Exploring functional cyclophellitol analogues as human retaining beta-glucosidase inhibitors

Article abstract of DOI:10.1039/c4ob01611d

The natural product, cyclophellitol and its aziridine analogue are potent mechanism-based retaining β-glucosidase inhibitors. In this paper we explore the inhibitory potency of a number of cyclophellitol analogues against the three human retaining β-glucosidases, GBA, GBA2 and GBA3. We demonstrate that N-alkyl cyclophellitol aziridine is at least equally potent in inhibiting the enzymes evaluated as its N-acyl congener, whereas the N-sulfonyl analogue is a considerably weaker inhibitor. Our results complement the literature on the inhibitory potency of cyclophellitol analogues and hold promise for the future design of more effective activity-based retaining glycosidase probes with respect to probe stability in physiological media.

Full text of DOI:10.1039/c4ob01611d

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Exploring functional cyclophellitol analogues as
human retaining beta-glucosidase inhibitors
Cite this: Org. Biomol. Chem., 2014,
2, 7786
1
a
a
b
c
Kah-Yee Li,† Jianbing Jiang,† Martin D. Witte, Wouter W. Kallemeijn,
c
c
a,c
Wilma E. Donker-Koopman, Rolf G. Boot, Johannes M. F. G. Aerts,
Jeroen D. C. Codée, Gijsbert A. van der Marel and Herman S. Overkleeft*
a
a
a
The natural product, cyclophellitol and its aziridine analogue are potent mechanism-based retaining
β-glucosidase inhibitors. In this paper we explore the inhibitory potency of a number of cyclophellitol
analogues against the three human retaining β-glucosidases, GBA, GBA2 and GBA3. We demonstrate that
N-alkyl cyclophellitol aziridine is at least equally potent in inhibiting the enzymes evaluated as its N-acyl
congener, whereas the N-sulfonyl analogue is a considerably weaker inhibitor. Our results complement
the literature on the inhibitory potency of cyclophellitol analogues and hold promise for the future design
of more effective activity-based retaining glycosidase probes with respect to probe stability in physio-
logical media.
Received 29th July 2014,
Accepted 19th August 2014
DOI: 10.1039/c4ob01611d
www.rsc.org/obc
Introduction
Cyclophellitol 1, first isolated in 1990 from Phellinus sp., is a
potent and selective mechanism-based inhibitor of retaining
1
β-glucosidases. Cyclophellitol 1 is a close structural mimic
of β-glucopyranose that upon binding reacts covalently and
irreversibly with the active site nucleophile of retaining
β-glucosidases.
Retaining β-glucosidases, and retaining β-glycosidases in
general, employ a catalytic mechanism that is commonly
2
referred to as the Koshland double displacement mechanism.
This process (see Fig. 1A), involves two acidic amino acid resi-
dues (Glu, Asp) residing in the enzyme active site. In retaining
β-glucosidases, these acidic residues are positioned about
Fig. 1 Mechanism of retaining β-exoglucosidase-mediated hydrolysis
of β-glucosidic linkages (A) and mechanism-based, covalent and irre-
versible inhibition of β-glucosidases by the natural product, cyclo-
5
–6 Å apart such that one can act as the nucleophile and the
other as a general acid/base catalyst. When a retaining β-exo-
glucosidase binds to cyclophellitol 1, the inhibitor is posi- phellitol 1 (B).
tioned in such a way that the epoxide can undergo trans-
diaxial opening by consecutive protonation and nucleophilic
3
attack of the two catalytic carboxylic acid residues (Fig. 1B).
linkages. As a result, hydrolysis of the covalently bound enzyme-
An ester linkage is formed that is considerably more stable cyclophellitol adduct is slowed down to a point that the enzyme
than the corresponding acylal linkage that would result from is irreversibly inhibited.
the initial step in the enzymatic cleavage of beta-glucosidic
Following the discovery of cyclophellitol 1, Tatsuta and co-
workers reported the synthesis of cyclophellitol thiirane 2 and
cyclophellitol aziridine 3 (Fig. 2) as well as a number of
4
a
Leiden Institute of Chemistry, Leiden University, Einsteinweg 55, 2300 RA Leiden,
N-alkyl- and N-acyl cyclophellitol aziridine derivatives. Inhi-
the Netherlands. E-mail: h.s.overkleeft@chem.leidenuniv.nl
bition studies that were performed on almond β-glucosidase
revealed that the nature of the N-substituent on cyclophellitol
aziridine influences inhibitory potency. Following these initial
studies, research on cyclophellitol derivatives as glycosidase
b
Stratingh Institute for Chemistry, University of Groningen, Nijenborgh 7,
9
747 AG Groningen, the Netherlands
c
Department of Medical Biochemistry, Academic Medical Center, Meibergdreef 15,
1105 AZ Amsterdam, the Netherlands
5
†
These authors contributed equally to this work.
