Total synthesis of grassystatin A, a probe for cathepsin e function

Article abstract of DOI:10.1016/j.bmc.2012.05.077

The linear depsipeptide grassystatin A, a valuable probe for the study of cathepsin E function, has been synthesized by a [4+6] strategy. It exhibited specific inhibitory activity against cathepsin E with an IC₅₀ value of 0.8 nM. Our studies in

Full text of DOI:10.1016/j.bmc.2012.05.077

Bioorganic & Medicinal Chemistry 20 (2012) 4774–4780
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Total synthesis of grassystatin A, a probe for cathepsin E function
Siming Yang a, Wei Zhang a, Ning Ding a, Jeannette Lo b, Yanxia Liu c, Michael J. Clare-Salzler b,
a,
Hendrik Luesch c, , Yingxia Li
⇑
⇑
a School of Pharmacy, Fudan University, Shanghai 201203, China
b Department of Pathology, Immunology, and Laboratory Medicine, University of Florida, Gainesville, FL 32610, USA
c Department of Medicinal Chemistry, University of Florida, Gainesville, FL 32610, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The linear depsipeptide grassystatin A, a valuable probe for the study of cathepsin E function, has been
synthesized by a [4+6] strategy. It exhibited specific inhibitory activity against cathepsin E with an
IC50 value of 0.8 nM. Our studies indicated that inhibition of cathepsin E did not have an impact on oval-
bumin antigen processing and peptide presentation, unique from studies of other aspartic protease
inhibitors.
Received 27 April 2012
Revised 30 May 2012
Accepted 30 May 2012
Available online 7 June 2012
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
Total synthesis
Grassystatin A
Cathepsin E inhibitor
Antigen processing and presentation
1. Introduction
2. Results and discussion
Cathepsin E (CE) is an intracellular aspartic protease that is
mainly present in immune cells and is highly homologous to the
aspartic protease cathepsin D (CD). Like CD, CE has also been
shown to have a role in antigen processing, but is localized in dif-
ferent cell compartments and tissues and has broader substrate
specificity.1,2 CE deficiency and over-expression have been shown
to result in a variety of pathological conditions; however, its exact
functional role remains to be elucidated.3,4 Inhibitors developed to
study the role of CE have not been forthcoming due to difficulties
with purification or preparation in quantities sufficient for func-
tional studies, and their non-specificity further complicates the
progress in discriminating its unique role.
The retrosynthetic analysis of 1 is outlined in Figure 2.
We envisioned that the molecule can be assembled via amida-
tion between the Asn and Leu residues because of the versatility
in the construction of the precursors (4, 5), low hindrance and ab-
sence of racemization during fragment coupling.
First, we focused on the synthesis of fragment 4 as summarized
in Scheme 1.
After the preparation of
protocol,6 the synthesis was started with coupling of the commer-
cially available Boc- -Val and -Hiva-OBn to afford 6 in 89% yield.
L- and D-Hiva-OBn through established
L
L
Considering the possibility of racemization while using violent
reaction conditions, we thought that 7 would easily be obtained
Grassystatins A–C were isolated from a cyanobacterium, identi-
fied as Lyngbya cf. confervoides by Luesch and co-workers.5 During
the biological evaluation of the three linear modified peptides, the
major compound, grassystatin A (1, Fig 1) was 30-fold more selec-
tive for CE over CD with sub-nanomolar potency. It contains some
notable structural features, including two N-methylamides and a
(3S,4S)-statine unit. It also consists of several hydroxy acids linked
through sterically hindered ester bonds which are very rare in the
marine peptides. Due to the difficulties mentioned above and ow-
ing to the need of higher amounts of grassystatin A for biological
assays, the total synthesis of grassystatin A (1) was undertaken
and synthetic material subjected to further biological evaluation.
by condensation between 6 and D-Hiva-OBn with a coupling re-
agent. However, the outcome did not appear as we expected. Var-
ious coupling reagents were tested, but the reactions failed. No
product was observed and the reactants were not consumed. Since
the large steric hinderance might have resulted in the low reactiv-
ity of the reactants, we made an attempt to find an alternative
route for 7. The preferred method was Yamaguchi esterification,
which furnished 7 in 99% yield and was mild enough not to cause
racemization. The intermediate was deprotected by 50% TFA/
CH2Cl2 and underwent smooth reductive amination upon treat-
ment with Na(BH3)CN to give 8 in 70% yield. Compound 8 was re-
moved of the benzyl protecting group and then coupled with H-
Leu-Obzl to afford fragment 4 in 75% yield.
The next effort was to synthesize fragment 5 as shown in
Scheme 2. H-Pro-OMe and Boc-N-Me-D-Phe were prepared
⇑
Corresponding authors.
E-mail addresses: luesch@cop.ufl.edu (H. Luesch), liyx417@fudan.edu.cn (Y. Li).
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.05.077

S. Yang et al. / Bioorg. Med. Chem. 20 (2012) 4774–4780
4775
N,N-Me2-L-Val
D-Hiva-1
L-Asn
H2N
L-Thr
N-Me-D-Phe
O
OH
O
O
OH
O
O
H
H
H
O
N
N
N
N
N
O
N
H
N
N
H
O
O
O
O
R
O
O
O
L-Hiva-2
L-Leu
(3S,4S)-Sta
L-Ala/2Aba
O-Me-L-Pro
1 R=Me Grassystatin A
2 R=Et Grassystatin B
(2R,3S)-Hmpa
N-Me-L-Gln
NH2
L-Ile
N-Me-D-Phe
O
O
OH
O
O
H
H
N
H
N
N
N
N
HO
N
N
H
O
O
O
O
O
O
L-Leu
(3S,4S)-Sta
3 Grassystatin C
Gly
O-Me-L-Pro
Figure 1. Grassystatins A-C.
