 mmol of Bsmoc-
Leu-F in 20 mL of dry DCM in the normal manner followed
by treatment of the crude protected dipeptide with 50% TFA/
DCM at room temperature for 2 h gave in 78.5% yield after
recrystallization from DCM/hexane the dipeptide acid as a
white solid, mp 91-93 °C, 1H NMR (CDCl3) δ 1.05 (d, 6), 1.2-
1.3 (m, 3), 3.05 (d, 2), 4.1-4.5 (m, 2), 5.08 (s, 2), 5.3 (s, 1), 7.3-
8.1 (m, 10). Anal. Calcd for C25H28N2O7S: C, 60.00; H, 5.60;
N, 5.60. Found: C, 59.64; H, 5.65; N, 5.54.
H-Leu -P h e-OF m ‚TF A. By treatment of H-Phe-OFm‚TFA
with Boc-Leu-F there was obtained in the normal manner the
protected dipeptide which with 50% TFA/DCM gave the TFA
salt in 79.3% yield as a white solid, mp 186-187 °C; 1H NMR
(CDCl3-DMSO-d6) δ 1.0 (d, 6), 1.2-1.4 (m, 3), 3.1 (d, 2), 4.1-
4.6 (m, 5), 7.1-7.9 (m, 13). Anal. Calcd for C31H33F3N2O5‚1/2
H2O: C, 64.25; H, 5.69; N, 4.84. Found: C, 64.68; H, 5.65; N,
4.84.
Octa p ep tid e 20 a s TF A Sa lt. The protected octapeptide
27 (25 mg, 0.0157 mmol) was treated with a deblocking cocktail
having the composition 88% TFA, 5% phenol, 5% water, 2%
triisopropylsilane. The course of the reaction was followed by
HPLC. Deblocking was complete after 1 h, the reaction mixture
was evaporated in vacuo, and the residue was precipitated
with excess ether and collected by centrifugation. The washing
process was repeated three times. The light brown octapeptide
salt was dried in a vacuum. HPLC conditions used for analysis
were the same as described for the protected octapeptide. The
tR of the main peak was 17.45 min with integration giving an
approximate purity of 70%. MS(FAB): calcd m/e 988.12; found
m/e 988.4 (M + Na)+. Amino acid analysis: found (expected):
Asp 0.63 (1), Glu 1.10 (1), Thr 2.05 (2), Pro 1.13 (1), Leu 1.00
(1), Phe 1.08 (1).
Bsm oc-Leu -Leu -Leu -OF m . To a solution of 0.196 g (1
mmol) of FmOH and 0.178 mL (1 mmol) of DIEA in 10 mL of
dry DCM was added 0.427 g (1.2 mmol) of Bsmoc-Leu-F. After
the reaction mixture had been stirred at room temperature
for 5 h, TLC analysis showed no trace of alcohol remaining.