inhibitors has been relatively limited until recently, especially
7786 | Org. Biomol. Chem., 2014, 12, 7786–7791
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Results and discussion
Cyclohexene derivative 8 was reported in 2005 by Madsen and
1
3
co-workers in their work on the total synthesis of the natural
product, cyclophellitol 1. This intermediate, accessible in gram
1
3,14
quantities following essentially the literature route,
proved
suitable for the synthesis of all target compounds. We pre-
viously reported that mCPBA-mediated epoxidation of the
homo-allylic alcohol in 8 furnished the required β-epoxide in
the natural product, cyclophellitol
1 with good stereo-
1
4
selectivity. The synthesis of β-thiirane 2 however required
access to 1,6-epi-cyclophellitol 9, which we could obtain in
good quantities through epoxidation of 8 with in situ formed
methyl(trifluoromethyl)-dioxirane. In this way, the desired pro-
tected 1,6-epi-cyclophellitol 9 was obtained as the major
isomer (46%), with partially protected cyclophellitol 10 as the
minor isomer. Pd(OH) -catalysed hydrogenolysis followed by
perbenzoylation gave 1,6-epi-cyclophellitol 11. Transformation
of epoxide 11 into thiirane analogue 2 with complete inver-
2
Fig. 2 Target structures subject of the here-presented studies.
when compared to the multitude of reports describing the syn- sion of configuration was performed by refluxing 11 with
thesis and evaluation of deoxynojirimycin-type competitive glyco- N,N-dimethylthioformamide (DMTF) and trifluoroacetic acid.
6
sidase inhibitors. Arguably, this lack of interest stems from Ensuing deacetylation under Zemplen conditions furnished
the covalent irreversible mode of action exerted by this com- cyclophellitol thiirane 2 in poor overall yield, but in sufficient
pounds class – a feature not normally considered to be ideal quantities to conduct the planned inhibition studies. It should
for drug development. Indeed, research on deoxynojirimycin be noted that in their original synthesis of thiirane 2, Tatsuta
derivatives is often conducted with drug discovery aims in and co-workers employed a per-(methoxy)benzylated 1,6-epi-
4
mind, with a number of N-alkyl deoxynojirimycin derivatives cyclophellitol derivative. We attempted to adopt this scheme
7
currently in clinical use or in development. Cyclophellitol but could not reproduce this intermediate following the litera-
derivatives and more in general mechanism-based glycosidase ture procedure (Scheme 1).
inactivators are however receiving increasing attention as dis-
The synthesis of substituted cyclophellitol aziridines 4–7
covery tools in chemical glycobiology research. 2-Deoxy- started from partially protected cyclophellitol aziridine 12,
-fluoroglycosides have proven indispensable tool to unravel which we prepared from the Madsen intermediate as we
8
2
14
the mode of action of retaining glycosidases and to confirm reported recently. Selective N-alkylation of 12 with iodopentane
the double displacement mechanism originally proposed by
9
Koshland. Cyclophellitol derivatives in turn proved good starting
points for the development of reagents for activity-based profil-
1
0
ing of retaining glycosidases.
We have previously shown that substitution of cyclophellitol
at the primary alcohol with a BODIPY reporter group yielded
1
a highly potent and specific activity-based probe to monitor
the human lysosomal retaining β-glucosidase, GBA, in cell
1
1
extracts, in situ and in vivo. N-acyl-cyclophellitol aziridine
with (BODIPY or biotin) reporter groups occupying the space
normally reserved for the aglycon within an exoglucosidase
active site in turn proved broad-spectrum retaining β-glucosi-
1
2
dase probes. These activity-based retaining β-glucosidase
probes efficiently modified all four known human retaining
β-glucosidases, GBA, GBA2, GBA3 and LPH. These latter
results invite a more in-depth study of modified cyclophellitol
derivatives as mechanism-based inhibitors, and possible start-
Scheme 1 Reagents and conditions: (a) oxone, CF
EDTA (46% 9, 35% 10); (b) (i) H
O, DMAP, pyridine (26%); (c) (i); N,N-dimethylthio-
formamide, TFA, CH Cl , 40 °C; (ii) NaOMe, MeOH (7%); (d) (i) iodo-
pentane, K CO , DMF, 90 °C (61%); (ii) Li, NH , THF, −60 °C (70%); (e) see
ref. 14; (f) valeric acid, 2-ethoxy-1-ethoxycarbonyl-1,2-dihydroxyquinoline
EEDQ), DMF, 0 °C (23%); (g) benzoic acid, EEDQ, DMF, 0 °C (23%);
(h) butanesulfonyl chloride, NaHCO , DMF (36%).
3
OCH
3
, NaHCO
3
,
ing points for activity-based probe development for retaining acetonitrile, 4 mM Na
2
2
, Pd(OH)
2
/C,
MeOH (97%); (ii) Bz
2
β-glucosidases. We here report on the synthesis and inhibitory
potential of a number of N-substituted cyclophellitol aziridine
derivatives on the human retaining β-glucosidases, GBA, GBA2
and GBA3, in comparison with cyclophellitol (1), cyclophellitol
thiirane (2) and cyclophellitol aziridine (3).