H2N
O
OH
O
O
O
OH
O
O
H
H
H
O
N
N
N
N
N
O
N
H
N
H
N
O
O
O
O
O
O
1
H2N
OH
O
O
OH
O
O
O
O
H
H
N
H
N
N
N
O
O
BocHN
N
H
N
N
O
OBn
+
O
O
O
O
O
4
5
Figure 2. Retrosynthetic analysis of 1.
according to the literature.7,8 Their condensation was then carried
out by using EDC/HOAt as coupling reagents to provide dipeptide 9
in 88% yield. Removal of the Boc group of 8 with 50% TFA/CH2Cl2
and incorporation with Boc-Ala (EDC, HOAt, DIEA) afforded the
tripeptide 10 smoothly in 85% yield. With the same procedure,
the condensation of 10 and Boc-Thr produced 11 in 75% yield.
In order to construct fragment 13, (S,S)-statine unit (12) had to
be prepared first. Though several routes have been reported, most
of them require specific reagents or too many steps to result in the
product. Considering the situations above, we chose the method
below.9 The product of reduction and Dess–Martin oxidation of
commercially available Boc-Leu-OH and then Claisen condensation
with lithio ethyl acetate afforded Boc-Sta-OEt in 64% yield. After
saponification of Boc-Sta-OEt, 12 was obtained. The method has a
low stereoselectivity (anti:syn = 60: 40), but the fragment can be
obtained in large amounts efficiently and economically.
the condensation between 13 and Boc-Asn to afford the fragment
5 with a yield of 71% over two steps.
With fragment 4 and 5 in hand, we arrived at the final stage of
the synthesis. First, the benzyl group and Boc of 4 and 5 were re-
moved under catalytic hydrogenation and 50% TFA/CH2Cl2. The
coupling was carried out with HATU/HOAt as coupling agents to
provide grassystatin A (1, 45% for three steps). However, the puri-
fication did not work smoothly. After being subjected to column
chromatography, the product decomposed into several fragments
of larger polarity. Upon MS analysis, we presumed that the ester
linkage between L-Hiva and D-Hiva was broken due to the acidity
of the silica gel. Therefore, we changed the standard workup proce-
dure and directly chromatographed and purified the target by
HPLC afterwards to afford product 1 as a white solid (purity:
98.8%). Spectroscopic data of compound 1 were in full agreement
with those reported for the natural product isolated from the cya-
nobacterium. Grassystatin A was found to be particularly prone to
hydrolysis catalyzed by acid, and required storage under inert
atmosphere at ꢀ20 °C. In fact, NMR spectra of samples stored at
With fragment 11 and 12 in hand, their coupling could then be
carried out. Compound 12 was activated by HATU/HOAt and incor-
porated into the peptide chain 11 to obtain 13. The last step was

4776
S. Yang et al. / Bioorg. Med. Chem. 20 (2012) 4774–4780
O
O
O
HO
1. EDC, DIEA, DMAP
O
OH
OBn
+
BocHN
OH
BocHN
2. H2, Pd/C, 89%
O
O
6
Yamaguchi reaction
EDC, DIEA, DMAP
O
OBn
OBn
BocHN
O
HO
O
O
O
7
O
O
1. CF3COOH, CH2Cl2
2. Na(BH3)CN, CH2O
O
OBn
N
70%
O
O
8
O
O
O
O
H
N
H2N
1. H2, Pd/C
2. EDC, HOAt, DIEA
75%
N
O
OBn
OBn
O
O
4
Scheme 1. Synthesis of fragment 4.
HN
EDC, HOBt, DIEA
+
N
O
Boc
OH
88%
BocN
N
O
O
O
O
O
O
H-Pro-OMe
9
OH
OH
O
BocHN
BocHN
N
1.CF3COOH
OH
BocHN
N
2.EDC, HOBt, DIEA
O
O
85%
O
O
10
OH
1. CF3COOH
2. EDC, HOAt,
O
DIEA, 75%
BocHN
OH
OH
O
12
OH
O
OH
O
O
O
H
N
H
N
BocHN
N
O
N
N
N
H
N
BocHN
O
O
O
83%
O
O
11
13
H2N
H2N
OH
O
O
OH
O
O
H
H
N
N
N
OH
BocHN
N
N
BocHN
2.EDC, HOAt, DIEA
71%
H
O
O
O
O
O
O
5
Scheme 2. Synthesis of fragment 5.

S. Yang et al. / Bioorg. Med. Chem. 20 (2012) 4774–4780
4777
antigen is cleaved into 9–22 mer peptide fragments is a critical
step of processing before peptide presentation occurs.16 However,
the requirement for Ag processing may be bypassed to still allow
T cell stimulation if APC are exposed to synthetic peptide frag-
ments.17 In this current study, we aimed to identify whether the
observed grassystatin-associated Ag presentation was stunted
due to inhibited Ag processing mechanisms using the OVA TCR
transgenic MHCII-restricted T cell system. We pretreated DC with
drug before incubation with whole OVA for co-culture with CD4+
T cells and found that both control and drug-treated DC were able
to stimulate comparable T cell proliferation suggesting that anti-
gen processing function remained intact. It is possible we did not
see an effect because grassystatin A only induces a transient effect
on cells, and the withdrawal of cells from drug exposure allowed
the antigen processing and presentation machinery to recover
function, an important observation for clinical considerations in
determining drug half-life and dosage. Interestingly, a study using
cathepsin E-deficient mice found that the antigen presentation
ability of DC was enhanced, possibly due to increased phagocytic
activity and higher expression of costimulatory molecules.15 We
also observed a minor enhancement in T cell response using
drug-treated DC incubated with whole Ag or peptide though it is
unknown whether this was due to enhanced processing or other
heightened functions within the endocytic pathway. It has been
suggested that inhibition of cathepsin E may not necessarily shut
down antigen processing as compensatory functions by similar en-
zymes can occur depending on which compartment is affected.