The solution was diluted with 20 mL of DCM and washed with
1 M HCl solution (2 × 10 mL), saturated NaHCO3 solution (2
× 10 mL), and saturated NaCl solution (2 × 10 mL). The dried
(MgSO4) solution was rotoevaporated and the residual sticky
residue dissolved in 10 mL of 2% TAEA in DCM. After stirring
at room temperature for 0.5 h, TLC analysis showed the
absence of starting material. Addition of 30 mL of DCM was
followed by washing with saturated NaCl (2 × 15 mL), water
(2 × 15 mL), and saturated NaCl (15 mL). The dried (MgSO4)
solution was filtered and rotavaped and the residue dissolved
in 10 mL of dry DCM and treated with 0.178 g (1 mmol) of
DIEA followed by 1.1 mmol of Bsmoc-Leu-F. The mixture was
stirred for 1 h at room temperature, diluted with 20 mL of
DCM, and worked up as described above. After removing the
solvent, the resulting dipeptide was deblocked by means of 2%
TAEA/DCM as described previously and coupled with Bsmoc-
Leu-F, again as described above. Following the usual workup
and removal of solvent, the crude tripeptide was purified by
column chromatography on silica gel with elution by EtOAc/
hexane (6/4) and final crystallization from DCM/hexane to give
0.51 g (67.4%) of the protected tripeptide as a white powder,
mp 94-95 °C; IR (KBr) 3282 (NH), 1737 (CO, ester), 1707 (CO,
urethane), 1643 cm-1 (CO, amide); 1H NMR (CDCl3) δ 0.8-
1.05 (m, 18), 1.4-1.8 (m, 9), 4.1-4.3 (m, 3), 4.45-4.55 (m, 3),
5.2 (q, 2), 5.5 (d, 1), 6.7 (t, 2), 7.15-7.8 (m, 13); HPLC (linear
gradient 40/90 CH3CN/H2O/0.1% TFA in 20 min, Nova Pak
C-18 column, 4 µm, 60 Å, 3.9 × 150 mm, flow rate 1 mL/min,
254 nm): single peak at tR ) 15.05 min. Anal. Calcd for
Select ive Deblock in g of Bsm oc-Leu -P h e-OF m . (A)
Rem ova l of th e F m Resid u e. A solution of 0.25 mmol of the
protected dipeptide in 2 mL of 10% diisopropylamine in DMF
was allowed to stand at room temperature with samples being
removed from time-to-time, diluted with CH3CN/H2O, and
injected onto an HPLC column using the conditions described
above which gave the following retention times: Bsmoc-Leu-
Phe-OFm, 18.1 min; Bsmoc-Leu-Phe-OH, 12.3 min; H-Leu-
Phe-OFm, 14.8 min; H-Leu-Phe-OH, 9.2 min. After 5 min the
starting material peak at 18.1 min began to drop and a new
band at 17.1 due to dibenzofulvene began to increase along
with the desired product at 12.3 min. These bands underwent
gradual changes until after 60 min the starting dipeptide had
disappeared completely with only two major peaks being
present: that for product, Bsmoc-Leu-Phe-OH at 12.3 min and
the dibenzofulvene band at 17.1 min. The HPLC picture did
not change after 24 h, showing that the Bsmoc residue was
not affected by diisopropylamine.
(B) Rem ova l of th e Bsm oc Resid u e. As already described
in the case of the leucine trimer, 2% TAEA/DCM gave, after
about 20 min, mainly H-Leu-Phe-OFm (tR ) 14.8 min) along
with a peak at 17.6 min believed to be due to the TAEA/Bsmoc
adduct. With 2% piperidine both Bsmoc and Fm groups were
completely removed after 5 min.
F m oc-Leu -OBsm . To a solution of 1.1 g (3 mmol) of Fmoc-
Leu-OH and 0.59 g (3 mmol) of benzothiophenesulfone-2-
methanol in 30 mL of dry THF was added 0.618 g (3 mmol) of
DCC at 0 °C. After stirring at 0 °C for 1 h and at room-
temperature overnight, the solvent was removed by means of
a rotary evaporator, 10 mL of EtOAc added, the mixture
filtered to remove DCU, the organic layer washed with 10%
citric acid, saturated NaHCO3, and saturated NaCl (2 × 20
mL each) and dried (MgSO4), the solvent again evaporated,
and the residue flash chromatographed using EtOAc/hexane
(6/4) to give 1.23 g (77.3%) of the ester as a foamy white solid,
mp 65-68 °C, IR(KBr) 3486 (NH), 1751 (CO, ester), 1720 cm-1
(CO, urethane); 1H NMR (CDCl3) δ 0.9 (d, 6), 1.6-0.18 (m, 3),
4.25-4.50 (m, 4), 5.16 (s, 2), 5.21 (d, 1), 7.1-7.7 (m, 13). Anal.
Calcd for C30H29NO6S: C, 67.80; H, 5.46; N, 2.64. Found: C,
67.89; H, 5.60; N, 2.50.