2
2
2
3
3
(
3
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 7786–7791 | 7787

View Article Online
Paper
Organic & Biomolecular Chemistry
and potassium carbonate, followed by the removal of the corresponding aziridine derivative is however a viable retaining
benzyl ethers under Birch reduction afforded 7. The N-acyl and β-glucosidase inhibitor and N-alkyl, N-acyl and N-sulfonyl sub-
N-sulfonyl derivatives 4–6 were prepared from cyclophellitol stituents are all tolerated. A subtle though important improve-
aziridine 3, which was prepared from 12 by Birch reduction as ment is found in changing an N-acyl substituent to an N-alkyl
1
4
reported previously. Reaction of 12 with 1-butanesulfonyl- substituent. This modification leads to an improved inhibitor
chloride and sodium bicarbonate in dimethylformamide with respect to potency, but also to an intrinsically more stable
furnished sulfonamide 6. Treatment of 12 with a pre-activated compound. Apart from inhibitor design, our results may have
solution of valeric acid or benzoic acid with 2-ethoxy-1-ethoxy- a significant impact on the future design of activity-based
carbonyl-1,2-dihydroquinoline (EEDQ) furnished 4 and 5, probes directed to retaining β-glucosidases, and to retaining
respectively. As we noted before, acylated cyclophellitol aziri- glycosidases in general. Our current activity-based retaining
dines are prone to nucleophilic ring opening and in order to glycosidase probe design is based on N-acylation of aziridines.
suppress this unwanted process, compounds 4 and 5 were pur- While these probes are amenable for in vitro and in situ label-
ified by HPLC using either a neutral (acetonitrile–water) or ling of such enzymes, their intrinsic reactivity in both slightly
basic (25–50 mM aqueous NH HCO –acetonitrile) gradient. In acidic and slightly basic aqueous media make them relatively
4
3
contrast to the acylated cyclophellitol aziridines 4 and 5, difficult to handle. We are currently exploring the synthesis
N-sulfonyl aziridine 6 and N-alkyl aziridine 7 proved more stable and evaluation of the corresponding N-alkyl activity-based
and could be purified to homogeneity using standard silica gel probes and will report on the outcome of these studies in the
column chromatography.
near future.
The inhibitory potential of cyclophellitol 1 and analogues
2
–7 towards the three human retaining β-glucosidases GBA,
GBA2 and GBA3 was evaluated by determining their apparent
IC50 values (Table 1). To this end, the enzymes were incubated
with a concentration series of the inhibitor for 30 minutes fol-
Experimental
General
lowed by measuring the residual enzymatic activity using the All reagents and solvents were a commercial grade and used as
fluorogenic substrate 4-methylumbelliferyl β-D-glucopyranoside. received unless stated otherwise. THF and methylene chloride
4
In agreement with the literature values on almond β-gluco- were stored over flamed-dried 3 Å molecular sieves. All reac-
sidase, cyclophellitol thiirane 2 did not inhibit the three tions were performed under an inert atmosphere unless stated
human retaining β-glucosidases up to 100 μM. The corres- otherwise. Solvents used for flash chromatography were of pro
ponding aziridine 3 inhibits all enzymes with equal or slightly analysis quality. Reactions were monitored by TLC analysis
higher potency than 1 and subtle, though persistent, differ- using aluminum sheets pre-coated with silica gel 60 with
ences are observed in the series 4–7 when compared to 1. Acyl- detection by UV-absorption (254 nm) and by spraying with
−
1
ation as well as alkylation, but not sulphonylation, of the a solution of (NH ) Mo O ·H O (25 g L ) and (NH4)4-
4
6
7
24
2
4 4 2
aziridine results in compounds that effectively inhibit GBA3. Ce(SO ) ·H
O (10 g L− ) in 10% sulphuric acid followed by
1
Whereas N-sulfonylation, as in 6, results in a small drop in charring at ∼150 °C or by spraying with 20% sulphuric acid in
GBA/GBA2 inhibitory activity, N-acylation (4, 5) yields inhibi- ethanol followed by charring at ∼150 °C. Column chromato-
tors with potency about equal to that of unsubstituted aziri- graphy was performed using Screening Device silica gel and
1
13
dine 3. The most potent inhibitor of this series however turned the indicated solvents. H NMR, C NMR, COSY and HSQC
out to be N-alkyl derivative 7, which inhibits GBA and GBA2 spectra were recorded on a Bruker DMX-400 (400/100 MHz),
1
0–100 fold more effectively when compared to 1. This result Bruker AV-400 (400/100 MHz) and Bruker AV-600 (600/
is of importance since compound 7 appeared considerably 150 MHz) spectrometer in the given solvent. Chemical shifts
more stable during its synthesis and purification, and is pre- are reported as δ-values in ppm relative to the chloroform
sumably also more stable under physiological conditions.
residual solvent peak or tetramethylsilane (TMS) as internal
OD. Coupling
In conclusion, we have shown that modifications of the standard or the deuterated solvent signal for CD
3
1
3
epoxide in cyclophellitol (1) can have a profound effect on the constants are given in Hz. All given C spectra are proton
inhibitory potency. As is evident, substitution of the epoxide decoupled. High-resolution mass spectra were recorded with a
oxygen for sulphur is detrimental for inhibitory activity. The LTQ Orbitrap (Thermo Finnigan). HPLC-MS purifications were
performed on an Agilent Technologies 1200 series automated
HPLC system with a Quadrupole MS 6130, equipped with a
Table 1 Inhibitory potency (apparent IC50 values in μM) of compounds semi-preparative Gemini C18 column (Phenomenex, 250 × 10,
1
–7 against GBA, GBA2 and GBA3
5μ particle size).