Furthermore, cathepsin E’s role in Ag processing may be an indirect
modulation of the cellular endosomal microenvironment rather
than a direct influence on processing.15 It remains to be deter-
mined where grassystatin exerts its effects along the endocytic
pathway as our observations suggest its influence may require its
presence within the cathepsin-E compartments during peptide–
MHC complex assembly.
Figure 3. CD4+ T cell proliferation in response to antigen-loaded DC. Purified
CD11c+ dendritic cells from spleen were treated with grassystatin A, then pulsed
with whole OVA or OVAp 323–339 and cocultured with enriched CD4+ T cells.
DMSO-treated DC were used as solvent controls.
room temperatures for a few days showed a gradual increase of the
decomposed products.
We previously found that grassystatin A exhibited specific
inhibitory activity against cathepsin E, an aspartic protease which
has been shown to have an integral role in immune function.
Consistent with our previous report,5 synthetic grassystatin A also
potently inhibited cathepsin E activity in vitro (IC50 0.8 nM) and to
a lesser extent cathepsin D activity (IC50 40 nM). Our previous
studies examining T cell response to exogenous antigen suggested
that grassystatin A may alter antigen presentation,5 and Yee et al.
demonstrated that negative regulation of cathepsin E expression
by CIITA (MHC class II transactivator) decreased antigen process-
ing.10 However, a clear mechanism in how cathepsin E inhibition
may alter APC function remains unknown. We sought to determine
whether grassystatin A may inhibit MHC class II-dependent anti-
gen processing using the DO11.10 TCR transgenic mouse model,
which possess MHC class II-restricted CD4+ T cells with a TCR that
recognize the OVA peptide 323–339 (OVAp). OVAp-pulsed den-
dritic cells (DC) presenting this synthetic peptide to CD4+ T cells
will result in T cell proliferation; alternatively DC that effectively
process whole OVA antigen to present this OVA fragment can also
elicit a T cell response. To assess the effect of grassystatin A on
antigen processing, we measured the ability of drug-treated, whole
antigen-pulsed DC to elicit a T cell response. We found that the T
cell proliferative response to OVA-pulsed grassystatin-treated DC
was comparable to that of OVA-pulsed DMSO control-treated DC,
suggesting that grassystatin A does not have an impact on ovalbu-
min antigen processing (Fig. 3). Furthermore, grassystatin-treated
DC that were loaded with OVAp were able to mount an equally ro-
bust response from T cells as controls, demonstrating that peptide
presentation was unaffected (Fig. 3).
This finding is unique from studies of other aspartic protease
inhibitors which implicated a role of cathepsin E in MHC class II
antigen processing.10–14 However, the MHCII antigen processing
and presentation pathway involves a complex system of proteins
and endocytic compartments, and the specific target within this
pathway may vary between cathepsin E inhibitors. In dendritic
cells, cathepsin E is present mainly in early endosomes and par-
tially in lysosomes,11,12,15 thus the effect of CE inhibitors may de-
pend on its presence during the period of peptide–MHCII
complex assembly within those compartments where both Ag pro-
cessing and presentation activities occur. Our previous study dem-
onstrated that T cell proliferation to peptide-APC stimulation was
reduced in the presence of grassystatin A, suggesting impaired
antigen presentation, but did not specifically address whether
the upstream antigen processing machinery was also affected.5
The routing of antigen through endocytic vesicles where whole
3. Conclusions
In summary, we have described the first total synthesis of gras-
systatin A, a promising linear depsipeptide which is a valuable
probe for the study of cathepsin E function. We validated the po-
tent in vitro cathepsin E inhibitory activity and have continued
preliminary biological studies to study potential cellular effects
on antigen processing and presentation. Further studies of the
structure–activity relationship will be reported in due course. Spe-
cifically, it will be of interest to modulate the selectivity profile
against various aspartic proteases, for example, to generate even
more selective cathepsin E inhibitors or specifically enhance the
inhibitory activity against cathepsin D based on the grassystatin
scaffold. Our initial molecular docking studies with grassystatins
suggested that the polar amino acid (Asn) in the P2 pocket adjacent
to the statine unit (N-end) may contribute to the cathepsin E pref-
erence and that the N,N-dimethyl group is a likely contributor to
the potency of these compounds against both cathepsins D and E.5
4. Experimental section
4.1. General information
Solvents were purified by standard methods. TLCs were carried
out on Merck 60 F254 silica gel plates and visualized by UV irradi-
ation or by staining with iodine absorbed on silica gel, ninhydrin
solution or with aqueous acidic ammonium molybdate solution
as appropriate. Flash column chromatography was performed on
silica gel (200–300 mesh, Qingdao, China). Optical rotations were
obtained using a JASCO P-1020 digital polarimeter. NMR spectra

4778
S. Yang et al. / Bioorg. Med. Chem. 20 (2012) 4774–4780
were recorded on JEOLJNM-ECP 400 MHz spectrometers. Chemical
shifts are reported in parts per million (ppm), relative to the
signals due to the solvent. Data are described as followed: chemical
shift, multiplicity (s = singlet, d = doublet, t = triplet, q = quartet,
br = broad, m = multiplet), coupling constants (Hz), integration,
and assignment. Mass spectra were obtained on a Q-Tof Ultima
Global mass spectrometer.
for 18 h at rt. The reaction was subsequently concentrated in vacuo
and the residue redissolved in EtOAc. The organic fraction was then
washed with saturated aqueous Na2CO3 solution and brine before
drying over Na2SO4 and concentrating in vacuo. The residue was
purified by column chromatography with petroleum–EtOAc (3:1)
to obtain 8 as colorless oil (304.7 mg, 70%).