F m oc-P h e-Leu -OBsm . A solution of 0.531 g (1 mmol) of
Fmoc-Leu-OBsm in 10 mL of 30% tert-butylamine in CH3CN
was allowed to stand at room temperature for 10-15 min (a
precipitate separated during the deblocking process). The
solvent was removed with a rotary evaporator at room tem-
perature and the oily residue dissolved in 10 mL of a mixture
of DCM and DMF (1/1). The solution was cooled to 0 °C and
0.387 g (1 mmol) of Fmoc-Phe-OH, 0.35 mL (2 mmol) of DIEA,
and 0.38 g (1 mmol) of HATU added. The reaction mixture
was stirred at 0 °C for 30 min and at room temperature for 30
min. The solution was diluted with 50 mL of EtOAc, washed
with 10% citric acid, saturated NaHCO3, and saturated NaCl
(2 × 10 mL each), and dried (MgSO4), and the solvent was
removed in vacuo. The residue was dissolved in 3 mL of DCM
and the solution diluted with 100 mL of hexane. The precipi-
C
42H51N3O8S: C, 66.55; H, 6.78; N, 5.54. Found: C, 66.55; H,
6.71; N, 5.51.
Bsm oc-Leu -P h e-OF m . Following the method described for
the synthesis of Bsmoc-Leu-OFm, BOC-Phe-OFm, mp 131-
132 °C, 1H NMR (CDCl3) δ 1.5 (s, 9), 3.1 (d, 2), 4.1-4.6 (m, 4),
6.9-7.8 (m, 13), was obtained from BOC-Phe-F and FmOH.
The crude product was treated with 50% TFA/DCM at room
temperature for 2 h to give after recrystallization from EtOH/
1
ether the TFA salt of H-Phe-OFm, mp 151-152 °C, H NMR
(CDCl3-DMSO-d8) δ 3.3 (d, 2), 4.1-4.6 (m, 4), 6.1 (br s, 1), 7.2-
8.1 (m, 13). To the crude TFA salt (5 mmol) in 30 mL of dry
DCM was added 5 mmol of DIEA followed by 5 mmol of Bsmoc-
Leu-F. After workup as described for the Leu-Leu analogue,
chromatographic purification on silica gel with elution by
EtOAc/hexane (6/4) gave in 73.4% yield the dipeptide as a
yellowish white solid, mp 135-137 °C, IR (KBr) 3418 (NH),
1
1735 (CO, ester), 1678 (CO, amide); H NMR (CDCl3) δ 1.05
(d, 6), 1.3-1.5 (m, 3), 3.1 (d, 2), 4.1-4.6 (m, 5), 5.1 (s, 2), 5.4
(d, 1), 6.5 (d, 1), 7.0-7.9 (m, 18); HPLC (linear 10/90 CH3CN/
H2O/0.1% TFA in 20 min followed by 5 min at 90% CH3CN,
Nova Pak C-18 column, 4 µm, 60 Å, 3.9 × 150 mm, flow rate
1 mL/min, 254 nm): single peak at tR ) 18.1 min. Anal. Calcd

4338 J . Org. Chem., Vol. 64, No. 12, 1999
Carpino et al.
tate was filtered and washed with 30 mL of hexane to remove
DBF. Since TLC showed an extra spot due to unreacted Fmoc-
Phe-OH, the product was flash chromatographed using EtOAc/
hexane (6/4) as eluent. Following removal of solvent, recrys-
tallization of the residue gave 0.49 g (72.3%) of the dipeptide
as a white solid, mp 78 °C; 1H NMR (CDCl3) δ 0.86 (d, 6), 1.5-
1.7 (m, 3), 3.1 (d, 2), 4.1-4.6 (m, 5), 5.12 (s, 2), 5.3 (d, 1), 6.4
(d, 1), 7.0-7.7 (m, 19). Anal. Calcd for C39H38N2O7S: C, 69.03;
H, 5.61; N, 4.13. Found: C, 69.02; H, 5.62; N, 4.07.
(m, 2), 7.0-7.8 (m, 13). Anal. Calcd for C41H41N3O8S‚1/2 H2O:
C, 66.13; H, 5.65; N, 5.65. Found: C, 66.13; H, 5.77; N, 5.62.