,3-Di-O-benzyl-1,6-epi-cyclophellitol 9. A solution of
.4 mM Na EDTA solution in H O (21 mL) and trifluoro-
acetone (9.32 mL, 104 mmol) were added to 8 (2.36 g, 6.94 mmol)
in acetonitrile (6.7 mL). mixture of oxone (21.3 g,
4.7 mmol) and NaHCO (4.08 g, 48.6 mmol) was added to the
2
Inhibitor
0
2
2
Enzyme
1
2
3
4
5
6
7
A
GBA
GBA2
GBA3
0.35
2
>100
>100
>100
>100
0.5
0.4
>100
0.07
0.2
0.5
1.4
0.055
20
2.5
0.3
0.3
0.017
0.003
0.7
3
3
solution over a period of 15 min at 0 °C. After stirring at 0 °C
7788 | Org. Biomol. Chem., 2014, 12, 7786–7791
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
for 3 h, the reaction mixture was diluted with H
2
O, extracted reduced pressure and the resulting crude product was dis-
with EtOAc, dried over MgSO4 and concentrated under solved in MeOH (1 mL). A catalytic amount of NaOMe was
reduced pressure. Purification by silica gel column chromato- added to the solution and stirred for 5 h at ambient tempera-
graphy (hexanes–Et
2
O, 12 : 88 → 10 : 90 and 8 : 92 → 0 : 100) ture. The reaction mixture was neutralized with Amberlite
+
furnished 10 (767 mg, 2.15 mmol, 31%) and 9 (1.14 g, IR-120 H , filtered and concentrated in vacuo. Purification by
1
3
.19 mmol, 46%) as white crystals. H NMR (400 MHz, CDCl
3
): silica column chromatography (methylene chloride–MeOH,
δ 7.41–7.17 (m, 10H, HArBn), 4.97 (d, J = 11.2 Hz, 1H, CH Bn), 98 : 2 → 96 : 4) afforded 3 (3.7 mg, 19 μmol, 7.0%) as a white
2
1
4
1
1
3
.79 (dd, J = 12.0, 21.2 Hz, 2H, CH Bn), 4.64 (d, J = 11.2 Hz, solid. H NMR (400 MHz, MeOD): δ 4.09 (dd, J = 4.4, 10.4 Hz,
2
H, CH
2
Bn), 3.86–3.77 (m, 3H, H-2, H-8), 3.56 (t, J = 10.0 Hz, 1H, H-8), 3.97 (d, J = 7.2 Hz, 1H, H-2), 3.53 (dd, J = 8.4, 9.2 Hz,
H, H-3), 3.41 (t, J = 10.0 Hz, 1H, H-4), 3.34–3.33 (m, 1H, H-6), 1H, H-8), 3.50 (dd, J = 2.4, 4.8 Hz, 1H, H-6), 3.21–3.16 (m, 2H,
.10 (d, J = 3.6 Hz, 1H, H-1), 2.88 (br s, 1H, OH), 2.54 (br s, 1H, H-3, H-4), 3.07 (d, J = 6.4 Hz, 1H, H-1), 2.33–2.27 (m, 1H, H-5).
1
3
13
OH), 2.15–2.10 (m, 1H, H-5). C NMR (100 MHz, CDCl
δ 138.2, 137.9, 128.5, 128.4, 128.0, 127.9, 127.8, 80.8, 79.4, 41.3. HRMS: found 193.05346 [M + H] , calculated for
5.4, 72.5, 70.2, 62.5, 54.4, 54.0, 43.5. HRMS: found 357.16989 [C H O S] 193.05291.
3
):
3
C NMR (100 MHz, CDCl ): δ 79.9, 75.6, 68.7, 66.0, 45.9, 41.6,
+
7
7
13 4
+
[M + H] , calculated for [C21
H
25
O
5
] 357.16965.