Rf 0.4 (5:1, petroleum–EtOAc); ½a D23
ꢂ
= +9.8° (c 0.06, CHCl3); 1H
NMR (CDCl3, 400 MHz): d: 7.31–7.36 (m, 5H), 5.18 (q, 2H,
J = 12.4, 14.4 Hz), 5.06 (d, 1H, J = 4.0 Hz), 4.96 (d, 1H, J = 4.8 Hz),
2.85 (d, 1H, J = 10.4 Hz), 2.35 (s, 6H), 2.23–2.34 (m, 2H), 2.01–
2.06 (m, 1H), 0.92–1.07 (m, 18H). 13C NMR (CDCl3, 100 MHz): d:
171.12, 169.31, 168.80, 135.37, 128.52, 128.41, 128.33, 77.70,
76.23, 74.27, 66.93, 41.27, 30.20, 30.16, 27.62, 19.80, 19.40,
19.14, 18.58, 17.38, 17.14; ESIHRMS calcd for C24H38N1O6 [M+H]+
436.2694; found 436.2708.
4.2. Chemistry
4.2.1. N-Boc-Val-
L
-Hiva-OH (6)
A
solution of Boc-
L-Val (521.0 mg, 2.4 mmol) and L-Hiva
(416.2 mg, 2.0 mmol) in anhydrous CH2Cl2 (10 ml) was treated
sequentially with EDC (591.9 mg, 3.0 mmol), DIEA (1.01 mL,
6.0 mmol) and DMAP (24.4 mg, 0.2 mmol) at 0 °C. The mixture
was warmed to room temperature and stirred overnight. After
dilution with EtOAc (100 ml), the whole mixture was washed with
10% citric acid (2 ꢁ 30 ml), 5% NaHCO3 (2 ꢁ 30 ml) and brine
(2 ꢁ 30 ml), dried over Na2SO4, and concentrated in vacuo. The res-
idue was purified by flash column chromatography with petroleum
ether–EtOAc (9:1) to afford 6* as a colorless oil (730.5 mg, 90%).
Hydrogenation of 6* was carried out in CH3OH–CH2Cl2 (1:1,
15 ml) in the presence of a catalytic amount of Pd-C (10%) under
hydrogen at room temperature. Pd-C was removed by filtration
and concentrated under reduced pressure to yield the acid 6, which
was used directly in the next step.
4.2.4. N,N-Me-Val-L-Hiva-D-Hiva-Leu-OBn (4)
Hydrogenation of 8 (435.3 mg, 1.0 mmol) was carried out in
CH3OH–CH2Cl2 (1:1,15 ml) in the presence of a catalytic amount
of Pd-C (10%) under hydrogen at room temperature. Pd-C was re-
moved by filtration and concentrated under reduced pressure to
yield the acid 8, which was used directly in the next step.
To a solution of acid 8 and H-Leu-OBnꢃTosOH (471.8 mg,
1.2 mmol) in CH2Cl2 (10 mL) was added DIEA (1.00 mL, 6.0 mmol),
HOAt (326.4 mg, 2.4 mmol) and EDC (473.5 mg, 2.4 mmol) at 0 °C.
The mixture was warmed to room temperature and stirred over-
night. After dilution with EtOAc (100 ml), the whole mixture was
washed with 10% citric acid (2 ꢁ 30 ml), 5% NaHCO3 (2 ꢁ 30 ml)
and brine (2 ꢁ 30 ml), dried over Na2SO4, and concentrated in va-
cuo. The residue was purified by flash column chromatography
with petroleum ether–EtOAc (4:1) to afford 4 as a colorless oil
(411.2 mg, 75%).
Rf 0.4 (10:1, CHCl3–MeOH); ½a D23
ꢂ
= +11.4° (c 0.04, CHCl3); 1H
NMR (CDCl3, 400 MHz): d: 5.06 (s, 1H), 4.93 (s, 1H), 4.32 (dd, 1H,
J = 3.5, 8.6 Hz), 2.29 (d, 2H, J = 6.6 Hz), 1.44 (s, 9H), 0.93–1.05 (m,
9H), 0.89 (d, 3H, J = 6.5 Hz). 13C NMR (CDCl3, 100 MHz): d:
174.00, 172.50, 155.91, 80.06, 76.89, 58.24, 31.06, 30.00, 28.29,
18.97, 18.80, 17.13, 16.98; ESIHRMS calcd for
C15H27N1O6
[M+Na]+ 340.1731; found 340.1729.
Rf 0.3 (5:1, petroleum–EtOAc); ½a D23
ꢂ
= +19.6° (c 0.07, CHCl3); 1H
NMR (CDCl3, 400 MHz): d: 7.33–7.36 (m, 5H), 6.90 (d, 1H,
J = 8.4 Hz), 5.12–5.21 (m, 3H), 4.69–4.72 (m, 1H), 4.66 (d, 1H,
J = 6.0 Hz), 2.85 (d, 1H, J = 10.4 Hz), 2.42–2.51 (m, 1H), 2.31 (s,
6H), 2.16–2.26 (m, 1H), 1.91–2.01 (m, 1H), 1.57–1.72 (m, 3H),
1.06 (q, 6H, J = 4.4, 6.8 Hz), 0.98 (t, 6H, J = 6.8 Hz), 0.93 (d, 3H,
J = 6.8 Hz), 0.85–0.90 (m, 9H). 13C NMR (CDCl3, 100 MHz): d:
172.51, 172.09, 169.37, 168.90, 135.57, 128.51, 128.27, 128.18,
78.36, 77.68, 73.80, 66.87, 50.64, 41.07, 40.80, 30.28, 29.84,
29.70, 27.62, 24.59, 22.76, 21.06, 19.57, 19.24, 18.85, 17.86,
16.19; ESIHRMS calcd for C30H49N2O7 [M+H]+ 549.3534; found
549.3557.