Selective F m oc-Deblock in g of F m oc-P h e-Leu -OBsm .
To a solution of 34 mg of the dipeptide dissolved in 0.5 mL of
a solvent containing a specific deblocking amine HPLC analy-
sis using a Nova Pak 4 µm, 60 Å, C-18 silica column (3.9 ×
150 mm) using a PDA detector, 220 and 254 nm, flow rate 1
mL/min with a linear gradient 10/90 in 20 min (CH3CN/H2O/
0.1% TFA). Under these conditions, the retention times in
minutes observed for authentic samples of the starting mate-
rial and the various reaction products were as follows: Fmoc-
Phe-Leu-OBsm, 20.1; DBF, 19.7; Fmoc-Phe-Leu-OH, 17.1;
H-Phe-Leu-OBsm, 11.9; H-Phe-Leu-OH, 7.8. Authentic samples
of these various deblocking products were obtained as fol-
lows: (a) Fmoc-Phe-Leu-OCMe3/TFA; (b) Boc-Phe-Leu-OBsm/
TFA; (c) Fmoc-Phe-Leu-OH/Et2NH/CH3CN. For the results of
the deblocking tests, see Table 5.
F m oc-Gly-P h e-Leu -OBsm . A solution of 0.339 g (0.5
mmol) of Fmoc-Phe-Leu-OBsm in 5 mL of 30% tert-butylamine
in acetonitrile was allowed to stand at room temperature for
10 min. TLC and HPLC analysis showed that the Fmoc group
had been completely deblocked at this point. Solvent was
removed immediately using an efficient double-trapped oil
pump. DCM was added twice and the solvent removed a second
and a third time in the same manner. The residue was
dissolved in 5 mL of DCM/DMF (1/1) and the mixture cooled
to 0 °C and treated with 0.15 g (0.5 mmol) of Fmoc-Gly-OH,
0.174 mL (1 mmol) of DIEA, and 0.19 g (0.5 mmol) of HATU.
After 30 min at 0 °C and 30 min at room temperature, 50 mL
of EtOAc was added and the solution washed with 10% citric
acid, saturated NaHCO3, and saturated NaCl (2 × 10 mL
each). The solvent was removed with a rotary evaporator in
vacuo, the residue dissolved in 2 mL of DCM, and 50 mL of
hexane added. The precipitate was filtered and washed with
hexane (2 × 10 mL) to give 0.295 g (80.4%) of the tripeptide
as a white solid for which HPLC analysis showed in addition
to the tripeptide at tR 18.5 min (89.6%), two extra peaks at tR
13.2 min (8.5%) and 19.5 min (residual DBF not removed by
the hexane washes, amount 1.9%). An analytical sample was
obtained by flash column chromatography over silica gel using
as eluent EtOAc/hexane (7/3). Following chromatography,
recrystallization from DCM/hexane gave the pure tripeptide
in 72.4% yield as a white solid, mp 108-109 °C; 1H NMR
(CDCl3) δ 0.84 (d, 6), 1.5-1.8 (m, 3), 3.06 (d, 2), 3.8 (d, 2), 4.2
(t, 1), 4.36 (d, 2), 4.5 (m, 1), 4.7 (m, 1), 5.09 (s, 2), 5.6 (t, 1), 7.1
Ack n ow led gm en t. We are indebted to the National
Science Foundation (NSF CHE-9003192), the National
Institutes of Health (GM-09706), and Research Corpo-
ration Technologies for support of this work. Millipore,
Inc. and Perseptive Biosystems, Inc. kindly supplied the
HOAt and HATU used in this study as well as various
PEG-PS resins. The National Science Foundation is also
thanked for grants used to purchase the high-field NMR
spectrometers used in this research. The Nebraska
Center for Mass Spectrometry carried out the FAB mass
spectral analyses.
Su p p or tin g In for m a tion Ava ila ble: Model experimental
procedures, stability studies, Figure 2, and confirmatory IR,
NMR, MS, UV, and amino acid analysis data. This material
is available free of charge via the Internet at http://pubs.acs.org.
J O982140L