2,3-Di-O-benzyl-7-pentylcyclophellitol aziridine. A solution
1
,6-epi-Cyclophellitol. A catalytic amount of Pd(OH)
2
was of 12 (91 mg, 0.21 mmol), 1-iodopentane (60 μL, 0.46 mmol)
added to a solution of 9 (170 mg, 0.48 mmol) in MeOH and potassium carbonate (87 mg, 0.63 mmol) in DMF was
2.5 mL). The solution was stirred under H atmosphere for stirred at 90 °C for 18 h. The reaction mixture was diluted with
h and next filtered over a small pad of celite and con- water and subsequently extracted with Et O, dried over MgSO
(
2
6
2
4
centrated under reduced pressure. The crude product was and concentrated under reduced pressure. Purification by
resuspended in chloroform and filtered to yield 1,6-epi-cyclo- silica column chromatography (methylene chloride–MeOH,
1
phellitol as white crystals (82.0 mg, 0.47 mmol, 97%). H NMR 97 : 3 → 96 : 4) gave the title compound (56 mg, 0.13 mmol,
1
(
400 MHz, MeOD): δ 3.93–3.89 (m, 2H, H-2, H-8), 3.78 (dd, J = 61%) as colourless oil. H NMR (400 MHz, CDCl ): δ 7.39–7.26
3
5
.6, 11.2 Hz, 1H, H-8), 3.46 (dd, J = 2.0, 4.4 Hz, 1H, H-6), 3.41 (m, 10H, HArBn), 4.97 (d, J = 11.2 Hz, 1H, CH
2
Bn), 4.77 (d, J =
Bn),
(dd, J = 8.4, 10.0 Hz, 1H, H-3), 3.36–3.31 (m, 2H, H-1, H-4), 11.6 Hz, 1H, CH Bn), 4.65 (dd, J = 9.2, 11.6 Hz, 2H, CH
2
2
1
3
2
.05–2.02 (m, 1H, H-5). C NMR (100 MHz, CDCl ): δ 73.0, 3.95 (dd, J = 6.4, 10.8 Hz, 1H, H-8), 3.87 (dd, J = 4.4, 10.8 Hz,
3
7
1.2, 69.3, 60.3, 57.5, 55.1, 44.1. HRMS: found 177.07576 1H, H-8), 3.71 (d, J = 8.4 Hz, 1H, H-2), 3.53–3.45 (m, 2H, H-4,
+
[M + H] , calculated for [C
7
H
12
O
5
] 177.07575.
OH), 3.31 (dd, J = 8.4, 10.0 Hz, 1H, H-3), 2.78 (br s, 1H, OH),
2
,3,4,8-Tetra-O-benzoyl-1,6-epi-cyclophellitol 11. 1,6-epi-Cyclo- 2.27 (dt, J = 7.2, 14.8 Hz, 1H, CH alkyl), 2.12–2.05 (m, 1H, CH
2
2
phellitol was taken up in pyridine (2.5 mL). Benzoic anhydride alkyl), 2.02–1.97 (m, 1H, H-5), 1.72 (dd, J = 3.6, 6.0 Hz, 1H,
1.09 g, 4.8 mmol) and a catalytic amount of DMAP were H-6), 1.55 (d, J = 6.0 Hz, 1H, H-1), 1.47 (q, J = 7.2 Hz, 2H, CH
added to the solution and the reaction mixture was stirred at alkyl), 1.31–1.19 (m, 4H, CH alkyl), 0.90 (t, J = 6.4 Hz, 3H,
(
2
2
1
3
5
0 °C for 18 h. The mixture was quenched with 1 M HCl, CH
extracted with EtOAc, dried over MgSO and concentrated 128.5, 128.4, 128.0, 127.9, 127.8, 127.7, 84.1, 81.0, 74.7, 72.1,
under reduced pressure. Purification by silica gel column 69.6, 68.1, 65.0, 61.0, 42.5, 41.5, 40.3, 29.5, 29.0, 22.5, 14.0.
3 3
alkyl). C NMR (100 MHz, CDCl ): δ 138.4, 137.8, 128.8,
4
+
4
chromatography (hexanes–EtOAc, 84 : 16 → 80 : 20) afforded 11 HRMS: found 427.26619 [M + H] , calculated for [C26H36NO ]
1
(73 mg, 0.12 mmol, 26%) as a clear oil. H NMR (400 MHz, 427/26724.
CDCl ): δ 8.14–8.02 (m, 5H, H Bz), 7.86 (d, J = 7.2 Hz, 1H,
7-Pentylcyclophellitol aziridine 7. Ammonia (5 mL) was con-
3
Ar
H
H
ArBz), 7.78 (d, J = 7.2 Hz, 1H, HArBz), 7.65–7.59 (m, 1H, densed at −60 °C. Lithium (22 mg) was added and the mixture
ArBz), 7.57–7.33 (m, 8H, HArBz), 7.31–7.27 (m, 2H, HArBz), was stirred for 30 min at −60 °C until lithium was completely
7
.21 (t, J = 7.6 Hz, 2H, H Bz), 5.99 (dd, J = 9.2, 10.4 Hz, 1H, dissolved. To this solution was added a solution of 2,3-di-O-
Ar
H-3), 5.82 (dd, J = 2.4, 9.2 Hz, 1H, H-2), 5.67 (t, J = 10.4 Hz, 1H, benzyl-7-pentylcyclophellitol aziridine (56 mg, 0.13 mmol) in
H-4), 4.67 (dd, J = 3.6, 11.6 Hz, H-8), 4.49 (dd, J = 5.6, 11.6 Hz, THF (3.0 mL). The reaction mixture was stirred for 30 min at
1
1
H, H-8), 3.79 (dd, J = 1.6, 3.6 Hz, 1H, H-6), 3.48 (d, J = 4.0 Hz, −60 °C and subsequently quenched with milliQ-water (2 mL).