4.2.2. N-Boc-Val-
L-Hiva-D-Hiva-OBn (7)
To a solution of acid 6 (380.6 mg, 1.2 mmol) and
D-Hiva
(208.1 mg, 1.0 mmol) in THF (10 mL) were added DIEA (1.01 mL,
6.0 mmol) and 2,4,6-trichlorobenzoyl chloride (1.20 mL, 1.8 mmol)
at room temperature. The reaction mixture was allowed to stir at
room temperature for 4 h before adding the tolulene (60 mL) to di-
lute the solution, then DMAP (5.10 g, 36 mmol) in toluene (30 mL)
was added. After stirring at room temperature for 5 h, the solvent
was removed under reduced pressure and the residue was chro-
matographed on silica gel with petroleum–EtOAc (9:1) to afford
7 (500.7 mg, 99%) as a yellow oil.
Rf 0.4 (5:1, petroleum–EtOAc); ½a D23
ꢂ
= +7.1° (c 0.12, CHCl3); 1H
4.2.5. N
A solution of
Boc-N -Me- -phenylalanine (2.80 g, 10 mmol) in anhydrous
a
-Boc-N
a
-Me-
D-Phe-L-Pro-OMe (9)
NMR (CDCl3, 400 MHz): d: 7.26–7.36 (m, 5H), 5.18 (dd, 2H,
J = 5.2, 17.6 Hz), 5.05 (d, 1H, J = 4.4 Hz), 4.94 (d, 1H, J = 4.8 Hz),
4.84 (d, 1H, J = 4.0 Hz), 4.34 (dd, 1H, J = 4.4, 9.2 Hz), 2.23–2.30 (m,
3H,), 1.44 (s, 9H), 0.92–1.04 (m, 18H). 13C NMR (CDCl3,
100 MHz): d: 171.77, 168.95, 168.75, 155.66, 135.32, 128.54,
128.42, 128.39, 79.74, 77.52, 77.44, 77.23, 66.97, 31.34, 30.29,
30.11, 28.33, 19.05, 18.81, 18.66, 17.41, 17.34, 16.96; ESIHRMS
calcd for C27H41N1O8 [M+Na]+ 530.2724; found 530.2743.
L
-proline methyl ester (1.30 g, 10 mmol) and Na-
a
D
CH2Cl2 (50 ml) was treated sequentially with HOBt (1.63 g,
12 mmol), EDC (2.37 g, 12 mmol), and DIEA (4.45 mL, 26.4 mmol)
at 0 °C. The mixture was warmed to room temperature and stirred
overnight. After dilution with EtOAc (300 mL), the whole mixture
was washed with 10% citric acid (100 mL), 5% NaHCO3 (100 mL),
and brine (100 mL), dried over Na2SO4, and concentrated in vacuo.
The residue was purified by flash column chromatography with
petroleum ether–EtOAc (5:1) to afford 9 as a white solid (3.43 g,
88%).
4.2.3. N,N-Me-Val-L-Hiva-D-Hiva-OBn (8)
To a solution of 7 (507.3 mg, 1.0 mmol) in CH2Cl2 (10 mL) at
0 °C, TFA (10 mL) was added. The mixture was stirred for 30 min
and then evaporated, the residual oil was dissolved twice in CH2Cl2
(15 mL) with evaporation each time to give TFA salt 7*, which was
used directly in the next step.
1H NMR (CDCl3, 400 MHz): d: 7.12–7.26 (m, 10H), 5.16 (t, 1H,
J = 7.4 Hz), 4.86 (dd, 1H, J = 5.1, 9.4 Hz), 4.42–4.48 (m, 2H), 3.73
(s, 6H), 3.39–3.54 (m, 4H), 3.20 (dd, 1H, J = 7.4, 14.1 Hz), 3.11
(dd, 1H, J = 5.1, 13.7 Hz), 2.86 (s, 4H), 2.81 (s, 4H), 2.16–2.22 (m,
2H), 1.83–2.00 (m, 6H), 1.61 (s, 9H), 1.20 (s, 9H)
To a solution of acid 7* in MeCN was added DIEA (201.5 lL,
1.2 mmol) at 0 °C, then 37% aqueous formaldehyde solution fol-
lowed by AcOH (0.1 mL). The reaction was stirred for 1 h before
Na(BH3)CN (188.5 mg, 3.0 mmol) was added. AcOH was added
periodically to maintain a pH of 5–7, and the reaction was stirred
4.2.6. N-Boc-Ala-N-Me-D-Phe-L-Pro-OMe (10)
To a solution of 9 (780.4 mg, 2.0 mmol) in CH2Cl2 (10 mL) at
0 °C, TFA (10 mL) was added. The mixture was stirred for 30 min

S. Yang et al. / Bioorg. Med. Chem. 20 (2012) 4774–4780
4779
and then evaporated, the residual oil was dissolved twice in CH2Cl2
(15 mL) with evaporation each time to give TFA salt 9*, which was
used directly in the next step.
A solution of 9* and Boc-Ala-OH (378.2 mg, 2.0 mmol) in anhy-
drous CH2Cl2 (15 mL) was treated sequentially with HOAt
(326.4 mg, 2.4 mmol), EDC (473.5 mg, 2.4 mmol), and DIEA
3.61 (m, 1H), 3.41–3.45 (m, 1H), 3.34–3.39 (m, 1H), 3.22 (dd, 1H,
J = 6.4, 14.0 Hz), 3.02 (s, 3H), 2.96 (q, 1H, J = 9.2, 14.0 Hz), 2.36–
2.51 (m, 2H), 2.21–2.24 (m, 1H), 1.77–2.00 (m, 3H), 1.63–1.70
(m, 1H), 1.50–1.57 (m, 1H), 1.44 (s, 9H), 1.31–1.36 (m, 1H), 1.10
(d, 3H, J = 6.4 Hz), 0.92 (d, 6H, J = 6.4 Hz), 0.86 (d, 3H, J = 5.2 Hz).
13C NMR (CDCl3, 100 MHz): d: 173.10, 172.56, 172.46, 169.81,
168.08, 156.31, 136.68, 129.58, 128.29, 126.70, 79.37, 70.51,
66.98, 59.27, 56.97, 55.73, 52.40, 52.22, 47.02, 45.91, 41.62,
40.16, 34.82, 30.38, 28.89, 28.42, 25.26, 24.83, 23.14, 22.20,
18.14, 16.88; ESIHRMS calcd for C36H57N5O10 [M+Na]+ 742.3998;
found 742.3973.
(892.7 lL, 5.3 mmol) at 0 °C. The solution was stirred for 15 min,
warmed to room temperature and stirred for another 12 h. The
mixture was then treated as described for 8. The residue was puri-
fied by flash column chromatography with petroleum ether–EtOAc
(3:1) to afford 10 as a white solid (784.0 mg, 85%).
Rf 0.3 (2:1, petroleum–EtOAc); ½a D23
ꢂ
= +56.4° (c 0.03, CHCl3); 1H
NMR (CDCl3, 400 MHz): d: 7.13–7.26 (m, 5H), 5.71 (dd, 1H, J = 6.4,
9.2 Hz), 5.17 (d, 1H, J = 8.4 Hz), 4.45 (t, 2H, J = 8.0 Hz), 3.74 (s, 3H),
3.47–3.49 (m, 1H), 3.35–3.40 (m, 1H), 3.22 (dd, 1H, J = 6.4, 14.4 Hz),
3.00 (s, 3H), 2.95 (dd, 1H, J = 9.6, 14.4 Hz), 2.20–2.25 (m, 1H), 1.82–
1.97 (m, 3H), 1.39 (s, 9H), 0.82 (d, 3H, J = 7.2 Hz). 13C NMR (CDCl3,
100 MHz): d: 173.05, 172.44, 168.35, 155.38, 136.90, 129.62,
128.20, 126.58, 79.50, 59.36, 55.63, 52.16, 30.22, 28.87, 28.32,
25.37, 17.54; ESIHRMS calcd for C24H35N3O6 [M+Na]+ 484.2418;
found 484.2410.
4.2.9. N-Boc-Asn-Sta-Thr-Ala-Na-Me-D-Phe-L-Pro-OMe (5)
To a solution of 13 (719.4 mg, 1.0 mmol) in CH2Cl2 (10 mL) at
room temperature was added TFA (10 mL). The reaction mixture
was stirred for 30 min and then reconcentrated from CH2Cl2 twice
to remove excess TFA. A solution of the TFA salt 13* and Boc-
OH (132.1 mg, 1.0 mmol) in CH2Cl2 (10 ml) was treated sequen-
tially with DIEA (646.8 l, 3.8 mmol), HATU (456.0 mg, 1.2 mmol),
L-Asn-
l
and HOAt (163.2 mg, 1.2 mmol). The reaction mixture was stirred
at 0 °C for 30 min and then stirred at room temperature for another
8 h. The reaction mixture was treated as described for 9. The resi-
due was purified by flash column chromatography with petroleum
ether–EtOAc (1:1) to afford 5 as a white foam solid (591.7 mg,
71%).
4.2.7. N-Boc-Thr-Ala-Na-Me-D-Phe-L-Pro-OMe (11)
To a solution of 10 (627.2 mg, 1.36 mmol) in CH2Cl2 (20 ml) at
room temperature was added TFA (20 ml). The reaction mixture
was stirred for 30 min and then reconcentrated from CH2Cl2 two
times to remove excess TFA. A solution of the TFA salt 10* and
Rf 0.2 (15:1, CHCl3–MeOH); ½a D23
ꢂ
= ꢀ25.0° (c 0.02, CHCl3); 1H
NMR (MeOD, 400 MHz): d: 7.12–7.27 (m, 5H), 5.67 (dd, 1H,
J = 3.6, 9.6 Hz), 4.68 (dd, 1H, J = 7.2, 14.0 Hz), 4.38–4.46 (m, 2H),
4.34 (d, 1H, J = 4.0 Hz), 4.08–4.18 (m, 1H), 3.88–4.02 (m, 2H),
3.72 (s, 3H), 3.37–3.52 (m, 2H), 3.13 (dd, 1H, J = 8.0, 14.4 Hz),
3.06 (s, 3H), 2.92 (dd, 1H, J = 4.4, 14.0 Hz), 2.73 (dd, 1H, J = 8.0,
15.2 Hz), 2.58 (dd, 1H, J = 8.8, 15.2 Hz), 2.40 (d, 2H, J = 6.0 Hz),
2.19–2.31 (m, 1H), 1.94–2.02 (m, 2H), 1.81–1.91 (m, 2H), 1.54–
1.71 (m, 2H), 1.44 (s, 9H), 1.16 (d, 3H, J = 1.6 Hz), 0.85–0.94 (m,
9H). 13C NMR (MeOD, 100 MHz): d: 173.72, 173.02, 172.85,
172.60, 172.34, 170.75, 168.90, 156.21, 136.86, 129.34, 127.88,
126.22, 79.54, 70.03, 67.12, 59.50, 58.38, 58.38, 55.89, 53.40,
51.61, 51.30, 51.14, 45.58, 40.17, 39.99, 36.62, 34.21, 29.36,
28.60, 27.37, 24.84, 24.45, 22.48, 21.06, 18.57, 15.40; ESIHRMS
calcd for C40H63N7O12 [M+Na]+ 856.4427; found 856.4402.
Boc-
sequentially with HOAt (222.2 mg, 1.63 mmol), EDC (313.1 mg,
1.63 mmol), and DIEA (570.9 l, 3.4 mmol). The reaction mixture
L-Thr-OH (298.0 mg, 1.36 mmol) in CH2Cl2 (10 ml) was treated
l
was stirred at 0 °C for 30 min and then stirred at room temperature
for another 8 h. The reaction mixture was treated as described for
9. The residue was purified by flash column chromatography with
petroleum ether–EtOAc (20:1–3:1) to afford 11 as a white foam so-
lid (573.6 mg, 75 %).