H, H-1), 3.00–2.96 (m, 1H, H-5). C NMR (100 MHz, CDCl ): The solution was allowed to come to ambient temperature and
3
1
3
δ 166.2, 166.1, 165.8, 165.6, 133.6, 133.4, 133.4, 133.3, stirred until all ammonia had evolved. Next, the solution was
1
1
1
33.1, 133.0, 130.1, 129.9, 129.8, 129.7, 129.7, 129.5, 129.4, concentrated in vacuo, dissolved in milliQ-water and neutral-
29.0, 128.9, 128.8, 128.7, 128.5, 128.5, 128.4, 128.3, 128.3, ized with Amberlite IR-120 H . Product bound to the resin was
+
4
28.2, 128.1, 127.8, 72.4, 70.2, 69.3, 62.9, 54.4, 53.6, 40.7. washed with water and subsequently eluted with a 1 M NH OH
+
HRMS: found 593.18069 [M + H] , calculated for [C H O ] solution and evaporated under reduced pressure. The resulting
3
5
29 9
+
5
93.18061.
Cyclophellitol thiirane 2. N,N-Dimethylthioformamide milliQ-water as eluent until the eluate was neutral. Evaporation
24.5 μL, 0.29 mmol) and trifluoroacetic acid (10.7 μL, of the combined eluate under reduced pressure gave crude 7,
.14 mmol) was added to a solution of 11 (73 mg, 0.12 mmol) which was purified by silica column chromatography (methyl-
in anhydrous methylene chloride (7.2 mL). After stirring for ene chloride–MeOH, 92 : 8) yielding 7 (8.2 mg, 34 μmol, 26%)
solid was again purified on Amberlite IR-120 NH
4
using
(
0
1
1
8 h at 40 °C, the reaction mixture was concentrated under as a white solid. H NMR (400 MHz, MeOD): δ 3.97 (dd, J = 4.4,
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 7786–7791 | 7789

View Article Online
Paper
Organic & Biomolecular Chemistry
1
9
1
0.0 Hz, 1H, H-8), 3.62–3.54 (m, 2H, H-2, H-8), 3.08 (t, J =
7-(N-Butylsulfonyl)-cyclophellitol aziridine 6. A solution of
.6 Hz, 1H, H-3), 3.00 (t, J = 9.6 Hz, 1H, H-4), 2.33 (dt, J = 8.0, cyclophellitol aziridine 3 (18 mg, 0.1 mmol), sodium hydrogen
1.6 Hz, 1H, CH alkyl), 2.16–2.10 (m, 1H, CH alkyl), 1.98 carbonate (56 mg, 0.67 mmol) and 1-butanesulfonyl chloride
2
2
(dd, J = 3.6, 6.0 Hz, 1H, H-6), 1.89–1.83 (m, 1H, H-5), 1.63 (d, (13 μmol, 0.1 mmol) in DMF was stirred at ambient tempera-
J = 6.4 Hz, H-1), 1.56 (q, J = 7.6 Hz, 2H, CH alkyl), 1.32–1.27 ture for 18 h. The reaction mixture was filtered over a small
2
1
3
(m, 4H, CH
2 3
alkyl), 0.90 (t, J = 1.6 Hz, 3H, CH alkyl). C NMR pad of celite and the filtrate was concentrated under reduced
(
3
100 MHz, MeOD): δ 79.0, 73.9, 70.1, 63.7, 62.2, 45.5, 43.0, pressure. Purification by silica column chromatography
+
0.7, 30.1, 23.7, 14.4. HRMS: found 246.16987 [M + H] , calcu- (methylene chloride–MeOH, 92 : 8) afforded 6 (11 mg, 36 μmol,
1
lated for [C12
H
24NO
4
] 246.16998.
36%) as a colourless oil. H NMR (400 MHz, MeOD): δ 4.02
7
-(N-Pentoyl)-cyclophellitol aziridine 4. A pre-activated (dd, J = 4.4, 10.4 Hz, 1H, H-8), 3.68 (d, J = 8.0 Hz, 1H, H-2),
mixed anhydride solution (1 M) was prepared by dissolving 3.54 (t, J = 9.6 Hz, 1H, H-8), 3.32–3.23 (m, 2H, CH sulfonyl
2
EEDQ (209 mg, 0.85 mmol) and valeric acid (0.09 mL, and MeOD solvent peak), 3.21–3.15 (m, 2H, H-3, H-6), 3.03
0
.85 mmol) in DMF (0.85 mL) and the reaction mixture was (t, J = 10.0 Hz, 1H, H-4), 2.87 (d, J = 6.8 Hz, 1H, H-1), 2.03–1.96
stirred at ambient temperature for 2 h before use. Cyclophelli- (m, 1H, H-5), 1.87–1.79 (m, 2H, CH2 sulfonyl), 1.53–1.44
1
3
tol aziridine 3 (21 mg, 0.12 mmol) was dissolved in DMF (m, 2H, CH
2 3
sulfonyl), 0.95 (t, 3H, CH sulfonyl). C NMR
(
0.6 mL) and the solution was cooled to 0 °C before 1 M pre- (100 MHz, MeOD): δ 78.6, 73.0, 69.2, 63.0, 52.3, 44.5, 43.7,
+
activated mixed anhydride solution (0.06 mL, 0.06 mmol) was 26.3, 22.6, 13.9. HRMS: found 308.11279 [M + H] , calculated
added. The reaction mixture was stirred at 0 °C for 30 min and for [C11
additional pre-activated solution (0.06 mL, 0.06 mmol) was
6
H22NO S] 308.11623.
Enzyme activity assays – determination of IC50 values
added. After stirring for 30 min at 0 °C, the reaction mixture
was quenched with MeOH and concentrated under reduced
pressure. Purification by semi-preparative reversed phase
HPLC (linear gradient: 13% → 16%, 3CV, solutions used: A:
The activity of GBA was assayed at 37 °C by incubating
with 3.75 mM 4-methylumbelliferyl-β-D-glucopyranoside (4MU-
β-D-Glc) as substrate in 150 mM McIlvain buffer, pH 5.2, con-
taining 0.1% (w/v) BSA, 0.2% (w/v) sodium taurocholate, and
0.1% (v/v) Triton X-100. Purified recombinant GBA (Cerezyme,
Genzyme, USA) was employed as enzyme source. GBA3 activity
was measured at 37 °C in 150 mM McIlvain buffer, pH 6.0,
containing 0.1% (w/v) BSA and 3.75 mM substrate. A human
splenectomized N370S/RecNcI Gaucher spleen was homo-
genized by sonication in cold Nanopure water, centrifuged for
30 min at 12 000 rpm (Sorvall RC-5b, Dupont Instruments,
rotor SS-34), and supernatant was used as source for GBA3.
The pellet was homogenized by sonication in cold water,
washed thrice by centrifugation, vide supra, resuspended in
2
H O, B: acetonitrile) gave the title compound (7.1 mg,
1
2
7 μmol, 23%) as a white solid. H NMR (400 MHz, MeOD):
δ 4.05 (dd, J = 4.4, 10.4 Hz, 1H, H-8), 3.70–3.65 (m, 2H, H-2,
H-8), 3.19 (dd, J = 8.0, 10.0 Hz, 1H, H-3), 3.06 (t, J = 9.6 Hz, 1H,
H-4), 3.01 (dd, J = 3.2, 6.0 Hz, 1H, H-6), 2.71 (d, J = 6.0, Hz, 1H,
H-1), 2.49 (dt, J = 1.2, 7.2 Hz, 2H, CH alkyl), 1.99–1.93 (m, 1H,
2
H-5), 1.62–1.55 (m, 2H, CH
alkyl), 0.93 (t, J = 7.2 Hz, CH
MeOD): δ 188.6, 79.1, 73.4, 69.3, 63.5, 62.2, 45.3, 42.4, 41.0,
2 2
alkyl), 1.40–1.30 (m, 2H, CH
1
3
3
alkyl). C NMR (100 MHz,
+
3
6.7, 28.3, 23.4, 14.2. HRMS: found 260.14926 [M + H] , calcu-
lated for [C12 ] 260.14925.
-(N-Benzoyl)-cyclophellitol aziridine 5. A solution of EEDQ
84 mg, 0.34 mmol) and benzoic acid (41 mg, 0.34 mmol) were
H
22NO
5
50 mM potassium phosphate buffer, pH 5.8, and incubated
7
with 5 mM conduritol B epoxide for 30 min at 4 °C before
being used as GBA2 source. Activity of GBA2 was measured at
(
dissolved in DMF (0.34 mL) and the solution was stirred at
room temperature for 2 h. Pre-activated mixed anhydride solu-
tion (0.17 μL, 0.17 mmol) was added to cyclophellitol aziridine
37 °C in 150 mM McIlvaine buffer, pH 5.8, containing 0.1%
(
w/v) BSA and 3.75 mM substrate. To determine the apparent
IC50 value, enzyme was pre-incubated with a range of inhibitor
dilutions for 30 min at 37 °C, prior to the addition of 4MU-
β-D-Glc. After quenching the substrate reaction with excess NaOH-
glycine (pH 10.3), fluorescence was measured with a fluori-
meter LS30 (Perkin Elmer) using λEX 366 nm and λEX 445 nm.
3
(46 mg, 0.26 mmol) in DMF (1.0 mL) at 0 °C. After stirring
for 30 min at 0 °C, an additional amount of pre-activated
mixed anhydride solution (0.17 μL, 0.17 mmol) was added and
the reaction mixture was stirred for 2 h at 0 °C and then
quenched with MeOH (1 mL) and concentrated in vacuo. Puri-
fication by semi-preparative reversed HPLC (linear gradient:
1
0% → 16%, 3CV, solutions used A: H
2
O, B: acetonitrile) and
Acknowledgements
lyophilization yielding 5 (17 mg, 0.06 mmol, 23%) as a white
1
powder. H NMR (400 MHz, MeOD): δ 7.97 (d, J = 7.2 Hz, 2H, The China Scholarship Council (CSC, to JBJ), the Netherlands
ArBz), 7.59 (t, J = 7.2 Hz, 1H, HArBz), 7.48 (t, J = 7.6 Hz, 2H, Organization for Scientific Research (NWO-CW, to HSO) and
H Bz), 4.04–4.00 (m, 2H, H-2, H-8), 3.73 (dd, J = 8.0, 10.4 Hz, the European Research Council (ERC, Advanced Grant to HSO)
H
Ar
1
1
H, H-8), 3.31–3.22 (m, 2H, H-3, H-4), 3.19 (dd, J = 3.2, 6.0 Hz, are acknowledged for the financial support.