Rf 0.2 (2:1, petroleum–EtOAc); ½a D23
ꢂ
= +34.5° (c 0.08, CHCl3); 1H
NMR (CDCl3, 400 MHz): d: 7.17–7.24 (m, 5H), 6.83–6.97 (br, 1H),
5.69 (q, 1H, J = 6.4, 9.2 Hz), 5.37 (d, 1H, J = 8.4 Hz), 4.71 (t, 1H,
J = 8.0 Hz), 4.17–4.34 (m, 1H), 4.02–4.16 (m, 1H), 3.74 (s, 3H),
3.48–3.54 (m, 1H), 3.27–3.41 (m, 1H), 3.21 (dd, 1H, J = 6.8,
14.4 Hz), 3.02 (s, 3H), 2.97 (q, 1H, J = 9.6, 14.4 Hz), 2.21–2.24 (m,
1H), 1.80–1.97 (m, 3H), 1.44 (s, 9H), 1.11 (d, 3H, J = 6.4 Hz), 0.85
(d, 3H, J = 6.8 Hz). 13C NMR (CDCl3, 100 MHz): d: 172.51, 172.42,
170.56, 168.09, 155.98, 136.66, 129.58, 128.26, 126.66, 80.16,
67.27, 59.29, 57.93, 55.66, 52.24, 47.03, 45.76, 34.75, 30.32,
28.87, 28.28, 25.29, 17.96, 16.96; ESIHRMS calcd for C28H42N4O8
[M+Na]+ 585.2895; found 585.2886.
4.2.10. Grassystatin A (1)
To a solution of 4 (83.3 mg, 0.1 mmol) in CH2Cl2 (10 mL) at
room temperature was added TFA (10 mL). The reaction mixture
was stirred for 30 min and then reconcentrated from CH2Cl2 twice
to remove excess TFA. Hydrogenation of 5 (54.8 mg, 0.1 mmol) was
carried out in CH3OH–CH2Cl2 (1:1,15 ml) in the presence of a cat-
alytic amount of Pd-C (10%) under hydrogen at room temperature.
Pd-C was removed by filtration and concentrated under reduced
pressure to yield the acid 5, which was used directly in the next
step. A solution of the TFA salt 4* and acid 5 in CH2Cl2 (10 ml)
4.2.8. N-Boc-Sta-Thr-Ala-Na-Me-D-Phe-L-Pro-OMe (13)
To a solution of 10 (562.3 mg, 1.0 mmol) in CH2Cl2 (10 mL) at
room temperature was added TFA (10 mL). The reaction mixture
was stirred for 30 min and then reconcentrated from CH2Cl2 twice
to remove excess TFA. A solution of the TFA salt 11* and 12
(275.2 mg, 1.0 mmol) in CH2Cl2 (10 ml) was treated sequentially
was treated sequentially with DIEA (64.7 ll, 0.38 mmol), HATU
(456.0 mg, 0.12 mmol), and HOAt (16.3 mg, 0.12 mmol). The reac-
tion mixture was stirred at 0 °C for 30 min and then stirred at room
temperature for another 8 h. The reaction mixture was treated as
described for 8. The residue was purified by HPLC to afford 1 as a
white foam solid (53.8 mg, 45%).
with DIEA (646.8 ll, 3.8 mmol), HATU (456.0 mg, 1.2 mmol), and
HOAt (163.2 mg, 1.2 mmol). The reaction mixture was stirred at
0 °C for 30 min and then stirred at room temperature for another
8 h. The reaction mixture was treated as described for 8. The resi-
due was purified by flash column chromatography with petroleum
ether–EtOAc (2:1) to afford 13 as a white foam solid (597.1 mg,
83%).
Rf 0.3 (10:1, CHCl3–MeOH); ½a D23
ꢂ
= ꢀ5.3° (c 0.1, CHCl3); 1H NMR
(CDCl3, 400 MHz): d: 7.54 (d, 1H, J = 7.8 Hz), 7.09–7.24 (m, 5H),
7.03 (d, 1H, J = 5.8 Hz), 6.36 (br, 1H), 6.04 (br, 1H), 5.65 (dd, 1H,
J = 6.6, 9.1 Hz), 5.12 (d, 1H, J = 3.3 Hz), 4.70–4.79 (m, 2H), 4.66 (d,
1H, J = 4.2 Hz), 4.44 (dd, 1H, J = 5.8, 8.0 Hz), 4.24–4.37 (m, 4H),
4.01–4.08 (m, 1H), 3.86–3.97 (m, 1H), 3.73 (s, 3H), 3.42–3.50 (m,
1H), 3.31–3.35 (m, 1H), 3.21 (dd, 1H, J = 6.6, 14.4 Hz), 3.02 (s,
3H), 2.95 (dd, 1H, J = 9.1, 14.2 Hz), 2.84–2.89 (m, 1H), 2.83 (d, 1H,
J = 10.6 Hz), 2.65 (dd, 1H, J = 6.0, 15.2 Hz), 2.57 (dd, 1H, J = 8.9,
Rf 0.15 (2:1, petroleum–EtOAc); ½a D23
ꢂ
= +26.5° (c 0.02, CHCl3); 1H
NMR (CDCl3, 400 MHz): d: 7.12–7.24 (m, 5H), 6.94 (d, 1H,
J = 7.6 Hz), 6.77 (d, 1H, J = 7.6 Hz), 5.67 (q, 1H, J = 6.8, 9.2 Hz),
4.78 (d, 1H, J = 9.6 Hz), 4.69 (t, 1H, J = 7.2 Hz), 4.40–4.47 (m, 2H),
4.20–4.24 (m, 1H), 3.97 (d, 1H, J = 10.4 Hz), 3.37 (s, 3H), 3.56–

4780
S. Yang et al. / Bioorg. Med. Chem. 20 (2012) 4774–4780
14.0 Hz), 2.37–2.44 (m, 2H), 2.31 (s, 6H), 2.18–2.24 (m, 2H), 1.95–
2.01 (m, 2H), 1.81–1.91 (m, 2H), 1.56–1.72 (m, 5H), 1.05–1.11 (m,
6H), 0.85–1.01 (m, 27H). 13C NMR (CDCl3, 100 MHz): d: 173.17,
172.68, 172.45, 172.38, 171.80, 170.6, 170.33, 170.09, 169.53,
168.13, 136.68, 129.51, 128.24, 126.64, 78.08, 77.62, 73.78, 70.38,
67.44, 59.24, 58.05, 55.69, 52.78, 52.24, 51.91, 50.38, 47.00,
45.62, 41.12, 40.65, 40.05, 39.84, 37.33, 34.71, 30.18, 29.78,
28.87, 27.57, 25.26, 24.74, 24.47, 23.10, 22.61, 22.23, 22.01,
19.53, 19.17, 18.92, 18.82, 17.76, 16.96, 16.42; ESIHRMS calcd for
automated cell harvester and radioactivity was analyzed with a li-
quid scintillation counter. Uptake of 3H-thymidine is reported as
counts per minute (cpm).
Acknowledgements
This work was supported by the National Natural Science Foun-
dation of China (81172975) and the University of Florida College of
Pharmacy.
C
58H95N9O16 [M+Na]+ 1196.6794; found 1196.6800.
Supplementary data
4.3. Biology
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2012.05.077.
4.3.1. Protease inhibition assays
In vitro enzymatic assays were carried out as described.5
References and notes
4.3.2. Animals
1. Hewitt, E. W.; Treumann, A.; Morrice, N.; Tatnell, P. J.; Kay, J.; Watts, C. J.
Immunol. 1997, 159, 4693.
2. Cathepsin, J. J.; Kay, E. In Handbook of Proteolytic Enzyme; Barrett, A. J., Rawlings,
N. D., Woessner, J. F., Eds.; Elsevier Academic Press: London, 2004; pp 33–38.
3. Zaidi, N.; Kallbacher, H. Biochem. Biophys. Res. Commun. 2008, 367, 517.
Twelve-week old male and female DO11.10 mice were kindly
provided as a gift by Lyle Moldawer, Department of Surgery, Uni-
versity of Florida College of Medicine.
´
4. Chlabicz, M.; Gacko, M.; Worowska, A.; Lapinski, R. Folia Histochem. Cytobiol.
4.3.3. Isolation of DC and T Cells from spleen
2011, 49, 547.
Spleens were processed using 100-micron nylon filters and de-
pleted of red blood cells using ACK ammonium chloride lysis buf-
fer. DC were first purified using CD11c+ magnetic beads separation
(Miltenyi Biotec), then untouched CD4+ T cells were enriched from
the flow-through negative fraction using a CD4+ T cell isolation kit
that depletes positively-labeled CD8a, CD11b, CD11c, CD19, CD45R
(B220), CD49b (DX5), CD105, anti-MHC-class II, and Ter-119 cells
(Miltenyi Biotec).
5. Kwan, J. C.; Eksioglu, E. A.; Liu, C.; Paul, V. J.; Luesch, H. J. Med. Chem. 2009, 52,
5732.
6. Peng, Y.; Pang, H. W.; Ye, T. Org. Lett. 2004, 6, 3781.
7. Bernard, F. E.; Howard, S.; Erwin, B. J. Am. Chem. Soc. 1954, 76, 1806.
8. McDermott, J. R.; Benoiton, N. L. Can. J. Chem. 1973, 51, 1915.
9. Kenneth, E. R.; Carl, F. H.; Gerald, S. P.; Ben, E. E. J. Org. Chem. 1982, 47, 3016.
10. Yee, C. S.; Yao, Y.; Li, P.; Klemsz, M. J.; Blum, J. S.; Chang, C. H. J. Immunol. 2004,
172, 5528.
11. Zaidi, N.; Burster, T.; Sommandas, V.; Herrmann, T.; Boehm, B. O.; Driessen, C.;
Voelter, W.; Kalbacher, H. Biochem. Biophys. Res. Commun. 2007, 364, 243.
12. Chain, B. M.; Free, P.; Medd, P.; Swetman, C.; Tabor, A. B.; Terrazzini, N. J.
Immunol. 2005, 174, 1791.
4.3.4. DC
13. Bennett, K.; Levine, T.; Ellis, J. S.; Peanasky, R. J.; Samloff, I. M.; Kay, J.; Chain, B.
M. Eur. J. Immunol. 1992, 22, 1519.
4.3.4.1. T cell co-culture.
Drug-treated DC were incubated in
14. Nishioku, T.; Hashimoto, K.; Yamashita, K.; Liou, S. Y.; Kagamiishi, Y.; Maegawa,
H.; Katsube, N.; Peters, C.; von Figura, K.; Saftig, P.; Katunuma, N.; Yamamoto,
K.; Nakanishi, H. J. Biol. Chem. 2002, 2002(277), 4816.
15. Kakehashi, H.; Nishioku, T.; Tsukuba, T.; Kadowaki, T.; Nakamura, S.;
Yamamoto, K. J. Immunol. 2007, 179, 5728.
16. Wilson, N. S.; Villadangos, J. A. Adv. Immunol. 2005, 86, 241.
17. Shimonkevitz, R.; Colon, S.; Kappler, J. W.; Marrack, P.; Grey, H. M. J. Immunol.
1984, 133, 2067.
complete RPMI with grassystatin A at 37 °C for 1 h, and solvent
control DC were incubated in cRPMI with 0.1% DMSO. DC were
then washed 3ꢁ with PBS, pulsed with either whole ovalbumin
(500lg/ml, Sigma) or ovalbumin peptide 323–339 (200 lg/ml,
GenScript) for 1–2 h in cRPMI at 37 °C and washed 3ꢁ. DC were
cocultured with 10^5 CD4+ T cells in 96-well round bottom plates
for 72 h followed by 17 h 3H-thymidine (1
lCi/well, Perkin Elmer)
incorporation. Cells were harvested and washed using an