H, H-6), 2.72 (d, J = 6.0 Hz, 1H, H-1), 2.08–2.00 (m, 1H, H-5).
1
3
C NMR (100 MHz, MeOD): δ 181.5, 134.3, 134.1, 134.0, 130.1,
References
1
29.7, 129.5, 128.5, 79.3, 73.5, 69.7, 63.7, 45.3, 45.1, 40.8.
+
HRMS: found 280.11765 [M + H ], calculated for [C14
H
17NO
4
]
1 (a) S. Atsuma, K. Umezawa, H. Iinuma, H. Naganawa,
Y. Iitaka and T. Takeuchi, J. Antibiot., 1990, 43, 49;
2
80.11795.
7790 | Org. Biomol. Chem., 2014, 12, 7786–7791
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
(
b) S. Atsumi, H. Iinuma, C. Nosaka and K. Umezawa,
Chem. Sci., 2013, 4, 897; (d) D. Koshland, Biol. Rev., 1953,
28, 416.
J. Antibiot., 1990, 43, 1579.
2
3
D. Koshland, Biol. Rev., 1953, 28, 416.
10 L. I. Willems, J. Jiang, K.-Y. Li, M. D. Witte, W. W. Kallemeijn,
T. J. N. Beenakker, S. P. Schröder, J. M. F. G. Aerts,
G. A. van der Marel, J. D. C. Codée and H. S. Overkleeft,
Chem. – Eur. J., 2014, DOI: 10.1002/chem.201404014.
(a) S. G. Withers and K. Umezawa, Biochim. Biophys. Res.
Commun., 1991, 177, 532; (b) T. M. Gloster, R. Madsen and
G. J. Davies, Org. Biomol. Chem., 2007, 5, 444.
4
(a) M. Nakata, C. Chong, Y. Nitawa, K. Toshima and 11 M. D. Witte, W. W. Kallemeijn, J. Aten, K.-Y. Li, A. Strijland,
K. Tatsuta, J. Antibiot., 1993, 46, 1919; (b) K. Tatsuta, Y. Niwata,
K. Umezawa, K. Toshima and M. Nakata, J. Antibiot., 1991, 44,
W. E. Donker-Koopman, A. M. C. H. van den Nieuwendijk,
B. Bleijlevens, G. Kramer, B. I. Florea, B. Hooibrink,
C. E. M. Hollak, R. Ottenhoff, R. G. Boot, G. A. van der
Marel, H. S. Overkleeft and J. M. F. G. Aerts, Nat. Chem.
Biol., 2010, 6, 907.
912; (c) K. Tatsuta, Pure Appl. Chem., 1996, 68, 1341.
5
6
J. Marco-Contelles, Eur. J. Org. Chem., 2001, 1607.
(a) N. Ishida, K. Kumagai, T. Niida, T. Tsuruoka and
H. Yumoto, J. Antibiot., 1967, 20, 66–71; (b) S. Inouye, 12 W. W. Kallemeijn, K.-Y. Li, M. D. Witte, A. R. A. Marques,
T. Tsuruoka, T. Ito and T. Niida, Tetrahedron, 1968, 23, 2125.
(a) T. Wennekes, R. J. B. H. N. van den Berg, R. G. Boot,
G. A. van der Marel, H. S. Overkleeft and J. M. F. G. Aerts,
Angew. Chem., Int. Ed., 2009, 48, 8848; (b) T. D. Butters,
R. A. Dwek and F. M. Platt, Curr. Top. Med. Chem., 2003, 3, 561.
J. Aten, S. Scheij, J. Jiang, L. I. Willems, T. M. Voorn-
Brouwer, C. P. A. A. van Roomen, R. Ottenhoff, R. G. Boot,
H. van den Elst, M. T. C. Walvoort, B. I. Florea,
J. D. C. Codée, G. A. van der Marel, J. M. F. G. Aerts and
H. S. Overkleeft, Angew. Chem., Int. Ed., 2012, 51, 12529.
7
8
9
(a) M. D. Witte, G. A. van der Marel, J. M. F. G. Aerts and 13 F. G. Hansen, E. Bundgaart and R. Madsen, J. Org. Chem.,
H. S. Overkleeft, Org. Biomol. Chem., 2011, 9, 5908;
b) K. A. Stubbs, Carbohydr. Res., 2014, 390, 9.
2005, 70, 10139.
(
14 K.-Y. Li, J. Jiang, M. D. Witte, W. W. Kallemeijn, H. van den
Elst, C.-S. Wong, S. Chanders, S. Hoogendoorn,
T. Beenakker, J. D. C. Codée, J. M. F. G. Aerts,
G. A. van der Marel and H. S. Overkleeft, Eur. J. Org. Chem.,
DOI: 10.1002/ejoc.201402588.
(a) D. J. Vocadlo, G. J. Davies, R. Laine and S. G. Withers,
Nature, 2001, 412, 835; (b) B. P. Rempel and S. G. Withers,
Glycobiology, 2008, 18, 570; (c) M. T. C. Walvoort,
G. A. van der Marel, H. S. Overkleeft and J. D. C. Codée,
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 7786–7791 | 7791